Power grid control model construction method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a power grid control model construction method and device, computer equipment, a readable storage medium and a program product. The method comprises the steps that historical power grid data are acquired, and the historical power grid data are used for representing a power grid fault condition and a power grid operation state when a power grid system breaks down; constructing a power grid knowledge graph based on the historical power grid data; a power grid control rule set is generated according to the power grid knowledge graph, and the power grid control rule set comprises at least one power grid control rule; and training to obtain a power grid control model according to the historical power grid data and the power grid control rule set. By adopting the method, the power grid control accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the field of power grid technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for constructing a power grid control model. Background Art

[0002] With the development of power grid technology, the application of power grid systems is becoming more and more extensive. Currently, the power grid is usually controlled by operation and maintenance personnel based on their experience. However, the accuracy of power grid control depends on the professional ability and experience level of operation and maintenance personnel, which easily leads to low power grid control accuracy. Summary of the Invention

[0003] Based on this, it is necessary to provide a grid control model construction method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of grid control in order to address the above technical problems.

[0004] In a first aspect, the present application provides a method for constructing a power grid control model, comprising:

[0005] Acquiring historical power grid data, wherein the historical power grid data is used to characterize power grid fault conditions and power grid operating conditions when power grid system faults occur;

[0006] Building a power grid knowledge graph based on the historical power grid data;

[0007] generating a power grid control rule set according to the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule;

[0008] A power grid control model is trained based on the historical power grid data and the power grid control rule set.

[0009] In a second aspect, the present application further provides a power grid control model construction device, comprising:

[0010] An acquisition module is used to acquire historical power grid data, wherein the historical power grid data is used to characterize the power grid fault condition and the power grid operation state when the power grid system fails;

[0011] A construction module, configured to construct a power grid knowledge graph based on the historical power grid data;

[0012] A generating module, configured to generate a power grid control rule set according to the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule;

[0013] A training module is used to train a power grid control model based on the historical power grid data and the power grid control rule set.

[0014] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0015] Acquiring historical power grid data, wherein the historical power grid data is used to characterize power grid fault conditions and power grid operating conditions when power grid system faults occur;

[0016] Building a power grid knowledge graph based on the historical power grid data;

[0017] generating a power grid control rule set according to the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule;

[0018] A power grid control model is trained based on the historical power grid data and the power grid control rule set.

[0019] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0020] Acquiring historical power grid data, wherein the historical power grid data is used to characterize power grid fault conditions and power grid operating conditions when power grid system faults occur;

[0021] Building a power grid knowledge graph based on the historical power grid data;

[0022] generating a power grid control rule set according to the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule;

[0023] A power grid control model is trained based on the historical power grid data and the power grid control rule set.

[0024] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the following steps:

[0025] Acquiring historical power grid data, wherein the historical power grid data is used to characterize power grid fault conditions and power grid operating conditions when power grid system faults occur;

[0026] Building a power grid knowledge graph based on the historical power grid data;

[0027] generating a power grid control rule set according to the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule;

[0028] A power grid control model is trained based on the historical power grid data and the power grid control rule set.

[0029] The above-mentioned power grid control model construction method, device, computer equipment, computer-readable storage medium and computer program product can realize the causal reasoning process of power grid system failure by constructing a knowledge graph of historical power grid data that can characterize power grid fault conditions and the power grid operating status when the power grid system fails. Then, based on the power grid knowledge graph, a more accurate power grid control rule set can be generated, and then based on the power grid control rule set and historical power grid data, a power grid control model can be constructed, so that the power grid control model can output corresponding power grid control rules according to the power grid operating status, thereby improving the accuracy of power grid control. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0031] Figure 1 A diagram illustrating an application environment of a method for constructing a power grid control model in one embodiment;

[0032] Figure 2 A schematic flow chart of a method for constructing a power grid control model in one embodiment;

[0033] Figure 3 1. A flowchart illustrating steps for training a power grid control model based on historical power grid data and a set of power grid control rules in one embodiment;

[0034] Figure 4 A structural block diagram of a power grid control model building device in one embodiment;

[0035] Figure 5 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0037] It should be noted that the terms "first", "second", etc. used in this application may be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "including" and "having" used in this application and any variations thereof are intended to cover non-exclusive inclusions. The term "plurality" used in this application refers to two or more. The term "and / or" used in this application refers to one of the solutions or any combination of multiple solutions.

