An organic Rankine cycle for two different heat sources: steam and hot water

Available Available online online at at www.sciencedirect.com www.sciencedirect.com

ScienceDirect

Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Available Energy Energy Procedia Procedia 00 00 (2017) (2017) 000–000 000–000

ScienceDirect ScienceDirect

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

Energy (2017) 000–000 883–890 EnergyProcedia Procedia129 00 (2017) www.elsevier.com/locate/procedia

IV International Seminar on ORC Power Systems, ORC2017 13-15 September 2017, Milano, Italy 15th International Symposium on District heat Heating and Cooling An organicThe Rankine cycle for two different sources: steam and water the heat demand-outdoor Assessing the feasibilityhot of using aa Taehong , Kyung Chun Kima,a,heat * temperature function for aSung long-term district demand forecast aa

School School of of Mechanical Mechanical Engineering, Engineering, Pusan Pusan National National University, University, Busan Busan 609-735, 609-735, South South Korea Korea

I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc a

IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France c Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France

Abstract Abstract

Direct Direct use use of of aa steam steam heat heat source source for for the the organic organic Rankine Rankine cycle cycle system system is is one one of of the the key key challenges challenges in in various various plant plant industries industries where where low-grade low-grade steam steam is is available available and and is is becoming becoming difficult difficult as as the the parent parent system system has has aa strong strong variability. variability. In In addition, addition, changing changing the type the type of of the the heat heat source source fluid fluid among among steam, steam, hot hot water, water, thermal thermal oil oil and and others others from from the the parent parent plant plant could could increase increase the the operation operation Abstract time time of of the the heat heat recovery recovery system, system, which which increases increases system system economics economics and and makes makes it it more more attractive. attractive. Although Although aa variety variety of of ORC ORC systems have been developed mainly using single heat source as hot water heat source or a heat transfer loop. An ORC system with systems have been developed mainly using single heat source as hot water heat source or a heat transfer loop. An ORC system with District heating networks are commonly addressed in the literature as one of the most effective solutions for decreasing the the direct use of a steam heat source has been rarely reported, and a system using two different heat sources have not been reported the direct use of a steam heat source has been rarely reported, and a system using two different heat sources have not been reported greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat yet. Here we the characteristics of system two different heat source fluids: steam and yet. Here wetoevaluate evaluate the performance performance characteristics of an an ORC ORC system using using two heat different heat in source fluids: could steam decrease, and hot hot sales. Due the changed climate conditions and building renovation policies, demand the future water. The target ORC system was originally developed for the hot water heat source and has a simple cycle configuration with water. The target ORC system was originally developed for the hot water heat source and has a simple cycle configuration with prolonging the investment return period. R245fa as and four components as: two-stage radial and plateR245fa as working working fluid, and is isiscomposed composed of typical four of components as: aademand two-stage radial turbine turbine and aa coupled coupled generator, plateThe main scope offluid, this paper to assess of thetypical feasibility using the heat – outdoor temperature functiongenerator, for heat demand type heat exchangers with a refrigerant tank, and a multi-stage centrifugal pump. The nominal net power output at the design point type heat exchangers with a refrigerant tank, and a multi-stage centrifugal pump. The nominal net power output at the design point forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 is 187.9 kW with the turbine expansion ratio of 9.5. The heat exchanger analysis showed that the steam heat source can be applied isbuildings 187.9 kWthat withvary the turbine ratio of 9.5.and Thetypology. heat exchanger that the steam heathigh) source canthree be applied in both expansion construction period Three analysis weather showed scenarios (low, medium, and district to this The heat and heat coefficient at region the showed aa large torenovation this system. system. The isothermal isothermal heat exchange exchange and high high heat transfer transfer coefficient at the the two-phase two-phase region of of heat the steam steam showed large scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained demand values were overdesign for the current application. We tested the ORC system using a steam generated from an incineration plant. The overdesign for the current application. We tested the ORC system using a steam generated from an incineration plant. The compared with results from a dynamic heat demand model, previously developed and validated by the authors. temperature pressure of steam were 302 the originally for temperature and pressure of the the only steamweather were 143.5℃ 143.5℃ and 302 kPa. kPa. We We showed that the ORC ORC system originally developed for the the The results and showed that when change and is considered, the showed margin that of error couldsystem be acceptable fordeveloped some applications hot water heat source could be used for the steam heat source without any major system changes. hot source demand could be was usedlower for thethan steam heat withoutscenarios any major system changes. (thewater errorheat in annual 20% forsource all weather considered). However, after introducing renovation © 2017 Published by Ltd. ©scenarios, 2017 The The Authors. Authors. Published by Elsevier Elsevier Ltd. (depending on the weather and renovation scenarios combination considered). error value increased up to 59.5% © 2017 Thethe Authors. Published by Elsevier Ltd. committee Peer-review under responsibility of the scientific of the IV International Seminar on ORC Power Systems. Peer-review under responsibility of the scientific committee of the the IV International Seminar onper ORC Powerthat Systems. The value of slope coefficient of increased on average within theIV range of 3.8% Seminar up to 8%on decade, corresponds to the Peer-review under responsibility the scientific committee of International ORC Power Systems. decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and Keywords: Organic Rankine cycle, different heat sources, waste recovery Keywords: Organic Rankine cycle, site site test, test, heat sources, waste heat heat recovery renovation scenarios considered). On different the other hand, function intercept increased for 7.8-12.7% per decade (depending on the coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and improve the accuracy of heat demand estimations. © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Corresponding author. author. Tel.: Tel.: +82 +82 515102324; 515102324; fax: fax: +82 +82 51557866. 51557866. ** Corresponding Cooling. E-mail address: address: [email protected] [email protected] E-mail

