基于随机森林回归的火电机组供电煤耗遗传优化模型

火电机组供电煤耗与电厂产能息息相关,为提高火电产能及其收益,有必要对火电机组供电煤耗进行优化,应用随机森林回归算法可从历史数据中挖掘机组供电煤耗与主蒸汽压力、主蒸汽温度等相关参数的回归模型,并以机组供电煤耗最低为目标条件,提出使用遗传算法优化方法搜索实时工况下最优运行参数控制策略。通过对某机组进行测试,结果表明,基于随机森林回归算法的火电机组供电煤耗遗传优化模型可有效减少机组供电煤耗,为发电企业提供优化建议,实现对发电成本的优化控制。

Genetic optimization model of power supply coal consumption for thermal power unit based on random forest

Since the coal consumption of thermal power plants is related to their capacity,it is necessary to optimize coal consumption in order to increase production capacity and revenue.The random forest regression algorithm can mine the regression model of coal consumption,main steam pressure,main steam temperature and other related parameters from historical data,and then propose the optimal operation strategy with the goal of the lowest coal consumption.Through the test of one unit,the results show that the genetic optimization model of coal consumption for power supply of thermal power units based on random forest regression algorithm can reduce coal consumption for power supply.