Novel high accurate sensorless dual-axis solar tracking system controlled by maximum power point tracking unit of photovoltaic systems

Applied Energy 173 (2016) 448–459

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Novel high accurate sensorless dual-axis solar tracking system controlled by maximum power point tracking unit of photovoltaic systems Hassan Fathabadi School of Electrical and Computer Engineering, National Technical University of Athens (NTUA), Athens, Greece

h i g h l i g h t s
Novel high accurate sensorless dual-axis solar tracker.
It has the advantages of both sensor based and sensorless solar trackers.
It does not have the disadvantages of sensor based and sensorless solar trackers.
Tracking error of only 0.11° that is less than the tracking errors of others.
An increase of 28.8–43.6% depending on the seasons in the energy efficiency.

a r t i c l e

i n f o

Article history: Received 19 December 2015 Received in revised form 6 March 2016 Accepted 29 March 2016

Keywords: Solar tracking system Maximum power point tracking Energy efficiency Solar energy

a b s t r a c t In this study, a novel high accurate sensorless dual-axis solar tracker controlled by the maximum power point tracking unit available in almost all photovoltaic systems is proposed. The maximum power point tracking controller continuously calculates the maximum output power of the photovoltaic module/panel/array, and uses the altitude and azimuth angles deviations to track the sun direction where the greatest value of the maximum output power is extracted. Unlike all other sensorless solar trackers, the proposed solar tracking system is a closed loop system which means it uses the actual direction of the sun at any time to track the sun direction, and this is the contribution of this work. The proposed solar tracker has the advantages of both sensor based and sensorless dual-axis solar trackers, but it does not have their disadvantages. Other sensorless solar trackers all are open loop, i.e., they use offline estimated data about the sun path in the sky obtained from solar map equations, so low exactness, cloudy sky, and requiring new data for new location are their problems. A photovoltaic system has been built, and it is experimentally verified that the proposed solar tracking system tracks the sun direction with the tracking error of 0.11° which is less than the tracking errors of other both sensor based and sensorless solar trackers. An increase of 28.8–43.6% depending on the seasons in the energy efficiency is the main advantage of utilizing the proposed solar tracking system. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction A maximum power point tracking (MPPT) unit is an essential part of almost all photovoltaic (PV) systems [1]. It adjusts the operating point of the PV module used in the system to the maximum power point (MPP) of the PV module [2]. Thus, a MPPT controller significantly increases the energy conversion efficiency of the PV system by extracting as much as possible instant power from the PV module [3]. Different MPPT methods have been reported in the literature [4]. Some methods are offline or model-based such

E-mail address: [email protected] http://dx.doi.org/10.1016/j.apenergy.2016.03.109 0306-2619/Ó 2016 Elsevier Ltd. All rights reserved.

as short-circuit current (SCC) and open-circuit voltage (OCV) because the PV module is regularly disconnected, and a specific physical parameter such as short-circuit current or open-circuit voltage is measured [5]. Other MPPT techniques such as extremum seeking control (ESC), perturb and observe (P&O), and incremental conductance (IC) use the output voltage or current of the PV module under actual operating condition, so they are online or modelfree methods [6]. The OCV technique uses the open-circuit voltage of the PV module to estimate the MPP voltage [7]. The temperature technique is an online version of the OCV method which estimates the open-circuit voltage of the PV module using its temperature under operating condition [8]. The SCC method measures the

H. Fathabadi / Applied Energy 173 (2016) 448–459

short-circuit current to estimate the MPP current [9,10]. Fuzzy methods first convert the PV module current and voltage into fuzzy parameters [11]. They provide some fuzzy outputs based on the fuzzy roles defined for the system, and then, the fuzzy outputs are converted into real outputs by the defuzzification unit [12]. A modified version of the fuzzy technique is adaptive-fuzzy technique which utilizes a fuzzy unit having an adapted gain [13]. Artificial neural network (ANN) based techniques use different trained neural networks to track the MPP [14,15]. The P&O technique first perturbs the PV module voltage or current, and then, the output power of the PV module is compared with the previous value to determine how to produce the perturbation in next step [16]. A modified version of P&O method which determines the perturbation direction using three points is known as three-point weighted technique [17]. Another improved version of the P&O method employing dynamic perturbations was proposed in [18]. Particle swarm optimization adaptive neuro-fuzzy inference system (PSO-ANFIS) and P&O-ANFIS are the two hybrid online MPPT techniques reported in [19]. The slope of P–V characteristic is used to track the MPP in the IC method [20,21]. An improved version of the IC technique having zero oscillation of the maximum output power was reported in [22]. The ESC technique uses a nonlinear closed loop mechanism the dynamic of which has been adapted well to the PV module operating point to find the MPP [23–25]. Ripple-based ESC is a modified version of the ESC technique which can be used in grid-connected PV systems, it uses the DC-link voltage ripple to find the MPP [26]. Some other MPPT techniques such as power management maximum power point tracking (PMMPPT) [27], Cuckoo Search (CS) [28], and modified genetic algorithm (GA) [29] have been also reported in the literature. The MPPT controller used in this study is implemented based on the high accurate MPPT technique presented in [30] which concurrently uses PV current and voltage deviations to precisely track the MPP. Solar energy is an important renewable energy source getting popular in many countries day by day. The main defect concerning solar energy conversion systems is their low efficiency, so increasing the energy efficiency of this type of renewable energy systems has been the subject of many research projects. A PV system uses a PV module/panel/array to convert solar energy into electric energy. To extract the maximum output power from the PV module, a solar tracker can be used to track the sun direction, so that, sunbeam strikes the PV module surface perpendicularly. In fact, previous researches showed that about 20–50% more solar energy can be captured depending on the geographic position by adding a solar tracker to a PV system [31]. Solar trackers are divided into two types: single-axis and dual-axis [32]. The sole axis of a singleaxis solar tracker is aligned along the local north meridian, it has only one freedom degree, so it can only track the sun in one direction which is the daily path of the sun [33]. Dual-axis type has two freedom degrees, so it can track the sun path in two directions which are daily and seasonal motions of the sun [34]. A singleaxis solar tracking system increases the daily output power of the PV module up-to about 20% compared to a fixed PV module [35]. It is clear that a dual-axis solar tracking system is more accurate to track the sun direction compared to a single-axis type [36]. The output power of the PV module can be increased up-to about 33% compared to a fixed PV module by utilizing a dual-axis solar tracker [37]. Dual-axis trackers are classified into two types: sensor based and sensorless solar trackers. A sensor based solar tracker acts as a closed loop system in which photo-sensors are used to provide appropriate feedback signals for tracking the sun direction using a feedback control system [38]. The sensors equipped with the radiance limiting tubes are carried and oriented by a dualaxis mechanical system to find the sun direction such as that shown in Fig. 1, and then, the correct angles of the sun position obtained by the sensors are used by the solar tracker to orient