[0038] The power grid control model construction method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store data that server 104 needs to process. The data storage system can be integrated with server 104 or placed on a cloud or other network server. Server 104 obtains historical power grid data, where the historical power grid data is used to characterize power grid fault conditions and the power grid operating status when a power grid system fault occurs. Based on the historical power grid data, a power grid knowledge graph is constructed. Based on the power grid knowledge graph, a power grid control rule set is generated, where the power grid control rule set includes at least one power grid control rule. A power grid control model is trained based on the historical power grid data and the power grid control rule set. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart car devices, projectors, etc. Portable wearable devices can include smart watches, smart bracelets, head-mounted devices, etc. Head-mounted devices can include virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server 104 may be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0039] In an exemplary embodiment, Figure 2 As shown in the figure, a method for constructing a power grid control model is provided, which is applied to Figure 1 The server 104 in the example is used as an example to illustrate the process, including the following steps 202 to 208.

[0040] Step 202: Acquire historical power grid data, wherein the historical power grid data is used to characterize power grid fault conditions and power grid operating states when power grid system faults occur.

[0041] The historical power grid data in step 202 includes at least one of power grid standard documents, historical power grid fault record data, and historical power grid system operation data; the power grid standard documents include at least one of voltage level classification specification data, operation abnormality determination threshold, relay protection requirement configuration data, power quality limit, and accident handling process data; the historical power grid fault record data includes at least one of historical power grid fault cause data (for example, power grid equipment aging, power grid load mutation, etc.), historical power grid fault location data (for example, busbar, switchgear, transformer, etc.), and historical power grid fault condition data (for example, voltage sag, harmonic excess, tripping, etc.); and the historical power grid system operation data includes at least one of the operating time, current, voltage, power factor, load rate, temperature rise, and frequency deviation of various types of power grid equipment.

[0042] Exemplarily, step 202 includes querying historical power grid data from at least one of a power grid dispatching platform, a power grid knowledge base, a station control layer, and a sensor data storage file, wherein the power grid knowledge base may be a knowledge base of a single power grid system or a knowledge base of multiple power grid systems.

[0043] Optionally, after step 202, the method further includes: performing data cleaning on the historical power grid data, wherein the data cleaning method includes but is not limited to removing duplicate data, removing erroneous data, completing missing data, etc.

[0044] Optionally, after data cleaning of the historical power grid data, the above method further includes: standardizing the historical power grid data, wherein the standardization method includes but is not limited to unifying the character encoding format, using the same description form for the same semantic representation, and sentence segmentation of complex sentences.

[0045] Step 204: construct a power grid knowledge graph based on historical power grid data.

[0046] Exemplarily, step 204 includes: extracting entity relationships from historical power grid data to obtain at least one set of entity relationship information; for each set of entity relationship information, constructing a triple structure based on the entity relationship information; fusing the triple structures corresponding to each set of entity relationship information to obtain a power grid knowledge graph, wherein the fusion method can be a splicing fusion method.

[0047] Furthermore, entity relationships are extracted from historical power grid data to obtain at least one group of entity relationship information, including: determining historical power grid system operation data belonging to the same operation time as the same group of historical operation data; generating fault diagnosis results and fault-oriented relationship information corresponding to each group of historical operation data based on power grid standard documents and historical power grid fault record data, and determining each group of historical operation data and its corresponding fault diagnosis results and fault-oriented relationship information as a group of entity relationship information.

[0048] In this way, by determining the historical power grid system operation data belonging to the same operation time as the same group of historical operation data, and subsequently performing corresponding fault diagnosis and fault-oriented relationship identification on each group of historical operation data, the time window alignment of the historical power grid data is achieved, ensuring the accuracy of the determination of entity relationship information.

[0049] As an embodiment, for each set of entity relationship information, a triple structure is constructed based on the entity relationship information, including: each set of triple structures includes a head entity vector, a relationship vector and a tail entity vector, and the fault diagnosis result in each set of entity relationship information is determined as the tail entity vector in each set of triple structures, and the fault-oriented relationship information corresponding to each set of entity relationship information is determined as the relationship vector in each set of triple structures; the historical operation data corresponding to each set of entity relationship information is determined as the head entity vector in each set of triple structures.

