22 February 2016
Advanced Process Control (MPC) is a mature technology and has been become the standard approach in industry for improving product quality, operation efficiency and enterprise profit. In this talk, I will address the recent development of economic assessment of Two-layer Industrial MPC system from two aspects, (i) An optimization approach for economic performance assessment of MPC system based on LQG benchmark and (ii) An Iterative Learning Control (ILC) approach for continuous economic performance improvement of MPC system. Industrial MPC system typically consists of Steady State Optimization (SSO) layer and Dynamic Optimal Control (DOC) layer. By explicitly incorporating uncertainty into the performance assessment problem, economic performance evaluation can be formulated as a stochastic optimization problem by integrating SSO and DOC layers. This helps to identify the opportunity to improve proï¬tability of the process by taking appropriate risk levels. Using the LQG benchmark to estimate achievable variability reduction through control system improvement, the proposed method provides an estimate of both the performance that can be expected from the improved control system and the operating condition that delivers the improved performance. The above LQG based economic performance assessment requires accurate process model and computationally demanding. In order to online improve the economic performance of MPC, we further developed an ILC approach to optimize the MPC performance iteratively. With the iterative learning control (ILC) strategy, SSO problem is solved at each trial to update the tuning parameter and designed condition of DOC, then DOC is conducted in the condition guided by SSO. The ILC strategy is proposed to adjust the tuning parameter of DOC based on the sensitivity analysis. The convergence of EPD by the proposed ILC has also been proved. The performance of the proposed method is illustrated via an SISO numerical system as well as an MIMO industry process.