The exploitation of performance monitoring and fault early warning system based on gas turbine operating data is conducive to mastering the overall performance of gas turbine

early warning equipment faults before the measurement point data reaches the alarm threshold

and predicting the trend of efficiency degradation.The performance monitoring model was developed with zero-code and configuration modeling technology

the early warning diagnosis models for compressor blade fouling and turbine exhaust temperature anomaly were built with artificial intelligence and deep learning algorithms

and these models have achieved engineering demonstration application in a GE 9E heavy-duty gas turbine in Guangdong Province. The results show that

this performance monitoring model can monitor the key performance parameters of the gas turbine in real time

and the calculation error is less than 1%

the early warning diagnosis models can monitor the deviation between the actual operating value and the normal operating value of the compressor outlet pressure and turbine exhaust temperature in real time

capture the changing trend of the deviation

and realize early warning of faults. The gas turbine performance monitoring and fault early warning system developed in this paper has a good engineering application effect and can be further popularized and applied.