Power dynamic scheduling method and system based on smart power grid

Abstract

The invention discloses an electric power dynamic scheduling method for a smart power grid, and relates to the field of power grid scheduling, and the method comprises the steps: collecting the power consumption data and meteorological data of a power consumption area and the operation data of the smart power grid in a first scheduling period, and obtaining the predicted meteorological data of the power consumption area in a second scheduling period, generating predicted electricity consumption data in the second scheduling period according to the electricity consumption data, the meteorological data and the predicted meteorological data of the electricity consumption area; and solving a preset target optimization function under a plurality of preset constraint conditions by adopting a whale algorithm to obtain a power supply scheme of each power supply device of the smart power grid in a second scheduling period, so that a control center issues a corresponding power supply instruction to each power supply device and controls the power supply of the devices. And calculating a power supply scheme of each power supply device through the whale algorithm, finally issuing a control instruction according to the power supply scheme, and repeating the steps in each scheduling period to realize dynamic scheduling of power.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid dispatching, and more specifically to a method and system for dynamic power dispatching based on a smart grid. Background Art

[0002] With the development of cities, the scale of power supply is increasing. Power dispatching must not only meet the requirements of stable operation of the power grid, but also consider the cost of power dispatching. Conventional manual dispatching can no longer meet the requirements, and there is an urgent need to achieve refined dispatching management through scientific means.

[0003] For example, Chinese patent application publication number CN114629180A provides a method and system for intelligent power grid scheduling. This invention predetermines the power supply to a power consumption area at a specific time based on the weather forecast for that time and the power load of the power consumption area at the same time on the previous day or the previous several days corresponding to that time. The system then adjusts the power supply to the power consumption area based on the power load of the power consumption area. However, this invention only considers the impact of power load when adjusting the power supply power, which is not comprehensive and can easily affect the accuracy of the power supply plan.

[0004] Therefore, the existing technology cannot accurately realize the power dispatching of the smart grid, which is insufficient. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a dynamic power scheduling method based on smart grid. This method not only takes into account the impact of weather factors on power consumption, but also establishes the optimization objectives and constraints of the power supply plan. Finally, the whale algorithm is used to solve the target optimization function, so as to obtain a more accurate and reasonable power supply plan.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for dynamic power dispatching based on a smart grid, comprising:

[0008] During a first scheduling cycle, power consumption data, meteorological data, and smart grid operation data of the power consumption area are collected, wherein the first scheduling cycle is a scheduling cycle with the current time as the end point of the interval;

[0009] Acquiring forecasted meteorological data for the power consumption area within a second scheduling period, wherein the second scheduling period is a scheduling period starting from the current time;

[0010] generating predicted power consumption data within the second scheduling period based on the power consumption data, meteorological data and predicted meteorological data of the power consumption area;

[0011] A whale algorithm is used to solve a preset target optimization function under preset multiple constraints to obtain a power supply plan for each power supply device of the smart grid during the second scheduling period, wherein the multiple constraints are determined based on the predicted power consumption data and the operating principle of the smart grid, and the target optimization function is determined based on the membership of the optimization functions of multiple scheduling objectives, and the membership is determined according to a trapezoidal function;

[0012] The calculated power supply scheme is uploaded to a control center, so that the control center issues corresponding power supply instructions to each power supply device to control the power supply of the device.

[0013] As a further improvement of the present invention, the whale algorithm is used to solve the preset target optimization function under multiple preset constraints, including:

[0014] Inputting operation data of the smart grid in the power consumption area, wherein the operation data of the smart grid includes operation data of each power supply device in the smart grid;

[0015] Initialize the relevant parameters of the whale algorithm and set the maximum number of iterations T max , whale population size H, whale individual dimension v and probability factor p;

[0016] An iterative operation is performed to output the power supply scheme.

[0017] As a further improvement of the present invention, the iterative operation includes:

[0018] Update the coefficient vector g, o and convergence factor a, find the optimal whale individual by calculating the target optimization function, retain the current optimal whale position, mutate the optimal whale position, update the whale position, and if the whale position is better than the optimal whale position, update the current optimal whale position until the number of iterations reaches the maximum number of iterations T max , output the current optimal whale position, which is the power supply solution.

[0019] As a further improvement of the present invention, the method of updating the convergence factor a is: Where s is a random number between (0,1).

[0020] As a further improvement of the present invention, the method for varying the optimal whale position is Cauchy variation, specifically: where X * (t) is the optimal whale position in the tth generation, and cauchy(0,1) is the Cauchy operator.

