Method and device for determining operation situation of low-voltage distribution area

Abstract

The invention discloses a method and a device for determining the operation situation of a low-voltage distribution area. The method comprises the following steps: acquiring historical power data output by a plurality of transformer area devices in a low-voltage power distribution transformer area in a historical time period; inputting the historical power data into a power parameter prediction model, and predicting by using the power parameter prediction model to obtain a plurality of power parameters corresponding to different nodes in the low-voltage distribution area in a future time period; carrying out risk perception on the operation situations of the different nodes based on the plurality of electric power parameters to obtain a voltage out-of-limit risk level and a power reverse transmission risk level of the low-voltage power distribution area in a future time period; and determining the operation situation of the low-voltage distribution area based on the voltage out-of-limit risk level and the power reverse transmission risk level. According to the invention, the technical problem of low accuracy of determining the operation situation of the low-voltage distribution area in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the field of low-voltage distribution station area situation prediction, and in particular to a method and device for determining the operating situation of a low-voltage distribution station area. Background Art

[0002] With the large-scale access of rooftop photovoltaics to low-voltage distribution substations, the voltage over-limit problem and power backflow problem at each node in the substation caused by excess photovoltaic output are becoming increasingly prominent. It is necessary to analyze the operating status of photovoltaic output based on short-term rolling forecasts. However, most of the current methods for determining the operating status cannot accurately show users the operating status of low-voltage distribution substations in future time periods, and the prediction accuracy is low.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a method and device for determining the operating status of a low-voltage distribution station area, so as to at least solve the technical problem of low accuracy in determining the operating status of a low-voltage distribution station area in the related art.

[0005] According to one aspect of an embodiment of the present invention, a method for determining the operating status of a low-voltage distribution station is provided, comprising: obtaining historical power data output by multiple substation equipment in the low-voltage distribution station in a historical time period, wherein the multiple substation equipment includes: multiple rooftop photovoltaic equipment, multiple single-phase power equipment and multiple three-phase power equipment; inputting the historical power data into a power parameter prediction model, and using the power parameter prediction model to predict multiple power parameters corresponding to different nodes in the low-voltage distribution station in a future time period; performing risk perception on the operating status of different nodes based on multiple power parameters, and obtaining a voltage over-limit risk level and a power reverse transmission risk level of the low-voltage distribution station in the future time period, wherein the voltage over-limit risk level is used to characterize the degree of deviation between multiple power parameters and standard parameters, and the power reverse transmission risk level is used to characterize the degree of danger when the power flow direction between different nodes is opposite to the expected direction; based on the voltage over-limit risk level and the power reverse transmission risk level, the operating status of the low-voltage distribution station is determined.

[0006] Furthermore, the power parameter prediction model includes: a data prediction module, an error determination module and a flow calculation module; historical power data is input into the power parameter prediction model, and the power parameter prediction model is used to predict multiple power parameters corresponding to different nodes in the low-voltage distribution station area in the future time period, including: historical power data is input into the data prediction module, and the data prediction module is used to predict the initial power data corresponding to different nodes in the future time period; the initial power data and preset confidence are input into the error determination module, and the error determination module is used to determine the prediction error corresponding to the initial power data, wherein the prediction error is used to characterize the initial power data and the error between the actual power data output by multiple substation equipment in the future time period; the initial power data and the prediction error are input into the flow calculation module, and the flow calculation module is used to determine multiple power parameters.

[0007] Furthermore, the historical power data is input into a data prediction module, and the data prediction module is used to predict the initial power data corresponding to different nodes in the future time period, including: dividing the historical power data according to the time axis of the historical power data to obtain training data, test data and prediction data; training the initial prediction module based on the training data to obtain a trained prediction module; testing the trained prediction module based on the test data to obtain a module test result, wherein the module test result includes the accuracy of the trained initial prediction module; in response to the module test result being that the accuracy of the trained prediction module is greater than a preset threshold, determining that the trained prediction module is a data prediction module, and processing the prediction data based on the data prediction module to obtain the initial power data.

[0008] Furthermore, the initial power data and preset confidence levels are input into an error determination module, and the error determination module is used to determine a prediction error corresponding to the initial power data, including: determining the worst probability distribution corresponding to the initial power data based on the initial power data; obtaining the difference between the upper and lower bounds of the probability distribution confidence level based on the worst probability distribution and preset confidence levels to obtain a prediction error, wherein the probability distribution confidence level is determined by the preset confidence levels.

[0009] Furthermore, the initial power data and the predicted error are input into the power flow calculation module, and the power flow calculation module is used to determine multiple power parameters, including: obtaining system parameters of the low-voltage distribution substation, and a power flow calculation model matching the low-voltage distribution substation, wherein the system parameters include: the grid structure and line parameters of the low-voltage distribution substation, and the active output data and reactive output data of the passive power generation equipment in the low-voltage distribution substation; adjusting the initial power data based on the predicted error to obtain the target power data; inputting the target power data and the system parameters into the power flow calculation model, and using the power flow calculation model to determine multiple power parameters.

[0010] Furthermore, the multiple power parameters include at least: predicted voltage values ​​of different nodes; risk perception of the operating status of different nodes based on the multiple power parameters to obtain the voltage over-limit risk level of the low-voltage distribution station area in the future time period, including: obtaining the reference voltage value of different nodes in the low-voltage distribution station area; obtaining the voltage per unit value of different nodes based on the ratio of the predicted voltage value to the reference voltage value; obtaining a first difference based on the absolute value of the difference between the voltage per unit value and a preset value; matching the first difference with multiple difference ranges to determine the voltage over-limit risk level, wherein different difference ranges correspond to different voltage over-limit risk levels.

[0011] Furthermore, the multiple power parameters also include: reverse power of different nodes; risk perception of the operating status of different nodes based on multiple power parameters, and obtaining the power backflow risk level of the low-voltage distribution station area in the future time period, including: obtaining the maximum forward power that different nodes in the low-voltage distribution station area can withstand, wherein the flow direction of the maximum forward power is the expected direction, and the flow direction of the reverse power is opposite to the expected direction; obtaining the power backflow coefficient based on the quotient of the reverse power and the maximum forward power; matching the power backflow coefficient with multiple coefficient ranges to determine the power backflow risk level, wherein different coefficient ranges correspond to different power backflow risk levels.

[0012] Furthermore, the historical power data includes: active output data and reactive output data of the rooftop photovoltaic equipment, three-phase load data consumed by a single electrical device, and three-phase load data consumed by a three-phase electrical device.

