Reservoir optimization scheduling method and device, computer scale storage medium and program product
The reservoir scheduling model is optimized by using the isosurface translation method, which solves the problems of computational expansion and low iteration efficiency in existing technologies, and achieves efficient and stable reservoir optimization scheduling, applicable to multiple time periods and multiple constraints.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-31
- Publication Date
- 2026-04-03
AI Technical Summary
Existing reservoir optimization scheduling methods suffer from increased computational scale and high computational complexity when dealing with mixed-type constraints. The simplex method suffers from low optimization efficiency due to increased iteration count, and the selection of principal components lacks clear engineering orientation.
The isosurface translation method is adopted. By constructing a linear programming model without introducing artificial variables, and combining the geometric advancement idea of the isosurface of the objective function, the principal components with consistent signs are selected and rotated to optimize the iteration direction, avoid model size expansion, and maintain the consistency of decision dimensions and physical constraints.
It improves the solution efficiency and engineering applicability of reservoir optimization scheduling, reduces computational complexity, and ensures computational stability and accuracy under multiple time periods and multiple constraints.
Smart Images
Figure CN121787848A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of water resource system scheduling and operation optimization technology, and relates to an isosurface translation method for linear programming models for reservoir optimization scheduling. More specifically, it relates to a reservoir optimization scheduling method, device, computer scale storage medium, and program product. Background Technology
[0002] Optimal reservoir scheduling is a core issue in the operation of water resource systems. Its essence can generally be reduced to coordinating and optimizing objectives such as power generation, flood control, and water supply under multiple operational constraints. In engineering practice, reservoir scheduling problems are often modeled as linear programming models, or transformed into equivalent linear programming models by piecewise linearizing the nonlinear relationships such as reservoir capacity-water level and head-output, thereby achieving a unified solution while maintaining the scheduling constraint structure.
[0003] The simplex method is a classic and widely used approach for solving linear programming problems. However, it requires the problem to be in standard form and have an initial basic feasible solution. When dealing with mixed-type constraints commonly encountered in reservoir scheduling, artificial variables are often introduced to construct the initial basis. While this approach is generally applicable, it often significantly increases the number of variables and the scale of constraints in reservoir scheduling scenarios, thereby expanding the computational scale and significantly increasing computational complexity, especially when the scheduling cycle is long or there are many constraints. Existing methods can reduce the use of artificial variables to address the scale expansion caused by their introduction, but their efficiency remains limited in complex reservoir scheduling models. Furthermore, in terms of principal component selection, the simplex method typically makes decisions based on the ratio of the maximum to minimum test numbers, failing to adequately consider the differences in the overall objective function improvement of different candidate directions. This can easily lead to low optimization efficiency, increased iteration counts, and consequently, negatively impact the overall performance of the scheduling computation. Summary of the Invention
[0004] To address the aforementioned deficiencies and improvement needs of existing technologies, this invention provides a reservoir optimization scheduling method, apparatus, computer-calibrated storage medium, and program product. Its purpose is to improve solution efficiency and engineering applicability while ensuring the simplicity of the reservoir scheduling algorithm.
[0005] To achieve the above objectives, according to one aspect of the present invention, a reservoir optimization scheduling method is provided, comprising: Obtain various hydrological data of the reservoir during the scheduling cycle, construct an optimized scheduling model, and represent the optimized scheduling model of the reservoir as a linear programming model. The linear programming model includes a scheduling model in which the decision variables, objective function and constraints themselves are linearly related, or an equivalent linear programming model obtained after piecewise linearization of the reservoir capacity-water level relationship and the head-output relationship. The linear programming model is solved using an isosurface translation solution mechanism, specifically including: S1: Canonical Model The linear programming model for optimal reservoir scheduling is as follows:
[0006]
[0007] Where Z represents the overall benefit index of reservoir operation and scheduling; x is the scheduling decision variable vector, whose elements represent the operation decision quantities for each time period; and c is the corresponding benefit weight coefficient vector.
