Intelligent substation fault classification method based on integrated monitoring platform

Abstract

The invention provides an intelligent substation fault classification method based on an integrated monitoring platform, and the method specifically comprises the steps: obtaining an operation plan of an intelligent substation, and determining an effective judgment range of a power grid; the method comprises the following steps: constructing an integrated monitoring platform, calling historical panoramic data of a power grid within an effective judgment range, setting a fault type according to a main wiring mode and a protection configuration condition of an intelligent substation, taking the historical panoramic data and the set fault type as training samples, constructing a GCN model, and setting cross entropy loss as a corresponding loss function. And training the GCN model, obtaining panoramic data of the intelligent substation in the effective judgment range in real time through the integrated monitoring platform, and detecting the fault type of the intelligent substation in real time through the GCN model. The data source is guaranteed through the integrated monitoring platform, the accuracy of fault classification analysis is improved, the validity of the applied data is guaranteed by delimiting the effective judgment range, and the analysis efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of substation monitoring, and in particular to an intelligent substation fault classification method based on an integrated monitoring platform. Background Art

[0002] Smart substation is an important part of smart grid and the foundation of strong smart grid. Therefore, the operation safety monitoring of smart substation is particularly important. The fault classification analysis of smart substation can assist in the rapid handling of smart substation faults and is an important part of the operation safety monitoring of smart substation.

[0003] Most of the fault classification of existing smart substations is analyzed by traditional fault type judgment algorithms that rely on thresholds and logical relationships. This method requires pre-calibration of thresholds and has high labor costs. With the extensive research and application of artificial intelligence, artificial intelligence algorithms gradually replace traditional fault type judgment algorithms to realize fault classification analysis of smart substations to solve the problem of high labor costs. However, artificial intelligence algorithms have high requirements for the validity, real-time performance of data sources, and standardization and unification of data. In terms of design methods, existing smart substations still have problems such as numerous subsystems, independent construction, and low integration. The numerous independent constructions and low-integration subsystems will have problems such as inconsistent information models, large differences in interface methods, and inability to interoperate between different manufacturers. This also makes it impossible to guarantee the validity of data sources and retrieved data when using artificial intelligence algorithms for smart substation fault classification analysis. The efficiency and accuracy of smart substation fault classification analysis obtained by artificial intelligence algorithms are both low. Summary of the invention

[0004] The purpose of the present invention is to overcome the problems in the prior art that when using artificial intelligence algorithms to perform fault analysis on smart substations, the smart substations have inconsistent information models, large differences in interface methods, and cannot achieve interoperability between different manufacturers. The validity of data sources and retrieved data cannot be guaranteed, and the efficiency and accuracy of smart substation fault classification analysis are low. A smart substation fault classification method based on an integrated monitoring platform is provided, which can define an effective judgment range according to the operation plan of the smart substation, and only use data within the effective judgment range to perform smart substation fault classification analysis, thereby improving the efficiency of smart substation fault classification analysis, and constructing an integrated monitoring platform to uniformly monitor power grid data, providing a unified and standardized data source for smart substation fault classification analysis, thereby improving the accuracy of subsequent smart substation fault classification analysis.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A fault classification method for intelligent substation based on an integrated monitoring platform, comprising:

[0007] Obtain the current operation plan of the smart substation, and determine the effective judgment range of the power grid based on the operation plan of the smart substation;

[0008] Build an integrated monitoring platform to uniformly monitor power grid data;

[0009] Based on the integrated monitoring platform, historical panoramic data of the power grid within the effective judgment range can be retrieved;

[0010] Set the fault type according to the main wiring mode and protection configuration of the smart substation;

[0011] Use historical panoramic data and set fault types as training samples to build a GCN model;

[0012] Set the cross entropy loss as the loss function for GCN model training and train the GCN model;

[0013] Through the integrated monitoring platform, the panoramic data of the smart substation within the effective judgment range is obtained in real time, and the trained GCN model is input to detect the fault type of the smart substation in real time.

