The Application Study Of Multivariable Intelligent Control Strategy In The Steam Temperature System

Control optimization of thermal power plants plays a decisive role in realizing the target of Chinese overall energy saving, the steam temperature control system is one of the most important part of the boiler control system. As a typical example of the complex systems with great inertia, wide operating range, non-linear and time varying, the steam temperature control system is rather complex. Traditional control strategy is difficult to obtain satisfactory control performance, which can not meet the growing control requirements, a number of scholars have engaged in the study of the steam temperature intelligent control strategy.To solve the steam temperature control problem of a600MW unit in Tuoketuo power plant, the researching object is the steam temperature (superheated and reheat steam temperature) control system, a multi-variable intelligent control strategy is proposed in this paper and then integrates it into the DCS (Distributed Control System) platform, the work can be summarized as the following:1. Analysis the coupling characteristics of steam temperature variables. Based on a massive field running data, a method of analysis and processing to extract the control variables and disturbance variables of the steam temperature multivariable system is proposed in this paper, based on the concept of the maximum information compression index, a common and effective steam temperature coupling characteristics table has been developed which provides a reference for engineering application.2. Steam temperature system TITO Model. Based on real-time operating data of the Tuoketuo power plant, a two-input and two-output system(TITO) is proposed to express superheated steam and reheat steam temperature system, an improved λ-sectional fuzzy weighted GK clustering algorithm is proposed to build steam temperature system multivariable fuzzy model, the obtained fuzzy model can be validated to prove its reliability.3. The controller design of typical load. At first, Combination of fuzzy multivariable systems, a kind of compensation programme is designed in this paper, main steam temperature of cascade dual loop steam temperature TITO system is controlled by fuzzy PID composite controller, parameters optimization of the main steam temperature controller and the reheat steam temperature controller is finished by Matlab Compensator design tool, steam temperature control under typical load can achieve satisfactory control performance.4. Steam temperature control of the full range of operating conditions. In order to solve the problem that model parameters change with Load, Multi-model control strategy is proposed for steam temperature mode control loops of three typical loads, a improved fuzzy supervisory is applied to realize switching without disturbance of steam temperature control in order to achieve the temperature intelligent control of the full conditions.5. Semi-physical simulation platform built based on Matlab-DCS. Based on the research aboved, semi-physical simulation platform based on Matlab-DCS is developed this paper. Intelligent controller is developed into EDPF-NT system of Guodianzhishen, combined with the simulation of the steam temperature model in Matlab, the communication between Matlab and DCS is UDP/Modbus protocol. For convenience, interactive interface and integrated toolbox are designed both in DCS and Matlab, the feasibility and accuracy of the simulation platform can be seen from the operated result.