厌氧发酵工艺的BP神经网络建模与粒子群算法优化

BP Neural Network Modeling and Particle Swarm Algorithm Optimization of Anaerobic Fermentation Process

为了获得厌氧发酵装置的最佳工艺条件参数,以发酵工程产气的量及质量参数为响应值,建立其BP神经网络模型;同时利用粒子群算法对网络模型进行全局寻优,最终获得最佳厌氧发酵工艺参数。本研究表明,采用BP神经网络模拟结合粒子群算法的优化方法,对厌氧发酵工艺具有较好的优化效果,为厌氧发酵过程控制提供理论依据。

In order to obtain the optimum process condition parameters of anaerobic fermentation device, with the gas vo｜ume and quality parameters of fermentation engineering as response value, and establish the BP neural network model. At the same time, network model for global optimization by particle swarm optimization algorithm. Finally achieve the best anaerobic fermentation process parameters. This study shows that using BP neural network simulation with optimization of particle swarm algorithm has better optimization effect on anaerobic fermentation process, to provide the theory basis for the anaerobic fermentation process control.