Power capacity configuration method and system for new energy power station group

Abstract

The invention provides a new energy power station group power supply capacity configuration method and system, and relates to the technical field of power system planning, and the method comprises the steps: building a multi-objective optimization model based on a multi-objective function, and adding a constraint condition for the multi-objective optimization model; inputting real-time power generation data of the new energy power station group into the multi-objective optimization model added with the constraint conditions, and solving by using a solving method to obtain a non-dominated solution set; evaluating a performance index of each target function based on the non-dominated solution set, and dynamically adjusting an initial weight of each target function based on the performance index to obtain a variable weight; according to the variable weight, fusing each objective function through a weighted average method to obtain a comprehensive objective function; the real-time power generation data of the new energy power station group is solved based on the comprehensive objective function, power capacity configuration information is obtained, power capacity is configured, multiple objectives and constraint conditions are comprehensively considered, and balance and optimization among the multiple objectives are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system planning, and in particular to a power capacity configuration method and system for a new energy power station group. Background Art

[0002] With the continuous growth of global energy demand and the increasing awareness of environmental protection, the construction and operation of new energy power station clusters as an important means to replace traditional fossil energy has become a research hotspot in the current energy field. New energy power station clusters usually include solar power stations, wind power stations, hydropower stations and other types. These power stations have the advantages of being renewable, environmentally friendly and low-carbon, and are of great significance to promoting the optimization of energy structure and sustainable development.

[0003] However, the power capacity configuration of the new energy power station group is a complex and critical issue. On the one hand, the reasonable configuration of power capacity can ensure the stable operation of the power station group, meet the power demand, and reduce the construction and operation costs. On the other hand, reasonable power capacity configuration can also improve the utilization rate of renewable energy, reduce dependence on traditional energy, and achieve the goals of optimizing the energy structure and protecting the environment.

[0004] In the existing power capacity configuration methods of renewable energy power station groups, a single objective function is usually used for optimization, such as minimizing total cost or maximizing the proportion of renewable energy. However, this method often ignores the trade-offs and coordination between multiple objectives, resulting in the configuration results that may not meet actual needs. For example, simply pursuing the minimization of total cost may result in a too low proportion of renewable energy, while simply pursuing the maximization of the proportion of renewable energy may increase construction and operation costs.

[0005] Therefore, it is necessary to provide a power capacity configuration method and system for a new energy power station group to solve the above technical problems. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a method and system for configuring the power capacity of a new energy power station group, which comprehensively considers multiple objectives and constraints and achieves balance and optimization among multiple objectives.

[0007] The present invention provides a method for configuring power capacity of a new energy power station group, which is applied to a new energy power station group and comprises the following steps:

[0008] Building a multi-objective optimization model based on the multi-objective function, and adding constraints to the multi-objective optimization model;

[0009] Inputting the real-time power generation data of the renewable energy power station group into the multi-objective optimization model to which constraints have been added, and solving the model using a solution method to obtain a non-dominated solution set;

[0010] Evaluating the performance index of each objective function based on the non-dominated solution set, and dynamically adjusting the initial weight of each objective function based on the performance index to obtain a variable weight;

[0011] According to the variable weights, each objective function is fused by a weighted average method to obtain a comprehensive objective function;

[0012] Based on the comprehensive objective function, the real-time power generation data of the new energy power station group is solved to obtain power capacity configuration information, and the power capacity is configured based on the power capacity configuration information.

[0013] Preferably, the multi-objective optimization model is constructed based on the multi-objective function, comprising:

[0014] Two objective functions are defined, namely, minimizing the total cost function Z1 and maximizing the renewable energy proportion function Z2. The mathematical expression for minimizing the total cost function Z1 is:

[0015]

[0016] Among them, C i is the construction cost per unit capacity of the i-th power station, O i The operation and maintenance cost per unit time of the i-th power station, y i is the installed capacity of the ith power station, and N is the total number of power stations;

[0017] The mathematical expression for maximizing the renewable energy proportion function Z2 is:

[0018]

[0019] Among them, E i (t) is the energy conversion efficiency of the ith power station at time t, R is the index set of renewable energy power stations, and E∈N.

[0020] Preferably, adding constraints to the multi-objective optimization model includes:

[0021] Add the target constraint that the total power generation of all selected power plants is not less than the demand forecast at any time point t, where the mathematical expression of the target constraint is:

[0022]

[0023] Among them, x i is whether the i-th power station is selected, and is represented by 0 or 1, and D(t) is the demand forecast at time t;

[0024] Add a capacity limit constraint to ensure that the installed capacity of each power station does not exceed the maximum installed capacity, and the mathematical expression is:

[0025]

[0026] Among them, P i is the maximum installed capacity of the ith power station.

[0027] Preferably, the real-time power generation data of the renewable energy power station group is input into the multi-objective optimization model to which constraints have been added, and a solution method is used to obtain a non-dominated solution set, including:

[0028] Collecting power generation data of a group of new energy power stations in real time, wherein the power generation data includes the actual energy conversion efficiency and actual power generation of each power station at time t;

[0029] Substituting the actual energy conversion efficiency and the actual power generation as input parameters into a multi-objective optimization model to which constraints have been added;

[0030] The multi-objective optimization model is solved by using a pre-configured mixed integer linear programming solver to obtain a set of non-dominated solution sets, wherein each non-dominated solution represents a candidate power capacity configuration information.

