摘要
为准确预测煤层底板突水量,提出了一种基于IPSO-SVR(改进的粒子群算法以优化支持向量回归机算法)的煤层底板突水量的预测模型。针对矿井底板突水这种非线性、小样本问题,通过改变粒子群算法的惯性权重因子定义以及引入混沌映射思想的方式,避免算法陷入局部最优值,强化全局搜索。结合王家岭等煤矿突水实例,将水压、含水层、隔水层厚度、底板破坏深度以及断层落差作为影响煤层底板突水量的特征因素,将该预测模型算法与PSO-SVR预测模型算法进行比较。仿真结果表明:该预测模型算法的预测值更接近实际值,具有一定实际应用价值。
In order to accurately predict the amount of water inrush from the coal floor,a prediction model of coal floor water inrush based on IPSO-SVR(the improved particle swarm algorithm is used to optimize the support vector regression machine algorithm)is proposed.Aiming at the non-linear and small sample problem of water inrush from coal floor,the definition of the inertia weight factor of the particle swarm algorithm is changed and the method of chaotic mapping is introduced to avoid the algorithm from falling into the local optimum and strengthen the global search.Combined with examples of coal mines such as Wangjialing,the water pressure,aquifer thickness,water barrier thickness,depth of floor failure,and fault drop are taken as the characteristic factors affecting the amount of water inrush from the coal floor.Comparing the prediction model algorithm with the PSO-SVR prediction model algorithm,the simulation results show that the predicted value of the prediction model algorithm is closer to the actual value and has certain practical application value.
作者
王鹏
朱希安
王占刚
刘德民
WANG Peng;ZHU Xi′an;WANG Zhangang;LIU Demin(School of Information and Communication Engineering,Beijing Information Science&Technology University,Beijing 100101,China;School of Safety Engineering,North China Institute of Science and Technology,Beijing 101601,China)
出处
《北京信息科技大学学报（自然科学版）》
2021年第1期40-44,共5页
Journal of Beijing Information Science and Technology University
基金
国家重点研发计划项目(2017YFC0804108)
北京市教委科研计划项目(KM201811232010)
北京市科技创新服务能力建设-基本科研业务费(市级)(科研类)(PXM2019_014224_000026)。
关键词
矿井突水
IPSO-SVR模型
煤层底板突水量
参数优化
mine water inrush
IPSO-SVR model
water inrush quantity from coal floor
parameter optimization