[0008] This indicates the various upper bound constraints that must be met during the operation of the reservoir. The constraint coefficient matrix, For the corresponding running boundary; This represents the physical constraints, such as water balance, that a reservoir system must satisfy. This indicates the various lower bound constraints that must be met during the operation of the reservoir; The constraints of the linear programming model for reservoir optimization scheduling are transformed into a form less than or equal to, and the initial basis of the scheduling model is constructed without introducing artificial variables. S2: Optimality Judgment Calculate the test numbers corresponding to each scheduling decision variable based on the current scheduling basis solution, and determine whether the current scheduling scheme satisfies the optimality condition or has a feasible improvement direction according to the linear programming discrimination criterion; S3: Determine the principal component While maintaining the feasibility of reservoir scheduling constraints, the rotating principal components used to update the scheduling basis solution are determined by comprehensively considering the feasible domain boundary constraints and the degree of improvement of the scheduling objective function isosurface. S4: Rotation Transformation Perform a rotation transformation centered on the pivot element, update the scheduling basis variables, and obtain a new reservoir scheduling basis solution; S5: Repeat steps S2 to S4 until a reservoir scheduling scheme that satisfies the optimality condition is obtained.
[0009] Furthermore, in step S1, a sign consistency judgment is introduced. When the reservoir water balance constraint or other constraints appear in the form of an equation, the initial scheduling basis solution is constructed using any of the following methods: (a) Perform equivalent splitting, splitting into an equivalent greater than or equal to constraint and a less than or equal to constraint, and process them separately; (b) For water balance type equality constraints, in combination with the characteristics of reservoir optimal scheduling, under the premise of ensuring that the physical meaning of scheduling remains unchanged, the water balance type equality constraints are equivalently expressed as constraints in the form of less than or equal to.
[0010] Furthermore, when all test numbers All constraint coefficients in the constraints If the answer is no, then the problem is determined to have no solution; where, For the constant term of the scheduling constraints, These are the constraint coefficients for scheduling decision variables.
[0011] Furthermore, in step S3, only if the constant term of the scheduling constraint... Constraint coefficients of scheduling decision variables and corresponding test numbers The three symbols are consistent and , Only when the value is not zero will the corresponding element be considered as a candidate for scheduling principal and participate in principal determination.
[0012] Further, in step S3, when all the test numbers If the current problem has reached its optimal state, then a feasibility test is performed on the scheduling constraints; if a constant term exists in the scheduling constraints... If the current solution vector does not meet the scheduling feasibility requirements, then the scheduling system will remove the check number when performing the principal component selection. Strict non-zero requirement, i.e., selection of test number The columns are used as the principal selection columns to drive the solution vector to adjust in the direction that satisfies the scheduling constraints.
[0013] Furthermore, in step S3, by constructing an isosurface function value increment index that reflects the degree of improvement of the scheduling objective function, the candidate element with the largest function value increment is selected as the pivot element in the global scope, thereby guiding the isosurface of the reservoir scheduling objective function to shift in the direction of improvement.
[0014] Furthermore, when the current reservoir scheduling fundamental solution is a degraded solution and does not meet the optimality condition, the test number is first selected. In the column, the test number Calculate the minimum positive intercept of the largest column. The element corresponding to the minimum positive intercept is selected as the pivot, and rotation updates are performed using this pivot; where... For the constant term of the scheduling constraints, These are the constraint coefficients for scheduling decision variables.
[0015] To achieve the above objectives, according to another aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the reservoir optimization scheduling method as described in any of the preceding claims.
[0016] To achieve the above objectives, according to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of reservoir optimization scheduling as described in any of the preceding claims.
[0017] To achieve the above objectives, according to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the steps of the reservoir optimization scheduling method as described in any of the preceding claims.
[0018] In general, compared with the prior art, the present invention can achieve the following beneficial effects: 1. This invention proposes an isosurface translation method for linear programming models of reservoir optimization scheduling. Based on the geometric advancement idea of the objective function isosurface, it comprehensively considers the feasible region boundary position and the increment of the isosurface function value during the principal component selection stage, selecting a more efficient iterative direction from the perspective of global objective improvement. Simultaneously, through model normalization and feasibility judgment mechanisms that do not introduce artificial variables, the solution process is made closer to the original constraint structure of reservoir scheduling, improving solution efficiency and engineering applicability while ensuring algorithm simplicity.
[0019] 2. This invention completes the initial basis construction of a linear programming model for reservoir optimization scheduling without introducing artificial variables, avoiding model expansion and maintaining consistency between scheduling decision variables and engineering physical meaning. Simultaneously, by introducing the concept of target isosurface translation and a principal component selection strategy based on minimum positive intercept and test number, the iterative process has a clear objective for improvement and effectively constrains the iteration direction, avoiding unfavorable situations such as unbounded growth. This method improves the directional controllability and solution efficiency of reservoir scheduling problems without requiring strict feasibility throughout the entire process, and is suitable for reservoir optimization scheduling engineering applications under multi-time period and multi-constraint conditions.