[0031] Preferably, the evaluating the performance index of each objective function based on the non-dominated solution set, and dynamically adjusting the initial weight of each objective function based on the performance index to obtain a variable weight, comprises:

[0032] The performance indicators of all non-dominated solutions are evaluated, where the performance indicators include cost efficiency and renewable energy proportion, and their mathematical expressions are:

[0033]

[0034] Where E(s) is the cost efficiency of the non-dominated solution s, Z1(s) is the total cost of the non-dominated solution s, E is the expected minimum cost, and y i (s) is the installed capacity of the i-th power plant in the non-dominated solution s, R(s) is the renewable energy proportion of the non-dominated solution s;

[0035] The comprehensive performance index P is calculated, and a non-dominated solution with the highest comprehensive performance index is selected from the non-dominated solution set as a reference solution, wherein the calculation formula of the comprehensive performance index P is:

[0036] P=(1-E(s))+R(s).

[0037] Based on the performance index of the reference solution, the initial weight of each objective function is dynamically adjusted to obtain a variable weight.

[0038] Preferably, dynamically adjusting the initial weights of each objective function based on the performance metrics of the reference solution to obtain variable weights includes:

[0039] If E(s) > θ and R(s) < A, then adjust the initial weight ω1 of the total cost minimization function Z1 and the initial weight ω2 of the renewable energy proportion maximization function Z2 according to the adjustment factor α. The adjustment formulas are:

[0040] ω’1 = ω1·(1 - α)

[0041] ω’2 = 1 - ω’1

[0042] If E(s) > θ and R(s) ≥ A, then adjust the initial weight ω1 and the initial weight ω2 according to the adjustment factor β. The adjustment formulas are:

[0043] ω’1 = ω1·(1 - β)

[0044] ω’2 = 1 - ω’1

[0045] If E(s) ≤ θ and R(s) < A, then adjust the initial weight ω1 and the initial weight ω2 according to the adjustment factor γ. The adjustment formulas are:

[0046] ω’2 = ω2·(1 + γ)

[0047] ω’1 = 1 - ω’2

[0048] If E(s) ≤ θ and R(s) ≥ A, then keep the initial weight ω1 and the initial weight ω2 unchanged, specifically:

[0049] ω’1 = ω1

[0050] ω’2 = ω2;

[0051] Wherein, A represents a preset renewable energy proportion threshold, and θ represents a preset cost efficiency threshold.

[0052] Preferably, solving the real-time power generation data of the new energy power station group based on the comprehensive objective function to obtain power capacity configuration information, and configuring the power capacity based on the power capacity configuration information includes:

[0053] Taking the real-time power generation data of the new energy power station group as input parameters and substituting them into a multi-objective optimization model with added constraint conditions and integrated comprehensive objective function;

[0054] Using a pre-configured mixed integer linear programming solver to solve the comprehensive objective function to obtain an optimal solution, where the optimal solution represents the final power capacity configuration plan;

[0055] generating power capacity configuration information based on the final power capacity configuration plan, wherein the power capacity configuration information includes the selection status and actual power generation of each power station;

[0056] Based on the generated power capacity configuration information, the power capacity of the new energy power station group is actually configured to ensure that each power station operates according to the requirements of the optimal solution to achieve a balance between the lowest total cost and the maximum proportion of renewable energy.

[0057] The present invention also provides a power capacity configuration system for a new energy power station group, which is used to execute a power capacity configuration method for a new energy power station group. The system is applied to a new energy power station group, and comprises:

[0058] A model building module, used to build a multi-objective optimization model based on a multi-objective function and add constraints to the multi-objective optimization model;

[0059] A non-dominated solution set generation module, used for inputting the real-time power generation data of the renewable energy power station group into the multi-objective optimization model to which constraints have been added, and solving the model using a solution method to obtain a non-dominated solution set;

[0060] A weight adjustment module, used for evaluating the performance index of each objective function based on the non-dominated solution set, and dynamically adjusting the initial weight of each objective function based on the performance index to obtain a variable weight;

[0061] An objective function fusion module is used to fuse the objective functions by weighted average method according to the variable weights to obtain a comprehensive objective function;

[0062] The power capacity configuration module is used to solve the real-time power generation data of the new energy power station group based on the comprehensive objective function, obtain power capacity configuration information, and configure the power capacity based on the power capacity configuration information.

[0063] Compared with the related art, the method and system for configuring the power capacity of a new energy power station group provided by the present invention have the following beneficial effects:

[0064] The present invention proposes a method for configuring the power capacity of a renewable energy power station group. The method constructs a multi-objective optimization model based on a multi-objective function and considers multiple constraints to achieve balance and optimization among multiple objectives.

[0065] Specifically, the present invention defines two objective functions, namely, minimizing the total cost function and maximizing the renewable energy proportion function, and adds the target constraint condition that the total power generation is not less than the demand forecast and the capacity limit constraint to ensure the rationality and feasibility of the configuration results.

