Power equipment disposal method, device, terminal and storage medium based on data middle platform

Through the data middle platform and neural network model, based on the operating data and parameters of the power equipment, the probability values ​​for maintenance and replacement are generated, which solves the problem of inaccurate power equipment disposal management in the existing technology, and realizes accurate and timely disposal of power equipment. Improves grid safety and equipment life.

CN118589693BActive Publication Date: 2024-11-29MINGXING ELECTRIC SICHUAN
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411067615.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-11-29
Estimated Expiration
2044-08-06

AI Technical Summary

Technical Problem

The existing power equipment disposal and management methods rely on regular inspections and experience maintenance, resulting in inaccurate judgment of the operating conditions of power equipment, which may lead to untimely disposal and create grid safety hazards.

Method used

The power equipment disposal method based on the data middle platform and neural network model is adopted to collect power equipment operation data through data acquisition equipment, combine equipment parameters, generate probability values ​​for maintenance and replacement, and make handling decisions based on the probability threshold to achieve predictive maintenance.

Benefits of technology

It improves the accuracy and timeliness of power equipment disposal, reduces grid safety hazards, extends the service life of the equipment, and reduces operating costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118589693B_ABST
    Figure CN118589693B_ABST
Patent Text Reader

Abstract

The present application discloses a method, device, terminal and storage medium for handling electric equipment based on a data middle station, which is applied to a data middle station, wherein the data middle station includes a data acquisition device and a server, wherein the data acquisition device and the server are both in the same network; the method includes: collecting the current operation data of the electric equipment through the data acquisition device; obtaining the equipment parameters of the electric equipment from the server; generating a first probability value and a second probability value according to the equipment parameters and the current operation data through an electric equipment judgment model; and processing the electric equipment based on the comparison result between the first probability value, the second probability value and a preset probability threshold. The present application aims to accurately provide adaptive disposal measures for the current status of the electric equipment through a data middle station and a neural network model, so as to dispose of the electric equipment in a timely manner.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, terminal and storage medium for handling electric power equipment based on a data middle station. Background Art

[0002] The disposal management of power equipment involves a series of activities such as equipment maintenance, overhaul, upgrade, decommissioning and scrapping, aiming to ensure the safe and reliable operation of the equipment while maximizing the equipment service life and reducing operating costs.

[0003] However, the current power equipment disposal management methods mainly include the following defects: the maintenance plan may rely too much on regular inspections rather than being based on the actual operating status of the equipment; the current management method may rely more on experienced technicians for preventive maintenance rather than predictive maintenance based on the equipment's operating data. This may lead to inaccurate judgments on the operating status of power equipment, especially in the diagnosis and maintenance of relatively complex power equipment, and may lead to untimely disposal of power equipment, thereby creating safety hazards for the power grid. Summary of the invention

[0004] The main purpose of this application is to provide a method, device, terminal and storage medium for handling power equipment based on a data middle platform, aiming to accurately provide adaptive handling measures for the current status of the power equipment through the data middle platform and the neural network model, so as to handle the power equipment in a timely manner.

[0005] To achieve the above-mentioned purpose, the present application provides a method for handling electric equipment based on a data middle station, which is applied to the data middle station, wherein the data middle station includes a data acquisition device and a server, and the data acquisition device and the server are both in the same network;

[0006] The method comprises:

[0007] Collecting current operation data of the power equipment through the data acquisition device;

[0008] Acquiring equipment parameters of the electric equipment from the server;

[0009] Generate a first probability value and a second probability value according to the equipment parameters and the current operation data through the electric equipment determination model, wherein the first probability value is used to characterize the probability that the electric equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to characterize the probability that the electric equipment is in a condition that requires immediate replacement;

[0010] The electric power device is processed based on a comparison result among the first probability value, the second probability value and a preset probability threshold.

[0011] Specifically, the electric power equipment determination model includes an input layer, a hidden layer and an output layer, and the output layer has two neurons;

[0012] The generating of the first probability value and the second probability value according to the equipment parameters and the current operation data by the electric power equipment determination model includes:

[0013] Generate a one-dimensional feature vector corresponding to the power equipment according to the equipment parameters and the current operation data through the input layer;

[0014] Generate a one-dimensional output vector according to the one-dimensional feature vector through the hidden layer;

[0015] The first probability value and the second probability value are generated according to the one-dimensional output vector through the two neurons of the output layer.

[0016] Specifically, the step of generating a one-dimensional feature vector corresponding to the power equipment according to the equipment parameters and the current operation data through the input layer includes:

[0017] Combining the device parameters and the current operation data into an input array;

[0018] The input array is input to the input layer, and the one-dimensional feature vector is obtained as output.