449

Fig. 1. Sensor of sensor based dual-axis solar tracker equipped with a radiance limiting tube, and mounted on an independent dual-axis mechanical system [39].

the PV module face toward the sun [39]. In fact, two independent dual-axis mechanical systems are needed; one for carrying the sensors, and another for PV module. It is clear that the reference points of the two mechanical systems should be identical, and this increases the complicacy of the solar tracker. Thus, the advantages and disadvantages of sensor based dual-axis solar trackers can be summarized as follows. They are high accurate, so that, their tracking error is about 0.15° [40], but they need one extra dual-axis mechanical system together with several sensors and radiance limiting tubes that significantly increase the cost and complicacy. A sensorless dual-axis solar tracker acts as an open loop system. It uses the offline estimated geographic data about the sun path in the sky obtained from different sun path charts or solar map equations [41]. Similarly, the advantages and disadvantages of sensorless dual-axis solar trackers can be summarized as follows. They are cheaper but their tracking error is averagely about 0.4°, so they are less accurate compared to the sensor based type because of using a set of offline estimated data rather than the online data indicating the actual position of the sun in the sky at the moment [42]. A new set of data is also needed by changing the geographical latitude and/or longitude of the PV module location. Furthermore, they follow the routine path of the sun in a cloudy sky because as mentioned, they operate as an open loop system, so there is not any feedback signal informing the actual environmental conditions. In this study, a novel high accurate sensorless dual-axis solar tracking system controlled by the internal MPPT unit is proposed. The MPPT controller continuously calculates the maximum output power of the PV module at any time, and then, it uses the deviations of the altitude and azimuth angles to find the correct direction of the sun where the greatest value of the maximum output power is obtained. In fact, the PV module itself plays the role of a sensor, so the proposed sensorless dual-axis solar tracker operates as a closed loop system which uses the online data indicating the actual position of the sun in the sky at the moment. Thus, it has the advantages of both sensor based and sensorless dual-axis solar

450

H. Fathabadi / Applied Energy 173 (2016) 448–459

trackers, but it does not have their disadvantages. In other words, on the one hand, it is a very high accurate solar tracker, so that, its tracking error is about 0.11° which is even less than that in sensor based trackers because it uses only one dual-axis mechanical system rather than the two independent mechanical systems used in sensor based trackers. On the other hand, similar to sensorless trackers, its cost and complication are less than the sensor based type because it does not need any sensors and extra dual-axis mechanical system for orienting the sensors. The properties of the proposed sensorless dual-axis solar tracker are compared with sensor based and sensorless dual-axis solar trackers in Table 1. The comparison explicitly shows that this study introduces a novel category of the solar trackers having the advantages of both sensor based and sensorless trackers, and this is the contribution of this work. A PV system has been constructed to implement the proposed solar tracker. Real experimental results are presented that not only validate theoretical results but also verify that the proposed sensorless dual-axis solar tracker is a very high accurate tracker which has the advantages of both the sensor based and sensorless trackers. The rest of this paper is organized as follows. The analysis of the proposed solar tracking system is presented in Section 2. Section 3 deals with the implementation of the proposed solar tracking system. Experimental results are presented in Sections 4 and 5 concludes the paper. 2. Analysis of the proposed solar tracking system The schematic diagram of the proposed solar tracking system is shown in Fig. 2. The solar tracker section consists of the stepper motor 1 which adjusts the altitude angle of the PV module/panel, the stepper motor 2 that adjusts azimuth angle, the altitude gear box that rotates the PV module/panel in the vertical plane around the altitude axis, and the azimuth gear box which similarly rotates

the PV module/panel in the horizon plane around the azimuth axis. To make clear the point, the horizon plane, vertical plane, altitude axis, azimuth axis, altitude angle symbolized by a, and azimuth angle symbolized by b are shown in Fig. 3. The output power of the PV module shown in Fig. 2 can be expressed as:

Ppv ¼ V pv Ipv

ð1Þ

where Ppv ; V pv and Ipv are respectively the PV module output power, voltage and current. The P–V characteristic of a PV module is shown in Fig. 4. It can be seen that the conditions which should be satisfied at the MPP are as:

