Design and experimental validation of an adaptive control law to maximize the power generation of a small-scale waste heat recovery system

Increasing the energy efficiency of industrial processes is a challenge that involves, not only improving the methodologies for design and manufacturing, but optimizing performance during part-load operation and transient conditions. A well-adopted solution consists of developing waste heat recovery (WHR) systems based on Organic Rankine Cycle (ORC) power units. The highest efficiency for such cycle is obtained at low superheating values, corresponding to the situation where the system exhibits time-varying nonlinear dynamics, triggered by the fluctuating nature of the waste heat source. In this paper, an adaptive control law using the Model Predictive Control (MPC) framework is proposed. This work goes a step beyond most of the existing scientific works in the field of ORC power systems, since the MPC controller is implemented in a lab-scale prototype, and its performance compared against a gain-scheduled PID strategy. The experimental results show that the adaptive MPC outperforms the gain-scheduled PID based strategy, as it allows to accurately regulate the evaporating temperature, while keeping vapor condition at the inlet of the expander i.e., the superheating, in a safe operating range, thus increasing the net power generation.