Water resource multi-time scale configuration method based on typical years

Abstract

The invention provides a water resource multi-time scale configuration method based on typical years, and belongs to the technical field of basin water resource fine configuration. And according to the objective function and the constraint condition, respectively making an annual month-by-month plan scheduling process line, a monthly ten-day plan scheduling process line and a ten-day plan scheduling process line. The method is suitable for medium and small watershed with high error sensitivity to data, and error accumulation caused by over-fitting of the model is reduced. The year-by-month-by-month-by-day multi-time scale nesting method provided by the invention can flexibly cope with incoming water fluctuation and enhance the adaptability and robustness of water resource allocation. According to the method, a POA algorithm and a dynamic constraint mechanism are fused, and refined scheduling and engineering benefit maximization are realized; reference is provided for fine scheduling of small and medium-sized basin reservoirs, and technical support is provided for safe and high-benefit operation of the reservoirs.

Description

Technical Field

[0001] This invention belongs to the field of watershed water resource fine allocation technology, and more specifically, it relates to a method for water resource allocation at multiple time scales based on a typical year. Background Technology

[0003] With the increase in population and arable land area, rapid urbanization and industrialization have further exacerbated water resource shortages, highlighting the contradictions in the balanced allocation of water resources between upstream and downstream areas and between left and right banks. Research on water resource allocation started relatively early, and multiple disciplines, including water resource system theory, optimization theory, ecological theory, and economic theory, have proposed multi-objective decision-making, multi-level optimization, and optimization theories, and have conducted extensive research on various aspects such as water resource supply and demand balance, water rights trading, and water resource allocation.

[0004] However, research focuses on the spatial joint allocation of complex water conservancy projects and reservoir groups, which is less applicable to small and medium-sized river basins in northern China with underdeveloped information systems, incomplete data, and a lack of allocation mechanisms. Furthermore, it faces problems such as untimely allocation and large tracking errors. Therefore, a universal, easy-to-use, and precise water resource allocation method is urgently needed. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-timescale allocation method for water resources based on a typical year, in order to solve the shortcomings of the existing technology, such as untimely allocation and regulation and large planning tracking errors. This invention provides a multi-timescale allocation method for reservoir water resources based on a typical year, which performs nested coordinated scheduling for three scheduling periods: annual monthly, monthly ten-day, and ten-day daily.

[0006] To achieve the above objectives, the present invention provides a method for multi-timescale allocation of water resources based on a typical year, comprising the following steps: The total annual water inflow is calculated based on historical ten-day water volume data. Based on the principle of the most unfavorable conditions for water inflow, a set of reference years is determined; the similarity between the water inflow process and the reference year is evaluated year by year, and the reference year with the highest similarity is selected as the typical year for water inflow. By combining the typical annual water allocation ratio and the total inflow ratio of the annual forecast, the water volume of the year is proportionally allocated and optimized, and a monthly inflow process line based on a typical year is produced. The annual total water consumption index and irrigation quota are determined based on the total annual water inflow, and the total water demand of the basin is calculated and the monthly water demand process line is plotted. Clearly define the reservoir capacity to be controlled at the beginning and end of the year. Based on the objective function and constraints, use the POA algorithm to iterate and optimize the monthly water volume allocation according to the monthly water inflow and demand, and create a monthly planning and scheduling process line. Based on the annual and monthly planned scheduling process line, the control capacity of the reservoir at the beginning and end of the month is determined. Based on the objective function and constraints, the POA algorithm is used to traverse and optimize the monthly water volume allocation according to the monthly inflow and demand, and to create the monthly and ten-day planned scheduling process line. Obtain the hydrological forecast for the next ten days, determine the control reservoir capacity at the beginning and end of each ten-day period based on the monthly ten-day plan scheduling process line, and use the POA algorithm to traverse and optimize the daily water volume allocation for each ten-day period based on the objective function and constraints, and create the ten-day daily plan scheduling process line.

