摘要
劣质煤水分与灰分含量大,可磨性差,相同负荷下所需给煤量增加,针对近年由于火电机组掺烧劣质煤易造成磨煤机堵塞,对磨煤机安全稳定运行十分不利的问题,提出了一种多尺度主元分析方法。现场未堵磨时的数据经小波多尺度分解后在多个尺度上建立主元模型,再对实时信号分解从而对多个模型同时进行检测,判断磨煤机是否发生堵磨,并给出堵磨的主要原因。通过对某火电厂磨煤机堵塞状况的检测判断,结果表明:该方法是有效可行的,在经小波多尺度分解后对磨煤机堵磨判别更为全面,绘制实时数据贡献图可得出引起磨煤机堵磨的主要信号。
Low-quality coal has large moisture and ash content, poor grindability, and the required amount of coal increased under the same load. In view of the blockage of the coal mill due to mix low- quality coal in the thermal power units which is very harmful to the safe and stable operation of the coal mill, a multi-scale principal component analysis method was proposed. The principal component model is established on several scales by wavelet multi-scale decomposition when the data are not blocked in the field, and then the real-time signals are decomposed so that multiple models can be detected simultaneously. According to the data we can judge whether the coal mill is blocked or not, and analyze the main signals that affect the blockage of coal mill when the mill blockage occurs. Judging by the blockage condition of a coal mill in a thermal power plant, the results show that the method is effective, after wavelet multi-scale decomposition, it is more comprehensive method to judge the blockage of coal mill, drawing contribution graph of real time data can obtain the main signals which cause the blockage of coal mill.
出处
《电力科学与工程》
2018年第1期41-45,共5页
Electric Power Science and Engineering
基金
国家重点研发计划资助项目(2017YFB0902102)
关键词
主元分析
小波变换
磨煤机堵塞
principal component analysis
wavelet transform
blockage of coal mill
