Method and device for determining early warning strategy of power distribution network and storage medium

Abstract

The invention discloses a method and device for determining an early warning strategy of a power distribution network and a storage medium, and relates to the technical field of power dispatching. The method comprises the steps that monitoring data of the power distribution network are acquired, and the monitoring data at least comprise one of three-phase current, three-phase voltage, frequency, power factor, environment data and unbalance degree of the power distribution network; the monitoring data are input into a target prediction model for analysis, a prediction result of the power distribution network is obtained, the prediction result is used for indicating the probability that the power distribution network is in a three-phase imbalance state, and the target prediction model is obtained by training an initial prediction model through historical monitoring data; and in response to the prediction result that the power distribution network is in the three-phase imbalance state, an early warning strategy of the power distribution network is determined, and the early warning strategy is used for indicating a rule for carrying out early warning on the power distribution network. The technical problem that the early warning monitoring precision of the power distribution network is low is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of early warning monitoring of distribution networks, and in particular to a method, device and storage medium for determining an early warning strategy of a distribution network. Background Art

[0002] At present, with the development of smart grid, the monitoring and maintenance of low-voltage distribution system has become particularly important. Three-phase imbalance is one of the common problems in low-voltage distribution network, which can cause equipment overheating, reduced efficiency, and even cause serious electrical failures. Therefore, effective monitoring and early warning of three-phase imbalance are particularly necessary.

[0003] In the related art, the three-phase imbalance monitoring and early warning methods for low-voltage power distribution systems mainly rely on traditional statistical analysis methods. However, the traditional statistical analysis method has limited monitoring accuracy for three-phase imbalance and cannot accurately judge the severity and scope of the imbalance. At the same time, the traditional method only relies on simple current differences to judge three-phase imbalance, ignoring the influence of other key factors, such as voltage fluctuations, frequency changes, environmental conditions, etc., which may lead to low accuracy of early warning and easy to generate false alarms or missed reports. Therefore, there is a technical problem of low accuracy of early warning monitoring of distribution networks.

[0004] With regard to the above-mentioned technical problem of low accuracy of early warning monitoring of distribution networks, no effective solution has been proposed so far. Summary of the invention

[0005] The embodiments of the present invention provide a method, device and storage medium for determining an early warning strategy of a distribution network, so as to at least solve the technical problem of low accuracy of early warning monitoring of the distribution network.

[0006] According to one aspect of an embodiment of the present invention, a method for determining an early warning strategy for a distribution network is provided. The method may include: obtaining monitoring data of the distribution network, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance; inputting the monitoring data into a target prediction model for analysis to obtain a prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training an initial prediction model using historical monitoring data; in response to the prediction result that the distribution network is in a three-phase unbalanced state, determining an early warning strategy for the distribution network, wherein the early warning strategy is used to indicate a rule for early warning of the distribution network.

[0007] Optionally, the monitoring data is input into a target prediction model for analysis to obtain prediction results of the distribution network, including: inputting the monitoring data into a target prediction model for analysis to determine initial prediction results of multiple distribution networks; and determining the prediction results of the distribution network based on each initial prediction result.

[0008] Optionally, based on each initial prediction result, determining the prediction result of the distribution network includes: determining the highest prediction result among multiple initial prediction results; comparing the highest prediction result with a prediction result threshold to obtain a comparison result; and determining the prediction result of the distribution network based on the comparison result.

[0009] Optionally, a method for determining an early warning strategy for a distribution network includes: obtaining historical monitoring data of the distribution network; and inputting the historical monitoring data into an initial model for training to obtain a target prediction model.

[0010] Optionally, before using historical monitoring data to train an initial prediction model to obtain a target prediction model, the method for determining a warning strategy for a distribution network further includes: preprocessing the historical monitoring data.

[0011] Optionally, the method for determining the early warning strategy of the distribution network also includes: obtaining a loss function, wherein the loss function is used to train an initial prediction model, the loss function contains physical constraint terms, and the physical constraint terms are used to optimize the corresponding historical monitoring data when the distribution network is in a three-phase unbalanced state; using the loss function to train the initial prediction model to obtain a target prediction model.