[0007] Furthermore, the statistical analysis of the total annual water inflow includes the following steps: A medium- to long-term inflow prediction model was established using time-series predictive analysis. The monthly runoff inflow for the coming year was predicted using the historical hydrological sequence at the ten-day scale of the inflow section. Draw the inflow process line from January to December based on the predicted monthly runoff; The runoff data for the next year are summed based on the aforementioned inflow process curve.

[0008] Furthermore, the medium- to long-term water inflow prediction model is as follows: ; The summation formula is: ; In the formula, i represents the historical year, j represents 1-12 months, and fx is the time series forecasting analysis function. The water volume for the j-th month of year i. Let the predicted water inflow be for month j. The total annual water inflow is as stated in the statistics.

[0009] Furthermore, the process of determining the typical year includes the following steps: Select the hydrological years in the historical sequence that satisfy the P=90% guarantee rate as the benchmark; Based on existing measured data, the water volume for each year was analyzed. Select certain years where the historical total inflow is within {90%~100%} of the total inflow of the stated year as reference years r: In the formula, Let r be the total water inflow in year r. For reference year set; In the reference year, the correlation coefficient method is used to assess the similarity between the inflow process and the inflow in the reference year on an annual basis: Select the reference year with the highest similarity as the typical year T; In the formula, This refers to the water volume in the jth month of year r.

[0010] Furthermore, the proportional allocation of water volume for the current year includes the following steps: The method of proportional water allocation is adopted, which allocates the water volume for the forecast year according to the water distribution ratio of each month in a typical year, and obtains the inflow process line based on the typical year. : ; In the formula, The water volume for the j-th month of year t. Let t be the total water inflow in year t.

[0011] Furthermore, the statistical analysis of total water demand in the basin and the plotting of monthly water demand curves include the following steps: First, based on the total annual water inflow, the years are divided into high-water years, normal-water years, low-water years, and extremely low-water years using the anomaly percentage method, and then determined according to different levels of the year; The monthly water demand of the basin is determined based on the annual total water consumption index and irrigation quota for the horizontal year. Based on the monthly water demand of the basin, draw the water demand process curve. .

[0012] Furthermore, the monthly water demand of the basin is the sum of agricultural water demand, industrial water demand, domestic water demand, and ecological water demand.

[0013] Furthermore, the process of creating the annual monthly plan scheduling line includes the following steps: Based on the actual water storage of the reservoir and the reservoir's flood season plan, and under the condition that the agricultural irrigation period, the main flood season, and the autumn and winter irrigation restrictions are met each month, a water-use scheduling plan is formulated in combination with the total water demand of the basin. During the allocation and adjustment process, two objectives must be met simultaneously: maximizing total water supply and minimizing negative water storage volume for the current month. ; By combining different periods such as irrigation season and flood season, a limit water level is set for each period, and the limit water level is converted into reservoir capacity through the relationship between water level and reservoir capacity. Thus, there is a reservoir capacity constraint every month. ; Based on the water storage level, the water distribution for each month is adjusted cyclically. When adjusting the model, the maximum water supply is used as the indicator to determine the control indicators at the end of the month. In the formula, For the water supply in month j, This represents the minimum annual water storage capacity. The reservoir's water demand for month j. This represents the maximum annual water storage capacity.

[0014] Furthermore, the POA control process in the monthly ten-day period and the ten-day period includes the following steps: Based on the water volume control for each month or ten-day period within the year, determine the water level and reservoir capacity at the beginning or end of the period, and use the POA algorithm to calculate the water distribution situation in each region for each ten-day period or day within the period. The reservoir water level is used as the decision variable for optimization, and the water level is discretized within a reasonable range at the beginning or end of the month with an accuracy of 0.01m. In optimal control, two objectives must be met: minimizing water supply deficit and minimizing waste of available water. ; In the formula, △Q n,t For the nth water user, the water supply deficit flow rate during time period t, where Δt is the length of the unit time period; Δq t The portion of the reservoir's discharge during time period t that is neither used to supply water to various water users nor belongs to the ecological flow; The reservoir water level must be controlled to ensure that it does not exceed the dynamic flood control limit level or fall below the minimum water level each month. In the formula, Q i,t Let q be the discharge flow of the i-th reservoir during time period t. i,min Let be the flow rate corresponding to the minimum water storage level of the i-th reservoir.

