Hydropower station optimal scheduling method coupled with medium-and-long-term runoff forecast information based on total forecast scheduling graph

Abstract

The invention discloses a hydropower station optimal dispatching method based on coupling medium and long term runoff forecast information of a total forecast dispatching graph, and belongs to the technical field of reservoir dispatching. The method comprises the following steps: firstly, designing a total forecast scheduling graph and decision modes of different partitions in each sub forecast scheduling graph contained in the total forecast scheduling graph; secondly, establishing a hydropower station total forecast scheduling graph optimization model; thirdly, performing multi-year long-series simulation on the power generation scheduling process of the hydropower station, and optimizing the total forecast scheduling graph; and finally, calculating the decision output of the hydropower station by taking the total forecast scheduling graph as the hydropower station scheduling rules under different long-term runoff forecast levels and taking the medium-term and long-term runoff forecast information as input. According to the method, a traditional single forecast scheduling graph is replaced by a plurality of sub-forecast scheduling graphs, a medium-term scheduling decision space is determined according to long-term runoff forecast grading information, a specific medium-term scheduling decision is made according to'reservoir capacity + medium-term inflow ', coupling utilization of medium-term and long-term runoff forecast information is achieved, and then more efficient utilization of water resources is achieved. The power generation dispatching benefit is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of reservoir dispatching, and relates to a hydropower station optimization dispatching method based on a total forecast dispatching diagram coupled with medium- and long-term runoff forecast information. Background Art

[0002] The task of medium- and long-term power generation dispatching of hydropower stations is to maximize the power generation benefits while ensuring the completion of the reservoir's end-of-flood water storage task. The decision-making results provide important boundary conditions for the daily dispatching production of hydropower stations and are an important basis for the formulation of daily power generation plans for hydropower stations. The dispatch chart has become an important tool for carrying out medium- and long-term power generation dispatching of hydropower stations because it is easy to understand and easy to operate. With reference to the power generation dispatch chart, decision makers can reasonably determine the power generation output based on the available water volume of the reservoir at each moment, and maximize the power generation benefits while ensuring power generation stability. The basic form of the traditional dispatch chart is as follows: Figure 1 As shown in the figure, it is divided into different zones such as reduced output zone, guaranteed output zone, increased output zone and full output zone by different control lines. The decision is guided by the water level or reservoir capacity at the decision time and the dispatching rules in different zones. Since the runoff forecast information is not considered, it is difficult to give full play to the power generation benefits of the hydropower station by making decisions based on the traditional dispatching chart. In recent years, with the development of runoff forecasting technology, considering runoff forecasting information has become an important means to improve the power generation benefits of hydropower stations. However, affected by many factors, the runoff forecasting information inevitably has uncertainty, and the forecast uncertainty increases with the extension of the forecast period. The forecast information with a long forecast period has important guiding value for the long-term regulation of hydropower stations, but the forecast error is large and the feasibility of direct application is low; while the forecast information with a short forecast period has high accuracy, but due to the short dispatching period of guidance, the operational space is limited. Therefore, how to develop a reasonable coupling utilization method based on the characteristics of different forecast information and give full play to the role of different forecast information is the key to improving the medium- and long-term power generation dispatching benefits of hydropower stations.

[0003] Existing studies have coupled runoff forecast information into the medium- and long-term power generation scheduling of hydropower stations. For example, Xu et al. (2013) improved the basic form of the traditional two-dimensional reservoir power generation scheduling diagram, coupled the ten-day runoff forecast information into the scheduling diagram, and proposed a forecast scheduling diagram model considering medium-term runoff forecast information (Xu W., et al. Evaluation of optimization operation models for cascaded hydropower reservoirs to utilize medium range forecasting inflow, Science China technological sciences, 2013, 56, 2540-2552.), the basic form of which is shown in the attached Figure 2 . Zhang et al. (2019) used the Bayesian stochastic dynamic programming method to couple the runoff forecast information of different forecast periods, and proposed an optimal dispatching method for hydropower stations considering long-, medium- and short-term runoff forecast information (Zhang X., et al. An optimal operation model for hydropower stations considering inflow forecasts with different lead-times, Water resources management, 2019, 33, 173-188.). Feng et al. (2023) used the fuzzy clustering method to classify the samples after the deterministic optimization dispatch was completed, and used the twin SVM to extract the reservoir power dispatching rules under different modes (Feng Z., el al. Deriving hydropower reservoir operation policy using data-driven artificial intelligence model based on pattern recognition and metaheuristic optimizer, Journal of hydrology, 2023, 624, 129916.). However, the above studies either only consider medium-term runoff forecast information and fail to comprehensively consider medium- and long-term forecast information, or use explicit random or implicit random optimization scheduling frameworks to couple multi-scale forecast information, facing obvious "dimensionality disaster" or lack of actual physical meaning of rule extraction targets. This patent combines the advantages of the scheduling diagram and comprehensively considers the characteristics of medium- and long-term runoff forecast information. By using multiple sub-forecast scheduling diagrams to replace the traditional single forecast scheduling diagram, different sub-forecast scheduling diagrams represent different decision spaces, determine the medium-term scheduling decision space based on the long-term runoff forecast classification information, and make specific scheduling decisions based on "reservoir capacity + medium-term inflow", it provides a hydropower station optimization scheduling method based on the total forecast scheduling diagram coupled with medium- and long-term runoff forecast information. Summary of the invention

