BP Neural Network Campus Interval Water Demand Prediction And Optimal Dispatching Method Based On Bayesian Criterion

With the rapid growth of economy and the gradual improvement of people's living standard,the domestic demand for domestic water is increasing day by day,the contradiction between supply and demand of water resources is becoming more and more prominent,and the rational utilization of water resources is very important.For the water supply network system,accurate and reliable water resources demand forecasting method is the important foundation of urban water supply dispatching,and reasonable and effective water resources dispatching method is the necessary link to realize the sustainable development of water resources.Accurate prediction and optimal scheduling of water resources are conducive to the rational planning of water supply,water use and water saving,thus promoting the efficient operation of water supply network system.This paper combines the data of water use in campus history,and uses Bayesian BP neural network interval prediction method to predict the water use in campus.This method can improve the problem of local minimization and slow convergence of BP neural network.Meanwhile,it can effectively predict the fluctuation range of the daily water consumption of the campus,and the accuracy of the forecast is 96.7%,thus providing an effective reference for water resource allocation.This paper first introduces the specific method of forecasting the water demand interval,then optimizes the BP neural network with Bayes rule,then simulates the interval prediction,and compares it with the traditional BP neural network prediction method.The simulation results show that most of the predicted values are basically consistent with the actual water consumption.The absolute value of the relative error is 1.6%,and the maximum time prediction error is 4.2%,thus verifying the effectiveness and accuracy of the proposed prediction method.In addition,combining the campus water data,taking the water supply,water demand and water cost as the constraint conditions,taking the population goal and the water environment target as the comprehensive objective function,the paper optimizes the relevant parameters of the water supply pump by genetic algorithm.Thus,a water resources scheduling method based on comprehensive objective genetic algorithm is established.This method can not only ensure that the teachers and students make full use of water resources,but also protect the water environment effectively,save water resources,avoid the traditional water use and ignore the use of comfort,butignore the water resources and the waste of water resources.In addition,the throttling device is introduced to further economize water resources under the control of synthetic objective genetic algorithm.Finally,the results of an example show that the method is reasonable and effective in optimal water resources regulation,and provides an effective reference for the allocation of water resources.