8 dPpv < dV ¼0 pv

ð2Þ

: dPpv ¼ 0 dIpv

In fact, the MPPT controller shown in Fig. 2 adjusts the PV module voltage and current, so that, the operating point of the PV module is ideally placed at the MPP. Thus, by utilizing the MPPT controller, the output power of the PV module and its operating point can be expressed as:

8 > < Ppv ¼ V mpp Impp ¼ Pmpp V pv ¼ V mpp > : Ipv ¼ Impp

ð3Þ

where Pmpp ; V mpp and Impp are respectively the PV module output power, voltage and current at the MPP. In the rest of this study, it is shown that the MPPT controller can be also used to produce appropriate control signals to track the sun direction by adding a novel feature to the MPPT controller that is explained in detail. The I–V and P–V characteristics of a typical PV module for a constant temperature (T ¼ 25  C) and the different solar irradiance impinging on the PV module are shown in Fig. 5. The P–V curves depicted

Table 1 Comparison between the proposed dual-axis solar tracker, sensor based trackers, and sensorless trackers. Dual-axis solar tracker type

Cost & complication

Accuracy

Tracking error

Sensors used in tracker

Number of dual-axis mechanical systems used

Sensor based Sensorless Proposed solar tracker

High Low Low

High Medium High

0.15° 0.4° 0.11°

Yes No No

2 1 1

Fig. 2. Schematic diagram of the proposed solar tracking system.

451

H. Fathabadi / Applied Energy 173 (2016) 448–459

Fig. 3. Coordinate axises: altitude axis, azimuth axis, altitude angle, and azimuth angle.

Fig. 5. I–V and P–V characteristics of a typical PV module for a constant temperature (T ¼ 25  C) and the different solar irradiance impinging on the PV module.

The goal is to find the correct tracking angles aT and bT . When the altitude and azimuth angles of the PV module/panel reach the correct tracking angles aT and bT , i.e., the PV module/panel is positioned at the correct tracking position, the following conditions mathematically appear:

( @f ða

T ;bT Þ @a @f ðaT ;bT Þ @b

¼0

ð5Þ

¼0

Using Eq. (4), the deviation of the maximum output power is obtained as:

DPmpp ¼

ð6Þ

where Da and Db are respectively the deviation of the altitude and azimuth angles. At the correct tracking position, the deviation of the maximum output power is obtained by substituting Eq. (5) in Eq. (6), so

Fig. 4. P–V characteristic of a typical PV module.

in Fig. 5 verify this clear point that an increase/decrease in the solar irradiance impinging on a PV module causes an increase/decrease in the maximum output power (Pmpp ) of the PV module, so the value of the maximum output power can be used as a parameter to track the sun direction. When sunbeam strikes the PV module surface perpendicularly, the solar irradiance impinging on the PV module is maximum, and so the greatest value of the maximum output power can be observed by the MPPT controller. In fact, the maximum output power of a PV module increases when the perpendicular axis of the PV module surface approaches to the sun direction, and finally gets its maximum value when the PV module reaches the position where its perpendicular axis is parallel with the sun direction. In this case, the PV module surface is exactly perpendicular to sunray that is the correct tracking position, and so the sun direction has been completely tracked by the solar tracker. It can be summarized that the value of the maximum output power depends on the altitude angle (a) and azimuth angle (b) of the PV module/panel, and gets its maximum value when the altitude angle reaches the correct tracking altitude angle aT , and similarly, the azimuth angle is equal to the correct tracking azimuth angle bT . Thus, the maximum output power of the PV module can be mathematically expressed by a bounded continuous function as:

Pmpp ¼ f ða; bÞ

@f ða; bÞ @f ða; bÞ Da þ Db @a @b

ð4Þ

DPmpp ¼

@f ðaT ; bT Þ @f ðaT ; bT Þ Da þ Db ¼ 0 @a @b

ð7Þ

Using Eq. (6), two slopes of the maximum output power are defined as:

8 < M 1 ¼ DPmpp ¼ @f ða;bÞ þ @f ða;bÞ Da @a @b : M2 ¼

D P mpp Db

¼

@f ða;bÞ Da @a Db

þ

Db ; Da

@f ða;bÞ ; @b

where Da – 0 where Db – 0

ð8Þ

where M1 and M2 are respectively the maximum output power slopes with respect to the altitude and azimuth angles. Substituting Eq. (7) in Eq. (8) verify that the maximum output power slopes M1 and M2 both become zero when the PV module/panel reaches the correct tracking position. The deviations of the maximum output power, altitude angle, and azimuth angle are computed by the MPPT controller as:

8 > < DPmpp ¼ Pmpp ðkÞ  P mpp ðk  1Þ Da ¼ aðkÞ  aðk  1Þ > : Db ¼ bðkÞ  bðk  1Þ

ð9Þ

where Pmpp ; a, and b are respectively the kth sample of the maximum output power, altitude angle, and azimuth angle during analog to digital conversion in the MPPT controller. Thus, the two

452

H. Fathabadi / Applied Energy 173 (2016) 448–459

slopes M1 and M2 defined by Eq. (8) can be expressed based on the kth samples as: 8 < M 1 ¼ Pmpp ðkþ1ÞPmpp ðkÞ ; where Da ¼ aðk þ 1Þ  aðkÞ – 0 aðkþ1ÞaðkÞ k ¼ 0; 1; 2; 3;. .. : M 2 ¼ Pmpp ðkþ1ÞPmpp ðkÞ ; where Db ¼ bðk þ 1Þ  bðkÞ – 0 bðkþ1ÞbðkÞ