[0015] Compared with the prior art, the present invention has the following technical effects: This invention presents a multi-timescale water resource allocation method based on a typical year. It determines the inflow process curve based on a typical year and is suitable for small and medium-sized watersheds with low robustness to data and high error sensitivity, effectively reducing error accumulation caused by model overfitting. The method proposes a nested approach across multiple timescales: year-monthly, month-ten-day, and ten-day-day, which can flexibly respond to fluctuations in inflow and enhance the adaptability and robustness of water resource allocation. This invention integrates the Proof-of-Action (POA) algorithm with a dynamic constraint mechanism to achieve refined scheduling and maximize engineering benefits; it provides a reference for refined scheduling of reservoirs in small and medium-sized watersheds and offers technical support for the safe and efficient operation of reservoirs. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart of a method for configuring water resources across multiple time scales based on a typical year, provided in an embodiment of the present invention; Figure 2A schematic diagram of the annual monthly water inflow curve for a typical year provided in this embodiment of the invention; Figure 3 This is a schematic diagram of the annual and monthly planning and scheduling process provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the technical problem to be solved, the technical solution, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0019] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0020] This invention provides a method for allocating water resources across multiple time scales based on typical years in reservoirs of small and medium-sized river basins. The process is as follows: Figure 1 As shown, it includes the following steps: Step S1: Establish a medium- to long-term water inflow prediction model using time series predictive analysis. Predict the monthly water inflow for the coming year using the historical hydrological sequence at the ten-day scale of the inflow section, draw the water inflow process line, and calculate the total annual water inflow.

[0021] Step S2: Based on the principle of the most unfavorable conditions for 90% of the incoming water, determine the set of reference years; use the Euclidean distance method to evaluate the similarity between the incoming water process and the incoming water of the reference years year by year, and select the reference year with the lowest similarity as the typical year of incoming water.

[0022] Step S3: Combine the typical annual water allocation ratio and the total inflow ratio predicted for the year to optimize the proportional allocation of water volume for the year, and create a monthly inflow process line based on the typical year.

[0023] Step S4: Determine the annual total water consumption index and irrigation quota based on the total annual water inflow, and calculate the total water demand of the basin and draw the monthly water demand process line.

[0024] Step S5: Based on the determined objective function and constraints, with the goal of minimizing the deficit, the POA algorithm is used to iterate and optimize the monthly water allocation based on the monthly water inflow and demand, and to create the monthly planned scheduling process line.

[0025] Step S6: Based on the annual and monthly planned scheduling process line, determine the control capacity at the beginning and end of the month. Based on the objective function and constraints in step S5, with the goal of minimizing the deficit, use the POA algorithm to iterate and optimize the monthly ten-day water volume allocation according to the monthly inflow and demand, and create the monthly ten-day planned scheduling process line.

[0026] Step S7: Obtain the hydrological forecast for the next ten days, determine the control capacity at the beginning and end of the ten-day period based on the monthly ten-day plan scheduling process line, and based on the objective function and constraints in Step S5, with the goal of minimizing the deficit, use the POA algorithm to iterate and optimize the daily water volume allocation for the ten-day period based on the water inflow and water demand for the next ten days, and create the ten-day daily plan scheduling process line.