[0050] Among them, the relationship vector represents the logical causal relationship between the head entity vector and the tail entity vector, such as triggering, leading to, affecting, etc.

[0051] In this way, the graph representation of the "device-state-event" triple structure is realized, and the knowledge vectorization representation is realized through the embedding algorithm.

[0052] Optionally, after constructing the triple structure according to the entity relationship information, the method further comprises: removing the triple structures whose corresponding triple relevance does not satisfy a preset relevance condition.

[0053] Furthermore, the triple structures whose corresponding triple correlation does not meet the preset correlation conditions are eliminated, including: obtaining the head entity vector and the relationship vector in the triple structure and fusing them to obtain a fusion vector, wherein the fusion method includes a summation fusion method; based on the distance between the fusion vector and the tail entity vector, the triple correlation of the triple structure is evaluated, and the triple structures whose corresponding triple correlation is less than the preset distance threshold are eliminated.

[0054] The preset distance threshold can be set by the user as needed or can be an empirical value, which is not limited here.

[0055] As an embodiment, the triple relevance of the triple structure is evaluated based on the distance between the fusion vector and the tail entity vector, including: determining the square of the L2 norm of the difference between the fusion vector and the tail entity vector as the triple relevance of the triple structure.

[0056] Optionally, the square of the L2 norm of the difference between the fusion vector and the tail entity vector is determined as the triplet correlation of the triplet structure, which can be expressed by the formula:

[0057]

[0058] in, is the triple association of the triple structure, is the head entity vector, is the relationship vector, is the tail entity vector.

[0059] In this way, considering that the constructed triple structure may have weak correlation, the correlation of the triple structure is quantified by taking the square of the L2 norm of the difference between the fusion vector and the tail entity vector, thereby achieving accurate extraction of the triple structure.

[0060] Step 206 : Generate a power grid control rule set based on the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule.

[0061] The grid control rule in step 206 is used to characterize at least one of the control action type, action object, control parameter, and response timing; the control action type includes but is not limited to ratio adjustment, harmonic compensation, and cooling start and stop.

[0062] Exemplarily, step 206 includes: generating control strategy information for solving each tail entity vector in the power grid knowledge graph according to the power grid knowledge graph, and determining the control strategy information of each tail entity vector as a power grid control rule set.

[0063] Step 208: Train a power grid control model based on historical power grid data and a set of power grid control rules.

[0064] Exemplarily, step 208 includes: constructing a training sample based on historical power grid data and a set of power grid control rules; and training a power grid control model based on the training sample.

[0065] Optionally, after training the grid control model based on historical grid data and a set of grid control rules, the above method also includes: obtaining real-time grid data of the target grid system, wherein the real-time grid data is used to characterize the grid operation status; and outputting the target grid control rules of the target grid system based on the real-time grid data through the grid control model.

[0066] The real-time grid data includes at least one of the operating time, current, voltage, power factor, load rate, temperature rise and frequency deviation of various types of grid equipment.

[0067] Optionally, after outputting the target grid control rules of the target grid system based on real-time grid data through the grid control model, the above method also includes: controlling the target grid system to execute the target grid control rules; collecting response information of the target grid system, and updating the grid control model based on the response information.

[0068] The response information is used to characterize at least one of a voltage response condition, a harmonic suppression response condition, and a response delay.

[0069] Furthermore, the power grid control model is updated according to the response information, including: when the response information does not meet the preset response conditions, the power grid control model is updated, and the adjustment method includes adjusting the model parameters of the power grid control model, the control priority of each power grid control rule and the comparative loss parameters corresponding to each group of power grid data sets.

[0070] In this way, by adopting an incremental update method driven by response feedback, the continuous adaptive evolution capability of the power grid control model is achieved, thereby improving the adaptive adjustment capability of the power grid control model.

[0071] In the above-mentioned power grid control model construction method, by constructing a knowledge graph of historical power grid data that can characterize the power grid fault condition and the power grid operating status when the power grid system fails, the causal reasoning process of power grid system failure can be realized. Then, based on the power grid knowledge graph, a more accurate power grid control rule set can be generated, and then based on the power grid control rule set and historical power grid data, a power grid control model is constructed, so that the power grid control model can output corresponding power grid control rules according to the power grid operating status, thereby improving the accuracy of power grid control.