[0004] In order to solve the problem that the existing medium- and long-term power generation forecast scheduling diagram of a hydropower station considering runoff forecast information cannot be coupled with medium- and long-term runoff forecast information at the same time due to dimensional limitations, thus making it difficult to exert the power generation scheduling benefits of the hydropower station, the present invention provides a method for optimizing the scheduling of a hydropower station based on a total forecast scheduling diagram coupled with medium- and long-term runoff forecast information. This method aims at the situation that the uncertainty of long-term runoff forecast is large at this stage, and there are certain difficulties in directly using the forecast results for reservoir scheduling. The method designs the basic form of the total forecast scheduling diagram, and adopts multiple sub-forecast scheduling diagrams to replace the traditional single forecast scheduling diagram. Different sub-forecast scheduling diagrams represent different decision spaces, and the medium-term scheduling decision space is determined based on the long-term runoff forecast classification information. The specific scheduling decision is made based on "reservoir capacity + medium-term inflow", and the coupling utilization of medium- and long-term runoff forecast information is realized. The Huanren Hydropower Station in Hunjiang River is taken as an example to verify the rationality of the present invention.

[0005] To achieve the above object, the technical solution of the present invention is as follows:

[0006] A method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information. The method first designs the basic form of the general forecast dispatch diagram and the decision-making methods of different partitions in each sub-forecast dispatch diagram it contains; secondly, an optimization model of the general forecast dispatch diagram of the hydropower station is established, that is, the objective function and constraints of the optimization algorithm are determined; thirdly, a multi-year series of simulations are performed on the power generation dispatching process of the hydropower station, and the general forecast dispatch diagram is optimized using a step-by-step optimization algorithm; finally, the optimized general forecast dispatch diagram is used as the dispatching rule for the hydropower station under different long-term runoff forecast levels, and the medium- and long-term runoff forecast information is used as input to calculate the decision-making output of the hydropower station. Specifically, the following steps are included:

[0007] S1. Basic form of design general forecast dispatch diagram and decision-making method:

[0008] According to the dispatching needs of the hydropower station and the runoff characteristics of the basin where it is located, a general forecast dispatching diagram consisting of multiple sub-forecast dispatching diagrams is used to replace the traditional single forecast dispatching diagram, and the number of control lines, regional divisions and power generation output calculation methods of each sub-forecast dispatching diagram in the general forecast dispatching diagram are clarified. Specifically:

[0009] The basic form of the general forecast dispatch diagram designed by the present invention is as shown in the attached Figure 3 As shown, it contains M sub-forecast scheduling diagrams, M represents the number of levels into which the long-term runoff is divided, and its value range is generally 3 to 5. Each sub-forecast scheduling diagram is a power generation scheduling rule under a long-term runoff forecast level. Each sub-forecast scheduling diagram is composed of the highest control line, the increased output line, the guaranteed output line, the reduced output line and the lowest control line. Each control line is composed of control points at different times; the horizontal coordinate of each sub-forecast scheduling diagram is time, and the vertical coordinate is the available water volume OCS t, the available water OCS t It is obtained by converting the reservoir capacity and the time period inflow into the same dimension and then superimposing them. The calculation method is shown in formula (1) and formula (2); each sub-forecast scheduling diagram is divided into a reduced output area, a guaranteed output area, an increased output area and a full output area by the above-mentioned highest control line, an increased output line, a guaranteed output line, a reduced output line and a lowest control line. Specifically: the area below the highest control line and above the increased output line is the full output area, the area below the increased output line and above the guaranteed output line is the increased output area, the area below the guaranteed output line and above the reduced output line is the guaranteed output area, and the area below the reduced output line and above the lowest control line is the reduced output area; for any sub-forecast scheduling diagram in the overall forecast scheduling diagram, such as the mth (1≤m≤M) sub-forecast scheduling diagram, each of its partitions is based on the OCS at the time. t Make power generation decisions. The calculation method is shown in formula (3).

[0010]

[0011] Where, OCS t is the available water at time t; Q t s V is the period flow converted from the reservoir storage at time t; t V is the reservoir capacity corresponding to the water level at time t; dead is the storage capacity corresponding to the dead water level; Δt is the time period length; Q t is the inflow flow in the tth period. In the process of optimizing the total forecast degree graph, Q t Use measured flow, and use forecast flow instead in actual dispatch; N t m N is the decision output of the mth sub-forecast scheduling diagram at the tth moment; e To ensure the output of hydropower station; N M is the installed capacity of the hydropower station; m,0,t , O m,1,t , O m,2,t , O m,3,t , O m,4,t is the control point corresponding to the tth time on each control line in the mth sub-forecast scheduling diagram, specifically: m,0,t represents the control point corresponding to the lowest control line at time t in the mth sub-forecast scheduling diagram; O m,1,t represents the control point corresponding to the reduced output line at time t in the mth sub-forecast scheduling diagram; m,2,t represents the control point corresponding to the output line at time t in the mth sub-forecast dispatch diagram; m,3,t represents the control point corresponding to the increased output line at time t in the mth sub-forecast dispatch diagram; O m,4,tRepresents the control point corresponding to the highest control line in the mth sub-forecast scheduling diagram at time t.

