Virtual power plant topology method, system, device and medium based on energy consumption optimization

Abstract

The present invention discloses a virtual power plant topology method, system, device and medium based on energy consumption optimization, which relates to the technical field of virtual power plant energy consumption optimization, and includes the following steps: obtaining the power line information between the power equipment of the virtual power plant, and collecting the historical power transmission data and historical environmental data of the power lines; performing feature calculation and processing to establish an energy consumption feature vector and a transmission loss relationship model; obtaining the energy consumption feature vector at a future moment and inputting it into the transmission loss relationship model to obtain the predicted line loss information; and dynamically adjusting and optimizing the selection of the power lines of the virtual power plant. The present invention is used to solve the problem that the existing virtual power plant energy consumption optimization technology cannot dynamically predict the transmission loss while comprehensively considering the combined effects of power factors and environmental factors on the transmission loss, and cannot adjust the power lines in advance based on the real-time changing transmission loss to adapt to the dynamically changing operating environment.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plant energy consumption optimization, and specifically to a virtual power plant topology method, system, device and medium based on energy consumption optimization. Background Art

[0002] The virtual power plant energy consumption optimization technology refers to the aggregation and coordinated optimization of resources such as distributed energy, energy storage devices and electrical equipment by using advanced information and communication technologies and software systems, so as to improve energy utilization efficiency, enhance the stability and reliability of the power system, and achieve the lowest-cost and highest-benefit power production and supply.

[0003] When the existing virtual power plant energy consumption optimization technology optimizes the transmission loss of power lines, because the virtual power plant topology is mostly based on static design, the dynamic changes of various factors in the power transmission process are not considered; for example, when determining the line connection and node layout in the past, only the fixed power supply and demand were relied on, and the evaluation of transmission loss often only considered basic factors such as line resistance loss, ignoring the impact of real-time environmental conditions on transmission loss; for example, in the patent application with the publication number of CN116505595A, a power monitoring and dispatching management system based on a virtual power plant is disclosed. This solution monitors and dispatches power according to fixed power supply and demand, reducing transmission loss to a certain extent; however, the environmental factors are ignored and it cannot adapt to the dynamically changing operating environment; moreover, the existing virtual power plant energy consumption optimization technology often collects and adjusts in real time, leaving very limited time for the system to analyze and judge; and the data obtained in real time may be delayed and inaccurate; during the operation of the virtual power plant, the data collected by the sensors need to go through transmission, processing and other links before being used for decision-making; during this process, data delay may occur, resulting in low effectiveness of energy consumption optimization; therefore, when the existing virtual power plant energy consumption optimization technology optimizes the transmission loss of power lines, it cannot dynamically predict the transmission loss while comprehensively considering the combined effects of power factors and environmental factors on transmission loss, and adjust the power lines in advance based on the real-time changing transmission loss to adapt to the dynamically changing operating environment. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems in the prior art to a certain extent. By obtaining the power line information between the power equipment of the virtual power plant, collecting the historical power transmission data and historical environmental data of the power lines; and performing feature calculation and processing to establish an energy consumption feature vector and a power transmission loss relationship model; obtaining the energy consumption feature vector at a future moment and inputting it into the power transmission loss relationship model to obtain the predicted line loss information; and dynamically adjusting and optimizing the selection of the power lines of the virtual power plant; so as to solve the problem that the existing virtual power plant energy consumption optimization technology cannot dynamically predict the power transmission loss while comprehensively considering the combined effects of power factors and environmental factors on the power transmission loss, and adjust the power lines in advance based on the real-time changing power transmission loss to adapt to the dynamically changing operating environment.

[0005] To achieve the above object, in a first aspect, the present application provides a virtual power plant topology method based on energy consumption optimization, including the following steps:

[0006] Obtain the power line information between the power equipment of the virtual power plant, and collect the historical power transmission data and historical environmental data of the power lines;

[0007] Based on the historical power transmission data and historical environmental data, perform feature calculation and processing, and establish an energy consumption feature vector and a power transmission loss relationship model;

[0008] Obtain the energy consumption feature vector of the power line at a future moment, and input it into the power transmission loss relationship model to obtain the predicted line loss information;

[0009] Based on the predicted line loss information, dynamically adjust and optimize the selection of the power lines of the virtual power plant.

[0010] Further, obtaining the power line information between all the power equipment of the virtual power plant and collecting the historical power transmission data and historical environmental data of the power lines includes the following sub-steps:

[0011] For all the power equipment managed by the virtual power plant, obtain all the power lines that can be connected between any two power equipment, and obtain all the connection switch nodes on the power lines. Denote any power line segment without intermediate connection switch nodes and power equipment as a transmission line segment; repeat to obtain all the transmission line segments of the virtual power plant, denoted as the power line information;

[0012] For any transmission line segment, obtain the line length of the transmission line segment, and obtain the transmission loss, input voltage, and input current of the transmission line segment at a first time interval, and obtain the environmental temperature, environmental humidity, and rainfall at the location of the transmission line segment, and record the acquisition time, the first time interval is k1;

[0013] Mark the line length, transmission loss, input voltage, input current, output voltage, output current of the obtained transmission line segment and the acquisition time as the first transmission data, and mark the ambient temperature, ambient humidity, rainfall of the location where the obtained transmission line segment is located and the acquisition time as the first environmental data;

[0014] Repeatedly obtain the first transmission data of all transmission line segments, denoted as historical transmission data, and repeatedly obtain the first environmental data of all transmission line segments, denoted as historical environmental data.

[0015] Furthermore, the feature calculation and processing based on the historical transmission data and the historical environmental data include the following sub-steps:

[0016] Merge the historical transmission data and the historical environmental data according to the corresponding transmission line segments and acquisition times, denoted as the first transmission feature data;

[0017] For the ambient temperature, ambient humidity and rainfall at any time under the same transmission line segment in the first transmission feature data, calculate the comprehensive environmental feature through the first feature formula and calculate the environmental correlation feature through the second feature formula. The first feature formula is as follows: , where ES is the comprehensive environmental feature, T is the collected ambient temperature, H is the collected ambient humidity, R is the collected rainfall, Tr is the historical average temperature, Hr is the historical average humidity, and Rr is the historical average rainfall; the second feature formula is as follows: , where DH is the environmental correlation feature;

[0018] Repeatedly obtain the comprehensive environmental feature and the environmental correlation feature corresponding to all acquisition times under all transmission line segments in the first transmission feature data, and merge and store them with the corresponding transmission line segments and acquisition times, marked as the second transmission feature data;

[0019] Perform normalization processing on the second transmission feature data according to the data type, and scale the size of all data in the first transmission feature data to [0, 1]; after completion, obtain the third transmission feature data.

