Control Technology, Based On Improved Pid Neural Network Algorithm

Composite material is a new type materia,which has many Prominent merits and fine performances. It applies chiefly in fields of requirement of high intensity, toughness and stabilization. Composite material is an integrated technology, including macromolecule chemistry, mathematics, mechanics, electronics and computer, etc.This paper researches in tension control of filament winding system, which is an important control parameter and affects the quality of winding products directly. The main works in this paper are as follows:1. This paper details producing mothed for tension control and addresses the framework of tension control system. Bsaed on it, we design a constant tension control system with data analyse performance. This framework contains software design, bus design, data analyse, tension detection and date filter, in order to support tension control system.2. This paper analysis the principle of digital PID control. Then simulation validates the control performance of proportion control, proportion and integration control, proportion integration differential control in variation parameter.3. A detailed description of neural networks, in particular the BP neural network design principles and learning algorithms, and its functions are verified by simulation4. Designed and implemented based on BP neural network PID controller, the ultimate by simulation of the system dynamic and static performance.Finally, Summaries are made and the priorities and direction of future research are prospected.