Keywords: Heat demand; Forecast; Climate change 1876-6102 © 1876-6102 © 2017 2017 The The Authors. Authors. Published Published by by Elsevier Elsevier Ltd. Ltd. Peer-review Peer-review under under responsibility responsibility of of the the scientific scientific committee committee of of the the IV IV International International Seminar Seminar on on ORC ORC Power Power Systems. Systems.

1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.

1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the scientific committee of the IV International Seminar on ORC Power Systems. 10.1016/j.egypro.2017.09.251

884 2

Taehong Sung et al. / Energy Procedia 129 (2017) 883–890 Taehong Sung and Kyung Chun Kim / Energy Procedia 00 (2017) 000–000

1. Introduction In many energy consumption areas, the thermal efficiency has been improved up to its achievable maximum using available economic technologies [1]. However, there is still a room for further improvement using other technologies which are not fully mature yet or still uneconomic. Organic Rankine cycle (ORC) is one of the potential candidates for this purpose because ORC was already commercialized [2] and has a typical payback period of 6-12 years depends on the heat source temperature [3]. Energy price and application area can change the economics. About 40-90 ORC units were globally introduced every year [4]. However, the application area is limited to the field which can be operated stably. This is because the economic performance of the ORC system is a bit ambiguous to invest and is greatly influenced by operation hours and partial-load operation. When the ORC system is used as the main engine such as geothermal and biomass power plants, the operation hour is long and the system can be driven by almost full load so that the system has a sufficient economics to install. Some bottoming systems also have a good economic performance if the parent system shows a stable and patterned operation like ORC for a steel arc furnace and internal combustion engines for continuous power production. However, the waste heat does not occur only in stable sources. Therefore, it is needed to find a way to increase the economic performance by decreasing the system cost, improving the partial-load performance, increasing the operation hours and others. The most fundamental factor for the economics is its initial cost which accounts for 70-80% of the total investment cost [5]. The ORC cost showed a power law tendency for the system size [6] but the cost has been unchanged recent time because the reduced cost of main components was offset by an additional cost from new regulations [7]. Energy price also plays an important role in the system economics, but they are not controllable. Maintenance cost is another important factor for the economics due to the high cost of mechanics and the production loss during the shutdown. Therefore, systems with less maintenance cycles are preferred. Financial cost has little impact on the system economics. In comparison, improving the total electricity production can be a key for the fine economics. ORC operated at partialload by controlling the pump and turbine speeds but this not provides the exergy optimized operation. For optimum operation, it is necessary to control the system characteristics for the relation of the mass flow rate and pressure ratio, which mainly depends on the turbine. The turbine can be optimized for the fluctuating heat source [8]. A variable inlet guide vane can improve the partial-load performance [9]. Multiple expanders in parallel also provide a similar performance improvement in the partial-load condition [10]. In this study, we propose a new concept for improving system total electricity production: mixing different heat sources. In this system, the ORC system can continue to run changing the heat source depending on its availability. The heat source can be the same or different type. Considering chemical plants, we designed a system that can use both the pressurized hot water and the steam. The ORC system was designed for the hot water heat source and it was tested with a steam heat source. 