[0012] According to another aspect of an embodiment of the present invention, a device for determining an early warning strategy of a distribution network is also provided. The device may include: an acquisition unit, used to acquire monitoring data of the distribution network, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance; an analysis unit, used to input the monitoring data into a target prediction model for analysis to obtain a prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training an initial prediction model using historical monitoring data; a determination unit, used to determine an early warning strategy of the distribution network in response to the prediction result that the distribution network is in a three-phase unbalanced state, wherein the early warning strategy is used to indicate a rule for early warning of the distribution network.

[0013] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein when the program is executed by a processor, the device where the storage medium is located is controlled to execute the method for determining the early warning strategy of the distribution network in an embodiment of the present invention.

[0014] According to another aspect of an embodiment of the present invention, a processor is provided, wherein the processor is used to run a program, wherein when the program is run, the method for determining the early warning strategy of the power distribution network in the embodiment of the present invention is executed.

[0015] According to another aspect of the embodiments of the present invention, a computer program product is provided, which includes computer instructions, and when the computer instructions are executed by a processor, the method for determining the early warning strategy of the power distribution network in the embodiments of the present invention is implemented.

[0016] In an embodiment of the present invention, monitoring data of the distribution network is obtained, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance; the monitoring data is input into the target prediction model for analysis to obtain the prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training the initial prediction model using historical monitoring data; in response to the prediction result that the distribution network is in a three-phase unbalanced state, the early warning strategy of the distribution network is determined, wherein the early warning strategy is used to indicate the rules for early warning of the distribution network. That is to say, the embodiment of the present invention analyzes the monitoring data of the distribution network through the target prediction model to obtain the prediction result, thereby determining the early warning strategy of the distribution network according to the prediction result, avoiding the problem of limited monitoring accuracy of the three-phase imbalance of the distribution network using traditional statistical methods, thereby solving the technical problem of low early warning monitoring accuracy of the distribution network, and achieving the technical effect of improving the early warning monitoring accuracy of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. 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:

[0018] Figure 1 is a flow chart of a method for determining an early warning strategy of a distribution network according to an embodiment of the present invention;

[0019] Figure 2 is a flow chart of a monitoring and early warning method for three-phase imbalance in low-voltage power distribution according to an embodiment of the present invention;

[0020] Figure 3 is a schematic diagram of a monitoring and early warning system for three-phase imbalance in low-voltage power distribution according to an embodiment of the present invention;

[0021] Figure 4 It is a schematic diagram of a device for determining an early warning strategy of a distribution network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] 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.

[0023] 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, functional component or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, functional components or devices.

[0024] According to an embodiment of the present invention, an embodiment of a method for determining an early warning strategy for a distribution network 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.

[0025] Figure 1 is a flow chart of a method for determining an early warning strategy for a distribution network according to an embodiment of the present invention. Figure 1 As shown, the method may include the following steps:

[0026] Step S101, obtaining monitoring data of the distribution network.

[0027] In the technical solution provided in the above step S101 of the present invention, the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance. The monitoring data may also be real-time data obtained by real-time monitoring of the distribution network.

[0028] In this embodiment, monitoring data of the distribution network is obtained. For example, the real-time data of the distribution network may include any one or more of real-time three-phase current, real-time three-phase voltage, real-time frequency, real-time power factor, real-time environmental conditions, and real-time imbalance, which are only illustrative examples and do not limit the specific content of the monitoring data.

[0029] Step S102: input the monitoring data into the target prediction model for analysis to obtain the prediction result of the distribution network.

[0030] In the technical solution provided in the above step S102 of the present invention, the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training the initial prediction model using historical monitoring data.

[0031] In this embodiment, after the monitoring data of the distribution network is acquired in step S101, the monitoring data is input into the target prediction model for analysis to obtain the prediction result of the distribution network.

[0032] Optionally, the monitoring data is input into the target prediction model to determine the initial prediction results of multiple distribution networks; the prediction values ​​corresponding to the initial prediction results are compared to determine the highest prediction value; the highest prediction value is compared with the prediction threshold to obtain a comparison result; and the prediction result of the distribution network is determined according to the comparison result. Among them, the monitoring data can also be called real-time data, and the prediction value can also be called a three-phase imbalance situation or probability.

[0033] For example, real-time data of the distribution network is obtained and input into the target prediction model, that is, in the trained deep learning model, the three-phase imbalance with the highest probability is used as the prediction result of the model. During the prediction process, the prediction label will generate multiple probabilities, and the model will select the result with the highest probability as the prediction result.