[0020] 3. The isosurface translation solution mechanism does not require the introduction of artificial variables in the process of solving the reservoir scheduling model, thereby reducing the scale of scheduling calculations and improving the computational efficiency and stability of large-scale or multi-period reservoir scheduling problems. Attached Figure Description
[0021] Figure 1 This is a flowchart of the isosurface translation method constructed according to a preferred embodiment of the present invention.
[0022] Figure 2This is a flowchart illustrating the determination of the solution in the optimality judgment in a preferred embodiment of the present invention.
[0023] Figure 3 This is a diagram of the ten-library structure in a preferred embodiment of the present invention. Detailed Implementation
[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.
[0025] To address the practical needs regarding the solution efficiency and stability of linear programming models formed during reservoir optimization scheduling, especially under constraints such as multi-period water balance, reservoir capacity boundaries, and outflow, the simplex method suffers from problems such as reliance on artificial variables, complex initial basis construction, and lack of clear engineering direction in principal component selection. This invention provides a linear programming solution method based on the idea of isosurface translation. This method aims to directly construct the initial basis without introducing artificial variables and guide the iterative process steadily along the direction of improving the scheduling objective through the geometric advancement mechanism of the target isosurface, thereby improving the solution efficiency of reservoir optimization scheduling problems.
[0026] Specifically, this invention provides a standardization method applicable to linear programming models for reservoir optimal scheduling. This method processes different types of constraints in the scheduling model, transforming "greater than or equal to" constraints into "less than or equal to" forms through symbol transformation. Equality constraints are decomposed into two equivalent inequality constraints, and slack variables are introduced for each. Alternatively, for equality constraints formed by water balance relationships or operating conditions, considering the characteristics of reservoir optimal scheduling, they can be equivalently represented as inequality constraints suitable for standardization without changing the physical meaning of the scheduling. Through these processes, an initial basis suitable for iteration can be directly constructed within the original scheduling decision variable space without introducing artificial variables, thus maintaining the decision dimension of the reservoir scheduling model unchanged and avoiding the problem of model size expansion caused by the introduction of artificial variables.
[0027] Preferably, the present invention also provides a principal component selection mechanism based on target isosurface translation. This mechanism interprets the solution process of the linear programming for optimal reservoir scheduling as a gradual translation process of the isosurface of the scheduling objective function. By calculating the minimum positive intercept in each potential adjustment direction and combining the test number to evaluate the gain of the corresponding direction on the scheduling objective (such as power generation benefits, water supply security, or comprehensive benefits), the direction with the greatest improvement in the scheduling objective is preferentially selected as the input variable, so that each iteration corresponds to a scheduling scheme improvement with clear engineering significance.
[0028] Based on the aforementioned principal component selection mechanism, this invention also provides a principal component selection condition for reservoir optimization scheduling problems, used to constrain the iterative adjustment direction of decision variables. This condition filters potential principal components based on the sign relationship between constraint coefficients, right-hand side constants, and test numbers, allowing only those with consistent signs to participate in iterative updates. In the reservoir optimization scheduling model, the constraint coefficients... With the constant term on the right-hand side The consistent sign of the decision variables corresponds to the direction of change, which can improve the scheduling objective function, but is also inevitably limited by physical constraints such as reservoir capacity limits, water balance, or discharge capacity, thus forming a finite and reachable scheduling boundary when calculating the minimum positive intercept. The constraint coefficients With test number Consistent sign usage can effectively avoid selecting iterative directions that could lead to infinite growth of decision variables, uncontrolled storage capacity, or unbounded expansion of the objective function.
[0029] In a preferred embodiment, the isosurface translation method proposed in this invention mainly includes the following: 1. Standardization Processing Construct and standardize a linear programming model for optimal reservoir scheduling:
[0030]
[0031] Where Z represents the overall benefit index of reservoir operation and scheduling; x is the scheduling decision variable vector, whose elements represent the operation decision quantities for each time period; and c is the corresponding benefit weight coefficient vector.
[0032] This indicates the various upper bound constraints that must be met during the operation of the reservoir. The constraint coefficient matrix, For the corresponding running boundary; This represents the physical constraints, such as water balance, that a reservoir system must satisfy. This indicates the various lower bound constraints that must be met during the operation of the reservoir; The model optimizes scheduling decisions while ensuring operational safety and physical balance.