2. The hot water ORC for the steam heat source The proposed ORC system increases the operation hours by choosing one of the heat sources based on its availability. In this study, steam and pressurized hot water were used as the heat source options. The ORC system, especially the evaporator, needs to be designed for one of the two heat sources. A system suitable for one heat source can be over-designed or under-designed for another heat source because steam and water have different heat transfer characteristics and coefficients. The optimal design of a system using various heat sources will require a thermoeconomic analysis considering a detailed state of the heat sources. We studied an ORC system designed for the hot water heat source (we call it as a hot water ORC or hot water mode) without considering the specific economics to investigate the technical feasibility of the concept. The hot water ORC can directly use the steam heat source (a steam ORC or steam mode) without an additional evaporator overdesign due to a high heat transfer coefficient of the phase change of the steam when a sufficient amount of the heat is supplied. There are some considerations in steam mode because the steam has different heat transfer characteristics in the evaporator. First, the evaporator outlet temperature of the working fluid can be overheated beyond the acceptable limit due to the evaporator overdesign in the steam mode when a high-pressure steam (superheated steam) is supplied. Steam Rankine cycle systems, which have similar problems due to unstable steam flow from the boiler, are solving these problems by installing control valves at both the inlet of the evaporator for the steam and the outlet of the



Taehong Sung et al. / Energy Procedia 129 (2017) 883–890 Taehong Sung and Kyung Chun Kim / Energy Procedia 00 (2017) 000–000

885 3

evaporator for the working fluid. The former decreases the saturation temperature of the steam so that the working fluid temperature is limited to this. The steam temperature after the throttle valve is higher than the saturation temperature of the steam but the amount of heat from this superheating is smaller than that of phase change, so the influence is not great. Therefore, we can avoid the overheating problem by controlling the throttling valve properly. The later directly decreases the working fluid temperature, however, the pressure is also decreased which decreases the thermal performance of the ORC system. The ORC system has a valve to open and close the bypass which can be used for this purpose. Second, the direct heat exchange with the steam heat source can cause a damage to the turbine due to an unintentional acceleration. The compressible characteristics of the steam change the heat transfer characteristics which can cause a sudden increase in the working fluid mass flow rate and accelerate the expander. The above-mentioned feedback steam pressure control system can be a cause of this problem if it is not properly adjusted. 3. Experimental setup In this study, the operability and performance of the ORC system which was originally developed for the hot water heat source of 140℃ [11] were analyzed using the steam heat source. The ORC system has a simple configuration with a back-to-back two-stage turbine, a multistage centrifugal pump, a plate preheater, a plate evaporator, a plate condenser and a refrigerant tank. For the evaporation, two plate heat exchangers were used due to the size limitation of a single heat exchanger. As a working fluid, R245fa was used due to its good thermodynamic and safety properties. However, the global warming potential is as high as 1030, so it will be banned in the future. The design point evaporation and condensation pressures are 2090 kPa and 220 kPa. The nominal net power output is 198.3 kW with pump and turbine isentropic efficiencies of 60% and 70%. The detailed design process and cycle and components information were reported in the previous paper [5, 12]. We moved the ORC to the steam generating incineration plant to test the feasibility as shown in Fig. 1. The steam supply line has two feedback control valves to adjust the steam pressure and temperature and an Ogden pump was installed for steam trapping after the ORC evaporator and preheater. The steam source condition is shown in Table 1. The evaporator increases the working fluid temperature from 30.8℃ to 129.6℃ and requires the design point heat loads of 1830 kW. The steam saturation temperature of 164.95℃ is higher than the working fluid saturation temperature of 124.08℃. When we use the original steam for the ORC, the required steam mass flow rate for the ORC system is 0.9 kg/s and the ORC can operate at its design point.

