Abstract: For improving boiler combustion performance of coal-fired power plant, a new optimized method based on constrained fuzzy association rule is proposed to determine boiler operation parameters. The associated relationship can be discovered between primary operation parameters and performance indexes by mining an ocean of historical operation data with different working conditions. According to the association relationship, the desired values of boiler parameters which are crucial to high-efficiency and low-emission operation can be acquired. Taking the data mining result of oxygen content in flue gas as an example, its optimization values in different working conditions are demonstrated. Multi-objective boiler combustion optimization based on constrained fuzzy association rule doesn’t only provide the theoretical basis for regulating boiler parameters at present, but also establishes the foundation on set points of important parameters for closed-loop control in the future.

Key words: Coal-fired power plant, combustion optimization, data mining, oxygen content in flue gas; optimal target values

摘要： 为了进一步提升燃煤电站锅炉燃烧性能，提出了基于约束模糊关联规则的锅炉运行参数目标值优化方法。通过挖掘锅炉不同工况下海量历史运行数据，发现锅炉稳态运行条件下主要运行参数与性能指标之间的关联关系，并据此得到锅炉高效低污染运行时对应的参数运行目标值，进而指导锅炉燃烧优化。数据挖掘结果以烟气含氧量为例，给出了锅炉运行不同工况下的优化目标值。基于约束模糊关联规则的多目标锅炉燃烧优化，不仅为当下锅炉运行参数调整提供了理论依据，也为日后进一步的闭环控制铺垫了参数设定值基础。

关键词: 燃煤电站, 燃烧优化, 数据挖掘, 烟气含氧量, 目标值优化