[0020] Furthermore, establishing the energy consumption feature vector and the transmission loss relationship model also includes the following sub-steps:

[0021] Based on the third transmission characteristic data, the input voltage, input current, ambient temperature, ambient humidity, rainfall, ambient comprehensive characteristics, and ambient correlation characteristics at the same acquisition moment of the same transmission line segment are combined with the line length of the corresponding transmission line segment to form an energy consumption feature vector, denoted as M = {m1, m2, m3, m4, m5, m6, m7, m8}, where m1, m2, m3, m4, m5, m6, m7, and m8 represent the line length, input voltage, input current, ambient temperature, ambient humidity, rainfall, ambient comprehensive characteristics, and ambient correlation characteristics in sequence;

[0022] Obtain the energy consumption feature vectors of all transmission line segments at all acquisition moments in the third transmission characteristic data, and combine and store the energy consumption feature vectors with the corresponding acquisition moments, transmission losses, output voltages, and output currents, which are marked as the fourth transmission characteristic data;

[0023] Use a multi-layer perceptron to construct a first relationship model. The first relationship model includes: a model input layer, a first hidden layer, a second hidden layer, and a model output layer; the model input layer includes 8 neurons, the model output layer includes 3 neurons, set the number of neurons in the first hidden layer as b1, and set the number of neurons in the second hidden layer as b2;

[0024] Use the fourth transmission characteristic data to train the first relationship model, and after completion, obtain the transmission loss relationship model.

[0025] Furthermore, obtaining the energy consumption feature vector of the power line at a future moment and inputting it into the transmission loss relationship model to obtain the line prediction loss information also includes the following sub-steps:

[0026] For any two power lines that can be connected by all power equipment, the input voltage and input current of the initial transmission line segments of all power lines that can be connected are respectively denoted as the initial voltage and the initial current;

[0027] Based on historical transmission data, obtain the initial voltage and initial current of the first time length, and sort them in chronological order from near to far, which are respectively denoted as the initial voltage sequence and the initial current sequence; the first time length is E1;

[0028] Sort the initial voltage sequence and the initial current sequence according to the data size respectively, which are respectively denoted as the voltage magnitude sequence and the current magnitude sequence. Remove the smallest f% data and the largest f% data of the voltage magnitude sequence and the current magnitude sequence respectively, and then calculate the average value and standard deviation of the remaining voltage magnitude sequence and current magnitude sequence respectively. Denote the average value and standard deviation corresponding to the remaining voltage magnitude sequence as WU and BU in sequence, and denote the average value and standard deviation corresponding to the remaining current magnitude sequence as WA and BA in sequence;

[0029] Perform anomaly screening processing on the initial voltage sequence and the initial current sequence respectively, including: denoting any initial voltage data in the initial voltage sequence as UR i , denoting any initial current data in the initial current sequence as AR j ; if |UR i - WU| > 3 * BU, then determine that UR i is voltage anomaly data, and use the average value of the two data closest to UR i in the time scales before and after UR i to replace UR i ; if |AR j - WA| > 3 * BA, then determine that AR j is current anomaly data, and use the average value of the two data closest to AR j in the time scales before and after AR j to replace AR j ; after completion, obtain the initial screened voltage sequence and the initial screened current sequence;

[0030] Denote any voltage data in the initial screened voltage sequence as U i , and denote any current data in the screened current sequence as A j ;

[0031] Calculate the average value of U i-1 , U i and U i+1 and replace U i ; calculate the average value of A j-1 , A j and A j+1 and replace A j ;

[0032] Repeat replacing all the data in the initial screened voltage sequence and the initial screened current sequence. After completion, obtain the initial smoothed voltage sequence and the initial smoothed current sequence.

[0033] Further, obtain the energy consumption feature vector of the power line at future moments and input it into the transmission loss relationship model to obtain the line prediction loss information, including the following sub-steps:

[0034] Denote any voltage data in the initial smoothed voltage sequence as PU i , and denote any current data in the initial smoothed current sequence as PA j ;

[0035] For the initial smoothed voltage sequence, repeatedly calculate the difference between adjacent voltage data through the voltage difference formula and average the obtained differences to obtain the voltage average change trend QU. The voltage difference formula is as follows: , where CUi Representing PU i+1 The difference from PU i ;

[0036] For the initial smoothed current sequence, repeatedly calculate the differences between adjacent current data through the current difference formula, and average the obtained differences to obtain the average current change trend QA. The current difference formula is as follows: , where CA j Representing PA j The difference from PA j ;

[0037] Set the second time length as E2, and calculate the average initial voltage and average initial current for the next second time length according to the first voltage formula and the first current formula respectively. The first voltage formula is as follows: , where YU represents the average initial voltage for the next second time length, and U1 represents the first voltage data in the initial screening voltage sequence. The first current formula is as follows: , where YA represents the average initial current for the next second time length, and A1 represents the first current data in the initial screening current sequence;

[0038] Based on historical environmental data, obtain the ambient temperature, ambient humidity, and rainfall of the most recent collection as the ambient temperature, ambient humidity, and rainfall for the next second time length, and perform feature calculation and processing. Then, combine them with the average initial voltage and average initial current to form the energy consumption feature vector of the initial transmission line segment for the next second time length, denoted as the prediction feature vector;

[0039] Input the prediction feature vector into the transmission loss relationship model to obtain the transmission loss, output voltage, and output current of the initial transmission line segment;

[0040] Use the output voltage and output current of the initial transmission line segment as the input voltage and input current of the next transmission line segment connected to the initial transmission line segment, construct the corresponding prediction feature vector, and input it into the transmission loss relationship model; obtain the corresponding transmission loss, output voltage, and output current;

[0041] Repeat to obtain the transmission losses of all transmission line segments included in all power lines that can be connected between the two power devices, denoted as the line prediction loss information.

[0042] Furthermore, the dynamic adjustment and optimization of the power line selection of the virtual power plant based on the line prediction loss information includes the following sub-steps:

[0043] For any two power devices, obtain all the power lines that can be connected, and obtain the corresponding line prediction loss information at the second time interval k2. Then calculate the total transmission loss of each power line based on the line prediction loss information, and compare their magnitudes. Select the power line with the minimum total transmission loss for power transmission.

[0044] In a second aspect, the present application provides a virtual power plant topology system based on energy consumption optimization, including a data collection module, a power transmission feature module, a loss prediction module, and a line adjustment module.

[0045] The data collection module is used to obtain the power line information between the power devices of the virtual power plant, and collect the historical power transmission data and historical environmental data of the power lines.

[0046] The power transmission feature module includes a feature calculation unit and a model establishment unit. The feature calculation unit performs feature calculation processing based on the historical power transmission data and historical environmental data. The model establishment unit is used to establish an energy consumption feature vector and a power transmission loss relationship model.

[0047] The loss prediction module is used to obtain the energy consumption feature vector of the power line at a future moment, and input it into the power transmission loss relationship model to obtain the line prediction loss information.