Fig. 1. A photograph and schematics of the ORC system connected to the steam heat source.

Taehong Sung et al. / Energy Procedia 129 (2017) 883–890 Taehong Sung and Kyung Chun Kim / Energy Procedia 00 (2017) 000–000

886 4

Table 1. Steam source condition. Parameters

Source

Throttled

Measured

Steam temperature (℃)

165

148.3

143.5

Steam pressure (kPa)

700

250

302

Steam saturation temperature (℃)

164.95

127.41

133.75

Steam mass flow rate (kg/s)

>2.5

>2.5

-

However, as mentioned above the turbine inlet temperature needs to be limited by 150℃ to keep the bearing temperature limitation. Lowering the steam pressure has another advantage. The required steam mass flow rate is decreased because the low steam pressure increases the latent heat. So, the amount of vapor vent in the open tank is small because the steam condensate has a less sensible heat. Note that when the saturated steam expands from less than 3200 kPa, the steam temperature decreases but the vapor quality increases. We achieved the throttling using two valves. Assuming that the throttling process is an isenthalpic process, the steam temperature is dropped to 148.3℃ when the steam pressure is decreased to 250 kPa. In this case, the turbine inlet temperature cannot exceed the limitation when an excess amount of steam is supplied. In the experiment, the steam valves have successfully lowered the steam temperature below the limitation. However, the steam incoming conditions were changed as 302 kPa and 143.5℃ which provides enough heat for the net power output of 186.6 kW. This mismatch was occurred due to the design error of the steam valves and an unintentional temperature loss in the supply line. The evaporator heat transfer rate is calculated by multiplying the working fluid mass flow rate and the enthalpy difference between the evaporator inlet and outlet as Eq. (1). The turbine power is directly measured using a power meter and the net power output was calculated as Eq. (2). The superheating and subcooling temperature are calculated by subtracting the fluid temperature by the saturation temperature as Eq. (3). Detailed information for the measurement devices was reported in the previous article [5].

Q ev = m wf (hwf ,ev ,out − hwf ,ev ,in )  W W − W =

(1)

Tsuper= T − Tsat

(3)

net

turb

pump

(2)

4. Results and discussion 4.1. Operating characteristics of the steam ORC Fig. 2a shows the operating characteristics of the pump and the turbine. The ORC system was started as follows. First, the steam was supplied to the evaporator and the preheater. The steam pressure was regulated as low as possible to balance the heat balance between the steam source and the ORC because the extra heat in the steam is released to the atmosphere and fill the test site with the released steam. The working fluid is initially circulated through the bypass line to avoid the turbine failure from the liquid entering. The bypass line was closed after 120 seconds and the working fluid was supplied to the turbine. The turbine rotating speed was indirectly controlled by adjusting the load bank which is connected to the generator. During the partial-load conditions, the turbine speed was controlled at a lower speed than the rated speed of 15000 RPM. The controlled turbine speed range is 8000-14000 RPM. We avoided the resonance frequency and secured 1000 RPM margin for preparing a sudden acceleration. Fig. 3a shows the increasing tendency of superheating of the turbine inlet flow during the bypass circulation. We observed three different types of oscillations. The first one was occurred during the starting period. The working fluid mass flow rate showed a large oscillation due to no pressure drop in the bypass line. This variation can be suppressed by installing an element for the pressure drop such as an orifice. The second one was occurred after each pump control. As shown in Fig. 2a, the working fluid mass flow rate shows small, sudden fluctuations during the constant pump speed control but the effect on the turbine speed is small. Fig. 2b shows that the corresponding turbine acceleration



