Flow data differentiation analysis method applied to transformer substation

By building a digital twin model of the substation and a deep learning differentiated evaluation model, the accuracy of differentiated analysis of flow data is improved, the problem of inaccurate analysis of substation flow data in the existing technology is solved, and timely monitoring of abnormal states of substations is achieved and prevention.

CN119066577AActive Publication Date: 2024-12-03SHIZUISHAN POWER SUPPLY COMPANY OF STATE GRID NINGXIA ELECTRIC POWER +1
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202411123376.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-12-03
Estimated Expiration
2044-08-15

AI Technical Summary

Technical Problem

The accuracy of the differentiated analysis of existing substation flow data is poor, resulting in inaccurate monitoring of abnormal states, affecting the stability and safety of the power grid.

Method used

Build a digital twin model for the operation process of the substation, combine it with the deep learning differentiated evaluation model, extract the time and space characteristics of the traffic data, and generate differentiated analysis results, thereby improving the accuracy of differentiated analysis of the traffic data.

Benefits of technology

Through real-time dynamic online monitoring, abnormal states of the substation can be discovered in a timely manner, damage and safety risks caused by aggravation of faults and improve grid stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066577A_ABST
    Figure CN119066577A_ABST
Patent Text Reader

Abstract

The invention provides a flow data differentiation analysis method applied to a transformer substation, and relates to the technical field of electric power and artificial intelligence, and the method comprises the steps: building a digital twinborn model of the operation process of the transformer substation in advance; wherein the digital twinborn model comprises each flow monitoring point position; acquiring flow data acquired by flow monitoring points which are arranged in the transformer substation and cover each piece of power equipment, and importing the flow data into the digital twinborn model; generating power equipment abnormal information representing that the power equipment monitored by the flow monitoring point position is abnormal according to the flow data; obtaining a target flow monitoring point position for generating the abnormal information of the power equipment from the digital twin model, and obtaining target flow data corresponding to the target flow monitoring point position; and performing differentiation analysis on the target traffic data by using a differentiation evaluation model pre-trained based on deep learning to obtain a differentiation analysis result. According to the scheme, the accuracy of differential analysis of the flow data of the transformer substation can be improved, and the accuracy of abnormal state monitoring of the transformer substation is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of electric power and artificial intelligence technology, and in particular to a flow data differentiation analysis method applied to a substation. Background Art

[0002] Substations are places where voltage is changed. There are various power equipment in substations, such as transformers, reactors, switchgear, voltage transformers, current transformers, etc. For example, power equipment such as transformers will not stop working immediately at the beginning of some faults. They will not be able to operate normally until the faults gradually worsen to a certain extent. When the faults of power equipment worsen, it may cause multiple operating systems to be paralyzed, resulting in power supply failures and affecting people's lives. If the existence of the fault can be discovered at the early stage of the fault, the damage to the power equipment caused by the worsening of the fault can be avoided. Therefore, monitoring the substation can timely detect abnormal conditions in the substation, so that timely taking corresponding countermeasures is of great significance for preventing potential safety risks and improving the stability of the power grid.

[0003] The current main monitoring method is to monitor the abnormal conditions of substations based on the flow data of substations. For example, threshold monitoring technology is used to analyze the flow data of substations, and then the abnormal state of substations is determined based on the differential analysis of flow data. However, the differentiation accuracy of existing substation flow data is poor, which reduces the accuracy of abnormal state analysis of substations. Summary of the invention

[0004] In view of this, in response to the above shortcomings, it is necessary to propose a flow data differential analysis method applied to substations to improve the accuracy of differential analysis of substation flow data, and thus improve the accuracy of abnormal state monitoring of substations.

[0005] The present invention provides a flow data differentiation analysis method applied to a substation, comprising:

[0006] Pre-constructing a digital twin model of the substation operation process; wherein the digital twin model includes various flow monitoring points; wherein each flow monitoring point is used to monitor at least one type of power data of the power equipment;

[0007] Obtaining flow data collected by flow monitoring points covering various power equipment set up in the substation, and importing the data into the digital twin model;

[0008] Generate power equipment abnormality information indicating that the power equipment monitored by the flow monitoring point is abnormal according to the flow data;

[0009] Acquire the target flow monitoring point for generating the abnormal information of the electric power equipment from the digital twin model;

[0010] Acquire target flow data corresponding to the target flow monitoring point from the flow data;

[0011] The target traffic data is subjected to differential analysis using a differentiated evaluation model pre-trained based on deep learning to obtain a differentiated analysis result.

[0012] Preferably, the pre-building of the digital twin model of the substation operation process includes:

[0013] Construct a three-dimensional model corresponding to the physical entities of each power equipment during the current substation operation;

[0014] Obtain the communication link relationship between each flow monitoring point during the operation of the substation;

[0015] The directed topology graph is constructed by taking the three-dimensional model corresponding to each power equipment as the node of the directed topology graph and taking the communication link relationship between each flow monitoring point as the connection relationship between the nodes;

[0016] The flow data collected at each flow monitoring point is matched with the three-dimensional model corresponding to each node of the directed topological graph to obtain the digital twin model.