[0021] As a further improvement of the present invention, updating the whale position includes:

[0022] If the probability factor p is less than 50%, the whale position is updated according to the bubble net predation mechanism, which is an Archimedean screw update, specifically: X(t+1)=D'bl·cos(2πl)+X*(t)l, where D' is the distance between the current whale position and the optimal whale position, b is a constant coefficient, and l is a random number between [-1, 1];

[0023] If the probability factor p is greater than or equal to 50% and the coefficient vector |g| is less than 1, the whale position is updated according to the prey encirclement mechanism, specifically: X(t+1)=X*(t)-g·D1, D1=|o·X*(t)-X(t)|, where X(t) is the position of the individual whale in generation t;

[0024] If the probability factor p≥50% and the coefficient vector |g|≥1, the position is updated according to the global search mechanism, specifically: X(t+1)=X rand (t)-g|o·X rand (t)-X(t)|, where X rand (t) is a randomly selected whale individual in the tth generation.

[0025] As a further improvement of the present invention, generating predicted power consumption data within the second scheduling period based on the power consumption data, meteorological data, and predicted meteorological data of the power consumption area includes:

[0026] generating a plurality of sample matrices and a plurality of prediction matrices based on the meteorological data and the predicted meteorological data, wherein each sample matrix and each prediction matrix corresponds to a sub-time period, the sub-time periods of all the sample matrices constitute the first scheduling period, and the sub-time periods of all the prediction matrices constitute the second scheduling period;

[0027] The predicted power consumption data is determined based on all sample matrices and all prediction matrices.

[0028] As a further improvement of the present invention, determining the predicted power consumption data based on all sample matrices and all prediction matrices includes:

[0029] According to the comprehensive similarity index, the correlation between each prediction matrix and each sample matrix is ​​calculated respectively to obtain a similarity group corresponding to each prediction matrix, wherein the comprehensive similarity index is determined according to the grey relational degree and the cosine similarity, and the similarity group includes the first N sample matrices in the similarity index sequence, wherein the similarity index sequence is a sequence in which the sample matrices are sorted in descending order according to the similarity index with the prediction matrix, wherein N is a positive integer;

[0030] According to the time corresponding to each of the sample matrices in the similar group, obtaining the power consumption corresponding to each time in the power consumption data;

[0031] The predicted power consumption at the corresponding moment of the prediction matrix is ​​calculated based on the corresponding power consumption.

[0032] As a further improvement of the present invention, the step of determining the target optimization function includes:

[0033] Calculating the membership of the optimization functions of the plurality of scheduling objectives using a trapezoidal function to obtain the membership functions of the plurality of scheduling objectives;

[0034] The target optimization function is determined according to the membership function.

[0035] As a further improvement of the present invention, the optimization functions of the multiple scheduling objectives include:

[0036] The objective optimization function of scheduling cost: Where A is the total cost and I is the total benefit;

[0037] Objective optimization function of environmental governance cost: Where k represents the type of pollutant, K represents the total number of pollutant types, and C k represents the cost of treating type k pollutants, where d represents the dth moment in the total duration of the scheduling period, and D represents the total duration of the scheduling period;

[0038] Objective optimization function of user satisfaction: f c =λ1O1+λ2O2+λ3O3, where λ1, λ2 and λ3 represent weight coefficients, O1 represents the user's electricity consumption, O2 represents the user's electricity consumption time, and O3 represents the user's satisfaction.

[0039] As a further improvement of the present invention, the multiple constraints include:

[0040] The output power constraints of each power supply device are: in is the minimum output power of the i-th power supply device, is the maximum output power of the i-th power supply device, P i (d) represents the output power of the i-th power supply device at the d-th moment;

[0041] The output power constraint of the smart grid is: Among them, P grid (d) is the interaction power between the active distribution network and the main grid dispatching center at the dth moment, P loda (d) the electric load power required at the dth moment, the required electric load power being determined based on the predicted power consumption data;

[0042] Pollutant emission constraints: E k (P i )≤L x , where E k (P i ) is the mass of the kth type of pollutant emitted by the i-th power supply equipment, L x The maximum mass of the kth type of pollutant allowed to be discharged.

