Probabilistic power flow analysis method and system based on source-load interaction mechanism

Abstract

The invention discloses a probabilistic power flow analysis method and system based on a source-load interaction mechanism, and the method comprises the steps: employing a source-load interaction relation recognition model to recognize a source-load interaction relation based on the source-load data corresponding to a target power grid, and enabling the model to be obtained through the iterative training of a large model; probabilistic power flow analysis is carried out based on the source-load interaction relation and the source-load data; each round of training comprises the following steps: acquiring sample source load data and a label thereof; sample source load characteristics of the sample source load data are determined through a large model, and the sample source load characteristics are mapped into interaction relation characteristics; obtaining target prompt information matched with the interaction relation characteristics in a database; using a large model to generate a second sample source-load interaction relationship based on the target prompt information and the interaction relationship characteristics; the current round of training is performed on the large model based on the similarity between the first sample source load interaction relationship and the second sample source load interaction relationship, so that the analysis result precision can be improved, and more accurate reference information is provided for planning and scheduling of a power system.

Description

Technical Field

[0001] The present application relates to the field of electrical technology, and in particular to a probabilistic power flow analysis method and system based on a source-load interaction mechanism. Background Art

[0002] In recent years, the proportion of renewable energy in total power generation has been increasing. With the large-scale access of renewable energy with intermittent power generation, the uncertainty of the power system has increased dramatically. The increase in uncertainty in the power system is likely to cause an impact on the power grid, threatening the safe and stable operation of the power grid. In the field of uncertainty analysis of modern power systems, probabilistic power flow analysis is often used as an important analysis tool. Probabilistic power flow analysis can fully consider various random factors and use the analysis results to provide comprehensive and important reference information for power system planning and operation scheduling, so as to facilitate subsequent planning and scheduling of the power system.

[0003] In the related art, the collected historical source and load data are usually used directly for probabilistic power flow analysis. In this process, a large number of uncertain factors are generated by random simulation based on the historical source and load data for further analysis. However, the uncertain factors of random simulation may not conform to the actual situation of the power grid to be analyzed, resulting in low accuracy of the analysis results for the power grid to be analyzed. Summary of the invention

[0004] In order to solve the above technical problems, the embodiment of the present application proposes a probabilistic power flow analysis method and system based on a source-load interaction mechanism, which can make the probabilistic power flow analysis results more consistent with the actual situation of the power grid to be analyzed, thereby improving the accuracy of the analysis results.

[0005] In a first aspect, an embodiment of the present application provides a probabilistic power flow analysis method based on a source-load interaction mechanism, including: Based on the source-load data corresponding to the target power grid, a pre-trained source-load interaction relationship identification model is used to identify the source-load interaction relationship, wherein the source-load interaction relationship identification model is obtained by iteratively training a large model; Based on the source-load interaction relationship and the source-load data, a probabilistic power flow analysis is performed; Wherein, each round of training in the iterative training includes: Acquire sample source-load data and its label, wherein the label indicates a first sample source-load interaction relationship; Determine the sample source-load characteristics of the sample source-load data through the large model, and map the sample source-load characteristics into interactive relationship characteristics; Acquire target prompt information matching the interactive relationship feature in a prompt information database; Based on the target prompt information and the interactive relationship characteristics, using the large model to generate a second sample source-load interactive relationship; Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, the large model is trained in a current round.

[0006] Optionally, the large model includes a pre-trained source-load feature extraction network, and the sample source-load feature is obtained by extracting features from the sample source-load data using the source-load feature extraction network; The training method of the source-load feature extraction network includes: Acquire sample features and corresponding data thereof, wherein the modality of the data corresponding to the sample features is the same as the modality of the sample source-load data; Using the source-load feature extraction network to be trained, extracting features from the data corresponding to the sample feature to obtain a first feature; Based on the difference between the first feature and the sample feature, the source-load feature extraction network to be trained is trained.

[0007] Optionally, the large model includes a mapping relationship network to be trained; The sample source-load characteristics are mapped into interactive relationship characteristics through the large model, including: Inputting the sample source-load characteristics into the mapping relationship network, so that the mapping relationship network maps the sample source-load characteristics into the interaction relationship characteristics according to its current network parameters; The performing a current round of training on the large model based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship includes: Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, a current round of adjustment is performed on the current network parameters of the mapping relationship network.

[0008] Optionally, the inputting the sample source-load feature into the mapping relationship network so that the mapping relationship network maps the sample source-load feature into the interactive relationship feature according to its current network parameters includes: Inputting the sample source-load characteristics into the mapping relationship network; The mapping relationship network maps the sample source-load feature to at least one interactive relationship sub-feature according to its current network parameters, wherein the at least one interactive relationship sub-feature corresponds one-to-one to at least one source-load interactive relationship type, the at least one source-load interactive relationship type is included in a plurality of preset source-load interactive relationship types, and an interactive relationship sub-feature characterizes the interactive relationship embodied by the sample source-load data under the corresponding source-load interactive relationship type; The mapping relationship network performs fusion processing on the at least one interactive relationship sub-feature according to its current network parameters to obtain the interactive relationship feature.