[0066] In the solution process, the present invention uses a mixed integer linear programming solver to solve the multi-objective optimization model and obtain a set of non-dominated solution sets. Then, the performance indicators of each objective function are evaluated based on the non-dominated solution set, and the initial weights of each objective function are dynamically adjusted to obtain variable weights. Finally, the objective functions are fused by the weighted average method according to the variable weights to obtain a comprehensive objective function, and the real-time power generation data of the new energy power station group is solved to obtain the final power capacity configuration plan.

[0067] In summary, the power capacity configuration method of a new energy power station group of the present invention comprehensively considers multiple objectives and constraints, and achieves balance and optimization among multiple objectives. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 A schematic diagram of a flow chart of a method for configuring power capacity of a renewable energy power station group provided by the present invention;

[0069] Figure 2 A schematic diagram of the module structure of a power capacity configuration system for a new energy power station group provided by the present invention. DETAILED DESCRIPTION

[0070] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only the parts related to the present invention, rather than all structures, are shown in the accompanying drawings. In addition, the embodiments of the present invention and the features in the embodiments may be combined with each other without conflict.

[0071] It should also be noted that, for ease of description, only the part relevant to the present invention but not all content is shown in the accompanying drawings. It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processing or methods depicted as flow charts. Although the flow chart describes each operation (or step) as sequential processing, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. When its operation is completed, the processing can be terminated, but it can also have additional steps not included in the accompanying drawings. The processing can correspond to methods, functions, procedures, subroutines, subprograms, etc.

[0072] Embodiment 1

[0073] The present invention provides a method for configuring the power capacity of a new energy power station group, the method is applied to a new energy power station group, and refers to Figure 1 As shown, the following steps are included:

[0074] S1: construct a multi-objective optimization model based on the multi-objective function, and add constraints to the multi-objective optimization model.

[0075] In the power capacity configuration of renewable energy power station groups, the multi-objective optimization model guides the optimization process through two main objective functions: minimizing the total cost and maximizing the proportion of renewable energy. These two objective functions correspond to different optimization purposes, and there may be conflicts between them, so it is necessary to find a balance point to achieve the optimal solution.

[0076] Specifically, in step S1, the construction process of the multi-objective optimization model includes:

[0077] Two objective functions are defined, namely, minimizing the total cost function Z1 and maximizing the renewable energy proportion function Z2. The mathematical expression for minimizing the total cost function Z1 is:

[0078]

[0079] Among them, C i is the construction cost per unit capacity of the i-th power station, O i The operation and maintenance cost per unit time of the i-th power station, y i is the installed capacity of the ith power station, and N is the total number of power stations.

[0080] In this embodiment, the objective function aims to reduce the overall construction, operation and maintenance costs of the renewable energy power plant group. Cost is a key factor in any commercial project, especially in the energy field, where both initial investment and long-term operating costs can be very high. Therefore, minimizing costs is crucial to improving the economic benefits of the project.

[0081] By minimizing costs, resources can be allocated more efficiently and unnecessary expenditures can be avoided, ensuring that funds are spent where they are most needed.

[0082] The mathematical expression for maximizing the renewable energy proportion function Z2 is:

[0083]

[0084] Among them, E i (t) is the energy conversion efficiency of the ith power station at time t, which reflects how much input the power station can convert into usable electricity at a specific point in time, R is the index set of renewable energy power stations, and E∈N.

[0085] In this embodiment, the objective function is to increase the proportion of renewable energy in the total energy output. The denominator represents the total power generation of all power plants at time t, while the numerator only includes the contribution of renewable energy power plants. Therefore, the larger the value of Z2, the higher the proportion of renewable energy in the entire power generation mix.

[0086] Minimizing the total cost function and maximizing the renewable energy proportion function work together to help decision makers find an optimal solution that is both economical and environmentally friendly when building a new energy power station group. In practical applications, a trade-off between the two objectives may be encountered, because reducing costs sometimes sacrifices a certain proportion of renewable energy, and vice versa. Therefore, in the multi-objective optimization process, it is usually necessary to combine other factors (such as policy orientation, market demand, etc.) to adjust the importance of each goal in order to achieve the best comprehensive effect.

[0087] In step S1, constraints are added to the multi-objective optimization model, specifically including:

[0088] Add the target constraint that the total power generation of all selected power plants is not less than the demand forecast at any time point t, where the mathematical expression of the target constraint is:

[0089]

[0090] Among them, x i is whether the i-th power station is selected, and is represented by 0 or 1, and D(t) is the demand forecast at time t.

[0091] Add a capacity limit constraint to ensure that the installed capacity of each power station does not exceed the maximum installed capacity, and the mathematical expression is:

[0092]

[0093] Among them, P i is the maximum installed capacity of the i-th power station, that is, the installed capacity of each power station shall not exceed its maximum installed capacity.

[0094] S2: inputting the real-time power generation data of the renewable energy power station group into the multi-objective optimization model to which constraints have been added, and solving the model using a solving method to obtain a non-dominated solution set.

[0095] Specifically, step S2 includes the following steps:

[0096] S21: Collecting power generation data of the new energy power station group in real time, wherein the power generation data includes the actual energy conversion efficiency and actual power generation of each power station at time t.

[0097] In this embodiment, it is necessary to collect power generation data from each new energy power station in real time. The core of this step is to ensure the timeliness and accuracy of the data, because real-time data directly reflects the current power generation efficiency and power generation, which are the basis for evaluating and optimizing the performance of power stations. To this end, a stable data acquisition system is established, which can automatically and continuously collect energy conversion efficiency and power generation information from each power station and transmit it to the central control system. The technical effect of this is that the most accurate snapshot of the power generation status can be obtained, providing a solid data foundation for subsequent optimization calculations.

