基于最小二乘支持向量机和粒子群法的水煤浆性能优化 -

采用最小二乘支持向量机(LSSVM)进行水煤浆(CWM)浓度建模,并利用粒子群法(PSO)对运行工况寻优,获得优化水煤浆浓度的调整方式。以优化调整方式的相应参数作为当前负荷下的基准值,能很好地解决了制浆变工况下运行参数基准的确定问题。以该基准为中心分析了给水流量和分散剂流量对水煤浆浓度的敏感性影响。通过现场试验证明,按此基准值运行,可以提高水煤浆性能。 The LSSVM(Least Square Support Vector Machines)was proposed to construct optimization model for CWM(coal water mixture)concentration and PSO(particle swarm optimization)is used to perform a search for determining the optimum solutions,from which the parameters of optimum adjustment mode could be the reference value of current load and could make certain reference value of variable condition operation.A sensitivity analysis of the effect of feed water flow and dispersant flow on CWM concentration is carried based on reference value.The performance of CWM could be heightened based on the reference value by scene test