[0014] Furthermore, the obtaining of the current operation plan of the smart substation and determining the effective determination range in combination with the operation plan of the smart substation include:

[0015] Determine the operating equipment and corresponding operating strategies according to the operation plan of the smart substation, and define the primary power supply scope of the smart substation based on the topological relationship of the operating equipment;

[0016] Retrieve historical power consumption data within the primary power supply range of the smart substation, and predict the power demand within the primary power supply range based on the retrieved historical power consumption data, adjust the primary power supply range based on the power demand results, and obtain the secondary power supply range of the smart substation;

[0017] The interconnection line information of the smart substation is extracted, the operating equipment is matched with each interconnection line, and the secondary power supply range is adjusted according to the operating status, capacity and configuration method of the corresponding interconnection line to obtain the effective judgment range corresponding to the smart substation.

[0018] Furthermore, the construction of the integrated monitoring platform includes:

[0019] Obtain the topological relationship, service type and information interaction characteristics of the smart substation within the effective determination range;

[0020] Construct the public information model corresponding to the smart substation according to the business type, and determine the professional extension model connected to the public information model according to the topological relationship;

[0021] According to the information interaction characteristics, the inter-layer connection relationship within the public information model and the connection relationship between the public information model and the professional extension model are constructed to obtain the full power grid model corresponding to the smart substation;

[0022] Unify the data interface and monitoring standards of the entire power grid model, and build an integrated monitoring platform based on the entire power grid model.

[0023] Furthermore, the building of the GCN model includes the expression of the output function of the GCN model being:

[0024]

[0025] Among them, Z k is the output value of the GCN model for the historical panoramic data of the kth node in the power grid, X k To effectively determine the historical panoramic data of the kth node in the power grid within the range, W f and b f are the weight matrix and bias matrix of the fully connected layer respectively, and A is the adjacency matrix of the power grid topology, I is the identity matrix, For the matrix The degree matrix, H (l +1) and H (l) are the outputs of the l+1th graph convolutional layer and the lth graph convolutional layer of the GCN model, respectively. l ) is the weight matrix of the lth graph convolutional layer of the GCN model, and where F h is the output dimension of the lth graph convolutional layer, and σ() is the activation function of the GCN model.

[0026] Furthermore, the expression of the activation function of the GCN model is:

[0027] σ(x)=LeakyReLU(x,β)=max(0,x)+β×min(0,x);

[0028] Among them, σ(x) is the function value of the activation function of the graph convolution layer of the GCN model, LeakyReLU() is the LeakyReLU activation function, x is the input data of the graph convolution layer of the GCN model, β is a constant, and β∈(0,1), max() is the maximum value function, and min() is the minimum value function.

[0029] Furthermore, the cross entropy loss is set as the loss function of the GCN model training, including the expression of the cross entropy loss is:

[0030]

[0031] Where L is the cross entropy loss value, N is the number of power grid nodes within the effective judgment range and the output dimension of the fully connected layer, and the number of power grid nodes within the effective judgment range is the same as the output dimension of the fully connected layer, k is the kth node of the power grid, y k is the fault type label value of the i-th node, Z k and Z j They are the outputs of the GCN model of the k-th node and the j-th node respectively, onehot() is the one-hot vectorization function, and exp() is the exponential function with the natural constant e as the base.

[0032] Furthermore, the fault types include primary fault types and secondary fault types. The primary fault types include 220kV line fault, 110kV line fault, main transformer fault, 220kV bus fault and 110kV bus fault. The secondary fault types include instantaneous single-phase grounding fault, permanent single-phase grounding fault, phase-to-phase fault or three-phase fault, manual closing fault, tripping on this side due to receiving remote trip signal, conversion fault within reclosing time, single-phase fault and fault phase failure, phase-to-phase fault and fault phase failure, instantaneous fault, permanent fault, line fault and switch failure, main transformer fault, internal fault and heavy gas action, internal fault and pressure release action and bus fault.

[0033] Furthermore, the panoramic data includes steady-state data, transient data and dynamic data.

[0034] The beneficial effects of the present invention are:

[0035] It can define the effective judgment range according to the operation plan of the smart substation, and only use the data within the effective judgment range for smart substation fault classification analysis, thereby improving the efficiency of smart substation fault classification analysis. It also builds an integrated monitoring platform to uniformly monitor power grid data, providing a unified and standardized data source for smart substation fault classification analysis, thereby improving the accuracy of subsequent smart substation fault classification analysis.