[0043] As a further improvement of the present invention, the present invention also provides a power dynamic dispatching system based on a smart grid, comprising:

[0044] Acquisition module: collects power consumption data, meteorological data, and smart grid operation data of the power consumption area during a first scheduling period, wherein the first scheduling period is a scheduling period with the current time as the end point of the interval; obtains predicted meteorological data of the power consumption area during a second scheduling period, wherein the second scheduling period is a scheduling period with the current time as the starting point of the interval;

[0045] Prediction module: generating predicted power consumption data within the second scheduling period based on the power consumption data, meteorological data and predicted meteorological data of the power consumption area;

[0046] An analysis module: employing a whale algorithm to solve a preset target optimization function under a plurality of preset constraints to obtain a power supply plan for each power supply device of the smart grid during the second scheduling period, wherein the plurality of constraints are determined based on the predicted power consumption data and the operating principle of the smart grid, and the target optimization function is determined based on membership of optimization functions of a plurality of scheduling objectives, wherein the membership is determined according to a trapezoidal function;

[0047] Transmission module: uploads the calculated power supply scheme to the control center, so that the control center issues corresponding power supply instructions to each power supply device to control the power supply of the device.

[0048] The present invention predicts power consumption through a prediction model, and uses a target optimization function and a whale algorithm to reasonably arrange the power supply plan of each power supply device in the smart grid according to the predicted power consumption and other constraints. Compared with the existing technology, the present invention can obtain a more accurate power supply plan, and can upload the power supply plan to the control center, which will issue corresponding power supply instructions to realize power scheduling. Moreover, the present invention can automatically calculate and generate different power supply plans based on data collected at different times to achieve the purpose of dynamic scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of the method steps of the present invention;

[0050] Figure 2 Schematic diagram of the steps for generating the target optimization function;

[0051] Figure 3 This is a flowchart of the whale algorithm;

[0052] Figure 4 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0053] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations of the technical solution of the present invention.

[0054] The term "and / or" in the following text simply describes an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Additionally, the character " / " generally indicates an "or" relationship between the related objects.

[0055] like Figure 1 As shown, this embodiment provides a method for dynamic power dispatch based on a smart grid, comprising:

[0056] During a first scheduling cycle, power consumption data, meteorological data, and smart grid operation data of the power consumption area are collected, wherein the first scheduling cycle is a scheduling cycle with the current time as the end point of the interval;

[0057] Acquiring forecasted meteorological data for the power consumption area within a second scheduling period, wherein the second scheduling period is a scheduling period starting from the current time;

[0058] generating predicted power consumption data within the second scheduling period based on the power consumption data, meteorological data and predicted meteorological data of the power consumption area;

[0059] A whale algorithm is used to solve a preset target optimization function under preset multiple constraints to obtain a power supply plan for each power supply device of the smart grid during the second scheduling period, wherein the multiple constraints are determined based on the predicted power consumption data and the operating principle of the smart grid, and the target optimization function is determined based on the membership of the optimization functions of multiple scheduling objectives, and the membership is determined according to a trapezoidal function;

[0060] The calculated power supply scheme is uploaded to a control center, so that the control center issues corresponding power supply instructions to each power supply device to control the power supply of the device.

[0061] Since the current moment is constantly changing, this embodiment can calculate the optimal power supply plan for each power supply device in the smart grid based on the data automatically collected in each scheduling cycle through the prediction model, target optimization function and whale algorithm, and directly issue power supply instructions through the control center to achieve dynamic scheduling of electricity.

[0062] Furthermore, in order to obtain the final power supply solution, this embodiment further provides a method for solving the target optimization function under constraints using the whale algorithm to obtain the power supply solution of each power supply device. The specific steps are as follows: Figure 3 Shown, including:

[0063] Inputting operation data of the smart grid in the power consumption area, wherein the operation data of the smart grid includes operation data of each power supply device in the smart grid;

[0064] Initialize the relevant parameters of the whale algorithm and set the maximum number of iterations T max , whale population size H, whale individual dimension v and probability factor p;

[0065] An iterative operation is performed to output the power supply scheme.

[0066] As a further improvement of the present invention, the iterative operation includes: updating the coefficient vectors g, o and the convergence factor a, finding the optimal whale individual by calculating the target optimization function, retaining the current optimal whale position, and mutating the optimal whale position, updating the whale position, and if the whale position is better than the optimal whale position, updating the current optimal whale position until the number of iterations t reaches the maximum number of iterations T max , output the optimal whale position, which is the power supply solution.

[0067] This embodiment takes into account the complexity of smart grid scheduling, as it involves multiple power supply devices and the objective function also includes multiple data types and large amounts of data. Therefore, the solution algorithm must possess strong optimization capabilities and high solution accuracy. Algorithms such as neural network algorithms, particle swarm algorithms, and genetic algorithms have drawbacks such as complex parameters, low optimization accuracy, long solution processes, and a tendency to fall into extreme values. Consequently, the optimal solutions obtained for solving grid scheduling problems are often poor.