[0009] Optionally, the current network parameters of the mapping relationship network include the current weights of the multiple source-load interaction relationship types, and the mapping relationship network performs fusion processing on the at least one interaction relationship sub-feature according to its current network parameters to obtain the interaction relationship feature, including: The mapping relationship network performs weighted fusion processing on the at least one interactive relationship sub-feature according to the current weight of each of the at least one source-load interactive relationship types to obtain the interactive relationship feature.

[0010] Optionally, the prompt information database contains a plurality of prompt information associated with the plurality of source-load interaction relationship types one by one, a prompt information includes a text representation for indicating a corresponding source-load interaction relationship type, and the target prompt information includes the prompt information associated with each of the at least one source-load interaction relationship types; The step of generating a second sample source-load interaction relationship based on the target prompt information and the interaction relationship feature using the large model includes: The target prompt information and the interactive relationship features are input into the large model, so that the large model uses the target text representation as the inference prompt and the interactive relationship features as the inference basis to perform inference, thereby generating the second sample source-load interactive relationship, wherein the target text representation refers to the text representation contained in the prompt information associated with each of the at least one source-load interactive relationship types.

[0011] Optionally, the similarity is determined by the similarity between a second feature and a third feature, wherein the second feature is obtained by extracting features of the source-charge interaction relationship of the first sample, and the third feature is obtained by extracting features of the source-charge interaction relationship of the second sample.

[0012] Optionally, the performing probabilistic power flow analysis based on the source-load interaction relationship and the source-load data includes: Constructing a probability distribution model according to the source load data; Based on the probability distribution model and the source-load interaction relationship, a pre-configured probability power flow calculation method is used to perform probability power flow analysis, wherein the probability power flow calculation method includes any one of the following: Monte Carlo simulation method, analytical method, and semi-invariant method.

[0013] Optionally, the probabilistic power flow calculation method at least includes a Monte Carlo simulation method, and the probabilistic power flow analysis is performed using a preconfigured probabilistic power flow calculation method based on the probability distribution model and the source-load interaction relationship, including: generating a random sample using the probability distribution model, wherein the random sample indicates a source-load correlation scenario; Modifying the random sample according to the source-load interaction relationship; Probabilistic power flow analysis is performed using Monte Carlo simulation based on modified random samples.

[0014] In a second aspect, an embodiment of the present application provides a probabilistic power flow analysis system based on a source-load interaction mechanism, including: A relationship identification module is used to identify the source-load interaction relationship based on the source-load data corresponding to the target power grid using a pre-trained source-load interaction relationship identification model, wherein the source-load interaction relationship identification model is obtained by iteratively training a large model; A probabilistic power flow analysis module, used for performing probabilistic power flow analysis based on the source-load interaction relationship and the source-load data; Wherein, each round of training in the iterative training includes: Acquire sample source-load data and its label, wherein the label indicates a first sample source-load interaction relationship; Determine the sample source-load characteristics of the sample source-load data through the large model, and map the sample source-load characteristics into interactive relationship characteristics; Acquire target prompt information matching the interactive relationship feature in a prompt information database; Based on the target prompt information and the interactive relationship characteristics, using the large model to generate a second sample source-load interactive relationship; Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, the large model is trained in a current round.

[0015] In summary, the embodiments of the present application have at least the following beneficial effects: According to an embodiment of the present application, a source-load interaction relationship is identified by using a pre-trained source-load interaction relationship identification model based on source-load data corresponding to a target power grid, wherein the source-load interaction relationship identification model is obtained by iteratively training a large model; a probabilistic power flow analysis is performed based on the source-load interaction relationship and the source-load data; wherein each round of training in the iterative training includes: obtaining sample source-load data and its label, wherein the label indicates a first sample source-load interaction relationship; determining a sample source-load feature of the sample source-load data through the large model, and mapping the sample source-load feature into an interaction relationship feature; obtaining a feature corresponding to the interaction relationship feature in a prompt information database; the target prompt information matching the characteristics of the target prompt information and the interactive relationship features, using the large model to generate a second sample source-load interactive relationship; based on the similarity between the first sample source-load interactive relationship and the second sample source-load interactive relationship, the large model is trained for the current round, so that the identified source-load interactive relationship can accurately reflect the actual situation between the load side and the power generation side of the target power grid, so that the actual situation can be fully taken into account when performing probabilistic power flow analysis, so that the probabilistic power flow analysis results are closer to the actual situation of the power grid to be analyzed, improve the accuracy of the analysis results, and provide more accurate reference information for power system planning and operation scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of a probabilistic power flow analysis method based on a source-load interaction mechanism provided in an embodiment of the present application; Figure 2 It is a flowchart of the training process of the source-load interaction relationship recognition model provided in the embodiment of the present application; Figure 3 It is a structural schematic diagram of a probabilistic power flow analysis system based on a source-load interaction mechanism provided in an embodiment of the present application; Figure 4 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] In the description of the present application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, features defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "multiple" is two or more. In the description of the present application, the term "including" and its variations are open inclusions, i.e., "including but not limited to". The term "based on" means "at least partially based on". The term "according to" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments".

