Ejector-performance-oriented model predictive control of MED-TVC systems under motive steam pressure disturbances
Year: 2026
Type: article
Abstract: This study addresses the challenge of maintaining high steam utilization efficiency in Multi-Effect Distillation with Thermal Vapor Compression (MED-TVC) systems under motive steam pressure disturbances. Owing to the high sensitivity of conventional fixed-geometry ejector entrainment performance to off-design pressure conditions, conventional temperature regulation strategies can maintain process stability but are generally unable to prevent the ejector from deviating from its optimal high-efficiency operating region. Consequently, system thermal efficiency may deteriorate under pressure disturbances. To address this problem, an ejector-performance-oriented model predictive control (EPO-MPC) strategy is proposed. The strategy combines a dynamic MED model, a data-driven ejector model constructed from computational fluid dynamics (CFD) data and thin-plate-spline (TPS) interpolation, and an ejector-performance-oriented setpoint optimizer embedded in a constrained MPC framework. The optimizer updates the final-effect evaporator temperature setpoint in real time so that the ejector remains close to its high-entrainment double-choked region. Simulations under motive steam pressure variations from 900 to 1300 kPa show that the EPO-MPC achieves a higher entrainment ratio (ER) and gain output ratio (GOR) compared to the PID controller and conventional MPC. The practical efficacy of the EPO-MPC is further validated through physical experiments. Results show that, compared to the PID and conventional MPC, the EPO-MPC achieves a 3.53% and 1.55% increase in GOR under realistic steam pressure fluctuations.
Institution: Shandong University
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,620
OpenAlex ID: W7171413880