[0034] Optionally, historical monitoring data of the distribution network is obtained; for example, historical electricity consumption data of users of branch switches is called up through the metering system to obtain historical three-phase current, historical three-phase voltage, historical frequency, historical power factor, historical environmental conditions, historical imbalance degree and historical imbalance state. This is only an illustrative example and does not limit the specific method and content of obtaining historical monitoring data of the distribution network.

[0035] Optionally, after acquiring historical monitoring data of the distribution network, the initial prediction model is trained using the historical monitoring data to obtain a target prediction model.

[0036] For example, the training set of the deep learning model is: historical three-phase current, historical three-phase voltage, historical frequency, historical power factor, historical environmental conditions, historical unbalance degree and historical unbalance state. Among them, the historical unbalance state is used as a prediction label, and the historical unbalance state includes the historical three-phase unbalance state, which is assigned the label 1, and also includes the historical three-phase balance state, which is assigned the label 0.

[0037] Step S103, in response to the prediction result that the distribution network is in a three-phase unbalanced state, determining an early warning strategy for the distribution network.

[0038] In the technical solution provided in the above step S103 of the present invention, the early warning strategy is used to indicate the rules for early warning of the distribution network.

[0039] In this embodiment, after the prediction result of the distribution network is obtained in step S102, when the prediction result is that the distribution network is in a three-phase unbalanced state, an early warning strategy for the distribution network is determined.

[0040] For example, a load shedding plan is formulated for users corresponding to the predicted three-phase imbalance, and the user list is transmitted to the low-voltage transparent management platform. The operation and maintenance management personnel review the load shedding plan and then carry out load shedding to complete the implementation of the early warning mechanism.

[0041] It should be noted that the above embodiment can be implemented by a monitoring and early warning system for three-phase imbalance in low-voltage power distribution.

[0042] In the above steps S101 to S103 of the present invention, monitoring data of the distribution network is obtained, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance; the monitoring data is input into the target prediction model for analysis to obtain the prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training the initial prediction model using historical monitoring data; in response to the prediction result that the distribution network is in a three-phase unbalanced state, the early warning strategy of the distribution network is determined, wherein the early warning strategy is used to indicate the rules for early warning of the distribution network. That is to say, the present invention analyzes the monitoring data of the distribution network through the target prediction model to obtain the prediction result, thereby determining the early warning strategy of the distribution network according to the prediction result, avoiding the problem of limited monitoring accuracy of the three-phase imbalance of the distribution network using traditional statistical methods, thereby solving the technical problem of low early warning monitoring accuracy of the distribution network, and achieving the technical effect of improving the early warning monitoring accuracy of the distribution network.

[0043] The above method of this embodiment is further introduced below.

[0044] As an optional implementation method, the monitoring data is input into the target prediction model for analysis to obtain the prediction results of the distribution network, including: inputting the monitoring data into the target prediction model for analysis to determine the initial prediction results of multiple distribution networks; based on each initial prediction result, determining the prediction result of the distribution network.

[0045] In this embodiment, the monitoring data is input into the target prediction model for analysis to determine the initial prediction results of multiple distribution networks. For example, the target prediction model generates multiple probabilities (initial prediction results) according to the prediction labels (monitoring data). This is only an illustrative example and does not limit the specific method for determining the prediction results of the distribution network.

[0046] Optionally, after determining the initial prediction result, the prediction result of the distribution network is determined according to each initial prediction result. For example, the model will select the result with the highest probability as the prediction result. This is only an illustrative example and does not limit the specific method for determining the prediction result of the distribution network.

[0047] Optionally, the prediction results are determined by a target prediction model, thereby avoiding the problem of limited monitoring accuracy of three-phase imbalance in the distribution network using traditional statistical methods, solving the technical problem of low accuracy of early warning monitoring of the distribution network, and achieving the technical effect of improving the accuracy of early warning monitoring of the distribution network.

[0048] As an optional implementation method, based on each initial prediction result, the prediction result of the distribution network is determined, including: determining the highest prediction result among multiple initial prediction results; comparing the highest prediction result with a prediction result threshold to obtain a comparison result; and determining the prediction result of the distribution network based on the comparison result.

[0049] In this embodiment, the highest prediction result is determined among multiple initial prediction results. The highest prediction result is also the highest probability result; the highest prediction result is compared with the prediction result threshold to obtain a comparison result; based on the comparison result, the prediction result of the distribution network is determined. The present invention determines the prediction result of the distribution network by comparing multiple initial prediction results, thereby avoiding the problem of limited monitoring accuracy of the three-phase imbalance of the distribution network using traditional statistical methods, thereby solving the technical problem of low early warning monitoring accuracy of the distribution network and achieving the technical effect of improving the early warning monitoring accuracy of the distribution network.