Taehong Sung et al. / Energy Procedia 129 (2017) 883–890 Taehong Sung and Kyung Chun Kim / Energy Procedia 00 (2017) 000–000

887 5

was less than 1000 RPM. These fluctuations were also suppressed as the system approaches the design point operating conditions. Another type of fluctuations was coming from the steam supply line. We used two throttling valve with the pressure feed-back control system for regulating both the incoming steam pressure and mass flow rate. The unoptimized feed-back control caused small, periodic fluctuations throughout the experiments. Fig. 3a shows that most of the heat transfer takes place in the preheater due to the heat exchanger oversizing. Fig. 4b also shows that the steam provided for the preheater was used. The collected steam after the preheater is a condensed steam but it for the evaporator is superheated steam. So, we can see that the evaporator worked as a superheater. However, the system already secured enough amount of the superheating temperature for the turbine safety (Fig. 4a), and the high superheating only threatens the turbine bearing safety with a small or no performance improvement. So, it would be better to bypass the evaporator in the steam mode to avoid additional pressure loss in the evaporator. Fig. 3b shows that the steam supply pressure increased as the system power increased. Compared to the steam supply design, the design point performance occurred at the steam pressure of 302 kPa. This is because the steam flow rate was lower than the design. In this case, a superheated working fluid of the temperature higher than 150℃ can be supplied to the turbine. Fig. 4a shows a gradual decrement in the working fluid superheating temperature. This is because the steam temperature has dropped in starting period to balance the heat rates between the steam and ORC system.

Fig. 2. (a) Pump speed and working fluid mass flow rate; (b) turbine speed and generator power output.

Fig. 3. (a) Preheater and evaporator heat rate and liquid level in the evaporator; (b) Temperature and pressure of the steam at the inlet and outlet of the preheater and evaporator.

888 6

Taehong Sung et al. / Energy Procedia 129 (2017) 883–890 Taehong Sung and Kyung Chun Kim / Energy Procedia 00 (2017) 000–000

Fig. 4. (a) Superheating temperature of the working fluid at the evaporator outlet and the turbine inlet; (b) Superheating temperature of steam at the inlet and outlet of the preheater and evaporator.

4.2. Comparison between the hot water ORC Fig. 5 shows the operating characteristics of the hot water ORC. When the pump speed was controlled, the hot water ORC also showed the mass flow rate fluctuations however, the variation is relatively small and the turbine rotation speed was not changed (Fig. 5a). The evaporator also showed the similar low liquid level in hot water mode (Fig. 5b). However, it is hard to distinguish that the liquid level indicates what quality point in the evaporator because we did not install sensors between them in the hot water line. For the comparison, two constant pump speed sections were sampled from the hot water and steam ORC tests as shown in Table 2. Fig. 6 shows the temperature-entropy diagram for the steam mode ORC system. In hot water mode, the preheater and the evaporator shared the heat loads as 1551 kW and 116 kW. In the steam mode, they shared the heat loads as 1729 kW and 16.6 kW. This, as expected, is the result of increased performance of the preheater located at the front as the heat transfer performance increases. Therefore, it was confirmed that the steam can be used as an alternative heat source if overheating and fluctuation problems are well treated. The steam ORC provides 3.7% higher isentropic efficiency. This is because the hot water ORC system was operated with cold cooling water which increases the pressure ratio up to 10.9. The used radial inflow turbine provides good isentropic efficiency over wide range of pressure ratio below the design point pressure ratio of 9.5. Over this pressure ratio, the turbine provides worse isentropic efficiency.

