Image Recognition Control Design And Implementation Based On The Network Monitoring In The Drainage System

City environmental protection and energy conservation have been always paid much attention. City sewage pumping system is of huge energy saving potential. Because of sewage pumping pipe network's characteristics such as nonlinearity, time-delay etc, improper control or fault signal would cause environmental pollution, so sewage pumping station's security monitoring is especially important.At present Hangzhou sewage pipe network system consists of two sewage treatment plants and 60 or more pumping stations, and in most pumping stations the pumps work at a constant speed or in the frequency control manner, and are supervised by the workers. The lack of fault detections would do harm to system security, and may results in high energy consumption. Once fault occurs, the information could not immediately be shared, which goes against global decision-making by the managers.Based on the sewage pumping pipe network system's charecteristics, web-based image recognition control system was discussed and analyzed in details. First, based on the proposed the security monitoring requirements of the drainage system network, a local video surveillance system was built. Image recognition control system's network structure and function of each module were proposed, the working principle was discussed, and solutions aiming at the involved key problem were partially achieved. Second, the system fault screen capture function was realized. Considered the influences taken by different environmental factors, the system key parameters region location method was suggested: pre-processing, separating based on projection and checking by prior knowledge, analyzed the experimental results, and proposed the improved methods. Then, based on the regional characteristics and the segmentation of the characters' commonly used methods' analysis, Ostu threshold segmentation and the vertical projection of the characters segmentation was used. After individual characters' thinning a classification and identification based on the character structure was adopted. Finally, a mathematical model was established according to the urban sewage pipe network system's time-delay, uncertain and non-linear features. Based on the sewage pumping station's principle of energy-saving, an adaptive fuzzy control strategy was proposed. The result of the instance simulation proved the effectiveness of the strategy.