[0027] In step S1 above, calculating the total annual water inflow includes the following steps: Step S11: Use time-series predictive analysis methods, such as regression analysis, LSTM, and ARIMA, to predict the monthly runoff for the next 12 months: (Equation 1) In the formula, i represents historical years (e.g., 1990-2024), j represents 1-12 months, and fx is the time series forecasting analysis function. The water volume for the j-th month of year i. This represents the predicted water inflow for month j.

[0028] Step S12: Obtain the predicted runoff inflow process curve for January to December. : Step S13: Sum the runoff data for the next 1-12 months: In the formula, This refers to the total annual water inflow.

[0029] In step S2 above, the process of determining a typical year includes: Step S21: The worst-case scenario principle includes selecting hydrological years in the historical sequence that satisfy the P=90% guarantee rate as the benchmark.

[0030] Step S22: Based on the existing measured data, iterate through the water volume of each year. Select the historical total water inflow within the statistical annual total water inflow. Certain years of {90%~100%} are used as reference years r (Equation 4).

[0031] In the formula, Let r be the total water inflow in year r. This is a reference year set.

[0032] Step S23: In the reference year, the correlation coefficient method is used to assess the similarity between the inflow process and the inflow in the reference year.

[0033] (Equation 5) In the formula, This refers to the water volume in the jth month of year r.

[0034] Step S24: Select the reference year with the lowest similarity as the typical year T, then the water inflow process curve for month j in year t is represented as follows: .

[0035] In step S3 above, the process of proportionally allocating the annual water volume includes: Step S31: Using the water volume proportional control method, the water volume for the forecast year is proportionally allocated according to the water distribution ratio of each month in a typical year, thus obtaining the inflow process curve based on the typical year. (Formula 6): In the formula, The water volume for the j-th month of year t. Let t be the total water inflow in year t.

[0036] In step S4 above, the process of calculating the total water demand of the basin and drawing the monthly water demand process line includes: Step S41: First, classify the years into high-water years, normal-water years, low-water years, and extremely low-water years based on the annual total water inflow using the anomaly percentage method, and determine the appropriate level of the year based on the different levels.

[0037] Step S42: Determine the monthly water demand of the basin based on the annual total water consumption index and irrigation quota for the horizontal year.

[0038] Step S43: The monthly water demand of the basin is the sum of agricultural water demand, industrial water demand, domestic water demand and ecological water demand.

[0039] Step S44: Draw the monthly water demand process curve based on the monthly water demand of the basin. .

[0040] In step S5 above, the process of creating the annual monthly plan scheduling process line includes: Step S51: Based on the actual water storage of the reservoir and the reservoir's flood season plan, and under the condition that the agricultural irrigation period, the main flood season, and the autumn and winter irrigation restrictions are met each month, and in combination with the total water demand of the basin, formulate a water-saving scheduling plan.

[0041] Step S52: During the allocation and adjustment process, two objectives must be met simultaneously: the maximum total water supply (Equation 7) and the minimum negative regulation water storage volume for the current month (Equation 8). (Equation 8) Step S53: Combining different periods such as irrigation season and flood season, set a limit water level for each period, and convert the limit water level into reservoir capacity through the water level-reservoir capacity relationship. Then there is a reservoir capacity constraint every month (Equation 9): In the formula, For the water supply in month j, This represents the minimum annual water storage capacity. The reservoir's water demand for month j. This represents the maximum annual water storage capacity.

[0042] Step S54: Based on the water storage level conditions, adjust the water distribution for each month. The maximum water supply is used as the indicator during model adjustment to determine the end-of-month control targets.

[0043] In steps S6 and S7 above, the process of POA control in monthly ten-day periods and ten-day periods includes: Step S61: Based on the water volume control for each month of the year (Step S6) / each ten-day period within the month (Step S7), determine the water level and reservoir capacity at the beginning or end of the period, and use the POA algorithm to calculate the water distribution situation in each region for each ten-day period / day within the period.

[0044] Step S62: In the optimization scheme, the reservoir water level is used as the decision variable for optimization, and the scheme is discretized within a reasonable water level range at the beginning or end of the month with an accuracy of 0.01m.