[0017] Preferably, the acquiring of flow data collected by flow monitoring points covering various power equipment arranged in the substation includes:

[0018] Obtaining process unit characteristics of each power equipment in the current substation; wherein the process unit characteristics include functional characteristics, working parameter characteristics, structural characteristics and safety and reliability characteristics;

[0019] Extracting process information according to the structural features of the process unit features; wherein the process information includes the assembly sequence and assembly relationship of the power equipment;

[0020] Determine the flow monitoring points of each power equipment according to the process information;

[0021] Determine the type of power data monitoring at each flow monitoring point according to the functional characteristics in the process unit characteristics;

[0022] Obtain the flow data of each power data monitoring type corresponding to each flow monitoring point in real time.

[0023] Preferably, generating the power equipment abnormality information indicating that the power equipment monitored by the flow monitoring point is abnormal according to the flow data includes:

[0024] Preset the flow threshold, flow deviation threshold and abnormal time threshold corresponding to each power data monitoring type at each flow monitoring point;

[0025] For the power data monitoring type of each flow monitoring point, the difference between the flow data of the power data monitoring type and the corresponding flow threshold is calculated to obtain the deviation flow;

[0026] Compare the deviation flow with the flow deviation threshold, and when the deviation flow is greater than the flow deviation threshold, accumulate the time during which the deviation flow corresponding to the current power data monitoring type is greater than the flow deviation threshold;

[0027] When the accumulated time is greater than the abnormal time threshold, the corresponding power data monitoring type is marked as an abnormal type, and power equipment abnormal information corresponding to the power equipment at the flow monitoring point where the abnormal type is located is generated.

[0028] Preferably, the differential analysis of the target traffic data using a differential evaluation model pre-trained based on deep learning includes:

[0029] Based on deep learning, a differentiated evaluation model is pre-trained for differentiated analysis of substation flow data;

[0030] Extracting the time and space characteristics of the target flow data respectively to obtain a flow time and space characteristic sequence;

[0031] The flow spatiotemporal feature sequence is input into the differential evaluation model, and the differential analysis result of the substation flow data is output.

[0032] Preferably, said extracting the time and space features of said target traffic data respectively comprises:

[0033] Pre-training LTSM neural network for predicting flow change trends of flow data;

[0034] Inputting the target flow data into the LTSM neural network to obtain the target flow change trend characteristics corresponding to the target flow monitoring point in the current monitoring period;

[0035] Importing the flow change trend characteristics into the nodes corresponding to the pre-trained substation directed topology graph; wherein each flow monitoring point in the substation directed topology graph constitutes a node of the topology graph;

[0036] Taking the flow monitoring point corresponding to the target flow data as the target node, determining the coupling weight between the target node and each adjacent node in the directed topological graph of the substation;

[0037] The feature aggregation representation after each adjacent node acts on the target node is determined according to the coupling weight to obtain the traffic spatiotemporal feature sequence.

[0038] Preferably, the determining of the coupling weights between the target node and each adjacent node in the directed topology graph of the substation includes:

[0039] Obtaining a first historical flow time series sequence collected by the target node and its adjacent nodes when the target node and its adjacent nodes in the directed topology graph of the substation do not generate power equipment abnormal information within a historical monitoring period; and obtaining a second historical flow time series sequence collected by the adjacent nodes when the target node in the directed topology graph of the substation generates power equipment abnormal information within a historical monitoring period;

[0040] Calculating the first average flow information of the power data monitoring type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence; and calculating the second average flow information of the power data monitoring type corresponding to the adjacent nodes according to the second historical flow time series sequence;

[0041] Determine a first flow amplitude time series sequence of the power data type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence and the first average flow information; and determine a second flow amplitude time series sequence of the power data type corresponding to the adjacent nodes according to the second historical flow time series sequence and the second average flow information;

[0042] Calculating, according to the first flow amplitude time series sequence and the second flow amplitude time series sequence, a differentiation coefficient of the power data monitoring type corresponding to the adjacent nodes;

[0043] The coupling weight is determined according to the differentiation coefficient.

[0044] Preferably,

[0045] When determining the first flow amplitude time series sequence of the power data type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence and the first average flow information, or determining the corresponding second flow amplitude time series sequence according to the second historical flow time series sequence and the second average flow information, the following calculation formula is used to calculate:

[0046]

[0047] Among them, H i represents the traffic amplitude time series of node i, X iL represents the flow amplitude of node i at the Lth moment, N represents the total number of moments, Q iL represents the traffic of node i at the Lth moment, Q iL A represents the average flow rate of node i, and δ is the conversion coefficient;

[0048] and / or,

[0049] When calculating the differentiation coefficient of the power data monitoring type corresponding to the adjacent nodes according to the first flow amplitude time series sequence and the second flow amplitude time series sequence, the following calculation formula is used to obtain:

[0050]

[0051] Among them, y(ij) represents the difference coefficient between node i and node j, X iL represents the traffic amplitude of node i at the Lth moment obtained from the first traffic amplitude time series, X jL represents the flow amplitude of node j at the Lth moment obtained from the first flow amplitude time series, Y iL represents the flow amplitude of node i at the Lth moment obtained from the second flow amplitude time series, Y jL represents the flow amplitude of node j at the Lth moment obtained from the second flow amplitude time series;

[0052] and / or,

[0053] When determining the coupling weight according to the differentiation coefficient, the following calculation formula is used:

[0054] e ij =θy(ij)

[0055] Among them, e ij represents the coupling weight between node i and node j, y(ij) represents the differentiation coefficient between node i and node j, and θ represents the conversion coefficient.