ð10Þ A flowchart depicting the detailed operation of the proposed solar tracker controlled by the MPPT controller is shown in Fig. 6 that is explained in detail as follows. The solar tracking system starts with appropriate initial altitude and azimuth angles a0 and b0 at daybreak, so at first (k ¼ 0), the angles are as að0Þ ¼ a0 and bð0Þ ¼ b0 . These initial angles are chosen based on the geographical location of the PV module/panel, so that, the PV module/panel approximately turns to the direction toward the point of the horizon where the sun rises at daybreak. Then, the initial maximum output power (P mpp ) associated with these initial angles is computed by the MPPT controller as P mpp ð0Þ ¼ V mpp ð0ÞImpp , where V mpp and Impp are respectively the MPP voltage and current at this step. At the next step (k ¼ 1), the altitude and azimuth angles are increased by the MPPT controller as:



að1Þ ¼ að0Þ þ Da; Da ¼ 0:12 bð1Þ ¼ bð0Þ þ Db;

Db ¼ 0:12

ð11Þ

Then, the maximum output power (P mpp ) associated with these angles is computed by the MPPT controller as Pmpp ð1Þ ¼ V mpp ð1ÞImpp , where V mpp and Impp are respectively the MPP voltage and current at this step., and the maximum output power slopes M1 and M 2 are calculated using Eq. (10) as:

8 < M1 ¼ Pmpp ð1ÞPmpp ð0Þ að1Það0Þ : M2 ¼ Pmpp ð1ÞPmpp ð0Þ bð1Þbð0Þ

ð12Þ

Based on the calculated maximum output power slopes M 1 and M2 , one of the following nine cases appears, and the altitude and azimuth angles are varied by the MPPT controller according to the following nine roles to find the appropriate values of these angles for the next step (k ¼ 2): Case 1: If M 1 ¼ M 2 ¼ 0, then the sun direction has been completely tracked, so there is no need to change the altitude and azimuth angles a and b, and thus



að2Þ ¼ að1Þ bð2Þ ¼ bð1Þ

ð13Þ

Case 2: If M 1 ¼ 0 and M2 > 0, then there is no need to change the altitude angle a, but the azimuth angle b should be increased, so

Fig. 6. Operational flowchart of the proposed solar tracking technique.

453

H. Fathabadi / Applied Energy 173 (2016) 448–459



að2Þ ¼ að1Þ bð2Þ ¼ bð1Þ þ Db; Db ¼ 0:12

ð14Þ

Case 3: If M 1 ¼ 0 and M2 < 0, then there is no need to change the altitude angle a, but the azimuth angle b should be decreased, so



að2Þ ¼ að1Þ bð2Þ ¼ bð1Þ  Db; Db ¼ 0:12

ð15Þ

Case 4: If M 1 > 0 and M2 ¼ 0, then there is no need to change the azimuth angle b, but the altitude angle a should be increased, so



að2Þ ¼ að1Þ þ Da; Da ¼ 0:12 bð2Þ ¼ bð1Þ

ð16Þ

Case 5: If M 1 > 0 and M 2 > 0, then the altitude and azimuth angles both should be increased as:



að2Þ ¼ að1Þ þ Da; Da ¼ 0:12 bð2Þ ¼ bð1Þ þ Db;

Db ¼ 0:12

ð17Þ

Case 6: If M 1 > 0 and M2 < 0, then the altitude/azimuth angle should be increased/decreased as:



að2Þ ¼ að1Þ þ Da; Da ¼ 0:12 bð2Þ ¼ bð1Þ  Db;

Db ¼ 0:12

ð18Þ

Case 7: If M 1 < 0 and M2 ¼ 0, then there is no need to change the azimuth angle b, but the altitude angle a should be decreased, so



að2Þ ¼ að1Þ  Da; Da ¼ 0:12 bð2Þ ¼ bð1Þ

ð19Þ

Case 8: If M 1 < 0 and M 2 > 0, then the altitude/azimuth angle should be decreased/increased as:



að2Þ ¼ að1Þ  Da; Da ¼ 0:12 bð2Þ ¼ bð1Þ þ Db;

Db ¼ 0:12

ð20Þ

Case 9: If M 1 < 0 and M 2 < 0, then the altitude and azimuth angles both should be decreased as:



að2Þ ¼ að1Þ  Da; Da ¼ 0:12 bð2Þ ¼ bð1Þ  Db;

Db ¼ 0:12

ð21Þ

The above-mentioned nine roles are repeated for all the next steps (k ¼ 3; 4; 5; . . .). Since the function Pmpp ¼ f ða; bÞ introduced by Eq. (4) is a bounded continuous function, it has a global maximum point, and so the process converges to the case 1. In other words, after a sequence of iterations, the altitude and azimuth angles both comply with the criteria mentioned in case 1 (M1 ¼ M2 ¼ 0), and thus, the sun direction is exactly tracked.

3. Implementation of the proposed solar tracking system The structure of the PV system constructed to implement the proposed solar tracker is shown in Fig. 7. It consists of a PV module, a DC/PWM converter, a MPPT controller, two stepper motor drivers, two stepper motors, and two gear boxes.

3.1. Electrical components 3.1.1. Stepper motors Two identical stepper motors having a step angle of 1.8°; one for adjusting altitude angle, and another for adjusting azimuth angle.

3.1.2. Stepper motor drivers Two identical stepper motor drivers; each drives the related stepper motor by delivering appropriate control signals and supply voltage, so that, the stepper motor rotates according to the direction and the steps number requested by the MPPT controller.