[0048] The line adjustment module dynamically adjusts and optimizes the selection of the power lines of the virtual power plant based on the line prediction loss information.

[0049] In a third aspect, the present application provides an electronic device, including a processor and a memory. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the above method are run.

[0050] In a fourth aspect, the present application provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method are run.

[0051] Advantages of the present invention: By obtaining the power line information between the power devices of the virtual power plant, and collecting the historical power transmission data and historical environmental data of the power lines; performing feature calculation processing based on the historical power transmission data and historical environmental data, and establishing an energy consumption feature vector and a power transmission loss relationship model; obtaining the energy consumption feature vector of the power line at a future moment, and inputting it into the power transmission loss relationship model to obtain the line prediction loss information; dynamically adjusting and optimizing the selection of the power lines of the virtual power plant based on the line prediction loss information; it is possible to comprehensively consider the combined influence of power factors and environmental factors on power transmission loss, dynamically predict power transmission loss, and adjust the power lines in advance based on the real-time changing power transmission loss to adapt to the dynamically changing operating environment.

[0052] The present invention collects temperature, humidity, and rainfall, and calculates the comprehensive environmental characteristics and environmental correlation characteristics, so that when optimizing energy consumption, it is no longer limited to static line parameters, and can better make relatively accurate loss predictions in different environments to adapt to the dynamically changing operating environment;

[0053] By predicting future transmission losses, the advantage is that the topology adjustment of the virtual power plant is made more stable, reducing the disturbance to the normal operation of the power system, allowing relevant personnel to have sufficient time for resource allocation and planning, which helps to reduce operating costs and improve economic benefits; by performing abnormal screening on voltage and current parameters and predicting future voltage and current, the advantage is that it effectively avoids the influence of abnormal values on the prediction accuracy, effectively processes the fluctuation characteristics of the data, and improves the accuracy and adaptability of the prediction results. Description of the Drawings

[0054] Figure 1 is the principle block diagram of the system of the present invention;

[0055] Figure 2 is the step flowchart of the method of the present invention;

[0056] Figure 3 is the schematic diagram of the transmission line segment of the present invention;

[0057] Figure 4 is the flowchart of obtaining the prediction feature vector of the present invention;

[0058] Figure 5 is the schematic diagram of the structure of the electronic device of the present invention. Detailed Embodiments

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0060] Example 1, please refer to Figure 1 As shown, the present application provides a virtual power plant topology system based on energy consumption optimization, including a data collection module, a transmission line feature module, a loss prediction module, and a line adjustment module;

[0061] The data collection module is used to obtain the power line information between the power equipment of the virtual power plant and collect the historical transmission data and historical environmental data of the power line;

[0062] The data collection module is configured with a data collection strategy, and the data collection strategy includes: Please refer to Figure 3 As shown, for all power equipment managed by the virtual power plant, obtain all connectable power lines between any two power equipment, and obtain all tie switch nodes on the power lines. Denote any power line segment without tie switch nodes and power equipment in the middle as a transmission line segment; repeat to obtain all transmission line segments of the virtual power plant, denoted as power line information; the tie switch node refers to the position where the tie switch is located in the virtual power plant grid topology, which is a key node connecting different lines, subnets or power regions; the tie switch node is the intersection point of multiple power lines and can achieve flexible switching between lines;

[0063] For any transmission line segment, obtain the line length of the transmission line segment, that is, the length of the transmission wire, and obtain the transmission loss, input voltage and input current of the transmission line segment at a first time interval, and obtain the ambient temperature, ambient humidity and rainfall at the location where the transmission line segment is located, and record the acquisition time. The first time interval is k1; in this embodiment, the first time interval k1 is 1 minute;

[0064] Mark the line length, transmission loss, input voltage, input current, output voltage, output current of the obtained transmission line segment and the acquisition time as the first transmission data, and mark the ambient temperature, ambient humidity, rainfall at the location where the obtained transmission line segment is located and the acquisition time as the first environmental data;

[0065] Repeat to obtain the first transmission data of all transmission line segments, denoted as historical transmission data, and repeat to obtain the first environmental data of all transmission line segments, denoted as historical environmental data;

[0066] In the specific implementation process, generally, the resistivity of metal wires changes with the ambient temperature, which leads to changes in transmission loss. The ambient temperature may affect the performance of the insulating materials of the transmission line; being in a high temperature for a long time causes the insulation resistance to decrease and the leakage current to increase, resulting in additional losses. Moreover, the ambient temperature also affects the heat dissipation of relevant equipment during the power transmission process; if the ambient temperature is too high, the heat dissipation efficiency will decrease, and its internal temperature will rise, leading to an increase in losses; the ambient humidity mainly affects the insulation performance of the transmission line; when the air humidity is high, it will reduce the surface resistance of the insulator and increase the leakage current, thereby increasing the transmission loss; rainfall will directly make the surface of the transmission line get wet, seriously reducing the insulation performance of the line, increasing the leakage current, and increasing the transmission loss. And during the rainfall process, the content of impurities and ions in the air will increase, resulting in an increase in the air conductivity; this will change the electric field distribution around the transmission line and increase the corona loss.

[0067] The power transmission feature module includes a feature calculation unit and a model establishment unit. The feature calculation unit performs feature calculation processing based on historical power transmission data and historical environmental data. The model establishment unit is used to establish an energy consumption feature vector and a power transmission loss relationship model;

[0068] The feature calculation unit is configured with a feature calculation strategy, which includes: merging the historical power transmission data and historical environmental data according to the corresponding power transmission line segments and acquisition times, and recording it as the first power transmission feature data;

[0069] For the environmental temperature, environmental humidity, and rainfall at any time under the same power transmission line segment in the first power transmission feature data, calculate the environmental comprehensive feature through the first feature formula and calculate the environmental correlation feature through the second feature formula. The first feature formula is as follows: , where ES is the environmental comprehensive feature, T is the collected environmental temperature, H is the collected environmental humidity, R is the collected rainfall, Tr is the historical average temperature, Hr is the historical average humidity, and Rr is the historical average rainfall; the historical average temperature, historical average humidity, and historical average rainfall can select data of the same month as the acquisition time in previous years; the second feature formula is as follows: , where DH is the environmental correlation feature;

[0070] The environmental comprehensive feature measures the deviation degree of the current environmental state from the historical average environmental state; during power transmission, when the environmental conditions deviate from the average value more, the comprehensive impact on the power transmission line is greater; for example, too high or too low temperature, too high humidity or too large rainfall may all have an adverse impact on the insulation performance, wire resistance, etc. of the power transmission line, thereby increasing the power transmission loss; this feature helps the model capture the impact of abnormal changes in environmental conditions on the power transmission loss.