[0056] Preferably, determining the feature aggregation representation of each adjacent node after acting on the target node according to the coupling weight includes obtaining it by using the following calculation formula:

[0057] F r * =ReLu(pF r +∑ z∈ur y(rz)pF z )

[0058] Among them, F r * represents the feature aggregation representation of node r, ReLu represents the activation function, p represents the parameter matrix of feature transformation, y(rz) represents the coupling weight of node r to node z, u r represents the set of adjacent nodes of node r, F r and F z They are the target traffic change trend characteristics of node r and node z respectively.

[0059] Preferably, after obtaining the differential analysis results of the substation flow data, the method further includes:

[0060] Obtaining an estimated flow time series sequence of the power data monitoring type of other flow monitoring points adjacent to the target flow monitoring point;

[0061] Compare the estimated flow time series of each adjacent flow monitoring point with the corresponding flow threshold, flow deviation threshold and abnormal time threshold;

[0062] Determine whether each adjacent flow monitoring point has a potential abnormal type and an estimated time stamp of the potential abnormal type according to the comparison result;

[0063] According to the adjacent traffic monitoring points and estimated timestamps that generate the potential anomaly type, corresponding anomaly prevention measures are generated.

[0064] It can be seen from the above technical scheme that in the fault monitoring method of the substation power equipment provided by the present invention, firstly, a digital twin model of the substation operation process is pre-constructed, and then the flow data collected by the flow monitoring points of each power equipment in the substation is obtained and imported into the digital twin model. Further, according to the flow data, abnormal information of the power equipment representing the abnormality of the power equipment at the flow monitoring point is generated, and displayed through the digital twin model, so that the target flow monitoring point for generating abnormal information of the power equipment can be further determined from the digital twin model, and then the target flow data corresponding to the target flow monitoring point can be obtained from the flow data. Further, the target flow data is subjected to differential analysis and processing using a pre-trained differential evaluation model, and the differential analysis results of the substation flow data can be obtained, and then the abnormal state of the substation can be analyzed. It can be seen that the digital twin model of the substation operation process constructed by this scheme can simulate the operation process of the substation in real time, thereby realizing dynamic online monitoring of each device in the substation. Moreover, a differentiated evaluation model based on deep learning is constructed on the basis of the digital twin model, which can dynamically and differentially analyze the flow data of the substation in real time based on the flow data collected at each flow monitoring point. In this way, the accuracy of the differentiated analysis of the flow data of the substation can be improved through real-time dynamic online analysis, and then it can be analyzed in time whether the substation is in an abnormal state. Furthermore, effective measures can be taken for the abnormal state of the substation before the fault matures, avoiding more serious accidents and losses after the early fault matures. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A flow chart of a flow data differential analysis method applied to a substation provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0067] like Figure 1 As shown, the present invention provides a flow data differentiation analysis method applied to a substation, and the method may include the following steps:

[0068] Step 101: pre-build a digital twin model of the substation operation process; wherein the digital twin model includes various flow monitoring points; wherein each flow monitoring point is used to monitor at least one type of power data of the power equipment;

[0069] Step 102: Obtain flow data collected by flow monitoring points covering various power equipment set in the substation, and import them into the digital twin model;

[0070] Step 103: generating power equipment abnormality information indicating that the power equipment monitored by the flow monitoring point is abnormal according to the flow data;

[0071] Step 104: Obtaining target flow monitoring points for generating abnormal information of power equipment from the digital twin model;

[0072] Step 105: Obtain target flow data corresponding to the target flow monitoring point from the flow data;

[0073] Step 106: Perform differential analysis on the target traffic data using a differentiated evaluation model pre-trained based on deep learning to obtain a differentiated analysis result.

[0074] In this embodiment, by constructing a digital twin model of the substation operation process, the operation process of the substation can be simulated in real time, thereby realizing dynamic online monitoring of each device in the substation. Moreover, a differentiated evaluation model is constructed based on deep learning on the basis of the digital twin model, which can perform dynamic differentiated analysis of the substation flow data in real time based on the flow data collected from each flow monitoring point. In this way, the accuracy of the differentiated analysis of the substation flow data can be improved through real-time dynamic online analysis, and then it can be analyzed in time whether the substation is in an abnormal state. Furthermore, effective measures can be taken for the abnormal state of the substation before the fault matures, so as to avoid more serious accidents and losses after the early fault matures.

[0075] For step 101, a digital twin model of the substation operation process is pre-built;

[0076] In this embodiment, it is considered to realize online monitoring of substation power equipment by constructing a digital twin model of the substation operation process, and at the same time, the flow data collected by each flow monitoring point is combined to perform differential analysis on the substation flow data, and then monitor the abnormal state of the substation. The constructed digital twin model includes various flow monitoring points; each flow monitoring point is used to monitor at least one type of power data of the power equipment. Specifically, when constructing a digital twin model, it can be achieved in the following ways:

[0077] Construct a three-dimensional model corresponding to the physical entities of each power equipment during the current substation operation;

[0078] Obtain the communication link relationship between each flow monitoring point during the operation of the substation;

[0079] The directed topology graph is constructed by taking the three-dimensional model corresponding to each power equipment as the node of the directed topology graph and taking the communication link relationship between each flow monitoring point as the connection relationship between the nodes;

[0080] The flow data collected at each flow monitoring point is matched with the three-dimensional model corresponding to each node of the directed topology graph to obtain a digital twin model.

