A power supply guaranteeing key equipment monitoring method and system, electronic device and medium

Abstract

The application provides a power supply guarantee key equipment monitoring method and system, an electronic device and a medium, and comprises the following steps: collecting real-time state data of a target power supply guarantee key equipment based on a sensor; the target power supply guarantee key equipment is any power supply guarantee key equipment in a power system, and the real-time state data comprises running state information, environment monitoring information and load bearing information at a current collection time t; controlling a target edge computing node to receive the real-time state data; inputting the running state information, the environment monitoring information and the load bearing information into an equipment fault monitoring model to obtain equipment monitoring information output by the equipment fault monitoring model; the equipment fault monitoring model is obtained based on sample running state information, sample environment monitoring information, sample load bearing information, corresponding alarm information and alarm category labels; and the running state of the target power supply guarantee key equipment is monitored based on the equipment monitoring information, so that the fault identification capability of the power supply guarantee key equipment is improved.

Description

Technical Field

[0001] This invention relates to the field of equipment monitoring technology, specifically to a method and system for monitoring key power supply equipment, as well as electronic equipment and media. Background Technology

[0002] The normal operation of critical power supply equipment is crucial for ensuring the safe and stable operation of the power system. Therefore, it is necessary to monitor the normal operating status of critical power supply equipment in real time. Existing methods for monitoring critical power supply equipment mainly rely on rule-based fault diagnosis, which involves setting fault judgment rules based on expert experience. When these rules are met, the monitored critical power supply equipment is considered to be in an abnormal state. However, rule-based fault diagnosis methods depend on fixed rules and thresholds to determine whether equipment is abnormal, making them unable to adapt to complex and changing equipment operating environments. This results in insufficient ability to identify new types of faults, thus reducing the fault identification capability of critical power supply equipment. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for monitoring key power supply equipment, as well as an electronic device and a computer-readable storage medium, to solve the technical problem that existing equipment cannot adapt to complex and ever-changing working environments, resulting in insufficient ability to identify new types of faults.

[0004] To achieve the objective of this invention, according to a first aspect of the invention, a method for monitoring key power supply equipment is provided, comprising:

[0005] Real-time status data of key power supply equipment is collected based on sensors; the key power supply equipment is any key power supply equipment in the power system, and the real-time status data includes operating status information, environmental monitoring information, and load information at the current acquisition time t.

[0006] The target edge computing node receives the real-time status data.

[0007] The operating status information, the environmental monitoring information, and the load information are input into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model. The equipment fault monitoring model is trained based on sample operating status information, sample environmental monitoring information, sample load information, and their corresponding alarm information and alarm category labels.

[0008] The operating status of the target power supply key equipment is monitored based on the equipment monitoring information.

[0009] According to a second aspect of the present invention, a monitoring system for critical equipment for ensuring power supply is provided, characterized in that it includes a module for performing the monitoring method for critical equipment for ensuring power supply as described in the first aspect above.

[0010] According to a third aspect of the present invention, an electronic device is provided, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby implementing the power supply critical equipment monitoring method described in the first aspect above.

[0011] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein a computer software program is stored therein, and when executed by a processor, the computer software program implements the power supply critical equipment monitoring method described in the first aspect above.

[0012] The power supply critical equipment monitoring method provided by this invention monitors each power supply critical equipment in the power system by training an equipment fault monitoring model based on operating status information, environmental monitoring information, load information and their corresponding alarm information and alarm category labels. This enables monitoring of the power supply critical equipment in different working environments, thus identifying the types of faults that occur in the power supply critical equipment in different working environments and improving the fault identification capability of the power supply critical equipment. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, obtaining other drawings based on these drawings without creative effort still falls within the scope of the present invention.

[0014] Figure 1 This is a flowchart of a method for monitoring key power supply equipment in an embodiment of the present invention;

[0015] Figure 2 This is a structural diagram of a power supply key equipment monitoring system according to an embodiment of the present invention;

[0016] Figure 3 This is an embodiment diagram of an electronic device according to an embodiment of the present invention;

[0017] Figure 4 This is an embodiment of a computer-readable storage medium according to an embodiment of the present invention. Detailed Implementation

[0018] The detailed description of the accompanying drawings is intended to illustrate the present embodiments of this application and is not intended to represent only the forms in which this application can be implemented. It should be understood that the same or equivalent functions can be accomplished by different embodiments intended to be included within the spirit and scope of this application.

