Research On Water Supply And Demand Prediction And Multi-Objective Optimal Allocation In Hillock And Hilly Region Of Loess

In this paper,Heshui County of Gansu Province,located in the loess Ridge hilly region,is selected as a typical area to carry out the research on water supply and demand forecast and its multi-objective optimal allocation.Based on the analysis of water resources situation in Heshui County,the paper makes a multi-method forecast on the future water supply and demand of this area,and selects the most reasonable result as the forecast value by using the"three-line"constraint and rationality analysis of water resources.Based on the forecast results of supply and demand water supply balance analysis,and establish a multi-purpose water resource optimal allocation model with the goal of maximizing economic benefits,social benefits and ecological benefits.Finally,from the model’s pareto optimal solution set,choose the solution that most accords with the local development planning and the actual situation as the allocation result.The main research methods include:Pettitt test,Mann-Kendall trend analysis,variational mode decomposition,GM(1,1)grey prediction,multi-objective particle swarm optimization algorithm,etc.The main achievements are as follows:(1)Analyze the basic situation and development trend of physical geography,climate,underlying surface conditions,economy,water resources of Heshui County.The average total amount of water resources over the years is 79 million m~3,including 76.52 million m~3of river runoff(surface water resources)and 16.46 million m~3of groundwater resources(phreatic water).The repeated calculation between river runoff and groundwater is 13.98million m~3,and the available water resources are 35.47 million m~3,with an average utilization rate of 44.9%.The current annual water supply of water resources is 20.4 million m~3,with surface water accounting for 73.7%and groundwater accounting for 26.3%.According to their proportion,they are mainly used for agricultural water,industrial water,and domestic water;Analyzing the runoff data of Heshui County in the past 30 years,it is found that the rainfall has shown an upward trend,while the runoff in the Heshui River and Hulu River basins has shown an upward trend,while the runoff in the main stream of Malian River and Gucheng River basins has shown a downward trend.(2)Forecast the water supply and demand in horizontal planning years.The water supply forecast is mainly based on the current water supply situation and future water conservancy planning.In 2025,the water supply of 50%,75%and 95%at different frequencies is 27.89 million m~3,27.28 million m~3and 26.71 million m~3,respectively.In2030,the water supply of 50%,75%and 95%at different frequencies will be 38.24 million m~3,37.63 million m~3and 37.07 million m~3respectively.Water demand prediction adopts the traditional method and VMD-GM(1,1)model to jointly forecast and choose the best(2020as the base year)by comprehensive comparison.The VMD-GM(1,1)model adopts the idea of"divide before combine",first decomposes the original sequence,then fits and predicts the decomposed new sequence respectively and then combines.The final result has a high degree of fitting,and the relative error of the base year is only-0.04%,and the predicted value is in line with the regional reality and the requirements of water conservancy development planning.(3)The forecast water supply and demand under different water supply conditions in different years of different levels are analyzed respectively.In 2025,there is no water shortage phenomenon under the frequency of 50%and 75%.When the frequency is 95%,the water shortage is 713,300 m~3,and the water shortage rate is 2.6%.In 2030,when the frequency is 50%and 75%,there will be no water shortage,while when the frequency is95%,there will be 726,700 m~3 of water shortage,with a water shortage rate of 1.92%.(4)According to the predicted water supply and demand value of each planning level year,a multi-objective water resource optimal allocation model was established based on the allocation relationship between different water sources and different departments,in which economic,social and ecological benefits were taken as the targets.The multi-objective particle swarm optimization algorithm is used to solve the model,and the optimal solution of pareto optimal solution set is selected as the result of optimal configuration according to the needs of regional development planning and actual situation.After analysis,the final optimal allocation result accords with the water conservancy development plan and the actual situation of the region and has practical guidance for the future allocation of water resources in Heshui County.The research results can be used as a reference for future water resources allocation in Heshui County,and can also provide theoretical basis and reference basis for further research.The research methods and results of this paper can also be used as reference for water resources allocation or planning in other regions.