[0081] In this embodiment, it is considered to obtain the physical entity parameters of each power equipment during the operation of the current substation, and then the physical entity of each power equipment is simulated and modeled to obtain a three-dimensional model of each power equipment entity. Further, the communication link connection relationship between each flow monitoring point during the operation of the current substation is obtained, and the three-dimensional model corresponding to each power equipment is used as the node of the directed topology graph, and the communication link relationship between each flow monitoring point is used as the connection relationship between the nodes to construct a directed topology graph. Finally, the flow data collected by each flow monitoring point is obtained, and the twin data is generated according to the flow, and the twin data is matched with the three-dimensional model corresponding to each node of the directed topology graph to generate a digital twin model. In this way, the constructed digital twin model can fully reflect the operation process of the actual substation, so that the online monitoring of the substation operation process can be realized based on the digital twin model, and at the same time, the differential analysis of the substation flow data can be realized on the basis of the digital twin model, and then the abnormal state of the substation can be monitored in real time, avoiding the failure of the substation to cause more serious losses.

[0082] For step 102, the flow data collected by the flow monitoring points covering each power equipment set in the substation is obtained and imported into the digital twin model.

[0083] In this step, the flow data collected at each flow monitoring point is imported into the digital twin model so that the digital twin model can display, store and analyze the data. Specifically, step 102 can be implemented as follows:

[0084] Obtaining process unit characteristics of each power equipment in the current substation; wherein the process unit characteristics include functional characteristics, working parameter characteristics, structural characteristics and safety and reliability characteristics;

[0085] Extracting process information based on the structural features of the process unit features; wherein the process information includes the assembly sequence and assembly relationship of the power equipment;

[0086] Determine the flow monitoring points of each power equipment based on process information;

[0087] According to the functional characteristics in the process unit characteristics, determine the type of power data monitoring at each flow monitoring point;

[0088] Obtain the flow data of each power data monitoring type corresponding to each flow monitoring point in real time.

[0089] In this embodiment, the power equipment included in the substation may include different types of power equipment such as transformers, circuit breakers, current transformers, voltage transformers, and grounding devices. The process unit characteristics of each power equipment in the substation may include functional characteristics, operating parameter characteristics, structural characteristics, and safety and reliability characteristics, which determine the application and performance of the power equipment in the power system.

[0090] Specifically, in terms of functional characteristics, each power equipment has its own specific functions, such as transformers for voltage conversion, circuit breakers for controlling the on and off status of circuits, current transformers and voltage transformers for monitoring current and voltage data, etc. These functions determine the role and status of power equipment in the power system.

[0091] Regarding working parameter characteristics, different types of power equipment have different working parameters, such as the rated capacity, rated voltage, short-circuit impedance of the transformer, the rated current, rated short-circuit current withstand capacity of the circuit breaker, etc. These working parameter characteristics determine the scope of use and performance of the power equipment in the power system.

[0092] As for structural characteristics, each type of electrical equipment has its own specific structural characteristics, such as the oil tank, winding, cooling system of the transformer, the contacts, spring mechanism, arc extinguishing chamber of the circuit breaker, etc. These structural characteristics determine the manufacturing process and maintenance methods of the equipment.

[0093] As for the safety and reliability characteristics, the safety and reliability of power equipment is a crucial point in its process unit characteristics. This includes the characteristics of the equipment's insulation performance, thermal stability, mechanical strength, etc., to ensure that the equipment can operate safely and stably under various working conditions.

[0094] For process information, it mainly represents the assembly sequence and assembly relationship of power equipment. In a substation, the assembly sequence and assembly relationship of each power equipment are usually determined according to the structure and functional requirements of the power system to ensure the safe and stable operation of the power system. For example, for circuit breakers, circuit breakers are usually installed between the power input terminal and each circuit to control the on and off state of the circuit. Circuit breakers are usually connected to transformers, cables and other equipment to achieve precise control and protection of the circuit.

[0095] For example, current transformers and voltage transformers are usually connected to circuit breakers or transformers to monitor current and voltage data and transmit the data to monitoring systems. They are usually installed at key locations in the power system to ensure accurate monitoring of the operating status of the power system.

[0096] Another example is the grounding device, which is used to ground the grounding point of the system to ensure the safe operation of the system. The grounding device is usually installed near the transformer or circuit breaker and connected to the metal casing of the power equipment.

[0097] Another example is control and monitoring equipment, which includes monitoring systems, protection devices, telesignaling and remote control devices, etc., which are used to monitor and control the operating status of the power system. These devices are usually installed in the control room or operation and maintenance center and connected to various power equipment to achieve centralized monitoring and control of the entire power system.

[0098] For the power data monitoring type, it can include but is not limited to electrical parameters such as rated voltage, rated current, rated capacity, operating parameters such as temperature, pressure, humidity, and power quality parameters such as voltage fluctuation, frequency fluctuation and harmonic content.