[0012] S2. Establish an optimization model for the overall forecast dispatch diagram of a hydropower station:

[0013] S2.1, using some control points in each sub-forecast dispatch diagram of the total forecast dispatch diagram in S1, namely: reducing the output line, ensuring the output line, and increasing the control points on the output line, as optimization variables, and taking the multi-year average power generation of the hydropower station as the objective function, construct an optimization model for the total forecast dispatch diagram of the hydropower station. The calculation formula of the objective function of the optimization model of the total forecast dispatch diagram of the hydropower station is as follows:

[0014]

[0015] Where E is the multi-year average power generation during the power generation scheduling simulation period; R is the number of years included in the power generation scheduling simulation period; N t is the decision output of the hydropower station in the tth period; c1 and c2 are penalty coefficients, which are greater than 0. t When it is less than the guaranteed output of the hydropower station, the penalty coefficient is used to penalize the power generation; N e To ensure the output of the hydropower station; Δt is the length of a time period; T is the total number of time periods in the scheduling period.

[0016] S2.2. Set basic constraints for hydropower station power generation, including water balance constraints, water storage level constraints, maximum discharge constraints, power generation flow constraints and unit output constraints, as follows:

[0017] ① Water balance constraints:

[0018] V t+1 =V t +(Q t -Q power,t -Q abandon,t )·Δt (5)

[0019] ②Water level constraints:

[0020] Z dead ≤Z t ≤Z normal (6)

[0021] ④ Power generation flow constraints:

[0022]

[0023] ⑤Unit output constraints:

[0024] N min ≤N t ≤N max (8)

[0025] Where V t 、V t+1 Indicates the storage capacity at the beginning and end of time period t; Q t , Q power,t , Q abandon,t represents the inflow, power generation and water abandonment flow in time period t; Z dead , Z normal and Z t They represent the dead water level, normal high water level and initial water level of the reservoir during period t respectively; and Respectively represent the minimum power generation flow and the maximum power generation flow of the hydropower station; N min 、N max and N t They represent the minimum output, maximum output and the decision output of the hydropower station in the tth period respectively.

[0026] S2.3. Set constraints (control line constraints) between different sub-forecast scheduling diagrams in the overall forecast scheduling diagram. In order to ensure that the higher the long-term runoff forecast level, the greater the decision-making output of the overall forecast scheduling diagram; the present invention adds control line constraints between different sub-forecast scheduling diagrams, specifically:

[0027] Assuming that the long-term runoff is divided into M levels from small to large, during the optimization process, the following constraints are added between the sub-forecast scheduling diagrams in the overall forecast scheduling diagram:

[0028] O 1,k,t ≥O 2,k,t ≥…≥O m,k,t ≥…≥O M,k,t (9)

[0029] In the formula, O m,k,t is the control point at time t on the kth control line in the mth sub-forecast scheduling diagram, 0≤k≤4; when the available water volume OCS at time t t At the same time, the constraint of formula (9) can make the decision output of each sub-forecast scheduling diagram in the overall forecast scheduling diagram satisfy the following relationship:

[0030]

[0031] That is: at the same time t and OCS t Under the condition, the decision output of the mth sub-forecast scheduling diagram at time t is greater than the decision output of the m-1th sub-forecast scheduling diagram, and less than the decision output of the m+1th sub-forecast scheduling diagram.

[0032] S3. Conduct a multi-year series of simulations on the power generation dispatching process of the hydropower station, and use a step-by-step optimization algorithm to solve the established optimization model of the total forecast dispatching diagram of the hydropower station:

[0033] S3.1. The forecast dispatch diagram of the hydropower station that only considers the medium-term runoff forecast information, that is, the traditional forecast dispatch diagram, is the initial solution of each sub-forecast dispatch diagram in the general forecast dispatch diagram, which guides the power generation dispatch of the hydropower station during the power generation dispatch simulation period. Specifically: When the general forecast dispatch diagram guides the power generation dispatch decision of the hydropower station at time t, it is necessary to first calculate the long-term runoff forecast level m and the available water volume OCS at that time. t Then select the mth sub-forecast scheduling diagram as the scheduling rule, according to OCS t The decision output of the sub-forecast scheduling diagram is calculated by formula (3): As the decision-making output of the hydropower station N t After the entire power generation scheduling simulation period is over, the objective function value of each total forecast scheduling diagram is calculated according to formula (4), that is, the multi-year average power generation E during the power generation scheduling simulation period; the length of the power generation scheduling simulation period is generally not less than 10 years, which can be specifically determined according to the length of the runoff sequence reserved in a specific basin.

[0034] S3.2. Discretize some control points on each sub-forecast dispatch diagram in the general forecast dispatch diagram at the first moment, namely, the control points on the output line of reducing, guaranteeing and increasing the output line. The number of discretization is L, and its value is generally 3 to 5. Remove the general forecast dispatch diagram that does not meet the constraint conditions shown in formula (9) after discretization to obtain the remaining general forecast dispatch diagrams. Gradually use the remaining general forecast dispatch diagrams to guide the power generation dispatch during the power generation dispatch simulation period of the hydropower station. The specific guidance process is the same as that described in S3.1. After the entire power generation dispatch simulation period is over, the objective function value of each general forecast dispatch diagram is calculated according to formula (4), namely, the multi-year average power generation E, and retain the general forecast dispatch diagram with the largest objective function value.