[0072] In an exemplary embodiment, Figure 3 As shown, step 208 includes steps 302 to 304. Among them:

[0073] Step 302: Constructing the control priority of each power grid control rule.

[0074] The control priority in step 302 is used to represent the control order of each power grid control rule. The higher the control priority of the power grid control rule, the earlier the power grid control rule is executed.

[0075] Optionally, as an embodiment, when the control priorities of multiple power grid control rules are the same or the control priority difference between the multiple power grid control rules is less than a preset difference threshold, it is determined that the multiple power grid control rules are executed at the same time.

[0076] The preset difference threshold can be set by the user as needed, or can be an empirical value, or can correspond to the control priority distribution of each power grid control rule. Specifically, the discrete degree of the control priority distribution is positively correlated with the preset difference threshold.

[0077] In this way, when the control priorities of multiple power grid control rules are the same or close, it means that the multiple power grid control rules need to be executed simultaneously to achieve coordinated control of the power grid system.

[0078] Exemplarily, step 302 includes: obtaining grid operation deviation information corresponding to all grid control rules, wherein the grid operation deviation information is used to characterize the degree to which the grid operation deviates from the normal operating state; and constructing the control priority of each grid control rule according to the grid operation deviation information corresponding to all grid control rules.

[0079] The grid operation deviation information is used to characterize at least one of the grid voltage deviation degree and the grid harmonic distortion degree.

[0080] Furthermore, based on the grid operation deviation information corresponding to all grid control rules, the control priority of each grid control rule is constructed respectively, including: for each grid control rule, determining the grid operation deviation information corresponding to the grid control rule, and the deviation ratio in the grid operation deviation information corresponding to all grid control rules; based on the deviation ratio, constructing the control priority of the grid control rule, wherein the deviation ratio is positively correlated with the control priority.

[0081] As one embodiment, the grid operation deviation information includes a grid voltage deviation value and a grid harmonic distortion rate; determining the grid operation deviation information corresponding to the grid control rule, and the deviation ratio in the grid operation deviation information corresponding to all grid control rules, including: fusing the grid voltage deviation value and the grid harmonic distortion rate corresponding to the grid control rule to obtain a grid deviation fusion value corresponding to the grid control rule; fusing the grid deviation fusion values ​​corresponding to all grid control rules to obtain a total grid deviation fusion value; determining the ratio between the grid deviation fusion value and the total grid deviation fusion value as the grid operation deviation information corresponding to the grid control rule, and the deviation ratio in the grid operation deviation information corresponding to all grid control rules.

[0082] Optionally, the ratio between the grid deviation fusion value and the total grid deviation fusion value is determined as the grid operation deviation information corresponding to the grid control rule. The deviation ratio in the grid operation deviation information corresponding to all grid control rules can be expressed as follows:

[0083]

[0084] in, Grid control rules The corresponding grid operation deviation information, the deviation ratio in the grid operation deviation information corresponding to all grid control rules, Grid control rules The corresponding grid voltage deviation value, Grid control rules The corresponding grid harmonic distortion rate.

[0085] In this way, considering that the larger the grid voltage deviation value, the more serious the voltage fluctuation corresponding to the grid control rule, the more urgent the need for grid system control; the larger the grid harmonic distortion rate, the more serious the harmonic pollution corresponding to the grid control rule, the more urgent the need for grid system control; therefore, the grid voltage deviation value and the grid harmonic distortion rate are set to have a positive correlation with the deviation ratio, and the deviation ratio is positively correlated with the control priority. Therefore, it is ensured that the more serious the grid fault corresponding to the grid control rule, the higher the priority of executing the grid control rule, thereby improving the accuracy of the priority setting of the grid control rule.

[0086] As one embodiment, determining the grid voltage deviation value corresponding to the grid control rule includes: obtaining a target voltage reference value, wherein the target voltage reference value can be set by the user as needed or determined in the grid standard document, and is not limited here; determining the deviation between the grid voltage corresponding to the grid control rule and the target voltage reference value as the grid voltage deviation value corresponding to the grid control rule.

[0087] As one embodiment, determining the grid harmonic distortion rate corresponding to the grid control rule includes: obtaining a target distortion rate reference value, wherein the target distortion rate reference value can be set by the user as needed or determined in the grid standard document, and is not limited here; determining the ratio between the actual harmonic distortion rate corresponding to the grid control rule and the target distortion rate reference value as the grid harmonic distortion rate corresponding to the grid control rule.