[0019] Reference Figure 1 , Figure 1 This is a flowchart of a method for monitoring key power supply equipment provided in an embodiment of the present invention. The executing entity of the method for monitoring key power supply equipment provided in an embodiment of the present invention can be an equipment monitoring system. Therefore, the method for monitoring key power supply equipment includes the following steps:

[0020] Step 10: Collect real-time status data of key power supply equipment based on sensors.

[0021] In this embodiment of the invention, each key power supply device is equipped with a corresponding sensor, so the status data of the key power supply device can be collected through the sensor.

[0022] Therefore, for any key power supply equipment in the power system, its real-time status data at each point in time can be collected through the sensors on the key power supply equipment.

[0023] In one embodiment, the sensors may include an operational status monitoring sensor, an environmental monitoring sensor, and a load-bearing monitoring sensor. Therefore, the collected real-time status data may include operational status information, environmental monitoring information, and load-bearing information of the target power supply critical equipment at each current acquisition time t. In this embodiment, the operational status information refers to operational parameters such as temperature, pressure, rotational speed, and current. The environmental monitoring information refers to the surrounding environmental data of the target power supply critical equipment, such as humidity, vibration, noise level, and external temperature. The load-bearing information refers to the operating status of the target power supply critical equipment under different loads, such as peak load, average load, and load fluctuation data.

[0024] Step 20: Control the target edge computing node to receive real-time status data.

[0025] It should be noted that the sensor in this embodiment of the invention has a corresponding communication module embedded in it, which can transmit the collected real-time status data to the device monitoring system.

[0026] Therefore, to improve data transmission rates and achieve rapid monitoring of critical power supply equipment, in this embodiment of the invention, the equipment monitoring system communicates with sensors by calling edge computing nodes in the data transmission network. In the data transmission network, edge computing nodes are constantly processing transmitted data. Since different edge computing nodes process different amounts of data, their data transmission rates differ. Therefore, the equipment monitoring system needs to determine the optimal edge computing node (target edge computing node) in the data transmission network—that is, the edge computing node with the fastest data transmission rate, i.e., the edge computing node that receives real-time status data in the shortest time—and control the target edge computing node to receive the real-time status data transmitted by the sensors.

[0027] Step 30: Input the operating status information, environmental monitoring information, and load information into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model.

[0028] Optionally, after receiving the operating status information, environmental monitoring information, and load information, the equipment monitoring system inputs the operating status information, environmental monitoring information, and load information into a pre-trained equipment fault monitoring model. The equipment fault monitoring model processes the data and performs monitoring and prediction on the operating status information, environmental monitoring information, and load information, and outputs the corresponding equipment monitoring information.

[0029] The equipment fault monitoring model in this embodiment of the invention is trained based on sample operating status information, sample environmental monitoring information, sample load information and their corresponding alarm information and alarm category labels.

[0030] Step 40: Monitor the operating status of key power supply equipment based on equipment monitoring information.

[0031] Optionally, in this embodiment of the invention, the equipment monitoring information output by the equipment fault monitoring model includes equipment status information, equipment fault type and fault occurrence probability. The equipment status information includes whether the target power supply key equipment is in normal operation or in an abnormal state, and the abnormal state means that the target power supply key equipment has failed.

[0032] Furthermore, the equipment monitoring system monitors the operating status of key power supply equipment based on equipment status information, the type of equipment failure, and the probability of failure.

[0033] This invention provides an equipment fault monitoring model trained based on operating status information, environmental monitoring information, load information, and corresponding alarm information and alarm category labels. This model monitors each key power supply equipment in the power system, enabling monitoring in different working environments. As a result, it can identify the types of faults that occur in key power supply equipment under different working environments, thus improving the fault identification capability of key power supply equipment.

[0034] In one embodiment, step 20, controlling the target edge computing node to receive real-time status data, includes:

[0035] Step 201: Determine the data processing time for each edge computing node based on its data processing capability, current data processing volume, and pending data volume in the data transmission network.

[0036] Step 202: Determine the data transmission delay time of each edge computing node based on the channel power gain and data transmission rate of each edge computing node;

[0037] Step 203: Determine the data transmission consumption time of each edge computing node based on the data processing time and data transmission latency of each edge computing node;

[0038] Step 204: Determine the target edge computing node in the data transmission network based on the data transmission consumption time of each edge computing node.