[0019] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0020] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this application have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0021] First, see Figure 1 , shows a flow chart of a probabilistic power flow analysis method based on a source-load interaction mechanism provided in an embodiment of the present application, the method includes steps S101-S102, which are as follows: S101, based on the source-load data corresponding to the target power grid, a source-load interaction relationship is identified using a pre-trained source-load interaction relationship identification model, wherein the source-load interaction relationship identification model is obtained by iteratively training a large model.

[0022] In one example, S101 may include: inputting source-load data corresponding to the target power grid into a pre-trained source-load interaction relationship identification model for identification, so as to obtain an identified source-load interaction relationship.

[0023] It should be noted that the source-load interaction relationship / sample source-load interaction relationship described in each embodiment of the present application is suitable for indicating at least one of the following: the interaction relationship between the load and the generator output in the target area corresponding to the target power grid, the interaction relationship between the load and the generator output in any sub-area contained in the target area, the interaction relationship between a type of load and different types of generator outputs in the target area or any sub-area, and the interaction relationship between a type of generator output and different types of loads in the target area or any sub-area.

[0024] S102: Performing a probabilistic power flow analysis based on the source-load interaction relationship and the source-load data.

[0025] In one example, S102 may include: constructing a probability distribution model based on the source-load interaction relationship and the source-load data; performing a probability power flow analysis using a pre-configured probability power flow calculation method according to the constructed probability distribution model, wherein the probability power flow calculation method includes any one of the following: Monte Carlo simulation method, analytical method, and semi-invariant method.

[0026] Furthermore, based on the source-load interaction relationship and the source-load data, constructing a probability distribution model may include: Based on the source data in the source load data, a first probability distribution model is constructed using a first distribution function.

[0027] Based on the load data in the source load data, a second probability distribution model is constructed using a second distribution function.

[0028] Based on the source-load interaction relationship and the source-load data, a third probability distribution model is constructed using a third distribution function.

[0029] Based on the first probability distribution model, the second probability distribution model and the third probability distribution model, a joint probability distribution model is constructed as the constructed probability distribution model.

[0030] Exemplarily, at least one of the first distribution function, the second distribution function, and the third distribution function can be determined according to at least one of the following: normal distribution, Weibull distribution, lognormal distribution, exponential distribution, and uniform distribution.

[0031] Among them, see Figure 2 , shows a schematic diagram of the process of each round of training in the iterative training provided in an embodiment of the present application, and each round of training includes steps S201-S205, which are specifically as follows: S201, obtaining sample source-load data and its label, wherein the label indicates a first sample source-load interaction relationship.

[0032] In one example, the sample source-load data may include historical load data, generator output data, weather data, holiday information, etc. The label may be annotated in the sample source-load data by manual annotation.

[0033] In one example, before S202, the acquired historical source load data may be preprocessed to use the preprocessed historical source load data as new historical source load data, wherein the data preprocessing may include at least one of the following: data cleaning, data standardization, and feature engineering. Among them, data cleaning is suitable for processing missing values, outliers, and noise to ensure the quality and integrity of the data. Data standardization is suitable for standardizing or normalizing the data to facilitate model training. Feature engineering is suitable for extracting useful features, such as time features (hours, dates, seasons), weather features (temperature, humidity), holiday signs, etc.

[0034] In one example, before S202, a suitable large language model can also be selected as the large model required for training. For example, an existing large language model can be selected, such as Transformer, BERT (Bidirectional Encoder Representations from Transformers), GPT (Generative Pre-trained Transformer), etc. Then the acquired / preprocessed historical source load data is converted into a text format that matches the selected large language model, and the converted data is serialized to obtain a long sequence as the sample source load data for training. The sample source load data used for training is then divided into a training set and a validation set. Exemplarily, the ratio of the training set to the validation set can be 80%:20%.

[0035] S202, determining the sample source-load characteristics of the sample source-load data through the large model, and mapping the sample source-load characteristics into interactive relationship characteristics.

[0036] In one example, S202 may include: encoding the sample source-load data through an encoder included in the large model to generate corresponding sample source-load features, and mapping the sample source-load features into interactive relationship features through a mapping relationship table pre-configured in the large model, wherein the mapping relationship table may record the mapping relationship between various types of sample source-load features and various types of interactive relationship features.

[0037] In one example, the sample source load data may be converted into text form in advance, so that the large model can perform feature extraction on the sample source load data in text form.

[0038] S203: Acquire target prompt information matching the interactive relationship feature from a prompt information database.

[0039] In one example, the prompt information database may include multiple prompt information corresponding to various types of interactive relationship features, wherein an interactive relationship feature corresponds to at least one of the multiple prompt information. Thus, the target prompt information is all prompt information corresponding to the mapped interactive relationship features.

[0040] In one example, each prompt information contained in the prompt information database may be stored in a text form.

