A New Training Algorithm for the Process Neural Network and Its Application

• 作 者: GuanShouping;PengJun;
• 机 构: Key Laboratory of Process Industry Automation,Ministry of Education,Northeastern University,Shenyang
• 出 处: 中国通信学会第五届学术年会
• 发表日期: 2008年期

This paper presents a new training aigodthm for the process neural network(PNN)when it is used to model an industrial process.On the base of the pretreatment for the process discrete data considering their including some pseudo ones，a new training algorithm based on discrete Walsh conversion Was used to convert the discrete data to be the direct inputs of PNN，which can shorten the PNN training time and improve the PNN mapping capability.The PNN model with the new training algorithm and two hidden-layers structure Was appfied tO forecast the mycefium density of the glutamate fermentation process，and the simulation results were excellent.