基于粗糙集的电站运行优化目标值研究

一直以来电站锅炉面临着降低运行成本的要求,锅炉效率作为反映电站机组经济运行的重要参数之一,它的优化问题受到广泛的关注.由于锅炉系统结构及运行条件复杂,为了降低运行成本,机组负荷变化时往往需要调整很多的运行参数,这对于现场的监控和调整难度较高.借助粗糙集方法来发现数据间隐含的关系,对影响锅炉效率的参数进行约简,从而确定出目标优化参数.利用某600 MW燃煤电站机组历史数据中提取得到的稳定运行工况下的数据,通过模糊聚类方法进行离散化,利用粗糙集方法进行属性约简,并从约简后的结果提取其中重要的决策规则,并采用置信度和支持度作为评价指标,获得目标参数的优化区间,给出可行的烟气含氧量的优化调整方案.

The Study of Data - Driven Optimization Methods of Power Plant Operation

The operation of coal - fired boiler has long had the requirement to reduce its operation cost.The boiler system and loading condition are complicated.In order to improve the efficiency of the utility boilers,many factors should be adjusted when the load changes Rough set is applied to retrieve hidden knowledge from data.The data of stable operation are collected from a 600MW coal - fired power plant and are discretized by the fuzzy clustering algorithm.Rough set is used to reduce the parameters according to the boiler efficiency. The most important decision rules are extracted from the reduction results and evaluated by measures such as supportiveness and confidence.Finally,the solution gives the optimal interval of practical oxygen content.