[0045] Step S63: In the optimization control, two objectives must be met: minimizing the water supply deficit (Equation 10) and minimizing the waste of available water (Equation 11).

[0046] (Equation 10) (Equation 11) △Q n,t For the nth water user, the water supply deficit flow c during time period t, where Δt is the length of the unit time period; Δq t This refers to the portion of the reservoir's discharge during time period t that is neither used to supply water to various water users nor belongs to the ecological flow.

[0047] Step S64: In addition to satisfying general constraints such as reservoir water balance constraints, reservoir capacity constraints, water level constraints, flow constraints, and boundary constraints during the model solution process, the following are added: monthly dynamic flood limit water level regulation, ecological flow regulation of the two reservoirs, and monthly minimum water level control of the reservoirs; the monthly reservoir water level must not exceed the dynamic flood limit water level, nor be less than the minimum water level control.

[0048] (Equation 12) Q i,tLet q represent the discharge flow of the i-th reservoir during time period t. i,min Let be the flow rate corresponding to the minimum water storage level of the i-th reservoir.

[0049] This invention provides a multi-timescale water resource allocation method based on a typical year. It determines the inflow process line based on a typical year and is suitable for small and medium-sized watersheds with low robustness to data and high error sensitivity, effectively reducing error accumulation caused by model overfitting. The method proposes a nested approach across multiple timescales: year-monthly, month-ten-day, and ten-day-day, which can flexibly respond to fluctuations in inflow and enhance the adaptability and robustness of water resource allocation. This invention integrates the Proof-of-Action (POA) algorithm with a dynamic constraint mechanism to achieve refined scheduling and maximize engineering benefits; it provides a reference for refined scheduling of reservoirs in small and medium-sized watersheds and offers technical support for the safe and efficient operation of reservoirs.

[0050] The following specific embodiment illustrates a method for multi-timescale allocation of water resources based on a typical year according to an embodiment of the present invention.

[0051] This invention takes the water resource scheduling of the Toutun River Basin as an example. The scheduling plan is formulated according to the rule of uniform spatial and temporal distribution of water resources. Compared with the scheduling plan without water intake rules, it significantly improves the balance of water supply between upstream and downstream water users in the basin.

[0052] The Toutun River basin is located in the middle section of the northern slope of the Tianshan Mountains in Xinjiang. It originates from the Tengger Peak of the Tengger Mountains within the Tianshan range. The Toutun River is 144 km long, with a total drainage area of ​​2553 km². 2 The Toutun River suffers from uneven spatial and temporal distribution of water resources. Its inflow primarily relies on snowmelt from mountainous areas and precipitation, with the flood season occurring from June to August, accounting for over 65% of the annual total. Furthermore, due to weak water conservancy project regulation capabilities, limited water supply capacity, and over-extraction of groundwater resources within the basin, the lower reaches experience severe water shortages in spring. Spring is the agricultural planting season, and water scarcity significantly impacts agricultural production. Traditional monthly water allocation methods are insufficient to meet the current and future needs for refined water resource allocation management in the basin. Therefore, a multi-scale nested allocation method is urgently needed to improve water resource allocation.

[0053] See Figure 1 The present invention provides a method for multi-timescale allocation of water resources based on a typical year, which specifically includes the following steps: Step S1: Combine the compiled data of the Toutun River Basin to obtain the ten-day water supply data for the long time series from 1990 to 2024.

[0054] Predicting January 2025 water inflow using non-stationary series stepwise regression periodic analysis. Approximately 5.31 million cubic meters, water inflow in February 2025. Approximately 4.44 million cubic meters, water inflow in March 2025. Approximately 5.81 million cubic meters, water inflow in April 2025. Approximately 12.57 million cubic meters, water inflow in May 2025. Approximately 23.47 million cubic meters, water inflow in June 2025. Approximately 30.72 million cubic meters, water inflow in July 2025. Approximately 44.77 million cubic meters, water inflow in August 2025. Approximately 35.14 million cubic meters, water inflow in September 2025. Approximately 16.06 million cubic meters, water inflow in October 2025. Approximately 9.26 million cubic meters, water inflow in November 2025. Approximately 7.28 million cubic meters, water inflow in December 2025. It is approximately 5.34 million cubic meters.