[0068] Therefore, this embodiment uses the whale optimization algorithm with better optimization performance to solve the problem. The basic idea of ​​the whale algorithm is to find the optimal solution by simulating the behavior of whales. In the algorithm, each whale represents a solution, and the entire whale group represents the solution space. The algorithm updates the position of each whale by simulating the behavior of whales, thereby gradually approaching the optimal solution. The principle of this method is easy to understand, the initial parameter value is simple, the robustness is strong, and the convergence accuracy is accurate. It can generate the optimal scheduling plan while achieving the optimization goal.

[0069] Specifically, the algorithm parameters that need to be updated in the whale algorithm are the coefficient vectors g, o and the convergence factor a.

[0070] g=2ar1-a

[0071] o=2r2

[0072]

[0073] Among them, r1 and r2 are random numbers between [0,1].

[0074] In the standard whale algorithm, the convergence factor a decreases linearly from 2 to 0 with the number of iterations. This decreasing pattern causes the algorithm to tend to global search in the early stages, but the search for prey is not thorough and the convergence speed is slow. In the later stages of the iteration, it tends to local search, at which time the optimization rate is very high and it is easy to fall into the local optimum. Therefore, the method for updating the convergence factor a in this implementation is: Where s is a random number between (0, 1). This method reduces the convergence factor in proportion to the number of iterations, which can alleviate the impact of the above problems.

[0075] During the optimization process of the whale algorithm, each whale in the group moves towards the optimal whale, which will cause the diversity of the group to disappear. If the current optimal whale is a local optimal solution, all whales will be concentrated in the area of ​​the local optimal solution, causing the whale algorithm to converge early. To avoid this situation, this embodiment adopts the Cauchy mutation strategy to perform Cauchy interference on the current optimal whale position. The Cauchy operator can generate a variable step size. When an individual whale falls into a local optimum, the Cauchy operator will generate a larger step size to help the whale jump out of the local optimum.

[0076] Specifically, the method for mutating the optimal whale position is: Among them, is the new optimal whale position after Cauchy mutation, X * (t) is the optimal whale position in the tth generation, and cauchy(0,1) is the Cauchy operator.

[0077] The methods for updating whale positions based on their feeding behavior can be categorized into three types, depending on the values ​​of the probability factor p and the coefficient vector g: bubble net feeding, encirclement, and global search. Both the encirclement and bubble net feeding mechanisms rely on the optimal whale position. The encirclement mechanism involves the whale moving toward the optimal whale position, continuously encircling and shrinking its prey. The bubble net feeding mechanism involves the whale emitting countless bubbles and swimming in a spiral. The global search mechanism means that individual whales no longer update their positions based on the optimal whale position, but instead randomly search for prey based on their current position.

[0078] Specifically, if the probability factor p ≥ 50%, the whale position is updated according to the bubble net predation mechanism, which is an Archimedean spiral update. Specifically, the bubble net predation mechanism is: X(t+1) = D'cl·cos(2πl)+X*(t)l, where D' is the distance between the current whale position and the optimal whale position, c is a constant coefficient, and l is a random number between [-1, 1]. Compared with the logarithmic spiral update in the existing technology, the Archimedean spiral can fully search the nearby space and has more advantages when facing problems with dense local optimal solutions.

[0079] If the probability factor p < 50% and the coefficient vector |g| < 1, the whale position is updated according to the prey encirclement mechanism, specifically: X(t+1) = X*(t) - g·D1, D1 = |o·X*(t) - X(t)|, X(t) is the position of the individual whale in the tth generation;

[0080] If the probability factor p < 50% and the coefficient vector |g| ≥ 1, the position is updated according to the global search mechanism, specifically: X(t+1) = X rand (t)-g|o·X rand (t)-X(t)|, where X rand (t) is the position of a randomly selected whale individual in generation t.

[0081] In this embodiment, the position of the whale is the solution of the target optimization function under the constraint conditions, so the optimal whale position finally output is the optimal power supply solution for each power supply device that meets the optimization target under the constraint conditions.

[0082] In order to establish a more accurate prediction model, this embodiment further provides a method for generating predicted power consumption data within the second scheduling period based on the power consumption data, meteorological data, and predicted meteorological data of the power consumption area, including:

[0083] generating a plurality of sample matrices and a plurality of prediction matrices based on the meteorological data and the predicted meteorological data, wherein each sample matrix and each prediction matrix corresponds to a sub-time period, the sub-time periods of all the sample matrices constitute the first scheduling period, and the sub-time periods of all the prediction matrices constitute the second scheduling period;

[0084] The predicted power consumption data is determined based on all sample matrices and all prediction matrices.

