Cement production quality prediction method integrating CSA and H-ELM

Abstract

The invention provides a cement production quality prediction method integrating CSA and H-ELM. The method comprises the following steps: S1, acquiring data related to cement production quality prediction; S2, preliminarily screening the data acquired in the step S1 by using a grey correlation analysis method; S3, initializing parameters of a hierarchical extreme learning machine, optimizing the number of hidden layer neurons of the hierarchical extreme learning machine by using a graffiti algorithm when the hierarchical extreme learning machine is trained by using the data screened in the step S2, obtaining the optimal number of hidden layer neurons of the hierarchical extreme learning machine, and obtaining a prediction model; and S4, predicting the cement production quality by using the prediction model. According to the method, the number of neurons in an H-ELM hidden layer is optimized through the CSA algorithm, the tedious operation of manual parameter adjustment is abandoned, the practicability and prediction precision of the model are improved, and meanwhile the optimized model can intelligently predict the production quality of products in the rotary kiln of the cement sintering system in real time and with high precision.

Description

technical field [0001] The invention relates to the technical field of cement production, and more particularly, to a cement production quality prediction method integrating CSA and H-ELM. Background technique [0002] Cement and plastic, metal and wood are commonly known as the four major materials of the national economy. The cement industry is an important pillar industry for national economic and social development, and an important supporting force for the continuous growth of my country's economy. Its degree of automation directly demonstrates the overall basic industrial strength of a country. As the country vigorously promotes urbanization and accelerates the progress of industrial development, as of 2016, the average annual output of cement in my country is about 2.5 billion tons, accounting for more than 50% of global cement output. However, my country's cement industry has a relatively low degree of automation. The energy consumption per unit of cement clinker is...

AI Technical Summary

Problems solved by technology

[0007] The invention provides a cement production quality prediction method that integrates CSA and H-ELM, uses the crow algorithm to optimize the prediction model, achieves the purpose of improving the prediction accuracy, and solves the problem of real-time prediction of cement quality in the rotary kiln of the cement firing system

Images