基于大数据驱动案例匹配的电站锅炉燃烧优化

Combustion optimization in power station based on big data-driven case-matching

电站锅炉燃烧是一个复杂的多相反应过程。针对基于计算智能的燃烧优化算法复杂度高、难以进行实时在线优化的问题,同时考虑提高锅炉效率与降低污染物排放的双重要求,提出了基于数据驱动案例匹配的电站锅炉燃烧优化方法。依托电厂的SIS大数据库,利用数据挖掘技术进行深层次的分析,离线建立燃烧案例库。在线优化时,基于机组DCS中的实时运行数据进行当前工况计算,基于案例库进行案例工况匹配,进而得到适于当前工况运行的最优参数值。对某机组的优化计算结果表明,基于数据驱动案例匹配的锅炉燃烧优化系统算法复杂度低、稳定性高,是一种相对简单、有效而实用的电站锅炉燃烧优化方法。

Boiler combustion is a complex multi-phase reaction process. Computational intelligence-based combustion optimization algorithm is of high complexity,and it is difficult to do real-time on-line optimization. A boiler combustion optimization method,based on data-driven case-matching,is presented by simultaneously considering dual requirements of improving boiler efficiency and reducing emissions. Relying on historical database SIS,data mining technology is used for in-depth analysis,and the combustion case library is built off-line. Real time operation data from DCS is used to calculate the current operating condition. The best match case in the case library is found to get the optimal parameters value for the current running condition. Results of several optimization examples show that the calculation of the proposed combustion optimization system is much less,and it is with high stability. Thus it is more suitable for online use to improve boiler efficiency and reduce emissions.