Microwave drying prediction method through BP (back-propagation) neural network based on incremental improvement

Abstract

The invention relates to a microwave drying prediction method through a BP (back-propagation) neural network based on incremental improvement, and the method is used for predicting relative dehydration rate and temperature in the production process of microwave drying of selenium-enriched slag through the neural network, wherein the three-layer BP neural network is selected as a prototype, the incremental learning and the L-M (Levenberg-Marquard) algorithm are adopted for improving the network, and a BP neural network model based on the incremental improvement is established for predicting the relative dehydration rate and the temperature during the microwave drying of the selenium-enriched slag. Through the simulation testing of a computer, the relative dehydration rate and temperature of the selenium-enriched slag in the microwave drying test process can be accurately and quickly predicted, the number of trial tests can be reduced, and the theoretical basis is provided for follow-up production.

Description

technical field [0001] The invention relates to a microwave drying prediction method based on an incrementally improved BP neural network, a method for predicting the relative dehydration rate and temperature in the production process of microwave-dried selenium-enriched slag by using the neural network. It belongs to the technical field of metallurgical engineering computer neural network control. Background technique [0002] In the production process of microwave drying of selenium-rich slag, the factors that affect the microwave drying effect include microwave input power, microwave action time, material rotation speed, etc., which have different influences in the drying process, resulting in a long test cycle and a large amount of testing in the microwave drying process. And the parameters are not easy to optimize. For this reason, the BP neural network with nonlinear mapping capability is selected to establish a simulation model for the microwave drying process and pr...

AI Technical Summary

Problems solved by technology

However, the known BP neural network algorithm is based on the gradient descent method, and the network weight is corrected by calculating the gradient of the objective function to the network weight and the threshold, and there are problems of slow convergence and local minimum in the training process; and for complex problems , during the training process, it will fall into a local minimum point, so that it cannot converge. Compared with the known BP neural network algorithm, the Levenberg-Marquardt (L-M) algorithm is used to improve the BP neural network, which improves the convergence speed of the neural network. In the process of training the neural network, the required training samples cannot be provided at one time, and when the training sample size is large, the limitation of the system memory makes the training of all samples infeasible. A BP neural network based on incremental learning is proposed. A nonlinear system prediction model of incremental BP neural network was established to predict the production results of microwave-dried selenium-enriched slag

Images