论文部分内容阅读
由于主成分分析(PCA)方法是一种线性算法,基于PCA的故障检测方法若直接运用于非线性系统的传感器故障检测和重构,会导致明显的故障误报和数据重构错误。为了使基于PCA的传感器故障检测和重构方法适用于非线性较严重的热工过程,对该方法进行了有效的改进。应用不同负荷下的历史数据,分别建立机组不同负荷下的局部PCA模型,再根据机组当前实际运行负荷选择相应的PCA模型进行传感器故障检测和重构,并结合相邻负荷PCA模型的计算结果进行数据融合,从而进一步提高了故障检测的准确性和重构精度。理论分析和现场实际应用表明,该算法能够对非线性较为严重的电厂热工过程进行精确的传感器故障检测和重构。
Since PCA method is a linear algorithm, the PCA-based fault detection method, if applied directly to sensor fault detection and reconstruction in nonlinear systems, will lead to obvious fault false alarms and data reconstruction errors. In order to make PCA-based sensor fault detection and reconstruction methods suitable for more serious non-linear thermal processes, this method is effectively improved. Applying the historical data under different loads, the local PCA model under different loads is established, and then the corresponding PCA model is selected to detect and reconstruct the sensor fault according to the actual operation load of the unit. The PCA model with adjacent load is used to calculate Data fusion, thereby further improving the accuracy of fault detection and reconstruction accuracy. The theoretical analysis and field practical application show that the algorithm can detect and reconstruct the accurate sensor fault in the thermal process of the plant with serious non-linearity.