Interval Observer based Extended Predictive Controller for Heavy Duty Gas Turbine

As the main component of gas-steam combined cycle power plant, heavy-duty gas turbine (HDGT) with the property of safety, reliability and high efficiency, which can greatly improve the performance of deep-peak shaving capability and fast frequency modulation performance of power system. With respect to the variable load control strategy for HDGT, an interval observer based extended model predictive control is proposed in this study by combining the interval observer and model predictive control, which is built upon the stability analysis of linear state feedback control. The salient property of proposed method lies in the reduction of negative effect caused by external in the process of variable load control. The typical study cases are selected for verifying the effect of the proposed methods, which outperforms the others methods in aspect of stable and safe.

1 Introduction

With the advantages of high efficiency, low pollution, and good peak shaving, the gas-steam combined cycle power plant (CCPP) is of significance for achieving energy conservation, emission reduction and low-carbon development of the electric power industry [1]. Among them, the heavy-duty gas turbine (HDGT) is an important component of CCPP with the property of high nonlinearity and complex component, etc., which has a significant impact on the operation performance of economic cost [2]. New energy, mainly wind power and photovoltaic, is connected to the power grid, and the uncertainty and fluctuation of new energy output will affect the energy regulation of the power system [3]. PID based HDGT needs to operate under variable working conditions frequently to satisfy the above-mentioned challenges [4]. Considering the variation of working conditions, the model of equipment can be modeled by using data-driven mechanism dual drive [5]. Therefore, it is urgent to develop a fast flexible and robust controller for HDGT controller to satisfy the demand for “deep” and “flexible” peak shaving of power system.