[0098] S22: Substituting the actual energy conversion efficiency and the actual power generation as input parameters into a multi-objective optimization model to which constraints have been added.

[0099] In this embodiment, the actual energy conversion efficiency and actual power generation obtained in S21 are used as input parameters and integrated into the multi-objective optimization model to which constraints have been added. This means that real-time data is combined with a pre-set mathematical model to reflect the optimal working state of the power station under current conditions.

[0100] Specifically, once the latest power generation data is obtained, the next thing to do is to integrate this data into the multi-objective optimization model. This means updating the relevant parameters in the model, especially those time-related variables, such as the actual energy conversion efficiency and actual power generation of each power station at a specific point in time. At this stage, the new data is also checked to see if it meets the preset constraints, such as whether the power generation meets the predicted demand or whether it exceeds the maximum installed capacity of a power station. If any outliers or violations of constraints are found, appropriate adjustments or flags are made for further investigation.

[0101] In addition, in order to ensure the stability and convergence of the model, some parameters need to be normalized so that they are in the same order of magnitude to avoid instability in numerical calculations.

[0102] S23: Solving the multi-objective optimization model using a pre-configured mixed integer linear programming solver to obtain a set of non-dominated solution sets, wherein each non-dominated solution represents a candidate power capacity configuration information.

[0103] In this embodiment, when all necessary data are ready and the model parameters have been updated, the solution phase is entered. A mixed integer linear programming (MILP) solver is used here, which is a powerful mathematical tool specifically used to solve optimization problems involving discrete decision variables.

[0104] In this application, discrete variables refer to which power plants should be selected (represented by 0 or 1), while continuous variables include the installed capacity of each power plant. The task of the solver is to find a set of such variable values ​​so that both objective functions - minimizing total cost and maximizing the proportion of renewable energy - can achieve the best possible results while satisfying all constraints. Since this is a multi-objective optimization problem, there is usually not a single optimal solution; instead, a non-dominated solution set, the Pareto front, is obtained. Each solution in this solution set represents a possible power capacity configuration scheme, and none of them is absolutely better than another. In order to make a selection from these candidate solutions, their performance indicators need to be further evaluated, which will be completed in subsequent steps. The solution process involves an iterative algorithm, and with each iteration of the algorithm, the solver will gradually approach the optimal solution until the predetermined accuracy requirement or the maximum number of iterations is reached.

[0105] More specifically, the mixed integer linear programming (MILP) solver consists of the following main components:

[0106] Input module: Responsible for receiving and parsing the input data of the optimization model, including the objective function, constraints and variable type (discrete or continuous). For the power capacity configuration problem of the renewable energy power station group, the input module will receive the real-time collected power generation data, the pre-set multi-objective optimization model and its constraints.

[0107] Preprocessing module: Perform preliminary processing on input data to simplify the problem size and improve solution efficiency. This includes:

[0108] Redundant constraint elimination: remove constraints that do not affect the solution space.

[0109] Variable domain reduction: By analyzing the constraints, the range of variables can be narrowed and the search space can be reduced.

[0110] Parameter normalization: Adjust parameters of different magnitudes to the same magnitude to avoid instability in numerical calculations.

[0111] Branch and Bound (B&B): This is the core algorithm of the MILP solver, which is used to handle discrete variables. B&B recursively decomposes the problem into subproblems, selects a discrete variable in each subproblem for branching (i.e. fixes it to 0 or 1), and then solves the corresponding linear programming (LP) relaxation problem. By comparing the solutions of each branch, the global optimal solution is gradually approached.

[0112] Cutting Planes: Used to strengthen the constraints of the linear programming relaxation problem, so as to more closely approximate the integer solution. Cutting plane methods include but are not limited to Gomory cutting planes, Chv -Gomory cutting plane, etc. These methods cut off non-integer solutions by adding additional linear inequalities, making the solution space more compact.

[0113] Post-processing module: responsible for organizing and outputting the solution results. This includes:

[0114] Solution verification: Check whether the final solution satisfies all constraints and calculate its corresponding objective function value.

[0115] Solution interpretation: Converting the mathematical solution into an understandable form, such as generating specific power capacity configuration information.

[0116] Based on the above structure, the specific solution process of the MILP solver can be divided into the following steps:

[0117] Step 1: Input data preparation

[0118] Receive real-time data: Get the latest power generation data, including the actual energy conversion efficiency and actual power generation of each power station at time t.

[0119] Update model parameters: Substitute these real-time data into the multi-objective optimization model and update relevant variables, such as the installed capacity and power generation of each power station.

[0120] Set initial weights: Set initial weights for the two objective functions. The initial weights can be set based on historical experience.

[0121] Step 2: Preprocessing

[0122] Redundant constraint elimination: Check and remove constraints that do not affect the solution space, such as the situation where the maximum installed capacity of some power plants already meets the demand forecast.

[0123] Variable domain reduction: By analyzing the constraints, determine the reasonable value range of each variable and reduce the search space.

