Development & Optimization Of Intelligent Monitoring System And Application Research On Leakage Detection For Water Supply Networks

Nowadays,the urban water supply networks are faced with problems such as high rate of leakage,frequent pipe burst accidents,unreasonable scheduling and secondary water pollution.So it is urgent to establish effective monitoring and scientific data analysis methods to efficiently solve above problems.However,the urban water supply networks are usually large-scale and buried underground which result in difficult to monitor.Neither the SCADA(Supervisory Control and Data Acquisition)system nor the commercial monitoring equipment is up to the networks monitoring task due to their high cost,standalone system structure and installation site restriction.Meanwhile,water supply networks are laid in complex urban environment resulting in monitoring data inevitably mixed with great amount of noise.While the traditional way of monitoring is lack of targeted denoising approach and it is unfavorable for accurate data analysis.In response to these problems,firstly,this paper introduced wireless sensor networks to develop an intelligent water supply networks monitoring system to meet the large-scale monitoring need.Then a research of pressure monitoring data denoising method was carried out on basis of above system.Finally,this paper took pipe leakage,the toughest task for water supply networks management,as an example to develop leakage identification technique based on water pressure wave analysis.For intelligent monitoring system,this paper developed low-power and low-cost terminal monitoring devices and gateway monitoring devices respectively.A dynamic ad-hoc network policy based on ZigBee was designed to ensure the wireless sensor networks establishment and its reliable operation.The gateway device,in the topmost level of the network,undertook gateway function between ZigBee network and mobile communication network for remote communication.A software platform in data center was developed to realize the matched functions such as data organization,strorage,publishing,control and device debugging.And a matched installation manal was designed to guide the actual monitoring application.For the monitoring system denoising optimization,this paper analyzed thecharacteristics of water pressure monitoring signals and adopted wavelet transform to implement data denoising optimal research from respects of threshold function,decomposition scale and basis function.In this way the water supply networks water pressure monitoring signal denosing model was constructed.For the monitoring system application research,this paper developed pipe leakage identification technique from respect of transient pressure wave analysis.The experiment adopted method of pipe scanning by artificial pressure wave in conditions of perfect pipes,single hole leakage,single crack leakage,branch outflow,multi-leakage,leakage plus outflow and so on.First,leakage charateristics were analyzed qualitatively and leakage location was calculated precisely by means of time-domain analysis.Second,by means of optimal local wave decomposition,the monitored pressure wave was adaptively decomposed into several intrinsic mode function components.To further describe the pressure wave characteristics of different pipe conditions,the instantaneous frequency was calculated for each component.Finally,by means of quantitative analysis both in values of instanceeous frequency RMS(eneygy)and approximate entropy for each component,the interference of various items(outflow,multi-leakage,etc.)to identify leakage were avoided effectively and precise leakage characteristics identification in complex conditions were achived successfully.