[0036] In terms of artificial intelligence algorithms, the GCN model was introduced. The GCN model is a graph convolutional network. Its network structure is highly unified with the topological structure of the power grid graph. It is very suitable for solving power grid data-driven problems and can further improve the accuracy of fault classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of a process of the present invention;

[0038] Figure 2 It is a structural schematic diagram of an integrated monitoring platform according to an embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of the structure of a panoramic data center according to an embodiment of the present invention;

[0040] Figure 4 It is a GCN model structure diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0042] Example:

[0043] A fault classification method for smart substations based on an integrated monitoring platform, such as Figure 1 As shown, including,

[0044] Obtain the current operation plan of the smart substation, and determine the effective judgment range of the power grid based on the operation plan of the smart substation;

[0045] Build an integrated monitoring platform to uniformly monitor power grid data;

[0046] Based on the integrated monitoring platform, historical panoramic data of the power grid within the effective judgment range can be retrieved;

[0047] Set the fault type according to the main wiring mode and protection configuration of the smart substation;

[0048] Use historical panoramic data and set fault types as training samples to build a GCN model;

[0049] Set the cross entropy loss as the loss function for GCN model training and train the GCN model;

[0050] Through the integrated monitoring platform, the panoramic data of the smart substation within the effective judgment range is obtained in real time, and the trained GCN model is input to detect the fault type of the smart substation in real time.

[0051] The obtaining of the current operation plan of the smart substation and determining the effective determination range in combination with the operation plan of the smart substation include:

[0052] Determine the operating equipment and corresponding operating strategies according to the operation plan of the smart substation, and define the primary power supply scope of the smart substation based on the topological relationship of the operating equipment;

[0053] Retrieve historical power consumption data within the primary power supply range of the smart substation, and predict the power demand within the primary power supply range based on the retrieved historical power consumption data, adjust the primary power supply range based on the power demand results, and obtain the secondary power supply range of the smart substation;

[0054] The interconnection line information of the smart substation is extracted, the operating equipment is matched with each interconnection line, and the secondary power supply range is adjusted according to the operating status, capacity and configuration method of the corresponding interconnection line to obtain the effective judgment range corresponding to the smart substation.

[0055] When conducting fault classification analysis of smart substations, in order to reduce the amount of data processing of artificial intelligence algorithms, the effective judgment range is defined in combination with the operation plan of the smart substation, and only the power grid data within the effective judgment range is called for fault classification analysis to improve the analysis efficiency of fault classification.

[0056] The delineation of the effective judgment range is essentially to ensure the validity of the called power grid data and the recognition efficiency and accuracy of the subsequent GCN model. Therefore, the operating equipment involved in the operation plan of the smart substation and the corresponding operation strategies, such as the capacity and efficiency of transformers, switchgear and transmission lines, the schedule and arrangement of equipment inspection, maintenance and upgrade, etc., are used to determine the power supply range corresponding to the smart substation. The power grid data generated within this power supply range is more correlated with the smart substation, and its data validity can be guaranteed.

[0057] The power supply range of a smart substation will change due to factors such as power demand, equipment operation, and the operation of the interconnection lines involved. Therefore, after determining the primary power supply range according to its operation plan, the power supply range can be adjusted based on power demand and interconnection line information to determine the effective judgment range corresponding to the smart substation.

[0058] Among them, the interconnection line is an important channel connecting different substations or power grids. Its status, capacity and configuration are directly related to the determination of the power supply range and the reliability of power transmission.

[0059] Specifically, the state of the tie line directly affects the continuity and stability of power supply. If the tie line fails or requires maintenance, it may cause power outage or restriction in some areas. The capacity of the tie line determines the ability to transmit power. The capacity of different tie lines may vary, which will affect the distribution and transmission of power between different substations. If the capacity of the tie line is insufficient, it may lead to limited power transmission, which in turn affects the size of the power supply range. The number and distribution of tie lines will affect the topology of the power grid and the flexibility of power transmission. The configuration of the tie lines directly affects the demarcation of the power supply range.

[0060] Secondly, in this embodiment, after retrieving the historical electricity consumption data, the electricity demand is predicted by combining the historical electricity consumption data through time series analysis.