[0088] Optionally, when the actual harmonic distortion rate corresponding to the grid control rule is greater than the target distortion rate reference value, 1 is determined as the grid harmonic distortion rate corresponding to the grid control rule.

[0089] Step 304 : training a power grid control model based on the control priority of each power grid control rule, historical power grid data, and a set of power grid control rules.

[0090] Exemplarily, step 304 includes: screening multiple groups of power grid data sets belonging to the same operating conditions from historical power grid data; determining the comparative loss parameters corresponding to each group of power grid data sets; for each group of power grid data sets, performing feature extraction on the power grid data sets according to the comparative loss parameters corresponding to the power grid data sets to obtain power grid characteristics; training a power grid control model according to the power grid characteristics corresponding to each group of power grid data sets, the control priority of each power grid control rule, and the power grid control rule set.

[0091] It is understandable that historical power grid data can be in different modalities, including but not limited to textual modalities (e.g., data, text, records, etc.), waveform modalities (e.g., current, voltage waveforms, etc.), and image modalities (e.g., thermal images of power grid equipment and monitoring images, etc.).

[0092] In this way, by screening multiple groups of power grid data sets belonging to the same operating conditions from historical power grid data, historical power grid data under different modes and belonging to the same operating conditions can belong to the same group of power grid data sets, ensuring input generalization.

[0093] Furthermore, the comparative loss parameters corresponding to each group of power grid data sets are determined, including: for each group of power grid data sets, a balance parameter is determined based on the similarity between the historical power grid data included in the power grid data set, wherein the similarity is positively correlated with the balance parameter; through the balance parameter, the historical power grid data included in the power grid data set are corrected and balanced to obtain corrected power grid data; based on the corrected power grid data, the comparative loss parameters corresponding to the power grid data set are determined, and the corrected power grid data and the comparative loss parameters are inversely correlated.

[0094] As one embodiment, the historical power grid data included in the power grid data set are corrected and balanced through balancing parameters to obtain corrected power grid data, including: taking any two historical power grid data as a power grid data pair, and correcting each power grid data pair through balancing parameters to obtain a corrected power grid data pair; for each power grid data pair, the ratio between the corrected power grid data pair corresponding to the power grid data pair and the fused data pair of the corrected power grid data pairs corresponding to all power grid data pairs is determined as the corrected power grid data.

[0095] Furthermore, each power grid data pair is corrected by using the balance parameter to obtain a corrected power grid data pair, including: fusing the two historical power grid data contained in the power grid data pair to obtain historical power grid fusion data, wherein the fusion method includes a quadrature fusion method; determining the ratio between the historical power grid fusion data and the balance parameter as a balance index; performing a power operation with a natural constant as the base and the balance index as the exponent to obtain a corrected power grid data pair.

[0096] As one embodiment, based on the corrected power grid data, the comparative loss parameter corresponding to the power grid data set is determined, including: performing a logarithmic operation with a natural constant as the base and the corrected power grid data as the power to obtain a logarithmic result, and determining the opposite of the logarithmic result as the comparative loss parameter corresponding to the power grid data set.

[0097] Optionally, a logarithmic operation is performed with the natural constant as the base and the modified power grid data as the power to obtain a logarithmic result. The opposite of the logarithmic result is determined as the comparative loss parameter corresponding to the power grid data set. The formula can be expressed as follows:

[0098]

[0099] in, is the contrast loss parameter corresponding to the power grid data set, For power grid data alignment, it belongs to modal Historical grid data, For power grid data alignment, it belongs to modal Historical grid data, For power grid data alignment, it belongs to modal Historical grid data, is the balance parameter.

[0100] In this way, by constructing the power grid features based on the comparison of loss parameters, the features of the power grid data of different modes belonging to the same operating state can be brought closer in the feature space, while the features of the power grid data not belonging to the same operating state are pulled apart in the feature space, thereby ensuring the consistency of cross-modal feature representation and improving the generalization of the power grid control model.

[0101] In this embodiment, taking into account that different power grid control rules may have a control sequence, the control sequence of the power grid control rules is defined by constructing the control priority of each power grid control rule, and then a power grid control model is constructed based on the control priority of each power grid control rule, historical power grid data and a set of power grid control rules, so that the constructed power grid control model can output control rules with a control sequence, thereby improving the accuracy of power grid control.