[0013] According to one aspect of an embodiment of the present invention, there is also provided a device for determining the operating status of a low-voltage distribution station, comprising: a data acquisition module for acquiring historical power data output by multiple substation devices in the low-voltage distribution station in a historical time period, wherein the multiple substation devices include: multiple rooftop photovoltaic devices, multiple single-phase power devices and multiple three-phase power devices; a parameter prediction module for inputting the historical power data into a power parameter prediction model, and using the power parameter prediction model to predict multiple power parameters corresponding to different nodes in the low-voltage distribution station in a future time period; a risk perception module for performing risk perception on the operating status of multiple nodes based on multiple power parameters, and obtaining a voltage over-limit risk level and a power reverse transmission risk level of the low-voltage distribution station in the future time period, wherein the voltage over-limit risk level is used to characterize the degree of deviation between multiple power parameters and standard parameters, and the power reverse transmission risk level is used to characterize the degree of danger when the power flow direction between different nodes is opposite to the expected direction; a situation determination module for determining the operating status of the low-voltage distribution station based on the voltage over-limit risk level and the power reverse transmission risk level.

[0014] Furthermore, the power parameter prediction model includes: a data prediction module, an error determination module and a flow calculation module; the parameter prediction module is also used to: input historical power data into the data prediction module, and use the data prediction module to predict the initial power data corresponding to different nodes in the future time period; input the initial power data and preset confidence into the error determination module, and use the error determination module to determine the prediction error corresponding to the initial power data, wherein the prediction error is used to characterize the initial power data and the error between the actual power data output by multiple substation equipment in the future time period; input the initial power data and the prediction error into the flow calculation module, and use the flow calculation module to determine multiple power parameters.

[0015] Furthermore, the parameter prediction module is also used to: divide the historical power data according to the time axis of the historical power data to obtain training data, test data and prediction data; train the initial prediction module based on the training data to obtain a trained prediction module; test the trained prediction module based on the test data to obtain a module test result, wherein the module test result includes the accuracy of the trained initial prediction module; in response to the module test result being that the accuracy of the trained prediction module is greater than a preset threshold, determine that the trained prediction module is a data prediction module, and process the prediction data based on the data prediction module to obtain initial power data.

[0016] Furthermore, the parameter prediction module is also used to: determine the worst probability distribution corresponding to the initial power data based on the initial power data; obtain the difference between the upper and lower bounds of the probability distribution confidence interval based on the worst probability distribution and the preset confidence level to obtain the prediction error, wherein the probability distribution confidence interval is determined by the preset confidence level.

[0017] Furthermore, the parameter prediction module is also used to: obtain system parameters of the low-voltage distribution station area, including the grid structure, line parameters and active and reactive output data of the passive power generation equipment of the distribution station area, and a power flow calculation model matching the low-voltage distribution station area, wherein the system parameters include: the grid structure, line parameters of the low-voltage distribution station area, and active output data and reactive output data of the passive power generation equipment in the low-voltage distribution station area; adjust the initial power data based on the prediction error to obtain the target power data; input the target power data and the system parameters into the power flow calculation model, and use the power flow calculation model to determine multiple power parameters.

[0018] Furthermore, the multiple power parameters include at least: predicted voltage values ​​of different nodes; the risk perception module is also used to: obtain reference voltage values ​​of different nodes in the low-voltage distribution station area; obtain voltage per unit values ​​of different nodes based on the ratio of the predicted voltage value to the reference voltage value; obtain a first difference based on the absolute value of the difference between the voltage per unit value and a preset value; match the first difference with multiple difference ranges to determine the voltage over-limit risk level, wherein different difference ranges correspond to different voltage over-limit risk levels.

[0019] Furthermore, the multiple power parameters also include: reverse power of different nodes; the risk perception module is also used to: obtain the maximum forward power that different nodes in the low-voltage distribution station area can withstand, wherein the flow direction of the maximum forward power is the expected direction, and the flow direction of the reverse power is opposite to the expected direction; based on the quotient of the reverse power and the maximum forward power, obtain the power reverse power coefficient; match the power reverse power coefficient with multiple coefficient ranges to determine the power reverse power risk level, wherein different coefficient ranges correspond to different power reverse power risk levels.

[0020] Furthermore, the historical power data includes: active output data and reactive output data of the rooftop photovoltaic equipment, three-phase load data consumed by a single electrical device, and three-phase load data consumed by a three-phase electrical device.

[0021] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present invention is executed when the program is running.

[0022] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0023] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0024] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0025] According to another aspect of the embodiments of the present invention, a computer program is further provided. When the computer program is executed by a processor, the methods in the embodiments of the present invention are implemented.

[0026] In an embodiment of the present invention, a method is adopted in which historical power data output by multiple substation equipment in a low-voltage distribution station area in a historical time period is obtained; the historical power data is input into a power parameter prediction model, and the power parameter prediction model is used to predict multiple power parameters corresponding to different nodes in the low-voltage distribution station area in a future time period; risk perception is performed on the operating status of different nodes based on multiple power parameters to obtain the voltage over-limit risk level and power backfeed risk level of the low-voltage distribution station area in the future time period; based on the voltage over-limit risk level and the power backfeed risk level, the operating status of the low-voltage distribution station area is determined. By performing risk perception on the operating status of different power nodes on the basis of the predicted multiple power parameters of different nodes in the low-voltage distribution station area, the accuracy of the perceived risk level, such as the voltage over-limit risk level and the power backfeed risk level, can be ensured, thereby ensuring the accuracy of the operating status determined according to the voltage over-limit risk level and the power backfeed risk level, thereby solving the technical problem of low accuracy in determining the operating status of the low-voltage distribution station area in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0028] Figure 1 is a flow chart showing a method for determining the operation status of a low-voltage distribution station area according to an embodiment of the present invention;

[0029] Figure 2 is a schematic diagram showing a process of determining the operation status of a low-voltage distribution station area according to an embodiment of the present invention;

[0030] Figure 3 is a schematic diagram showing a low-voltage distribution substation system structure according to an embodiment of the present invention;

[0031] Figure 4 is a schematic diagram showing a photovoltaic output data prediction result according to an embodiment of the present invention;

[0032] Figure 5 is a schematic diagram showing a three-phase load data prediction result according to an embodiment of the present invention;

[0033] Figure 6 The present invention is a structural block diagram of a device for determining the operating status of a low-voltage distribution station area according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0036] According to an embodiment of the present invention, an embodiment of a method for determining the operating status of a low-voltage distribution station is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0037] Figure 1 FIG. 1 is a flow chart showing a method for determining the operation status of a low-voltage distribution station area according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0038] Step S102, obtaining historical power data output by multiple substation equipment in a low-voltage distribution substation in a historical time period.

[0039] Among them, multiple substation equipment includes: multiple rooftop photovoltaic equipment, multiple single-phase power equipment and multiple three-phase power equipment.

[0040] Among them, historical power data include: active and reactive output data of rooftop photovoltaic equipment in low-voltage distribution substations, and three-phase load data consumed by single / three-phase electrical equipment in low-voltage distribution substations.