[0035] S3.3. Taking the total forecast dispatch diagram retained in S3.2 as the initial solution, discretize some of the control points at the second moment in each of its sub-forecast dispatch diagrams, namely, the control points on the output line of reducing the output line, ensuring the output line, and increasing the output line. The discrete number L is the same as that in S3.2. Remove the total forecast dispatch diagrams that do not meet the constraints shown in formula (9) after discretization to obtain the remaining total forecast dispatch diagrams. Use the remaining total forecast dispatch diagrams to guide the power generation dispatch of the hydropower station during the power generation dispatch simulation period. Calculate the objective function value of each total forecast dispatch diagram according to formula (4) and retain the total forecast dispatch diagram with the largest objective function value.

[0036] S3.4, and so on, until the discretization of the control point at the last moment on the general forecast scheduling diagram is completed, that is, a round of iteration is completed. After the completion of this round of iteration, the optimized control point on the general forecast scheduling diagram is:

[0037]

[0038] In the formula, O m,k,tT represents the control point value of the kth control on the mth sub-forecast scheduling diagram at time t in the overall forecast scheduling diagram; D The total number of time periods in the total forecast schedule.

[0039] S3.5. Take the local optimal total forecast dispatch diagram obtained in S3.4 as the initial solution for the new round of iteration, and record the corresponding objective function value of the initial solution as E old Repeat steps S3.2 to S3.4 to complete a new round of iterations and obtain the local optimal total forecast scheduling graph objective function value E after the new round of iterations, and assign it to E new , and compare the latest round of iterative objective function values ​​with the previous round. If E old and E new If the absolute value of the difference is less than the predetermined range, the iteration is stopped and the final optimized total forecast scheduling diagram is output; otherwise, steps S3.2 to S3.4 are repeated. The predetermined range is preferably E old 0.01% to 0.1%.

[0040] S4. Calculate the decision output of the hydropower station based on the medium- and long-term runoff forecast information and the optimized total forecast dispatch diagram:

[0041] S4.1. Based on the long-term runoff grade forecast result, for example, grade m, select the mth sub-forecast scheduling diagram in the overall forecast scheduling diagram as the scheduling rule.

[0042] S4.2. Based on the mid-term runoff forecast results Q t Calculate the current available water volume (OCS) of the reservoir using the current water storage volume of the reservoir t , the specific calculation method is shown in formula (1) and formula (2).

[0043] S4.3. Use the sub-forecast dispatch chart selected in S4.1 as the dispatch rule and the available water OCS calculated in S4.2 t is the decision variable, and N is calculated using formula (3) t m , as the hydropower station decision output N in the tth period t .

[0044] The present invention has the following beneficial effects:

[0045] Traditional forecast dispatch diagram (i.e. Figure 2) can only consider the runoff forecast information of a single forecast period, while the present invention adopts multiple sub-forecast scheduling diagrams to replace the traditional single forecast scheduling diagram. Different sub-forecast scheduling diagrams represent different decision spaces. The medium-term scheduling decision space is determined according to the long-term runoff forecast classification information, and the specific medium-term scheduling decision is made according to "reservoir capacity + medium-term inflow". The coupled utilization of medium- and long-term runoff forecast information is realized, so that decision makers can use the medium- and long-term runoff forecast information at the same time and formulate a more reasonable power generation plan, thereby achieving more efficient utilization of water resources and improving the power generation scheduling efficiency of hydropower stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the basic form of the traditional dispatch diagram without considering the runoff forecast information;

[0047] Figure 2 It is a schematic diagram of the basic form of the traditional forecast dispatch diagram that only considers the medium-term runoff forecast information;

[0048] Figure 3 A schematic diagram of the basic form of the general forecast scheduling diagram designed for the present invention; Figure 3 (a) is a schematic diagram of the basic form of the sub-forecast scheduling diagram when the long-term runoff forecast level is 1; Figure 3 (b) is a schematic diagram of the basic form of the sub-forecast scheduling diagram when the long-term runoff forecast level is m; Figure 3 (c) is a schematic diagram of the basic form of the sub-forecast scheduling diagram when the long-term runoff forecast level is M;

[0049] Figure 4 It is a schematic diagram of the content of the present invention;

[0050] Figure 5 This is the traditional forecast dispatch diagram of Huanren Hydropower Station which only considers the medium-term runoff forecast information;

[0051] Figure 6 Flowchart of the overall forecast scheduling diagram for stepwise optimization algorithm optimization;

[0052] Figure 7 This is a basic diagram of the general forecast dispatch diagram of Huanren Power Station; Figure 7 (a) is a schematic diagram of the basic form of the sub-forecast scheduling diagram when the long-term runoff forecast level is 1; Figure 7 (b) is a schematic diagram of the basic form of the sub-forecast scheduling diagram when the long-term runoff forecast level is 2; Figure 7 (c) is a schematic diagram of the basic form of the sub-forecast scheduling diagram when the long-term runoff forecast level is 3;

[0053] Figure 8 A general forecast dispatch diagram considering medium- and long-term runoff forecast information for Huanren Hydropower Station; Figure 8(a) is the sub-forecast scheduling diagram when the long-term runoff forecast level of Huanren Hydropower Station is 1; Figure 8 (b) is the sub-forecast dispatch diagram when the long-term runoff forecast level of Huanren Hydropower Station is 2; Figure 8 (c) in the figure is the sub-forecast scheduling diagram when the long-term runoff forecast level of Huanren Hydropower Station is 3. DETAILED DESCRIPTION

[0054] In order to make the technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and taking Huanren Hydropower Station as an example. It should be understood that the specific implementation methods described herein are only for explanation and do not limit the scope of protection of the present invention.