[0039] Optionally, considering the different data transmission rates of different edge computing nodes, the device monitoring system acquires the data processing capability, current data processing volume, and pending data volume of each edge computing node in the data transmission network. Based on the data processing capability, current data processing volume, and pending data volume of each edge computing node, the system calculates the data processing time of each edge computing node. The specific calculation formula for the data processing time is as follows:

[0040]

[0041] in, Indicates data processing time. Indicates data processing capability. Indicates the current amount of data being processed. Indicates the amount of data to be processed. and This indicates a preset percentage weight. Generally, a preset percentage weight... Set to 0.6-0.8, preset percentage weight. Set it to 0.3-0.5.

[0042] Furthermore, the device monitoring system acquires the channel power gain and data transmission rate of each edge computing node. Based on the channel power gain and data transmission rate of each edge computing node, it calculates the data transmission delay time of each edge computing node. The specific calculation formula for the data transmission delay time is as follows:

[0043]

[0044] in, B represents the data transmission delay time, and K represents the channel bandwidth of the edge computing node. Indicates channel power gain. Indicates the data transmission rate. This represents the channel gain factor.

[0045] Furthermore, the equipment monitoring system calculates the data transmission consumption time of each edge computing node based on the data processing time and data transmission latency of each edge computing node. The specific calculation formula for the data transmission consumption time is as follows:

[0046]

[0047] in, Indicates the time consumed by data transmission. and This represents the weight of the time spent, where the weight of the time spent is... Weighting based on time consumption The sum equals 1. Generally, the weight of time consumption percentage is... Greater than the weight of time consumption Furthermore, there are no identical situations between the two.

[0048] Furthermore, the equipment monitoring system compares the data transmission time of each edge computing node and identifies the edge computing node with the shortest data transmission time as the target edge computing node in the data transmission network.

[0049] This invention determines the target edge computing node with the shortest data transmission time in the transmission network by considering the data processing capability, current data volume, pending data volume, channel power gain, and data transmission rate of each edge computing node. Real-time status data is received through the target edge computing node, thereby improving the data transmission rate and enabling rapid monitoring of critical power supply equipment.

[0050] In one embodiment, the training steps of the equipment fault monitoring model include:

[0051] Step 50: Collect sample operation status information, sample environment monitoring information, and sample load information;

[0052] Step 60: Based on the sample collection time, sample operation status information, sample environment monitoring information, and sample load information, generate multidimensional data features of the sample.

[0053] Step 70: Based on the multidimensional data features of the samples and their corresponding alarm information and alarm category labels, train the preset neural network model to obtain the equipment fault monitoring model.

[0054] Optionally, the equipment monitoring system collects sample operating status information, sample environmental monitoring information, and sample load information.

[0055] Furthermore, the equipment monitoring system extracts features from the sample operating status information, sample environmental monitoring information, and sample load information, as in the process of step 301 of the embodiment.

[0056] Furthermore, the equipment monitoring system extracts the sample data features from the sample operating status information, sample environment monitoring information, and sample load information, and adjusts the dimensions of the sample data features with the sample collection time to generate multi-dimensional sample data features, as in the embodiments of steps 302 and 3021 to 3024.

[0057] Furthermore, the equipment monitoring system inputs the multidimensional data features of the samples, along with their corresponding alarm information and alarm category labels, into a pre-set neural network model. The pre-set neural network model is then trained to obtain the equipment fault monitoring model. The pre-set neural network model can be a feedforward neural network (FNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory network (LSTM), or similar.

[0058] It should be noted that the neural network model in this embodiment of the invention is a Long Short-Term Memory (LSTM) network, that is, the equipment fault monitoring model in this embodiment of the invention is trained based on the Long Short-Term Memory (LSTM) network.

