Jaakko Larjola

Organic Rankine Cycle Power Systems: From the Concept to Current Technology, Applications, and an Outlook to the Future

The cumulative global capacity of organic Rankine cycle (ORC) power systems for the conversion of renewable and waste thermal energy is undergoing a rapid growth and is estimated to be approx. 2000 MWe considering only installations that went into operation after 1995. The potential for the conversion of the thermal power coming from liquid-dominated geothermal reservoirs, waste heat from primary engines or industrial processes, biomass combustion, and concentrated solar radiation into electricity is arguably enormous. ORC technology is possibly the most flexible in terms of capacity and temperature level and is currently often the only applicable technology for the conversion of external thermal energy sources. In addition, ORC power systems are suitable for the cogeneration of heating and/or cooling, another advantage in the framework of distributed power generation. Related research and development is therefore very lively. These considerations motivated the effort documented in this article, aimed at providing consistent information about the evolution, state, and future of this power conversion technology. First, basic theoretical elements on the thermodynamic cycle, working fluid, and design aspects are illustrated, together with an evaluation of the advantages and disadvantages in comparison to competing technologies. An overview of the long history of the development of ORC power systems follows, in order to place the more recent evolution into perspective. Then, a compendium of the many aspects of the state of the art is illustrated: the solutions currently adopted in commercial plants and the main-stream applications, including information about exemplary installations. A classification and terminology for ORC power plants are proposed. An outlook on the many research and development activities is provided, whereby information on new high-impact applications, such as automotive heat recovery is included. Possible directions of future developments are highlighted, ranging from efforts targeting volume-produced stationary and mobile mini-ORC systems with a power output of few kWe, up to large MWe base-load ORC plants.

Power Limits of High-Speed Permanent-Magnet Electrical Machines for Compressor Applications

The maximum-power limits for high-speed permanent-magnet (PM) electrical machines for air compressor applications are determined in the speed range 20000-100000 r/min. For this purpose, five PM machines are designed and the electromagnetic, thermal, and mechanical designs of each machine are simultaneously performed. The critical values of the thermal and mechanical constraints are considered in order to obtain the maximum powers of the electrical machines. The electromagnetic losses generated in the machine are the output parameters of the electromagnetic design and input parameters for the thermal design. The thermal design is performed using a multiphysics method, which couples computational-fluid-dynamics equations with heat-transfer equations. The mechanical design considers the retention of the rotor elements against the huge centrifugal forces that arise during the high-speed operation and also the rotor dynamics properties of the rotor. The reliability of these design techniques is experimentally validated in the paper. The obtained maximum-power limit defines the speed-power region, in which the high-speed PM electrical machines intended for compressor applications can have a safe operation.

Numerical Simulation of Real-Gas Flow in a Supersonic Turbine Nozzle Ring

In small Rankine cycle power plants, it is advantageous to use organic media as the working fluid. A low-cost single-stage turbine design together with the high molecular weight of the fluid leads to high Mach numbers in the turbine. Turbine efficiency can be improved significantly by using an iterative design procedure based on an accurate CFD simulation of the flow. For this purpose, an existing Navier-Stokes solver is tailored for real gas, because the expansion of an organic fluid cannot be described with ideal gas equations. The proposed simulation method is applied for the calculation of supersonic flow in a turbine stator. The main contribution of the paper is to demonstrate how a typical ideal-gas CFD code can be adapted for real gases in a very general, fast, and robust manner.

Siloxanes as Working Fluids for Mini-ORC Systems Based on High-Speed Turbogenerator Technology

This paper presents a study aimed at evaluating the use of siloxanes as the working fluid of a small-capacity (≈10kWe) ORC turbogenerator based on the “high-speed technology” concept, combining the turbine, the pump, and the electrical generator on one shaft, whereby the whole assembly is hermetically sealed, and the bearings are lubricated by the working fluid. The effects of adopting different siloxane working fluids on the thermodynamic cycle configuration, power output, and on the turbine and component design are studied by means of simulations. Toluene is included into the analysis as a reference fluid in order to make comparisons between siloxanes and a suitable low molecular weight hydrocarbon. The most influential working fluid parameters are the critical temperature and pressure, molecular complexity and weight, and, related to them, the condensation pressure, density and specific enthalpy over the expansion, which affect the optimal design of the turbine. The fluid thermal stability is also extremely relevant in the considered applications. Exhaust gas heat recovery from a 120 kW diesel engine is considered in this study. The highest power output, 13.1 kW, is achieved with toluene as the working fluid, while, among siloxanes, D4 provides the best simulated performance, namely 10.9 kW. The high molecular weight of siloxanes is beneficial in low power capacity applications, because it leads to larger turbines with larger blade heights at the turbine rotor outlet, and lower rotational speed if compares, for instance, to toluene.