[0071] The environmental correlation feature comprehensively considers the non-linear relationship between humidity, rainfall, and temperature; humidity and rainfall are related to the moisture content, and temperature not only affects the physical state of moisture but also affects the physical characteristics of the power transmission line; for example, in a high-temperature environment, the increase in humidity and rainfall may exacerbate the oxidation corrosion of the wire or make the moisture on the surface of the insulator more likely to form a conductive path, thereby increasing the power transmission loss; this feature reflects the potential impact of this complex dynamic correlation on the power transmission loss by dividing the product of humidity and rainfall by the square of the temperature.

[0072] Repeatedly obtain the environmental comprehensive feature and environmental correlation feature corresponding to all acquisition times under all power transmission line segments in the first power transmission feature data, and merge and store them with the corresponding power transmission line segments and acquisition times, and mark them as the second power transmission feature data;

[0073] Normalize the second transmission feature data according to the data type respectively, and scale the magnitudes of all the data in the first transmission feature data to [0, 1]; after completion, the third transmission feature data is obtained; the normalization formula is as follows: , where X0 is the data after normalization, X is the data before normalization, Xmi is the minimum value of the same type of data corresponding to X, and Xma is the maximum value of the same type of data corresponding to X;

[0074] The model establishment unit is configured with a model establishment strategy, and the model establishment strategy includes: based on the third transmission feature data, combine the input voltage, input current, ambient temperature, ambient humidity, rainfall, ambient comprehensive feature, and ambient correlation feature at the same acquisition moment of the same transmission line segment, plus the line length of the corresponding transmission line segment to form an energy consumption feature vector, denoted as M = {m1, m2, m3, m4, m5, m6, m7, m8}, where m1, m2, m3, m4, m5, m6, m7, and m8 represent the line length, input voltage, input current, ambient temperature, ambient humidity, rainfall, ambient comprehensive feature, and ambient correlation feature in sequence;

[0075] Obtain the energy consumption feature vectors of all transmission line segments at all acquisition moments in the third transmission feature data, and merge and store the energy consumption feature vectors with the corresponding acquisition moments, transmission losses, output voltages, and output currents, marked as the fourth transmission feature data;

[0076] Construct a first relationship model using a multi-layer perceptron. The first relationship model includes: a model input layer, a first hidden layer, a second hidden layer, and a model output layer; the model input layer includes 8 neurons, and there are 8 feature components in the energy consumption feature vector M. The model output layer includes 3 neurons for outputting the transmission loss, output voltage, and output current. Set the number of neurons in the first hidden layer as b1, and set the number of neurons in the second hidden layer as b2; in this embodiment, b1 = b2 = 24. The number of hidden layers and the number of neurons in the hidden layers of the first relationship model can be set according to the actual application scenario, and appropriate numbers of hidden layers and neurons in each hidden layer can be selected; generally, you can start with a simple structure, and you can set one to three hidden layers and try the number of neurons between 10 and 100; for example, first set one hidden layer with 30 neurons;

[0077] Train the first relationship model using the fourth transmission feature data, and after completion, obtain the transmission loss relationship model;

[0078] In the specific implementation process, by calculating the comprehensive environmental characteristics and environmental correlation characteristics, these interrelated factors can be incorporated into the subsequent model in a comprehensive manner, more accurately reflecting the internal relationship between environmental factors and transmission loss; compared with the model that only considers a single factor, it can predict transmission loss more accurately. Secondly, in terms of the generalization ability of the subsequent model, considering these combined characteristics helps the model adapt to different environmental conditions. Since the environment where the virtual power plant is located is complex and changeable, it is difficult for a simple fixed-parameter model to adapt to various weather conditions. By introducing the comprehensive environmental characteristics and environmental correlation characteristics, the model can learn the loss change rules under different combinations of environmental factors, so that when facing new and unseen combinations of weather data, it can also reasonably predict transmission loss. Finally, in terms of the dynamic adaptability of the model, over time and with the change of the environment, the comprehensive environmental characteristics and environmental correlation characteristics can timely reflect the dynamic impact of environmental changes on loss. The operating environment of the virtual power plant is not static. By continuously collecting data and updating the combined characteristics, the model can adjust the prediction of transmission loss in real time, better track and adapt to the changes in actual operation, and provide strong support for the energy consumption optimization of the virtual power plant.

[0079] The loss prediction module is used to obtain the energy consumption feature vector of the power line at a future moment and input it into the transmission loss relationship model to obtain the predicted line loss information.

[0080] The loss prediction module is configured with a loss prediction strategy, and the loss prediction strategy includes: Please refer to Figure 4 As shown, for all power lines that can be connected between any two power devices, the input voltage and input current of the initial transmission line segments of all power lines that can be connected are respectively denoted as the initial voltage and the initial current; that is, the initial input voltage and input current of two connected power devices. For example, when a power generation device and a power consumption device are connected, the initial voltage and the initial current are the voltage and current output by the power generation device.

[0081] Based on historical transmission data, the initial voltage and the initial current for the first time length are obtained and sorted in chronological order from near to far, and are respectively denoted as the initial voltage sequence and the initial current sequence; the first time length is E1; in this embodiment, E1 is two hours.

[0082] Sort the initial voltage sequence and the initial current sequence according to the data size respectively, and denote them as the voltage magnitude sequence and the current magnitude sequence. Remove the smallest f% of the data and the largest f% of the data from the voltage magnitude sequence and the current magnitude sequence respectively. In this embodiment, f% is 5%, that is, keep the middle 90% of the data. For example, a certain current magnitude sequence is {4.3, 5.5, 5.6, 5.6, 5.8, 5.8, 5.8, 5.8, 5.9, 5.9, 6.0, 6.0, 6.1, 6.2, 6.3, 6.3, 6.3, 6.4, 6.4, 7.5}, the total number is 20, f% is 5%, 20 * 5% = 1, then remove the smallest 1 data and the largest 1 data, that is, 4.3 and 7.5. By discarding 5% of the data at both ends, this method can effectively avoid the interference of extreme outliers on the calculation of the standard deviation. In the data, occasionally there may be extremely large or extremely small data points caused by sensor failures, sudden electromagnetic interference or other temporary factors. If the standard deviation is calculated using all the data, these extreme values will increase the standard deviation, resulting in too loose criteria for subsequent outlier judgment. However, using the middle 90% of the data to calculate the standard deviation can exclude the influence of these extreme values, thereby ensuring the accuracy of subsequent data screening and prediction.

[0083] Then calculate the mean and standard deviation of the remaining voltage magnitude sequence and current magnitude sequence respectively. Denote the mean and standard deviation corresponding to the remaining voltage magnitude sequence as WU and BU in order, and denote the mean and standard deviation corresponding to the remaining current magnitude sequence as WA and BA in order.