[0124] Parameter normalization: adjust parameters of different magnitudes (such as cost and scale) to the same magnitude to ensure the stability of numerical calculations.

[0125] Step 3: Branch and Bound (B&B)

[0126] Initialization: Initialize all discrete variables (such as the selected state of the power plant) to an undetermined state and solve the initial linear programming (LP) relaxation problem.

[0127] Branching: Select an undetermined discrete variable and fix it to 0 and 1 respectively, generating two sub-problems.

[0128] Solve subproblems: Solve the corresponding LP relaxation problem for each subproblem to obtain its optimal solution and objective function value.

[0129] Pruning: Compare the solutions of each branch. If the solution of a branch is obviously inferior to other branches, prune the branch and do not explore further.

[0130] Update upper and lower bounds: Update the upper and lower bounds of the global optimal solution based on the best integer solution currently found, and gradually approach the global optimal solution.

[0131] Step 4: Cutting Plane Method

[0132] Add cutting planes: After each solution of the LP relaxation problem, check whether the solution is an integer solution. If not, add appropriate cutting planes to cut off non-integer solutions and make the solution space more compact.

[0133] Resolve: After adding the cutting planes, resolve the LP relaxation problem and continue the branch and bound process until an integer solution is found or the predetermined accuracy requirement is met.

[0134] Step 5: Post-processing

[0135] Solution verification: Check whether the final solution satisfies all constraints and calculate its corresponding objective function value.

[0136] Solution interpretation: Convert the mathematical solution into an understandable form, such as generating specific power capacity configuration information, including the selection status and actual power generation of each power plant.

[0137] S3: Evaluate the performance index of each objective function based on the non-dominated solution set, and dynamically adjust the initial weight of each objective function based on the performance index to obtain a variable weight.

[0138] Specifically, step S3 includes the following steps:

[0139] The performance indicators of all non-dominated solutions are evaluated, where the performance indicators include cost efficiency and renewable energy proportion, and their mathematical expressions are:

[0140]

[0141] Where E(s) is the cost efficiency of the non-dominated solution s, Z1(s) is the total cost of the non-dominated solution s, E is the expected minimum cost, and y i (s) is the installed capacity of the i-th power station in the non-dominated solution s, and R(s) is the proportion of renewable energy in the non-dominated solution s.

[0142] In this embodiment, in multi-objective optimization, there is usually not a single optimal solution, but a set of non-dominated solutions (i.e., Pareto front). These solutions reach different balance points between different objectives. In order to select the most appropriate solution from these candidate solutions, each non-dominated solution needs to be evaluated to understand their performance on each objective. This step is the basis for dynamically adjusting the weights, ensuring that the subsequent optimization process can flexibly adjust the importance of each objective according to the actual situation.

[0143] By quantitatively evaluating the cost efficiency and renewable energy proportion of each non-dominated solution, we can fully understand the performance of each solution on the two main objectives and provide a basis for subsequent selection.

[0144] The comprehensive performance index P is calculated, and a non-dominated solution with the highest comprehensive performance index is selected from the non-dominated solution set as a reference solution, wherein the calculation formula of the comprehensive performance index P is:

[0145] P=(1-E(s))+R(s).

[0146] In this embodiment, in order to select an optimal reference solution from multiple non-dominated solutions, a comprehensive evaluation criterion is required. The comprehensive performance index is a commonly used measurement method that combines multiple performance indicators and can more comprehensively reflect the overall performance of each solution. Selecting the non-dominated solution with the highest comprehensive performance index as the reference solution can ensure that the solution we select achieves the best balance between multiple objectives.

[0147] The calculation formula of the comprehensive performance index P is designed to be: when the cost efficiency is close to 1 (that is, the cost is close to the minimum), 1-E(s) is close to 0; and the higher the proportion of renewable energy, the larger R(s). Therefore, the higher the comprehensive performance index, the better the balance between the two objectives.

[0148] The solution with the highest comprehensive performance index is selected from the non-dominated solution set as the reference solution. This solution will play a key role in the subsequent weight adjustment process because it represents the current optimal balance point.

[0149] By introducing the comprehensive performance index, an optimal reference solution can be found among multiple non-dominated solutions, ensuring that the selected solution achieves the best balance between multiple objectives.

[0150] Based on the performance index of the reference solution, the initial weight of each objective function is dynamically adjusted to obtain a variable weight.

[0151] In multi-objective optimization, there may be conflicts between different objectives. Therefore, it is necessary to dynamically adjust the weights of each objective according to the actual situation to achieve the optimal balance. By evaluating the performance indicators of the reference solutions, it is possible to determine which objective the current optimization should focus on and adjust the weights accordingly. This dynamic adjustment mechanism enables the optimization process to better adapt to the changing environment and requirements.