[0102] As a detailed embodiment, historical power grid data is obtained, wherein the historical power grid data is used to characterize power grid fault conditions and power grid operating conditions when power grid system faults occur; entity relationship extraction is performed on the historical power grid data to obtain at least one set of entity relationship information; for each set of entity relationship information, a triple structure is constructed based on the entity relationship information; the triple structures corresponding to each set of entity relationship information are fused to obtain a power grid knowledge graph; a set of power grid control rules is generated based on the power grid knowledge graph, wherein the set of power grid control rules includes at least one power grid control rule; power grid operation deviation information corresponding to all power grid control rules is obtained, wherein the power grid operation deviation information is used to characterize the degree to which power grid operation deviates from normal operating conditions; control priorities of each power grid control rule are respectively constructed based on the power grid operation deviation information corresponding to all power grid control rules; multiple sets of power grid data sets belonging to the same operating condition are selected from the historical power grid data; a comparative loss parameter corresponding to each set of power grid data sets is determined; for each set of power grid data sets, feature extraction is performed on the power grid data sets based on the comparative loss parameter corresponding to the power grid data sets to obtain power grid features; and a power grid control model is trained based on the power grid features corresponding to each set of power grid data sets, the control priorities of each power grid control rule, and the set of power grid control rules.

[0103] Furthermore, real-time grid data of the target grid system is obtained, wherein the real-time grid data is used to characterize the grid operation status; and target grid control rules of the target grid system are output according to the real-time grid data through the grid control model.

[0104] In this way, by constructing a knowledge graph based on historical grid data that can characterize grid fault conditions and the grid operating state when a grid system fails, a causal reasoning process for grid system failures can be implemented. Based on the grid knowledge graph, a more accurate set of grid control rules can be generated. Based on the grid control rule set and historical grid data, a grid control model can be constructed. This allows the grid control model to output corresponding grid control rules according to the grid operating state, thereby improving grid control accuracy. Furthermore, considering that different grid control rules have a control priority order, the control order of each grid control rule is defined by constructing a control priority for each grid control rule. Based on the control priority of each grid control rule, historical grid data, and the grid control rule set, a grid control model can be constructed. This allows the constructed grid control model to output control rules that have a control priority order, thereby improving grid control accuracy.

[0105] It should be understood that, although the various steps in the flowcharts involved in the above embodiments are displayed in sequence according to the instructions of the arrows, these steps are not necessarily performed in sequence in the order indicated by the arrows. Unless clearly stated herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of the steps or stages in other steps or other steps. It is understandable that the various steps in different embodiments can be freely combined as needed, and the various non-contradictory schemes formed by the combination all fall within the scope of protection of this application.

[0106] Based on the same inventive concept, embodiments of the present application also provide a power grid control model construction device for implementing the power grid control model construction method involved above. The implementation solution provided by this device is similar to the implementation solution described in the above method. Therefore, the specific limitations of one or more power grid control model construction device embodiments provided below can be found in the above-mentioned limitations of the power grid control model construction method and will not be repeated here.

[0107] In an exemplary embodiment, Figure 4 As shown, a power grid control model construction device 400 is provided, comprising: an acquisition module 402, a construction module 404, a generation module 406 and a training module 408, wherein:

[0108] The acquisition module 402 is used to acquire historical power grid data, wherein the historical power grid data is used to characterize the power grid fault condition and the power grid operation state when the power grid system fails.

[0109] The construction module 404 is used to construct a power grid knowledge graph based on historical power grid data.

[0110] The generation module 406 is configured to generate a power grid control rule set according to the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule.

[0111] The training module 408 is used to train a power grid control model based on historical power grid data and a set of power grid control rules.

[0112] In one embodiment, the training module 408 is further configured to construct a control priority for each grid control rule; and train a grid control model based on the control priority of each grid control rule, historical grid data, and a grid control rule set.

[0113] In one embodiment, the training module 408 is also used to obtain grid operation deviation information corresponding to all grid control rules, wherein the grid operation deviation information is used to characterize the degree to which the grid operation deviates from the normal operating state; and based on the grid operation deviation information corresponding to all grid control rules, the control priority of each grid control rule is constructed respectively.

