Research on Prediction Model of Surrounding Rock Loose Circle Based on BP Neural Network and Genetic Algorithm
Aiming at the problems of heavy workload and high test cost in the test of loose circle of sur-rounding rock in mine roadway,combined with the actual project of Baixiangshan Iron Mine,it is proposed to use BP neural network and genetic algorithm to construct the prediction model of loose circle to realize the prediction of loose circle range.SPSS software was used to statistically analyze the impact indicators of the loose circle,and the key influencing factors were determined by frequency analysis.On this basis,the indi-cators for the prediction of the loose circle were selected.Based on BP neural network and genetic algorithm,a GA-BP loose circle thickness prediction model is constructed.Based on the collected loose circle data,training samples and test samples are formed.The training samples are used to train the prediction model,and the test samples are used to test the accuracy of the prediction model.The results show that the fitness of genetic algorithm is very close to the best fitness after 30 iterations.Among the five sets of test data,the maxi-mum error is 15.2 cm and the minimum error is 1.13 cm,which are not more than 20 cm.It shows that the prediction model has high prediction accuracy and certain reliability.
underground roadwayprediction of loose circleBP neural networkgenetic algorithm
万方数据
维普