[0061] The construction of the integrated monitoring platform includes:

[0062] Obtain the topological relationship, service type and information interaction characteristics of the smart substation within the effective determination range;

[0063] Construct the public information model corresponding to the smart substation according to the business type, and determine the professional extension model connected to the public information model according to the topological relationship;

[0064] According to the information interaction characteristics, the inter-layer connection relationship within the public information model and the connection relationship between the public information model and the professional extension model are constructed to obtain the full power grid model corresponding to the smart substation;

[0065] Unify the data interface and monitoring standards of the entire power grid model, and build an integrated monitoring platform based on the entire power grid model.

[0066] Specifically, in this embodiment, an integrated monitoring platform is constructed based on the IEC 61970 CIM model, i.e., the common information model. When constructing the integrated monitoring platform, in addition to the above-mentioned model construction and connection relationship determination steps, it is necessary to first establish a development environment for the integrated monitoring platform, thereby providing a multi-level software interface, and the established development environment supports business customization and adjustment, thereby adapting to the business needs of different types of smart substations.

[0067] At the same time, it is also necessary to establish an integrated environment so that the integrated monitoring platform can have the capabilities of system integration and business integration. After building the full power grid model, it can ensure that the subsystems contained in the full power grid model can be connected normally, and ensure that after connecting to the full power grid model, it can receive all kinds of data generated to achieve unified monitoring of power grid data.

[0068] Secondly, it is necessary to establish an operating environment and a maintenance environment that can provide system management and security management mechanisms to achieve resource monitoring and optimization while ensuring that the constructed integrated monitoring platform can meet business needs and provide a security protection system.

[0069] The integrated monitoring platform constructed can access various subsystems and receive various data generated. It can also retrieve corresponding power grid data from the dispatching center through advanced applications to achieve comprehensive monitoring of power grid data.

[0070] Each subsystem is connected to the integrated monitoring platform through the IEC 61850 protocol, and various data are connected to the integrated monitoring platform through IEC61850-9-2GOOSE.

[0071] Specifically, the structural diagram of the constructed integrated monitoring platform is as follows: Figure 2 shown.

[0072] After building an integrated monitoring platform, in order to optimize the subsequent retrieval and application of power grid data, it is also necessary to build a corresponding panoramic data center to facilitate the rapid retrieval of corresponding panoramic data during the subsequent intelligent substation fault classification analysis.

[0073] For the panoramic data center, its structure diagram is as follows Figure 3 As shown, when building a panoramic data center, it is necessary to unify the coding and standardize the semantic encapsulation, that is, to set up a unified information service to provide analysis, calculation, early warning and auxiliary decision-making of the full power grid model after the full power grid model is connected, and provide data storage and management functions for various applications.

[0074] Secondly, integrated monitoring standards are formulated to provide a unified data bus and message interface, that is, a unified data interface (API) that supports multi-source data, and application function modularization and standardization specifications are formulated to achieve information sharing and collaborative interaction between different application function modules, further optimizing the subsequent data call efficiency and the data monitoring efficiency after each subsystem is connected to the integrated monitoring platform.

[0075] Specifically, the panoramic data center is equipped with an equipment parameter library, a steady-state database, a dynamic database, a transient database, an equipment monitoring database, and an environment and auxiliary database to achieve comprehensive collection of power grid data.

[0076] After ensuring the data source, it is also necessary to set the fault type according to the specific operating conditions of the smart substation to optimize the analysis efficiency of subsequent smart substation fault classification.

[0077] The fault types include primary fault types and secondary fault types. The primary fault types include 220kV line fault, 110kV line fault, main transformer fault, 220kV bus fault and 110kV bus fault. The secondary fault types include instantaneous single-phase grounding fault, permanent single-phase grounding fault, phase-to-phase fault or three-phase fault, manual closing fault, tripping on this side due to receiving remote trip signal, conversion fault within reclosing time, single-phase fault and fault phase failure, phase-to-phase fault and fault phase failure, instantaneous fault, permanent fault, line fault and switch failure, main transformer fault, internal fault and heavy gas action, internal fault and pressure release action and bus fault.

[0078] The first-level fault and the second-level fault are corresponding. Specifically, the fault types set in this embodiment are shown in Table 1:

[0079] Table 1 Fault type table

[0080]

[0081] After setting the fault type, set the corresponding label for each fault type to facilitate model training.

