Abstract:To solve the problem that the fault nodes can be diagnosed quickly and effectively in large-scale multiprocessor systems,fireworks algorithm is applied to solving the systemlevel fault diagnosis problem for the first time as an effective diagnosis algorithm.Fireworks algorithm has a good global search ability and local search ability of self-regulation mechanism.Taking the most of this advantage of fireworks algorithm,the constraint equations are designed according to the characteristics of the PMC model,and a new fitness function is proposed.At the same time,the mutation operator and the selection strategy are optimized.The simulation result shows that the algorithm has good stability and convergence.Finally,the validity of the algorithm is proved.