Model predictive control of a heat recovery steam generator during cold start-up operation using piecewise linear models. (5th June 2017)

• Record Type: Journal Article
• Authors: Sindareh-Esfahani, Peyman; Tabatabaei, S. Sepehr; Pieper, Jeff K.
• Is Part Of: Applied thermal engineering. Volume 119(2017)
• Page Start: 516
• Page End: 529
• Publication Date: 2017-06-05
• Subjects: Heat recovery steam generator -- Cold start-up -- Prediction error method -- Model predictive control -- Piecewise systems; Heat engineering -- Periodicals; Heating -- Equipment and supplies -- Periodicals; Periodicals; 621.40205
• DOI: 10.1016/j.applthermaleng.2017.03.041
• Languages: English
• ISSNs: 1359-4311

Abstract

Highlights: Piecewise linear model for cold start-up. Experimentally validated data from simulations. Cold start-up optimization using both linear and nonlinear models. Application of model predictive control. Optimization of the modeling and control processes. Abstract: Proposed is a new optimization scenario for the cold start-up of a heat recovery steam generator in a combined cycle power plant. A nonlinear model consisting of all components of a heat recovery steam generator including economizer, drum, and superheater is considered as the main plant. Another system is identified based on prediction error method to design the control inputs of the system. The model obtained based on prediction error method is a piecewise linear model validated using a wide set of experimental data acquiring from a heat recovery steam generator during cold start-up. In order to improve the performance of the heat recovery steam generator cold start-up, a quadratic cost function is proposed based on wall temperature of superheater, drum pressure, exhaust gas temperature and water flow in the economizer. During cold start-up, some variables and rates should remain bounded and are constrained in the cost function, directly improving the lifespan of the heat recovery steam generator. The optimization targets are met using model predictive control. Simulation results are compared with actual plant data showing the effectiveness and performance of the proposed control strategy.