Abstract: Aiming at the problems that the particle size can’t be measured online and the offline analysis by lab sample existing in large-time delay, by combining the characteristics of the one stage grinding circuit, the soft sensor model of particle size was proposed by the combination of improved chaotic self-adaptive particle swarm optimization and BP neural network algorithm. Taking advantages of chaotic theory ergodicity and PSO global optimal searching ability, the algorithm above couldadjust the weights of BP network adaptively and avoid falling into the local optimum. As a result of MATLAB simulation, the measurement accuracy of the improved CSAPSO-BP NN is higher than the PSO-BP NN and CPSO-BP NN, and it also has better ability of convergence and optimization performance. To sum up, the proposed soft sensor approach is efficient.

Key words: chaos, particle swarm optimization, neural network, particle size, soft-sensor