Research And Development Of Water Supply Dispatching Remote Monitoring System For Intelligent Water Affairs

In recent years,the state has attached great importance to the informatization construction of township water conservancy facilities.The government proposes to enhance technology support for the development of water supply in townships.It is urgently needed to construct more smart water affairs and smart pumping stations.However,at present,the automation level of water supply in rural areas is relatively low,and the scheduling control of water supply relies on manual labor which leads to poor real-time water supply in townships and high management costs.Meanwhile,water consumption statistics have not been scientifically used.Based on the practical application needs of water supply enterprises,this paper studies the short-term water consumption forecasting algorithm and designs a remote monitoring system for water supply dispatching for rural areas in my countryFirstly,the problems in the existing remote monitoring systems and water consumption prediction algorithms are analyzed.The architecture of the remote monitoring system for water supply scheduling is designed for the requirements of water supply enterprises,which includes a general Io T terminal,a short-term water consumption prediction algorithm and an information platform.Secondly,the terminal hardware system for water supply scheduling based on the NB-Io T is developed.The system is capable of multi-mode data acquisition to collect 4-20 m A signal,pulse signal and serial bus communication and other types of sensor data,and the control of peripherals according to the cloud commands.The terminal improves the real-time performance of the water supply scheduling system.Thirdly,inspired by data-driven methods,a short-term water consumption prediction model is established.The raw data is preprocessed by filling missing values and correcting outliers.The LSTM-Attention water consumption prediction model is established by combing the LSTM model with the Attention mechanism,and tuning hyperparameters.By comparing with other models,the prediction accuracy of the LSTM-Attention model is verified.Then,the characteristics of water consumption data are analyzed and the water consumption prediction model is improved.By analyzing the principle of CEEMDAN method and comparing its advantages and disadvantages with EMD method,a CEEMDAN-LSTMAttention water consumption prediction model is proposed.The accuracy of the model predictions was verified by linear regression analysis of the actual water data.Compared with other models,the comparative experimental results show that the combined model proposed in this paper has higher fitting accuracy and lower prediction error for water consumption prediction,which greatly improves prediction performance.Finally,the information platform of water supply scheduling remote monitoring system is designed.The platform realizes Data management,real-time control,water consumption prediction and data display.The water supply scheduling remote monitoring system developed in this paper has been put into operation in March 2022.At present,the system runs reliably and achieves the goal of system design.The system provides guidance for the water supply scheduling strategy of water supply enterprises which has great application value.