Fig. 5. Operating characteristics of the hot water ORC. (a) Working fluid mass flow rate and turbine rotating speed; (b) Total heat rate of the evaporator and the liquid level in the evaporator.



Taehong Sung et al. / Energy Procedia 129 (2017) 883–890 Taehong Sung and Kyung Chun Kim / Energy Procedia 00 (2017) 000–000

889 7

Table 2. Operating parameters of the ORC systems. Numbers indicated in Fig. 6. *Total heat rate and pressure losses. The hot water conditions between the evaporator and the preheater were not measured. **Water condensate increased the pressure. ***The working fluid conditions between two heat exchanger was erroneously measured. Parameters Working fluid side (1) Pump inlet (Tank outlet) (2) Pump outlet (Preheater inlet) (3) Preheater outlet (Evaporator inlet) (4) Evaporator outlet (5) Turbine inlet (6) Turbine outlet (Condenser inlet) (7) Condenser inlet (8) Condenser outlet (Tank inlet) Heat source side (9) Evaporator and preheater inlet (10) Preheater outlet (11) Evaporator outlet Heat sink side (12) Condenser inlet (13) Condenser outlet Turbine Mass flow rate Pressure ratio Rotational speed Generator power Generator efficiency Isentropic efficiency Preheater Heat rate (source/working fluid) Pressure loss (source/working fluid) Evaporator Heat rate (source/working fluid) Pressure loss (source/working fluid) Condenser Heat rate (sink/working fluid) Pressure loss (sink/working fluid) Pump Calculated motor power Pump and motor efficiency System Net power output Electric cycle efficiency

Hot water ORC

Steam ORC

197 kPa / 22.4 ℃ 2176 kPa / 24.1 ℃ (2124 kPa)*** / 128.2 ℃ 2131 kPa / 141.1 ℃ 2115 kPa / 139.8 ℃ 195 kPa / 80.4 ℃ 192 kPa / 80.0 ℃ 182 kPa / 28.9 ℃

276 kPa / 40.4 ℃ 2176 kPa / 41.4 ℃ 2126 kPa / 134.3 ℃ 2110 kPa / 135.7 ℃ 2098 kPa / 134.0 ℃ 275 kPa / 79.2 ℃ 269 kPa / 79.1 ℃ 262 kPa / 41.4 ℃

510 kPa / 142.5 ℃ -/479 kPa / 120.1 ℃

302 kPa / 143.1 ℃ 284 kPa / 129.6 ℃ 306 kPa** / 135.5 ℃

175 kPa / 17.9 ℃ 159 kPa / 27.8 ℃

227 kPa / 32.6 ℃ 126 kPa / 36.6 ℃

5.97 kg/s 10.9 14950 RPM 210.1 kW 0.983 0.717

7.01 kg/s 7.62 14051 RPM 202.5 kW 0.984 0.754

1650 kW* / 1551 kW 31.5 kPa* / 52.8 kPa*,***

- / 1729 kW 17.9 kPa / 51 kPa

- / 116 kW -/-

- / 16.6 kW -4.3 kPa** / 16 kPa

1437 kW / 1417 kW 16.2 kPa / 10 kPa

- / 1531 kW 101 kPa / 7.6 kPa

14.59 kW 0.601

15.92 kW 0.645

195.5 kW 0.117

186.6 kW 0.107

Fig. 6. Temperature-entropy diagram for the steam mode ORC system at near design point.