[0085] As a further improvement of the present invention, determining the predicted power consumption data based on all sample matrices and all prediction matrices includes:

[0086] According to the comprehensive similarity index, the correlation between each prediction matrix and each sample matrix is ​​calculated respectively to obtain a similarity group corresponding to each prediction matrix, wherein the comprehensive similarity index is determined according to the grey relational degree and the cosine similarity, and the similarity group includes the first N sample matrices in the similarity index sequence, wherein the similarity index sequence is a sequence in which the sample matrices are sorted in descending order according to the similarity index with the prediction matrix, wherein N is a positive integer;

[0087] According to the time corresponding to each of the sample matrices in the similar group, obtaining the power consumption corresponding to each time in the power consumption data;

[0088] The predicted power consumption at the corresponding moment of the prediction matrix is ​​calculated based on the corresponding power consumption.

[0089] Meteorological data may include temperature, humidity, wind speed, etc.

[0090] For example, a scheduling period is set to include 24 hours, each sub-time period is 1 hour, and the sample matrix corresponding to the i-th hour (or the i-th moment) can be expressed as:

[0091]

[0092] Where T i,1 ~T i,60 represents the temperature per minute in the i-th hour, H i,1 ~H i,60 represents the humidity per minute in the i-th hour, B i,1 ~B i,60 Represents the wind speed per minute in the i-th hour, i = 1, 2, …, 24.

[0093] The prediction matrix for the jth hour (jth) can be expressed as:

[0094]

[0095] Where Tj,1 ~T j,60 represents the temperature per minute in the jth hour, H j,1 ~H j,60 represents the humidity per minute in the jth hour, B j,1 ~B j,60 Represents the wind speed per minute in the jth hour, j = 1, 2, …, 24.

[0096] The comprehensive similarity index is: S ij =qr ij +(1-qω ij ), S ij It represents the comprehensive similarity between the i-th sample matrix and the j-th prediction matrix. q is the empirical weight coefficient, which is determined based on the specific weather conditions. When meteorological factors fluctuate violently, the value of q should be close to 0, otherwise it should be close to 1. ij Represents the grey correlation between the i-th sample matrix and the j-th prediction matrix, which is used to measure the overall correlation between the two matrices. ij It represents the cosine similarity between the i-th sample matrix and the j-th prediction matrix, and is used to measure the similarity of the changing trends of the column vectors in the two matrices.

[0097] Specifically, M represents the number of meteorological features, that is, the number of column vectors. For example, the meteorological features selected in this embodiment are temperature, humidity and wind speed, that is, M=3, ∈ ij (m) represents the correlation coefficient between the mth meteorological feature in the i-th sample matrix and the mth meteorological feature in the j-th prediction matrix.

[0098] Specifically:

[0099]

[0100] where Y' j (m) represents the normalized value of the mth meteorological characteristic vector in the jth prediction matrix, Y' i (m) represents the normalized value of the mth meteorological feature in the i-th sample matrix, the mth meteorological feature vector is the m-th column vector, ψ is a resolution coefficient between 0 and 1, preferably ψ=0.5.

[0101]

[0102] where Y im and Y jm Represents the mth meteorological feature vector in the i-th sample matrix and the j-th prediction matrix.

[0103] According to the above steps, the comprehensive similarity index between the j-th prediction matrix and each sample matrix can be calculated, and the N sample matrices with the largest comprehensive similarity index are selected as the similarity group of the j-th prediction matrix. The number of selected sample matrices can be adjusted according to actual conditions.

[0104] According to the time corresponding to each sample matrix in the j-th prediction matrix similarity group, the power consumption corresponding to each time is obtained in the power consumption data. The average power consumption corresponding to each time is the predicted power consumption of the j-th hour. Repeat the above steps for each prediction matrix to obtain the predicted power consumption for 24 hours, that is, the predicted power consumption data within a scheduling cycle with the current time as the starting point of the interval.

[0105] This embodiment takes into account that meteorological factors are the key factors that cause changes in power load. Therefore, this embodiment can effectively obtain predicted power consumption data based on the correlation between meteorological data. The above prediction process is simple, the prediction speed is fast, and the extrapolation is good. It is suitable for short-term prediction. The above method can quickly obtain predicted power consumption data as the basis for subsequent data analysis.