Fig. 7. Structure of the PV system constructed to implement the proposed solar tracker.

454

H. Fathabadi / Applied Energy 173 (2016) 448–459

switches with a constant switching period of T i ¼ 1=f i , and a duty ratio of DS ¼ t Son =T i , where f i and tSon are the switching frequency and the switch S turn-on time, respectively. When S is turned on, the load current IL flows through S and arrives to the load RL . When S is turned off, IL immediately reaches zero, so during tSon , the load current IL is expressed as:

IL ðtÞ 

Fig. 8. Proposed simple DC/PWM converter.

V pv ¼ IB Rds þ RL

ð22Þ

where Rds is the static drain to source on-resistance of the MOSFET switch S. In steady state, V pv is approximately constant, and the voltage loss across the filter inductor Lpv is negligible, so the load current IL can be considered as a constant current IB during t Son . Thus, the load current IL can be estimated as the PWM waveform shown in Fig. 9. The DC term (average) of the load current symbolized by ILDC can be obtained as:

ILDC ¼ DS

V pv  DS I B Rds þ RL

ð23Þ

Since the filter capacitor C pv is large enough, the voltage across it can be considered as a constant voltage. Thus, the current flowing through C pv can be ignored, and the average load current ILDC is approximately equal to the average PV current Ipv DC , and so, it is found using Eq. (23) that Fig. 9. Waveform of the load current IL ðtÞ.

Table 2 Parameters of the PV circuit, and the specifications of the PV module KC200GT connected to the constructed PV system. PV module KC200GT Current at MPP Ipv mpp (A) Voltage at MPP V pv mpp (V) Output power at MPP P pv mpp (W) Short-circuit current ISC (A) Open-circuit voltage V OC (V) Stepper motor NEMA 23

DC/PWM converter 7.61 26.3 200.1430 8.21 32.9

10 f i (kHz) MOSFET switch S IRF1407 PV filter C pv (l F) 470 Lpv (l H) 820 Stepper motor driver AMIS-30543

3.1.3. DC/PWM converter The simple DC/PWM converter used in the constructed system is shown in Fig. 8. It consists of only one MOSFET switch S that

Ipv DC  ILDC  DS IB

ð24Þ

Eq. (24) shows that the duty ratio DS is a control signal, so that, the PV current can be controlled and adjusted to a specific level by varying it. 3.1.4. MPPT controller In this study, the MPPT technique presented in [30] has been used to track the MPP. It adjusts the operating point of the PV module to the MPP by varying the duty ratio DS . 3.1.5. PV module One commercial PV module Kyocera KC200GT has been used in this study. The technical specifications of PV module Kyocera KC200GT under STC (standard test condition: Solar irradiance G ¼ 1000 W m2 , air mass (AM) 1.5 solar radiation spectrum, cell temperature T ¼ 25  C, and solar angle hz ¼ 48:19 ) extracted from its datasheet is summarized in Table 2.

Fig. 10. Mechanical components of the constructed solar tracker: (a) Azimuth gear box, (b) Altitude gear box, (c) Whole of the assembled mechanical part together with the PV module KC200GT installed on it.

H. Fathabadi / Applied Energy 173 (2016) 448–459

455

Fig. 11. Electrical circuit of the constructed PV system including the electrical part of the proposed solar tracker.

3.2. Mechanical components As shown in Fig. 7, the mechanical components are the two identical gear boxes; an altitude gear box for rotating the PV module in the vertical plane around the altitude axis, and an azimuth gear box for rotating the PV module in the horizon plane around the azimuth axis. Thus, the altitude gear box has been installed horizontally on the PV module axis while the azimuth gear box has been placed vertically as shown in Fig. 10(a) and (b). Each gear box consists of two gearwheels; one primary gearwheel, and one secondary gearwheel with the gear ratio of N 1 =N 2 ¼ 1=15. Since each stepper motor has a step angle of 1.8°, and the gearbox interconnected to it has a gear ratio of 1/15, so each step of the rotation of a stepper motor (1.8°) is converted into a rotation of 0.12° by the related gearbox. It can be summarized that the rotation of the altitude/azimuth stepper motor during one step (1.8°) is converted into a 0.12°-rotation of the PV module around the altitude/azimuth axis by the altitude/azimuth gearbox. The whole of the assembled mechanical part of the solar tracking system together with the PV module KC200GT installed on it is shown in Fig. 10(c).

proposed solar tracking technique depicted in Fig. 6. The PV current is measured by INA 168, and then, is delivered to the analog to digital conversion (ADC) pin of the microcontroller. The PV voltage is similarly sensed by the ADC pin of the microcontroller after scaling it using the potentiometer RV1. The PV voltage and current both are sampled with a sampling period of 100 ls. The opto-diac S delivers a periodic switching pulse with the period of T i ¼ 100 ls (f i ¼ 10 kHz) and the duty ratio of DS produced and adjusted by the microcontroller to track the MPP of the PV module [30]. In each step, the microcontroller MC68HC11A8 (MPPT controller) calculates the two maximum output power slopes M 1 and M 2 ,

4. Experimental results A PV system has been built to experimentally validate the theoretical results, and to evaluate the performance of the proposed solar tracker controlled by the MPPT unit. The electrical circuit of the constructed PV system is shown in Fig. 11. The MPPT controller has been implemented using a microcontroller MC68HC11A8. The microcontroller has been programmed according to the MPPT algorithm presented in [30] and the operational flowchart of the

Fig. 12. Variation of the solar irradiance impinging on the PV module provided using four 20% filmy glasses (T ¼ 25  C).

