Implementation and Improvement of Simulated Annealing Algorithm in Neural Net

Neural net is one net structure used to simulate the structure of human brain, which has a number of realization methods such as the simulated annealing algorithm, the BP algorithm and genetic algorithm and so on. The simulated annealing algorithm is a kind of calculation precision of random search algorithm, which can be applied to many of little premise information questions and converged to the optimal value gradually. This paper introduces the simulated annealing algorithm, discusses the principle of simulated annealing algorithm, improves the algorithm and shows the results of the calculation experiment.