[0082] Historical panoramic data also includes three types of data: steady-state data, transient data, and dynamic data. When retrieving historical panoramic data, only the global data when the corresponding fault type occurs in the smart substation is retrieved to realize subsequent GCN model training.

[0083] The historical panoramic data and fault type labels are used as training samples, the specific historical panoramic data are the input data in the training samples, and the fault type labels are the output data.

[0084] The input data is specifically shown in the following matrix X:

[0085]

[0086] Where N represents the number of nodes, F represents the number of features, is the eigenvalue of the i-th feature of the j-th node.

[0087] The output data, i.e. the fault type label, is as follows: k As shown:

[0088]

[0089] Among them, yk is the fault type label corresponding to the kth node, m is the number of fault types, is a set of natural numbers.

[0090] Build as Figure 4 The GCN model shown includes two graph convolutional layers and one fully connected layer.

[0091] The expression of the activation function of the GCN model is:

[0092] σ(x)=LeakyReLU(x,β)=max(0,x)+β×mmin(0,x);

[0093] Among them, σ(x) is the function value of the activation function of the graph convolution layer of the GCN model, LeakyReLU() is the LeakyReLU activation function, x is the input data of the graph convolution layer of the GCN model, β is a constant, and β∈(0,1). In this embodiment, β is specifically set to 0.01, max() is the maximum value function, and min() is the minimum value function.

[0094] The output of the graph convolutional layer is shown below:

[0095]

[0096] Combining the above formula, it can be deduced that after the final fully connected layer, the output function of the GCN model is expressed as:

[0097]

[0098] Among them, Z k is the output value of the GCN model for the historical panoramic data of the kth node in the power grid, X k To effectively determine the historical panoramic data of the kth node in the power grid within the range, W f and b f are the weight matrix and bias matrix of the fully connected layer respectively, and A is the adjacency matrix of the power grid topology, I is the identity matrix, For the matrix The degree matrix, H (l +1) and H (l) are the outputs of the l+1th graph convolutional layer and the lth graph convolutional layer of the GCN model, respectively. W(l) is the weight matrix of the lth graph convolutional layer of the GCN model, and Among them, F h is the output dimension of the lth graph convolutional layer, and σ() is the activation function of the GCN model.

[0099] And set the cross entropy loss as the loss function for GCN model training, as shown in the following formula:

[0100]

[0101] Among them, L is the cross entropy loss value, N is the number of power grid nodes in the effective judgment range and the output dimension of the fully connected layer. The number of power grid nodes in the effective judgment range is the same as the output dimension of the fully connected layer. k is the kth node of the power grid, y k is the fault type label value of the i-th node, Z k and Z j They are the outputs of the GCN model of the k-th node and the j-th node respectively, onehot() is the one-hot vectorization function, and exp() is the exponential function with the natural constant e as the base.

[0102] Then use the gradient descent algorithm to train the GCN model until the loss converges to complete the training of the GCN model.

[0103] By inputting the panoramic data of the smart substation, including steady-state data, transient data and dynamic data, into the trained GCN model in real time, fault classification of the smart substation can be achieved.

[0104] The above-described embodiment is only a preferred solution of the present invention and does not limit the present invention in any form. There are other variations and modifications without exceeding the technical solution described in the claims.

Claims

1. A fault classification method for intelligent substation based on an integrated monitoring platform, characterized in that: Including, obtaining the current operation plan of the smart substation, and determining the effective determination range of the power grid in combination with the operation plan of the smart substation; Build an integrated monitoring platform to uniformly monitor power grid data; Based on the integrated monitoring platform, historical panoramic data of the power grid within the effective judgment range can be retrieved; Set the fault type according to the main wiring mode and protection configuration of the smart substation; Use historical panoramic data and set fault types as training samples to build a GCN model; Set the cross entropy loss as the loss function for GCN model training and train the GCN model; Through the integrated monitoring platform, the panoramic data of the smart substation within the effective judgment range is obtained in real time, and the trained GCN model is input to detect the fault type of the smart substation in real time.