[0059] Considering the equipment fault monitoring model's predictions for different categories, and taking into account class imbalance and prediction uncertainty, the total loss function of the equipment fault monitoring model in this embodiment of the invention... Including classification loss function Weighted loss function and uncertainty loss function ;

[0060] Therefore, the total loss function of the equipment fault monitoring model It can be represented as:

[0061]

[0062] Among them, the classification loss function It can be represented as:

[0063]

[0064] Among them, the weighted loss function It can be represented as:

[0065]

[0066] Among them, the uncertainty loss function It can be represented as:

[0067]

[0068] Therefore, the final total loss function of the equipment fault monitoring model It can be represented as:

[0069]

[0070] Where N represents the number of samples and C represents the number of categories. This represents the alarm category label of the c-th category for the i-th sample. Let represent the probability predicted by the model for the c-th class of the i-th sample. This represents the weight for the i-th sample, which is dynamically set based on the sample's running state information. This represents the prediction variance for the i-th sample. This represents the first preset hyperparameter. This indicates the second preset hyperparameter.

[0071] This invention trains a fault monitoring model applicable to complex and ever-changing environments by collecting time, operating status information, environmental monitoring information, and load information, along with corresponding alarm information and alarm category labels. This improves the model's ability to identify new types of faults. Therefore, by using the fault monitoring model to monitor each key power supply equipment in the power system, it is possible to monitor key power supply equipment in different working environments. This allows for the identification of fault types occurring in key power supply equipment in different working environments, thus improving the fault identification capability of key power supply equipment.

[0072] In one embodiment, the equipment fault monitoring model includes a feature extraction layer, a data dimension processing layer, an equipment status prediction layer, a fault type determination layer, and a fault occurrence probability prediction layer. The feature extraction layer extracts data features from real-time status data. The data dimension processing layer concatenates the extracted data features to obtain multi-dimensional data features. The equipment status prediction layer predicts the equipment status of critical power supply equipment. The fault type determination layer predicts the fault type of critical power supply equipment. The fault occurrence probability prediction layer predicts the probability of fault occurrence for critical power supply equipment.

[0073] Therefore, step 30 inputs the operating status information, environmental monitoring information, and load information into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model, including:

[0074] Step 301: Input the operating status information, environmental monitoring information, and load information into the feature extraction layer for feature extraction, and obtain the operating status features, environmental monitoring features, and load characteristics output by the feature extraction layer.

[0075] The feature extraction layer in this embodiment of the invention is a multi-channel convolutional feature extraction layer, implemented based on a self-attention mechanism. In one embodiment, the feature extraction layer may include a convolutional layer, a ReLU activation function, and a pooling layer. Specifically, the process is as follows: Operational status information is input to the first channel convolutional feature extraction layer of the feature extraction layer. Features are extracted from the operational status information by the first channel convolutional feature extraction layer to obtain the operational status features output by the first channel convolutional feature extraction layer. Similarly, environmental monitoring information is input to the second channel convolutional feature extraction layer for feature extraction, and load information is input to the third channel convolutional feature extraction layer for feature extraction, resulting in the environmental monitoring features output by the second channel convolutional feature extraction layer and the load-bearing features output by the third channel convolutional feature extraction layer.

[0076] Step 302: Input the operating status characteristics, environmental monitoring characteristics, and load characteristics into the data dimension processing layer for dimension adjustment to obtain the M*4 dimensional data characteristics output by the data dimension processing layer.

[0077] Optionally, the equipment monitoring system inputs the operating status characteristics, environmental monitoring characteristics, and load characteristics to the data dimension processing layer for dimension adjustment, obtaining M*4 dimensional data characteristics output by the data dimension processing layer, where M is determined based on the dimension of the current acquisition time t. In one embodiment, the current acquisition time t includes t1, t2, t3, and t4. The current acquisition time t1 corresponds to operating status characteristic a1, environmental monitoring characteristic a2, and load characteristic a3; the current acquisition time t2 corresponds to operating status characteristic b1, environmental monitoring characteristic b2, and load characteristic b3; the current acquisition time t3 corresponds to operating status characteristic c1, environmental monitoring characteristic c2, and load characteristic c3; and the current acquisition time t4 corresponds to operating status characteristic d1, environmental monitoring characteristic d2, and load characteristic d3. Therefore, the M*4 dimensional data characteristics can be represented as {(t1*a1*a2*a3), (t2*b1*b2*b3), (t3*c1*c2*c3), (t4*d1*d2*d3)}.

[0078] Step 303: Input the M*4 dimensional data features into the equipment status prediction layer to obtain the equipment status information output by the equipment status prediction layer.

