Distributed performance monitoring and analysis of industrial plants

摘要

Provided is a technique for providing early fault detection using process control data approved by control devices in a process plant. The technique determines an early indicator of a condition within the process plant, such as an error, an anomaly, or a power loss. The leading indicator can be determined using a principal component analysis. A process signal indicating a process variable corresponding to the leading indicator is then retrieved and analyzed. A rolling fast Fourier transform (FFT) may be performed on the process signal to generate time series data to monitor the process plant. If the presence of the leading indicator is detected in the time series data, a warning or other prediction of the condition may be generated. Thus, process errors using variations and anomalies can be identified as early predictions.