[0106] Furthermore, in order to make the final power supply plan more reasonable, this embodiment establishes an optimization function for multiple scheduling objectives with the minimum scheduling cost, the minimum environmental governance cost, and the highest user satisfaction as scheduling objectives, specifically including:

[0107] The objective optimization function of scheduling cost: Where A is the total cost and I is the total benefit;

[0108] Objective optimization function of environmental governance cost: Where k represents the type of pollutant, K represents the total number of pollutant types, and C k represents the cost of treating type k pollutants, where d represents the dth moment in the total duration of the scheduling period, and D represents the total duration of the scheduling period;

[0109] Objective optimization function of user satisfaction: f c =λ1O1+λ2O2+λ3O3, where λ1, λ2 and λ3 represent weight coefficients, O1 represents the user's electricity consumption, O2 represents the user's electricity consumption time, and O3 represents the user's satisfaction.

[0110] Where A=A1+A2+A3-(A4+A5), where A1 represents the dispatching cost of configurable power resources, A2 represents the power purchase cost of the power grid, A3 represents the power supply operation and maintenance cost, A4 represents the power generation subsidy, and A5 represents the user charging fee. The total benefit is V d represents the interactive electricity price between the active distribution network and the main grid dispatching center at the dth moment, represents the total load value of the smart grid at the dth moment, and Δd represents the duration corresponding to each moment;

[0111] Specifically, the cost of purchasing electricity from the power grid where Ψ d It represents the interaction power between the active distribution network and the main grid dispatching center at the dth moment.

[0112] Power supply operation and maintenance costs where Γ1 d represents the wind power absorbed at the dth moment, Γ2 d represents the photovoltaic power consumed at the dth moment, Γ3 d represents the thermal power consumed at the dth moment, ι1 is the wind turbine operation and maintenance cost, ι2 is the photovoltaic operation and maintenance cost, and ι3 is the thermal power unit operation and maintenance cost.

[0113] The data used in this method can be obtained from the operation data of the smart grid in the power consumption area, wherein the operation data of the smart grid includes the operation data of each power supply device in the smart grid, and the operation data includes operation cost data and operation parameter data.

[0114] Furthermore, in order to reduce the computational difficulty, this embodiment further processes the optimization functions of multiple scheduling objectives. The specific steps are as follows: Figure 2 Shown, including:

[0115] Calculating the membership of the optimization functions of the plurality of scheduling objectives using a trapezoidal function to obtain the membership functions of the plurality of scheduling objectives;

[0116] The target optimization function is determined according to the membership function.

[0117] Specifically, the analytic hierarchy process may be used to calculate the weight coefficient of each membership function, and the target optimization function may be calculated based on the weight coefficient.

[0118] The role of the hierarchical analysis method is to extract complex problems, transform the overall comprehensive problem into a single simple problem, reduce the dimension, and thus perform mathematical modeling and solution through logical association. Therefore, this embodiment selects the hierarchical analysis method to calculate the weight coefficient. The specific steps of the hierarchical analysis method are common knowledge and will not be described in detail in this embodiment.

[0119] Preferably, in order to make the calculation results more accurate, the optimization function f of multiple scheduling objectives can be a 、f b and f c Normalization processing is performed to obtain normalized target optimization functions δ1, δ2 and δ3, and then the normalized target optimization functions are used to replace the optimization functions of multiple scheduling targets in the above steps for data analysis.

[0120] The normalized target optimization functions δ1, δ2, and δ3 are specifically:

[0121]

[0122] in is the scheduling cost before optimization, is the environmental governance cost before optimization, is the user satisfaction before optimization, and can be obtained from the operating cost data. It indicates that multi-objective optimization should meet the requirements of minimizing scheduling costs, minimizing environmental governance costs and maximizing user satisfaction.

[0123] At this time, the membership function δ i,min ≤δ i ≤δ i,max .

[0124] Among them, δ i,min and δ i,max They respectively represent the minimum setting value and the maximum setting value corresponding to the target optimization function of the i-th normalization process, which can be set according to historical data.

[0125] Furthermore, to ensure the safe operation of the power supply solution, it is necessary to determine constraints based on the operating principles of the smart grid. The operating principles of the smart grid include the operating principles of each power supply device. This embodiment further provides a method for determining constraints, including:

[0126] Output power constraints of each power supply device: in is the minimum output power of the i-th power supply device, is the maximum output power of the i-th power supply device, P i (d) represents the output power of the i-th power supply device at the d-th moment;

[0127] Output power constraints of smart grids: Among them, P grid (d) is the interaction power between the active distribution network and the main grid dispatching center at the dth moment, P loda (d) is the electric load power required at the dth moment, which is determined based on the predicted power consumption trend;

[0128] Pollutant emission constraints: E k (P i )≤L x , where E k (P i) is the mass of the kth type of pollutant emitted by the i-th power supply equipment, L x The maximum mass of the kth type of pollutant allowed to be discharged.