[0041] The above-mentioned low voltage distribution substation area may refer to an active low voltage distribution substation area. The acquisition parameters of the above-mentioned historical power data may be a measurement frequency of 15 minutes / time, 96 sampling points per day, and a total of two months of sampling, but is not limited thereto.

[0042] In an optional scheme of the present embodiment, in order to accurately predict the operating status of the low-voltage distribution station area, the system for determining the operating status of the low-voltage distribution station area (referred to as the status determination system) can first predict the power parameters of different power nodes in the low-voltage distribution station area in the future time period, and then determine the operating status of the low-voltage distribution station area in the future time period based on the changes in the power parameters. Based on this, in order to accurately determine the power parameters of different power nodes in the low-voltage distribution station area in the future time period, the status determination system can first obtain the historical power data output by each rooftop photovoltaic and each single / three-phase user in the low-voltage distribution station area in the historical time period, such as obtaining the active and reactive output data output by the rooftop photovoltaic equipment in the low-voltage distribution station area, and the three-phase load data consumed by the single / three-phase electrical equipment in the low-voltage distribution station area, so as to predict the power parameters of different nodes in the low-voltage distribution station area in the future time period based on the acquired historical power data. In order to ensure the accuracy of power parameters in future time periods predicted based on the acquired historical power data, the acquired photovoltaic output data may include data in three characteristic dimensions, such as photovoltaic output power, solar irradiance, humidity and other data; the acquired three-phase load data may include data in five characteristic dimensions, such as load data, rainfall, temperature, holidays, humidity and other data.

[0043] Step S104, inputting historical power data into a power parameter prediction model, and using the power parameter prediction model to predict a plurality of power parameters corresponding to different nodes and points in the low-voltage distribution station area in the future time period.

[0044] The above-mentioned different nodes may refer to source nodes in the low-voltage distribution station area.

[0045] In an optional scheme of the present embodiment, after obtaining the historical power data situation determination system, the system can use a pre-trained prediction model, i.e., the above-mentioned power parameter prediction model, to process the acquired historical power data to predict the power parameters corresponding to different nodes in the low-voltage distribution station area in the future time period, such as the voltage, power and other parameters of different nodes.

[0046] Step S106, performing risk perception on the operating status of multiple nodes based on multiple power parameters, and obtaining the voltage over-limit risk level and power reverse transmission risk level of the low-voltage distribution station area in the future time period.

[0047] Among them, the voltage over-limit risk level is used to characterize the degree of deviation between multiple power parameters and standard parameters, and the power reverse flow risk level is used to characterize the degree of danger when the power flow direction between different nodes is opposite to the expected direction.

[0048] In an optional scheme of the present embodiment, taking into account that the operating status of a low-voltage distribution substation can usually reflect the operating status of the low-voltage distribution substation in a future time period, the corresponding status determination system can predict multiple power parameters corresponding to different nodes, and then perform risk perception on the predicted multiple power parameters to determine the voltage over-limit risk level and power backflow risk level of the low-voltage distribution substation in the future time period.

[0049] Step S108, determining the operation status of the low voltage distribution station area based on the voltage over-limit risk level and the power backflow risk level.

[0050] In an optional scheme of the present embodiment, after determining the power over-limit risk level and power backflow risk level of the low-voltage distribution substation in the future time period, the operating status of the low-voltage distribution substation can be determined based on these two risk levels, thereby predicting the risk situation of the low-voltage distribution substation in the future time period to ensure the power quality and power supply reliability of the low-voltage distribution substation.

[0051] In an embodiment of the present invention, a method is adopted in which historical power data output by multiple substation equipment in a low-voltage distribution station area in a historical time period is obtained; the historical power data is input into a power parameter prediction model, and the power parameter prediction model is used to predict multiple power parameters corresponding to different nodes in the low-voltage distribution station area in a future time period; risk perception is performed on the operating status of different nodes based on multiple power parameters to obtain the voltage over-limit risk level and power backfeed risk level of the low-voltage distribution station area in the future time period; based on the voltage over-limit risk level and the power backfeed risk level, the operating status of the low-voltage distribution station area is determined. By performing risk perception on the operating status of different power nodes on the basis of the predicted multiple power parameters of different nodes in the low-voltage distribution station area, the accuracy of the perceived risk level, such as the voltage over-limit risk level and the power backfeed risk level, can be ensured, thereby ensuring the accuracy of the operating status determined according to the voltage over-limit risk level and the power backfeed risk level, thereby solving the technical problem of low accuracy in determining the operating status of the low-voltage distribution station area in the related art.

[0052] Furthermore, the power parameter prediction model includes: a data prediction module, an error determination module and a flow calculation module; historical power data is input into the data prediction module, and the data prediction module is used to predict the initial power data corresponding to different nodes in the future time period; the initial power data and the preset confidence are input into the error determination module, and the error determination module is used to determine the prediction error corresponding to the initial power data, wherein the prediction error is used to characterize the initial power data and the error between the actual power data output by multiple substation equipment in the future time period; the initial power data and the prediction error are input into the flow calculation module, and the flow calculation module is used to determine multiple power parameters.

[0053] In an optional scheme of the present embodiment, taking into account that in the actual process of predicting power parameters, there may be many factors that affect the accuracy of the prediction results, resulting in certain prediction errors in the power parameters predicted by the power parameter prediction model. Therefore, in order to ensure the accuracy of the predicted power parameters, at least a corresponding data prediction module, an error determination module and a flow calculation module may be configured in the power parameter prediction model, so as to use the data prediction module to perform data prediction, and then use the error determination module to determine the possible errors in the predicted data, and finally use the flow calculation module to calculate multiple power parameters after error correction. Based on this, when predicting multiple power parameters corresponding to different power nodes, the situation determination system can first input the acquired historical power data into the data prediction module, so as to use the data prediction module to preliminarily predict the data that may be output by different power nodes in the low-voltage distribution station area in the future time period based on the historical power data, that is, the above-mentioned initial power data, and then use the error determination module to determine the prediction error corresponding to the initial power data based on the determined initial power data and the preset confidence, that is, the error between the initial power data and the actual power data output by each rooftop photovoltaic and each single / three-phase user in the future time period, and finally input the initial power data and the prediction error into the value flow calculation module, so that the flow calculation module can be used to determine multiple power parameters of different power nodes with higher accuracy.

[0054] Furthermore, the historical power data is input into a data prediction module, and the data prediction module is used to predict the initial power data corresponding to different nodes in the future time period, including: dividing the historical power data according to the time axis of the historical power data to obtain training data, test data and prediction data; training the initial prediction module based on the training data to obtain a trained prediction module; testing the trained prediction module based on the test data to obtain a module test result, wherein the module test result includes the accuracy of the trained initial prediction module; in response to the module test result being that the accuracy of the trained prediction module is greater than a preset threshold, determining that the trained prediction module is a data prediction module, and processing the prediction data based on the data prediction module to obtain the initial power data.