[0079] Optionally, the equipment monitoring system inputs M*4 dimensional data features into the equipment status prediction layer to obtain the equipment status information output by the equipment status prediction layer. The equipment status information includes whether the target power supply key equipment is in normal operation or in an abnormal state. An abnormal state means that the target power supply key equipment has failed.

[0080] Step 304: Input the M*4 dimensional data features into the fault type determination layer to obtain the equipment fault type output by the fault type determination layer.

[0081] Furthermore, the equipment monitoring system inputs M*4 dimensional data features to the fault type determination layer, obtaining the fault type output by the fault type determination layer. Fault types include mechanical faults, electrical faults, abnormal temperature, communication faults, software faults, and environmental factors. Mechanical faults include bearing damage, gear wear or breakage, and motor failure. Electrical faults include short circuits or open circuits, insulation aging or breakdown, and current overload. Abnormal temperature includes overheating (causing equipment protection shutdown) and cooling system failure. Communication faults include signal loss between the control system and the equipment, and data mistransmission. Software faults include control system program crashes and incorrect parameter settings. Environmental factors include excessive humidity leading to corrosion, and dust or pollutants affecting equipment operation.

[0082] Step 304: Input the M*4 dimensional data features into the fault occurrence probability prediction layer to obtain the fault occurrence probability output by the fault occurrence probability prediction layer.

[0083] Step 305: Determine the equipment status information, the type of equipment failure, and the probability of failure as equipment monitoring information.

[0084] Furthermore, the equipment monitoring system inputs the M*4 dimensional data features into the fault occurrence probability prediction layer to obtain the fault occurrence probability output by the fault occurrence probability prediction layer.

[0085] Furthermore, the equipment monitoring system defines equipment status information, equipment fault type, and fault occurrence probability as equipment monitoring information. In one embodiment, the equipment monitoring information is {equipment status information: normal operating status; equipment fault type; fault occurrence probability}, indicating that the critical power supply equipment is normal and has a very low probability of malfunction. Equipment monitoring information is {equipment status information: normal operating status; equipment fault type: overheating; fault occurrence probability: 20%}, indicating that the critical power supply equipment is normal, but there is a 20% probability that overheating will cause malfunction. Equipment monitoring information is {equipment status information: abnormal status; equipment fault type: bearing damage; fault occurrence probability: 100%}, indicating that the critical power supply equipment has experienced a mechanical fault due to bearing damage.

[0086] This invention uses an equipment fault monitoring model to monitor each key power supply equipment in a power system, enabling monitoring of these equipment in different working environments. This allows for the identification of fault types in these key power supply equipment under different working conditions, thus improving the fault identification capability of these equipment.

[0087] In one embodiment, the data dimension processing layer includes a first fully connected layer, a second fully connected layer, two residual convolutional layers, and a feedforward network layer. The first and second fully connected layers are used to fully connect the data features. The two residual convolutional layers are used to align and concatenate the data features after the full connection by the fully connected layers. The feedforward network layer is used to map the data features concatenated by the two residual convolutional layers to time.

[0088] Step 302 inputs the operating status characteristics, environmental monitoring characteristics, and load-bearing characteristics into the data dimension processing layer for dimension adjustment, obtaining the M*4 dimensional data characteristics output by the data dimension processing layer, including:

[0089] Step 3021: Input the operating status features and environmental monitoring features into the first fully connected layer, and perform a full connection on the operating status features and environmental monitoring features based on the first fully connected layer to obtain the first fully connected features output by the first fully connected layer.

[0090] Step 3022: Input the operating status features and load characteristics into the second fully connected layer, and perform full connection on the operating status features and load characteristics based on the second fully connected layer to obtain the second fully connected features output by the second fully connected layer.

[0091] In this embodiment of the invention, the fully connected layer uses convolutional layers and pooling layers. During the fully connected processing, data features are used as input to the convolutional layer, and a bias is added to improve the generalization ability of the features. Then, the features are activated by a non-linear activation function to perform full connectivity of the features, resulting in fully connected features. Therefore, the fully connected process of the fully connected features is as follows: .

[0092] in, Indicates fully connected features. Represents the nonlinear activation function of the Rectified Linear Unit (ReLU). This represents the kernel weight matrix of the convolutional layer. This represents the data features input to the convolutional layer, where n is determined based on the number of data features input to the convolutional layer. This represents the bias vector of the nonlinear activation function.

