- Email: [email protected]
Performance analysis of an automatic idle speed control system with a hydraulic accumulator for pure electric construction machinery
Automation in Construction 84 (2017) 184–194
Contents lists available at ScienceDirect
Automation in Construction journal homepage: www.elsevier.com/locate/autcon
Performance analysis of an automatic idle speed control system with a hydraulic accumulator for pure electric construction machinery
MARK
Tianliang Lin, Lang Wang, Weiping Huang, Haoling Ren⁎, Shengjie Fu, Qihuai Chen College of Mechanical Engineering and Automation, Huaqiao University, 361021 Xiamen, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Construction machinery Hydraulic excavator Energy saving Idle speed Hydraulic accumulator Pure electric system
To reduce the energy consumption and emissions of a hydraulic excavator (HE), an electric motor (EM) is employed to replace the internal combustion engine (ICE) that powers the hydraulic pumps. Owing to the excellent control characteristics and high efficiency of the EM, a two-level idle speed control system with a hydraulic accumulator (HA) for a HE is proposed to reduce energy consumption and improve the control performance of the actuator when the idle speed control (ISC) is switched off. A mathematical model is established and key parameters are analyzed and optimized. A simulation is performed using AMESim, and a control strategy for the two-level idle speed control is developed by using a co-simulation between AMESim and Simulink. A test rig is built based on the optimized parameters and simulation results. Experimental results show that the EM speed can be automatically switched between the first idle speed, second idle speed, and normal operating speed. Although the idle speed of the EM in the novel ISC system can be reduced more than that in a conventional ISC system, the proposed ISC system can still build actuator pressure more quickly in a working mode when the ISC is switched off. Compared to a system without idle speed control, the energy saving of the proposed system is approximately 36.06%. The proposed two-level idle speed control system with a HA can achieve high energy efficiency and excellent control performance, and it can be also applied to engine-driven construction machinery.
1. Introduction Recently, energy efficiency and a low environmental impact have become the basic requirements for construction machinery to meet increasingly stringent emission regulations. Thus, research on energy saving methods for construction machinery, especially hydraulic excavators (HEs), is necessary and urgent owing to their high energy consumption and emissions. Technologies that aim to reduce energy consumption and emissions, such as positive and negative flow control systems, load sensing controls, hybrid systems, pure electric systems, energy regeneration, and idle speed control (ISC) have been proposed and utilized in various models. Both flow-control systems [1,2] and load-sensing controls [3,4] are to adjust the output of pumps to meet the requirements of a load and avoid extra energy loss. Hybrid systems utilize more than one power train to allow the engine to operate at an ideal output level to reduce fuel consumption and emissions [5–7], but they cannot achieve zero emissions owing to the use of internal combustion engines (ICEs). Energy recovery systems (ERSs) are typically used to regenerate potential energy when the load is down or kinetic energy when the actuator is braking [8–11].
⁎
The traditional energy saving methods used in HEs cannot significantly improve energy savings if there are no further technological breakthroughs. The successful application of pure electric systems in vehicles provides a new method for HEs to achieve energy savings. A pure electric system can achieve zero emissions and lower energy losses [12]. Equipping HEs with pure electric systems has become the major trend in energy saving research. There have been only a few studies on pure electric HE systems. Because ISC is considered a good method for fundamentally reducing energy consumption [13], this paper focuses on a new ISC system for a pure electric HE. The working process of construction machinery is typically periodic. Idle time accounts for approximately 30% of a typical working cycle [14]. ISC is adapted to reduce energy loss and emissions during idle time. Traditional ISC is applied to control the speed switch between two values in ICEs. Generally, idle speed should be as low as possible to improve fuel economy and reduce emissions [15]. He et al. investigated a new engine and reduced the idle speed from the original 800 rpm to 700 rpm [16]. Li et al. set the idle speed in an engine to 611 rpm instead of the normal (production) idle speed of 740 rpm and used a sliding mode control for ISC. Although a low idle speed can reduce fuel
Corresponding author. E-mail address: [email protected] (H. Ren).
http://dx.doi.org/10.1016/j.autcon.2017.09.001 Received 2 September 2016; Received in revised form 10 August 2017; Accepted 2 September 2017 0926-5805/ © 2017 Published by Elsevier B.V.
Automation in Construction 84 (2017) 184–194
T. Lin et al.
The working mode recognition process of the ISC system in Fig. 1 is presented in Fig. 3. When the joystick returns to the middle position, the controller detects that the pressure difference between the two pressures on the joystick is under a preset small positive value and sends signals to the multi-way valve to turn it to the middle position. The proposed ISC system is used to control the speed of the EM as it switches between different values, including the switch from high speed to low speed and from low speed to high speed. When reducing the speed of the EM, the main consideration is energy saving. The value of the idle speed is the primary factor that influences energy consumption. When the working mode of the ISC system is switched off, control performance is the main consideration. Therefore, the time required for building pressure in the pump should be smooth and short. By taking the self-suction capacity of the pump into account and referring to Fig. 2, the first level idle speed is set to 800 rpm and the second level idle speed is set to 500 rpm because the HA can help the pump build pressure. During the various processes, the control strategy changes accordingly. 1) First level ISC When the time that the joystick resides in the middle position is longer than time T1, the controller reduces the EM speed to the first level idle speed n1 to reduce energy loss. Meanwhile, the maximum load pressure is used to determine whether the HA should charge the pump. When the pressure difference between the HA and the maximum load pressure is larger than a set value, solenoid directional valves 1 and 2 are powered on. The pump then charges the HA through solenoid directional valve 1. After the HA pressure increases to the set value, solenoid directional valves 1 and 2 are powered off. 2) Second level ISC The control target is to achieve minimum energy consumption. The output power of the system should outweigh the total energy loss and minimize energy consumption. When the time that the joystick resides in the middle position is longer than time T1 + T2 and the pressure difference between the maximum actuator pressure and the HA pressure is under a threshold value, the controller switches the EM to the second level idle speed n2, which is lower than the first level idle speed n1. 3) ISC is switched off When the actuator recovers from the idle speed and the EM returns to normal speed, which is approximately 1800 rpm, the global positive flow control is used to make the flow rate supplied by the pump and the HA match the requirement of the load. Solenoid directional valves 1 and 2 are powered on. The stored oil in the HA is released through solenoid directional valve 1 to the inlet of the multi-way valve to help the pump build pressure quickly and drive the actuator. When the HA pressure is lower than the pump pressure, solenoid directional valve 1 is powered off and the actuator is driven by the pump only. The proportion of the flow rate supplied by the HA and the pump is scheduled to achieve smooth and quick movement of the actuator.
consumption, it carries an increased risk of misfiring and engine stalling [17,18]. Thus, the transition to and from idle speed must be smooth and carefully controlled [18]. Many researchers have devoted themselves to solving ISC problems with various control strategies [19–21]. Most of the research has focused on ISC for automobiles. However, owing to the intrinsic properties of an ICE, the speed range is narrow and the idle speed remains high, which limits the potential energy savings of ISC. However, there have been a few studies on construction machinery. Xiong et al. analyzed construction machinery working principles and implemented methods for ISC in a rotary drilling rig [22]. They tested fuel consumption under various speeds and found that the speed with the lowest fuel consumption was the preset idle speed. Liu designed an ISC system based on working conditions, speed sensing, and engine power matching [23]. The adjustment time for the engine between idle speed and the rated speed was approximately two seconds. Hao optimized the duty ratio of pulse wave modulation, minimum ramping time, and idle speed through an adaptive control method and achieved superior energy savings [24]. However, these studies were still based on the ICE and the idle speed could not be set to a low value, causing relatively low energy efficiency. Furthermore, the pump could not build pressure quickly enough to drive the actuator when the working mode of the ISC was switched off. ISC is adapted to reduce energy loss and emissions during idle time. When ISC is applied to ICEs, it can also reduce fuel consumption and greenhouse gas emissions. However, owing to the intrinsic properties of ICEs, the speed range is narrow and the idle speed remains high, which limits the energy saving potential of ISC and makes it difficult to meet requirements for environmental protection. When ISC is applied to pure electric systems, the control performance of the electric motor (EM) is different from the ICE, particularly the speed range, dynamic response, and efficiency. Therefore, in this research, a two-ton class pure electric HE was built in our lab for experimentation. A novel ISC with a hydraulic accumulator (HA) for the two-ton HE is the focus of this research. The remainder of this paper is organized as follows: The structure and working principle of the automatic ISC system with a HA is described in Section 2. A mathematical model of the proposed ISC system is described in Section 3. The influence of various parameters on the performance characteristics of the ISC system is discussed in detail in Section 4. The experimental results are analyzed in Section 5. A summary and our final conclusions are provided in Section 6. 2. Structure and working principle of the novel automatic ISC system Fig. 1 presents a schematic of the proposed automatic ISC system. Compared to traditional ISC systems, the proposed ISC system has the following improvements: 1) Pure electric drive enables zero emissions for the driving system. An EM is used in place of an ICE for driving the hydraulic pumps. The efficiency of the EM is presented in Fig. 2 and can reach 95% when the EM speed is under 500 rpm, and up to 96%–98% when the EM speed is above 800 rpm. The EM has higher efficiency at low speeds, a wider speed range, and a quicker dynamic response than an ICE. 2) A HA is connected to the outlet of the pump via solenoid directional valve 1. The HA can provide auxiliary power to drive the actuator when the working mode of ISC is switched off. The pump can then build pressure quickly enough to drive the actuator. This means that the idle speed can be set to a low value to obtain good energy savings and decrease noise. 3) There is a pressure loading unit at the return port when the multiway valve is in the middle working position. It can separate the pump from the tank via solenoid directional valve 2 and charge the HA based on the control strategy. 4) All pressures of the key components are detected and used to recognize the working mode of the ISC system.