[0084] Perform outlier screening processing on the initial voltage sequence and the initial current sequence respectively, including: Denote any initial voltage data in the initial voltage sequence as UR i , and denote any initial current data in the initial current sequence as AR j ; If |UR i - WU| > 3 * BU, then determine that UR i is voltage abnormal data, and use the average value of the two data closest to UR i in the time scales before and after UR i to replace UR i ; If |AR j - WA| > 3 * BA, then determine that AR j is current abnormal data, and use the average value of the two data closest to AR j in the time scales before and after AR j to replace AR jReplace; after completion, obtain the initial screening voltage sequence and the initial screening current sequence; for example, if AR5 = 4.1, WA = 6.1, BA = 0.6, and |AR5 - WA| > 3 * BA, then AR5 is abnormal current data. If AR4 and AR6 are not abnormal current data, the average value of AR4 and AR6 can be calculated for replacement; when calculating the average value, the two closest data are preferably on both sides of the abnormal data respectively.

[0085] Record any voltage data in the initial screening voltage sequence as U i , and record any current data in the screening current sequence as A j ;

[0086] Calculate the average value of U i-1 , U i and U i+1 and replace U i ; calculate the average value of A j-1 , A j and A j+1 and replace A j ; for example, if A4 = 5.9, A5 = 6.2, A6 = 6.2, then (5.9 + 6.2 + 6.2) / 3 = 6.1, and 6.1 is used to replace A5 = 6.2, that is, the replaced A5 = 6.1;

[0087] Repeat replacing all the data in the initial screening voltage sequence and the initial screening current sequence. After completion, obtain the initial smoothed voltage sequence and the initial smoothed current sequence;

[0088] Record any voltage data in the initial smoothed voltage sequence as PU i , and record any current data in the initial smoothed current sequence as PA j ;

[0089] For the initial smoothed voltage sequence, repeatedly calculate the difference between adjacent voltage data through the voltage difference formula, and average the obtained differences to obtain the average voltage change trend QU. The voltage difference formula is as follows: , where CU i represents the difference between PU i+1 and PU i ;

[0090] For the initial smoothed current sequence, repeatedly calculate the difference between adjacent current data through the current difference formula, and average the obtained differences to obtain the average current change trend QA. The current difference formula is as follows: , where CA j represents the difference between PA j and PA j ;

[0091] For example, for an initial smoothed current sequence of 6.2, 6.1, 6.3, 6.4, 6.3, 6.5, then 6.1 - 6.2 = -0.1, 6.3 - 6.2 = 0.1, 6.4 - 6.3 = 0.1, 6.3 - 6.4 = -0.1, 6.5 - 6.3 = 0.2, and the average current change trend QA = [-0.1 + 0.1 + 0.1 + (-0.1) + 0.2] / 5 = 0.04;

[0092] Set the second time length to E2. In this embodiment, E2 is 0.5 hours. Calculate the average initial voltage and average initial current for the next second time length according to the first voltage formula and the first current formula respectively. The first voltage formula is as follows: , where YU represents the average initial voltage for the next second time length, and U1 represents the first voltage data in the initial screened voltage sequence. The first current formula is as follows: , where YA represents the average initial current for the next second time length, and A1 represents the first current data in the initial screened current sequence;

[0093] Based on historical environmental data, obtain the most recently collected environmental temperature, environmental humidity, and rainfall as the environmental temperature, environmental humidity, and rainfall for the next second time length. Since the changes in environmental temperature, environmental humidity, and rainfall are relatively slow, when E2 is set to a relatively small value, for example, E2 = 0.5 hours, the most recently collected environmental temperature, environmental humidity, and rainfall can be used as the environmental temperature, environmental humidity, and rainfall for the next 0.5 hours. If E2 is set to a larger value, the future environmental temperature, environmental humidity, and rainfall can be obtained through weather forecasts; then perform feature calculation and processing, and together with the average initial voltage and average initial current, form the energy consumption feature vector for the initial transmission line segment for the next second time length, denoted as the prediction feature vector;

[0094] Input the prediction feature vector into the transmission loss relationship model to obtain the transmission loss, output voltage, and output current of the initial transmission line segment;

[0095] Use the output voltage and output current of the initial transmission line segment as the input voltage and input current of the next transmission line segment connected to the initial transmission line segment, construct the corresponding prediction feature vector, and input it into the transmission loss relationship model; obtain the corresponding transmission loss, output voltage, and output current;

[0096] Repeatedly obtain the transmission losses of all transmission line segments included in all power lines that can be connected between two power devices, which is denoted as line prediction loss information; since there are often multiple power lines that can be connected between two power devices, for some transmission line segments of some power lines, there may be no latest input voltage and input current for reference because they are not commonly used. However, the initial starting points of all power lines are the same, that is, the initial input voltage and input current are the same. And the environmental data of each transmission line segment can be collected in real time, and the line length remains unchanged. Therefore, the transmission losses of all transmission line segments can be obtained by gradually predicting downward along the direction of the power line from the initial input voltage and input current, so as to traverse all power lines.

[0097] In the specific implementation process, the second time length E2 is the predicted time length, and the first time length E1 is the historical time length used for prediction. E1 > E2, and the two settings should correspond to each other. For example, to predict data for the next 0.5 hours, 2 hours of previous data are required.

[0098] The line adjustment module dynamically adjusts and optimizes the power line selection of the virtual power plant based on the line prediction loss information.

[0099] The line adjustment module is configured with a line adjustment strategy. The line adjustment strategy includes: for any two power devices, obtain all power lines that can be connected, and obtain the corresponding line prediction loss information at the second time interval. Then calculate the total transmission loss of each power line based on the line prediction loss information, and compare their magnitudes. Select the power line with the minimum total transmission loss for power transmission. The second time interval is k2. In this embodiment, the second time interval k2 is 0.5 hours, which is the same as the predicted time length. After each prediction, select the power line with the minimum total transmission loss for power transmission.

[0100] In the specific implementation process, by making early adjustments based on predictions and always selecting the line with the minimum loss for transmission, the loss of electric energy during transmission can be directly reduced, which helps to balance the load of each line; because the line with the minimum loss can usually transmit electric energy more efficiently within its carrying capacity, avoiding the situation where some lines or devices are overloaded due to long-term high-load operation; overload may cause the line or device to overheat, accelerate insulation aging, or even cause failures. By reasonably selecting the line, this risk can be reduced and the stable operation of the power system can be ensured.

