An Optimized BP Neuron Network Algorithm Based on GASA Hybrid Learning Strategy

To overcome the drawbacks of falling into local minimum and slow convergence during Back- Propgation(BP)neural network learning,a neuron network optimization algorithm based on GASA hy- brid strategy,which combines the parallel searching structure of genetic algorithm with the probabilistic jumping property of simulated annealing,is proposed.When the optimized BP neural network was applied to crop pest prediction,the result indicated that it greatly improved neural network convergence perform- ance and speed and,to a certain degree,simplified the complexity of the algorithm.