3. Mathematical model of the ISC system with a HA The simplified structure of the proposed ISC system presented in Fig. 4 is used to analyze the control performance of the novel ISC system. The assumptions made are as follows: 1) The influence of the solenoid directional valve on the velocity characteristics of the actuator is not considered. 2) Only the piston-out movement of the actuator is considered. The influence of piston movements on the volume between the non-rod chamber, HA, and the pump is ignored. 3) The actuator and pump have no elastic loads or disturbance torque. 4) The safety valve, which is connected to the pump output port, does not work and overflow during the working process. 5) The pressure of the inlet port of the pump is zero.
185
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 1. Schematic of the proposed automatic ISC system.
where qp is the displacement of the pump in m3/rad. Cip and Cep are the internal and external leakage coefficient of the pump, respectively, in m3/(Pa·s). Ctp is the total leakage coefficient of the pump in m3/(Pa·s) and Ctp = Cip + Cep. pp is the output port pressure of the pump in Pa. (3) Torque equation for the load of the pump and EM As mentioned in the assumptions, there are no elastic loads or disturbance torques. Thus, the load torque equation for the pump is
TL = qp⋅pp
(3)
and the torque equilibrium equation is
Tm − TL = J
(4) The flow equation for the proportional directional valve (multi-way valve) is
(1) Electromagnetic torque equation of the inverter-EM The EM employed in this paper is asynchronous. Considering that the setup time for the electromagnetic torque of the inverter-EM is much shorter than the mechanical response time of the pump-EM, the torque of the inverter and EM can be simplified as a proportion:
QC = K Q x v + K C (pp − pLb ),
(5)
where QC is the flow rate through the proportional directional valve in m3/s. KQ is the flow gain of the proportional directional valve in m2/s. KC is the flow-pressure coefficient of the proportional directional valve in m3/(Pa·s). xv is the opening displacement of the proportional directional valve port in m. pLb is the non-rod chamber pressure of the actuator in Pa.
(1)
where Tm is the electromagnetic torque of the EM in N·m. Km is the proportional factor. ωt and ω are the target angular velocity and realtime angular velocity of the EM, respectively, in rad/s.
(5) The flow equation for the HA is
(2) Output flow rate of the pump:
Qp = qp ω − (Cip + Cep ) pp = qp ω − Ctp pp ,
(4)
where TL is the load torque of the pump in TLN·m. J is the total inertial moment of the pump, EM, and coupling in kg·m2. bm is the viscous damping coefficient when the pump rotates in N·s/rad.
Fig. 2. Efficiency map of the EM.
Tm = Km (ωt − ω)
dω + bm ω, dt
Qac =
(2) 186
V0 dpac , p0 dt
(6)
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 3. Working mode recognition process of the ISC system.
where Qac is the flow rate of the HA in m3/s. p0 and pac are the precharge pressure and working pressure of the HA, respectively, in Pa. V0 is the rated volume of the HA in m3.
chamber is zero, specifically, pLs = 0. Thus, Eq. (7) can be rewritten as
K Q x v + K C pp − (K C + Cta ) pLb − A1 va =
VLV dpLb . βe dt
(9)
(6) Continuity equation for the oil (7) Force balance equation for the actuator: The continuity equation for the actuator is
m VLV dpLb + A1 va = Qc − Cia (pLb − pLs ) − Cea pLb . βe dt
βe
dt
= Qp + Qac − Qc ,
(10)
where m is the total mass of the piston and the equivalent mass of the load on the piston of the actuator in kg. bc is the viscous damping coefficient of the piston and the load of the actuator in N/(m/s). Fcom is the coulomb friction when the piston moves out in N. Based on Eqs. (1)–(10), a Laplace transformation can be obtained as follows:
(7)
The continuity equation for the pump is
VpaV dpp
dva = pLb A1 − bc va − Fcom, dt
(8)
where βe is the effective bulk modulus in Pa. VpaV is the volume between the pump, HA, and proportional directional valve in m3. VLV is the volume between the non-rod chamber of the actuator and the proportional directional valve in m3. A1 is the effective area of the nonrod chamber of the actuator in m2. va is the velocity of the actuator in m/s. Cia and Cea are the internal and external leakage coefficients of the actuator, respectively, in m3/(Pa·s). Cta is the total leakage coefficient of the actuator in m3/(Pa·s) and Cta = Cia + Cea. When the piston of the actuator moves out, the pressure in the rod
Km ωt − qp⋅pp (s ) = (Js + Km + bm ) ω (s )
qp ω (s ) − K Q x v (s ) + K C pLb (s ) +
V0 s V s p (s ) = ⎜⎛ LV + Ctp + K C ⎟⎞ pp (s ) p0 ac ⎝ βe ⎠ (12)
VpaV s K Q x v (s ) + K C pp (s ) − A1 va (s ) = ⎜⎛ + Ctc + K C ⎟⎞ pLb (s ) ⎝ βe ⎠ 187
(11)
(13)
Automation in Construction 84 (2017) 184–194
T. Lin et al.
A1 pLb (s ) = (ms + bc ) va (s ) + Fcom
(14)
Eqs. (11)–(14) can then be integrated as
1
va (s ) = A1 +
(
VLV s βe
+ Cta
)
⎧ qp2 ⎪ ⎡ VpaV s −⎢ + (ms + bc ) ⎨ Js + Km + bm βe ⎣ ⎪ A1 ⎩
⎤ + Ctp ⎥ pp (s ) ⎦ Km qp ωt (s ) Vs V F ⎫ + + 0 pac (s ) − ⎜⎛ LV s + Cta ⎞⎟ com Js + Km + bm p0 β ⎝ e ⎠ A1 ⎬ ⎭
(15)
In this study, the influence of the characteristics of the EM on actuator velocity is not considered. Additionally, the coulomb friction is constant and is not considered when analyzing actuator velocity. The leakages of the pump and actuator can be neglected. Therefore, Eq. (15) can be simplified as
va (s ) = =
−βe A1 VpaV s A12 βe + VLV (ms + bc ) s
pp (s ) +
βe A1 V0 p0 ⋅s [A12 βe + VLV (ms + bc ) s]
pac (s )
A1 ⎡−V dp (s ) + βe V0 dp (s ) ⎤ paV p p0 ac ⎥ A12 βe + VLV (ms + bc ) s ⎢ ⎦ ⎣ (16)
where dpp and dpac are the pressure gradient of the pump and the HA, respectively. When not considering the influence of the HA on actuator velocity, the actuator velocity of the tradition ISC system is written as follows: Fig. 4. Simplified structure of the proposed ISC system.
va (s ) =
−A1 VpaV dpp (s ) A12 βe
+ VLV (ms + bc ) s
(17)
The hydraulic natural frequency and damping ratio of the ISC system with and without the HA are similarly written as follows: Actuator
Hydraulic accumulator
Solenoid directional valve 1 AMESim/Simulink cosimulation interface
From joystick
Fig. 5. ISC system with the HA.
188
Solenoid directional valve 2
Multi-way valve Pressure loading unit
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 6. Co-simulation model used in the simulation.