[0101] Embodiment 2, please refer to Figure 2 As shown, the present application provides a virtual power plant topology method based on energy consumption optimization, including the following steps:

[0102] Step S1: Obtain the power line information among the power equipment of the virtual power plant, and collect the historical power transmission data and historical environmental data of the power lines. Step S1 includes the following sub-steps:

[0103] Step S101: For all the power equipment managed by the virtual power plant, obtain all the power lines that can be connected between any two power equipment, and obtain all the tie switch nodes on the power lines. Denote any power line segment without tie switch nodes and power equipment in the middle as a transmission line segment. Repeat to obtain all the transmission line segments of the virtual power plant, denoted as power line information;

[0104] Step S102: For any transmission line segment, obtain the line length of the transmission line segment, and obtain the transmission loss, input voltage, and input current of the transmission line segment at the first time interval. Also obtain the ambient temperature, ambient humidity, and rainfall at the location of the transmission line segment, and record the acquisition time. The first time interval is k1;

[0105] Step S103: Mark the line length, transmission loss, input voltage, input current, output voltage, output current, and acquisition time of the obtained transmission line segment as the first power transmission data, and mark the ambient temperature, ambient humidity, rainfall, and acquisition time at the location of the obtained transmission line segment as the first environmental data;

[0106] Step S104: Repeat to obtain the first power transmission data of all transmission line segments, denoted as historical power transmission data, and repeat to obtain the first environmental data of all transmission line segments, denoted as historical environmental data.

[0107] Step S2: Perform feature calculation and processing based on the historical power transmission data and historical environmental data, and establish an energy consumption feature vector and a power transmission loss relationship model. Step S2 includes the following sub-steps:

[0108] Step S201: Merge the historical power transmission data and historical environmental data according to the corresponding transmission line segments and acquisition times, denoted as the first power transmission feature data;

[0109] Step S202: For the ambient temperature, ambient humidity, and rainfall at any time under the same transmission line segment in the first power transmission feature data, calculate the comprehensive environmental feature through the first feature formula and calculate the environmental correlation feature through the second feature formula. The first feature formula is as follows: , where ES is the comprehensive environmental feature, T is the collected ambient temperature, H is the collected ambient humidity, R is the collected rainfall, Tr is the historical average temperature, Hr is the historical average humidity, and Rr is the historical average rainfall; The second feature formula is as follows: , where DH is the environmental correlation feature;

[0110] Step S203: Repeatedly obtain the comprehensive environmental features and environmental correlation features corresponding to all acquisition times under all transmission line segments in the first transmission feature data, and merge and store them with the corresponding transmission line segments and acquisition times, marked as the second transmission feature data;

[0111] Step S204: Perform normalization processing on the second transmission feature data according to the data type respectively, and scale the magnitudes of all data in the first transmission feature data to [0, 1]; after completion, obtain the third transmission feature data;

[0112] Step S205: Based on the third transmission feature data, combine the input voltage, input current, ambient temperature, ambient humidity, rainfall, comprehensive environmental features, and environmental correlation features at the same acquisition time of the same transmission line segment, plus the line length of the corresponding transmission line segment to form an energy consumption feature vector, denoted as M = {m1, m2, m3, m4, m5, m6, m7, m8}, where m1, m2, m3, m4, m5, m6, m7, and m8 represent the line length, input voltage, input current, ambient temperature, ambient humidity, rainfall, comprehensive environmental features, and environmental correlation features in sequence;

[0113] Step S206: Obtain the energy consumption feature vectors at all acquisition times of all transmission line segments in the third transmission feature data, and merge and store the energy consumption feature vectors with the corresponding acquisition times, transmission losses, output voltages, and output currents, marked as the fourth transmission feature data;

[0114] Step S207: Use a multi-layer perceptron to construct a first relationship model. The first relationship model includes: a model input layer, a first hidden layer, a second hidden layer, and a model output layer; the model input layer includes 8 neurons, the model output layer includes 3 neurons, set the number of neurons in the first hidden layer as b1, and set the number of neurons in the second hidden layer as b2;

[0115] Step S208: Use the fourth transmission feature data to train the first relationship model, and after completion, obtain the transmission loss relationship model.

[0116] Step S3: Obtain the energy consumption feature vector of the power line at a future time, and input it into the transmission loss relationship model to obtain the line prediction loss information; Step S3 includes the following sub-steps:

[0117] Step S301: For all power lines that can be connected between any two power devices, denote the input voltage and input current of the initial transmission line segments of all power lines that can be connected as the initial voltage and the initial current respectively;

[0118] Step S302: Obtain the initial voltage and initial current for the first time length based on historical power transmission data, and sort them in chronological order from near to far, denoted as the initial voltage sequence and the initial current sequence respectively; the first time length is E1.

[0119] Step S303: Sort the initial voltage sequence and the initial current sequence respectively according to the data size, denoted as the voltage magnitude sequence and the current magnitude sequence respectively, and remove the smallest f% data and the largest f% data of the voltage magnitude sequence and the current magnitude sequence respectively.

[0120] Step S304: Then calculate the average value and standard deviation of the remaining voltage magnitude sequence and current magnitude sequence respectively. Denote the average value and standard deviation corresponding to the remaining voltage magnitude sequence as WU and BU in order, and denote the average value and standard deviation corresponding to the remaining current magnitude sequence as WA and BA in order.

[0121] Step S305: Perform anomaly screening processing on the initial voltage sequence and the initial current sequence respectively, including: Denote any initial voltage data in the initial voltage sequence as UR i , and denote any initial current data in the initial current sequence as AR j ; If |UR i - WU| > 3 * BU, then determine that UR i is voltage anomaly data, and use the average value of the two data closest to UR i in the time scales before and after UR i to replace UR i .

[0122] Step S306: If |AR j - WA| > 3 * BA, then determine that AR j is current anomaly data, and use the average value of the two data closest to AR j in the time scales before and after AR j to replace AR j ; After completion, obtain the initial screened voltage sequence and the initial screened current sequence.

[0123] Step S307: Denote any voltage data in the initial screened voltage sequence as U i , and denote any current data in the screened current sequence as A j ;

[0124] Step S308: Calculate the average value of U i-1 , U i and U i+1 and replace U i ; Calculate the value related to A j-1 , Aj and A j+1 Take the average value of and replace A j ;

[0125] Step S309: Repeatedly replace all the data in the initial screening voltage sequence and the initial screening current sequence. After completion, obtain the initial smoothed voltage sequence and the initial smoothed current sequence;

[0126] Step S310: Denote any voltage data in the initial smoothed voltage sequence as PU i , and denote any current data in the initial smoothed current sequence as PA j ;

[0127] Step S311: For the initial smoothed voltage sequence, repeatedly calculate the difference between adjacent voltage data through the voltage difference formula, and take the average of the obtained differences to get the average voltage change trend QU. The voltage difference formula is as follows: , where CU i represents the difference between PU i+1 and PU i ;

[0128] Step S312: For the initial smoothed current sequence, repeatedly calculate the difference between adjacent current data through the current difference formula, and take the average of the obtained differences to get the average current change trend QA. The current difference formula is as follows: , where CA j represents the difference between PA j and PA j ;

[0129] Step S313: Set the second time length as E2, and calculate the average initial voltage and the average initial current for the next second time length respectively according to the first voltage formula and the first current formula. The first voltage formula is as follows: , where YU represents the average initial voltage for the next second time length, and U1 represents the first voltage data in the initial screening voltage sequence; The first current formula is as follows: , where YA represents the average initial current for the next second time length, and A1 represents the first current data in the initial screening current sequence;

[0130] Step S314: Based on the historical environmental data, obtain the most recently collected environmental temperature, environmental humidity, and rainfall as the environmental temperature, environmental humidity, and rainfall for the next second time length, and perform feature calculation and processing. Then, together with the average initial voltage and the average initial current, form the energy consumption feature vector of the initial transmission line segment for the next second time length, denoted as the prediction feature vector;

[0131] Step S315: Input the predicted feature vector into the transmission loss relationship model to obtain the transmission loss, output voltage, and output current of the initial transmission line segment.