[0055] In an optional scheme of the present embodiment, considering that in actual scenarios, the photovoltaic output data output by the photovoltaic equipment in the low-voltage distribution station area is usually affected by factors such as radiation, and the corresponding three-phase load data output by the three-phase power supply equipment may be affected by factors such as rainfall, temperature, holidays, humidity, etc. Therefore, in the present invention, the initial power data of the low-voltage distribution station area in the future time period can be short-term rolling forecast to accurately and quickly determine the above initial power data. Considering that in the process of short-term rolling forecast, the situation determination system usually does not replace the parameters used in the power parameter prediction model on a large scale, therefore, the prediction module used in the current prediction process can be the prediction module used in the last prediction process, and considering that in the short-term prediction process, the accuracy of the data predicted by the prediction module is high, the accuracy of the corresponding prediction module is also high, therefore, if the prediction module used in the last prediction process needs to be adjusted to further improve the accuracy of the prediction module in the current prediction process, it is actually not necessary to change too many parameters contained in the prediction module, and the overall efficiency of adjusting the prediction module is high. Therefore, in order to improve the accuracy of the determined initial power data, the situation determination system can adjust the parameters contained in the prediction module currently used before each prediction. Based on this, after obtaining the historical power data, the situation determination system can first divide the historical power data into three parts: training data, test data and prediction data according to the time axis of the historical power data, and then use the training data to train the prediction module used in the last prediction process, that is, the above-mentioned initial prediction module, to obtain a trained prediction module, and then use the test data to test the trained prediction module to obtain the corresponding module test results. The situation determination system can determine the training module through the module test results. The accuracy of the trained initial prediction module. If the module test result shows that the accuracy of the trained prediction module is high, for example, greater than the preset threshold, the currently trained prediction module can be determined as the data prediction module used in the current prediction process, and the prediction data can be processed according to the data prediction module to obtain the initial power data; if the module test result shows that the accuracy of the trained prediction module is low, for example, less than or equal to the preset threshold, the situation determination system can use the divided training data again to train the above-mentioned initial prediction module multiple times until the accuracy of the trained prediction module is greater than the preset threshold, or the accuracy of the trained prediction module is ensured to be greater than the accuracy of the initial prediction module.

[0056] In an optional scheme of the present embodiment, a GRU (Gated Recurrent Unit) network with a dual attention mechanism can be used as the data prediction module used above to perform short-term rolling predictions on the initial power data of the low-voltage distribution station area in the future time period. Correspondingly, when predicting the initial power data, the situation determination system can first input the acquired historical power data, i.e., the photovoltaic output data and three-phase load data mentioned above, and divide the historical power data into a training set, a test set, and a prediction set. For example, the first 80% of the historical power data can be divided into a training set, the 80% to 96% of the historical power data can be divided into a test set, and the last 4% of the historical power data can be divided into a prediction set. Then, the divided training set and test set are used to adjust the GRU network used in the previous prediction process, and finally, the prediction set is input into the adjusted GRU network to use the adjusted GRU network to predict the initial power data with higher accuracy.

[0057] Among them, the GRU network of the dual attention mechanism may include input vectors, feature attention layers, 3-layer GRU networks, temporal attention layers, and fully connected layer outputs. In the GRU network, the multidimensional feature vector corresponding to the training set can be combined with the hidden state of the GRU network at the previous moment, and then calculated through the feature attention layer to obtain the influence weight of each feature vector on the current prediction situation, and obtain the corrected input features. After learning the features through the two-layer GRU network, the influence weight of the output information at each historical moment is calculated through the temporal attention layer, thereby obtaining the improved current hidden state output, and finally inputting it into the fully connected layer to obtain the final prediction result;

[0058] Correspondingly, the document relationship of the GRU network structure parameters can be shown as follows:

[0059] z t =σ(W z ×[X t ,h t-1 ]);

[0060] r t =σ(W r ×[X t ,h t-1 ]);

[0061]

[0062] Where: z t is the update gate, r t To reset the gate, X t is the current input, is the sum of the input and past hidden states, h tis the hidden layer output, W z , W r , are all trainable parameter matrices.

[0063] In the GRU network, the hidden state h of the GRU network at the previous moment of the feature time series can be t-1 And the current input features As the input of the feature attention mechanism, the weight obtained at the current moment is multiplied by the corresponding feature by calculating the weight, etc., to output the corresponding X t , the corresponding relationship is as follows:

[0064]

[0065] in, These are the neuron weights that the multilayer perceptron needs to learn. is the bias parameter.

[0066] In the GRU network, the temporal attention mechanism can be used to analyze the correlation between the historical state and the current state and assign influence weights by taking the last hidden state containing the information of each historical moment as the input of the temporal attention mechanism. The calculation formula is as follows:

[0067]

[0068] Among them, the score function uses dot product, α t，i is the attention weight of the hidden layer state of the historical input to the current input, c t is the middle vector, Represents the hidden layer state value at the current moment of the final output.

[0069] In the GRU network, the mean square error δ can be used MSE As the loss function, the calculation formula is as follows:

[0070]

[0071] Among them, y i is the actual value, is the predicted value, N is the total number of samples;

[0072] In the GRU network, the Adam algorithm can be used as the optimization algorithm to update the weights. The algorithm formula is as follows:

[0073]

[0074] in, is the weight of the Adam algorithm to be updated, δ is the learning rate, μ is the smoothing parameter, are the mean and variance of the gradient after error correction, and the calculation formulas for these parameters can be shown as follows:

[0075]

[0076] Among them, θ t , τ t are the mean and variance of the gradient, σ1 and σ2 are the attenuation factors.

[0077] In the GRU network, the mean absolute error δ can be used MAE and the coefficient of determination δ R2 As an evaluation indicator of the deep learning model, the feasibility of the model is judged. The corresponding calculation formulas are as follows:

[0078]

[0079] in, is the average of the true values.

[0080] Furthermore, the initial power data and preset confidence levels are input into an error determination module, and the error determination module is used to determine a prediction error corresponding to the initial power data, including: determining the worst probability distribution corresponding to the initial power data based on the initial power data; obtaining the difference between the upper and lower bounds of the probability distribution confidence level based on the worst probability distribution and preset confidence levels to obtain a prediction error, wherein the probability distribution confidence level is determined by the preset confidence levels.

[0081] In an optional scheme of the present embodiment, when using the error determination module to determine the prediction error corresponding to the initial power data, the situation determination system can first determine the worst probability distribution corresponding to the initial power data based on the above-mentioned initial power data, and then use the worst probability distribution and preset confidence to obtain the difference between the upper and lower bounds of the probability distribution confidence interval to determine the prediction error corresponding to the initial power data.