[0050] For example, assuming that the highest predicted value is 85% and the prediction threshold is 50%, since 85%>50%, that is, the highest predicted value is greater than the prediction threshold, based on this, it can be determined that the prediction result of the distribution network is that the distribution network is in a three-phase unbalanced state.

[0051] As an optional implementation manner, the method for determining the early warning strategy of the distribution network also includes: obtaining historical monitoring data of the distribution network; inputting the historical monitoring data into the initial prediction model for training to obtain a target prediction model.

[0052] In this embodiment, historical monitoring data of the distribution network is obtained; the historical monitoring data is input into the initial prediction model for training to obtain a target prediction model. For example, the historical monitoring data of the distribution network is obtained through the metering system. This is only an example and does not limit the specific method of obtaining the historical monitoring data of the distribution network.

[0053] For example, the historical electricity consumption data of the users of the shunt switch is called up through the metering system to obtain the historical three-phase current, historical three-phase voltage, historical frequency, historical power factor, historical environmental conditions, historical imbalance and historical imbalance state. Among them, the historical frequency refers to the historical frequency of the distribution network; the historical imbalance is a quantitative indicator used to measure the degree of imbalance of the three-phase current or voltage. The historical imbalance is obtained by calculating the new imbalance factor. The specific calculation method is shown in the following formula (1):

[0054]

[0055] Among them, Δ represents the historical imbalance, I A ,I B ,I C Represent the amplitudes of the three-phase currents respectively.

[0056] As an optional implementation manner, before using historical monitoring data to train the initial prediction model to obtain the target prediction model, the method for determining the early warning strategy of the distribution network also includes: preprocessing the historical monitoring data.

[0057] In this embodiment, the historical monitoring data is preprocessed by data cleaning, standardization, and filling missing values.

[0058] For example, historical three-phase current, historical three-phase voltage, historical frequency, historical power factor, historical environmental conditions, historical unbalance degree and historical unbalance state are preprocessed to obtain preprocessed data. Preprocessing mainly involves data cleaning, standardization and filling missing values ​​for the above data.

[0059] As an optional implementation method, the method for determining the early warning strategy of the distribution network also includes: obtaining a loss function, wherein the loss function is used to train an initial prediction model, the loss function contains physical constraint terms, and the physical constraint terms are used to optimize the corresponding historical monitoring data when the distribution network is in a three-phase unbalanced state; using the loss function to train the initial prediction model to obtain a target prediction model.

[0060] In this embodiment, a loss function is obtained and used to train an initial prediction model, wherein the loss function is a function that balances the number of samples.

[0061] For example, the loss function is shown in the following formulas (2) to (4):

[0062] L=L i +γL p (2)

[0063]

[0064]

[0065] Among them, L is the loss function, L i is the loss term, L p is the physical constraint term, α is the positive sample weight. Since the model has few positive samples and is intended to predict the results of three-phase imbalance, α is preferably set to 0.8, and y is the actual value in the historical data. is the probability of the deep learning model predicting the three imbalanced situations (the probability that the model predicts that the sample is a positive sample), θ m represents the maximum allowable imbalance (determined based on the historical imbalance measured), and γ is the adjustment coefficient.

[0066] It should be noted that the above embodiment can be implemented by a monitoring and early warning system for three-phase imbalance in low-voltage power distribution.

[0067] In this embodiment, monitoring data of the distribution network is obtained, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance; the monitoring data is input into the target prediction model for analysis to obtain the prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training the initial prediction model using historical monitoring data; in response to the prediction result that the distribution network is in a three-phase unbalanced state, the early warning strategy of the distribution network is determined, wherein the early warning strategy is used to indicate the rules for early warning of the distribution network. That is to say, the present invention analyzes the monitoring data of the distribution network through the target prediction model to obtain the prediction result, thereby determining the early warning strategy of the distribution network according to the prediction result, avoiding the problem of limited monitoring accuracy of the three-phase imbalance of the distribution network using traditional statistical methods, thereby solving the technical problem of low early warning monitoring accuracy of the distribution network, and achieving the technical effect of improving the early warning monitoring accuracy of the distribution network.