[0114] In one embodiment, the training module 408 is also used to screen multiple groups of power grid data sets belonging to the same operating conditions from historical power grid data; determine the comparative loss parameters corresponding to each group of power grid data sets; for each group of power grid data sets, perform feature extraction on the power grid data sets according to the comparative loss parameters corresponding to the power grid data sets to obtain power grid features; and train a power grid control model based on the power grid features corresponding to each group of power grid data sets, the control priority of each power grid control rule, and the power grid control rule set.

[0115] In one embodiment, after training a grid control model based on historical grid data and a set of grid control rules, the above-mentioned device also includes: an output module for obtaining real-time grid data of the target grid system, wherein the real-time grid data is used to characterize the operating status of the grid; through the grid control model, the target grid control rules of the target grid system are output based on the real-time grid data.

[0116] In one embodiment, the construction module 404 is also used to extract entity relationships from historical power grid data to obtain at least one set of entity relationship information; for each set of entity relationship information, a triple structure is constructed based on the entity relationship information; and the triple structures corresponding to each set of entity relationship information are merged to obtain a power grid knowledge graph.

[0117] Each module in the aforementioned power grid control model construction device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0118] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 5As shown. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless means, and the wireless means can be implemented via Wi-Fi, mobile cellular networks, near-field communication (NFC), or other technologies. When executed by the processor, the computer program implements a method for constructing a power grid control model. The display unit of the computer device is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0119] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0120] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0121] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0122] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0123] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0124] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0125] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0126] The above embodiments merely illustrate several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for constructing a power grid control model, characterized in that: The method comprises: Acquiring historical power grid data, wherein the historical power grid data is used to characterize power grid fault conditions and power grid operating conditions when power grid system faults occur; Building a power grid knowledge graph based on the historical power grid data; generating a power grid control rule set according to the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule; A power grid control model is trained based on the historical power grid data and the power grid control rule set.

2. The method according to claim 1, characterized in that The training of a power grid control model based on the historical power grid data and the power grid control rule set includes: Establishing control priorities for each of the power grid control rules; A power grid control model is obtained by training according to the control priority of each power grid control rule, the historical power grid data and the power grid control rule set.

3. The method according to claim 2, characterized in that The control priority of each power grid control rule is constructed, including: Obtaining grid operation deviation information corresponding to all the grid control rules, wherein the grid operation deviation information is used to characterize the degree to which the grid operation deviates from a normal operating state; According to the power grid operation deviation information corresponding to all the power grid control rules, the control priority of each power grid control rule is respectively established.

4. The method according to claim 2, characterized in that The training of a power grid control model based on the control priority of each power grid control rule, the historical power grid data, and the power grid control rule set includes: Filtering multiple groups of power grid data sets belonging to the same operating status from the historical power grid data; Determining a comparative loss parameter corresponding to each set of the power grid data sets; For each group of the power grid data sets, extracting features from the power grid data sets according to the contrast loss parameters corresponding to the power grid data sets to obtain power grid features; A power grid control model is obtained by training according to the power grid characteristics corresponding to each group of the power grid data sets, the control priority of each power grid control rule, and the power grid control rule set.

5. The method according to any one of claims 1 to 4, characterized in that After training the grid control model based on the historical grid data and the grid control rule set, the method further includes: Acquiring real-time grid data of a target power grid system, wherein the real-time grid data is used to characterize the operating status of the power grid; The target grid control rules of the target grid system are outputted according to the real-time grid data through the grid control model.

6. The method according to claim 1, characterized in that The step of constructing a power grid knowledge graph based on the historical power grid data includes: Extracting entity relationships from the historical power grid data to obtain at least one set of entity relationship information; For each set of entity relationship information, construct a triple structure according to the entity relationship information; The triple structures corresponding to each group of entity relationship information are fused to obtain a power grid knowledge graph.

7. A power grid control model construction device, characterized in that: The device comprises: An acquisition module is used to acquire historical power grid data, wherein the historical power grid data is used to characterize the power grid fault condition and the power grid operation state when the power grid system fails; A construction module, configured to construct a power grid knowledge graph based on the historical power grid data; A generating module, configured to generate a power grid control rule set according to the power grid knowledge graph, wherein the power grid control rule set includes at least one power grid control rule; A training module is used to train a power grid control model based on the historical power grid data and the power grid control rule set.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images