k-means和QGA优化RBF神经网络模型在导水裂缝带高度预测方面的应用

导水裂缝带高度选取精度的高低直接影响到水体下采煤系统的安全性。为准确预计导水裂缝带高度,本文构建了RBF神经网络基础模型,采用QGA量子遗传算法和K-means算法对基础模型进行优化,获得了K-means和QGA优化的RBF神经网络导水裂缝带高度预计模型。模型经过训练学习和检验,发现模型预计精度满足工程精度需求,且与PSO-RBF神经网络相比,精度更高、收敛速度更快。

Application of RBF neural network model optimized by K-means and QGA in Water conducted fractured zone height prediction

The selection accuracy of the water conducted fractured zone height directly affects the safety of mining under the water body. To accurately predict the height of water conducted fractured zone, this paper first constructs the basic model of RBF neural network, then optimizes the basic model with QGA quantum genetic algorithm and K-means algorithm, and obtains K-means and QGA optimized RBF neural network height prediction model of water conducted fracture zone. The model is trained and tested. The results show that the accuracy of the model meets the requirements of engineering accuracy, and compared with the PSO-RBF neural network, the accuracy is higher and the convergence speed is faster.