456

H. Fathabadi / Applied Energy 173 (2016) 448–459

been also used; one for adjusting the altitude angle (a), and another for azimuth angle (b). According to Eqs. (14)–(21), in each step, the microcontroller produces and delivers ten control signals to the two drivers that cause a rotation of 1.8° in each stepper motor in the correct direction. As mentioned in Section 3.2, a rotation of 1.8° in the altitude/azimuth stepper motor is converted into a 0.12°-rotation of the PV module around the altitude/azimuth axis by the altitude/azimuth gearbox. The above process is repeated until the altitude and azimuth angles both comply with the criteria mentioned in case 1 (M 1 ¼ M2 ¼ 0), i.e., the altitude and azimuth angles both reach the correct tracking angles aT and bT , and thus, the sun direction is tracked with a tracking error of 0.11° in both altitude and azimuth angles. For convenience, the technical specifications of the different parts of the constructed PV system are also summarized in Table 2. Fig. 13. Experimental waveform of the PV output power: static-dynamic response of the MPPT controller under uniform and non-uniform irradiance conditions.

and varies the altitude and azimuth angles to find the correct tracking angles at any time based on the nine roles presented by Eqs. (13)–(21) in Section 2. Two stepper motor drivers AMIS30543 have been used, one for driving the altitude stepper motor and another for driving the azimuth stepper motor. Each deriver has been connected to the microcontroller through the five wires, and receives five control signals from the MPPT controller. Two stepper motors NEMA 23 with a holding torque of 9 kg cm have

4.1. Evaluation of the MPPT controller At first, one experiment was set up to evaluate and test the performance of the implemented MPPT controller to track the MPP. The static-dynamic response of the MPPT controller was obtained under STC, uniform and non-uniform irradiance conditions. The solar irradiance impinging on the PV module is 1000 W m2 under STC. Four 20% filmy glasses were used, each 20% filmy glass positioned on the PV module causes a 20% shading effect, so one filmy glass causes a 20% reduction in the solar irradiance while two 20%

Fig. 14. Comparing between the actual altitude and azimuth angles measured from hour to hour in daylight (6:00–21:00) on Aug. 19, 2015 and the values tracked by the proposed solar tracker.

457

H. Fathabadi / Applied Energy 173 (2016) 448–459

Fig. 15. Comparing between the daily output powers of the PV module on the tracking mode and fixed mode: (a) On Jan. 19, 2014 in winter, (b) on Aug. 19, 2015 in summer.

filmy glasses situated on the PV module provide a 40% reduction in the solar irradiance, etc. At first, the four 20% filmy glasses were positioned on the PV module KC200GT connected to the constructed PV system, so the solar irradiance impinging on the PV module was 200 W m2 . After each one minute, one filmy glass was brought out during about 5 s, so an increase of 200 W m2 in the solar irradiance together with uniform and non-uniform irradiance conditions appeared. The approximation of the solar irradiance variation is shown in Fig. 12. The real experimental waveform of the PV module output power showing the staticdynamic response of the implemented MPPT controller under uniform and non-uniform irradiance conditions is shown in Fig. 13. Comparing the static-dynamic response shown in Fig. 13 with the solar irradiance variation shown in Fig. 12 and the experimental data reported in the datasheet of PV module KC200GT verifies that the MPPT controller has a fast high accurate static-dynamic response under different conditions. For instance, on the one hand, Fig. 13 shows that for G ¼ 800 W m2 , the average output power is 160.11 W. On the other hand, the maximum output power of PV

module KC200GT for G ¼ 800 W m2 reported in its datasheet is 160.69 W, and so the MPPT efficiency is obtained as:

MPPT efficiency ¼

160:11  100 ¼ 99:64% 160:69

ð25Þ

This explicitly verifies that the constructed MPPT controller high accurately tracks the MPP. 4.2. Performance evaluation of the solar tracker In second experiment, the constructed solar tracking system was activated. The actual altitude and azimuth angles of the sun position in the sky observed from the PV module situated in the geographic position of (37.9667°N, 23.7167°E) were measured from hour to hour in daylight (6:00–21:00) on Aug. 19, 2015 using two digital protractors with the accuracy of 0.01°; one for measuring the altitude angle and another for the azimuth angle. Fig. 14 (a) and (b) compares the actual values of the altitude and azimuth angles of the sun position to the altitude and azimuth angles

458

H. Fathabadi / Applied Energy 173 (2016) 448–459

Table 3 Average daily electric energy produced by the PV module in the four seasons, and increase in energy efficiency resulted from utilizing the proposed solar tracker (tracking mode). Time

Spring Summer Autumn Winter During one year

Average daily electric energy produced by the PV module (kW h) Fixed mode

Tracking mode

1.5764 2.0220 1.7314 1.1422 1.6180

2.1060 2.9037 2.3945 1.4712 2.2189

Increase in energy efficiency (%)

