ORC 2017

14:20 Session 7A: Small Scale Systems (2)
Chair: Sylvain Quoilin

14:20 20 mins	Performance correlations for characterizing the optimal off-design operation of an ORC power system Rémi Dickes, Olivier Dumont, Sylvain Quoilin, Vincent Lemort Abstract: A common aspect of ORC power systems is the shifting nature of their operating conditions. Whether for waste heat recovery, solar or geothermal applications, the heat source (and sometimes the heat sink) conditions may fluctuate in time, leading the ORC system to adapt its working regime for performance (or safety) reasons. To this end, various control parameters may be adjusted, like the pump speed and the cooling load. For a given set of heat source and ambient conditions, these control parameters may be set to maximize various system outputs, like its net power generation or its net thermal efficiency. The goal of this work is to use a steady-state off-design model for optimizing the performance of a 2kW ORC unit over its complete range of operating conditions. The off-design model employed is charge-sensitive (i.e. it imposes the total mass of refrigerant in the system) and is able to derive the ORC internal conditions (pressure, temperature, mass flow rates, etc.) based on its boundary conditions only. Using this model, a complete mapping of the system optimal performance is generated. In order to exploit this performance mapping in higher-level simulations, a set of handy and polynomial-free correlations is developed to best fit the ORC performance. Ultimately, these correlations are aimed to be reused for characterizing other ORC power systems.
14:40 20 mins	Experimental investigation of the operating point of a 1-kW ORC system Chinedu Unamba, James Freeman, Steven Lecompte, Martin White, Oyeniyi Oyewunmi, Paul Sapin, Christos Markides Abstract: The organic Rankine cycle (ORC) is a promising technology for the conversion of waste heat from industrial processes as well as heat from renewable sources. Many efforts have been channeled towards maximizing the thermodynamic potential of ORC systems through the selection of working fluids and the optimal choice of operating parameters with the aim of improving overall system designs, and the selection and further development of key components. Nevertheless, experimental work has typically lagged behind modelling efforts. In this paper, we present results from tests on a small-scale (1 kWel) ORC engine consisting of a rotary-vane pump, a brazed-plate evaporator and a brazed-plate condenser, a scroll expander with a built-in volume ratio of 3.5, and using R245fa as the working fluid. An electric oil-heater acted as the heat source, providing hot oil at temperatures in the range 120-140 °C. The frequency of the expander was not imposed by an inverter or the electricity grid but depended directly on the attached generator load; both the electrical load on the generator and the pump rotational speed were varied in order to investigate the performance of the system. Based on the generated data, this paper explores the relationship between the operating conditions of the ORC engine and changes in the heat-source temperature, pump and expander speeds leading to working fluid flow rates between 0.0088 kg/s and 0.0337 kg/s, from which performance maps are derived. The experimental data is, in turn, used to assess the performance of both the individual components and of the system, with the help of an exergy analysis. In particular, the exergy analysis indicates that the expander accounts for the second highest loss in the system. Analysis of the results suggests that increased heat-source temperatures, working-fluid flow rates, higher pressure ratios and larger generator loads improve the overall cycle efficiency. Specifically, a 46% increase in pressure ratio from 2.4 to 4.4 allowed a 3-fold electrical power output increase from 180 W to 550 W, and an increase in the thermal efficiency of the ORC engine from 1 to 4%. Beyond reporting on important lessons learned in improving the performance of the system under consideration, comparisons will be shown for making proper choices with respect to the interplay between heat-source temperature, generator load, and pump speed in an ORC system.
15:00 20 mins	Experimental performance of a micro-ORC energy system for low grade heat recovery Michele Bianchi, Lisa Branchini, Andrea De Pascale, Francesco Melino, Valentina Orlandini, Saverio Ottaviano, Michele Pinelli, Pier Ruggero Spina, Alessio Suman Abstract: The state-of-the art of ORC energy systems is mainly dominated by large scale units in the MW range of power output, in the field of heat recovery at mid-high temperature levels (around 200-500°C), where multiple commercial realizations are available. Nevertheless, the cutting-edge niche of micro-ORC energy systems offers good solutions for low-temperature heat recovery. Many prototypes are currently under investigations, but a leading technology is not yet established. This work reports an experimental activity carried out for performance characterization of a prototypal micro-ORC energy system. In particular, the paper presents the test bench developed in the laboratories of the University of Bologna and the first obtained results in terms of thermodynamic performance and main components characteristics. The ORC system comprises a small reciprocating three-piston expander, run on R134a as operating fluid. Heat is provided to the ORC from an external source, via hot water at temperature below 100 °C, in order to simulate a low-enthalpy heat recovery process. The system rejects unused heat via a water cooled condenser. Thus, the investigated ORC is a plug and play system, requiring only to be connected to the hot and cold heat sources. The ORC system has been tested for prolonged operation at various thermal input conditions. In particular, the behavior of the key cycle parameters and performance indexes (e.g. max. and min. pressures, superheating temperature, expander isentropic efficiency, electric power output, etc.) are investigated as function of pump rotational speed (i.e. organic fluid mass flow rate), for three different set point values of the hot source (65 °C, 75 °C, 85 °C). The operating thermodynamic cycle has been completely characterized by means of a real-time measurement and acquisition tool, developed in LabVIEW environment. Performance variations of the system have been monitored: the electric power output ranges between 0.30 to 1.2 kW, with gross efficiency in the range 2.9-4.4 %, while the expander “electro-isentropic” efficiency results in the range of 35-42 %.