2. A method for classifying faults of intelligent substations based on an integrated monitoring platform according to claim 1, characterized in that: The obtaining of the current operation plan of the smart substation and determining the effective determination range in combination with the operation plan of the smart substation include: Determine the operating equipment and corresponding operating strategies according to the operation plan of the smart substation, and define the primary power supply scope of the smart substation based on the topological relationship of the operating equipment; Retrieve historical power consumption data within the primary power supply range of the smart substation, and predict the power demand within the primary power supply range based on the retrieved historical power consumption data, adjust the primary power supply range based on the power demand results, and obtain the secondary power supply range of the smart substation; The interconnection line information of the smart substation is extracted, the operating equipment is matched with each interconnection line, and the secondary power supply range is adjusted according to the operating status, capacity and configuration method of the corresponding interconnection line to obtain the effective judgment range corresponding to the smart substation.

3. The intelligent substation fault classification method based on the integrated monitoring platform according to claim 1 is characterized in that: The construction of the integrated monitoring platform includes: Obtain the topological relationship, business type and information interaction characteristics of the smart substation within the effective determination range; construct the public information model corresponding to the smart substation according to the business type, and determine the professional extension model connected to the public information model according to the topological relationship; According to the information interaction characteristics, the inter-layer connection relationship within the public information model and the connection relationship between the public information model and the professional extension model are constructed to obtain the full power grid model corresponding to the smart substation; Unify the data interface and monitoring standards of the entire power grid model, and build an integrated monitoring platform based on the entire power grid model.

4. The method for classifying faults of intelligent substations based on an integrated monitoring platform according to claim 1 is characterized in that: The building of the GCN model includes: the expression of the output function of the GCN model is: Among them, Z k is the output value of the GCN model for the historical panoramic data of the kth node in the power grid, X k To effectively determine the historical panoramic data of the kth node in the power grid within the range, W f and b f are the weight matrix and bias matrix of the fully connected layer respectively, and A is the adjacency matrix of the power grid topology, I is the identity matrix, For the matrix The degree matrix, H (l +1) and H (l) are the outputs of the l+1th graph convolutional layer and the lth graph convolutional layer of the GCN model, respectively. (l) is the weight matrix of the lth graph convolutional layer of the GCN model, and Among them, F h is the output dimension of the lth graph convolutional layer, and σ( ) is the activation function of the GCN model.

5. The method for classifying faults of intelligent substations based on an integrated monitoring platform according to claim 4 is characterized in that: The expression of the activation function of the GCN model is: o(x)=LeakyReLU(x,β)=max(0,x)+β×min(0,x); Among them, σ(x) is the function value of the activation function of the graph convolution layer of the GCN model, LeakyReLU( ) is the LeakyReLU activation function, x is the input data of the graph convolution layer of the GCN model, β is a constant, and β∈(0,1), max( ) is the maximum value function, and min( ) is the minimum value function.

6. A fault classification method for intelligent substation based on integrated monitoring platform according to claim 4, characterized in that: The setting of the cross entropy loss as the loss function of the GCN model training includes the expression of the cross entropy loss being: Among them, L is the cross entropy loss value, N is the number of power grid nodes in the effective judgment range and the output dimension of the fully connected layer. The number of power grid nodes in the effective judgment range is the same as the output dimension of the fully connected layer. k is the kth node of the power grid, y k is the fault type label value of the i-th node, Z k and Z j are the outputs of the GCN model of the k-th node and the j-th node respectively, onehot() is the one-hot vectorization function, and exp() is the exponential function with the natural constant e as the base.

7. The method for classifying faults of intelligent substations based on an integrated monitoring platform according to claim 1, characterized in that: The fault types include primary fault types and secondary fault types. The primary fault types include 220kV line fault, 110kV line fault, main transformer fault, 220kV bus fault and 110kV bus fault. The secondary fault types include instantaneous single-phase grounding fault, permanent single-phase grounding fault, phase-to-phase fault or three-phase fault, manual closing fault, tripping on this side due to receiving remote trip signal, conversion fault within reclosing time, single-phase fault and fault phase failure, phase-to-phase fault and fault phase failure, instantaneous fault, permanent fault, line fault and switch failure, main transformer fault, internal fault and heavy gas action, internal fault and pressure release action and bus fault.

8. The method for classifying faults of intelligent substations based on an integrated monitoring platform according to claim 1, characterized in that: The panoramic data includes steady-state data, transient data and dynamic data.

Images