33.6 43.6 38.3 28.8 37.1

tracked by the proposed solar tracker at the same times. The initial values of the altitude and azimuth angles were chosen 2° and 75°, respectively, i.e., að0Þ ¼ a0 ¼ 2 and bð0Þ ¼ b0 ¼ 75 . The comparison shows that the maximum value of the tracking error is 0.11°, so the tracking error of the proposed solar tracker is 0.11° both in the azimuth and altitude angles that experimentally validates the theoretical result presented in Section 2. Thus, it can be summarized that the proposed solar tracker high accurately tracks the sun direction in the sky. In other experiments, the daily output powers of the PV module were measured from hour to hour in daylight on Jan. 19, 2014 in winter and on Aug. 19, 2015 in summer. Each measurement was performed twice; at the first time, it was measured when the implemented solar tracker was active (tracking mode). At the second time, it was measured when the solar tracker was off, and the PV module was fixed at the noon position of the sun (fixed mode). The MPPT process has been active on both tracking and fixed modes. The daily output powers measured in winter and summer for both the tracking and fixed modes are shown in Fig. 15(a) and (b), respectively. It can be seen from Fig. 15(a) that, in winter, the daily electric energies produced by the PV module on the fixed and tracking modes are respectively about 1.1422 kW h and 1.4712 kW h, and thus, the solar energy converted on the tracking mode is 28.8% more than that on the fixed mode. Similarly, Fig. 15(b) shows that the daily electric energies obtained on the fixed and tracking modes are respectively about 2.0220 kW h and 2.9037 kW h in summer, so there is an increase of 43.6% on the tracking mode. The above-mentioned daily energy measurements and the other measurements related to spring and autumn are summarized in Table 3. 4.3. Cost analysis As shown in Fig. 7, the proposed solar tracker is implemented by adding only two stepper motors, two stepper motor drivers, and two gear boxes to the conventional system. In the constructed system shown in Fig. 11, the prices of the added components are $80 (two stepper motors NEMA 23), $40 (two stepper motor drivers AMIS-30543), and $70 (two gear boxes), and so the total price is about $190. Table 3 shows that, throughout one year, the PV module KC200GT used in the constructed system can produce the average daily electric energies of 1.6180 kW h and 2.2189 kW h on the fixed and tracking modes, respectively. Thus, the PV module daily produces an extra energy of 0.6 kW h by utilizing the proposed solar tracker. On the other hand, the average electricity price around the world is about 0.2 $/kW h, so the price of the extra daily electric energy produced by the PV module is $0.12. Comparing this price with the total price of the solar tracker components ($190) shows that utilizing the solar tracker returns its extra cost ($190) after only 1583 days ($190=$0:12  1583 days).

5. Conclusion In this paper, a novel high accurate sensorless dual-axis solar tracking system was proposed. It is controlled by the MPPT controller which calculates the maximum output power of the PV module, and uses the altitude and azimuth angles deviations to track the sun position in the sky. The proposed solar tracking system only uses the actual direction of the sun to track the sun direction in the sky. Thus, unlike all other sensorless solar tracker, it is a closed loop system which uses online actual data. The proposed solar tracking system has the advantages of both sensor based and sensorless dual-axis solar trackers, but it does not have their disadvantages. A photovoltaic system was built, and the real experimental results were presented that explicitly validated the theoretical results. It was shown that the proposed solar tracking system tracks the sun direction with the tracking error of 0.11°. The energy efficiency of a PV system can be increased by 28.8–43.6% depending on the seasons using the proposed solar tracking system.

References [1] Rezk H, Eltamaly AM. A comprehensive comparison of different MPPT techniques for photovoltaic systems. Sol Energy 2015;112:1–11. [2] Fathabadi H. Lambert W function-based technique for tracking the maximum power point of PV modules connected in various configurations. Renew Energy 2015;74:214–26. [3] Hong CM, Ou TC, Lu KH. Development of intelligent MPPT (maximum power point tracking) control for a grid-connected hybrid power generation system. Energy 2013;50:270–9. [4] Mellit A, Kalogirou SA. MPPT-based artificial intelligence techniques for photovoltaic systems and its implementation into field programmable gate array chips: review of current status and future perspectives. Energy 2014;70: 1–21. [5] Shen CL, Ko YX. Hybrid-input power supply with PFC (power factor corrector) and MPPT (maximum power point tracking) features for battery charging and HB-LED driving. Energy 2014;72:501–9. [6] Ozdemir S, Altin N, Sefa I. Single stage three level grid interactive MPPT inverter for PV systems. Energy Convers Manage 2014;80:561–72. [7] Enslin JHR, Wolf MS, Snyman DB, Swiegers W. Integrated photovoltaic maximum power point tracking converter. IEEE Trans Ind Electron 1997;44:769–73. [8] Park M, Yu In-Keun. A study on the optimal voltage for MPPT obtained by surface temperature of solar cell. 30th annual conference of IEEE, vol. 3. p. 2040–5. [9] Masoum MAS, Dehbonei H, Fuchs EF. Theoretical and experimental analyses of photovoltaic systems with voltage and current-based maximum power-point tracking. IEEE Trans Energy Convers 2002;17(4):514–22. [10] Noguchi T, Togashi S, Nakamoto R. Short-current pulse-based maximumpower-point tracking method for multiple photovoltaic-and-converter module system. IEEE Trans Ind Electron 2002;49(1):217–23. [11] Salah CB, Ouali M. Comparison of fuzzy logic and neural network in maximum power point tracker for PV systems. Electric Power Syst Res 2011;81:43–50. [12] Algazar MM, AL-monier H, EL-halim HA, El Kotb Salem ME. Maximum power point tracking using fuzzy logic control. Int J Electr Power Energy Syst 2012;39 (1):21–8. [13] Guenounou O, Dahhou B, Chabour F. Adaptive fuzzy controller based MPPT for photovoltaic systems. Energy Convers Manage 2014;78:843–50. [14] Mellit A, Sag˘lam S, Kalogirou SA. Artificial neural network-based model for estimating the produced power of a photovoltaic module. Renew Energy 2013;60:71–8. [15] Rizzo SA, Scelba G. ANN based MPPT method for rapidly variable shading conditions. Appl Energy 2015;145:124–32. [16] Abdelsalam AK, Massoud AM, Ahmed S, Enjeti PN. High-performance adaptive perturb and observe MPPT technique for photovoltaic-based microgrids. IEEE Trans Power Electron 2011;26(4):1010–21. [17] Jiang JA, Huang TL, Hsiao YT, Chen CH. Maximum power tracking for photovoltaic power systems. J Sci Eng 2005;8(2):147–53. [18] Ahmed J, Salam Z. An improved perturb and observe (P&O) maximum power point tracking (MPPT) algorithm for higher efficiency. Appl Energy 2015;150:97–108. [19] Muthuramalingam M, Manoharan PS. Comparative analysis of distributed MPPT controllers for partially shaded stand alone photovoltaic systems. Energy Convers Manage 2014;86:286–99. [20] Liu F, Duan S, Liu F, Liu B, Kang Y. A variable step size INC MPPT method for PV systems. IEEE Trans Ind Electron 2008;55(7):2622–8. [21] Mei Q, Shan M, Liu L, Guerrero JM. A novel improved variable step-size incremental-resistance MPPT method for PV systems. IEEE Trans Ind Electron 2011;58(6):2427–34. [22] Tey KS, Mekhilef S. Modified incremental conductance MPPT algorithm to mitigate inaccurate responses under fast-changing solar irradiation level. Sol Energy 2014;101:333–42.