[0068] The technical solution of the embodiment of the present invention is illustrated below in conjunction with preferred implementation modes.

[0069] At present, with the development of smart grid, the monitoring and maintenance of low-voltage distribution system has become particularly important. Three-phase imbalance is one of the common problems in low-voltage distribution network, which can cause equipment overheating, reduced efficiency, and even cause serious electrical failures. Therefore, effective monitoring and early warning of three-phase imbalance are particularly necessary.

[0070] In the related art, the three-phase imbalance monitoring and early warning methods for low-voltage power distribution systems mainly rely on traditional statistical analysis methods. However, the traditional statistical analysis method has limited monitoring accuracy for three-phase imbalance and cannot accurately judge the severity and scope of the imbalance. At the same time, the traditional method only relies on simple current differences to judge three-phase imbalance, ignoring the influence of other key factors, such as voltage fluctuations, frequency changes, environmental conditions, etc., which may lead to low accuracy of early warning and easy to generate false alarms or missed reports. Therefore, there is a technical problem of low accuracy of early warning monitoring of distribution networks.

[0071] With regard to the above-mentioned technical problem of low accuracy of early warning monitoring of distribution networks, no effective solution has been proposed so far.

[0072] However, the embodiment of the present invention proposes a monitoring and early warning method for three-phase imbalance in low-voltage power distribution, which obtains real-time data of the distribution network, including any one or more of real-time three-phase current, real-time three-phase voltage, real-time frequency, real-time power factor, real-time environmental conditions, and real-time imbalance, and inputs it into the trained deep learning model, outputs the probability of three-phase imbalance in the distribution network, and implements an early warning mechanism for the predicted three-phase imbalance. The technical problem of low accuracy of early warning monitoring of the distribution network is solved, and the technical effect of improving the accuracy of early warning monitoring of the distribution network is achieved.

[0073] The embodiments of the present invention are further described below.

[0074] Figure 2 1 is a flow chart of a monitoring and early warning method for three-phase imbalance in low-voltage power distribution according to an embodiment of the present invention. The analysis method comprises the following steps:

[0075] Step S201, calling for the user's historical electricity consumption data of the branch switch.

[0076] In this embodiment, the user's historical electricity consumption data of the branch switch is called up through the metering system to obtain the historical three-phase current, historical three-phase voltage, historical frequency, historical power factor, historical environmental conditions, historical imbalance degree and historical imbalance state.

[0077] Optionally, the historical frequency refers to the historical frequency of the distribution network. The historical unbalance degree is a quantitative index used to measure the degree of unbalance of three-phase current or voltage, and is obtained by calculating the new unbalance factor. The specific calculation method is shown in the above formula (1), which will not be repeated here.

[0078] Step S202: pre-process the user's historical electricity consumption data to obtain pre-processed data.

[0079] In this embodiment, the user's historical electricity consumption data in step S201 is preprocessed.

[0080] Optionally, the historical three-phase current, the historical three-phase voltage, the historical frequency, the historical power factor, the historical environmental conditions, the historical unbalance degree and the historical unbalance state are preprocessed to obtain preprocessed data.

[0081] Optionally, the preprocessing mainly involves data cleaning, standardization, and filling missing values ​​on the above data.

[0082] Step S203: predicting the three-phase imbalance situation through a deep learning model.

[0083] In this embodiment, the deep learning model is selected from but not limited to a long short-term memory network (Long Short-Term Memory, referred to as LSTM) or a gated recurrent unit (Gated Recurrent Unit, referred to as GRU).

[0084] Optionally, the training set of the deep learning model is: historical three-phase current, historical three-phase voltage, historical frequency, historical power factor, historical environmental conditions, historical unbalance degree and historical unbalance state. Among them, the historical unbalance state is used as a prediction label, and the historical unbalance state includes the historical three-phase unbalance state, which is assigned a label of 1, and also includes the historical three-phase balanced state, which is assigned a label of 0.

[0085] Optionally, the present application trains the deep learning model through a training set, and at the same time, constructs a loss function of the deep learning model to optimize the model to deal with the three-phase imbalance problem.

[0086] Optionally, during the data call test process, the distribution network is in a stable operating state. Only when encountering equipment abnormalities, failures, or sudden changes in load will three-phase imbalance occur. This will lead to an imbalance between positive and negative samples in step a and the data in the training set. The number of negative samples is much higher than the number of positive samples, which makes the model more biased towards generating prediction results with negative samples, which is likely to cause deviations in the prediction results. Therefore, the present application makes a special design for the loss function.