[0055] The present invention provides a method for optimizing the dispatching of hydropower stations based on the coupling of medium- and long-term runoff forecast information based on the overall forecast dispatching diagram. Figure 4 As shown, the specific implementation process includes the following five steps from S1 to S5, wherein S1 to S3 introduce the basic form design of the general forecast dispatching diagram of Huanren Power Station and the optimization process of each control point on the general forecast dispatching diagram, step S4 introduces how to make power generation decisions according to the medium- and long-term runoff forecast information at each moment after optimizing the general forecast dispatching diagram, and S5 evaluates the actual application effect of the present invention in Huanren Power Station, as follows:

[0056] S1. According to the dispatching demand of Huanren Power Station and the runoff characteristics of the basin where it is located, the long-term runoff forecast period is 3 months. Therefore, the long-term runoff (forecast period of 3 months) frequency curve is divided into 3 levels from small to large, namely: Level 1, Level 2 and Level 3, representing the "dry", "normal" and "abundant" water conditions respectively. According to the above division, the basic form of the general forecast dispatch diagram designed by the present invention in the application of Huanren Power Station is as shown in the attached figure. Figure 7 As shown in the figure, it includes three sub-forecast dispatching diagrams, that is, M takes the value of 3, which is used to distinguish the water supply status of "dry", "normal" and "abundant". Figure 7 Neutron forecast dispatch diagrams (a), (b), and (c) are used as the hydropower station dispatch rules when the long-term runoff forecast level of Huanren Reservoir is 1, 2, and 3, respectively.

[0057] S2. Establishment of the optimization model of the general forecast dispatch diagram of Huanren Power Station:

[0058] S2.1, with Figure 7 The control points on each sub-forecast dispatch diagram in the general forecast dispatch diagram shown in the figure, namely, the control points on the output line of reducing, guaranteeing and increasing, are used as optimization variables. The objective function is set to the multi-year average power generation of Huanren Power Station. The calculation method is shown in formula (4), where the penalty coefficients c1 and c2 are set to 10 and 4 respectively to guarantee the output N e It is 33MW.

[0059] S2.2. Set constraints, including water balance constraints, water level constraints, maximum discharge constraints, power generation flow constraints, unit output constraints, and constraints between sub-forecast scheduling diagrams in the overall forecast scheduling diagram. For details, see formula (5) to formula (9). The constraint indicators are shown in Table 1.

[0060] Table 1 Constraint index values ​​of Huanren Power Station

[0061]

[0062] S3. Refer to the attached Figure 6 , the established optimization model is solved by using stepwise optimization algorithm. During the optimization process, the values ​​of various parameters are shown in Table 2:

[0063] Table 2 Parameters of stepwise optimization algorithm

[0064]

[0065] S3.1. The forecast dispatch diagram considering the medium-term runoff forecast information of Huanren Power Station, that is, the traditional forecast dispatch diagram, is the initial solution of each sub-forecast dispatch diagram in the overall forecast dispatch diagram, as shown in the attached figure. Figure 5 As shown in the figure, it guides the power generation dispatch of Huanren Power Station from 1968 to 2010. Specifically, when guiding the power generation dispatch decision of Huanren Power Station at time t, it is necessary to first calculate the long-term runoff forecast level m and available water volume OCS with a forecast period of 3 months at that time. t Then select the mth sub-forecast scheduling diagram as the scheduling rule, according to OCS t The decision output of the sub-forecast scheduling diagram is calculated by formula (3): As the decision-making output of Huanren Power Station N t , the entire power generation scheduling simulation period, i.e. 1968 to 2010, after the simulation is completed, the objective function value of each total forecast scheduling diagram is calculated according to formula (4), i.e.: the multi-year average power generation E.

[0066] S3.2. Discretize some control points on each sub-forecast dispatch diagram in the first moment general forecast dispatch diagram, i.e., control points on the output line of reducing, guaranteeing and increasing output lines. The discrete number L is 3. Remove the general forecast dispatch diagram that does not meet the constraint conditions shown in formula (9) after discretization to obtain the remaining general forecast dispatch diagrams. Use the remaining general forecast dispatch diagrams to guide the power generation dispatch of Huanren Power Station from 1968 to 2010. The specific guidance process is the same as that in S3.1. After the whole power generation dispatch simulation period is completed, the objective function value of each general forecast dispatch diagram is calculated according to formula (4), i.e., the multi-year average power generation E. The general forecast dispatch diagram with the largest objective function value is retained.

[0067] S3.3. Based on the master forecast dispatch diagram retained in S3.2, some control points at the second moment in each of its sub-forecast dispatch diagrams, namely, the control points on the output line of reducing, guaranteeing and increasing, are discretized. The discrete number L is 3. The master forecast dispatch diagrams that do not meet the constraint conditions shown in formula (9) are removed after discretization to obtain the remaining multiple master forecast dispatch diagrams. The remaining master forecast dispatch diagrams are gradually used to guide the power generation dispatch of Huanren Power Station from 1968 to 2010. The specific guidance process is the same as that described in S3.1. After the entire power generation dispatch simulation period is completed, the objective function value of each master forecast dispatch diagram is calculated according to formula (4), namely, the multi-year average power generation E. The master forecast dispatch diagram with the largest objective function value is retained.