[0055] The runoff inflow data from January to December 2015 were summed to obtain a predicted runoff inflow of 200.18 million cubic meters.

[0056] Step 2: Based on the worst-case scenario principle, select a year with an inflow of 180 to 200 million cubic meters of water as a dry year.

[0057] Based on the long-term ten-day water supply data from 1990 to 2024, 2001, 2004, 2010, 2021 and 2023 were selected as reference years.

[0058] Calculate according to Equation 5 The correlation coefficients with 2001, 2004, 2010, 2021, and 2023 were 0.958, 0.962, 0.782, 0.938, and 0.871, respectively. 2004, with the highest similarity coefficient, was selected as the typical year. T。 Therefore, 2004 was selected as the typical year.

[0059] Step 3: Based on the typical annual water allocation ratio and the predicted total inflow ratio, the annual water volume is proportionally allocated, resulting in the following proportional inflows for January to December 2024: 3.33 million cubic meters, 2.77 million cubic meters, 3.39 million cubic meters, 9.46 million cubic meters, 21.12 million cubic meters, 24.62 million cubic meters, 44.62 million cubic meters, 44.06 million cubic meters, 21.92 million cubic meters, 3.04 million cubic meters, 6.98 million cubic meters, and 4.87 million cubic meters. The monthly inflow curve for a typical year is shown below. Figure 2 As shown, Figure 2 In the figure, the vertical axis is in units of 10,000 cubic meters.

[0060] Step 4: Calculate the water demand in the interval and downstream, and calculate the total water demand of the basin.

[0061] Step 5: Based on different periods such as irrigation season and flood season, set limiting water levels for each time period, and convert the limiting water levels into reservoir capacity using the water level-reservoir capacity relationship to obtain the reservoir capacity constraints for each month. Specific constraints are as follows: Reservoir water balance constraints: (Equation 13) Storage capacity constraints: (Equation 14) Water level constraints: (Equation 15) Flow constraints: (Equation 16) Boundary conditions: (Equation 17) (Equation 18) Other constraints: In addition to the above constraints, to meet the requirements, the following measures have been added: monthly dynamic flood control level regulation, ecological flow regulation of the two reservoirs, and monthly minimum reservoir water level control; the monthly reservoir water level must not exceed the dynamic flood control level or be lower than the minimum water level control. (Equation 12) In the above formula, V i,t VL i,t VU i,t Let I be the minimum and maximum values ​​of the water storage volume and water storage capacity of the i-th reservoir during time period t; i,t Q i,t These represent the inflow and outflow of the i-th reservoir during time period t; QL i,t QU i,t Z represents the lower and upper limits of the discharge flow from the i-th reservoir during time period t; i,min Z i,max Z represents the minimum and maximum water levels of the i-th reservoir, respectively; i,t Z represents the water level of the i-th reservoir during time period t; i,start Z i,end These represent the initial and final water levels of the i-th reservoir during its operation; q i,min Let be the minimum water level of the i-th reservoir.

[0062] Based on water storage level conditions, monthly water allocation is adjusted iteratively. Using the Power of Analysis (POA) algorithm, with maximum water supply as the indicator, monthly control targets are determined. Through continuous iterative calculations, an optimized annual monthly water allocation scheme is obtained, and an annual monthly planned scheduling process line is created, as shown below. Figure 3 As shown.