Computational and Experimental Comparison of Different Volute Geometries in a Radial Compressor

Two different volute geometries of a radial compressor at three different operating points have been analyzed using Computational Fluid Dynamics and detailed laboratory measurements. The performance of the volutes were compared using steady-state CFD-analysis, where the volute and the impeller with diffuser were modeled separately. In addition, a time dependent simulation of the complete compressor using the sliding mesh technique was performed for one operation point. Both volutes were manufactured and the overall performance of the compressor, the pressure distribution in the volute and the flow field in the volute inlet were measured with the respective volute geometries. The results obtained from steady, quasi-steady and time-accurate simulations are compared with experimental data.Copyright

CFD Analysis of a Centrifugal Compressor Impeller and Volute

In this study, centrifugal compressor performance was predicted using CFD. Three-dimensional time-averaged impeller and volute simulations were performed using a Navier–Stokes code. The presented performance prediction method has been divided into three phases. Firstly, the impeller was calculated with a vaneless diffuser. That gives inlet boundary conditions for the volute analysis and the pressure ratio at the diffuser exit. Next, the volute analysis was performed and a static pressure recovery coefficient obtained. Finally, that result was combined with the pressure ratio prediction from the impeller analysis, and the overall compressor performance thus obtained.Copyright

Electricity from waste heat using the organic rankine cycle (ORC)

In the conversion of low temperature heat into electricity the greatest efficiency is obtained in many cases by using an organic Rankine cycle (ORC). This paper deals with an ORC-design, in which a high-speed oil free turbogenerator-feed pump is used. Freedom from oil significantly increases the thermal stability of the organic fluid and makes a hermetic design possible.

Experimental and Numerical Study of Real-Gas Flow in a Supersonic ORC Turbine Nozzle

Using organic matter as the working fluid in small Rankine cycle power plants is beneficial. However, high molecular weight of the fluid and the single-stage design of the turbine lead to a supersonic flow in the turbine. An Organic Rankine Cycle (ORC) plant was designed and tested. Toluene was used as the working fluid and as lubricant. The turbine and the feed pump were placed on the same shaft as the high-speed generator in the designed 175 kW unit. CFD simulations were used in the design process. Toluene is behaving as a real gas in the nozzle. To ensure an accurate simulation, a real gas model of toluene was implemented in an existing Navier-Stokes flow solver. Polynomial and rational regression were used to achieve the functions for the gas properties. The pressure and temperature were measured at the nozzle inlet and outlet. In the CFD simulations the nozzle ring was modelled with and without a temperature probe in order to model the effect of the probe to the flow field and compare the simulated pressure and temperature values against the measurements. The nozzle geometry was also modelled in 2D and 3D in order to see the effect of the 3D in the flow field. There was quite a good agreement between the measured and simulated data. The agreement in the temperature was better than in the pressure. The effect of 3D on the simulation results was minor, which was expected. The simulated flow field revealed that the shock waves developing in the trailing edge of the nozzle were seen in the turbine rotor inlet.Copyright

Numerical Investigation of the Effect of Tip Clearance to the Performance of a Small Centrifugal Compressor

Numerical analysis is conducted for the 3-dimensional impeller and vaneless diffuser of a small centrifugal compressor. The influence of impeller tip clearance is investigated. A Navier-Stokes flow solver Finflo has been applied for the simulation. A practical real gas model has been generated for the calculation. Simulations with different sizes of tip clearance at different mass flow rates have been made. The results are compared to experimental results at a certain tip clearance and one operating point. Reasonable agreement has been obtained. The ideal gas model has also been applied to compare with the real gas model. The numerical results show that tip clearance has a significant effect on the performance of a small centrifugal compressor. As the size of tip clearance increases, both the pressure ratio and the efficiency decrease. The decreasing rate of efficiency is higher at higher mass flow rates and lower at lower mass flow rates. The input power of the compressor hardly changes with different sizes of tip clearance, but increases as the mass flow rate increases. The incidence of impeller and flow angle at the exit of the impeller increase as the size of tip clearance increases. Correlations of the size of tip clearance with the efficiency drop and change of flow angle at the exit of impeller are given. The detailed flow distribution shows that as the size of tip clearance increases, the tangential leaking flow at the tip clearance makes the low velocity flow region grow larger and move from the suction-shroud corner to the center of the flow channel. The main flow at the pressure side is compressed and accelerated. Therefore the uniformity of the flow in the whole channel decreases. The detailed flow distribution also shows that the leaking flow is stronger at higher mass flow rates.Copyright

Optimization of the Part-Load Control Strategy for a Two-Shaft Microturbine With Intercooling, Reheat and Recuperation

Microturbines have become popular among small-scale distributed energy systems. This paper focuses on a two-shaft arrangement where high efficiency is obtained through intercooling, reheat and recuperation. An optimized method for controlling the part-load performance via variable speed control of the generator shaft in addition to the turbine inlet temperature reduction is presented. The studied methods to reduce the power output were variable speed control of the generator shaft in combination with independent turbine inlet temperature control of both turbines. Optimization was performed by using a differential evolutionary algorithm to find a sufficient number of points at steadily reducing power settings to determine the optimal control curves for the three control parameters. In the microturbine model the operating values of the engine were obtained by solving the system of nonlinear equations formed by the governing relations. As a result an optimal part-load control method was found which provides better part-load efficiency than any of the studied control methods alone or in simple combinations could have provided. The optimal control strategy and the relative change of part-load electric efficiency were shown to be fairly independent of the design-point specifications for the turbomachinery and recuperator.Copyright