Aiming at the problem that current grid fault diagnosis algorithms are difficult to take into account both real-time and comprehensive performance, which makes it unable to meet the practical requirements of dispatching, a fast and intelligent diagnosis scheme using multi-source data for fault diagnosis is proposed in this paper. The remote coding data is mapped to the fault diagnosis space through fault coding technology to form a fault coding set, for the defect of using only the switch data, the fault code is corrected and added using the electrical quantity information in the wide area measurement system (WAMS). The labeled fault code is then input into a probabilistic neural network (PNN) for training to construct a PNN classification model. The fault analysis module is constructed to analyze the protection and circuit breaker misoperation and refusals to form a comprehensive intelligent diagnostic model. The analysis of examples shows that the proposed method is correct and feasible, which can meet the practical requirements of scheduling.