[0082] In an optional scheme of the present embodiment, the above-mentioned prediction error can be determined by a pre-constructed uncertainty model, wherein the uncertainty of the data may refer to the uncertainty of the difference between the predicted value of the photovoltaic output data and the predicted value of the three-phase load data and the actual value. In order to ensure the accuracy of the determined prediction error, the situation determination system may directly obtain the worst probability distribution of the data through a deep learning method based on a distributed robust optimization method of deep learning, and then determine the corresponding prediction error based on the acquired worst probability distribution.

[0083] Based on this, in the constructed uncertainty model, the predicted initial power data can be used as input, and the worst probability distribution of the typical scenario can be used as output. Correspondingly, when constructing the uncertainty model, the corresponding training samples can be constructed first, and the construction process can be as follows:

[0084] Considering a certain degree of photovoltaic output prediction error and load forecast error Subordinate to the normal distribution N(1,(σ PV ) 2 ) and N(1,(σ Pd ) 2 ), where σ PV , σ Pd is the standard deviation of photovoltaic output fluctuation and load fluctuation, as follows:

[0085]

[0086] in, is the predicted output of photovoltaics, is the load forecast value, from which we can obtain the continuous forecast error scenario and its initial probability distribution, that is, the normal distribution.

[0087] Then, based on the Wasserstein distance, we construct a probability distribution fuzzy set. The Wasserstein distance between any two probability distributions P1 and P2 is defined as:

[0088] W(P1, P2)=inf{∫d(ξ1, ξ2)∏(dξ1, dξ2)}

[0089] Where d(ξ1,ξ2)=||ξ1-ξ2|| is the Euclidean norm between sample arrays ξ1 and ξ2, Π(dξ1,dξ2) is the joint sample distribution with marginal distributions P1 and P2, and inf represents the infimum function;

[0090] The DRO fuzzy set D0 based on Wasserstein distance is based on the initial probability distribution P of uncertain variables. S The Wasserstein ball with ε as the center and radius contains all possible probability distributions including the true distribution, that is,

[0091] D0={P i ∈M(I)|W(P i , P S )≤ε}.

[0092] Among them, M(I) represents the total probability distribution on the random variable space / , and the radius ε reflects the confidence of the unknown probability distribution.

[0093] With the minimum voltage offset value as the goal, all probability distributions are traversed, and the sum of the predicted value and the prediction error is used as input to calculate the voltage offset value of each node. The probability distribution of the prediction error corresponding to the maximum node voltage offset value is recorded, and the predicted value is recorded. In this way, a large number of training samples consisting of the worst probability distributions of predicted values ​​and prediction errors can be obtained.

[0094] After constructing the training samples, the situation determination system can further construct a mapping relationship between the predicted value and the worst probability distribution of the prediction error through deep learning, as shown below:

[0095] First, the mapping relationship between the predicted value and the probability distribution of the prediction error is generated through the deep learning model:

[0096]

[0097] Among them, I is the model input, 0 is the model output, is the φ-phase active power injected into node i by photovoltaic, is the φ-phase load power of node i, is the worst probability distribution of the φ-phase active power injected into the photovoltaic node i, is the worst probability distribution of the φ-phase load power at node i.

[0098] At the same time, the ReLU function is used as the nonlinear activation function of the deep learning model. The function expression is as follows:

[0099]

[0100] in, is the output of the upper layer neurons, is the input of the next layer of neurons.

[0101] The mapping from input to output can be expressed as follows:

[0102] I=u 0 ;

[0103] r z =W z u z-1 +ρ z z∈Z;

[0104] u z =max(r z , 0)z∈Z;

[0105] O=W |Z|+1 u |Z| +ρ |Z|+1 .

[0106] Among them, u 0is the initial input, r z 、u z are the linear mapping of the hidden layer z and the output of the activation function, Z is the set of hidden layers, |Z| is the number of hidden layers, W and ρ are the weights and biases that need to be learned through training, and considering that the output is a probability distribution, the following constraints must be met:

[0107] 0≤O≤1.

[0108] In addition, the Adam algorithm can be used, and the mean square error δ MSE The deep learning model is trained as the loss function. The specific process is to calculate the product of the input and the corresponding weight in the forward propagation, use the activation function to act on the sum of these products, and return the relevant errors in the back propagation of the neural network to calculate the loss function and its gradient for each parameter, update the weight parameter W and the bias parameter ρ, realize the training of the model, and select the mean absolute error δ MAE and the coefficient of determination δ R2 As an evaluation indicator of deep learning models, it is used to judge the feasibility of the model.

[0109] Considering the significance level of α, calculate the upper and lower bounds of the (1-α) confidence interval under the worst probability distribution. The error value of the uncertainty is the difference between the upper and lower bounds of the confidence interval:

[0110]

[0111] in are the active and reactive error values ​​of photovoltaic and load respectively, are the lower bounds of the confidence intervals for active and reactive power of photovoltaic and load, respectively. are the upper limits of the confidence intervals for active and reactive power of PV and load respectively.

[0112] Furthermore, the initial power data and the predicted error are input into the power flow calculation module, and the power flow calculation module is used to determine multiple power parameters, including: obtaining system parameters of the low-voltage distribution substation, and a power flow calculation model matching the low-voltage distribution substation, wherein the system parameters include: the grid structure and line parameters of the low-voltage distribution substation, and the active output data and reactive output data of the passive power generation equipment in the low-voltage distribution substation; adjusting the initial power data based on the predicted error to obtain the target power data; inputting the target power data and the system parameters into the power flow calculation model, and using the power flow calculation model to determine multiple power parameters.

[0113] In an optional scheme of the present embodiment, in order to ensure the accuracy of the multiple power parameters of different power nodes determined, when using the prediction error to adjust the initial power data, the situation determination system can first obtain the system parameters of the low-voltage distribution station area, such as power grid parameters, power supply parameters, load parameters, topology parameters, etc., and at the same time, according to the system identification of the low-voltage distribution station area, obtain the power flow calculation model matching the low-voltage distribution station area, and then use the prediction error to adjust the predicted initial power data to obtain the corresponding template power data. Finally, the template power data and the obtained system parameters are input into the power flow calculation model, and the power flow calculation model can be used to determine the multiple power parameters corresponding to different nodes.

[0114] In view of the fact that the low-voltage distribution network is radial and there is a widespread phenomenon of three-phase load imbalance and line parameter imbalance, a three-phase unbalanced three-phase four-wire branch power flow model considering photovoltaic and load uncertainty can be constructed for the low-voltage distribution area, as follows:

[0115] For any node j at time t, there is a power balance equation:

[0116]

[0117] For the branch ij at time t, we have:

[0118]

[0119] Among them, φ∈{A,B,C}; are the active power and reactive power of the φ phase at the head end of branch ij, u(j) is the head end node set of the branch with node j as the end node in the low voltage distribution network, v(j) is the end node set of the branch with node j as the head end node in the low voltage distribution network, V i φ,t is the node voltage amplitude, P j φ,t , are the net active power and reactive power injected into node j, respectively. are the active and reactive power values ​​of the unit output respectively, are the load active and reactive power values, respectively. are the branch resistance and reactance considering the self-impedance and mutual impedance of the φ-phase line of the distribution network, respectively.