[0041] S204: Based on the target prompt information and the interactive relationship characteristics, use the large model to generate a second sample source-load interactive relationship.

[0042] It can be understood that adding target prompt information can enable the large model to generate the second sample source-load interaction relationship more efficiently and accurately.

[0043] In one example, the target prompt information may be text information suitable for describing the reasoning prompt process related to the interactive relationship feature, so as to prompt the large model how to more efficiently perform reasoning related to the interactive relationship feature. Thus, S204 may include: inputting the target prompt information and the interactive relationship feature into the large model, so that the large model first extracts the text features of the target prompt information, then uses the extracted text features as reasoning prompts, and performs reasoning based on the mapped interactive relationship features to obtain the second sample source-load interactive relationship.

[0044] In one example, before S205, the method may further include: setting training parameters for the current round of training of the large model, wherein the training parameters to be set may include at least one of the following: hyperparameters, loss functions, and optimizers.

[0045] Among them, the hyperparameters may include at least one of the following: learning rate, batch size, and training rounds.

[0046] The loss function may include at least one of the following: mean square error, cross entropy loss.

[0047] The optimizer may include at least one of the following: Adam (Adaptive Moment Estimation) and SGD (Stochastic Gradient Descent).

[0048] S205: Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, perform a current round of training on the large model.

[0049] In one example, the current round of training in S205 may include: calculating the loss function based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship to update the model parameters through back propagation, and / or, regularly (i.e., during a round of training in iterative training) evaluating the model performance on the validation set to adjust the hyperparameters and model structure according to the evaluation results, thereby completing the current round of training. The evaluation result may be determined by calculating at least one of the following indicators: mean square error, root mean square error, and determination coefficient.

[0050] In one example, the first sample source-load interaction relationship and the second sample source-load interaction relationship can both be expressed in text form. In this way, a general text similarity model can be used to determine the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship for characterizing the similarity, wherein the similarity can be obtained by weighted calculation of the corresponding cosine similarity, Euclidean distance, Manhattan distance and / or Pearson correlation coefficient.

[0051] In one example, the large model can also use any number of models in the above embodiments for integrated learning through model integration to improve the robustness and accuracy of the prediction.

[0052] In an optional implementation, the large model includes a pre-trained source-load feature extraction network, and the sample source-load feature is obtained by extracting features from the sample source-load data using the source-load feature extraction network.

[0053] It should be noted that the source-load feature extraction network in this embodiment is pre-trained before the large model is iteratively trained, that is, it is pre-trained before the current round of training of the large model. Usually, a general feature extraction network with the same modality as the sample source-load data can be directly configured into the large model for direct call, so that when the large model is iteratively trained, there is no need to train the source-load feature extraction network, making the training of the large model more efficient. This is because the form of the sample source-load data is usually fixed and has nothing to do with the region where the power grid is located.

[0054] The training method of the source-load feature extraction network includes: Sample features and corresponding data are obtained, wherein the modality of the data corresponding to the sample features is the same as the modality of the sample source-load data.

[0055] Using the source-load feature extraction network to be trained, feature extraction is performed on the data corresponding to the sample feature to obtain a first feature.

[0056] Based on the difference between the first feature and the sample feature, the source-load feature extraction network to be trained is trained.

[0057] In one example, the source-load feature extraction network to be trained may be composed of encoders of corresponding modalities.

[0058] In an example, the difference between the first feature and the sample feature can be determined by the similarity between the first feature and the sample feature, and the similarity can be obtained by weighted calculation of corresponding cosine similarity, Euclidean distance, Manhattan distance and / or Pearson correlation coefficient.

[0059] In an optional implementation, the large model includes a mapping relationship network to be trained.

[0060] The sample source-load characteristics are mapped into interactive relationship characteristics through the large model, including: The sample source-load characteristics are input into the mapping relationship network, so that the mapping relationship network maps the sample source-load characteristics into the interaction relationship characteristics according to its current network parameters.

[0061] The performing a current round of training on the large model based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship includes: Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, a current round of adjustment is performed on the current network parameters of the mapping relationship network.

[0062] In one example, the current network parameters of the mapping relationship network refer to the network parameters of the network before the current round of training begins, wherein the network parameters may include a preset mapping relationship table, which may record the mapping relationship between various types of sample source-load features and various types of interactive relationship features. It should be understood that the preset various types of sample source-load features may not necessarily include the extracted sample source-load features. Therefore, the mapping process can be completed by matching the sample source-load features with the highest similarity to the extracted sample source-load features from the preset various types of sample source-load features. In this way, the current round adjustment of the current network parameters of the mapping relationship network may include adjusting the mapping relationship.

[0063] In one example, the similarity between the source-load interaction relationship of a first sample and the source-load interaction relationship of a second sample can be determined based on the similarity between the source-load interaction relationship of the first sample and the source-load interaction relationship of the second sample, and the similarity can be obtained by weighted calculation of the corresponding cosine similarity, Euclidean distance, Manhattan distance and / or Pearson correlation coefficient.

