论文部分内容阅读
针对遗传算法(GA)存在早熟现象和局部寻优能力较差等缺陷,引入具有很强局部搜索能力的模拟退火算法(SA),组成改进的遗传模拟退火算法(GSA)提高优化问题的能力和求解质量。针对BP神经网络容易陷入局部最小和收敛速度慢等方面的不足,应用改进的遗传模拟退火算法搜索BP神经网络的最优权值和阀值,提高BP神经网络的预测精度,建立了围岩力学参数反分析的GSA-BP神经网络模型。将该模型应用于乌东德水电站右岸地下厂房围岩力学参数的反演分析中,根据监测围岩变形数据反演围岩力学参数,反演所得参数应用到正计算分析中,得出的计算位移与实测值吻合较好,说明该方法的有效性和应用于该工程的可行性。
Aiming at the shortcomings of premature genetic algorithm (GA) and poor local optimization ability, this paper introduces a simulated annealing algorithm (SA) with strong local search ability to form an improved Genetic Simulated Annealing Algorithm (GSA) to improve the optimization problem and Solve the quality. Aiming at the shortcoming that BP neural network is easy to fall into the local minimum and the convergence speed is slow, the improved genetic simulated annealing algorithm is used to search the optimal weights and thresholds of BP neural network to improve the prediction accuracy of BP neural network. Parameter Back Analysis of GSA-BP Neural Network Model. The model was applied to the back analysis of the mechanical parameters of the surrounding rock of the right bank of Wudongde hydropower station. The mechanical parameters of the surrounding rock were retrieved based on the deformation data of the surrounding rock. The inversion parameters were applied to the positive calculation and analysis. The displacement is in good agreement with the measured value, indicating the effectiveness of the method and the feasibility of the method applied to the project.