[0120] Furthermore, the multiple power parameters include at least: predicted voltage values ​​of different nodes; risk perception of the operating status of different nodes based on the multiple power parameters to obtain the voltage over-limit risk level of the low-voltage distribution station area in the future time period, including: obtaining the reference voltage value of different nodes in the low-voltage distribution station area; obtaining the voltage per unit value of different nodes based on the ratio of the predicted voltage value to the reference voltage value; obtaining a first difference based on the absolute value of the difference between the voltage per unit value and a preset value; matching the first difference with multiple difference ranges to determine the voltage over-limit risk level, wherein different difference ranges correspond to different voltage over-limit risk levels.

[0121] In an optional scheme of the present embodiment, the multiple determined power parameters may include at least predicted voltage values ​​corresponding to different power nodes. When determining the power over-limit risk level of the low-voltage distribution station area in the future time period, the situation determination system may first obtain the reference voltage values ​​of different nodes in the low-voltage distribution station area, and then determine the voltage per unit value of different power nodes based on the ratio of the predicted voltage value to the reference voltage value, and obtain the above-mentioned first difference based on the absolute value of the difference between the determined voltage per unit value and the preset value, and finally match the first difference with different over-limit level ranges, that is, the above-mentioned multiple difference ranges, so as to determine the voltage over-limit risk level of the low-voltage distribution station area in the future time period, wherein different difference ranges may correspond to different voltage over-limit risk levels.

[0122] For ease of understanding, the voltage over-limit risk level may refer to the deviation of each node voltage from the reference value. Based on this, the voltage over-limit situation may be represented by the difference between the actual node voltage and the reference value. The formula is as follows:

[0123] ΔU=|U i -U0|.

[0124] Among them U i represents the voltage per unit value of node i, U0 represents the node voltage reference value, which is taken as 1.00, and the normal range of the node voltage is taken as 0.94~1.06.

[0125] Before determining the voltage over-limit situation, the voltage over-limit risk level can be set first, for example, 5 levels can be set. When the node ΔU range is 0.03-0.05, it can be considered to be at level I risk; when it is 0.05-0.06, it can be considered to be at level II risk; when the above two levels of risk occur, it can be considered that there is no threat to the distribution network temporarily, but vigilance is required. When the node ΔU range is 0.06-0.07, it can be considered to be at level III risk; when it is 0.07-0.08, it can be considered to be at level IV risk; when it is greater than 0.08, it can be considered to be at level V risk. When the above three risks occur, it can be considered that the distribution network is in an emergency state, and the higher the risk level, the greater the risk of distribution network operation.

[0126] Furthermore, the multiple power parameters also include: reverse power of different nodes; risk perception of the operating status of different nodes based on multiple power parameters, and obtaining the power backflow risk level of the low-voltage distribution station area in the future time period, including: obtaining the maximum forward power that different nodes in the low-voltage distribution station area can withstand, wherein the flow direction of the maximum forward power is the expected direction, and the flow direction of the reverse power is opposite to the expected direction; obtaining the power backflow coefficient based on the quotient of the reverse power and the maximum forward power; matching the power backflow coefficient with multiple coefficient ranges to determine the power backflow risk level, wherein different coefficient ranges correspond to different power backflow risk levels.

[0127] In an optional scheme of the present embodiment, the multiple power parameters determined may also include the reverse power of different source nodes. When determining the power backflow risk level of the low-voltage distribution station area in the future time period, the situation determination system may first obtain the maximum forward power that different nodes in the low-voltage distribution station area can withstand, wherein the flow direction of the maximum forward power is the expected direction, and the flow direction of the reverse power is opposite to the expected direction. After obtaining the maximum forward power that different nodes can withstand, the situation determination system may obtain the corresponding power backflow coefficient based on the quotient of the reverse power and the maximum forward power, and then match the power backflow coefficient with multiple coefficient ranges to determine the power backflow risk level, wherein different coefficient ranges correspond to different power backflow risk levels.

[0128] For ease of understanding, the above-mentioned power reverse transmission risk level can refer to the situation where power is transmitted from low voltage to high voltage in the distribution network. When constructing the power flow calculation model, it is assumed that the power is positive when it flows from high voltage to low voltage and negative when it flows from low voltage to high voltage. Therefore, it is possible to determine whether the power reverse transmission risk occurs by judging whether the power direction of each node is correct. The power reverse transmission risk level can be divided according to the amount of power reverse transmission, with the maximum power P that the node power can withstand as the maximum power. max0 is the reference value, and the power reverse value can be expressed as:

[0129] P r =αP max0 .

[0130] Where α is the power reverse coefficient.

[0131] Before determining the voltage over-limit situation, the power backflow risk level can be set first. For example, 5 levels can be set. When α is in the range of 0 to 0.02, it can be considered to be at level I risk; when α is in the range of 0.02 to 0.05, it can be considered to be at level II risk; when α is in the range of 0.05 to 0.1, it can be considered to be at level III risk; when α is in the range of 0.1 to 0.2, it can be considered to be at level IV risk; when α is greater than 0.2, it can be considered to be at level V risk.

[0132] In the present invention, the voltage value of each node, the active power value and the direction of each node in the flow calculation results are used as input and substituted into the risk assessment index system to calculate the voltage over-limit risk level and the power backflow risk level, so that the operating status of each node in the low-voltage distribution substation area can be accurately perceived, thereby determining the operating status of the low-voltage distribution substation area in the future time period.

[0133] For ease of understanding, Figure 2 FIG. 1 is a schematic diagram showing a process of determining the operation status of a low-voltage distribution station area according to an embodiment of the present invention. Figure 2 As shown, the entire situation determination process can be divided into three parts, the first part is the situation awareness part, the second part is the situation understanding part, and the third part can be the situation prediction part. In the first part, the situation determination system can obtain the historical power data of the low-voltage distribution substation and the system parameters of the low-voltage distribution substation. In the second part, the situation determination system can successively construct a GRU network for data prediction, an uncertainty model for determining the prediction error, a flow calculation model for data correction, and a risk perception algorithm for risk perception. In the third part, the situation determination system can use the multiple algorithms and models constructed in the second part to perform operations such as data prediction, error determination, data correction and risk perception, and finally determine the operation status of the low-voltage distribution substation in the future time period based on the perceived risk level of the low-voltage distribution substation.