[0064] In an optional implementation, the inputting the sample source-load feature into the mapping relationship network so that the mapping relationship network maps the sample source-load feature into the interactive relationship feature according to its current network parameters includes: The sample source-load characteristics are input into the mapping relationship network.

[0065] The mapping relationship network maps the sample source-load feature into at least one interactive relationship sub-feature according to its current network parameters, wherein the at least one interactive relationship sub-feature corresponds one-to-one to at least one source-load interactive relationship type, and the at least one source-load interactive relationship type is included in a plurality of preset source-load interactive relationship types. An interactive relationship sub-feature characterizes the interactive relationship reflected by the sample source-load data under the corresponding source-load interactive relationship type.

[0066] The mapping relationship network performs fusion processing on the at least one interactive relationship sub-feature according to its current network parameters to obtain the interactive relationship feature.

[0067] In one example, the mapping relationship table preset in the above embodiment may also record the mapping relationship between each type of sample source-load feature and at least one type of interactive relationship sub-feature among the various types of interactive relationship sub-features.

[0068] In one example, the fusion process may include a splicing process.

[0069] In an example, the source-load interaction relationship type may be used to indicate at least one of the following: different sub-areas in the target area corresponding to the target power grid, and the relationship between various types of source devices connected to the target power grid and various types of load devices.

[0070] In an optional implementation, the current network parameters of the mapping relationship network include the current weights of the multiple source-load interaction relationship types, and the mapping relationship network performs fusion processing on the at least one interaction relationship sub-feature according to its current network parameters to obtain the interaction relationship feature, including: The mapping relationship network performs weighted fusion processing on the at least one interactive relationship sub-feature according to the current weight of each of the at least one source-load interactive relationship types to obtain the interactive relationship feature.

[0071] It can be understood that, when the current network parameters of the mapping relationship network include the current weights of each of the multiple source-load interaction relationship types, the process of performing a current round of adjustment on the current network parameters of the mapping relationship network may also include performing a current round of adjustment on the current weight so as to use the adjusted current weight as the current weight for the next round.

[0072] In an optional embodiment, the prompt information database contains multiple prompt information associated one by one with the multiple source-load interaction relationship types, one prompt information includes a text representation for indicating the corresponding source-load interaction relationship type, and the target prompt information includes the prompt information associated with each of the at least one source-load interaction relationship types.

[0073] The step of generating a second sample source-load interaction relationship based on the target prompt information and the interaction relationship feature using the large model includes: The target prompt information and the interactive relationship features are input into the large model, so that the large model uses the target text representation as the inference prompt and the interactive relationship features as the inference basis to perform inference, thereby generating the second sample source-load interactive relationship, wherein the target text representation refers to the text representation contained in the prompt information associated with each of the at least one source-load interactive relationship types.

[0074] In an optional embodiment, the similarity is determined by the similarity between a second feature and a third feature, wherein the second feature is obtained by extracting features of the source-charge interaction relationship of the first sample, and the third feature is obtained by extracting features of the source-charge interaction relationship of the second sample.

[0075] In an example, the similarity between the second feature and the third feature may be obtained by weighted calculation of corresponding cosine similarity, Euclidean distance, Manhattan distance and / or Pearson correlation coefficient.

[0076] In an optional implementation, the performing of probabilistic power flow analysis based on the source-load interaction relationship and the source-load data includes: A probability distribution model is constructed according to the source load data.

[0077] Based on the probability distribution model and the source-load interaction relationship, a pre-configured probability power flow calculation method is used to perform probability power flow analysis, wherein the probability power flow calculation method includes any one of the following: Monte Carlo simulation method, analytical method, and semi-invariant method.

[0078] In one example, based on the probability distribution model and the source-load interaction relationship, a probability power flow analysis is performed using a preconfigured probability power flow calculation method, which may include: using the probability power flow calculation method to perform calculations according to the probability distribution model, and using the source-load interaction relationship to correct the calculation results to complete the probability power flow analysis.

[0079] In an optional embodiment, the probabilistic power flow calculation method at least includes a Monte Carlo simulation method, and the probabilistic power flow analysis is performed using a preconfigured probabilistic power flow calculation method based on the probability distribution model and the source-load interaction relationship, including: A random sample is generated using the probability distribution model, wherein the random sample indicates a source-load correlation scenario.

[0080] The random sample is modified according to the source-load interaction relationship.

[0081] Probabilistic power flow analysis is performed using Monte Carlo simulation based on modified random samples.

[0082] On the second aspect, accordingly, the embodiments of the present application also provide a probabilistic power flow analysis system based on a source-load interaction mechanism, which can implement all processes of the probabilistic power flow analysis method based on a source-load interaction mechanism provided in the above embodiments.

[0083] See also Figure 3 , shows a schematic diagram of the structure of a probabilistic power flow analysis system based on a source-load interaction mechanism provided in an embodiment of the present application, the system comprising: The relationship identification module 301 is used to identify the source-load interaction relationship based on the source-load data corresponding to the target power grid using a pre-trained source-load interaction relationship identification model, wherein the source-load interaction relationship identification model is obtained by iteratively training a large model.

