摘要
针对Ricaaras曲线预测模型参数估计较难的问题,研究一种改进型变步长果蝇优化算法,对基本果蝇优化算法进行步长改进和流程优化,并将Ricaaras模型的参数估计转化成一个四维约束函数优化问题。遵循“最小线性二乘”的准则,实现Ricaaras模型的参数估计,将该模型应用于某市地铁沉降观测点的沉降趋势预测。为进一步验证算法准确性和有效性,将改进型变步长果蝇优化算法(改进型VS-FOA)与粒子群算法(PSO)、遗传算法(GA)从预测精度和收敛速度两个维度进行对比分析,结果表明,用改进型VS-FOA算法估计的Ricaaras模型预测精度更高。
In view of the difficulty of parameter estimation in the Richards curve prediction model,an improved fruit fly optimization algorithm was studied to improve the step size and to optimize the process,and the parameter estimation problem of Richards model was transformed into a four-dimensional constraint function optimization problem.Based on the principle of u linear least square method,parameter estimation of the Richards model was realized,and the model was applied to a city Metro engineering prediction.To further verify the accuracy and effectiveness of the algorithm,the improved VS-FOA algorithm was compared with the particle swarm optimization(PSO)algorithm and the genetic algorithm(GA)algorithm in two dimensions:prediction accuracy and convergence speed.The results show that the Richards model estimated by the improved VS-FOA algorithm has higher prediction accuracy.
作者
罗利娟
Luo Lijuan(School of Engineering and Technology,Xi’an Fanyi University,Xi’an Shaanxi 710105)
出处
《西安翻译学院论坛》
2020年第3期59-63,共5页
Forum of Xi'an Fanyi University
基金
西安翻译学院校级科研项目(19A03)
区域经济与产业发展研究团队(XFU17KYTDC02)。