[0087] Optionally, positive samples refer to data points where there is a three-phase imbalance phenomenon. In a power system, when the amplitude, phase or frequency of the three-phase current or voltage is inconsistent, there is a three-phase imbalance. Negative samples refer to data points where there is no three-phase imbalance phenomenon. In this case, the amplitude, phase and frequency of the three-phase current or voltage are consistent.

[0088] Optionally, the loss function can be calculated by the aforementioned formula (2) to formula (4), which will not be described in detail here.

[0089] Optionally, the main purpose of designing this loss function is to better handle the class imbalance problem, especially for the three-phase imbalance monitoring and early warning system, because the positive samples (i.e., the three-phase imbalance) are relatively small, so special attention needs to be paid to the prediction accuracy of such samples. Adding physical constraints L p This can make the model more in line with the needs of actual application scenarios. m When the threshold is reached, a penalty is imposed, which helps prevent the model from overconfidently predicting high-probability events while encouraging the model to make more accurate predictions near the threshold. This effectively reduces the false alarm rate, avoids frequent and unnecessary warning signals, and improves the practicality of the warning system.

[0090] Optionally, the deep learning model is trained using the above training set and loss function to obtain a trained deep learning model that can provide accurate early warnings in practical applications.

[0091] Alternatively, a long short-term memory network (LSTM) is a special type of recurrent neural network (RNN) used to process sequence data. LSTM has a stronger memory capacity than traditional RNN and can better capture long-term dependencies, thus performing better when processing time series data.

[0092] Optionally, LSTM controls the flow of information by using gating units, including forget gate, input gate, and output gate. The forget gate controls whether to forget the previous memory, the input gate controls the input of new information, and the output gate controls the output memory.

[0093] Optionally, due to the gating mechanism of LSTM, it can handle long-term dependencies more effectively and avoid problems such as gradient vanishing or gradient exploding. Therefore, LSTM has achieved great success in the fields of language modeling, machine translation, speech recognition, etc., and has become one of the most commonly used recurrent neural network structures.

[0094] Alternatively, a recurrent neural network is a neural network model that can process sequence data. Unlike traditional feedforward neural networks, RNNs have recurrent connections that can pass information between different time steps.

[0095] Alternatively, RNN is widely used in tasks such as natural language processing, speech recognition, and time series prediction. Its main advantage is that it can process input sequences of variable length and can capture the temporal correlation in the sequence.

[0096] Alternatively, RNN also has some problems, such as gradient disappearance and gradient explosion, which make training difficult. In order to solve these problems, some improved RNN models have emerged, such as long short-term memory network (LSTM) and gated recurrent unit (GRU).

[0097] Alternatively, a gated recurrent unit is a recurrent neural network structure commonly used to process sequence data. Compared with the traditional RNN structure, GRU has a simpler design and more efficient computing performance.

[0098] Optionally, the GRU contains an update gate and a reset gate, which control the flow and retention of information. The update gate determines how much of the previous memory should be retained in the current time step, while the reset gate determines how much new information should be ignored in the current time step.

[0099] Optionally, through this gating mechanism, GRU can better handle long sequence data and long-term dependencies, while reducing problems such as gradient disappearance and gradient explosion. Therefore, GRU is widely used in tasks such as natural language processing, speech recognition, and time series prediction.

[0100] Step S204: Acquire real-time data of the distribution network, input it into the trained deep learning model, and output the probability of three-phase imbalance in the distribution network.

[0101] In this embodiment, real-time data of the distribution network is obtained, which may include any one or more of real-time three-phase current, real-time three-phase voltage, real-time frequency, real-time power factor, real-time environmental conditions, and real-time imbalance, which are input into a trained deep learning model to output the probability of three-phase imbalance in the distribution network.

[0102] Optionally, the three-phase unbalanced condition with the highest probability is used as the prediction result of the model. During the prediction process, multiple probabilities are generated based on the prediction labels, and the model selects the result with the highest probability as the prediction result.

[0103] Step S205: Implementing an early warning mechanism for the predicted three-phase imbalance situation.