[0084] The probabilistic power flow analysis module 302 is used to perform probabilistic power flow analysis based on the source-load interaction relationship and the source-load data.

[0085] Wherein, each round of training in the iterative training includes: Sample source-load data and a label thereof are obtained, wherein the label indicates a first sample source-load interaction relationship.

[0086] The sample source-load characteristics of the sample source-load data are determined through the large model, and the sample source-load characteristics are mapped into interaction relationship characteristics.

[0087] Obtain target prompt information matching the interactive relationship feature from a prompt information database.

[0088] Based on the target prompt information and the interactive relationship characteristics, a second sample source-load interactive relationship is generated using the large model.

[0089] Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, the large model is trained in a current round.

[0090] In an optional implementation, the large model includes a pre-trained source-load feature extraction network, and the sample source-load feature is obtained by extracting features from the sample source-load data using the source-load feature extraction network.

[0091] The training method of the source-load feature extraction network includes: Sample features and corresponding data are obtained, wherein the modality of the data corresponding to the sample features is the same as the modality of the sample source-load data.

[0092] Using the source-load feature extraction network to be trained, feature extraction is performed on the data corresponding to the sample feature to obtain a first feature.

[0093] Based on the difference between the first feature and the sample feature, the source-load feature extraction network to be trained is trained.

[0094] In an optional implementation, the large model includes a mapping relationship network to be trained.

[0095] The sample source-load characteristics are mapped into interactive relationship characteristics through the large model, including: The sample source-load characteristics are input into the mapping relationship network, so that the mapping relationship network maps the sample source-load characteristics into the interaction relationship characteristics according to its current network parameters.

[0096] The performing a current round of training on the large model based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship includes: Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, a current round of adjustment is performed on the current network parameters of the mapping relationship network.

[0097] In an optional implementation, the inputting the sample source-load feature into the mapping relationship network so that the mapping relationship network maps the sample source-load feature into the interactive relationship feature according to its current network parameters includes: The sample source-load characteristics are input into the mapping relationship network.

[0098] The mapping relationship network maps the sample source-load feature into at least one interactive relationship sub-feature according to its current network parameters, wherein the at least one interactive relationship sub-feature corresponds one-to-one to at least one source-load interactive relationship type, and the at least one source-load interactive relationship type is included in a plurality of preset source-load interactive relationship types. An interactive relationship sub-feature characterizes the interactive relationship reflected by the sample source-load data under the corresponding source-load interactive relationship type.

[0099] The mapping relationship network performs fusion processing on the at least one interactive relationship sub-feature according to its current network parameters to obtain the interactive relationship feature.

[0100] In an optional implementation, the current network parameters of the mapping relationship network include the current weights of the multiple source-load interaction relationship types, and the mapping relationship network performs fusion processing on the at least one interaction relationship sub-feature according to its current network parameters to obtain the interaction relationship feature, including: The mapping relationship network performs weighted fusion processing on the at least one interactive relationship sub-feature according to the current weight of each of the at least one source-load interactive relationship types to obtain the interactive relationship feature.

[0101] In an optional embodiment, the prompt information database contains multiple prompt information associated one by one with the multiple source-load interaction relationship types, one prompt information includes a text representation for indicating the corresponding source-load interaction relationship type, and the target prompt information includes the prompt information associated with each of the at least one source-load interaction relationship types.

[0102] The step of generating a second sample source-load interaction relationship based on the target prompt information and the interaction relationship feature using the large model includes: The target prompt information and the interactive relationship features are input into the large model, so that the large model uses the target text representation as the inference prompt and the interactive relationship features as the inference basis to perform inference, thereby generating the second sample source-load interactive relationship, wherein the target text representation refers to the text representation contained in the prompt information associated with each of the at least one source-load interactive relationship types.

[0103] In an optional embodiment, the similarity is determined by the similarity between a second feature and a third feature, wherein the second feature is obtained by extracting features of the source-charge interaction relationship of the first sample, and the third feature is obtained by extracting features of the source-charge interaction relationship of the second sample.

[0104] In an optional implementation, the performing of probabilistic power flow analysis based on the source-load interaction relationship and the source-load data includes: A probability distribution model is constructed according to the source load data.

[0105] Based on the probability distribution model and the source-load interaction relationship, a pre-configured probability power flow calculation method is used to perform probability power flow analysis, wherein the probability power flow calculation method includes any one of the following: Monte Carlo simulation method, analytical method, and semi-invariant method.

[0106] In an optional embodiment, the probabilistic power flow calculation method at least includes a Monte Carlo simulation method, and the probabilistic power flow analysis is performed using a preconfigured probabilistic power flow calculation method based on the probability distribution model and the source-load interaction relationship, including: A random sample is generated using the probability distribution model, wherein the random sample indicates a source-load correlation scenario.

[0107] The random sample is modified according to the source-load interaction relationship.