[0033] Convert all constraints to "less than or equal to" constraints. The specific processing logic is as follows:
[0034] Compared to the simplex method, this approach offers significant advantages in solving reservoir optimization scheduling problems. The simplex method typically introduces artificial variables to construct initial basic feasible solutions. While this approach formally guarantees feasibility, it introduces additional decision variables and constraint columns, thus altering the dimensional structure of the original model. In high-dimensional engineering applications such as reservoir optimization scheduling, this dimensional expansion not only increases computational and storage burdens but may also weaken the correspondence between original decision variables and physical constraints, affecting the model's engineering interpretability. In contrast, this method avoids introducing artificial variables, thus preserving the decision dimension and constraint structure of the reservoir scheduling model. This characteristic allows the algorithm to avoid unnecessary scale expansion in multi-time-period, multi-constraint reservoir optimization scheduling problems, effectively reducing computational complexity while maintaining solution accuracy, and improving overall solution efficiency and engineering applicability.
[0035] 2. Optimality Judgment In the solution process of this invention, the criteria for determining a unique optimal solution, infinitely many optimal solutions, and unbounded solutions are consistent with those for linear programming problems. The mechanism for determining no solution is as follows: when the test numbers corresponding to all decision variables... If at this time there exists This situation indicates that there are no further adjustment directions to improve the scheduling objective under the current scheduling state, but some constraints have not yet been satisfied. In this case, instead of directly determining that the model has no solution, we choose... This row satisfies sign consistency and the increment of the isosurface function value. The largest element is the principal component, and the process continues iterating to gradually improve the engineering feasibility of the scheduling scheme. And when all... But it exists And all constraint coefficients in this row If the condition is met, it indicates that under the current hydrological conditions and operational constraints, adjusting any decision variable cannot restore the water balance or reservoir capacity to a physically feasible range for that period, meaning there is no feasible direction for scheduling and repair. In this case, the original reservoir optimal scheduling linear programming model is deemed to have no feasible solution. See details... Figure 2 .
[0036] 3. Principal Component Selection Mechanism for Isosurface Translation If the current solution does not fall into any of the above four categories, the following method is used to select the pivot element and perform a rotation transformation.
[0037] (1) Calculation of minimum positive intercept: Traverse each column of the decision variables Only when the sign consistency judgment is satisfied (i.e., the constant term corresponding to the constraint) Constraint coefficients Test number The three symbols are the same, and Calculate the ratio from the corresponding elements of )
[0038] The smallest ratio is taken as the minimum positive intercept of the variable, which is the minimum feasible step size that can be adjusted under the current scheduling state.
[0039] (2) Calculation of isosurface function value increment: For the minimum feasible step size corresponding to each decision variable, calculate its isosurface function value increment, that is, the gain of the scheduling objective function in that adjustment direction:
[0040] in This indicates the degree of improvement in scheduling objectives (such as increased power generation efficiency or increased overall revenue) that can be achieved when the corresponding decision variable is adjusted to the constraint boundary in the current direction, under the condition of satisfying engineering constraints.
[0041] Select The maximum value, i.e., the point with the largest objective function gain among all feasible propagation directions, is selected as the principal component of this rotation transformation. If multiple points correspond to... If the values are equal and both are maximum, then calculate their corresponding minimum positive intercepts. Then, the test numbers for this column were... Divide by the constraint coefficient corresponding to the minimum positive intercept Obtain the unit test number This is to improve the efficiency of the objective function value.
[0042] In the selection of principal components, this method no longer relies solely on the size of the test number, but instead comprehensively evaluates the selection process from an engineering perspective that progressively advances the isosurface of the reservoir's optimal scheduling objective. By introducing a sign consistency criterion, geometrically speaking, when the right-hand side is constrained... With constraint coefficient When the numbers are the same, This corresponds to the minimum step size that, under the current scheduling state, the decision variable adjusted along this direction can first reach the physical constraints such as the upper and lower limits of reservoir capacity, water balance, or discharge capacity. This direction is constrained and achievable in engineering. And when... and When the signs are the same, it means that shifting along that direction can improve the value of the scheduling objective function (such as power generation benefits or comprehensive operating benefits), thus avoiding unbounded growth caused by equivalent expansion in the "wrong direction".
