基于狼群算法的概率积分法模型参数反演方法研究

概率积分法模型属于典型的多元复杂非线性函数,且部分参数存在相关性,使得概率积分法模型参数反演一直是开采沉陷数据处理的热点问题和难点问题。WPA(狼群算法)作为一种新群体智能算法,已在多维背包求解问题、水电站水库优化调度以及旅行商等复杂非线性最优化问题中得到成功应用,但在开采沉陷数据处理领域尚未见文献报道。鉴于此,本文首次将WPA引入概率积分法模型参数反演中,构建了基于WPA的概率积分法模型参数反演方法。研究结果表明:WPA反演概率积分法模型参数具有较高的准确度、较好的精度、良好的稳健性,且满足工程应用标准。本文研究成果对概率积分法模型参数精准反演具有重要参考价值。

Research on Parameter Estimation of Probability Integral Model Based on Wolves Algorithm

The model of probabilistic integral method is a typical multivariate complex nonlinear function, and some parameters are correlated, which makes the parameter inversion of probabilistic integral method a hot and difficult problem in mining subsidence data processing. WPA(wolf pack algorithm), as a new swarm intelligence algorithm, has been successfully applied in multi-dimensional knap-solving problems, optimal operation of hydroelectric power stations and reservoirs, as well as complex nonlinear optimization problems such as travel agents. In view of this, this paper introduces WPA into parameter inversion of probabilistic integration method for the first time, and constructs a model parameter inversion method based on WPA.The research results show that the model parameters of WPA inversion probability integral method have higher accuracy, better accuracy, good robustness, and meet engineering application standards.The research results of this paper have important reference value for accurate parameter inversion of probabilistic integral method.