
Energy consumption prediction of cement production based on chaotic neural network-Markov chain
Author(s) -
Run Yin,
Chunxia Dou,
Dong Yue
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/675/1/012108
Subject(s) - energy consumption , chaotic , residual , artificial neural network , markov chain , production (economics) , computer science , energy (signal processing) , consumption (sociology) , markov model , control theory (sociology) , mathematical optimization , statistics , artificial intelligence , algorithm , mathematics , engineering , machine learning , sociology , social science , control (management) , electrical engineering , economics , macroeconomics
In order to improve the accuracy of energy consumption modeling and prediction in the cement production process, a cement production energy consumption prediction model based on chaotic neural network is proposed: 1. In the energy consumption modeling stage, chaotic neural network is used to reconstruct the phase of chaotic time series. Space, the chaotic neural network can still make high-precision predictions of the system even when the network input is incomplete or mutated, and the determination coefficient value is 0.019 higher than that of the RBF neural network. 2. In the energy consumption prediction stage, the introduction of Marko The residual error correction method is to correct the current forecast value based on the residual error between the historical predicted energy consumption value and the actual energy consumption value. The result shows that the relative residual error of the predicted value corrected by the Markov correction method decreases from -0.6% to -0.25 %, the predicted value of energy consumption is closer to the actual value. Based on the above description of the two stages of energy consumption modeling, the proposed cement production energy consumption prediction model has better prediction effects and higher prediction accuracy than traditional mechanism modeling in cement production energy consumption prediction.