[0132] Step S316: Use the output voltage and output current of the initial transmission line segment as the input voltage and input current of the next transmission line segment connected to the initial transmission line segment, construct the corresponding predicted feature vector, and input it into the transmission loss relationship model to obtain the corresponding transmission loss, output voltage, and output current.

[0133] Step S317: Repeat to obtain the transmission losses of all transmission line segments included in all power lines that can be connected between two power devices, which is recorded as the line prediction loss information.

[0134] Step S4: Dynamically adjust and optimize the selection of power lines in the virtual power plant based on the line prediction loss information. Step S4 includes the following sub-steps:

[0135] Step S401: For any two power devices, obtain all power lines that can be connected, and obtain the corresponding line prediction loss information at the second time interval.

[0136] Step S402: Calculate the total transmission loss of each power line according to the line prediction loss information, then compare their magnitudes, and select the power line with the minimum total transmission loss for power transmission. The second time interval is k2.

[0137] Example 3, please refer to Figure 5 as shown in Figure 5 illustrates a schematic structural diagram of an electronic device. The electronic device may include: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, it runs the steps in the virtual power plant topology method based on energy consumption optimization to achieve the following functions: obtain the power line information between the power devices in the virtual power plant, and collect the historical transmission data and historical environmental data of the power lines; perform feature calculation and processing based on the historical transmission data and historical environmental data, and establish an energy consumption feature vector and a transmission loss relationship model; obtain the energy consumption feature vector of the power line at a future moment, and input it into the transmission loss relationship model to obtain the line prediction loss information; dynamically adjust and optimize the selection of power lines in the virtual power plant based on the line prediction loss information.

[0138] In addition, when the logical instructions in the above-mentioned memory are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0139] Embodiment 4, this application also provides a computer-readable storage medium. This application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, it runs the steps in the virtual power plant topology method based on energy consumption optimization as described above to achieve the following functions: obtaining the power line information between the power equipment of the virtual power plant, and collecting the historical power transmission data and historical environmental data of the power lines; performing feature calculation processing based on the historical power transmission data and historical environmental data, and establishing an energy consumption feature vector and a power transmission loss relationship model; obtaining the energy consumption feature vector of the power line at a future moment and inputting it into the power transmission loss relationship model to obtain the line prediction loss information; dynamically adjusting and optimizing the selection of the power lines of the virtual power plant based on the line prediction loss information.

[0140] Through the description of the above embodiments, the embodiments of the present invention can be provided as a method, a system, or a computer program product. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disks, optical discs, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments.

[0141] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. For another example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of systems, modules, and units can be electrical, mechanical, or other forms.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A virtual power plant topology method based on energy consumption optimization, characterized in that, It includes the following steps: Obtain the power line information between the power equipment of the virtual power plant, and collect the historical power transmission data and historical environmental data of the power lines; Conduct feature calculation and processing based on the historical power transmission data and historical environmental data, and establish an energy consumption feature vector and a power transmission loss relationship model; Merge the historical power transmission data and historical environmental data according to the corresponding power line segments and acquisition times, and denote it as the first power transmission feature data; For the ambient temperature, ambient humidity, and rainfall at any moment under the same transmission line segment in the first power transmission characteristic data, calculate the ambient comprehensive characteristic through the first characteristic formula and calculate the ambient correlation characteristic through the second characteristic formula. The first characteristic formula is as follows: , where ES is the ambient comprehensive characteristic, T is the collected ambient temperature, H is the collected ambient humidity, R is the collected rainfall, Tr is the historical average temperature, Hr is the historical average humidity, and Rr is the historical average rainfall; the second characteristic formula is as follows: , where DH is the ambient correlation characteristic; Repeatedly obtain the environmental comprehensive features and environmental correlation features corresponding to all acquisition times under all power line segments in the first power transmission feature data, and merge and store them with the corresponding power line segments and acquisition times, and mark them as the second power transmission feature data; Perform normalization processing on the second power transmission feature data according to the data type, and scale the magnitudes of all data in the first power transmission feature data to [0, 1]; after completion, obtain the third power transmission feature data; Based on the third power transmission feature data, combine the input voltage, input current, environmental temperature, environmental humidity, rainfall, environmental comprehensive features, and environmental correlation features at the same acquisition time of the same power line segment, plus the line length of the corresponding power line segment to form an energy consumption feature vector, denoted as M={m1, m2, m3, m4, m5, m6, m7, m8}, where m1, m2, m3, m4, m5, m6, m7, and m8 represent the line length, input voltage, input current, environmental temperature, environmental humidity, rainfall, environmental comprehensive features, and environmental correlation features in sequence; Obtain the energy consumption feature vectors of all power line segments and all acquisition times in the third power transmission feature data, and merge and store the energy consumption feature vectors with the corresponding acquisition times, transmission losses, output voltages, and output currents, and mark them as the fourth power transmission feature data; Use a multi-layer perceptron to construct a first relationship model. The first relationship model includes: a model input layer, a first hidden layer, a second hidden layer, and a model output layer; the model input layer includes 8 neurons, the model output layer includes 3 neurons, set the number of neurons in the first hidden layer to b1, and set the number of neurons in the second hidden layer to b2; Train the first relationship model using the fourth power transmission feature data, and after completion, obtain the power transmission loss relationship model; Obtain the energy consumption feature vector of the power line at a future time, and input it into the power transmission loss relationship model to obtain the line predicted loss information; Dynamically adjust and optimize the selection of the power lines of the virtual power plant based on the line predicted loss information.

