Search

This paper proposes a real-time capable augmented control scheme for a parallel evaporator organic Rankine cycle (ORC) waste heat recovery system for a heavy-duty diesel engine, which ensures efficient and safe ORC system operation. Assuming a time constant separation between the thermal and pressure dynamics, a nonlinear model predictive control (NMPC) is designed to regulate the mixed working fluid (WF) outlet temperature and the differential temperature between the two parallel evaporator outlets. Meanwhile, the evaporator pressure is regulated by an external PID control. The NMPC is designed using a reduced order, moving boundary control model of the heat exchanger system. In the NMPC formulation, state feedback is constructed from the estimated state via an unscented Kalman filter based on temperature measurements of the exhaust gas and WF at the evaporator outlet. The performance of the proposed control scheme is demonstrated in simulation over an experimentally validated, high fidelity, and physics-based ORC plant model during a transient constant speed and variable load engine drive cycle. The performance of the proposed control scheme (NMPC plus PID) is further validated via comparison with a conventional, multiple-loop PID controlling both the mixed evaporator outlet WF temperature, and the evaporator pressure. The simulation results demonstrate that the proposed control scheme outperforms a multiple-loop PID control in terms of both safety and total recovered thermal energy by up to 12% and 9%, respectively.

Organic Rankine Cycle (ORC) waste heat recovery systems offer promising engine fuel economy improvements for heavy-duty on-highway trucks. An ORC test rig with parallel evaporators to recover both tailpipe and EGR waste heat from a 13L heavy duty diesel engine was developed and used in this work to demonstrate a novel control strategy based on Model-Predictive Control (MPC). The main control objectives for the ORC system are: (i) regulation of working fluid temperature, (ii) safe turbine operation - away from 2-phase region, and (iii) maximization of waste heat recovery. The MPC uses a built-in moving boundary evaporator model to predict future system response and generate optimal actuator reference commands. Two variants of MPC were considered in this work: an adaptive linear MPC (LMPC) and a nonlinear MPC (NPMC). Compared with the traditionally used PID controller, MPC demonstrates more accurate temperature control and improved disturbance rejection in simulation. Finally, the LMPC and NMPC controllers were implemented on the ORC test rig and showing promising initial test results.

This paper discusses the control challenges of a parallel evaporator organic Rankine cycle (ORC) waste heat recovery (WHR) system for a diesel engine. A nonlinear model predictive control (NMPC) is proposed to regulate the mixed working fluid outlet temperature of both evaporators, ensuring efficient and safe ORC system operation. The NMPC is designed using a reduced order control model of the moving boundary heat exchanger system. In the NMPC formulation, the temperature difference between evaporator outlets is penalized so that the mixed temperature can be controlled smoothly without exceeding maximum or minimum working fluid temperature limits in either evaporator. The NMPC performance is demonstrated in simulation over an experimentally validated, high fidelity, physics based ORC plant model. NMPC performance is further validated through comparison with a classical PID control for selected high load and low load engine operating conditions. Compared to PID control, NMPC provides significantly improved performance in terms of control response time, overshoot, and temperature regulation.

This paper presents steady state power optimization for an Organic Rankine Cycle (ORC) waste heat recovery (WHR) system in heavy-duty diesel (HDD) applications and examines attainable power levels. Both recirculated and tail pipe exhaust gas streams are utilized as heat sources via a parallel evaporator configuration. A dynamic, physics-based ORC model is utilized to determine optimized net power production. The optimization variables include: working fluid pump speed, mass flow distribution ratio for the parallel evaporators, turbine speed, and the pump speed of condenser cooling fluid. A sensitivity analysis is conducted for all optimization variables. Working fluid pump speed and expansion turbine speed are demonstrated as the most sensitive variables. Sensitivity results indicate increasing system power production as the working fluid mass flow rate increases. A map is developed to identify optimal and safe ORC operating regions for a given engine cycle. Based on the sensitivity analysis working fluid pump speed is selected as the variable of interest in the steady state optimization and results indicate that system power generation could be improved by as much as 25%. Analysis of two quasi steady operational cycles, a multi-mode and a constant speed variable load are conducted. ORC power results indicate that a speed-constrained, mechanically coupled turbine expander operates in its peak efficiency zone to produce maximum power output for the steady-state cycles tested. These steady state power optimization results can be used for future power optimization during engine transients.