[0099] For example, for transformers, the corresponding power data monitoring types include but are not limited to oil temperature monitoring, oil level monitoring, transformer winding temperature monitoring, partial discharge monitoring and humidity monitoring, etc. By monitoring the transformer oil temperature, it is ensured that the internal temperature of the transformer is within the safe range to prevent the oil temperature from being too high and causing equipment damage; by monitoring the oil level in the transformer oil tank, it is ensured that the amount of oil required for the normal operation of the transformer is ensured to prevent the insulation performance of the equipment from being affected by insufficient oil level; by monitoring the temperature of the transformer winding, it is ensured that the winding will not be damaged due to overheating; by monitoring the partial discharge that may occur inside the transformer, such as voltage fluctuations, frequency fluctuations, harmonic content and other power quality parameters, it is possible to prevent internal insulation failures of the transformer; by monitoring the humidity of the surrounding environment of the transformer, it is possible to ensure that the insulation performance of the transformer is not affected by the humid environment.

[0100] For step 103, power equipment abnormality information is generated based on the flow data, indicating that the power equipment monitored by the flow monitoring point is abnormal.

[0101] In this step, it is determined whether abnormal flow data continues to exist in the flow data corresponding to each power data monitoring type, and based on this, power equipment abnormality information indicating that the corresponding power equipment has abnormality is generated.

[0102] Specifically, step 103 can be implemented in the following manner:

[0103] Preset the flow threshold, flow deviation threshold and abnormal time threshold corresponding to each power data monitoring type at each flow monitoring point;

[0104] For the power data monitoring type of each flow monitoring point, the difference between the flow data of the power data monitoring type and the corresponding flow threshold is calculated to obtain the deviation flow;

[0105] The deviation flow is compared with the flow deviation threshold, and when the deviation flow is greater than the flow deviation threshold, the time during which the deviation flow corresponding to the current power data monitoring type is greater than the flow deviation threshold is accumulated;

[0106] When the accumulated time is greater than the abnormal time threshold, the corresponding power data monitoring type is marked as an abnormal type, and power equipment abnormal information of the power equipment at the flow monitoring point corresponding to the abnormal type is generated.

[0107] In this embodiment, the flow threshold, flow deviation threshold, and abnormal time threshold corresponding to the power data monitoring type of each flow monitoring point are first preset. Then, the flow data corresponding to the power data type collected at each flow monitoring point is compared with the corresponding flow threshold, and the deviation flow between the flow data corresponding to the power data monitoring type and the corresponding flow threshold can be obtained by calculating the difference. Further, by judging the size relationship between the deviation flow corresponding to each power data monitoring type and the flow deviation threshold, if the deviation flow corresponding to a certain power data monitoring type is greater than the deviation flow threshold, and the accumulated time is greater than the preset abnormal time threshold, it means that the power equipment corresponding to the current power data monitoring type is abnormal, and then the power equipment abnormal information corresponding to the power equipment is generated.

[0108] For steps 104 and 105, the target flow monitoring points for generating abnormal information of the power equipment are obtained from the digital twin model, and the target flow data corresponding to the target flow monitoring points are obtained from the flow data.

[0109] After generating the abnormal information of the power equipment, this step can query the target flow monitoring point corresponding to the generation of the abnormal information of the power equipment through the digital twin model, and then obtain the target flow data collected by the target flow monitoring point from the flow data collected by the target flow monitoring point, that is, obtain the flow data of the power equipment in the abnormal state.

[0110] For step 106, a differentiated analysis is performed on the target traffic data using a differentiated evaluation model pre-trained based on deep learning to obtain a differentiated analysis result.

[0111] This step considers building a differentiated evaluation model based on deep learning on the basis of the digital twin model, and using the differentiated evaluation model to perform differentiated analysis on the substation flow data, and then monitor the abnormal state of the substation in real time. Specifically, step 106 can be implemented by the following steps:

[0112] S61: Based on deep learning, a differentiated evaluation model is pre-trained for differentiated analysis of substation flow data;

[0113] S62: extracting the time and space characteristics of the target flow data respectively to obtain a flow time and space characteristic sequence;

[0114] S63: Input the flow spatiotemporal feature sequence into the differentiated evaluation model, and output the differentiated analysis results of the substation flow data.

[0115] For step S61, when pre-training a differentiated evaluation model for differential analysis of substation flow data based on deep learning, first obtain sample flow data within several historical monitoring cycles of each flow monitoring point, and use the sample flow data within these several historical monitoring cycles as the training set and test set of the differentiated evaluation model to train the differentiated evaluation model. Input the training set into the initial differentiated evaluation model for training until the loss function converges or meets the accuracy requirements of the model, and save the model parameters. Then test the differentiated evaluation model with the test set until it meets the reliability requirements of the model, and output the differentiated evaluation model.

[0116] For step S62, when extracting the time and space features of the target flow data respectively, it can be achieved in the following manner:

[0117] Pre-training LTSM neural network for predicting flow change trends of flow data;

[0118] The target flow data is input into the LTSM neural network to obtain the target flow change trend characteristics of the corresponding target flow monitoring point in the current monitoring period;

[0119] Importing the flow change trend characteristics into the nodes corresponding to the pre-trained substation directed topology graph; wherein each flow monitoring point in the substation directed topology graph constitutes a node of the topology graph;

[0120] Taking the flow monitoring point corresponding to the target flow data as the target node, determining the coupling weight between the target node and each adjacent node in the directed topological graph of the substation;

[0121] The feature aggregation representation of each adjacent node after acting on the target node is determined according to the coupling weight, and the flow spatiotemporal feature sequence is obtained.