[0063] Step 6: Based on the annual and monthly planned scheduling process line, determine the control capacity of the reservoir at the beginning and end of the month. According to the objective function and constraints in Step 5, with the goal of minimizing the water shortage, use the POA algorithm to traverse and optimize the monthly ten-day water allocation. During the optimization process, the reservoir water level is used as the decision variable, and the optimization is performed with an accuracy of 0.01m within the reasonable water level range at the beginning or end of the month. The optimization control must satisfy two objectives: minimizing the water supply shortage and minimizing the waste of available water. It must also satisfy general constraints such as reservoir water balance constraints, capacity constraints, water level constraints, flow constraints, and boundary constraints, as well as additional special constraints such as the regulation of the monthly dynamic flood control limit water level, the regulation of the ecological flow of the two reservoirs, and the monthly minimum water level control of the reservoirs. Finally, a monthly ten-day planned scheduling process line is created.

[0064] Step 7: Obtain the hydrological forecast for the next ten days. Based on the monthly ten-day plan scheduling process line, determine the control reservoir capacity at the beginning and end of the ten-day period. According to the objective function and constraints in Step 5, with the goal of minimizing the deficit, use the POA algorithm to traverse and optimize the daily water conservancy configuration for each ten-day period. Similarly, in the optimization process, the reservoir water level is used as the decision variable, discretized with an accuracy of 0.01m, and the relevant objective and constraint conditions are met to create the ten-day daily plan scheduling process line.

[0065] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A method for multi-timescale allocation of water resources based on a typical year, characterized in that, Includes the following steps: The total annual water inflow is calculated based on historical ten-day water volume data. Based on the principle of the most unfavorable conditions for incoming water, a set of reference years is determined; The similarity between the inflow process and the reference year is assessed year by year, and the reference year with the highest similarity is selected as the typical year for inflow. By combining the typical annual water allocation ratio and the total inflow ratio of the annual forecast, the water volume of the year is proportionally allocated and optimized, and a monthly inflow process line based on a typical year is produced. The annual total water consumption index and irrigation quota are determined based on the total annual water inflow, and the total water demand of the basin is calculated and the monthly water demand process line is plotted. Clearly define the reservoir capacity to be controlled at the beginning and end of the year. Based on the objective function and constraints, use the POA algorithm to iterate and optimize the monthly water volume allocation according to the monthly water inflow and demand, and create a monthly planning and scheduling process line. Based on the annual and monthly planned scheduling process line, the control capacity of the reservoir at the beginning and end of the month is determined. Based on the objective function and constraints, the POA algorithm is used to traverse and optimize the monthly water volume allocation according to the monthly inflow and demand, and to create the monthly and ten-day planned scheduling process line. Obtain the hydrological forecast for the next ten days, determine the control reservoir capacity at the beginning and end of each ten-day period based on the monthly ten-day plan scheduling process line, and use the POA algorithm to traverse and optimize the daily water volume allocation for each ten-day period based on the objective function and constraints, and create the ten-day daily plan scheduling process line.

2. The method for multi-timescale allocation of water resources based on a typical year as described in claim 1, characterized in that, The statistical analysis of the total annual water inflow includes the following steps: A medium- to long-term inflow prediction model was established using time-series predictive analysis. The monthly runoff inflow for the coming year was predicted using the historical hydrological sequence at the ten-day scale of the inflow section. Draw the inflow process line from January to December based on the predicted monthly runoff; The runoff data for the next year are summed based on the aforementioned inflow process curve.

3. The method for multi-timescale allocation of water resources based on a typical year as described in claim 2, characterized in that, The medium- to long-term water inflow prediction model is as follows: ； The summation formula is: ; In the formula, i represents the historical year, j represents 1-12 months, and fx is the time series forecasting analysis function. The water volume for the j-th month of year i. Let the predicted water inflow be for month j. The total annual water inflow is as stated in the statistics.