[0108] Probabilistic power flow analysis is performed using Monte Carlo simulation based on modified random samples.

[0109] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the probabilistic power flow analysis method based on the source-load interaction mechanism described in any one of the above are implemented.

[0110] In a fourth aspect, an embodiment of the present application provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of any of the above-mentioned probabilistic power flow analysis methods based on the source-load interaction mechanism.

[0111] In a fifth aspect, an embodiment of the present application provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the steps of the probabilistic power flow analysis method based on the source-load interaction mechanism as described above when executing the computer program.

[0112] See also Figure 4 The computer device of this embodiment includes: a processor 401, a memory 402, and a computer program stored in the memory 402 and executable on the processor 401, such as a probabilistic power flow analysis program based on a source-load interaction mechanism. When the processor 401 executes the computer program, the steps in the above-mentioned probabilistic power flow analysis method embodiments based on a source-load interaction mechanism are implemented, such as Figure 1 Steps S101-S102 are shown.

[0113] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory 402 and executed by the processor 401 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the computer device.

[0114] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may include, but is not limited to, a processor 401 and a memory 402. Those skilled in the art will appreciate that the schematic diagram is merely an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components than shown in the diagram, or may combine certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

[0115] The processor 401 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor 401 may also be any conventional processor, etc. The processor 401 is the control center of the computer device, and uses various interfaces and lines to connect various parts of the entire computer device.

[0116] The memory 402 can be used to store the computer program and / or module. The processor 401 implements various functions of the computer device by running or executing the computer program and / or module stored in the memory 402 and calling the data stored in the memory 402. The memory 402 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory 402 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0117] Wherein, if the module / unit integrated in the computer device 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 present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor 401. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0118] In summary, the embodiments of the present application have at least the following beneficial effects: According to an embodiment of the present application, a source-load interaction relationship is identified by using a pre-trained source-load interaction relationship identification model based on source-load data corresponding to a target power grid, wherein the source-load interaction relationship identification model is obtained by iteratively training a large model; a probabilistic power flow analysis is performed based on the source-load interaction relationship and the source-load data; wherein each round of training in the iterative training includes: obtaining sample source-load data and its label, wherein the label indicates a first sample source-load interaction relationship; determining a sample source-load feature of the sample source-load data through the large model, and mapping the sample source-load feature into an interaction relationship feature; obtaining a feature corresponding to the interaction relationship feature in a prompt information database; the target prompt information matching the characteristics of the target prompt information and the interactive relationship features, using the large model to generate a second sample source-load interactive relationship; based on the similarity between the first sample source-load interactive relationship and the second sample source-load interactive relationship, the large model is trained for the current round, so that the identified source-load interactive relationship can accurately reflect the actual situation between the load side and the power generation side of the target power grid, so that the actual situation can be fully taken into account when performing probabilistic power flow analysis, so that the probabilistic power flow analysis results are closer to the actual situation of the power grid to be analyzed, improve the accuracy of the analysis results, and provide more accurate reference information for power system planning and operation scheduling.

[0119] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary hardware platform, and of course, it can also be implemented entirely by hardware. Based on such an understanding, all or part of the contribution of the technical solution of the present application to the background technology can be embodied in the form of a software product, and the computer software product can be stored in a storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application or some parts of the embodiments.

[0120] The above is a preferred embodiment of the present application. 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 application. These improvements and modifications are also considered to be within the scope of protection of the present application.

Claims

1. A probabilistic power flow analysis method based on source-load interaction mechanism, characterized in that: include: Based on the source-load data corresponding to the target power grid, a pre-trained source-load interaction relationship identification model is used to identify the source-load interaction relationship, wherein the source-load interaction relationship identification model is obtained by iteratively training a large model; Based on the source-load interaction relationship and the source-load data, a probabilistic power flow analysis is performed; Wherein, each round of training in the iterative training includes: Acquire sample source-load data and its label, wherein the label indicates a first sample source-load interaction relationship; Determine the sample source-load characteristics of the sample source-load data through the large model, and map the sample source-load characteristics into interactive relationship characteristics; Acquire target prompt information matching the interactive relationship feature in a prompt information database; Based on the target prompt information and the interactive relationship characteristics, using the large model to generate a second sample source-load interactive relationship; Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, the large model is trained in a current round.

2. The probabilistic power flow analysis method based on source-load interaction mechanism according to claim 1 is characterized in that: The large model includes a pre-trained source-load feature extraction network, and the sample source-load feature is obtained by extracting features from the sample source-load data using the source-load feature extraction network; The training method of the source-load feature extraction network includes: Acquire sample features and corresponding data thereof, wherein the modality of the data corresponding to the sample features is the same as the modality of the sample source-load data; Using the source-load feature extraction network to be trained, extracting features from the data corresponding to the sample feature to obtain a first feature; Based on the difference between the first feature and the sample feature, the source-load feature extraction network to be trained is trained.

