Artificial fish swarm algorithm in industrial process alarm threshold optimization

Artificial fish-swarm algorithm is a novel method to search global optimum, which is typical application of behaviorism in artificial intelligence. Compared with traditional method it is more portability and stability. This paper, based on the loss function to increase the enterprise benefit, brings forward an optimized criterion of setting threshold to relieve the security officer work in the chemical industry. In addition, we add lose function into threshold optimization to explain the benefit of a program is suit for the actual environment. The simulation results show that the proposed algorithm has greatly improved the system performance.