[0122] In this embodiment, it is considered to obtain the historical flow time series corresponding to the power data type collected by each flow monitoring point in the historical monitoring period, and based on this, an LTSM neural network for predicting the change trend of flow data is constructed. Then, after the abnormal information of power equipment is generated at the flow monitoring point, it is considered to input the target flow data of the corresponding abnormal type into the trained LTSM neural network, and output the target flow change trend characteristics of the corresponding target flow monitoring point. Further, the directed topology graph of the substation can be learned through the graph attention network, and the target flow change trend characteristics obtained by each flow monitoring point in the current monitoring period are imported into the nodes corresponding to the directed topology graph of the substation, so that the flow monitoring point that generates the abnormal information of power equipment corresponds to the node in the directed topology graph of the substation as the target node, and the coupling weight of the target node to the adjacent node in the directed topology graph of the substation is determined. Further, the coupling weight is assigned to each adjacent node, and the feature aggregation representation after each adjacent node acts on the target node can be obtained by using the neighbor aggregation mechanism of the graph attention network, thereby obtaining the flow spatiotemporal feature sequence. In this way, by fully considering the influence of adjacent nodes on the target node, that is, the correlation between the flow data collected by the adjacent flow monitoring points and the flow data collected by the current flow monitoring point, the obtained data not only contains time characteristics, but also includes spatial characteristics, and fully considers the correlation between each flow monitoring point or each power equipment, so that differential analysis can be performed based on the obtained flow time-space feature sequence, which can obtain more accurate differential analysis results, and thus make the monitoring results of substation abnormal status monitoring more accurate.

[0123] Furthermore, when determining the coupling weights between the target node and each adjacent node in the directed topology graph of the substation, it can be achieved in the following manner:

[0124] Obtaining a first historical flow time series sequence collected by the target node and its adjacent nodes when the target node and its adjacent nodes in the directed topology graph of the substation do not generate power equipment abnormal information within a historical monitoring period; and obtaining a second historical flow time series sequence collected by the adjacent nodes when the target node in the directed topology graph of the substation generates power equipment abnormal information within a historical monitoring period;

[0125] Calculate the first average flow information of the power data monitoring type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence; and calculate the second average flow information of the power data monitoring type corresponding to the adjacent nodes according to the second historical flow time series sequence;

[0126] Determine a first flow amplitude time series sequence of the power data type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence and the first average flow information; and determine a second flow amplitude time series sequence of the power data type corresponding to the adjacent nodes according to the second historical flow time series sequence and the second average flow information;

[0127] Calculate the differentiation coefficient of the power data monitoring type corresponding to the adjacent nodes according to the first flow amplitude time series sequence and the second flow amplitude time series sequence;

[0128] According to the differentiation coefficient, the coupling weight is determined.

[0129] Specifically, when determining the first flow amplitude time series sequence of the power data type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence and the first average flow information, and determining the corresponding second flow amplitude time series sequence according to the second historical flow time series sequence and the second average flow information, the following calculation formula can be used to calculate:

[0130]

[0131] Among them, H i represents the traffic amplitude time series of node i, X iL represents the flow amplitude of node i at the Lth moment, N represents the total number of moments, Q iL represents the traffic of node i at the Lth moment, Q iL A represents the average flow rate of node i, and δ is the conversion coefficient.

[0132] When calculating the differentiation coefficient of the power data monitoring type corresponding to the adjacent nodes according to the first flow amplitude time series sequence and the second flow amplitude time series sequence, the following calculation formula can be used to calculate:

[0133]

[0134] Among them, y(ij) represents the difference coefficient between node i and node j, X iL represents the traffic amplitude of node i at the Lth moment obtained from the first traffic amplitude time series, X jL represents the flow amplitude of node j at the Lth moment obtained from the first flow amplitude time series, Y iL represents the flow amplitude of node i at the Lth moment obtained from the second flow amplitude time series, Y jL represents the flow amplitude of node j at the Lth moment obtained from the second flow amplitude time series;

[0135] When determining the coupling weight according to the differentiation coefficient, the following calculation formula can be used:

[0136] eij =θy(ij)

[0137] Among them, e ij represents the coupling weight between node i and node j, y(ij) represents the differentiation coefficient between node i and node j, and θ represents the conversion coefficient.

[0138] Furthermore, after determining the feature aggregation representation of each adjacent node acting on the target node according to the coupling weight, the following calculation formula can be used to obtain it:

[0139] F r * =ReLu(pF r +∑ z∈ur y(rz)pF z )

[0140] Among them, F r * represents the feature aggregation representation of node r, ReLu represents the activation function, p represents the parameter matrix of feature transformation, y(rz) represents the coupling weight of node r to node z, u r represents the set of adjacent nodes of node r, F r and F z They are the target traffic change trend characteristics of node r and node z respectively.