3. The probabilistic power flow analysis method based on source-load interaction mechanism according to claim 1 is characterized in that: The large model includes a mapping relationship network to be trained; The sample source-load characteristics are mapped into interactive relationship characteristics through the large model, including: Inputting the sample source-load characteristics into the mapping relationship network, so that the mapping relationship network maps the sample source-load characteristics into the interaction relationship characteristics according to its current network parameters; The performing a current round of training on the large model based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship includes: Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, a current round of adjustment is performed on the current network parameters of the mapping relationship network.

4. The probabilistic power flow analysis method based on source-load interaction mechanism according to claim 3 is characterized in that: The step of inputting the sample source-load feature into the mapping relationship network so that the mapping relationship network maps the sample source-load feature into the interactive relationship feature according to its current network parameters includes: Inputting the sample source-load characteristics into the mapping relationship network; The mapping relationship network maps the sample source-load feature to at least one interactive relationship sub-feature according to its current network parameters, wherein the at least one interactive relationship sub-feature corresponds one-to-one to at least one source-load interactive relationship type, the at least one source-load interactive relationship type is included in a plurality of preset source-load interactive relationship types, and an interactive relationship sub-feature characterizes the interactive relationship embodied by the sample source-load data under the corresponding source-load interactive relationship type; The mapping relationship network performs fusion processing on the at least one interactive relationship sub-feature according to its current network parameters to obtain the interactive relationship feature.

5. The probabilistic power flow analysis method based on source-load interaction mechanism according to claim 4 is characterized in that: The current network parameters of the mapping relationship network include the current weights of the multiple source-load interaction relationship types, and the mapping relationship network performs fusion processing on the at least one interaction relationship sub-feature according to its current network parameters to obtain the interaction relationship feature, including: The mapping relationship network performs weighted fusion processing on the at least one interactive relationship sub-feature according to the current weight of each of the at least one source-load interactive relationship types to obtain the interactive relationship feature.

6. The probabilistic power flow analysis method based on source-load interaction mechanism according to claim 4 is characterized in that: The prompt information database includes a plurality of prompt information associated with the plurality of source-load interaction relationship types, wherein one prompt information includes a text representation for indicating a corresponding source-load interaction relationship type, and the target prompt information includes the prompt information associated with each of the at least one source-load interaction relationship types; The step of generating a second sample source-load interaction relationship based on the target prompt information and the interaction relationship feature using the large model includes: The target prompt information and the interactive relationship features are input into the large model, so that the large model uses the target text representation as the inference prompt and the interactive relationship features as the inference basis to perform inference, thereby generating the second sample source-load interactive relationship, wherein the target text representation refers to the text representation contained in the prompt information associated with each of the at least one source-load interactive relationship types.

7. The probabilistic power flow analysis method based on source-load interaction mechanism according to claim 1 is characterized in that: The similarity is determined by the similarity between the second feature and the third feature, wherein the second feature is obtained by extracting the feature of the source-charge interaction relationship of the first sample, and the third feature is obtained by extracting the feature of the source-charge interaction relationship of the second sample.

8. The probabilistic power flow analysis method based on source-load interaction mechanism according to claim 1 is characterized in that: The performing of probabilistic power flow analysis based on the source-load interaction relationship and the source-load data includes: Constructing a probability distribution model according to the source load data; Based on the probability distribution model and the source-load interaction relationship, a pre-configured probability power flow calculation method is used to perform probability power flow analysis, wherein the probability power flow calculation method includes any one of the following: Monte Carlo simulation method, analytical method, and semi-invariant method.

9. The probabilistic power flow analysis method based on source-load interaction mechanism according to claim 8 is characterized in that: The probabilistic power flow calculation method at least includes a Monte Carlo simulation method. The probabilistic power flow analysis is performed using a pre-configured probabilistic power flow calculation method based on the probabilistic distribution model and the source-load interaction relationship, including: generating a random sample using the probability distribution model, wherein the random sample indicates a source-load correlation scenario; Modifying the random sample according to the source-load interaction relationship; Probabilistic power flow analysis is performed using Monte Carlo simulation based on modified random samples.

10. A probabilistic power flow analysis system based on source-load interaction mechanism, characterized in that: include: A relationship identification module is used to identify the source-load interaction relationship based on the source-load data corresponding to the target power grid using a pre-trained source-load interaction relationship identification model, wherein the source-load interaction relationship identification model is obtained by iteratively training a large model; A probabilistic power flow analysis module, used for performing probabilistic power flow analysis based on the source-load interaction relationship and the source-load data; Wherein, each round of training in the iterative training includes: Acquire sample source-load data and its label, wherein the label indicates a first sample source-load interaction relationship; Determine the sample source-load characteristics of the sample source-load data through the large model, and map the sample source-load characteristics into interactive relationship characteristics; Acquire target prompt information matching the interactive relationship feature in a prompt information database; Based on the target prompt information and the interactive relationship characteristics, using the large model to generate a second sample source-load interactive relationship; Based on the similarity between the first sample source-load interaction relationship and the second sample source-load interaction relationship, the large model is trained in a current round.

Images