[0104] In this embodiment, a load shedding plan is formulated for users corresponding to the predicted three-phase imbalance, and the user list is transmitted to the low-voltage transparent management platform. The operation and maintenance management personnel review the load shedding plan and then perform load shedding to complete the implementation of the early warning mechanism. According to an embodiment of the present invention, a device for determining an early warning strategy for a distribution network is also provided. It should be noted that the device for determining an early warning strategy for a distribution network can be used to execute the method for determining an early warning strategy for a distribution network in the method embodiment.

[0105] Figure 3Schematic diagram of a monitoring and early warning system for three-phase imbalance in low-voltage power distribution according to an embodiment of the present invention. Figure 3 As shown, the monitoring and early warning system 300 includes: a call module 301, a preprocessing module 302, a prediction module 303, an output module 304 and an early warning module 305.

[0106] The call test module 301 is used to call the user's historical power consumption data of the branch switch through the metering system to obtain the historical three-phase current, historical three-phase voltage, historical frequency, historical power factor, historical environmental conditions, historical imbalance degree and historical imbalance state.

[0107] The preprocessing module 302 is connected to the call detection module 301 and is used to preprocess the historical three-phase current, historical three-phase voltage, historical frequency, historical power factor, historical environmental conditions, historical unbalance degree and historical unbalance state to obtain preprocessed data.

[0108] The prediction module 30 is connected to the preprocessing module 302 and is used to predict the three-phase imbalance situation through a deep learning model.

[0109] Output module 304: connected to the prediction module 303, used to obtain real-time data of the distribution network, which may include any one or more of real-time three-phase current, real-time three-phase voltage, real-time frequency, real-time power factor, real-time environmental conditions, and real-time imbalance, and input it into the trained deep learning model to output the probability of three-phase imbalance in the distribution network.

[0110] Early warning module 305: connected to output module 304, used to implement early warning mechanism for the predicted three-phase imbalance situation.

[0111] In this embodiment, the real-time data of the distribution network is obtained, which may include any one or more of real-time three-phase current, real-time three-phase voltage, real-time frequency, real-time power factor, real-time environmental conditions, and real-time imbalance, and is input into the trained deep learning model, and the probability of three-phase imbalance in the distribution network is output, and the predicted three-phase imbalance is implemented with an early warning mechanism. The technical problem of low accuracy of early warning monitoring of the distribution network is solved, and the technical effect of improving the accuracy of early warning monitoring of the distribution network is achieved.

[0112] Figure 4 Schematic diagram of a device for determining an early warning strategy for a distribution network according to an embodiment of the present invention. Figure 4 As shown, the device 400 for determining the early warning strategy of the distribution network may include: an acquisition unit 401 , an analysis unit 402 and a determination unit 403 .

[0113] The acquisition unit 401 is used to acquire monitoring data of the distribution network, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance.

[0114] The analysis unit 402 is used to input the monitoring data into the target prediction model for analysis to obtain the prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training the initial prediction model using historical monitoring data.

[0115] The determination unit 403 is configured to determine an early warning strategy for the distribution network in response to the prediction result that the distribution network is in a three-phase unbalanced state, wherein the early warning strategy is used to indicate a rule for early warning of the distribution network.

[0116] Optionally, the analysis unit 402 may include: a first determination module, used to input the monitoring data into the target prediction model to determine the initial prediction results of multiple distribution networks; and a second determination module, used to determine the prediction results of the distribution network based on each initial prediction result.

[0117] Optionally, the determination unit 403 may include: a third determination module, used to determine the highest prediction result among multiple initial prediction results; a comparison module, used to compare the highest prediction result with a prediction result threshold to obtain a comparison result; and a fourth determination module, used to determine the prediction result of the distribution network based on the comparison result.

[0118] Optionally, the device 400 for determining the early warning strategy of the distribution network may further include: a first acquisition unit, used to acquire historical monitoring data of the distribution network; and a first training unit, used to input the historical monitoring data into an initial prediction model for training to obtain a target prediction model.

[0119] Optionally, the device 400 for determining an early warning strategy for a distribution network may further include: a preprocessing unit, configured to preprocess historical monitoring data.

[0120] Optionally, the device 400 for determining the early warning strategy of the distribution network may include: a second acquisition unit, used to obtain a loss function, wherein the loss function is used to train an initial prediction model, the loss function contains physical constraint terms, and the physical constraint terms are used to optimize the corresponding historical monitoring data when the distribution network is in a three-phase unbalanced state; a second training unit, used to use the loss function to train the initial prediction model to obtain a target prediction model.