ωh1 = A1
ξh1 =
bc 2A1
causes the EM to work at the first level idle speed, which is 800 rpm. After 25 s, if the joystick is still in the middle position, the controller causes the EM to work at the second level idle speed, which is 500 rpm, to reduce energy consumption. When the controller detects the joystick leaving the middle position, it causes the EM to accelerate to its normal working speed, which is 1800 rpm. As seen in Fig. 8, when the joystick returns to the middle position, the actuator stops. Therefore, the velocity and pressure of the actuator are zero and the pump is unloading. When the EM works at the first level idle speed, the actuator stays still, while the pump supplies oil to the HA and the pressures in both the accumulator and pump rise until the pressure of the HA reaches a preset value. The pump then begins unloading through solenoid directional valve 2. When the joystick leaves the middle position, both the HA and the pump supply oil to the actuator to make it work quickly. The pressure gradients of the pump and HA are the motivation for actuator movement. The pressure in the HA decreases with the movement of the actuator. The energy stored in the HA drops sharply when the EM recovers from idle speed and the actuator begins working, as shown in Fig. 8. Shortly after the pump and HA begin to supply oil to the actuator, the output power of the HA becomes smooth. This indicates that the HA can be used as an auxiliary energy source to build pressure in the actuator quickly. Both Fig. 7 and Fig. 8 indicate that the proposed ISC system can accomplish the preset goals and that the control strategy works well.
βe mVLV
(18)
VLV mβe
(19)
The difference between the two ISC systems is that the system with the HA is also affected by the pressure gradient of the HA, as seen in Eq. (16). It can be deduced from Eq. (16) that actuator velocity is influenced by the rated volume V0 and pre-charge pressure p0 of the HA, the volume between the actuator and proportional valve VLV, and the volume between the HA, pump, and proportional valve VpaV. The influence of these parameters will be discussed in the following sections. 4. Simulation study 4.1. Simulation model A control strategy based on the aforementioned requirements and the entire system with ISC is modeled in AMESim and presented in Fig. 5. Fig. 6 is the AMESim and Simulink co-simulation interface model used in Fig. 5. Fig. 7 illustrates the EM speed changes based on the state of the joystick and controlled by the proposed ISC system. The joystick returns to the middle position and the actuator begins to stop work at 3 s. When the controller detects that the joystick remains in the middle position for 8 s, meaning the actuator has stopped working for 8 s, the controller
4.2. Influence of the parameters on the performance characteristics of the ISC system Based on the EM speed control, the key parameters of the components for the performance of the actuator are analyzed to achieve better performance and a more rapid response to the signal of the joystick. The following section will discuss the influence of these parameters on the performance characteristics of the ISC system. 4.2.1. Influence of the rated volume of the HA Fig. 9 illustrates the influence of the rated volume of the HA on actuator performance and the pump. With a larger volume, it takes the HA and pump a longer time to reach the preset velocity, as shown in Fig. 9(a). However, when the actuator begins to work after the idle speed period, the actuator reaches a higher velocity, the pressure in the HA decreases more slowly, and the pump pressure rises more quickly, as shown in Fig. 9. The larger the pressure gradient of the pump and the HA, meaning the faster the pressure increases, the faster the actuator response to the joystick and the faster it reaches normal velocity. According to Fig. 9, a smaller volume for the HA reaches the preset
Fig. 7. EM speed and the signal from the joystick. ①-Stop work for 8 s; ②-First level idle speed for 25 s; ③-Second level idle speed; ④-ISC is switched off.
189
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 8. Key curves of the system during ISC.
velocity earlier and requires less energy for charging. Table 1 contains the energy required to charge the HA and the energy released from the HA when the actuator recovers from idle speed. The energy efficiency of the 1.6 L HA is approximately 81.2% while that of the 3 L HA is approximately 75.8%. Therefore, the 1.6 L HA has the highest efficiency. Owing to its quick response and high efficiency, the 1.6 L HA is the best choice.
Table 1 Energy efficiency with different HA volumes. Volume/L
1.6
2.0
2.5
3.0
Input energy/J Released energy/J Efficiency/%
2598.278 2111.355 81.2
3064.068 2400.395 78.3
3589.677 2740.145 76.3
4000.752 3032.4 75.8
lower, as shown in Fig. 10(c). This means that the efficiency of the HA is lower when the pre-charge pressure is lower. Considering the actuator response when recovering from idle speed and the energy efficiency of the HA, a quick response to the joystick is the primary goal. Thus, a 2 MPa pre-charge pressure for the HA is chosen.
4.2.2. Influence of the pre-charge pressure of the HA When the rated volume of the HA is 1.6 L, the influence of the precharge pressure on the actuator and pump is illustrated in Fig. 10. As seen from the actuator velocity, the actuator can reach normal velocity earlier when the pre-charge pressure in the HA is lower. During the first level idle speed period, the pump outputs significant energy to charge the HA when the pre-charge pressure is lower. Then, during the recovery period, as the actuator returns to work from the second level idle speed, the HA releases less energy when the pre-charge pressure is
4.2.3. Influence of the volume between the pump, HA, and multi-way valve When the rated volume of the HA is 1.6 L and the pre-charge pressure is 2 MPa, the volume VpaV between the pump, HA and multiFig. 9. Influence of the rated volume of the HA.
(a) Actuator and pump pressure
(b) Actuator velocity and HA output power when recovering from idle speed 190
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 10. Influence of the pre-charge pressure of the HA.
(a) Actuator and pump pressure
(b) Actuator velocity and HA output power when recovering from idle speed
(c) Output energy of the pump Fig. 11. Influence of the volume between the pump, HA, and multi-way valve.
4.2.4. Influence of the volume between the multi-way valve and actuator When the rated volume of the HA is 1.6 L and the pre-charge pressure is 2 MPa, the volume VLV between the non-rod chamber of the actuator and the proportional directional valve falls within the range [0.5, 1, 1.5]. From Fig. 12, there is little difference when the volume
way valve falls within the range [0.25, 0.5, 1, 1.5, 2]. From Fig. 11, when the volume is 0.25 VpaV, the actuator reaches normal velocity earlier and more stably. The pump pressure builds more quickly when the volume is smaller. Therefore, a short connection pipe between the three components is desired.
191
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 12. Influence of the volume between the multi-way valve and actuator.
Table 2 Key parameters used in the test rig. Key components
Parameter
Value
Actuator
Rod diameter/mm Cylinder diameter/mm Maximum stroke/mm Power/kW Speed/(rpm) Displacement/(mL·r− 1) Volume/L Pre-charge pressure/MPa
35 63 100 8 1800 16 1.6 2
Electric motor Pump Hydraulic accumulator
VLV changes. Therefore, it must only satisfy the requirement of convenience for the connection pipes between the components. Fig. 14. Experimental curves of the EM speed and joystick pressure.
5. Experimental research
1800
The basic parameters for the test rig are listed in Table 2, and the layout of the test rig is presented in Fig. 13. Fig. 14 illustrates the EM speed during ISC, based on the state of the joystick. The EM can switch between the normal speed, first level idle speed, and second level idle speed, based on the control strategy. The experimental results for the EM speed match the simulation results closely, as shown in Fig. 15. It can be seen that both results follow the same trend. The normal speed is 1800 rpm, the first level idle speed is 800 rpm, and the second level idle speed is 500 rpm. During the experiment, it took some time for the EM to reach its target speed, which
1200
600 0
10
20
30
40
50
Fig. 15. EM speed comparison between the simulation and experimental results.
Fig. 16. Pressures during the proposed ISC process. Fig. 13. Layout of the test rig.
192
Automation in Construction 84 (2017) 184–194
T. Lin et al.
led to the small differences between the simulation and experimental results. Fig. 16 illustrates the pressures of the HA, pump, and actuator during the proposed ISC process. When the EM enters the first level idle speed at approximately 10 s, due to the low pressure in the HA, the pump begins to charge it and the pressures in both the HA and the outlet of the pump increase. The high pressure in the HA will be used when the ISC is switched off to help increase the pump pressure. When the EM enters the second level idle speed at approximately 36 s, the pressure in the HA is nearly constant and the pump begins unloading. Both the outlet pressure and output flow rates decrease, which leads to low energy consumption by the pump. When the EM recovers from idle speed and the actuator returns to work, due to the delay in pressure building at the outlet of the pump, the pressured oil in the HA flows through the multi-way valve and into the actuator to help it overcome the resistance and move. This occurs from 42 s to 45 s. The pressure in the HA decreases as the pump pressure increases. When the actuator pressure becomes higher than the HA pressure at approximately 45 s, solenoid directional valve 1 is closed and the HA stops supplying oil to the actuator. The actuator is only driven by the pump after this point. With the HA, the inlet pressure of the actuator rises quickly when the ISC is switched off, which reduces vibrations in the actuator and helps achieve better control performance. The ISC system with and without the HA has some influence on the actuator and energy consumption. From Fig. 17, the actuator pressure and pump pressure increase more quickly and have approximately 2 s less delay, and rebuild to their target pressure faster by approximately 1 s in the ISC system with the HA than in the system without the HA, in the period from 42 s to 48 s. The inlet pressure of the actuator has a 1.2 MPa pressure drop and the inlet pressure reaches its maximum value 1 s later in the ISC system without the HA. Furthermore, the actuator pressure in the system with the HA rises smoothly while that in the system without HA has some 0.5 MPa fluctuations when the EM recovers from idle speed and returns to work, indicating that the proposed system with the HA can achieve better control performance. However, in the proposed system with the HA, the pump consumes more energy to charge the HA, which can be seen in Fig. 18. During idling, including first level and second level idling, the system without the ISC consumes 50.09 kJ of energy, the ISC system without the HA consumes 30.07 kJ of energy, and the ISC system with the HA consumes 32.03 kJ of energy. The energy saving efficiency of the system with the HA is 36.06% and that of the system without the HA is 39.97%. The extra energy consumption of approximately 3.91% is used to charge the HA and reused when the EM returns to work from the idle speed state.
