【作者】 曾德良； 刘继伟； 柳玉；

【Author】 ZENG De-liang,LIU Ji-wei,LIU Yu (1.North China Electric Power University,Key Laboratory of Measurement & Control New Technology and System for Industrial Process,Beijing 102206)

【机构】 华北电力大学工业过程测控新技术与系统北京市重点实验室；

【摘要】 针对设备状态难以准确获取的问题,将小波变换用于状态检测中,提出了多尺度建模、多尺度分析的方法。在建立多尺度标准预测模型的基础上,提取实际运行数据与模型残差的低频趋势分量,构建了磨煤机磨辊磨损指数,并将信息粒化的思想引入状态检测中,大幅降低了计算复杂度,节省了计算资源。利用最小二乘支持向量机对磨损指数进行预测,为实现设备的预知维修打下基础。通过对某机组运行数据的分析证明,该方法可以有效提取设备状态信息。更多还原

【Abstract】 In order to overcome the difficulty of obtaining equipment’s status,wavelet transform was used in condition detection. The method of multi-scale modeling and analysis was proposed.Based on multi-scale standard prediction model,low-frequency trend of residuals between model and actual data was extracted to construct roller abrasion index.Besides,information granulation was used to reduce computational complexity and save computing resources,and then ls-svm was used to predict the abrasion index which was the foundation for condition maintenance.The result shows that,the equipment’s status can be obtained.更多还原