[0152] Specifically, the dynamic adjustment process of the initial weights of each objective function includes:

[0153] If E(s) > θ and R(s) < A, it means that the cost is already close to the optimal value, but the proportion of renewable energy still needs to be increased. At this time, the weight of the cost objective should be reduced, and the weight of the renewable energy proportion objective should be increased. Then, the initial weight ω1 of the total cost minimization function Z1 and the initial weight ω2 of the renewable energy proportion maximization function Z2 are adjusted according to the adjustment factor α. The adjustment formula is:

[0154] ω’1 = ω1·(1 - α)

[0155] ω’2 = 1 - ω’1

[0156] If E(s) > θ and R(s) ≥ A, it means that the cost is already close to the optimal value, and the proportion of renewable energy has reached or exceeded the target. At this time, the weight of the cost objective should be further reduced to continue optimizing the cost. Then, the initial weight ω1 and the initial weight ω2 are adjusted according to the adjustment factor β. The adjustment formula is:

[0157] ω’1 = ω1·(1 - β)

[0158] ω’2 = 1 - ω’1

[0159] If E(s) ≤ θ and R(s) < A, it means that the cost has not yet reached the optimal value, but the proportion of renewable energy still needs to be increased. At this time, the weight of the renewable energy proportion objective should be increased, and the weight of the cost objective should be appropriately reduced to balance the two objectives. Then, the initial weight ω1 and the initial weight ω2 are adjusted according to the adjustment factor γ. The adjustment formula is:

[0160] ω’2 = ω2·(1 + γ)

[0161] ω’1 = 1 - ω’2

[0162] If E(s) ≤ θ and R(s) ≥ A, it means that both objectives have reached a good level. At this time, the weights are kept unchanged to avoid unnecessary adjustments. Specifically, the initial weight ω1 and the initial weight ω2 are kept unchanged, as follows:

[0163] ω’1 = ω1

[0164] ω’2 = ω2;

[0165] Wherein, A represents the preset renewable energy ratio threshold, and θ represents the preset cost efficiency threshold.

[0166] S4: According to the variable weights, each objective function is fused by a weighted average method to obtain a comprehensive objective function.

[0167] In this embodiment, in the multi-objective optimization process, the two objective functions of minimizing total cost and maximizing the proportion of renewable energy often conflict with each other. In order to find an optimal solution that can reduce cost and increase the proportion of renewable energy, a trade-off must be made between these two objectives. By dynamically adjusting the initial weights of each objective function, the importance of cost and renewable energy proportion in different situations can be flexibly reflected. However, adjusting the weights alone cannot directly guide the optimization solver on how to find the optimal solution, so these weights need to be integrated into a comprehensive objective function so that the optimization solver can search under a unified goal.

[0168] Specifically, the variable weights and obtained from step S3 reflect the relative importance of the cost efficiency and the proportion of renewable energy of the optimal solution in the current non-dominated solution set.

[0169] The two objective functions are combined into a new single objective function by weighted averaging. The values ​​of ω'1 and ω'2 determine the contribution of each objective function to the comprehensive objective function. For example, if ω'1 is larger, the comprehensive objective function tends to minimize the cost; if ω'2 is larger, it tends to maximize the proportion of renewable energy.

[0170] The comprehensive objective function replaces the two independent objective functions in the original multi-objective optimization model to form a new single-objective optimization model. This means that in the subsequent solution process, the optimization solver will only focus on maximizing or minimizing the comprehensive objective function Z instead of processing the two independent objective functions separately.

[0171] S5: Solving the real-time power generation data of the new energy power station group based on the comprehensive objective function to obtain power capacity configuration information, and configuring the power capacity based on the power capacity configuration information.

[0172] Specifically, step S5 includes the following steps:

[0173] S51: The real-time power generation data of the renewable energy power station group is used as an input parameter and substituted into a multi-objective optimization model to which constraints have been added and integrated into a comprehensive objective function.

[0174] S52: Use a pre-configured mixed integer linear programming solver to solve the comprehensive objective function to obtain an optimal solution, wherein the optimal solution represents a final power capacity configuration plan.

[0175] In this embodiment, after the real-time power generation data is prepared and the dynamically adjusted weights are applied, the solution phase is now entered. The key to this phase is to use the pre-configured MILP solver to solve the comprehensive objective function to obtain a more optimized power capacity configuration solution. The specific working principle and process are as follows:

[0176] Start the pre-configured MILP solver and set the initial parameters. These parameters include the maximum number of iterations, convergence accuracy, etc. to ensure that the solving process can find a high-quality solution in a reasonable time.

[0177] The solver starts solving the comprehensive objective function. Since this is a single-objective optimization problem, the solver will find a set of variable values ​​that optimizes the comprehensive objective function. In this process, the solver will consider all constraints to ensure that the solution found meets the requirements of cost and renewable energy content, as well as practical operating constraints.

[0178] For discrete variables (such as the selected state of a power plant), the solver uses branch-and-bound to gradually narrow the search space. By recursively decomposing the problem into subproblems and selecting a discrete variable in each subproblem for branching, the solver can effectively approach the global optimal solution. In addition, the solver also uses historical solution results (i.e., the set of non-dominated solutions obtained during the first solution) as heuristic information to speed up the solution process.

[0179] To enforce constraints on the relaxed linear programming problem, the solver uses the cutting plane method. By adding additional linear inequalities to cut off non-integer solutions, the solution space is made more compact, thereby finding integer solutions more quickly. This step helps improve the efficiency of the solution, especially when facing complex optimization problems.

[0180] Once a potential optimal solution is found, the solver verifies the solution to ensure that it satisfies all constraints and calculates the corresponding objective function value. If the quality of the solution is good enough, the solving process ends; otherwise, the solver will continue searching until the predetermined accuracy requirement or the maximum number of iterations is reached.