[0129] The data used in this method can also be obtained from the operation data of the smart grid in the power consumption area and the operation data of each power supply device in the smart grid.

[0130] This embodiment provides a method for dynamic power dispatch based on a smart grid. This method generates a power dispatch plan based on collected data. This plan, while meeting the operating principles of the smart grid and its power supply equipment, achieves the goals of minimizing dispatch costs, minimizing environmental governance costs, and maximizing user satisfaction. Compared to existing solutions, it is more accurate and effective. Furthermore, if the input data changes, the dispatch plan generated using this method will automatically change, thereby achieving dynamic power dispatch and avoiding the inconvenience of manually adjusting the dispatch plan.

[0131] Furthermore, the present invention provides a smart grid-based power dynamic dispatching system. Figure 4 Shown, including:

[0132] Collection module: collects power consumption data, meteorological data, and smart grid operation data of the power consumption area within a scheduling cycle with the current time as the end point of the interval. The scheduling cycle is 24 hours. Within a scheduling cycle with the current time as the starting point of the interval, obtains the predicted meteorological data of the power consumption area according to the meteorological forecast;

[0133] Prediction module: used to input the power consumption data, meteorological data and predicted meteorological data of the power consumption area into a preset prediction model, and the prediction model outputs the predicted power consumption data within a scheduling period with the current time as the starting point of the interval;

[0134] Analysis module: Using a whale algorithm to solve a preset target optimization function under multiple preset constraints to obtain a power supply plan for each power supply device of the smart grid within a scheduling cycle with the current time as the starting point of the interval, the multiple constraints are determined based on the predicted power consumption data and the working principle of the smart grid, and the target optimization function is determined based on the membership of the optimization functions of multiple scheduling objectives, and the membership is determined according to a trapezoidal function;

[0135] Transmission module: uploads the calculated power supply scheme to the control center, so that the control center issues corresponding power supply instructions to each power supply device to control the power supply of the device.

[0136] Specifically, the system includes a sensor and a cloud server. The sensor includes a collection module, and the cloud server includes a prediction module, an analysis module, and a transmission module.

[0137] The system can automatically collect data through sensors, process the data through cloud servers, and then output subsequent power supply plans. The power supply plans are uploaded to the control center so that the control center can issue corresponding power supply instructions to achieve dynamic power scheduling.

[0138] From the above analysis, it can be seen that the smart grid-based power dynamic scheduling method and system provided in the embodiment of the present application can generate a corresponding power supply plan through automatically collected data, thereby realizing dynamic scheduling of power. Compared with the existing technology, the present application uses more influencing factors and combines the target optimization function with the whale algorithm to generate a more accurate power supply plan.

[0139] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0140] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, 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 produce 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 flowcharts and / or block diagrams. 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.

[0141] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0142] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0143] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that improvements and modifications that do not depart from the principles of the present invention are within the scope of protection of the present invention.

Claims

1. A method for dynamic power dispatching based on smart grid, characterized in that: include: During a first scheduling cycle, power consumption data, meteorological data, and smart grid operation data of the power consumption area are collected, wherein the first scheduling cycle is a scheduling cycle with the current time as the end point of the interval; Acquiring forecasted meteorological data for the power consumption area within a second scheduling period, wherein the second scheduling period is a scheduling period starting from the current time; generating predicted power consumption data within the second scheduling period based on the power consumption data, meteorological data and predicted meteorological data of the power consumption area; A whale algorithm is used to solve a preset target optimization function under preset multiple constraints to obtain a power supply plan for each power supply device of the smart grid during the second scheduling period, wherein the multiple constraints are determined based on the predicted power consumption data and the operating principle of the smart grid, and the target optimization function is determined based on the membership of the optimization functions of multiple scheduling objectives, and the membership is determined according to a trapezoidal function; The calculated power supply scheme is uploaded to a control center, so that the control center issues corresponding power supply instructions to each power supply device to control the power supply of the device.