Fig. 18. Comparison of pump energy consumptions.
6. Summary and conclusions Based on the working cycle and characteristics of electrically powered construction machinery, a two-level ISC system with a HA is proposed, which can also be applied in engine-driven construction machinery. The major results achieved are as follows: (1) A HA with a small volume and low pre-charging pressure is a good selection for the proposed ISC system according to the mathematical model and simulation analysis. (2) The proposed ISC system can accomplish EM speed switching between the first level idle speed, second level idle speed, and normal speed based on a control strategy. The idle speed in the novel ISC system can be reduced further than that in a conventional ISC system, which can improve energy savings. (3) The ISC system with the HA can build the pressure of the actuator quickly and enable the actuator to follow the action of the joystick rapidly when the ISC is switched off. (4) The ISC system with the HA can save 36.06% energy compared to a system without ISC and achieve better control performance. Abbreviations EM ERS HA HE ICE ISC
electric motor energy recovery system hydraulic accumulator hydraulic excavator internal combustion engine idle speed control
Nomenclature A1 bc bm Cea Cep Cia Cip Cta Ctp Fcom J KC KQ Km m
Fig. 17. Comparison between the ISC with and without the HA.
193
effective area of the non-rod chamber of the actuator viscous damping coefficient of the piston and the load of the actuator viscous damping coefficient when the pump rotates external leakage coefficients of the actuator external leakage coefficient of the pump internal leakage coefficients of the actuator internal leakage coefficient of the pump total leakage coefficient of the actuator total leakage coefficient of the pump coulomb friction when the piston moves out total inertial moment of the pump, EM, and coupling flow-pressure coefficient of the proportional directional valve flow gain of the proportional directional valve proportional factor total mass of the piston and the equivalent mass of the load
Automation in Construction 84 (2017) 184–194
T. Lin et al.
p0 pac pjL pjR Δpj pLb pLs pLmax dpp dpac Δpac pp Qac QC qp t T1 T2 Tm TL V0 VLV VpaV va xv βe δ ωt ω
on the piston of the actuator pre-charge pressure of the HA working pressure of the HA left side output pressure of the joystick right side output pressure of the joystick pressure differential of the two sides of the joystick non-rod chamber pressure of the actuator rod chamber pressure of the actuator maximum pressure of the load pressure gradient of the pump pressure gradient of the HA pressure differential of HA output port pressure of the pump flow rate of the HA flow rate through the proportional directional valve displacement of the pump time time that the system stay at the first level idle speed time that the system stay at the second level idle speed electromagnetic torque of the EM load torque of the pump rated volume of the HA volume between the non-rod chamber of actuator and proportional directional valve volume between the pump, HA, and proportional directional valve velocity of the actuator opening displacement of the proportional directional valve port effective bulk modulus a small positive value target angular velocity of the EM real-time angular velocity of the EM
[2]
[3]
[4] [5] [6]
[7] [8]
[9]
[10] [11] [12] [13]
[14] [15] [16] [17] [18]
Acknowledgements
[19]
The authors acknowledge the support of National Natural Science Foundation of China (51205140 & 51505160), Promotion Program for Young and Middle-aged Teacher in Science and Technology Research of Huaqiao University (ZQN-YX201), Natural Science Foundation of Fujian Province (2015J01206) and Open Foundation of the State Key Laboratory of Fluid Power and Mechatronic Systems (GZKF-201517)
[20]
[21] [22] [23]
References
[24]
[1] Q. Chen, W. Wu, W. Liu, et al., Research on pump control signal of excavator
194
positive control system, Journal of Hefei University of Technology (Natural Science) 37 (6) (2014) 645–649. M. Cheng, B. Xu, H. Yang, Efficiency improvement for electrohydraulic flow sharing systems, 9th International Fluid Power Conference, RWTH Aachen University, Germany, 24–26, March, 2014. K. Sugimura, H.S. Murrenhoff, Hybrid load sensing – displacement controlled architecture for excavators, Proceedings of the 14th Scandinavian International Conference on Fluid Power (SICFP'15), Tampere University of Technology, Finland, 20–22, May, 2015, pp. 20–22. Jackson R S, Clanton R R, Pfaff J L. Hydraulic control valve system with electronic load sense control: U.S. Patent 7,089,733. 2006-8-15. T. Lin, Q. Wang, B. Hu, et al., Development of hybrid powered hydraulic construction machinery, Autom. Constr. 19 (1) (2010) 11–19. R. Hippalgaonkar, M. Ivantysynova, J. Zimmerman, Fuel savings of a mini-excavator through a hydraulic hybrid displacement controlled system, Proceedings of the 8th International Fluid Power Conference, Dresden, Germany, 26–28, March, 2012, pp. 139–153. M. Cipek, D. Pavković, J. Petrić, A control-oriented simulation model of a powersplit hybrid electric vehicle, Appl. Energy 101 (2013) 121–133. T. Wang, Q. Wang, Optimization design of a permanent magnet synchronous generator for a potential energy recovery system, IEEE Transactions on Energy Conversion 27 (4) (2012) 856–863. T. Wang, Q. Wang, T. Lin, Improvement of boom control performance for hybrid hydraulic excavator with potential energy recovery, Autom. Constr. 30 (2013) 161–169. T.A. Minav, A. Virtanen, L. Laurila, et al., Storage of energy recovered from an industrial forklift, Autom. Constr. 22 (2012) 506–515. T. Lin, Q. Wang, B. Hu, et al., Research on the energy regeneration systems for hybrid hydraulic excavators, Autom. Constr. 19 (8) (2010) 1016–1026. W.A. O'Brien, R.B. Stickel, G.J. May, Advancing electric-vehicle development with pure-lead-tin battery technology, J. Power Sources 67 (1) (1997) 151–155. S.D. Cairano, D. Yanakiev, A. Bemporad, et al., Model predictive idle speed control: design, analysis, and experimental evaluation, IEEE Trans. Control Syst. Technol. 20 (1) (2012) 84–97. Q. Wang, Research on key technology of oil-electric hybrid excavator, Chinese Journal of Mechanical Engineering 49 (20) (2013) 123–129. D. Hrovat, J. Sun, Models and control methodologies for IC engine idle speed control design, Control. Eng. Pract. 5 (8) (1997) 1093–1100. Y. He, F. Ma, J. Deng, et al., Reducing the idle speed of an SI CNG engine fueled by HCNG with high hydrogen ratio, Int. J. Hydrog. Energy 37 (10) (2012) 8698–8703. X. Li, S. Yurkovich, Sliding mode control of delayed systems with application to engine idle speed control, IEEE Trans. Control Syst. Technol. 9 (6) (2001) 802–810. X. Huang, H. Xie, K. Song, Idle speed control of FSAE racing engines based on Midranging ADRC, Proceedings of the 33rd Chinese Control Conference (CCC), 2014, pp. 201–206 Nanjing, China. C. Kandler, T. Koenings, S.X. Ding, et al., Stability investigation of an idle speed control loop for a hybrid electric vehicle, IEEE Trans. Control Syst. Technol. 23 (3) (2015) 1189–1196. T. Laurain, J. Lauber, R.M. Palhares, Observer design to control individual cylinder spark advance for idle speed management of a SI engine, IEEE International Conference on Industrial Electronics and Applications (ICIEA), 2015, pp. 262–267. T. Chen, H. Xie, L. Li, et al., Methods to achieve HCCI/CAI combustion at idle operation in a 4VVAS gasoline engine, Appl. Energy 116 (2014) 41–51. Y. Xiong, P. Zhang, S. Guo, Control and experiment research of XR220D rotary drilling Rig, Equipment Manufacturing Technology 10 (2015) 240–241. R. Liu, Research on the power matching control system for the hydraulic excavators, Machine Tool & Hydraulics 43 (10) (2015) 111–115. P. Hao, Q. He, X. Zhang, et al., Study on load and operating mode identification of excavator, Hydraulics Pneumatics & Seals 5 (2008) 005.
