Intelligent calculating prediction control method of thermal power generating unit coordination control system

Abstract

The invention discloses an intelligent calculating prediction control method of a thermal power generating unit coordination control system. The method comprises the following steps of step1, determining a sampling time interval Ts of prediction control; step2, through an unit characteristic test, acquiring a local model Gm described by an input-output difference equation form of a boiler steam turbine system under M typical load conditions; step3, using model prediction control to carry out square wave signal control simulation on all the local models Gms in the step2; step4, using control simulation data of all the local models Gms to train an intelligent calculating model so as to form an intelligent prediction controller IPC; step5, taking the intelligent prediction controller IPC acquired from the step4 through training as a real-time controller, collecting needed input signals and sending into the intelligent prediction controller IPC, wherein the intelligent prediction controller IPC automatically calculates and acquires a control increment instruction delta u (t) needed for adjusting the boiler steam turbine under any conditions. In the invention, accuracy is high, a reaction speed is fast and an on-line calculated amount is small.

Description

Technical field [0001] The invention relates to an intelligent calculation predictive control method of a thermal power generating unit coordinated control system, and belongs to the technical field of automatic control of a thermal power generating unit. Background technique [0002] The task of the power station boiler-turbine coordinated control system is to pass the fuel quantity u on the boiler side b And the opening of the turbine governing door u t Joint adjustment of the load, timely response to intermediate load command N 0 Required unit output N e , While maintaining the unit main steam pressure P t In the allowable setting value P 0 Scope, the coordination of the two is an indispensable link to improve the economic benefits of the generator set and ensure the safe operation of the generator set. It is increasingly being valued by power generation manufacturers and scientific researchers. [0003] Nowadays, almost all large-scale power plant units must participate in the p...

AI Technical Summary

Problems solved by technology

[0006] For example, the Chinese patent "Model-Based Dynamic Decoupling Coordinated Control Method for Thermal Power Units and Furnaces" (200910154934.X) proposed a control method based on model-based dynamic decoupling for the coupling problem between the characteristics of the coordination system of machines and furnaces. However, the main control scheme is still conventional PID control, so there are two problems: (1) The decoupling model with fixed parameters has poor robustness to unit load changes; (2) Conventional PID control is limited to three parameter degrees of freedom, and it is still difficult to achieve satisfactory control effects

However, because the system is affected by many factors at any time, the results of online identification often cannot reflect or even completely deviate from the real coordination system characteristics, and the control effect at this time cannot meet the expected control quality.

[0008] There are also some patents and documents that use segmented multi-model predictive control, but they all consider the problem from the perspective of the model describing the characteristics of the system itself, and still need to perform complex predictive control calculations online, which is not conducive to the widely used distributed control system at present. (DCS), even through a dedicated industrial controller such as PLC, these solutions need to occupy a large amount of memory overhead, which is very unfavorable for PLC devices with limited memory, thus limiting the popularization and application of control solutions

[0009] From the above analysis, it can be seen that the existing boiler-turbine coordinated control methods still have certain limitations or defects.

Images