[0068] S3.4, and so on, until the discretization of each control point at the last moment on the overall forecast scheduling diagram is completed, that is, a round of iteration is completed, and the local optimal overall forecast scheduling diagram is obtained.

[0069] S3.5. Take the local optimal total forecast dispatch diagram obtained in S3.4 as the initial solution for the new round of iteration, and record the corresponding objective function value of the initial solution as E old Repeat steps S3.2 to S3.4 to obtain the local optimal total forecast scheduling graph objective function value after a new round of iteration, denoted as E new , and compare the latest round of iterative objective function values ​​with the previous round. If E old and E new The absolute value of the difference is less than the predetermined range ΔE, where E old If the value is less than 0.01%, the iteration stops and the total forecast scheduling diagram is output; otherwise, steps S3.2 to S3.4 are repeated.

[0070] After optimization, the overall forecast dispatch diagram of Huanren Power Station considering medium- and long-term runoff forecast information is obtained, as shown in the attached figure. Figure 8 shown.

[0071] S4. Based on the medium- and long-term runoff forecast information at each time, determine the decision output N of Huanren Power Station in the tth period according to the total forecast dispatch diagram obtained in S3.5 t :

[0072] S4.1. Carry out long-term runoff grade forecast for Huanren Power Station with a forecast period of 3 months. According to the long-term runoff grade forecast result, for example, if the grade forecast result is m, the mth sub-forecast scheduling diagram in the overall forecast scheduling diagram is selected as the scheduling rule for the tth period.

[0073] S4.2. Carry out mid-term ten-day runoff forecast for Huanren Hydropower Station. Based on the mid-term runoff forecast results, Q t and the current water storage capacity of the reservoir V t Calculate the current available water volume (OCS) of Huanren Power Station t, the specific calculation formula is (1) and (2).

[0074] S4.3. Use the sub-forecast dispatch chart selected in S4.1 as the dispatch rule and the available water OCS calculated in S4.2 t is the decision variable, and the decision output calculated by formula (3) is used as the decision output of Huanren Power Station. For Huanren Power Station, its guaranteed output N e The installed capacity is 33MW. M It is 222MW.

[0075] S5. Taking 2011 to 2020 as the verification period, the water level of Huanren Power Station is set to 297m. The optimized general forecast dispatching diagram considering medium and long-term runoff forecast information, that is, the general forecast dispatching diagram proposed by the present invention, is the dispatching rules under different long-term runoff forecast levels. The runoff forecast of each time of Huanren Power Station and the long-term runoff level forecast with a forecast period of 3 months are used as inputs to simulate the power generation dispatching of Huanren Power Station from 2011 to 2020 to evaluate the actual application effect of the present invention in Huanren Power Station. Table 3 shows the multi-year average power generation and guarantee rate of power generation dispatching of Huanren Power Station from 2011 to 2020 when different hydropower station optimization dispatching methods are used. It can be seen that compared with the forecast dispatching diagram that only considers the medium-term runoff forecast information, that is, the traditional forecast dispatching diagram, after using the general forecast dispatching diagram proposed by the present invention to guide power generation dispatching, the multi-year average power generation of Huanren Power Station can be increased by 6.69 million kWh, and the power generation guarantee rate can be increased by 0.6%. The above results show that the hydropower station optimization scheduling method based on the total forecast scheduling diagram coupled with medium- and long-term runoff forecast information proposed in the present invention not only improves the reliability of hydropower station power generation, but also brings greater power generation benefits in the medium- and long-term power generation scheduling application of Huanren Power Station.

[0076] Table 3 Comparison of dispatch simulation results of Huanren Power Station using different methods from 2011 to 2020

[0077]

[0078] The above-described embodiment is only a specific implementation of the present invention and cannot be regarded as limiting the scope of the patent right of the present invention. It should be emphasized that technical professionals in this field may make several improvements without departing from the basic principles of the present invention, and these improvements are included in the scope of protection of the present invention.

Claims

1. A method for optimizing the dispatching of hydropower stations based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information, characterized in that: The hydropower station optimization dispatching method first designs the basic form of the general forecast dispatching diagram and the decision-making methods of different partitions in each sub-forecast dispatching diagram contained therein; secondly, establishes the optimization model of the general forecast dispatching diagram of the hydropower station, and determines the objective function and constraint conditions of the optimization algorithm; Third, a multi-year series simulation of the hydropower station power generation dispatching process is carried out, and the general forecast dispatching diagram is optimized by using a step-by-step optimization algorithm; finally, the optimized general forecast dispatching diagram is used as the hydropower station dispatching rule under different long-term runoff forecast levels, and the medium- and long-term runoff forecast information is used as input to calculate the decision-making output of the hydropower station; the specific steps include: S1. Design the basic form of the general forecast dispatch diagram and the decision-making method; S2. Establish an optimization model for the overall forecast dispatch diagram of a hydropower station; S3. Conduct a multi-year series simulation of the power generation dispatching process of the hydropower station, and use a step-by-step optimization algorithm to solve the established optimization model of the total forecast dispatching diagram of the hydropower station; S4. Calculate the decision-making output of the hydropower station based on the medium- and long-term runoff forecast information and the optimized total forecast dispatch diagram.

