论文部分内容阅读
针对水电工程地应力测量中测点较少、量测结果离散及计算的非线性问题,提出了基于人工神经网络(ANN)和遗传算法(GA)的应力测量压力反分析模型。该模型利用ANN预测井底压力,并结合井底量测压力值,建立一个适应度函数,然后利用GA在一个大的搜索空间中找到适应度函数的最优解,即所求的应力参数。最终利用糯扎渡水电工程实测水力压力值对该模型进行验证,结果表明模型识别应力值与实测值的相对误差在5%以内,验证了ANN-GA模型的有效性,为水电工程应力确定提供了新方法。
In order to solve the problem of less measurement points, discrepancy of measurement results and calculation of nonlinearity in geo-stress measurement of hydropower projects, an inverse stress analysis model of stress measurement based on artificial neural network (ANN) and genetic algorithm (GA) is proposed. The model uses ANN to predict the bottom hole pressure and builds a fitness function based on the pressure measured at the bottom of the well. Then GA finds the optimal solution of the fitness function in a large search space, namely the stress parameter. Finally, the model is verified by the measured pressure value of Nuozhadu Hydropower Project. The results show that the relative error between the model identification stress and the measured value is less than 5%, which verifies the validity of the ANN-GA model and provides the basis for the hydroelectric project’s stress determination A new method