Study On Fault Diagnosis Method Of Microgrid Based On Improved Temporal Cause-effect Network

At present,due to signal transmission and relay protection equipment malfunction and other reasons,short circuit fault diagnosis of micro grid will make fault alarm signals go wrong,which will make the diagnosis process lengthy and the accuracy of diagnosis results decline,which will cause difficulties in troubleshooting and daily maintenance of micro grid.Firstly,a Improved Temporal cause-effect Network(ITCEN)fault diagnosis method for micro grid is proposed,and modeling is carried out according to the installation location of over-current protection equipment of micro grid,so as to solve the problem of topology change caused by state transition of micro grid.Then through fault warning signal 0-1 value and matrix operations,according to the main circuit breaker judgment factor,protection and circuit breaker status factor and regional decision factor,quickly determine fault zone,ruled out without the influence of the main circuit breaker and the normal protection redundancy in the process of operation,reduce unnecessary and temporal consistency analysis by dimension reduction to shorten the logical reasoning and accurate diagnosis is given.Then,an adaptive PSO-BP(Particle Swarm Optimization-Error Back Propagation)neural network algorithm is presented to train the neural network by automatically adjusting parameters,improve the speed of operation and convergence,and work together with ITCEN to establish a diagnosis model suitable for the two operation states of micro grid grid-connection and island,and correct the fault result through the accident set.Finally,an example is given to verify the correctness of the improved cause-and-effect timing network algorithm.Then,a microgrid in liaoning province is taken as an example and PSCAD and MATLAB software are combined to carry out fault diagnosis of the improved cause-and-effect timing network and the adaptive pso-bp neural network for the microgrid.