System and method for predicting drum water level of circulating fluidized bed domestic garbage incineration boiler

Abstract

The invention discloses a system and a method for predicting the drum water level of a circulating fluidized bed (CFB) domestic garbage incineration boiler. A moderately complex drum water level dynamic characteristic mechanism model based on a differential equation form is established first from the mechanism of drum water level change characteristic based on reasonable simplifying assumption according to a mass conservation equation, an energy conservation equation and some basic equations in combination with the unique heat absorption distribution law of the wall-cooled wall of the CFB domestic garbage incineration boiler. Then, tacit knowledge in running historical data is dug using ANFIS (Adaptive Neuro-Fuzzy Inference System) modeling to compensate errors caused by model simplifying assumption, model reduction or linear processing, incomplete cognition on the mechanism of a thing change process, different object characteristics and internal and external disturbance in the mechanism modeling process of the drum water level. The advantages of mechanism modeling and ANFIS modeling are sufficiently exerted, and the prediction precision of the drum water level is prompted.

Description

technical field [0001] The invention relates to the field of energy engineering, in particular to a system and method for predicting the drum water level of a circulating fluidized bed domestic waste incineration boiler. Background technique [0002] Since waste incineration can well realize volume reduction, volume reduction, harmlessness and resource utilization of waste treatment technology, in the past ten years, under the guidance of relevant national industrial policies, the domestic waste incineration industry has achieved vigorous development. Circulating Fluidized Bed (CFB) domestic waste incineration technology, as one of the main domestic waste incineration treatment technologies, has been promoted and applied in many cities in China. By the end of 2014, 70 domestic waste incineration boilers had been built. Yu Taiwan, with a daily waste processing capacity of 64,000 tons, has made an important contribution to my country's waste incineration treatment industry. Dr...

AI Technical Summary

Problems solved by technology

[0005] 2) The model obtained by the mechanism modeling method is relatively complex, and it is often a high-order model or nonlinear. For better solution and practical application, order reduction or linearization is usually performed, thereby reducing the accuracy of the model;

[0006] 3) The formation process of the drum water level involves complex two-phase flow and bubble generation-expansion / contraction-burst process. People's research and understanding of these processes is not very complete, so it will not be fully understood in the modeling process. Express the characteristics of these processes, and even make wrong mathematical descriptions, which will bring errors to the model;

In addition, the characteristics of the production process also aggravate the inhomogeneity of the heat absorption of the water wall in the CFB waste incineration boiler

Therefore, these drum water level change models based on the uniform heating of the water wall are not necessarily applicable to CFB domestic waste incineration boilers

[0008] 5) In the actual production process, the modeling objects and physical processes are always faced with various known or unknowable disturbances, and these disturbances are usually ignored in the process of mechanism modeling, resulting in an overly idealized model

However, the pure ANFIS modeling method has shortcomings such as poor model extrapolation, weak model interpretation, dependence on training samples, and limited working conditions.

Images