基于算法的煤质发热量预测  

Calorific value prediction for coals based on a certain algorithm

作　　者：刘林 王朋 翟永杰[3] 周杰联 

机构地区：[1]广东电网有限责任公司佛山供电局,广东佛山528000 [2]广东电网有限责任公司电力科学研究院,广东广州510080 [3]华北电力大学自动化系,河北保定071003

出　　处：《热力发电》2015年第2期47-51,57,共6页Thermal Power Generation

基　　金：中央高校基本科研业务费专项资金资助项目(12MS118)

摘　　要：为了提高煤质发热量的预测精度,提出了基于参数自适应粒子群(PSO)和最小二乘支持向量机(LSSVM)算法的煤质发热量预测方法:建立LSSVM煤质发热量预测模型,通过参数自适应PSO算法优化LSSVM模型参数。使用测试数据对煤质发热量的预测效果进行了仿真比较。结果表明,基于参数自适应PSO-LSSVM发热量预测模型能够准确预测发热量的变化,且较参数自适应PSO-BP神经网络、基本PSO-LSSVM以及GA-LSSVM模型运算速度快,预测精度高,利于工程应用。In order to improve the prediction precision of calorific value of coals,a forecasting method based on parameters adaptive particle swarm optimization（PSO）and least squares support vector machine（LSSVM）algorithm was presented.Through optimizing the parameters of LSSVM algorithm by parameters adaptive PSO algorithm,aprediction model for coals＇ calorific value was established.Moreover,the predicted calorific values were compared with the test data.The simulation results show that,the method based on the parameter adaptive PSO-LSSVM prediction model can accurately predict the change of calorific value,which has faster calculation speed,higher precision and is suitable for engineering application,compared with the parameter self-adaptive PSO-BP model,the PSO-LSSVM model and the genetic algorithm（GA）-LSSVM model.

关 键 词：煤质 发热量 预测 参数自适应 粒子群算法 最小二乘支持向量机算法 

分 类 号：TK16[动力工程及工程热物理—热能工程] TK212