4. The method for multi-timescale allocation of water resources based on a typical year as described in claim 3, characterized in that, The process of determining the typical year includes the following steps: Select the hydrological years in the historical sequence that satisfy the P=90% guarantee rate as the benchmark; Based on existing measured data, the water volume for each year was analyzed. Select certain years where the historical total inflow is within {90%~100%} of the total inflow of the stated year as reference years r: In the formula, Let r be the total water inflow in year r. For reference year set; In the reference year, the correlation coefficient method is used to assess the similarity between the inflow process and the inflow in the reference year on an annual basis: Select the reference year with the highest similarity as the typical year T; In the formula, This refers to the water volume in the jth month of year r.

5. The method for multi-timescale allocation of water resources based on a typical year as described in claim 4, characterized in that, The proportional allocation of water volume for the current year includes the following steps: The method of proportional water allocation is adopted, which allocates the water volume for the forecast year according to the water distribution ratio of each month in a typical year, and obtains the inflow process line based on the typical year. : ； In the formula, The water volume for the j-th month of year t. Let t be the total water inflow in year t.

6. The method for multi-timescale allocation of water resources based on a typical year as described in claim 4, characterized in that, The process of calculating the total water demand of the watershed and plotting the monthly water demand curve includes the following steps: First, based on the total annual water inflow, the years are divided into high-water years, normal-water years, low-water years, and extremely low-water years using the anomaly percentage method, and then determined according to different levels of the year; The monthly water demand of the basin is determined based on the annual total water consumption index and irrigation quota for the horizontal year. Based on the monthly water demand of the basin, draw the water demand process curve. .

7. The method for multi-timescale allocation of water resources based on a typical year as described in claim 6, characterized in that, The monthly water demand of the basin is the sum of agricultural water demand, industrial water demand, domestic water demand and ecological water demand.

8. The method for multi-timescale allocation of water resources based on a typical year as described in claim 6, characterized in that, The process of creating the annual monthly plan scheduling line includes the following steps: Based on the actual water storage of the reservoir and the reservoir's flood season plan, and under the condition that the agricultural irrigation period, the main flood season, and the autumn and winter irrigation restrictions are met each month, a water-use scheduling plan is formulated in combination with the total water demand of the basin. During the allocation and adjustment process, two objectives must be met simultaneously: maximizing total water supply and minimizing negative water storage volume for the current month. ； By combining different periods such as irrigation season and flood season, a limit water level is set for each period, and the limit water level is converted into reservoir capacity through the relationship between water level and reservoir capacity. Thus, there is a reservoir capacity constraint every month. ; Based on the water storage level, the water distribution for each month is adjusted cyclically. When adjusting the model, the maximum water supply is used as the indicator to determine the control indicators at the end of the month. In the formula, For the water supply in month j, This represents the minimum annual water storage capacity. The reservoir's water demand for month j. This represents the maximum annual water storage capacity.

9. The method for multi-timescale allocation of water resources based on a typical year as described in claim 8, characterized in that, The process of POA control in the monthly and daily ten-day periods includes the following steps: Based on the water volume control for each month or ten-day period within the year, determine the water level and reservoir capacity at the beginning or end of the period, and use the POA algorithm to calculate the water distribution situation in each region for each ten-day period or day within the period. The reservoir water level is used as the decision variable for optimization, and the water level is discretized within a reasonable range at the beginning or end of the month with an accuracy of 0.01m. In optimal control, two objectives must be met: minimizing water supply deficit and minimizing waste of available water. ； In the formula, △Q n,t For the nth water user, the water supply deficit flow rate during time period t, where Δt is the length of the unit time period; Δq t The portion of the reservoir's discharge during time period t that is neither used to supply water to various water users nor belongs to the ecological flow; The reservoir water level must be controlled to ensure that it does not exceed the dynamic flood control limit level or fall below the minimum water level each month. In the formula, Q i,t Let q be the discharge flow of the i-th reservoir during time period t. i,min Let be the flow rate corresponding to the minimum water storage level of the i-th reservoir.

Images