The Research Of DCS And Intelligent Optimization Strategy For Sewage Treatment Plant Based On Carrousel-2000 Technology

On the basis of one certain sewage treatment plant, this paper approaches the realization of DCS and intelligent optimization control strategy for sewage treatment plant based on Carrousel-2000 technology.First, after analyzing the theory, characteristics, process flow and main technical parameters of Carrousel-2000 oxidation ditch technology, this paper determines the overall design scheme of DCS and the detailed control strategy in key links. According to the overall scheme, this paper finishes the design of device layer, field control station and central control room; develops the monitoring programs by Sunytech8.0 configuration software in supervisory computer; realizes the data communication among supervisory computer, field control station and device layer via network, ultimately achieves distributed control and concentrate management of this sewage treatment plant.Secondly, aiming at the importance and difficulty of DO concentration control, this paper introduces parameter self-tuning Fuzzy-PID control strategy into DO control system of this plant. On the basis of PID parameter control table which is built by off-line fuzzy inferencing, this paper write table-look program with ST language in configuration software. In actual project, according to the deviation, the system on-line amends PID parameters through table-look program to achieve accurate control of DO concentrations in different working conditions. For the reliable problem of DO concentration resulted by one point sensing, this paper adopt multi-point sensing method to collect the signal, use the data fusion algorithm which is writed by ST language in supervisory computer to analyze those data, and get the reliable DO concentration in aerobic zone for control system.Finally, aiming at the absence of self-learning capacity in fuzzy control, this paper constructs FNN-PID control strategy which is off-line optimized by improved Genetic Algorithm. The adjusting of rules and membership function in fuzzy control is converted into the optimization of connection weights and network structure in neutral network. Through the self-learning capacity of neutral network optimize the parameters of Fuzzy-PID controller.