Contents lists available at ScienceDirect
Automation in Construction journal homepage: www.elsevier.com/locate/autcon
Performance analysis of an automatic idle speed control system with a hydraulic accumulator for pure electric construction machinery
MARK
Tianliang Lin, Lang Wang, Weiping Huang, Haoling Ren⁎, Shengjie Fu, Qihuai Chen College of Mechanical Engineering and Automation, Huaqiao University, 361021 Xiamen, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Construction machinery Hydraulic excavator Energy saving Idle speed Hydraulic accumulator Pure electric system
To reduce the energy consumption and emissions of a hydraulic excavator (HE), an electric motor (EM) is employed to replace the internal combustion engine (ICE) that powers the hydraulic pumps. Owing to the excellent control characteristics and high efficiency of the EM, a two-level idle speed control system with a hydraulic accumulator (HA) for a HE is proposed to reduce energy consumption and improve the control performance of the actuator when the idle speed control (ISC) is switched off. A mathematical model is established and key parameters are analyzed and optimized. A simulation is performed using AMESim, and a control strategy for the two-level idle speed control is developed by using a co-simulation between AMESim and Simulink. A test rig is built based on the optimized parameters and simulation results. Experimental results show that the EM speed can be automatically switched between the first idle speed, second idle speed, and normal operating speed. Although the idle speed of the EM in the novel ISC system can be reduced more than that in a conventional ISC system, the proposed ISC system can still build actuator pressure more quickly in a working mode when the ISC is switched off. Compared to a system without idle speed control, the energy saving of the proposed system is approximately 36.06%. The proposed two-level idle speed control system with a HA can achieve high energy efficiency and excellent control performance, and it can be also applied to engine-driven construction machinery.
1. Introduction Recently, energy efficiency and a low environmental impact have become the basic requirements for construction machinery to meet increasingly stringent emission regulations. Thus, research on energy saving methods for construction machinery, especially hydraulic excavators (HEs), is necessary and urgent owing to their high energy consumption and emissions. Technologies that aim to reduce energy consumption and emissions, such as positive and negative flow control systems, load sensing controls, hybrid systems, pure electric systems, energy regeneration, and idle speed control (ISC) have been proposed and utilized in various models. Both flow-control systems [1,2] and load-sensing controls [3,4] are to adjust the output of pumps to meet the requirements of a load and avoid extra energy loss. Hybrid systems utilize more than one power train to allow the engine to operate at an ideal output level to reduce fuel consumption and emissions [5–7], but they cannot achieve zero emissions owing to the use of internal combustion engines (ICEs). Energy recovery systems (ERSs) are typically used to regenerate potential energy when the load is down or kinetic energy when the actuator is braking [8–11].
⁎
The traditional energy saving methods used in HEs cannot significantly improve energy savings if there are no further technological breakthroughs. The successful application of pure electric systems in vehicles provides a new method for HEs to achieve energy savings. A pure electric system can achieve zero emissions and lower energy losses [12]. Equipping HEs with pure electric systems has become the major trend in energy saving research. There have been only a few studies on pure electric HE systems. Because ISC is considered a good method for fundamentally reducing energy consumption [13], this paper focuses on a new ISC system for a pure electric HE. The working process of construction machinery is typically periodic. Idle time accounts for approximately 30% of a typical working cycle [14]. ISC is adapted to reduce energy loss and emissions during idle time. Traditional ISC is applied to control the speed switch between two values in ICEs. Generally, idle speed should be as low as possible to improve fuel economy and reduce emissions [15]. He et al. investigated a new engine and reduced the idle speed from the original 800 rpm to 700 rpm [16]. Li et al. set the idle speed in an engine to 611 rpm instead of the normal (production) idle speed of 740 rpm and used a sliding mode control for ISC. Although a low idle speed can reduce fuel
Corresponding author. E-mail address: [email protected] (H. Ren).
http://dx.doi.org/10.1016/j.autcon.2017.09.001 Received 2 September 2016; Received in revised form 10 August 2017; Accepted 2 September 2017 0926-5805/ © 2017 Published by Elsevier B.V.
Automation in Construction 84 (2017) 184–194
T. Lin et al.
The working mode recognition process of the ISC system in Fig. 1 is presented in Fig. 3. When the joystick returns to the middle position, the controller detects that the pressure difference between the two pressures on the joystick is under a preset small positive value and sends signals to the multi-way valve to turn it to the middle position. The proposed ISC system is used to control the speed of the EM as it switches between different values, including the switch from high speed to low speed and from low speed to high speed. When reducing the speed of the EM, the main consideration is energy saving. The value of the idle speed is the primary factor that influences energy consumption. When the working mode of the ISC system is switched off, control performance is the main consideration. Therefore, the time required for building pressure in the pump should be smooth and short. By taking the self-suction capacity of the pump into account and referring to Fig. 2, the first level idle speed is set to 800 rpm and the second level idle speed is set to 500 rpm because the HA can help the pump build pressure. During the various processes, the control strategy changes accordingly. 1) First level ISC When the time that the joystick resides in the middle position is longer than time T1, the controller reduces the EM speed to the first level idle speed n1 to reduce energy loss. Meanwhile, the maximum load pressure is used to determine whether the HA should charge the pump. When the pressure difference between the HA and the maximum load pressure is larger than a set value, solenoid directional valves 1 and 2 are powered on. The pump then charges the HA through solenoid directional valve 1. After the HA pressure increases to the set value, solenoid directional valves 1 and 2 are powered off. 2) Second level ISC The control target is to achieve minimum energy consumption. The output power of the system should outweigh the total energy loss and minimize energy consumption. When the time that the joystick resides in the middle position is longer than time T1 + T2 and the pressure difference between the maximum actuator pressure and the HA pressure is under a threshold value, the controller switches the EM to the second level idle speed n2, which is lower than the first level idle speed n1. 3) ISC is switched off When the actuator recovers from the idle speed and the EM returns to normal speed, which is approximately 1800 rpm, the global positive flow control is used to make the flow rate supplied by the pump and the HA match the requirement of the load. Solenoid directional valves 1 and 2 are powered on. The stored oil in the HA is released through solenoid directional valve 1 to the inlet of the multi-way valve to help the pump build pressure quickly and drive the actuator. When the HA pressure is lower than the pump pressure, solenoid directional valve 1 is powered off and the actuator is driven by the pump only. The proportion of the flow rate supplied by the HA and the pump is scheduled to achieve smooth and quick movement of the actuator.
consumption, it carries an increased risk of misfiring and engine stalling [17,18]. Thus, the transition to and from idle speed must be smooth and carefully controlled [18]. Many researchers have devoted themselves to solving ISC problems with various control strategies [19–21]. Most of the research has focused on ISC for automobiles. However, owing to the intrinsic properties of an ICE, the speed range is narrow and the idle speed remains high, which limits the potential energy savings of ISC. However, there have been a few studies on construction machinery. Xiong et al. analyzed construction machinery working principles and implemented methods for ISC in a rotary drilling rig [22]. They tested fuel consumption under various speeds and found that the speed with the lowest fuel consumption was the preset idle speed. Liu designed an ISC system based on working conditions, speed sensing, and engine power matching [23]. The adjustment time for the engine between idle speed and the rated speed was approximately two seconds. Hao optimized the duty ratio of pulse wave modulation, minimum ramping time, and idle speed through an adaptive control method and achieved superior energy savings [24]. However, these studies were still based on the ICE and the idle speed could not be set to a low value, causing relatively low energy efficiency. Furthermore, the pump could not build pressure quickly enough to drive the actuator when the working mode of the ISC was switched off. ISC is adapted to reduce energy loss and emissions during idle time. When ISC is applied to ICEs, it can also reduce fuel consumption and greenhouse gas emissions. However, owing to the intrinsic properties of ICEs, the speed range is narrow and the idle speed remains high, which limits the energy saving potential of ISC and makes it difficult to meet requirements for environmental protection. When ISC is applied to pure electric systems, the control performance of the electric motor (EM) is different from the ICE, particularly the speed range, dynamic response, and efficiency. Therefore, in this research, a two-ton class pure electric HE was built in our lab for experimentation. A novel ISC with a hydraulic accumulator (HA) for the two-ton HE is the focus of this research. The remainder of this paper is organized as follows: The structure and working principle of the automatic ISC system with a HA is described in Section 2. A mathematical model of the proposed ISC system is described in Section 3. The influence of various parameters on the performance characteristics of the ISC system is discussed in detail in Section 4. The experimental results are analyzed in Section 5. A summary and our final conclusions are provided in Section 6. 2. Structure and working principle of the novel automatic ISC system Fig. 1 presents a schematic of the proposed automatic ISC system. Compared to traditional ISC systems, the proposed ISC system has the following improvements: 1) Pure electric drive enables zero emissions for the driving system. An EM is used in place of an ICE for driving the hydraulic pumps. The efficiency of the EM is presented in Fig. 2 and can reach 95% when the EM speed is under 500 rpm, and up to 96%–98% when the EM speed is above 800 rpm. The EM has higher efficiency at low speeds, a wider speed range, and a quicker dynamic response than an ICE. 2) A HA is connected to the outlet of the pump via solenoid directional valve 1. The HA can provide auxiliary power to drive the actuator when the working mode of ISC is switched off. The pump can then build pressure quickly enough to drive the actuator. This means that the idle speed can be set to a low value to obtain good energy savings and decrease noise. 3) There is a pressure loading unit at the return port when the multiway valve is in the middle working position. It can separate the pump from the tank via solenoid directional valve 2 and charge the HA based on the control strategy. 4) All pressures of the key components are detected and used to recognize the working mode of the ISC system.