2. The virtual power plant topology method based on energy consumption optimization according to claim 1, wherein Obtain the power line information between all the power equipment of the virtual power plant, and collect the historical power transmission data and historical environmental data of the power lines, including the following sub-steps: For all the power equipment managed by the virtual power plant, obtain all the power lines that can be connected between any two power equipment, and obtain all the connection switch nodes on the power lines. Denote any power line segment without connection switch nodes and power equipment in the middle as a power line segment; Repeatedly obtain all the power line segments of the virtual power plant, and denote it as the power line information; For any transmission line segment, obtain the line length of the transmission line segment, and obtain the transmission loss, input voltage, and input current of the transmission line segment at the first time interval, and obtain the ambient temperature, ambient humidity, and rainfall at the location of the transmission line segment, and record the acquisition time. The first time interval is k1; Mark the line length, transmission loss, input voltage, input current, output voltage, output current, and acquisition time of the obtained transmission line segment as the first transmission data, and mark the ambient temperature, ambient humidity, rainfall, and acquisition time at the location of the obtained transmission line segment as the first environmental data; Repeat to obtain the first transmission data of all transmission line segments, denoted as historical transmission data, and repeat to obtain the first environmental data of all transmission line segments, denoted as historical environmental data.

3. The virtual power plant topology method based on energy consumption optimization according to claim 2, characterized in that Obtain the energy consumption feature vector of the power line at a future moment and input it into the transmission loss relationship model. The steps for obtaining the predicted line loss information also include the following sub-steps: For all power lines that can be connected between any two power devices, denote the input voltage and input current of the initial transmission line segments of all power lines that can be connected as the initial voltage and the initial current respectively; Obtain the initial voltage and initial current of the first time length based on the historical transmission data and sort them in the chronological order from near to far, denoted as the initial voltage sequence and the initial current sequence respectively. The first time length is E1; Sort the initial voltage sequence and the initial current sequence respectively according to the data size, denoted as the voltage magnitude sequence and the current magnitude sequence respectively. Remove the smallest f% data and the largest f% data of the voltage magnitude sequence and the current magnitude sequence respectively, and then calculate the average value and standard deviation of the remaining voltage magnitude sequence and current magnitude sequence respectively. Denote the average value and standard deviation corresponding to the remaining voltage magnitude sequence as WU and BU respectively, and denote the average value and standard deviation corresponding to the remaining current magnitude sequence as WA and BA respectively; Perform anomaly screening processing on the initial voltage sequence and the initial current sequence respectively, including: Denote any initial voltage data in the initial voltage sequence as UR i , and denote any initial current data in the initial current sequence as AR j ; If |UR i - WU| > 3 * BU, then determine that UR i is voltage anomaly data, and use the average value of the two data closest to UR i in the time scales before and after UR i to replace UR i ; If |AR j - WA| > 3 * BA, then determine that AR j is current anomaly data, and use the average value of the two data closest to AR j in the time scales before and after AR j to replace AR j ; After completion, obtain the initial screened voltage sequence and the initial screened current sequence; Denote any voltage data in the initial screening voltage sequence as U i and denote any current data in the screening current sequence as A j ; Calculate U i-1 、U i and U i+1 and replace U i ; Calculate the average value of A j-1 、A j and A j+1 and replace A j ; Repeat to replace all the data in the initial screened voltage sequence and the initial screened current sequence. After completion, obtain the initial smoothed voltage sequence and the initial smoothed current sequence.

4. The virtual power plant topology method based on energy consumption optimization according to claim 3, characterized in that Obtain the energy consumption feature vector of the power line at a future moment and input it into the transmission loss relationship model. The steps for obtaining the predicted line loss information include the following sub-steps: Denote any voltage data in the initial smoothed voltage sequence as PU i ,Denote any current data in the initial smoothed current sequence as PA j ; For the initial smoothed voltage sequence, repeatedly calculate the differences between adjacent voltage data through the voltage difference formula, and average the obtained differences to obtain the average voltage change trend QU. The voltage difference formula is as follows: , where CU i represents the difference between PU i+1 and PU i ; For the initial smoothed current sequence, repeatedly calculate the difference between adjacent current data through the current difference formula, and average the obtained differences to obtain the average current change trend QA. The current difference formula is as follows: , where CA j represents the difference between PA j and PA j ; Set the second time length as E2, and calculate the average initial voltage and average initial current for the future second time length according to the first voltage formula and the first current formula respectively; the first voltage formula is as follows: , where YU represents the average initial voltage for the future second time length, and U1 represents the first voltage data in the initial screening voltage sequence; the first current formula is as follows: , where YA represents the average initial current for the future second time length, and A1 represents the first current data in the initial screening current sequence; Obtain the ambient temperature, ambient humidity, and rainfall of the most recent collection as the ambient temperature, ambient humidity, and rainfall of the next second time length in the future based on the historical environmental data, and perform feature calculation and processing, and then form the energy consumption feature vector of the initial transmission line segment for the next second time length with the average initial voltage and the average initial current, denoted as the predicted feature vector; Input the predicted feature vector into the transmission loss relationship model to obtain the transmission loss, output voltage, and output current of the initial transmission line segment; Use the output voltage and output current of the initial transmission line segment as the input voltage and input current of the next transmission line segment connected to the initial transmission line segment, construct the corresponding predicted feature vector, and input it into the transmission loss relationship model; obtain the corresponding transmission loss, output voltage, and output current; Repeatedly obtain the transmission losses of all transmission line segments included in all power lines that can be connected between two power devices, which is denoted as line prediction loss information.

5. The virtual power plant topology method based on energy consumption optimization according to claim 4, characterized in that Based on the line prediction loss information, the dynamic adjustment and optimization of the power line selection of the virtual power plant includes the following sub-steps: For any two power devices, obtain all power lines that can be connected, and obtain the corresponding line prediction loss information at the second time interval. Calculate the total transmission loss of each power line according to the line prediction loss information, and then compare the sizes. Select the power line with the smallest total transmission loss for power transmission. The second time interval is k2.

6. The virtual power plant topology system based on energy consumption optimization is applicable to the virtual power plant topology method based on energy consumption optimization according to any one of claims 1-5, and is characterized in that It includes a data collection module, a power transmission feature module, a loss prediction module, and a line adjustment module; The data collection module is used to obtain the power line information between the power devices of the virtual power plant, and collect the historical power transmission data and historical environmental data of the power lines; The power transmission feature module includes a feature calculation unit and a model establishment unit. The feature calculation unit performs feature calculation and processing based on the historical power transmission data and historical environmental data. The model establishment unit is used to establish an energy consumption feature vector and a power transmission loss relationship model; The loss prediction module is used to obtain the energy consumption feature vector of the power line at a future moment, and input it into the power transmission loss relationship model to obtain the line prediction loss information; The line adjustment module dynamically adjusts and optimizes the power line selection of the virtual power plant based on the line prediction loss information.

7. An electronic device, characterized in that, It includes a processor and a memory. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the method according to any one of claims 1-5 are run.

8. A storage medium, on which a computer program is stored, characterized in that, When the computer program is executed by the processor, the steps in the method according to any one of claims 1-5 are run.

Images