[0043] (3) Handling of Degenerate Situations: During the principal component selection process, if all minimum positive intercepts satisfying the sign consistency judgment are 0, it indicates that in the current scheduling state, the adjustable margins of multiple engineering constraints in the corresponding adjustment directions have been exhausted. That is, adjustments along any decision variable direction cannot bring about actual improvement to the scheduling objective, and the scheduling scheme enters a degenerate state. To address this situation, the present invention introduces the following processing strategy: First, select the test number. In the column, the test number Calculate the minimum positive intercept of the largest column. The element corresponding to the smallest positive intercept is selected as the pivot. Rotation updates are performed using this pivot, which can effectively escape the degenerate cycle, avoid repeated oscillations between the same basic variable combinations, and maintain the convergence of the algorithm.
[0044] (4) Feasibility recovery: when all test numbers This means that when the current problem has reached its optimal state, a feasibility check is performed on the scheduling constraints. If a constant term exists in the scheduling constraints... If the current solution vector does not meet the scheduling feasibility requirements, then the scheduling system removes the strict non-zero requirement for the test number when performing the principal component selection, meaning it can choose the test number. The columns are used as the principal selection columns to drive the solution vector to adjust in the direction that satisfies the scheduling constraints.
[0045] After performing a rotation transformation on the selected principal component, repeat the optimality judgment steps until the final optimal solution is obtained or the problem is determined to be unsolvable.
[0046] By using the aforementioned isosurface translation principal component selection mechanism, this invention interprets the iterative process of reservoir optimal scheduling linear programming as a gradual adjustment process of the scheduling scheme within the engineering constraint boundary. Without introducing artificial variables, it takes into account both the direction of scheduling target improvement and engineering constraint limitations, effectively improving the stability and efficiency of solving multi-time period, multi-constraint reservoir scheduling problems.
[0047] To verify the superiority of the reservoir optimization scheduling algorithm of this invention, a ten-reservoir scheduling model with maximizing power generation efficiency as the objective function, as described in the literature "Constrained Differential Dynamic Programming and Its Application to Multireservoir Control," is selected as a demonstration case (see details). Figure 3This case study involves multiple reservoirs and multiple time periods. Considering constraints such as water balance, reservoir capacity, and outflow, it is a typical large-scale mixed-constraint linear programming problem. Because this example involves multiple reservoirs and multiple time periods with operating parameters, the relevant input data is large in scale. This paper does not elaborate on the specific numerical values, but only retains the model structure and constraint forms, and verifies the effectiveness of the proposed method through comparison of calculation results. The model is as follows:
[0048]
[0049] in, Indicates total power generation benefits. Indicates the amount of electricity generated during a given period. This indicates the power output coefficient of the hydropower station. express Time-of-use electricity price coefficient express The power generation flow rate during a given time period is the decision variable. This indicates the head of the water used for power generation.
[0050] st
[0051]
[0052]
[0053] In the formula, This represents the water storage at the end of the time period. This represents the initial water storage volume for the period. Water supply during specific time periods For power generation flow, For the discharge flow rate, For time intervals; This represents the reservoir capacity corresponding to the minimum water level during that period. This represents the reservoir capacity corresponding to the maximum water level during that period. This represents the maximum power generation flow rate.
[0054] Based on the standardization processing method proposed in this invention, the ten-database optimization scheduling model is standardized and solved using the isosurface translation method. The calculation results are compared with those of the simplex method to verify the effectiveness of the method of this invention. See the table below for details.
[0055]
[0056] The calculation results show that, under the same ten-reservoir optimization scheduling example, the isosurface translation method and the simplex method proposed in this invention achieve completely consistent optimal scheduling efficiency, indicating that the method of this invention maintains consistency with the classical method in terms of solution accuracy and optimality. Furthermore, the method of this invention demonstrates significant advantages in terms of the number of iterations and computation time, reducing the number of iterations from 243 in the traditional simplex method to 152, and the average computation time from 536 ms to 99 ms. These results demonstrate that, by introducing the concept of target isosurface translation, the method of this invention significantly reduces the number of iterations, lowers computational complexity, and improves solution efficiency while maintaining the quality of the optimal solution. These advantages have even more significant engineering application value in large-scale reservoir optimization scheduling problems with multiple reservoirs and multiple constraints.