[0181] S53: Generate power capacity configuration information based on the final power capacity configuration plan, wherein the power capacity configuration information includes the selection status and actual power generation of each power station.

[0182] In this embodiment, after the solver finds a new optimal solution, the next step is to convert the mathematical solution into power capacity configuration information that is easy to understand and execute, and apply it to actual operations. The specific working principle and process are as follows:

[0183] The optimal solution output by the solver is parsed to extract the selection status (0 or 1) and actual power generation of each power plant. This information can directly guide the operators of power plants on how to configure power capacity to achieve a balance between the lowest total cost and the maximum proportion of renewable energy.

[0184] Based on the parsed results, detailed power capacity configuration information is generated. This information is usually presented in the form of tables or graphs, which is convenient for operators to view and understand. For example, a table can list the name, selection status and expected power generation of each power station; a graph can intuitively show the contribution and relationship of each power station under the new configuration.

[0185] The generated configuration information is sent to the control systems of each power plant to ensure that they operate according to the requirements of the new optimal solution. Through the automated system, these instructions can be issued quickly and accurately, reducing the possibility of human intervention and improving the efficiency and accuracy of configuration.

[0186] S54: Based on the generated power capacity configuration information, the power capacity of the new energy power station group is actually configured to ensure that each power station operates according to the requirements of the optimal solution to achieve a balance between the lowest total cost and the maximum proportion of renewable energy.

[0187] In this embodiment, after generating and applying the power capacity configuration information, we need to ensure that each power station actually configures the power capacity according to the requirements of the optimal solution. The specific working principle and process are as follows:

[0188] Implementation configuration: Based on the power capacity configuration information generated in step S53, the operator or the automation system will actually adjust the power capacity of each power station. This includes:

[0189] Select power plants: Based on the selection status (0 or 1) in the optimal solution, decide which power plants should be activated or deactivated.

[0190] Set power generation plan: According to the actual power generation in the optimal solution, formulate a detailed power generation plan to ensure that the power generation of each power station in different time periods meets expectations.

[0191] Verify the configuration effect: After the configuration is implemented, immediately verify the actual operation of each power plant to ensure that they are working properly according to the requirements of the optimal solution. This can be done through a real-time monitoring system to check whether the energy conversion efficiency, actual power generation and other key indicators of each power plant meet expectations.

[0192] Embodiment 2

[0193] The present invention also provides a power capacity configuration system for a new energy power station group, which is used to execute a power capacity configuration method for a new energy power station group. The system is applied to a new energy power station group, referring to Figure 2As shown, including:

[0194] The model building module 100 is used to build a multi-objective optimization model based on a multi-objective function and add constraints to the multi-objective optimization model.

[0195] The non-dominated solution set generation module 200 is used to input the real-time power generation data of the renewable energy power station group into the multi-objective optimization model to which constraints have been added, and solve it using a solution method to obtain a non-dominated solution set.

[0196] The weight adjustment module 300 is used to evaluate the performance index of each objective function based on the non-dominated solution set, and dynamically adjust the initial weight of each objective function based on the performance index to obtain a variable weight.

[0197] The objective function fusion module 400 is used to fuse the objective functions by weighted average method according to the variable weights to obtain a comprehensive objective function.

[0198] The power capacity configuration module 500 is used to solve the real-time power generation data of the new energy power station group based on the comprehensive objective function, obtain power capacity configuration information, and configure the power capacity based on the power capacity configuration information.

[0199] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0200] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, the storage medium including a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable programmable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0201] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. A method for configuring power capacity of a new energy power station group, characterized in that: The method is applied to a group of new energy power stations, and comprises the following steps: Building a multi-objective optimization model based on the multi-objective function, and adding constraints to the multi-objective optimization model; Inputting the real-time power generation data of the renewable energy power station group into the multi-objective optimization model to which constraints have been added, and solving the model using a solution method to obtain a non-dominated solution set; Evaluating the performance index of each objective function based on the non-dominated solution set, and dynamically adjusting the initial weight of each objective function based on the performance index to obtain a variable weight; According to the variable weights, each objective function is fused by a weighted average method to obtain a comprehensive objective function; Based on the comprehensive objective function, the real-time power generation data of the new energy power station group is solved to obtain power capacity configuration information, and the power capacity is configured based on the power capacity configuration information.

2. A method for configuring power capacity of a new energy power station group according to claim 1, characterized in that: The multi-objective optimization model is constructed based on the multi-objective function, comprising: Two objective functions are defined, namely, minimizing the total cost function Z1 and maximizing the renewable energy proportion function Z2. The mathematical expression for minimizing the total cost function Z1 is: Among them, C i is the construction cost per unit capacity of the i-th power station, O i The operation and maintenance cost per unit time of the i-th power station, y i is the installed capacity of the ith power station, and N is the total number of power stations; The mathematical expression for maximizing the renewable energy proportion function Z2 is: Among them, E i (t) is the energy conversion efficiency of the ith power station at time t, R is the index set of renewable energy power stations, and E∈N.

3. A method for configuring power capacity of a new energy power station group according to claim 2, characterized in that: Adding constraint conditions to the multi-objective optimization model includes: Add the target constraint that the total power generation of all selected power plants is not less than the demand forecast at any time point t, where the mathematical expression of the target constraint is: Among them, x i is whether the i-th power station is selected, and is represented by 0 or 1, and D(t) is the demand forecast at time t; Add a capacity limit constraint to ensure that the installed capacity of each power station does not exceed the maximum installed capacity, and the mathematical expression is: Among them, P i is the maximum installed capacity of the ith power station.

