基于GA-SVM模型预测减水剂对充填体强度的影响  

Prediction of water-reducers influence on strength of backfill body using GA-SVM model

作　　者：张钦礼[1] 李浩[1] 刘吉祥[1] 刘群武 陈秋松[1] 

机构地区：[1]中南大学资源与安全工程学院,长沙410083

出　　处：《科技导报》2015年第11期44-48,共5页Science & Technology Review

基　　金：国家科技支撑计划项目(2012BAC09B02)

摘　　要：改善充填料浆的工作性能、提高充填体的强度和密实性是矿山充填法的研究方向之一。充填配比实验表明,在充填材料的制备过程中加入适量的高效减水剂可以提高充填体的抗压强度。为得到经济、高效的减水剂添加参数,以萘系、氨基、木钙和脂肪族4种减水剂结合新型充填胶凝材料,应用某矿山的全尾砂进行配比实验,建立GA-SVM预测模型进行优化选择。在优选过程中,以4种减水剂的添加量作为输入因子,以充填体28 d龄期单轴抗压强度作为综合输出因子,根据室内试验,建立训练、验证样本集;建立支持向量机(SVM)回归预测模型,通过遗传算法(GA)对SVM模型参数进行优化选择,当4种减水剂组合添加的质量分数依次为0、0.35%、0.30%、0.60%,抗压强度预测值为4.20 MPa。与实验对比,该模型预测结果的相对误差能控制在1%以下,精确度较高,为减水剂添加参数的优选提供了一种新思路。To improve the working performance of filling slurry, increasing strength and density of backfill body is a researchdirection of mine filling method. Filling ratio experiments showed that adding appropriate amount of water reducers during thepreparation of the filling material could increase the compressive strength of the filling body. In order to obtain economic and efficientwater reducers and parameters,, four kinds of water reducers, i.e., naphthalene, amino, wood, and calcium binding aliphatic were usedfor new filling materials. A match experiment with certain mine backfilling materials was carried out, and a GA-SVM predictionmodel was established to optimize the selection. The four kinds of water reducers were used as the input data and the 28 dayscompressive strengths of filling body were confirmed to be the synthesized output data. Some training and validating samples wereestablished through indoor experiment; a support vector machine （SVM） regression model was established. Then, the model parameterswere optimized through the genetic algorithm （GA）. The results show that the best tailing concentrations of the four kinds of waterreducers were 0, 0.35%, 0.30%, and 0.60%, and that the compressive strength of filling body could be 4.20 MPa. Compared with theexperiment results, the relative error of the prediction result can be controlled within about 1%. This model provides a new method tooptimize sedimentation parameters.

关 键 词：充填料浆 减水剂 单轴抗压强度 支持向量机 遗传算法 

分 类 号：TD853.34[矿业工程—金属矿开采]