[0134] A simulation analysis is conducted on a low-voltage distribution substation with 54 nodes and rooftop photovoltaics. Figure 3 is a schematic diagram of a low voltage distribution substation system structure according to an embodiment of the present invention. Figure 3As shown, the low-voltage distribution station area has 3 main feeders, 48 ​​single loads, 7 three-phase loads and 3 rooftop photovoltaics. The capacity of the distribution transformer is 48kW, the maximum capacity of phase a is 18kW, the maximum capacity of phase b is 10kW, and the maximum capacity of phase c is 20kW. Single-phase photovoltaic 1 is connected to node 4, located in phase a, with a capacity of 1200W, single-phase photovoltaic 2 is connected to node 32, located in phase b, with a capacity of 1200W, and single-phase photovoltaic 3 is connected to branch feeder B24, located in phase c, with a capacity of 1200W. There are 7 three-phase loads in total, of which two are connected to the main feeder 1, which are connected to branch feeders B15 and B16 respectively, two are connected to the main feeder 2, which are connected to branch feeders B22 and B23 respectively, and three are connected to the main feeder 3, which are connected to branch feeders B33, B34 and B35 respectively. The power data of each photovoltaic and single / three-phase user is sampled at a frequency of 15 minutes per time, and a total of two months of historical data is obtained. Figure 3 Based on the simulation test of the low voltage distribution network shown, it can be obtained Figure 4 and Figure 5 The results shown are Figure 4 is a schematic diagram showing a photovoltaic output data prediction result according to an embodiment of the present invention, wherein: Figure 4 The horizontal axis represents time, and the vertical axis represents photovoltaic power; Figure 5 is a schematic diagram showing a three-phase load data prediction result according to an embodiment of the present invention, wherein: Figure 5 The horizontal axis represents time, and the vertical axis represents the three-phase load power, including phase a load power, phase b load power and phase c load power.

[0135] Table 1

[0136]

[0137] Table 1 is a table comparing the results of the situation determination method shown in the present invention with the results of the conventional determination method. It can be seen from Table 1 that when the uncertainty in the power data prediction process is considered, the perception accuracy of the voltage state can reach 98.1%, and the voltage over-limit risk perception accuracy can reach 98.5%, which is 5% and 4.4% higher than that without considering the uncertainty, respectively. The voltage over-limit risk state perception of the rural low-voltage distribution station area can be better perceived. When considering the certainty, the power state perception accuracy and the power reverse transmission risk perception accuracy both reach 100%, which is 1.3% and 33.3% higher than that without considering the uncertainty, respectively. Therefore, the power reverse transmission risk state perception of the rural low-voltage distribution station area can be better perceived. Obviously, the above embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made on the basis of the above description. It is not necessary and impossible to list all the implementation methods here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the claims of the present invention.

[0138] According to an embodiment of the present invention, a device embodiment of a method for determining the operating status of a low-voltage distribution station area is provided. It should be noted that the device can be used to execute the above-mentioned method for determining the operating status of a low-voltage distribution station area. Figure 6 is a structural block diagram of a device for determining the operation status of a low-voltage distribution station area according to an embodiment of the present invention. Figure 6 As shown, the device may include: a data acquisition module 602 , a parameter prediction module 604 , a risk perception module 606 and a situation determination module 608 .

[0139] Among them, the data acquisition module 602 is used to obtain the historical power data output by each rooftop photovoltaic and each single / three-phase user in the low-voltage distribution station area in the historical time period; the parameter prediction module 604 is used to input the historical power data into the power parameter prediction model, and use the power parameter prediction model to predict multiple power parameters corresponding to different nodes in the low-voltage distribution station area in the future time period; the risk perception module 606 is used to perform risk perception on multiple power parameters corresponding to different nodes, and obtain the voltage over-limit risk level and power reverse transmission risk level of the low-voltage distribution station area in the future time period, wherein the voltage over-limit risk level is used to characterize the degree of deviation between multiple power parameters and standard parameters, and the power reverse transmission risk level is used to characterize the degree of danger when the power flow direction between different nodes is opposite to the expected direction; the situation determination module 608 is used to determine the operating situation of the low-voltage distribution station area based on the voltage over-limit risk level and the power reverse transmission risk level.

[0140] Furthermore, the power parameter prediction model includes: a data prediction module, an error determination module and a flow calculation module; the parameter prediction module is also used to: input historical power data into the data prediction module, and use the data prediction module to predict the initial power data corresponding to different nodes in the future time period; input the initial power data and preset confidence into the error determination module, and use the error determination module to determine the prediction error corresponding to the initial power data, wherein the prediction error is used to characterize the initial power data and the error between the actual power data output by each rooftop photovoltaic and each single / three-phase user in the future time period; input the initial power data and the prediction error into the flow calculation module, and use the flow calculation module to determine multiple power parameters.

[0141] Furthermore, the parameter prediction module is also used to: divide the historical power data according to the time axis of the historical power data to obtain training data, test data and prediction data; train the initial prediction module based on the training data to obtain a trained prediction module; test the trained prediction module based on the test data to obtain a module test result, wherein the module test result includes the accuracy of the trained initial prediction module; in response to the module test result being that the accuracy of the trained prediction module is greater than a preset threshold, determine that the trained prediction module is a data prediction module, and process the prediction data based on the data prediction module to obtain initial power data.

[0142] Furthermore, the parameter prediction module is also used to: determine the worst probability distribution corresponding to the initial power data based on the initial power data; obtain the difference between the upper and lower bounds of the probability distribution confidence interval based on the worst probability distribution and the preset confidence level to obtain the prediction error.

[0143] Furthermore, the parameter prediction module is also used to: obtain system parameters of the low-voltage distribution substation, including the grid structure of the distribution substation, line parameters and active and reactive output data of passive power generation equipment, and a power flow calculation model matching the low-voltage distribution substation; adjust the initial power data based on the prediction error to obtain the target power data; input the target power data and system parameters into the power flow calculation model, and use the power flow calculation model to determine multiple power parameters.

[0144] Furthermore, the multiple power parameters include at least: a predicted voltage value of each node; the risk perception module is also used to: obtain a reference voltage value of different nodes in the low-voltage distribution station area; obtain a voltage per unit value of different nodes based on a ratio of the predicted voltage value to the reference voltage value; obtain a first difference based on the absolute value of the difference between the voltage per unit value and a preset value; match the first difference with multiple difference ranges to determine the voltage over-limit risk level, wherein different difference ranges correspond to different voltage over-limit risk levels.

[0145] Furthermore, the multiple power parameters also include: reverse power of different source nodes; the risk perception module is also used to: obtain the maximum forward power that different nodes in the low-voltage distribution station area can withstand, wherein the flow direction of the maximum forward power is the expected direction, and the flow direction of the reverse power is opposite to the expected direction; based on the quotient of the reverse power and the maximum forward power, obtain the power reverse power coefficient; match the power reverse power coefficient with multiple coefficient ranges to determine the power reverse power risk level, wherein different coefficient ranges correspond to different power reverse power risk levels.

