基于改进ABC的LSSVM氧化还原电位预测研究

Research on prediction of oxidation reduction potential based on improved ABC and LSSVM algorithm

摘要:

针对氧化还原电位对于生物氧化提金预处理过程的控制和优化具有重要作用,提出了一种基于改进的ABC算法优化最小二乘支持向量机的预测方法。该算法是在标准人工蜂群算法的基础上,通过引入欧氏距离,使得在一定邻域内观察蜂采用不同于雇佣蜂的搜索策略。采用改进的ABC算法优化最小二乘支持向量机的参数,取得最优解并赋予最小二乘支持向量机进行预测。以新疆某金矿的生产数据进行仿真研究,结果表明：基于改进的ABC算法优化的最小二乘支持向量机具有较高的预测精度,该方法能使模型取得较好的预测效果。

Abstract:

Oxidation reduction potential plays an important role in the control and optimization of the biological oxidation pretreatment process of gold, a new LSSVM parameter optimization method based on improved antificial bee colony is proposed. Improved ABC algorithm is based on the standard ABC algorithm, and Euclidean distance is introduced to the improved ABC algorithm to identify the valid neighborhood, so that onlooked bees can use search strategy which is different from employed bees. Using improved ABC algorithm to optimize factors, then the optimal solution can be assigned to LSSVM to predict. Taking real data of a gold mine in Xinjiang to simulate, the result show that: LSSVM based on the improved ABC algorithm can improve the prediction accuracy of the model, and the method makes the model have a better prediction.