[0141] Furthermore, after step 106, corresponding abnormality prevention measures may be generated according to the results of the differential analysis. Specifically, after obtaining the results of the differential analysis of the substation flow data, the following may be further included:

[0142] Obtain an estimated flow time series sequence of the power data monitoring type of other flow monitoring points adjacent to the target flow monitoring point;

[0143] Compare the estimated flow time series of each adjacent flow monitoring point with the corresponding flow threshold, flow deviation threshold and abnormal time threshold;

[0144] Determine whether each adjacent flow monitoring point has a potential abnormal type and an estimated time stamp of the potential abnormal type according to the comparison result;

[0145] According to the adjacent traffic monitoring points and estimated timestamps that generate potential anomaly types, corresponding anomaly prevention measures are generated.

[0146] In this embodiment, after determining the differentiated analysis results of the substation flow data, the estimated flow time series of other adjacent nodes can also be obtained, and then compared with the corresponding flow threshold, flow deviation threshold and abnormal time threshold based on the method of step 103, so as to determine the potential abnormal type and the estimated timestamp of the abnormal type. In this way, corresponding abnormal prevention measures can be generated based on the potential abnormal type and the estimated timestamp, thereby further avoiding the losses and risks caused by the maturity of potential faults.

[0147] The present specification also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute a method in any one of the embodiments in the specification.

[0148] The present specification also provides a computing device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, the method in any embodiment of the specification is implemented.

[0149] Since the device embodiments provided by the present invention are based on the same inventive concept as the method embodiments of this specification, the specific contents can be found in the description of the method embodiments of this specification and will not be repeated here.

[0150] The modules or units in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs. The above disclosure is only the preferred embodiment of the present invention, and of course it cannot be used to limit the scope of the rights of the present invention. Those skilled in the art can understand that all or part of the processes of the above embodiment are implemented, and the equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A flow data differentiation analysis method applied to a substation, characterized in that: include: Pre-constructing a digital twin model of the substation operation process; wherein the digital twin model includes various flow monitoring points; wherein each flow monitoring point is used to monitor at least one type of power data of the power equipment; Obtaining flow data collected by flow monitoring points covering various power equipment set up in the substation, and importing the data into the digital twin model; Generate power equipment abnormality information indicating that the power equipment monitored by the flow monitoring point is abnormal according to the flow data; Acquire the target flow monitoring point for generating the abnormal information of the electric power equipment from the digital twin model; Acquire target flow data corresponding to the target flow monitoring point from the flow data; The target traffic data is subjected to differential analysis using a differentiated evaluation model pre-trained based on deep learning to obtain a differentiated analysis result.

2. The flow data differentiation analysis method applied to a substation according to claim 1 is characterized in that: The pre-building of the digital twin model of the substation operation process includes: Construct a three-dimensional model corresponding to the physical entities of each power equipment during the current substation operation; Obtain the communication link relationship between each flow monitoring point during the operation of the substation; The directed topology graph is constructed by taking the three-dimensional model corresponding to each power equipment as the node of the directed topology graph and taking the communication link relationship between each flow monitoring point as the connection relationship between the nodes; The flow data collected at each flow monitoring point is matched with the three-dimensional model corresponding to each node of the directed topological graph to obtain the digital twin model.

3. The flow data differentiation analysis method applied to a substation according to claim 1 is characterized in that: The method of obtaining the flow data collected by the flow monitoring points covering each power equipment set in the substation includes: Obtaining process unit characteristics of each power equipment in the current substation; wherein the process unit characteristics include functional characteristics, working parameter characteristics, structural characteristics and safety and reliability characteristics; Extracting process information according to the structural features of the process unit features; wherein the process information includes the assembly sequence and assembly relationship of the power equipment; Determine the flow monitoring points of each power equipment according to the process information; Determine the type of power data monitoring at each flow monitoring point according to the functional characteristics in the process unit characteristics; Obtain the flow data of each power data monitoring type corresponding to each flow monitoring point in real time.

4. The flow data differentiation analysis method applied to a substation according to claim 1 is characterized in that: The generating, according to the flow data, power equipment abnormality information indicating that the power equipment monitored by the flow monitoring point is abnormal, comprises: Preset the flow threshold, flow deviation threshold and abnormal time threshold corresponding to each power data monitoring type at each flow monitoring point; For the power data monitoring type of each flow monitoring point, the difference between the flow data of the power data monitoring type and the corresponding flow threshold is calculated to obtain the deviation flow; Compare the deviation flow with the flow deviation threshold, and when the deviation flow is greater than the flow deviation threshold, accumulate the time during which the deviation flow corresponding to the current power data monitoring type is greater than the flow deviation threshold; When the accumulated time is greater than the abnormal time threshold, the corresponding power data monitoring type is marked as an abnormal type, and power equipment abnormal information corresponding to the power equipment at the flow monitoring point where the abnormal type is located is generated.

5. The flow data differentiation analysis method applied to a substation according to claim 1 is characterized in that: The differential analysis of the target traffic data using the differentiated evaluation model pre-trained based on deep learning includes: Based on deep learning, a differentiated evaluation model is pre-trained for differentiated analysis of substation flow data; Extracting the time and space characteristics of the target flow data respectively to obtain a flow time and space characteristic sequence; The flow spatiotemporal feature sequence is input into the differential evaluation model, and the differential analysis result of the substation flow data is output.