[0121] In this embodiment, monitoring data of the distribution network is obtained, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance; the monitoring data is input into the target prediction model for analysis to obtain the prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training the initial prediction model using historical monitoring data; in response to the prediction result that the distribution network is in a three-phase unbalanced state, the early warning strategy of the distribution network is determined, wherein the early warning strategy is used to indicate the rules for early warning of the distribution network. That is to say, the present invention analyzes the monitoring data of the distribution network through the target prediction model to obtain the prediction result, thereby determining the early warning strategy of the distribution network according to the prediction result, avoiding the problem of limited monitoring accuracy of the three-phase imbalance of the distribution network using traditional statistical methods, thereby solving the technical problem of low early warning monitoring accuracy of the distribution network, and achieving the technical effect of improving the early warning monitoring accuracy of the distribution network.

[0122] According to an embodiment of the present invention, there is further provided a computer-readable storage medium, the storage medium comprising a stored program, wherein the program executes the method for determining the early warning strategy of the distribution network in the method embodiment.

[0123] According to an embodiment of the present invention, a processor is further provided. The processor is used to run a program, wherein when the program is run, the method for determining the early warning strategy of the distribution network in the method embodiment is executed.

[0124] According to an embodiment of the present invention, a computer program product is further provided. The computer program product includes computer instructions. When the computer instructions are executed by a processor, the method for determining the early warning strategy of the distribution network in the method embodiment is implemented.

[0125] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0126] 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.

[0127] In the several embodiments provided in this application, 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. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that 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.

[0128] 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.

[0129] 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.

[0130] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent functional component, 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 functional component, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or 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, disk or optical disk, etc., which can store program codes.

[0131] 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 an early warning strategy for a distribution network, characterized in that: include: Acquire monitoring data of the distribution network, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and imbalance; Inputting the monitoring data into a target prediction model for analysis to obtain a prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training an initial prediction model using historical monitoring data; In response to the prediction result that the distribution network is in the three-phase unbalanced state, an early warning strategy for the distribution network is determined, wherein the early warning strategy is used to indicate a rule for providing an early warning for the distribution network.

2. The method according to claim 1, characterized in that The monitoring data is input into the target prediction model for analysis to obtain the prediction result of the distribution network, including: Inputting the monitoring data into the target prediction model for analysis to determine initial prediction results of a plurality of the distribution networks; Based on each of the initial prediction results, the prediction result of the power distribution network is determined.

3. The method according to claim 2, characterized in that Determining the prediction result of the distribution network based on each of the initial prediction results includes: Determine the highest prediction result among the plurality of initial prediction results; Comparing the highest prediction result with the prediction result threshold to obtain a comparison result; Based on the comparison result, the prediction result of the power distribution network is determined.

4. The method according to claim 1, characterized in that The method further comprises: Acquiring the historical monitoring data of the distribution network; The historical monitoring data is input into the initial prediction model for training to obtain the target prediction model.

5. The method according to claim 4, characterized in that Before the initial prediction model is trained using the historical monitoring data to obtain the target prediction model, the method further includes: The historical monitoring data is preprocessed.

6. The method according to any one of claims 1 to 5, characterized in that The method further comprises: Obtaining a loss function, wherein the loss function is used to train the initial prediction model, the loss function includes a physical constraint term, and the physical constraint term is used to optimize the corresponding historical monitoring data when the distribution network is in the three-phase unbalanced state; The loss function is used to train the initial prediction model to obtain the target prediction model.

7. A device for determining an early warning strategy for a distribution network, characterized in that: include: An acquisition unit, configured to acquire monitoring data of a distribution network, wherein the monitoring data includes at least one of the following of the distribution network: three-phase current, three-phase voltage, frequency, power factor, environmental data, and unbalance; An analysis unit, used for inputting the monitoring data into a target prediction model for analysis to obtain a prediction result of the distribution network, wherein the prediction result is used to indicate the probability that the distribution network is in a three-phase unbalanced state, and the target prediction model is obtained by training an initial prediction model using historical monitoring data; A determination unit is used to determine an early warning strategy for the distribution network in response to the prediction result that the distribution network is in the three-phase unbalanced state, wherein the early warning strategy is used to indicate a rule for early warning of the distribution network.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed by a processor, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 6.

9. A processor, characterized in that: The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 6 when running.

10. A computer program product, characterized in that The computer program product comprises computer instructions, which implement the method according to any one of claims 1 to 6 when executed by a processor.

Images