3. Mathematical model of the ISC system with a HA The simplified structure of the proposed ISC system presented in Fig. 4 is used to analyze the control performance of the novel ISC system. The assumptions made are as follows: 1) The influence of the solenoid directional valve on the velocity characteristics of the actuator is not considered. 2) Only the piston-out movement of the actuator is considered. The influence of piston movements on the volume between the non-rod chamber, HA, and the pump is ignored. 3) The actuator and pump have no elastic loads or disturbance torque. 4) The safety valve, which is connected to the pump output port, does not work and overflow during the working process. 5) The pressure of the inlet port of the pump is zero.
185
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 1. Schematic of the proposed automatic ISC system.
where qp is the displacement of the pump in m3/rad. Cip and Cep are the internal and external leakage coefficient of the pump, respectively, in m3/(Pa·s). Ctp is the total leakage coefficient of the pump in m3/(Pa·s) and Ctp = Cip + Cep. pp is the output port pressure of the pump in Pa. (3) Torque equation for the load of the pump and EM As mentioned in the assumptions, there are no elastic loads or disturbance torques. Thus, the load torque equation for the pump is
TL = qp⋅pp
(3)
and the torque equilibrium equation is
Tm − TL = J
(4) The flow equation for the proportional directional valve (multi-way valve) is
(1) Electromagnetic torque equation of the inverter-EM The EM employed in this paper is asynchronous. Considering that the setup time for the electromagnetic torque of the inverter-EM is much shorter than the mechanical response time of the pump-EM, the torque of the inverter and EM can be simplified as a proportion:
QC = K Q x v + K C (pp − pLb ),
(5)
where QC is the flow rate through the proportional directional valve in m3/s. KQ is the flow gain of the proportional directional valve in m2/s. KC is the flow-pressure coefficient of the proportional directional valve in m3/(Pa·s). xv is the opening displacement of the proportional directional valve port in m. pLb is the non-rod chamber pressure of the actuator in Pa.
(1)
where Tm is the electromagnetic torque of the EM in N·m. Km is the proportional factor. ωt and ω are the target angular velocity and realtime angular velocity of the EM, respectively, in rad/s.
(5) The flow equation for the HA is
(2) Output flow rate of the pump:
Qp = qp ω − (Cip + Cep ) pp = qp ω − Ctp pp ,
(4)
where TL is the load torque of the pump in TLN·m. J is the total inertial moment of the pump, EM, and coupling in kg·m2. bm is the viscous damping coefficient when the pump rotates in N·s/rad.
Fig. 2. Efficiency map of the EM.
Tm = Km (ωt − ω)
dω + bm ω, dt
Qac =
(2) 186
V0 dpac , p0 dt
(6)
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 3. Working mode recognition process of the ISC system.
where Qac is the flow rate of the HA in m3/s. p0 and pac are the precharge pressure and working pressure of the HA, respectively, in Pa. V0 is the rated volume of the HA in m3.
chamber is zero, specifically, pLs = 0. Thus, Eq. (7) can be rewritten as
K Q x v + K C pp − (K C + Cta ) pLb − A1 va =
VLV dpLb . βe dt
(9)
(6) Continuity equation for the oil (7) Force balance equation for the actuator: The continuity equation for the actuator is
m VLV dpLb + A1 va = Qc − Cia (pLb − pLs ) − Cea pLb . βe dt
βe
dt
= Qp + Qac − Qc ,
(10)
where m is the total mass of the piston and the equivalent mass of the load on the piston of the actuator in kg. bc is the viscous damping coefficient of the piston and the load of the actuator in N/(m/s). Fcom is the coulomb friction when the piston moves out in N. Based on Eqs. (1)–(10), a Laplace transformation can be obtained as follows:
(7)
The continuity equation for the pump is
VpaV dpp
dva = pLb A1 − bc va − Fcom, dt
(8)
where βe is the effective bulk modulus in Pa. VpaV is the volume between the pump, HA, and proportional directional valve in m3. VLV is the volume between the non-rod chamber of the actuator and the proportional directional valve in m3. A1 is the effective area of the nonrod chamber of the actuator in m2. va is the velocity of the actuator in m/s. Cia and Cea are the internal and external leakage coefficients of the actuator, respectively, in m3/(Pa·s). Cta is the total leakage coefficient of the actuator in m3/(Pa·s) and Cta = Cia + Cea. When the piston of the actuator moves out, the pressure in the rod
Km ωt − qp⋅pp (s ) = (Js + Km + bm ) ω (s )
qp ω (s ) − K Q x v (s ) + K C pLb (s ) +
V0 s V s p (s ) = ⎜⎛ LV + Ctp + K C ⎟⎞ pp (s ) p0 ac ⎝ βe ⎠ (12)
VpaV s K Q x v (s ) + K C pp (s ) − A1 va (s ) = ⎜⎛ + Ctc + K C ⎟⎞ pLb (s ) ⎝ βe ⎠ 187
(11)
(13)
Automation in Construction 84 (2017) 184–194
T. Lin et al.
A1 pLb (s ) = (ms + bc ) va (s ) + Fcom
(14)
Eqs. (11)–(14) can then be integrated as
1
va (s ) = A1 +
(
VLV s βe
+ Cta
)
⎧ qp2 ⎪ ⎡ VpaV s −⎢ + (ms + bc ) ⎨ Js + Km + bm βe ⎣ ⎪ A1 ⎩
⎤ + Ctp ⎥ pp (s ) ⎦ Km qp ωt (s ) Vs V F ⎫ + + 0 pac (s ) − ⎜⎛ LV s + Cta ⎞⎟ com Js + Km + bm p0 β ⎝ e ⎠ A1 ⎬ ⎭
(15)
In this study, the influence of the characteristics of the EM on actuator velocity is not considered. Additionally, the coulomb friction is constant and is not considered when analyzing actuator velocity. The leakages of the pump and actuator can be neglected. Therefore, Eq. (15) can be simplified as
va (s ) = =
−βe A1 VpaV s A12 βe + VLV (ms + bc ) s
pp (s ) +
βe A1 V0 p0 ⋅s [A12 βe + VLV (ms + bc ) s]
pac (s )
A1 ⎡−V dp (s ) + βe V0 dp (s ) ⎤ paV p p0 ac ⎥ A12 βe + VLV (ms + bc ) s ⎢ ⎦ ⎣ (16)
where dpp and dpac are the pressure gradient of the pump and the HA, respectively. When not considering the influence of the HA on actuator velocity, the actuator velocity of the tradition ISC system is written as follows: Fig. 4. Simplified structure of the proposed ISC system.
va (s ) =
−A1 VpaV dpp (s ) A12 βe
+ VLV (ms + bc ) s
(17)
The hydraulic natural frequency and damping ratio of the ISC system with and without the HA are similarly written as follows: Actuator
Hydraulic accumulator
Solenoid directional valve 1 AMESim/Simulink cosimulation interface
From joystick
Fig. 5. ISC system with the HA.
188
Solenoid directional valve 2
Multi-way valve Pressure loading unit
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 6. Co-simulation model used in the simulation.