2. The method for dynamic power dispatch based on smart grid according to claim 1, characterized in that: The whale algorithm is used to solve the preset target optimization function under multiple preset constraints, including: Inputting operation data of the smart grid in the power consumption area, wherein the operation data of the smart grid includes operation data of each power supply device in the smart grid; Initialize the relevant parameters of the whale algorithm and set the maximum number of iterations T max , whale population size H, whale individual dimension v and probability factor p; An iterative operation is performed to output the power supply scheme.

3. The method for dynamic power dispatch based on smart grid according to claim 2, characterized in that: The iterative operation includes: Update the coefficient vector and convergence factor, find the optimal whale individual by calculating the target optimization function, retain the current optimal whale position, mutate the optimal whale position, update the whale position, and if the whale position is better than the optimal whale position, update the current optimal whale position until the number of iterations reaches the maximum number of iterations, output the current optimal whale position, and the power supply scheme is determined based on the current optimal whale position.

4. The method for dynamic power dispatch based on smart grid according to claim 3, characterized in that: The convergence factor decreases in proportion to the number of iterations during the updating process.

5. The method for dynamic power dispatch based on smart grid according to claim 4, characterized in that: The method for mutating the optimal whale position is Cauchy mutation.

6. The method for dynamic power dispatch based on smart grid according to claim 5, characterized in that: The updated whale position includes: If the probability factor p is less than 50%, the whale position is updated according to the bubble net predation mechanism, and the bubble net predation mechanism is an Archimedean spiral update; If the probability factor p ≥ 50% and the coefficient vector |g| < 1, the whale position is updated according to the prey encirclement mechanism; If the probability factor p≥50% and the coefficient vector |g|≥1, the position is updated according to the global search mechanism.

7. The method for dynamic power dispatch based on smart grid according to claim 1, characterized in that: Generating predicted power consumption data for the second scheduling period according to the power consumption data, meteorological data, and predicted meteorological data of the power consumption area includes: generating a plurality of sample matrices and a plurality of prediction matrices based on the meteorological data and the predicted meteorological data, wherein each sample matrix and each prediction matrix corresponds to a sub-time period, the sub-time periods of all the sample matrices constitute the first scheduling period, and the sub-time periods of all the prediction matrices constitute the second scheduling period; The predicted power consumption data is determined based on all sample matrices and all prediction matrices.

8. The method for dynamic power dispatch based on smart grid according to claim 7, characterized in that: The step of determining the predicted power consumption data based on all sample matrices and all prediction matrices includes: According to the comprehensive similarity index, the correlation between each prediction matrix and each sample matrix is ​​calculated respectively to obtain a similarity group corresponding to each prediction matrix, wherein the comprehensive similarity index is determined according to the grey relational degree and the cosine similarity, and the similarity group includes the first N sample matrices in the similarity index sequence, wherein the similarity index sequence is a sequence in which the sample matrices are sorted in descending order according to the similarity index with the prediction matrix, wherein N is a positive integer; According to the time corresponding to each of the sample matrices in the similar group, obtaining the power consumption corresponding to each time in the power consumption data; The predicted power consumption at the corresponding moment of the prediction matrix is ​​calculated based on the corresponding power consumption.

9. The method for dynamic power dispatch based on smart grid according to claim 1, characterized in that: The optimization functions of the multiple scheduling objectives include: a target optimization function of scheduling cost, a target optimization function of environmental governance cost, and a target optimization function of user satisfaction; the multiple constraint conditions include: the output power constraints of the various power supply devices, the output power constraints of the smart grid, and pollutant emission constraints.

10. A power dynamic dispatching system based on smart grid, characterized in that: include: Acquisition module: collects power consumption data, meteorological data, and smart grid operation data of the power consumption area during a first scheduling period, wherein the first scheduling period is a scheduling period with the current time as the end point of the interval; obtains predicted meteorological data of the power consumption area during a second scheduling period, wherein the second scheduling period is a scheduling period with the current time as the starting point of the interval; Prediction module: generating predicted power consumption data within the second scheduling period based on the power consumption data, meteorological data and predicted meteorological data of the power consumption area; An analysis module: employing a whale algorithm to solve a preset target optimization function under a plurality of preset constraints to obtain a power supply plan for each power supply device of the smart grid during the second scheduling period, wherein the plurality of constraints are determined based on the predicted power consumption data and the operating principle of the smart grid, and the target optimization function is determined based on membership of optimization functions of a plurality of scheduling objectives, wherein the membership is determined according to a trapezoidal function; Transmission module: uploads the calculated power supply scheme to the control center, so that the control center issues corresponding power supply instructions to each power supply device to control the power supply of the device.

Images