890 8

Taehong Sung et al. / Energy Procedia 129 (2017) 883–890 Taehong Sung and Kyung Chun Kim / Energy Procedia 00 (2017) 000–000

5. Conclusions In this study, we tested a feasibility of steam heat source mix for the ORC system which was originally developed for the hot water heat source. We tested the ORC system using a steam generated from an incineration plant. The test results are summarized as follows: 1) ORC system originally developed for the hot water heat source can be used for the steam heat source without any modification of the ORC system. The ORC system produced the net power output of 186.6 kW in steam mode. The lower power output is due to the low-pressure ratio in summer but the expander itself showed better isentropic efficiency with the higher turbine back pressure. The pumping power was slightly increased as 1.3 kW due to the high pump efficiency. The evaporator showed no difference between the hot water ORC and the steam ORC. There was no significance difference in the pressure loss on the working fluid side. 2) The steam valve has successfully lowered the steam temperature. Due to the design error of the steam valve, the steam supply pressure was increased to 302 kPa, but the steam supply temperature was controlled to 143.5℃ with an unintentional temperature loss in the supply line. 3) In the steam mode, the preheater showed enough heat transfer capacity for the ORC system. So, if a bypass is installed between the preheater and the turbine, the unnecessary pressure loss can be avoided. 4) In the steam mode, the ORC system showed two additional fluctuations. One is a sudden acceleration of the turbine rotating speed with a pump speed control which has a large effect during the starting period and was suppressed as the system approached the design point. The danger of the sudden acceleration could be avoided by securing a rotational speed margin. Another fluctuation was coming from the unoptimized feed-back control of the steam supply line. Acknowledgements This research was supported by the National Research Foundation of Korea (NRF) with a grant funded by the Korean government (MSIP) through the Global Core Research Center for Ships & Offshore Plants (GCRC-SOP, No. 2011-0030013). References [1] G. Bonvicini, Report for "Waste Heat Recovery for Power Valorisation with Organic Rankine Cycle Technology in Energy Intensive Industries", in, TASIO, 2015. [2] D. Maraver, J. Royo, V. Lemort, S. Quoilin, Systematic optimization of subcritical and transcritical organic Rankine cycles (ORCs) constrained by technical parameters in multiple applications, Appl Energ, 117 (2014) 11-29. [3] X.Q. Wang, X.P. Li, Y.R. Li, C.M. Wu, Payback period estimation and parameter optimization of subcritical organic Rankine cycle system for waste heat recovery, Energy, 88 (2015) 734-745. [4] H.X. Zhai, Q.S. An, L. Shi, V. Lemort, S. Quoilin, Categorization and analysis of heat sources for organic Rankine cycle systems, Renew Sust Energ Rev, 64 (2016) 790-805. [5] T. Sung, E. Yun, H.D. Kim, S.Y. Yoon, B.S. Choi, K. Kim, J. Kim, Y.B. Jung, K.C. Kim, Performance characteristics of a 200-kW organic Rankine cycle system in a steel processing plant, Appl Energ, 183 (2016) 623-635. [6] C. Walsh, P. Thornley, The environmental impact and economic feasibility of introducing an Organic Rankine Cycle to recover low grade heat during the production of metallurgical coke, J Clean Prod, 34 (2012) 29-37. [7] A. Guercio, R. Bini, 15 - Biomass-fired Organic Rankine Cycle combined heat and power systems, in: Organic Rankine Cycle (ORC) Power Systems, Woodhead Publishing, 2017, pp. 527-567. [8] S.Y. Cho, C.H. Cho, K.Y. Ahn, Y.D. Lee, A study of the optimal operating conditions in the organic Rankine cycle using a turbo-expander for fluctuations of the available thermal energy, Energy, 64 (2014) 900-911. [9] D.S. Hu, Y. Zheng, Y. Wu, S.L. Li, Y.P. Dai, Off-design performance comparison of an organic Rankine cycle under different control strategies, Appl Energ, 156 (2015) 268-279. [10] E. Yun, D. Kim, S.Y. Yoon, K.C. Kim, Experimental investigation of an organic Rankine cycle with multiple expanders used in parallel, Appl Energ, 145 (2015) 246-254. [11] T. Sung, K.C. Kim, Development of a 200-kW organic Rankine cycle power system for low-grade waste heat recovery, in press, J Clean Energ Technol, (2017). [12] S. Han, J. Seo, B.S. Choi, Development of a 200 kW ORC radial turbine for waste heat recovery, J Mech Sci Technol, 28 (2014) 5231-5241.