[0093] Therefore, the equipment monitoring system inputs the operating status characteristics and environmental monitoring characteristics into the first fully connected layer. Based on the first fully connected layer, the operating status characteristics and environmental monitoring characteristics are fully connected to obtain the first fully connected feature output by the first fully connected layer. Therefore, the full connection process of the first fully connected feature is as follows:

[0094]

[0095] in, This represents the first fully connected feature. Indicates the characteristics of the operating status. This indicates the characteristics of environmental monitoring.

[0096] Furthermore, the equipment monitoring system inputs the operating status characteristics and load characteristics to the second fully connected layer. Based on the second fully connected layer, a full connection is performed on the operating status characteristics and load characteristics to obtain the second fully connected characteristics output by the second fully connected layer. Therefore, the full connection process of the second fully connected characteristics is as follows:

[0097]

[0098] in, This represents the second fully connected feature. Indicates the characteristics of the operating status. This indicates the characteristics of environmental monitoring.

[0099] Step 3023: Input the first fully connected feature and the second fully connected feature into two residual convolutional layers, and perform residual concatenation on the first fully connected feature and the second fully connected feature based on the two residual convolutional layers to obtain the residual spliced ​​feature output by the two residual convolutional layers.

[0100] In this embodiment of the invention, the two residual convolutional layers are constructed based on the sigmoid residual activation function and the tanh residual activation function. Therefore, in the process of residual connection, the output of the sigmoid residual activation function and the output of the tanh residual activation function are multiplied together to obtain the residual concatenation feature output by the two residual convolutional layers.

[0101] Optionally, the equipment monitoring system inputs the first fully connected feature and the second fully connected feature into two residual convolutional layers, performs residual concatenation on the first and second fully connected features using the sigmoid residual activation function to obtain the first residual connected feature, and performs residual concatenation on the first and second fully connected features using the tanh residual activation function to obtain the second residual connected feature.

[0102] The specific process of the sigmoid residual activation function is as follows:

[0103]

[0104] The specific process of processing the tanh residual activation function is as follows:

[0105]

[0106] in, Indicates the first residual connectivity feature. The linear matrix representing the sigmoid residual activation function is... This represents the bias vector of the sigmoid residual activation function. This represents the sigmoid residual activation function. This indicates the second residual connectivity feature. The linear matrix representing the tanh residual activation function is... This represents the bias vector of the tanh residual activation function. This represents the tanh residual activation function.

[0107] Furthermore, the equipment monitoring system will incorporate the first residual connection characteristic. Second residual connection feature Multiplying them together yields the residual concatenated feature H from the outputs of the two residual convolutional layers.

[0108] Step 3024: Input the residual stitching features into the feedforward network layer. Based on the feedforward network layer, perform a linear mapping between the residual stitching features and the current acquisition time t to obtain the M*4 dimensional data features output by the feedforward network layer.

[0109] In this embodiment of the invention, the feedforward network layer is processed based on the sigmoid linear activation function. Therefore, the device monitoring system inputs the residual stitching features into the feedforward network layer. The feedforward network layer then performs a linear mapping between the residual stitching features and the current acquisition time t to obtain the M*4 dimensional data features output by the feedforward network layer. The processing procedure of the feedforward network layer in this embodiment of the invention is as follows:

[0110]

[0111] in, Represents M*4 dimensional data features. This represents the sigmoid linear activation function. The coefficients representing a linear relationship. This indicates the bias in a linear relationship.

[0112] The following describes the power supply key equipment monitoring system provided in the embodiments of the present invention. The power supply key equipment monitoring system described below can be referred to in correspondence with the power supply key equipment monitoring method described above.

[0113] Please see Figure 2 , Figure 2 This is a structural diagram of a power supply critical equipment monitoring system provided in an embodiment of the present invention. The power supply critical equipment monitoring system includes:

[0114] The acquisition module 210 is used to collect real-time status data of the target power supply key equipment based on sensors; the target power supply key equipment is any power supply key equipment in the power system, and the real-time status data includes operating status information, environmental monitoring information and load information;

[0115] The receiving module 220 is used to control the target edge computing node to receive real-time status data;

[0116] The intelligent prediction module 230 is used to input operating status information, environmental monitoring information, and load information into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model. The equipment fault monitoring model is trained based on sample operating status information, sample environmental monitoring information, sample load information, and their corresponding alarm information and alarm category labels.