ωh1 = A1
ξh1 =
bc 2A1
causes the EM to work at the first level idle speed, which is 800 rpm. After 25 s, if the joystick is still in the middle position, the controller causes the EM to work at the second level idle speed, which is 500 rpm, to reduce energy consumption. When the controller detects the joystick leaving the middle position, it causes the EM to accelerate to its normal working speed, which is 1800 rpm. As seen in Fig. 8, when the joystick returns to the middle position, the actuator stops. Therefore, the velocity and pressure of the actuator are zero and the pump is unloading. When the EM works at the first level idle speed, the actuator stays still, while the pump supplies oil to the HA and the pressures in both the accumulator and pump rise until the pressure of the HA reaches a preset value. The pump then begins unloading through solenoid directional valve 2. When the joystick leaves the middle position, both the HA and the pump supply oil to the actuator to make it work quickly. The pressure gradients of the pump and HA are the motivation for actuator movement. The pressure in the HA decreases with the movement of the actuator. The energy stored in the HA drops sharply when the EM recovers from idle speed and the actuator begins working, as shown in Fig. 8. Shortly after the pump and HA begin to supply oil to the actuator, the output power of the HA becomes smooth. This indicates that the HA can be used as an auxiliary energy source to build pressure in the actuator quickly. Both Fig. 7 and Fig. 8 indicate that the proposed ISC system can accomplish the preset goals and that the control strategy works well.
βe mVLV
(18)
VLV mβe
(19)
The difference between the two ISC systems is that the system with the HA is also affected by the pressure gradient of the HA, as seen in Eq. (16). It can be deduced from Eq. (16) that actuator velocity is influenced by the rated volume V0 and pre-charge pressure p0 of the HA, the volume between the actuator and proportional valve VLV, and the volume between the HA, pump, and proportional valve VpaV. The influence of these parameters will be discussed in the following sections. 4. Simulation study 4.1. Simulation model A control strategy based on the aforementioned requirements and the entire system with ISC is modeled in AMESim and presented in Fig. 5. Fig. 6 is the AMESim and Simulink co-simulation interface model used in Fig. 5. Fig. 7 illustrates the EM speed changes based on the state of the joystick and controlled by the proposed ISC system. The joystick returns to the middle position and the actuator begins to stop work at 3 s. When the controller detects that the joystick remains in the middle position for 8 s, meaning the actuator has stopped working for 8 s, the controller
4.2. Influence of the parameters on the performance characteristics of the ISC system Based on the EM speed control, the key parameters of the components for the performance of the actuator are analyzed to achieve better performance and a more rapid response to the signal of the joystick. The following section will discuss the influence of these parameters on the performance characteristics of the ISC system. 4.2.1. Influence of the rated volume of the HA Fig. 9 illustrates the influence of the rated volume of the HA on actuator performance and the pump. With a larger volume, it takes the HA and pump a longer time to reach the preset velocity, as shown in Fig. 9(a). However, when the actuator begins to work after the idle speed period, the actuator reaches a higher velocity, the pressure in the HA decreases more slowly, and the pump pressure rises more quickly, as shown in Fig. 9. The larger the pressure gradient of the pump and the HA, meaning the faster the pressure increases, the faster the actuator response to the joystick and the faster it reaches normal velocity. According to Fig. 9, a smaller volume for the HA reaches the preset
Fig. 7. EM speed and the signal from the joystick. ①-Stop work for 8 s; ②-First level idle speed for 25 s; ③-Second level idle speed; ④-ISC is switched off.
189
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 8. Key curves of the system during ISC.
velocity earlier and requires less energy for charging. Table 1 contains the energy required to charge the HA and the energy released from the HA when the actuator recovers from idle speed. The energy efficiency of the 1.6 L HA is approximately 81.2% while that of the 3 L HA is approximately 75.8%. Therefore, the 1.6 L HA has the highest efficiency. Owing to its quick response and high efficiency, the 1.6 L HA is the best choice.
Table 1 Energy efficiency with different HA volumes. Volume/L
1.6
2.0
2.5
3.0
Input energy/J Released energy/J Efficiency/%
2598.278 2111.355 81.2
3064.068 2400.395 78.3
3589.677 2740.145 76.3
4000.752 3032.4 75.8
lower, as shown in Fig. 10(c). This means that the efficiency of the HA is lower when the pre-charge pressure is lower. Considering the actuator response when recovering from idle speed and the energy efficiency of the HA, a quick response to the joystick is the primary goal. Thus, a 2 MPa pre-charge pressure for the HA is chosen.
4.2.2. Influence of the pre-charge pressure of the HA When the rated volume of the HA is 1.6 L, the influence of the precharge pressure on the actuator and pump is illustrated in Fig. 10. As seen from the actuator velocity, the actuator can reach normal velocity earlier when the pre-charge pressure in the HA is lower. During the first level idle speed period, the pump outputs significant energy to charge the HA when the pre-charge pressure is lower. Then, during the recovery period, as the actuator returns to work from the second level idle speed, the HA releases less energy when the pre-charge pressure is
4.2.3. Influence of the volume between the pump, HA, and multi-way valve When the rated volume of the HA is 1.6 L and the pre-charge pressure is 2 MPa, the volume VpaV between the pump, HA and multiFig. 9. Influence of the rated volume of the HA.
(a) Actuator and pump pressure
(b) Actuator velocity and HA output power when recovering from idle speed 190
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 10. Influence of the pre-charge pressure of the HA.
(a) Actuator and pump pressure
(b) Actuator velocity and HA output power when recovering from idle speed
(c) Output energy of the pump Fig. 11. Influence of the volume between the pump, HA, and multi-way valve.
4.2.4. Influence of the volume between the multi-way valve and actuator When the rated volume of the HA is 1.6 L and the pre-charge pressure is 2 MPa, the volume VLV between the non-rod chamber of the actuator and the proportional directional valve falls within the range [0.5, 1, 1.5]. From Fig. 12, there is little difference when the volume
way valve falls within the range [0.25, 0.5, 1, 1.5, 2]. From Fig. 11, when the volume is 0.25 VpaV, the actuator reaches normal velocity earlier and more stably. The pump pressure builds more quickly when the volume is smaller. Therefore, a short connection pipe between the three components is desired.
191
Automation in Construction 84 (2017) 184–194
T. Lin et al.
Fig. 12. Influence of the volume between the multi-way valve and actuator.
Table 2 Key parameters used in the test rig. Key components
Parameter
Value
Actuator
Rod diameter/mm Cylinder diameter/mm Maximum stroke/mm Power/kW Speed/(rpm) Displacement/(mL·r− 1) Volume/L Pre-charge pressure/MPa
35 63 100 8 1800 16 1.6 2
Electric motor Pump Hydraulic accumulator
VLV changes. Therefore, it must only satisfy the requirement of convenience for the connection pipes between the components. Fig. 14. Experimental curves of the EM speed and joystick pressure.
5. Experimental research
1800
The basic parameters for the test rig are listed in Table 2, and the layout of the test rig is presented in Fig. 13. Fig. 14 illustrates the EM speed during ISC, based on the state of the joystick. The EM can switch between the normal speed, first level idle speed, and second level idle speed, based on the control strategy. The experimental results for the EM speed match the simulation results closely, as shown in Fig. 15. It can be seen that both results follow the same trend. The normal speed is 1800 rpm, the first level idle speed is 800 rpm, and the second level idle speed is 500 rpm. During the experiment, it took some time for the EM to reach its target speed, which
1200
600 0
10
20
30
40
50
Fig. 15. EM speed comparison between the simulation and experimental results.
Fig. 16. Pressures during the proposed ISC process. Fig. 13. Layout of the test rig.
192
Automation in Construction 84 (2017) 184–194
T. Lin et al.
led to the small differences between the simulation and experimental results. Fig. 16 illustrates the pressures of the HA, pump, and actuator during the proposed ISC process. When the EM enters the first level idle speed at approximately 10 s, due to the low pressure in the HA, the pump begins to charge it and the pressures in both the HA and the outlet of the pump increase. The high pressure in the HA will be used when the ISC is switched off to help increase the pump pressure. When the EM enters the second level idle speed at approximately 36 s, the pressure in the HA is nearly constant and the pump begins unloading. Both the outlet pressure and output flow rates decrease, which leads to low energy consumption by the pump. When the EM recovers from idle speed and the actuator returns to work, due to the delay in pressure building at the outlet of the pump, the pressured oil in the HA flows through the multi-way valve and into the actuator to help it overcome the resistance and move. This occurs from 42 s to 45 s. The pressure in the HA decreases as the pump pressure increases. When the actuator pressure becomes higher than the HA pressure at approximately 45 s, solenoid directional valve 1 is closed and the HA stops supplying oil to the actuator. The actuator is only driven by the pump after this point. With the HA, the inlet pressure of the actuator rises quickly when the ISC is switched off, which reduces vibrations in the actuator and helps achieve better control performance. The ISC system with and without the HA has some influence on the actuator and energy consumption. From Fig. 17, the actuator pressure and pump pressure increase more quickly and have approximately 2 s less delay, and rebuild to their target pressure faster by approximately 1 s in the ISC system with the HA than in the system without the HA, in the period from 42 s to 48 s. The inlet pressure of the actuator has a 1.2 MPa pressure drop and the inlet pressure reaches its maximum value 1 s later in the ISC system without the HA. Furthermore, the actuator pressure in the system with the HA rises smoothly while that in the system without HA has some 0.5 MPa fluctuations when the EM recovers from idle speed and returns to work, indicating that the proposed system with the HA can achieve better control performance. However, in the proposed system with the HA, the pump consumes more energy to charge the HA, which can be seen in Fig. 18. During idling, including first level and second level idling, the system without the ISC consumes 50.09 kJ of energy, the ISC system without the HA consumes 30.07 kJ of energy, and the ISC system with the HA consumes 32.03 kJ of energy. The energy saving efficiency of the system with the HA is 36.06% and that of the system without the HA is 39.97%. The extra energy consumption of approximately 3.91% is used to charge the HA and reused when the EM returns to work from the idle speed state.