6. The flow data differentiation analysis method applied to a substation according to claim 5 is characterized in that: The extracting time and space features of the target traffic data respectively comprises: Pre-training LTSM neural network for predicting flow change trends of flow data; Inputting the target flow data into the LTSM neural network to obtain the target flow change trend characteristics corresponding to the target flow monitoring point in the current monitoring period; Importing the flow change trend characteristics into the nodes corresponding to the pre-trained substation directed topology graph; wherein each flow monitoring point in the substation directed topology graph constitutes a node of the topology graph; Taking the flow monitoring point corresponding to the target flow data as the target node, determining the coupling weight between the target node and each adjacent node in the directed topological graph of the substation; The feature aggregation representation after each adjacent node acts on the target node is determined according to the coupling weight to obtain the traffic spatiotemporal feature sequence.

7. The flow data differentiation analysis method applied to a substation according to claim 6 is characterized in that: The determining of the coupling weights between the target node and each adjacent node in the directed topological graph of the substation includes: Obtaining a first historical flow time series sequence collected by the target node and its adjacent nodes when the target node and its adjacent nodes in the directed topology graph of the substation do not generate power equipment abnormal information within a historical monitoring period; and obtaining a second historical flow time series sequence collected by the adjacent nodes when the target node in the directed topology graph of the substation generates power equipment abnormal information within a historical monitoring period; Calculating the first average flow information of the power data monitoring type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence; and calculating the second average flow information of the power data monitoring type corresponding to the adjacent nodes according to the second historical flow time series sequence; Determine a first flow amplitude time series sequence of the power data type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence and the first average flow information; and determine a second flow amplitude time series sequence of the power data type corresponding to the adjacent nodes according to the second historical flow time series sequence and the second average flow information; Calculating, according to the first flow amplitude time series sequence and the second flow amplitude time series sequence, a differentiation coefficient of the power data monitoring type corresponding to the adjacent nodes; The coupling weight is determined according to the differentiation coefficient.

8. The flow data differentiation analysis method applied to a substation according to claim 7 is characterized in that: When determining the first flow amplitude time series sequence of the power data type corresponding to the target node and its adjacent nodes according to the first historical flow time series sequence and the first average flow information, or determining the corresponding second flow amplitude time series sequence according to the second historical flow time series sequence and the second average flow information, the following calculation formula is used to calculate: Among them, H i represents the traffic amplitude time series of node i, X iL represents the flow amplitude of node i at the Lth moment, N represents the total number of moments, Q iL represents the traffic of node i at the Lth moment, Q iL A represents the average flow rate of node i, and δ is the conversion coefficient; and / or, When calculating the differentiation coefficient of the power data monitoring type corresponding to the adjacent nodes according to the first flow amplitude time series sequence and the second flow amplitude time series sequence, the following calculation formula is used to obtain: Among them, y(ij) represents the difference coefficient between node i and node j, X iL represents the traffic amplitude of node i at the Lth moment obtained from the first traffic amplitude time series, X jL represents the flow amplitude of node j at the Lth moment obtained from the first flow amplitude time series, Y iL represents the flow amplitude of node i at the Lth moment obtained from the second flow amplitude time series, Y jL represents the flow amplitude of node j at the Lth moment obtained from the second flow amplitude time series; and / or, When determining the coupling weight according to the differentiation coefficient, the following calculation formula is used: e ij =θy(ij) Among them, e ij represents the coupling weight between node i and node j, y(ij) represents the differentiation coefficient between node i and node j, and θ represents the conversion coefficient.

9. The flow data differentiation analysis method applied to a substation according to claim 6 is characterized in that: Determining the feature aggregation representation after each adjacent node acts on the target node according to the coupling weight includes obtaining by using the following calculation formula: F r * =ReLu(pF r +∑ z∈ur y(rz)pF z ) Among them, F r * represents the feature aggregation representation of node r, ReLu represents the activation function, p represents the parameter matrix of feature transformation, y(rz) represents the coupling weight of node r to node z, u r represents the set of adjacent nodes of node r, F r and F z They are the target traffic change trend characteristics of node r and node z respectively.

10. The flow data differentiation analysis method applied to a substation according to any one of claims 1 to 9, characterized in that: After obtaining the differentiated analysis results of the substation flow data, further analysis includes: Obtaining an estimated flow time series sequence of the power data monitoring type of other flow monitoring points adjacent to the target flow monitoring point; Compare the estimated flow time series of each adjacent flow monitoring point with the corresponding flow threshold, flow deviation threshold and abnormal time threshold; Determine whether each adjacent flow monitoring point has a potential abnormal type and an estimated time stamp of the potential abnormal type according to the comparison result; According to the adjacent traffic monitoring points and estimated timestamps that generate the potential anomaly type, corresponding anomaly prevention measures are generated.

Citation Information

Patent Citations

  • Transformer substation fault early warning system and method based on digital twinning

    CN114881292A

  • Enterprise carbon quota prediction method based on enterprise operation data

    CN116681187A

  • Water conservancy model construction method and system based on digital twinning

    CN117371337A

  • System and method for evaluating health state of power distribution equipment

    CN117556359A

  • Gas volume scheduling evaluation method based on big data natural gas transmission network

    CN117575341A