Byproduct gas generation amount prediction method based on combined weight

Abstract

The invention discloses a byproduct gas generation amount prediction method based on combined weight. The method comprises the following steps: acquiring coke oven gas generation amount, coal charging amount and volatile component data of mixed coal in a certain time interval, acquiring theoretical coke oven gas generation amount and coke oven gas yield according to a physical calculation formula, and taking coke oven gas generation quantity, coal loading quantity, volatile components of mixed coal and coke oven gas yield data as neural network input quantity; then preprocessing the data, training a left-end neural network prediction model and a right-end neural network prediction model respectively by the processed data, selecting an optimal weight ratio according to a training result, and performing linear weighting to obtain a combined weight prediction model; and finally, predicting the coke oven gas generation amount in specified time by using the combined weight prediction model. The method better integrates the advantages of prediction models at the left end and the right end, provides a reliable basis for coke oven gas generation amount prediction in combination with historical storage resources, and can more accurately and effectively predict the coke oven gas generation amount compared with traditional methods.

Description

Technical field [0001] The present invention relates to the field of artificial intelligence, and Background technique [0002] The by-product gas is an important secondary energy of steel enterprises. By optimizing the dynamic balance of gas systems, improve gas utilization efficiency, realizing the "zero dispersion" of gas, is the highest goal of gas optimization schedules for each steel enterprise. The basis for achieving this goal is to accurately predict the gas generation and consumption of the enterprise. The coke oven gas is a by-product of the coke process. It is a high-temperature value gas. my country is the largest coke production. In 2020, my country's coke production is 47116.1 million tons. According to this calculation, my country's coke oven gas production is very high. Therefore, it is necessary to provide technical support for the dynamic balance and optimization scheduling of coke oven gas gas systems. [0003] Complicated multi-change gas system conditions of...

AI Technical Summary

Problems solved by technology

However, there are relatively few studies on the prediction of coke oven gas in iron and steel enterprises in the field of deep learning. Therefore, it is necessary to propose an efficient prediction method for coke oven gas generation, which can better solve the problem of coke oven gas prediction.

Images