[0057] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for optimal reservoir scheduling, characterized in that, include: Obtain various hydrological data of the reservoir during the scheduling cycle, construct an optimized scheduling model, and represent the optimized scheduling model of the reservoir as a linear programming model. The linear programming model includes a scheduling model in which the decision variables, objective function and constraints themselves are linearly related, or an equivalent linear programming model obtained after piecewise linearization of the reservoir capacity-water level relationship and the head-output relationship. The linear programming model is solved using an isosurface translation solution mechanism, specifically including: S1: Canonical Model The linear programming model for optimal reservoir scheduling is as follows: Where Z represents the overall benefit index of reservoir operation and scheduling; x is the scheduling decision variable vector, whose elements represent the operation decision quantities for each time period; and c is the corresponding benefit weight coefficient vector. This indicates the various upper bound constraints that must be met during the operation of the reservoir. The constraint coefficient matrix, For the corresponding running boundary; This represents the physical constraints, such as water balance, that a reservoir system must satisfy. This indicates the various lower bound constraints that must be met during the operation of the reservoir; The constraints of the linear programming model for reservoir optimization scheduling are transformed into a form less than or equal to, and the initial basis of the scheduling model is constructed without introducing artificial variables. S2: Optimality Judgment Calculate the test numbers corresponding to each scheduling decision variable based on the current scheduling basis solution, and determine whether the current scheduling scheme satisfies the optimality condition or has a feasible improvement direction according to the linear programming discrimination criterion; S3: Determine the principal component While maintaining the feasibility of reservoir scheduling constraints, the rotating principal components used to update the scheduling basis solution are determined by comprehensively considering the feasible domain boundary constraints and the degree of improvement of the scheduling objective function isosurface. S4: Rotation Transformation Perform a rotation transformation centered on the pivot element, update the scheduling basis variables, and obtain a new reservoir scheduling basis solution; S5: Repeat steps S2 to S4 until a reservoir scheduling scheme that satisfies the optimality condition is obtained.
2. The reservoir optimization scheduling method according to claim 1, characterized in that: In step S1, a sign consistency check is introduced. When the reservoir water balance constraint or other constraints appear in the form of an equation, the initial scheduling basis solution is constructed using any of the following methods: (a) Perform equivalent splitting, splitting into an equivalent greater than or equal to constraint and a less than or equal to constraint, and process them separately; (b) For water balance type equality constraints, in combination with the characteristics of reservoir optimal scheduling, under the premise of ensuring that the physical meaning of scheduling remains unchanged, the water balance type equality constraints are equivalently expressed as constraints in the form of less than or equal to.
3. The reservoir optimization scheduling method according to claim 1, characterized in that: When all test numbers All constraint coefficients in the constraints If the answer is no, then it is determined that there is no solution; where, For the constant term of the scheduling constraints, These are the constraint coefficients for scheduling decision variables.
4. The reservoir optimization scheduling method according to claim 3, characterized in that: In step S3, only if the constant term of the scheduling constraint... Constraint coefficients of scheduling decision variables and corresponding test numbers The three symbols are consistent and , Only when the value is not zero will the corresponding element be considered as a candidate for scheduling principal and participate in principal determination.
5. The reservoir optimization scheduling method according to claim 1 or 4, characterized in that: In step S3, when all test numbers If the current problem has reached its optimal state, then a feasibility test is performed on the scheduling constraints; if a constant term exists in the scheduling constraints... If the current solution vector does not meet the scheduling feasibility requirements, then the scheduling system will remove the check number when performing the principal component selection. Strict non-zero requirement, i.e., selection of test number The columns are used as the principal selection columns to drive the solution vector to adjust in the direction that satisfies the scheduling constraints.
6. The reservoir optimization scheduling method according to claim 1 or 4, characterized in that: In step S3, by constructing an isosurface function value increment index that reflects the degree of improvement of the scheduling objective function, the candidate element with the largest function value increment is selected as the pivot element in the global scope, thereby guiding the isosurface of the reservoir scheduling objective function to shift in the direction of improvement.
7. The reservoir optimization scheduling method according to claim 1, characterized in that: When the current reservoir scheduling basic solution is a degraded solution and does not meet the optimality condition, the test number is selected first. In the column, the test number Calculate the minimum positive intercept of the largest column. The element corresponding to the minimum positive intercept is selected as the pivot, and rotations are performed using this pivot. For the constant term of the scheduling constraints, These are the constraint coefficients for scheduling decision variables.
8. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the reservoir optimization scheduling method according to any one of claims 1 to 7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of reservoir optimization scheduling of the method according to any one of claims 1 to 7.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the reservoir optimization scheduling method according to any one of claims 1 to 7.