Fig. 18. Comparison of pump energy consumptions.
6. Summary and conclusions Based on the working cycle and characteristics of electrically powered construction machinery, a two-level ISC system with a HA is proposed, which can also be applied in engine-driven construction machinery. The major results achieved are as follows: (1) A HA with a small volume and low pre-charging pressure is a good selection for the proposed ISC system according to the mathematical model and simulation analysis. (2) The proposed ISC system can accomplish EM speed switching between the first level idle speed, second level idle speed, and normal speed based on a control strategy. The idle speed in the novel ISC system can be reduced further than that in a conventional ISC system, which can improve energy savings. (3) The ISC system with the HA can build the pressure of the actuator quickly and enable the actuator to follow the action of the joystick rapidly when the ISC is switched off. (4) The ISC system with the HA can save 36.06% energy compared to a system without ISC and achieve better control performance. Abbreviations EM ERS HA HE ICE ISC
electric motor energy recovery system hydraulic accumulator hydraulic excavator internal combustion engine idle speed control
Nomenclature A1 bc bm Cea Cep Cia Cip Cta Ctp Fcom J KC KQ Km m
Fig. 17. Comparison between the ISC with and without the HA.
193
effective area of the non-rod chamber of the actuator viscous damping coefficient of the piston and the load of the actuator viscous damping coefficient when the pump rotates external leakage coefficients of the actuator external leakage coefficient of the pump internal leakage coefficients of the actuator internal leakage coefficient of the pump total leakage coefficient of the actuator total leakage coefficient of the pump coulomb friction when the piston moves out total inertial moment of the pump, EM, and coupling flow-pressure coefficient of the proportional directional valve flow gain of the proportional directional valve proportional factor total mass of the piston and the equivalent mass of the load
Automation in Construction 84 (2017) 184–194
T. Lin et al.
p0 pac pjL pjR Δpj pLb pLs pLmax dpp dpac Δpac pp Qac QC qp t T1 T2 Tm TL V0 VLV VpaV va xv βe δ ωt ω
on the piston of the actuator pre-charge pressure of the HA working pressure of the HA left side output pressure of the joystick right side output pressure of the joystick pressure differential of the two sides of the joystick non-rod chamber pressure of the actuator rod chamber pressure of the actuator maximum pressure of the load pressure gradient of the pump pressure gradient of the HA pressure differential of HA output port pressure of the pump flow rate of the HA flow rate through the proportional directional valve displacement of the pump time time that the system stay at the first level idle speed time that the system stay at the second level idle speed electromagnetic torque of the EM load torque of the pump rated volume of the HA volume between the non-rod chamber of actuator and proportional directional valve volume between the pump, HA, and proportional directional valve velocity of the actuator opening displacement of the proportional directional valve port effective bulk modulus a small positive value target angular velocity of the EM real-time angular velocity of the EM
[2]
[3]
[4] [5] [6]
[7] [8]
[9]
[10] [11] [12] [13]
[14] [15] [16] [17] [18]
Acknowledgements
[19]
The authors acknowledge the support of National Natural Science Foundation of China (51205140 & 51505160), Promotion Program for Young and Middle-aged Teacher in Science and Technology Research of Huaqiao University (ZQN-YX201), Natural Science Foundation of Fujian Province (2015J01206) and Open Foundation of the State Key Laboratory of Fluid Power and Mechatronic Systems (GZKF-201517)
[20]
[21] [22] [23]
References
[24]
[1] Q. Chen, W. Wu, W. Liu, et al., Research on pump control signal of excavator
194
positive control system, Journal of Hefei University of Technology (Natural Science) 37 (6) (2014) 645–649. M. Cheng, B. Xu, H. Yang, Efficiency improvement for electrohydraulic flow sharing systems, 9th International Fluid Power Conference, RWTH Aachen University, Germany, 24–26, March, 2014. K. Sugimura, H.S. Murrenhoff, Hybrid load sensing – displacement controlled architecture for excavators, Proceedings of the 14th Scandinavian International Conference on Fluid Power (SICFP'15), Tampere University of Technology, Finland, 20–22, May, 2015, pp. 20–22. Jackson R S, Clanton R R, Pfaff J L. Hydraulic control valve system with electronic load sense control: U.S. Patent 7,089,733. 2006-8-15. T. Lin, Q. Wang, B. Hu, et al., Development of hybrid powered hydraulic construction machinery, Autom. Constr. 19 (1) (2010) 11–19. R. Hippalgaonkar, M. Ivantysynova, J. Zimmerman, Fuel savings of a mini-excavator through a hydraulic hybrid displacement controlled system, Proceedings of the 8th International Fluid Power Conference, Dresden, Germany, 26–28, March, 2012, pp. 139–153. M. Cipek, D. Pavković, J. Petrić, A control-oriented simulation model of a powersplit hybrid electric vehicle, Appl. Energy 101 (2013) 121–133. T. Wang, Q. Wang, Optimization design of a permanent magnet synchronous generator for a potential energy recovery system, IEEE Transactions on Energy Conversion 27 (4) (2012) 856–863. T. Wang, Q. Wang, T. Lin, Improvement of boom control performance for hybrid hydraulic excavator with potential energy recovery, Autom. Constr. 30 (2013) 161–169. T.A. Minav, A. Virtanen, L. Laurila, et al., Storage of energy recovered from an industrial forklift, Autom. Constr. 22 (2012) 506–515. T. Lin, Q. Wang, B. Hu, et al., Research on the energy regeneration systems for hybrid hydraulic excavators, Autom. Constr. 19 (8) (2010) 1016–1026. W.A. O'Brien, R.B. Stickel, G.J. May, Advancing electric-vehicle development with pure-lead-tin battery technology, J. Power Sources 67 (1) (1997) 151–155. S.D. Cairano, D. Yanakiev, A. Bemporad, et al., Model predictive idle speed control: design, analysis, and experimental evaluation, IEEE Trans. Control Syst. Technol. 20 (1) (2012) 84–97. Q. Wang, Research on key technology of oil-electric hybrid excavator, Chinese Journal of Mechanical Engineering 49 (20) (2013) 123–129. D. Hrovat, J. Sun, Models and control methodologies for IC engine idle speed control design, Control. Eng. Pract. 5 (8) (1997) 1093–1100. Y. He, F. Ma, J. Deng, et al., Reducing the idle speed of an SI CNG engine fueled by HCNG with high hydrogen ratio, Int. J. Hydrog. Energy 37 (10) (2012) 8698–8703. X. Li, S. Yurkovich, Sliding mode control of delayed systems with application to engine idle speed control, IEEE Trans. Control Syst. Technol. 9 (6) (2001) 802–810. X. Huang, H. Xie, K. Song, Idle speed control of FSAE racing engines based on Midranging ADRC, Proceedings of the 33rd Chinese Control Conference (CCC), 2014, pp. 201–206 Nanjing, China. C. Kandler, T. Koenings, S.X. Ding, et al., Stability investigation of an idle speed control loop for a hybrid electric vehicle, IEEE Trans. Control Syst. Technol. 23 (3) (2015) 1189–1196. T. Laurain, J. Lauber, R.M. Palhares, Observer design to control individual cylinder spark advance for idle speed management of a SI engine, IEEE International Conference on Industrial Electronics and Applications (ICIEA), 2015, pp. 262–267. T. Chen, H. Xie, L. Li, et al., Methods to achieve HCCI/CAI combustion at idle operation in a 4VVAS gasoline engine, Appl. Energy 116 (2014) 41–51. Y. Xiong, P. Zhang, S. Guo, Control and experiment research of XR220D rotary drilling Rig, Equipment Manufacturing Technology 10 (2015) 240–241. R. Liu, Research on the power matching control system for the hydraulic excavators, Machine Tool & Hydraulics 43 (10) (2015) 111–115. P. Hao, Q. He, X. Zhang, et al., Study on load and operating mode identification of excavator, Hydraulics Pneumatics & Seals 5 (2008) 005.