Circulating fluid bed boiler online optimized self-learning control method

Abstract

The invention discloses a circulating fluid bed boiler online optimized self-learning control method, which is characterized in that: a load knowledge base, a bed temperature knowledge base, a secondary air volume knowledge base and a self-learning updating algorithm are provided, so a characteristic value and a running experience value of a boiler can be extracted, stored and utilized to improve the control performance. A heat efficiency online optimizer adjusting bed temperature and a second air volume are provided, a built-in Newton gradient optimization algorithm can be used to instantly optimize the combustion heat efficiency, so the combustion process of the boiler is approximate to an economical combustion best area. A single-loop-multi-impulse intelligent control algorithm can be adopted for controlling the temperature of a boiler bed layer, so multiple variables influencing the bed temperature can be coordinated and controlled after being effectively decoupled, and the integration of the safety target and the economic target can be realized. Due to the adoption of the control method, the safe, stable and high-efficient running of the circulating fluid bed boiler can be realized, the robustness is strong, fault-tolerant capacity for a preset experience value can be realized, and the debugging difficulty of the control system can be reduced.

Description

technical field [0001] The invention relates to a combustion automatic control method for a circulating fluidized bed boiler, in particular to an online optimization self-learning control method for a circulating fluidized bed boiler capable of ensuring safe operation and high-efficiency combustion of the circulating fluidized bed boiler. Background technique [0002] Circulating fluidized bed boiler (referred to as CFB boiler) has been the focus of domestic and foreign research in the past 20 years because of its wide fuel adaptability, low desulfurization cost, high efficiency and low pollution, and easy comprehensive utilization of ash and slag. In recent years, it has gradually replaced other boiler types and has become the most widely used combustion boiler technology in thermal power generation, industrial heating and other fields, and is developing rapidly towards larger-scale and supercritical CFB boilers. At present, the number and total installed capacity of CFB bo...

AI Technical Summary

Problems solved by technology

[0006] 1. The existing control algorithm focuses on online control optimization, self-contained system, it is difficult to combine the inherent parameter curve of the boiler and the experience data of the operator, resulting in long adjustment time and poor quality of dynamic process, etc.

[0007] 2. The bed temperature is greatly affected by coal quality (such as particle size, calorific value, volatile matter), etc., but the existing optimization algorithm adopts the same control algorithm for different coal quality conditions; Points cannot be self-learning, return to the same working conditions before the operation, still need to repeat the long self-optimization process

[0008] 3. If the initial experience value is inaccurate, it will seriously affect the performance of the entire control system, and the error value cannot be dynamically eliminated during the operation process; the high requirements of the experience parameters result in poor robustness of the entire control system, difficult debugging, and long construction time

Images