[0117] The equipment status monitoring module 240 is used to monitor the operating status of key power supply equipment based on equipment monitoring information.

[0118] The power supply critical equipment monitoring system provided in this embodiment of the invention monitors each power supply critical equipment in the power system by training an equipment fault monitoring model based on operating status information, environmental monitoring information, load information and their corresponding alarm information and alarm category labels. This enables monitoring of the power supply critical equipment in different working environments, thus identifying the types of faults that occur in the power supply critical equipment in different working environments and improving the fault identification capability of the power supply critical equipment.

[0119] Please see Figure 3 , Figure 3 An embodiment diagram of an electronic device provided in accordance with the present invention. For example... Figure 3 As shown, this embodiment of the invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, it performs the following steps:

[0120] Real-time status data of key power supply equipment is collected based on sensors; the key power supply equipment is any key power supply equipment in the power system, and the real-time status data includes the operating status information, environmental monitoring information and load information at the current acquisition time t;

[0121] Control the target edge computing node to receive real-time status data;

[0122] The operating status information, environmental monitoring information, and load information are input into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model. The equipment fault monitoring model is trained based on sample operating status information, sample environmental monitoring information, sample load information, and their corresponding alarm information and alarm category labels.

[0123] The operating status of key power supply equipment is monitored based on equipment monitoring information.

[0124] Please see Figure 4 , Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with an embodiment of the present invention is shown. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, it performs the following steps:

[0125] Real-time status data of key power supply equipment is collected based on sensors; the key power supply equipment is any key power supply equipment in the power system, and the real-time status data includes the operating status information, environmental monitoring information and load information at the current acquisition time t;

[0126] Control the target edge computing node to receive real-time status data;

[0127] The operating status information, environmental monitoring information, and load information are input into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model. The equipment fault monitoring model is trained based on sample operating status information, sample environmental monitoring information, sample load information, and their corresponding alarm information and alarm category labels.

[0128] The operating status of key power supply equipment is monitored based on equipment monitoring information.

[0129] On the other hand, embodiments of the present invention also provide a computer program product, which includes a computer program that can be stored on a computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the power supply critical equipment monitoring method provided by the above methods, the method including:

[0130] Real-time status data of key power supply equipment is collected based on sensors; the key power supply equipment is any key power supply equipment in the power system, and the real-time status data includes the operating status information, environmental monitoring information and load information at the current acquisition time t;

[0131] Control the target edge computing node to receive real-time status data;

[0132] The operating status information, environmental monitoring information, and load information are input into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model. The equipment fault monitoring model is trained based on sample operating status information, sample environmental monitoring information, sample load information, and their corresponding alarm information and alarm category labels.

[0133] The operating status of key power supply equipment is monitored based on equipment monitoring information.

[0134] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for monitoring key power supply equipment, characterized in that, include: Real-time status data of key power supply equipment for the target were collected based on sensors. The target power supply key equipment is any power supply key equipment in the power system, and the real-time status data includes the operating status information, environmental monitoring information and load information at the current acquisition time t. The target edge computing node receives the real-time status data. The operating status information, the environmental monitoring information, and the load information are input into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model. The equipment fault monitoring model is trained based on sample operating status information, sample environmental monitoring information, sample load information and their corresponding alarm information and alarm category labels. Monitor the operating status of the target power supply key equipment based on the equipment monitoring information; The control target edge computing node receives the real-time status data, including: Based on the data processing capability, current data processing volume, and pending data volume of each edge computing node in the data transmission network, the data processing time of each edge computing node is determined. Based on the channel power gain and data transmission rate of each edge computing node, the data transmission delay time of each edge computing node is determined. Based on the data processing time and data transmission latency of each edge computing node, the data transmission consumption time of each edge computing node is determined. The target edge computing node in the data transmission network is determined based on the data transmission consumption time of each edge computing node; The formula for determining the data processing time of each edge computing node is as follows: ； in, Indicates data processing time. Indicates data processing capability. Indicates the current amount of data being processed. Indicates the amount of data to be processed. and Indicates the preset percentage weight; The formula for calculating the data transmission latency of each edge computing node is as follows: ； in, B represents the data transmission delay time, and K represents the channel bandwidth of the edge computing node. Indicates channel power gain. Indicates the data transmission rate. Indicates the channel gain factor; The formula for determining the data transmission consumption time of each edge computing node is as follows: ； in, Indicates the time consumed by data transmission. and This indicates the weight of the time spent.