[0146] Furthermore, the historical power data includes: active and reactive output data of rooftop photovoltaic equipment in the low-voltage distribution station area, and three-phase load data consumed by single / three-phase power equipment in the low-voltage distribution station area.

[0147] An embodiment of the present invention further provides an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present invention is executed when the program is running.

[0148] An embodiment of the present invention further provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0149] An embodiment of the present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0150] An embodiment of the present invention further provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0151] An embodiment of the present invention further provides a computer program, which implements the methods in the above-mentioned embodiments of the present invention when executed by a processor.

[0152] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0153] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units can be a logical function division, and there can be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0154] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0155] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0156] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0157] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for determining the operating status of a low-voltage distribution station area, characterized in that: include: Obtain historical power data output by multiple substation equipment in a low-voltage distribution substation in a historical time period, wherein the multiple substation equipment includes: multiple rooftop photovoltaic equipment, multiple single-phase power equipment and multiple three-phase power equipment; Inputting the historical power data into a power parameter prediction model, and using the power parameter prediction model to predict a plurality of power parameters corresponding to different nodes in the low-voltage distribution substation area in a future time period; Based on the multiple power parameters, the operating status of the different nodes is perceived at risk, and the voltage over-limit risk level and the power backflow risk level of the low-voltage distribution station area in the future time period are obtained, wherein the voltage over-limit risk level is used to characterize the degree of deviation between the multiple power parameters and the standard parameters, and the power backflow risk level is used to characterize the degree of danger when the power flow direction between different nodes is opposite to the expected direction; Based on the voltage over-limit risk level and the power backflow risk level, the operating status of the low-voltage distribution station area is determined.

2. The method according to claim 1, characterized in that The power parameter prediction model includes: a data prediction module, an error determination module and a power flow calculation module; the historical power data is input into the power parameter prediction model, and the power parameter prediction model is used to predict multiple power parameters corresponding to different nodes in the low-voltage distribution substation area in the future time period, including: Inputting the historical power data into the data prediction module, and using the data prediction module to predict and obtain the initial power data corresponding to each node in the future time period; Inputting the initial power data and the preset confidence into the error determination module, and using the error determination module to determine the prediction error corresponding to the initial power data, wherein the prediction error is used to characterize the error between the initial power data and the actual power data output by the multiple substation devices in the future time period; The initial power data and the prediction error are input into the power flow calculation module, and the power flow calculation module is used to determine the plurality of power parameters.

3. The method according to claim 2, characterized in that Inputting the historical power data into the data prediction module, and using the data prediction module to predict and obtain the initial power data corresponding to each node in the future time period, including: Dividing the historical power data according to the time axis of the historical power data to obtain training data, test data and prediction data; Training the initial prediction module based on the training data to obtain a trained prediction module; Testing the trained prediction module based on the test data to obtain a module test result, wherein the module test result includes the accuracy of the initial prediction module after training; In response to the module test result that the accuracy of the trained prediction module is greater than a preset threshold, the trained prediction module is determined to be the data prediction module, and the prediction data is processed based on the data prediction module to obtain the initial power data.

4. The method according to claim 2, characterized in that: Inputting the initial power data and the preset confidence into the error determination module, and using the error determination module to determine the prediction error corresponding to the initial power data, comprises: Based on the initial power data, determining the worst probability distribution corresponding to the initial power data; The prediction error is obtained by calculating the difference between the upper and lower bounds of the probability distribution confidence interval based on the worst probability distribution and the preset confidence level, wherein the probability distribution confidence interval is determined by the preset confidence level.

5. The method according to claim 2, characterized in that: Inputting the initial power data and the prediction error into the power flow calculation module, and using the power flow calculation module to determine the multiple power parameters, including: Obtaining system parameters of the low-voltage distribution substation and a power flow calculation model matching the low-voltage distribution substation, wherein the system parameters include: the grid structure and line parameters of the low-voltage distribution substation, and active power output data and reactive power output data of the passive power generation equipment in the low-voltage distribution substation; Adjusting the initial power data based on the prediction error to obtain target power data; The target power data and the system parameters are input into the power flow calculation model, and the power flow calculation model is used to determine the plurality of power parameters.

6. The method according to any one of claims 1 to 5, characterized in that The multiple power parameters at least include: predicted voltage values ​​of different nodes; based on the multiple power parameters, risk perception of the operation status of the different nodes is performed to obtain the voltage over-limit risk level of the low-voltage distribution station area in the future time period, including: Obtaining reference voltage values ​​of different nodes in the low voltage distribution station area; Based on the ratio of the predicted voltage value to the reference voltage value, obtaining voltage per unit values ​​of different nodes; Obtaining a first difference value based on an absolute value of a difference between the voltage per unit value and a preset value; The first difference is matched with a plurality of difference ranges to determine the voltage over-limit risk level, wherein different difference ranges correspond to different voltage over-limit risk levels.

7. The method according to claim 6, characterized in that The multiple power parameters also include: reverse power of different nodes; risk perception of the operation status of the different nodes based on the multiple power parameters, Obtaining the power reverse transmission risk level of the low voltage distribution station area in the future time period includes: Obtaining the maximum forward power that different nodes in the low-voltage distribution station area can withstand, wherein the flow direction of the maximum forward power is the expected direction, and the flow direction of the reverse power is opposite to the expected direction; Obtaining a power reverse transmission coefficient based on the quotient of the reverse power and the maximum forward power; The power backflow coefficient is matched with a plurality of coefficient ranges to determine the power backflow risk level, wherein different coefficient ranges correspond to different power backflow risk levels.

8. The method according to any one of claims 1 to 5, characterized in that The historical power data includes: active output data and reactive output data output by the rooftop photovoltaic equipment, three-phase load data consumed by the single electrical equipment, and three-phase load data consumed by the three-phase electrical equipment.

9. A device for determining the operation status of a low-voltage distribution station area, characterized in that: include: A data acquisition module is used to acquire historical power data output by multiple substation equipment in a low-voltage distribution substation in a historical time period, wherein the multiple substation equipment includes: multiple rooftop photovoltaic equipment, multiple single-phase power equipment and multiple three-phase power equipment; A parameter prediction module, used to input the historical power data into a power parameter prediction model, and use the power parameter prediction model to predict a plurality of power parameters corresponding to different nodes in the low-voltage distribution substation area in a future time period; A risk perception module, used to perform risk perception on the operating status of the multiple nodes based on the multiple power parameters, and obtain the voltage over-limit risk level and power backflow risk level of the low-voltage distribution station area in the future time period, wherein the voltage over-limit risk level is used to characterize the degree of deviation between the multiple power parameters and the standard parameters, and the power backflow risk level is used to characterize the degree of danger when the power flow direction between different nodes is opposite to the expected direction; A situation determination module is used to determine the operating situation of the low-voltage distribution station area based on the voltage over-limit risk level and the power backflow risk level.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 8 when running.

Images