4. A method for configuring power capacity of a new energy power station group according to claim 3, characterized in that: The real-time power generation data of the renewable energy power station group is input into the multi-objective optimization model to which constraints have been added, and a solution method is used to obtain a non-dominated solution set, including: Collecting power generation data of a group of new energy power stations in real time, wherein the power generation data includes the actual energy conversion efficiency and actual power generation of each power station at time t; Substituting the actual energy conversion efficiency and the actual power generation as input parameters into a multi-objective optimization model to which constraints have been added; The multi-objective optimization model is solved by using a pre-configured mixed integer linear programming solver to obtain a set of non-dominated solution sets, wherein each non-dominated solution represents a candidate power capacity configuration information.

5. A method for configuring power capacity of a new energy power station group according to claim 4, characterized in that: The step of evaluating the performance index of each objective function based on the non-dominated solution set and dynamically adjusting the initial weight of each objective function based on the performance index to obtain a variable weight includes: The performance indicators of all non-dominated solutions are evaluated, where the performance indicators include cost efficiency and renewable energy proportion, and their mathematical expressions are: Where E(s) is the cost efficiency of the non-dominated solution s, Z1(s) is the total cost of the non-dominated solution s, E is the expected minimum cost, and y i (s) is the installed capacity of the i-th power plant in the non-dominated solution s, R(s) is the renewable energy proportion of the non-dominated solution s; The comprehensive performance index P is calculated, and a non-dominated solution with the highest comprehensive performance index is selected from the non-dominated solution set as a reference solution, wherein the calculation formula of the comprehensive performance index P is: P = (1-E(s)) + R(s); Based on the performance index of the reference solution, the initial weight of each objective function is dynamically adjusted to obtain a variable weight.

6. A method for configuring power capacity of a new energy power station group according to claim 5, characterized in that: The step of dynamically adjusting the initial weights of the objective functions based on the performance indicators of the reference solution to obtain variable weights includes: If E(s) > θ and R(s) < A, then the initial weight ω1 that minimizes the total cost function Z1 and the initial weight ω2 that maximizes the renewable energy proportion function Z2 are adjusted according to the adjustment factor α. The adjustment formulas are as follows: ω’1 = ω1·(1 - α) ω’2 = 1 - ω’1 If E(s) > θ and R(s) ≥ A, then the initial weight ω1 and the initial weight ω2 are adjusted according to the adjustment factor β. The adjustment formulas are as follows: ω’1 = ω1·(1 - β) ω’2 = 1 - ω’1 If E(s) ≤ θ and R(s) < A, then the initial weight ω1 and the initial weight ω2 are adjusted according to the adjustment factor γ. The adjustment formulas are as follows: ω’2 = ω2·(1 + γ) ω’1 = 1 - ω’2 If E(s) ≤ θ and R(s) ≥ A, then the initial weight ω1 and the initial weight ω2 remain unchanged, specifically: ω’1 = ω1 ω’2 = ω2; where A represents a preset renewable energy proportion threshold, and θ represents a preset cost - efficiency threshold.

7. A method for configuring power capacity of a new energy power station group according to claim 6, characterized in that: The method of solving the real - time power generation data of the new - energy power station group based on the comprehensive objective function to obtain power capacity configuration information and configuring the power capacity based on the power capacity configuration information includes: Taking the real - time power generation data of the new - energy power station group as input parameters and substituting them into a multi - objective optimization model with added constraint conditions and integrated comprehensive objective function; Using a pre - configured mixed - integer linear programming solver to solve the comprehensive objective function to obtain an optimal solution, where the optimal solution represents the final power capacity configuration plan; Generating power capacity configuration information based on the final power capacity configuration plan, where the power capacity configuration information includes the selection status and actual power generation of each power station; Based on the generated power capacity configuration information, actually configuring the power capacity of the new - energy power station group to ensure that each power station operates according to the requirements of the optimal solution, so as to achieve the balance of the lowest total cost and the maximum renewable energy proportion.

8. A power capacity configuration system for a new energy power station group, used to execute a power capacity configuration method for a new energy power station group as claimed in any one of claims 1 to 7, characterized in that: The system is applied to a new - energy power station group, including: A model - building module, which is used to build a multi - objective optimization model based on a multi - objective function and add constraint conditions to the multi - objective optimization model; A non - dominated solution set generation module, which is used to input the real - time power generation data of the new - energy power station group into the multi - objective optimization model with added constraint conditions and use a solving method to solve it to obtain a non - dominated solution set; A weight adjustment module, which is used to evaluate the performance indicators of each objective function based on the non - dominated solution set and dynamically adjust the initial weights of each objective function based on the performance indicators to obtain variable weights; An objective - function fusion module, which is used to fuse each objective function by a weighted - average method according to the variable weights to obtain a comprehensive objective function; A power - capacity configuration module, which is used to solve the real - time power generation data of the new - energy power station group based on the comprehensive objective function to obtain power capacity configuration information and configure the power capacity based on the power capacity configuration information.

Images