2. The method for monitoring key power supply equipment according to claim 1, characterized in that, The equipment fault monitoring model includes a feature extraction layer, a data dimension processing layer, an equipment status prediction layer, a fault type determination layer, and a fault occurrence probability prediction layer. The step of inputting the operating status information, the environmental monitoring information, and the load information into the equipment fault monitoring model to obtain the equipment monitoring information output by the equipment fault monitoring model includes: The operating status information, the environmental monitoring information, and the load-bearing information are input into the feature extraction layer for feature extraction, resulting in the operating status features, environmental monitoring features, and load-bearing features output by the feature extraction layer. The operating status characteristics, the environmental monitoring characteristics, and the load characteristics are input into the data dimension processing layer for dimension adjustment to obtain M*4 dimensional data characteristics output by the data dimension processing layer; wherein, M is determined based on the dimension of the current acquisition time t; The M*4 dimensional data features are input into the device state prediction layer to obtain the device state information output by the device state prediction layer. The M*4 dimensional data features are input into the fault type determination layer to obtain the equipment fault type output by the fault type determination layer. The M*4 dimensional data features are input into the fault occurrence probability prediction layer to obtain the fault occurrence probability output by the fault occurrence probability prediction layer. The device status information, the type of device failure, and the probability of failure occurrence are determined as the device monitoring information.

3. The method for monitoring key power supply equipment according to claim 2, characterized in that, The data dimension processing layer includes a first fully connected layer, a second fully connected layer, two residual convolutional layers, and a feedforward network layer; the step of inputting the operating status features, the environmental monitoring features, and the load-bearing features into the data dimension processing layer for dimension adjustment to obtain the M*4 dimensional data features output by the data dimension processing layer includes: The operating status features and the environmental monitoring features are input into the first fully connected layer. Based on the first fully connected layer, the operating status features and the environmental monitoring features are fully connected to obtain the first fully connected features output by the first fully connected layer. The operating status feature and the load-bearing feature are input to the second fully connected layer. The operating status feature and the load-bearing feature are fully connected based on the second fully connected layer to obtain the second fully connected feature output by the second fully connected layer. The first fully connected feature and the second fully connected feature are input into the two residual convolutional layers. The first fully connected feature and the second fully connected feature are residually concatenated based on the two residual convolutional layers to obtain the residual concatenated feature output by the two residual convolutional layers. The residual stitching feature is input into the feedforward network layer, and the residual stitching feature and the current acquisition time t are linearly mapped based on the feedforward network layer to obtain the M*4 dimensional data feature output by the feedforward network layer.

4. The method for monitoring key power supply equipment according to claim 1, characterized in that, The training steps for the equipment fault monitoring model include: Collect information on sample operation status, sample environment monitoring, and sample load. Based on the sample collection time, sample operation status information, sample environment monitoring information, and sample load information, multidimensional data features of the sample are generated. Based on the multidimensional data features of the samples and their corresponding alarm information and alarm category labels, a preset neural network model is trained to obtain the equipment fault monitoring model.

5. The method for monitoring key power supply equipment according to claim 4, characterized in that, The total loss function of the equipment fault monitoring model Including classification loss function Weighted loss function and uncertainty loss function ; The total loss function Represented as: ； The classification loss function Represented as: ； The weighted loss function Represented as: ； The uncertainty loss function Represented as: ； The total loss function The final form is expressed as: ； Where N represents the number of samples and C represents the number of categories. This represents the alarm category label of the c-th category for the i-th sample. Let represent the probability predicted by the model for the c-th class of the i-th sample. This represents the weight for the i-th sample, which is dynamically set based on the sample's running state information. This represents the prediction variance for the i-th sample. This represents the first preset hyperparameter. This indicates the second preset hyperparameter.

6. A monitoring system for key power supply equipment, characterized in that, It includes a module for performing the power supply critical equipment monitoring method according to any one of claims 1 to 5.

7. An electronic device, characterized in that, include: Memory, used to store computer software programs; A processor is used to read and execute the computer software program, thereby implementing the power supply critical equipment monitoring method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer software program, which, when executed by a processor, implements the power supply critical equipment monitoring method as described in any one of claims 1 to 5.

Images