15:20 20 mins	A turbine based domestic micro ORC system Piotr Klonowicz, Łukasz Witanowski, Łukasz Jędrzejewski, Tomasz Suchocki, Piotr Lampart Abstract: The paper presents an analysis of a turbine based micro ORC system for a domestic biomass boiler. The assumed nominal boiler capacity is in the range between 15 kW and 20 kW. Such a small thermal output is particularly difficult from the point of view of power generation. One of the major issues is the economic side of the system. In general, small systems are relatively more expensive. Another problem is poor electrical efficiency which results from the efficiencies of the individual components such as the expansion unit and the feed pump. Various working fluids are considered and also various methods of evaporation such as thermal oil loop, pressurized water loop and direct evaporation. The system efficiencies are examined. The main focused is laid on the turbine design. Due to low volume flow rates the rotational speeds must be high, even exceeding 100 krpm for some of the designs. Various turbine types are considered, however, the preferred choice is partially admitted highly loaded impulse stage [1]. Very small volume flow rates lead to blade channels that are difficult to manufacture, throats of the supersonic nozzles can be particularly problematic. Manufacturing difficulties impose worse tolerances which lead to bigger relative clearances and blade edge thicknesses which have a negative impact on the efficiency [2]. High speed require appropriate bearing system which must be reliable and, particularly for such a small system, simple. Gas bearings (e.g. foil bearings) seem to fulfill these requirements [3]. [1] P. Klonowicz, F. Heberle, M. Preißinger, and D. Brüggemann, “Significance of loss correlations in performance prediction of small scale, highly loaded turbine stages working in Organic Rankine Cycles,” Energy, vol. 72, pp. 322–330, 2014. [2] E. Macchi and A. Predichizzi, “Efficiency Prediction for Axial-Flow Turbines Operating with Nonconventional Fluids,” J. Eng. power, vol. 4, no. 103, pp. 718–724, 1981. [3] G. Zywica, P. Baginski, and S. Banaszek, “Experimental Studies on Foil Bearing With a Sliding Coating Made of Synthetic Material,” J. Tribol., vol. 138, no. 1, p. 11301, Sep. 2015.
15:40 20 mins	Low temperature heat recovery in engine coolant for stationary and road transport applications Pierre Leduc, Pascal Smague, Arthur Leroux, Gabriel Henry Abstract: IFPEN and ENOGIA have joined forces to develop compact, low-cost ORC turbines for heat recovery in engine coolant for trucks and passenger cars. Low temperature ORC (around 90°C – 363 K heat source) is a specialty of ENOGIA which has sold to date more than 40 stationary ORC systems, with electrical power outputs ranging from 5 to 40 kW, covering a large variety of applications: agricultural biogas combined heat and power engines, landfill biogas plants, biomass boilers, concentrated solar thermal systems and geothermal plants. With IFPEN, new developments have been made to deliver a 100 kWe stationary ORC model and to adapt ENOGIA's approach to small ORC turbines for truck and automotive applications. Most of R&D projects related to ORC systems propose performance optimized heat recovery solutions with recovery in engine exhaust gas. However recovery potential is important, real life is much more complex, with severe thermal constraints on the system and complex control strategy in transient behavior, generating important development and manufacturing costs and risks in terms of system reliability in the long term. For these new product developments, IFPEN and ENOGIA have investigated another way for heat recovery, aiming at reducing the cost of the system and facilitating its integration onboard a truck or a passenger car. As heat source, engine coolant is selected for its stable temperature condition; obviously exergy level of this source is lower than for exhaust gas, but improvement on system efficiency and engine thermo-management can be done to lower the difference. At the end, ORC running pressure and temperature are low, allowing a lightweight, compact and low-cost solution. The presentation deals with the R&D work that has been performed, combining ORC system simulation, 3D computations for turbine flow optimization, component mechanical design and lab testing. Results show a potential to enhance fuel efficiency of a truck by about 3%.
16:00 20 mins	Evaluating the quality of steady-state multivariate experimental data in various ORC experimental setups Sylvain Quoilin, Olivier Dumont, Rémi Dickes, Vincent Lemort Abstract: An increasing amount of experimental data is being released in the ORC field. This data is required to assess and compare the performance of different machines, to point out the main sources of losses, or to calibrate and to validate models. However, this data is subject to many sources of noise and errors, such as sensor malfunctions, transient phenomena, operator misuse of the test rig, noise in the data acquisition chain, unaccounted for external influences, etc. It is therefore important to assess the data quality, by evaluating the level of noise, the presence of outliers, or to measure the « explainability » of the acquired variables with respect to the externally imposed operating condition. This task is far from straightforward, especially if the data is sparse and multivariate. Gaussian Processes (GP) are an active field of research in machine learning and provide a powerful tool for the above purposes. Their Bayesian formulation allows predicting the variable of interest for new/unseen data points and provides coherent estimates of predictive uncertainty. In this work, we use GPExp, a library developed at the University of Liège, to evaluate experimental data in a GP framework. The main features of the tool are the following: - Provide a smooth (de-noised) multivariate operating map of the measured variable w.r. to the inputs. - Determine which inputs are relevant to predict a selected output (feature selection) - Provide a sensitivity analysis of the output with respect to the inputs - Provide the accuracy (confidence intervals) to predict the output with a given set of inputs. This interval should be function of the data density. - Detect the observation that are likely to be outliers This paper presents the development of the tool and illustrates, through examples, how it can be used to detect the main dependencies, shortcomings and outliers in experimental data. Examples are first described for univariate or bivariate processes, whose quality can be assessed visually, and then extended to real ORC processes with multiple input variables. The experimental data relative to three different test rigs is compared.