一种新型煤矿底板破坏深度预测模型

为有效预防煤矿底板突水风险,在传统的粒子群优化算法中增加自适应权重,结合遗传算法的交叉、变异步骤改进传统的粒子群优化算法,并用其优化SVM模型,建立改进的GA-PSO-SVM煤矿底板破坏深度预测模型,选取采深、煤层倾角、采高、工作面长度、煤层底板承压水水压和煤层底板损伤变量作为影响底板破坏深度的主控因素,通过15组煤炭生产单位采集底板破坏带深度相关数据,测试改进的GA-PSO-SVM模型的性能,并与FOA-SVM模型、BP模型的预测结果进行对比,研究表明:改进的GA-PSO-SVM模型预测结果与实测结果的误差范围为0.36%~5.22%,FOA-SVM模型预测结果的误差范围为1.60%~12.49%,BP模型预测结果的误差范围为1.01%~20%,改进的GA-PSO-SVM模型预测结果的误差范围更小,更适合煤矿现场的应用要求。

A New Model for Predicting the Deep Destruction of Coal Mine Floor

In order to effectively prevent the risk of water inrush from coal mine floor,the traditional particle swarm optimization(PSO)algorithm is improved by adding adaptive weights to the traditional particle swarm optimization(PSO)algorithm,combining the crossover and mutation steps of genetic algorithm,and using it to optimize the SVM model,the improved GA-PSO-SVM prediction model of coal mine floor failure depth is established,and the mining depth,coal seam dip angle,mining height,working face length and coal seam floor pressure are selected.Water pressure and damage variable of coal seam floor are the main factors affecting the floor failure depth.The performance of the improved GA-PSO-SVM model is tested by collecting relevant data of floor failure zone depth from 15 groups of coal production units.The results are compared with those of FOA-SVM model and BP model.The results show that the error range between the predicted results of the improved GA-PSO-SVM model and the measured results is 0.36.The error range of FOA-SVM model is 1.60%~12.49%,and that of BP model is 1.01%~20%.The improved GA-PSO-SVM model has a smaller error range and is more suitable for the application of coal mine.