H. Fathabadi / Applied Energy 173 (2016) 448–459 [23] Brunton SL, Rowley CW, Kulkarni SR, Clarkson C. Maximum power point tracking for photovoltaic optimization using ripple-based extremum seeking control. IEEE Trans Power Electron 2010;25(10):2531–40. [24] Lei P, Li Y, Seem JE. Sequential ESC-based global MPPT control for photovoltaic array with variable shading. IEEE Trans Sustain Energy 2011;2(3):348–58. [25] Bazzi AM, Krein PT. Concerning maximum power point tracking for photovoltaic optimization using ripple-based extremum seeking control. IEEE Trans Power Electron 2011;26(6):1611–2. [26] Munteanu I, Bratcu AI. MPPT for grid-connected photovoltaic systems using ripple-based extremum seeking control: analysis and control design issues. Sol Energy 2015;111:30–42. [27] Orabi M, Hilmy F, Shawky A, Qahouq JAA, Hasaneen E-S, Gomaa E. On-chip integrated power management MPPT controller utilizing cell-level architecture for PV solar system. Sol Energy 2015;117:10–28. [28] Ahmed J, Salam Z. A maximum power point tracking (MPPT) for PV system using Cuckoo Search with partial shading capability. Appl Energy 2014;119: 118–30. [29] Daraban S, Petreus D, Morel C. A novel MPPT (maximum power point tracking) algorithm based on a modified genetic algorithm specialized on tracking the global maximum power point in photovoltaic systems affected by partial shading. Energy 2014;74:374–88. [30] Fathabadi H. Novel fast dynamic MPPT (maximum power point tracking) technique with the capability of very high accurate power tracking. Energy 2016;94:466–75. [31] Quesada G, Guillon L, Rousse DR, Mehrtash M, Dutil Y, Paradis P-L. Tracking strategy for photovoltaic solar systems in high latitudes. Energy Convers Manage 2015;103:147–56.

459

[32] Sallaberry F, Pujol-Nadal R, Larcher M, Rittmann-Frank MH. Direct tracking error characterization on a single-axis solar tracker. Energy Convers Manage 2015;105:1281–90. [33] Chong KK, Wong CW. General formula for on-axis sun-tracking system and its application in improving tracking accuracy of solar collector. Sol Energy 2009;83(3):298–305. [34] Barker L, Neber M, Lee H. Design of a low-profile two-axis solar tracker. Sol Energy 2013;97:569–76. [35] Al-Mohammad A. Efficiency improvement of photovoltaic panels using a sun tracking system. Appl Energy 2004;79(3):345–54. [36] Eke R, Senturk A. Performance comparison of a double-axis sun tracking versus fixed PV system. Sol Energy 2012;86(9):2665–72. [37] Roth P, Georgiev A, Boudinov H. Cheap two axis sun following device. Energy Convers Manage 2005;46(7–8):1179–92. [38] Zhang P, Zhou G, Zhu Z, Li W, Cai Z. Numerical study on the properties of an active sun tracker for solar streetlight. Mechatronics 2013;23(8):1215–22. [39] Cañadaa J, Utrillasb MP, Martinez-Lozanob JA, Pedrósb R, Gómez-Amob JL, Maja A. Design of a sun tracker for the automatic measurement of spectral irradiance and construction of an irradiance database in the 330–1100 nm range. Renew Energy 2007;32(12):2053–68. [40] Yao Y, Hu Y, Gao S, Yang G, Du J. A multipurpose dual-axis solar tracker with two tracking strategies. Renew Energy 2014;72:88–98. [41] Duarte F, Gaspar PD, Goncalves LC. Two axes solar tracker based on solar maps controlled by a low-power microcontroller. J Energy Power Eng 2011;5(7): 671–6. [42] Tirmikci CA, Yavuz C. Comparison of solar trackers and application of a sensor less dual axis solar tracker. J Energy Power Eng 2015;9:556–61.