2. A method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 1, characterized in that: The S1 is specifically: according to the scheduling needs of the hydropower station and the runoff characteristics of the basin where it is located, a general forecast scheduling diagram composed of multiple sub-forecast scheduling diagrams is used to replace the traditional single forecast scheduling diagram, and the number of control lines, regional divisions and power generation output calculation methods of each sub-forecast scheduling diagram in the general forecast scheduling diagram are clarified.

3. The method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 2, characterized in that: The general forecast dispatch diagram includes M sub-forecast dispatch diagrams, where M represents the number of levels into which the long-term runoff is divided, and each sub-forecast dispatch diagram is a power generation dispatch rule under a long-term runoff forecast level; each sub-forecast dispatch diagram is composed of a maximum control line, an increase output line, a guarantee output line, a reduction output line and a minimum control line, and each control line is composed of control points at different times; The horizontal axis of each sub-forecast dispatch diagram is time, and the vertical axis is available water volume OCS t , the available water OCS t It is obtained by converting the reservoir capacity and the time period inflow into the same dimension and then superimposing them. The calculation method is shown in formula (1) and formula (2); each sub-forecast dispatch diagram is divided into a reduced output area, a guaranteed output area, an increased output area and a full output area by the highest control line, an increased output line, a guaranteed output line, a reduced output line and a lowest control line; For any sub-forecast schedule in the overall forecast schedule, such as the mth sub-forecast schedule, each partition is based on the OCS at the time t Make power generation decisions. The calculation method is shown in formula (3), where 1≤m≤M; Where, OCS t is the available water at time t; V is the period flow converted from the reservoir storage at time t; t V is the reservoir capacity corresponding to the water level at time t; dead is the storage capacity corresponding to the dead water level; Δt is the time period length; Q t is the inflow flow in the tth period. In the process of optimizing the total forecast degree graph, Q t The measured flow is used, and in actual dispatch, the forecast flow is used instead; N is the decision output of the mth sub-forecast scheduling diagram at the tth moment; e To ensure the output of hydropower station; N M is the installed capacity of the hydropower station; m,0,t , O m,1,t , O m,2,t , O m,3,t , O m,4,t is the control point corresponding to the tth time on each control line in the mth sub-forecast scheduling diagram, specifically: m,0,t represents the control point corresponding to the lowest control line at time t in the mth sub-forecast scheduling diagram; O m,1,t represents the control point corresponding to the reduced output line at time t in the mth sub-forecast scheduling diagram; O m,2,t represents the control point corresponding to the output line at time t in the mth sub-forecast dispatch diagram; m,3,t represents the control point corresponding to the increased output line at time t in the mth sub-forecast dispatch diagram; O m,4,t Represents the control point corresponding to the highest control line in the mth sub-forecast scheduling diagram at time t.

4. A method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 3, characterized in that: The value range of M is 3-5.

5. The method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 3, characterized in that: The rules for dividing the sub-forecast scheduling diagrams into reduced output areas, guaranteed output areas, increased output areas and full output areas are as follows: the area below the highest control line and above the increased output line is the full output area, the area below the increased output line and above the guaranteed output line is the increased output area, the area below the guaranteed output line and above the reduced output line is the guaranteed output area, and the area below the reduced output line and above the lowest control line is the reduced output area.

6. A method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 3, characterized in that: The S2 is specifically: S2.

1. Taking some control points in each sub-forecast dispatch diagram of the total forecast dispatch diagram in S1 as optimization variables and the multi-year average power generation of the hydropower station as the objective function, an optimization model of the total forecast dispatch diagram of the hydropower station is constructed; the calculation formula of the objective function of the optimization model of the total forecast dispatch diagram of the hydropower station is as follows: Where E is the multi-year average power generation during the power generation scheduling simulation period; R is the number of years included in the power generation scheduling simulation period; N t is the decision output of the hydropower station in the tth period; c1 and c2 are penalty coefficients, which are greater than 0. t When it is less than the guaranteed output of the hydropower station, the penalty coefficient is used to penalize the power generation; N e To ensure the output of the hydropower station; Δt is the length of a time period; T is the total number of time periods in the dispatch period; S2.

2. Set basic constraints for power generation at the hydropower station, including water balance constraints, water storage level constraints, maximum discharge constraints, power generation flow constraints and unit output constraints; S2.

3. Set control line constraints between different sub-forecast scheduling diagrams in the general forecast scheduling diagram; in order to ensure that the higher the long-term runoff forecast level, the greater the decision-making output of the general forecast scheduling diagram, add control line constraints between different sub-forecast scheduling diagrams: Assuming that the long-term runoff is divided into M levels from small to large, during the optimization process, the following constraints are added between the sub-forecast scheduling diagrams in the overall forecast scheduling diagram: THE 1,k,t ≥O 2,k,t ≥…≥O m,k,t ≥…≥O M,k,t (9) In the formula, O m,k,t is the control point at time t on the kth control line in the mth sub-forecast scheduling diagram, 0≤k≤4; when the available water volume OCS at time t t At the same time, the constraint of formula (9) can make the decision output of each sub-forecast scheduling diagram in the overall forecast scheduling diagram satisfy the following relationship: At the same time t and OCS t Under the condition, the decision output of the mth sub-forecast scheduling diagram at time t is greater than the decision output of the m-1th sub-forecast scheduling diagram, and less than the decision output of the m+1th sub-forecast scheduling diagram.

7. A method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 6, characterized in that: The basic constraints for power generation of the S2.2 hydropower station are as follows: ① Water balance constraints: 5 t+1 =V t +(Q t -Q power,t -Q abandon,t )·Δt (5) ②Water level constraints: WITH dead ≤Z t ≤Z normal (6) ④ Power generation flow constraints: ⑤Unit output constraints: N min ≤N t ≤N max (8) Where V t 、V t+1 Indicates the storage capacity at the beginning and end of time period t; Q t , Q power,t , Q abandon,t represents the inflow, power generation and water abandonment flow in time period t; Z dead , Z normal and Z t They represent the dead water level, normal high water level and initial water level of the reservoir during period t respectively; and Respectively represent the minimum power generation flow and the maximum power generation flow of the hydropower station; N min 、N max and N t They represent the minimum output, maximum output and the decision output of the hydropower station in the tth period respectively.

8. The method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 7, characterized in that: The S3 is specifically: S3.

1. The traditional forecast dispatch diagram of the hydropower station that only considers the medium-term runoff forecast information is used as the initial solution of each sub-forecast dispatch diagram in the general forecast dispatch diagram to guide the power generation dispatch of the hydropower station during the power generation dispatch simulation period. Specifically: When the general forecast dispatch diagram guides the power generation dispatch decision of the hydropower station at time t, it is necessary to first calculate the long-term runoff forecast level m and the available water volume OCS at that time. t Then select the mth sub-forecast scheduling diagram as the scheduling rule, according to OCS t The decision output N of the sub-forecast scheduling diagram is calculated by formula (3): t m As the decision-making output of the hydropower station N t , after the entire power generation scheduling simulation period is over, the objective function value of each total forecast scheduling diagram is calculated according to formula (4), and the multi-year average power generation E of the power generation scheduling simulation period; S3.

2. Discretize some control points on each sub-forecast dispatch diagram in the first moment total forecast dispatch diagram, the number of discrete points is L, and remove the total forecast dispatch diagram that does not meet the constraint conditions shown in formula (9) after discretization, and obtain the remaining multiple total forecast dispatch diagrams. Gradually use the remaining total forecast dispatch diagrams to guide the power generation dispatch during the power generation dispatch simulation period of the hydropower station. The specific guidance process is the same as that described in S3.

1. After the entire power generation dispatch simulation period is over, calculate the objective function value of each total forecast dispatch diagram according to formula (4), the multi-year average power generation E, and retain the total forecast dispatch diagram with the largest objective function value; S3.3, using the total forecast dispatch diagram retained in S3.2 as the initial solution, discretize some control points at the second moment in each of its sub-forecast dispatch diagrams, remove the total forecast dispatch diagrams that do not meet the constraint conditions shown in formula (9) after discretization, and obtain the remaining multiple total forecast dispatch diagrams, use the remaining total forecast dispatch diagrams to guide the power generation dispatch during the power generation dispatch simulation period of the hydropower station, calculate the objective function value of each total forecast dispatch diagram according to formula (4) and retain the total forecast dispatch diagram with the largest objective function value; S3.4, and so on, until the discretization of the control point at the last moment on the general forecast scheduling diagram is completed, that is, a round of iteration is completed. After the completion of this round of iteration, the optimized control point on the general forecast scheduling diagram is: In the formula, O m,k,t T represents the control point value of the kth control on the mth sub-forecast scheduling diagram at time t in the overall forecast scheduling diagram; D The total number of time periods of the total forecast dispatch diagram; S3.

5. Take the local optimal total forecast dispatch diagram obtained in S3.4 as the initial solution for the new round of iteration, and record the corresponding objective function value of the initial solution as E old Repeat steps S3.2 to S3.4 to complete a new round of iterations and obtain the local optimal total forecast scheduling graph objective function value E after the new round of iterations, and assign it to E new , and compare the latest round of iterative objective function values ​​with the previous round. If E old and E new If the absolute value of the difference is less than the predetermined range, the iteration is stopped and the final optimized total forecast scheduling diagram is output; otherwise, steps S3.2 to S3.4 are repeated.

9. A method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 8, characterized in that: In the S3: In S3.1, the length of the power generation scheduling simulation period is not less than 10 years, which is determined according to the length of the runoff sequence reserved in the specific basin; In S3.2, the discrete number L is 3 to 5; In S3.3, the discrete quantity value is the same as S3.2; In S3.5, the predetermined range may be old 0.01% to 0.1%.

10. The method for optimizing the dispatching of a hydropower station based on a general forecast dispatch diagram coupled with medium- and long-term runoff forecast information according to claim 8, characterized in that: The S4 is specifically: S4.

1. According to the long-term runoff grade forecast result, such as grade m, select the mth sub-forecast scheduling diagram in the overall forecast scheduling diagram as the scheduling rule; S4.

2. Based on the mid-term runoff forecast results Q t Calculate the current available water volume (OCS) of the reservoir using the current water storage volume of the reservoir t , the specific calculation method is shown in formula (1) and formula (2); S4.

3. Use the sub-forecast dispatch chart selected in S4.1 as the dispatch rule and the available water OCS calculated in S4.2 t is the decision variable, and N is calculated using formula (3) t m , as the hydropower station decision output N in the tth period t .

Images