[0019] Specifically, the hidden layer includes a first fully connected layer, a second fully connected layer, and a third fully connected layer, the first fully connected layer has a first number of neurons, the second fully connected layer has a second number of neurons, and the third fully connected layer has a third number of neurons, the second number is half of the first number, and the third number is half of the second number;

[0020] The step of generating a one-dimensional output vector according to the one-dimensional feature vector through the hidden layer includes:

[0021] Input the one-dimensional feature vector to the first fully connected layer, and output a first vector;

[0022] Input the first vector to the second fully connected layer, and output a second vector;

[0023] The second vector is input into the third fully connected layer, and output to obtain the one-dimensional output vector.

[0024] Specifically, the hidden layer includes a fourth fully connected layer, a first random dropout layer, a fifth fully connected layer, a second random dropout layer, a sixth fully connected layer, and a batch normalization layer;

[0025] The step of generating a one-dimensional output vector according to the one-dimensional feature vector through the hidden layer includes:

[0026] Inputting the one-dimensional feature vector into the fourth fully connected layer, and outputting a third vector;

[0027] Inputting the third vector to the first random dropout layer, and outputting an output vector corresponding to the third vector, wherein a value of the third vector is the same as a value of the output vector corresponding to the third vector;

[0028] Inputting the output vector corresponding to the third vector into the fifth fully connected layer, and outputting a fourth vector;

[0029] Inputting the fourth vector to the second random dropout layer, and outputting an output vector corresponding to the fourth vector, wherein a value of the fourth vector is the same as a value of the output vector corresponding to the fourth vector;

[0030] Inputting the output vector corresponding to the fourth vector into the sixth fully connected layer, and outputting a fifth vector;

[0031] The fifth vector is normalized by the batch normalization layer to obtain the one-dimensional output vector.

[0032] Specifically, the processing of the electric power device based on the comparison result between the first probability value, the second probability value and a preset probability threshold includes:

[0033] If the first probability value is not less than the preset probability threshold and the second probability value is less than the preset probability threshold, repairing the power equipment according to the actual condition of the power equipment;

[0034] If the first probability value is less than the preset probability threshold and the second probability value is not less than the preset probability threshold, replacing the power equipment;

[0035] If the first probability value and the second probability value are both smaller than the preset probability threshold, the power equipment continues to be used.

[0036] Specifically, after generating the first probability value and the second probability value according to the equipment parameters and the current operation data through the electric power equipment determination model, the method further includes:

[0037] receiving a model update instruction input by a user, and merging the current operation data with the historical operation data of the electric power equipment into updated historical operation data;

[0038] Based on the updated historical operation data and the equipment parameters, the electric power equipment determination model is updated to obtain an updated electric power equipment determination model.

[0039] To achieve the above-mentioned purpose, the present application also provides a power equipment handling device based on a data middle station, which is applied to the data middle station, wherein the data middle station includes a data acquisition device and a server, and the data acquisition device and the server are both in the same network;

[0040] The device comprises:

[0041] The first unit is used to collect current operation data of the power equipment through the data collection device;

[0042] A second unit is used to obtain equipment parameters of the power equipment from the server;

[0043] A third unit is used to generate a first probability value and a second probability value according to the equipment parameters and the current operation data through an electric equipment determination model, wherein the first probability value is used to represent the probability that the electric equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to represent the probability that the electric equipment is in a condition that requires immediate replacement;

[0044] The fourth unit is used to process the power device based on a comparison result among the first probability value, the second probability value and a preset probability threshold.

[0045] To achieve the above-mentioned purpose, the present application also provides a terminal, including a memory storing multiple instructions; the processor loads instructions from the memory to execute the steps in any one of the methods provided in the present application.

[0046] To achieve the above objectives, the present application also provides a storage medium, which stores multiple instructions, and the instructions are suitable for a processor to load to execute the steps in any method provided in the present application.

[0047] The present application provides a method, device, terminal and storage medium for handling power equipment based on a data middle station. First, the current operating data of the power equipment can be collected by the data acquisition device, and the equipment parameters of the power equipment can be obtained from the server; then, a first probability value and a second probability value can be accurately generated according to the equipment parameters and the current operating data through the power equipment judgment model, and the adaptive handling measures of the current status of the power equipment are reflected by the probability values, wherein the first probability value is used to characterize the probability that the power equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to characterize the probability that the power equipment is in a condition that requires immediate replacement; finally, the power equipment is handled based on the comparison result between the first probability value, the second probability value and a preset probability threshold.

[0048] Compared with the existing disposal methods of power equipment, the present application can accurately provide adaptive disposal measures for the current status of the power equipment through the data middle platform and neural network model, so as to dispose of the power equipment in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of a process flow provided for an embodiment of the present application;

[0050] Figure 2 A schematic diagram of the structure of an electric power equipment handling device based on a data middle station provided in an embodiment of the present application;

[0051] Figure 3 A schematic diagram of the structure of a terminal provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0053] Since the current management method may rely more on experienced technicians to perform preventive maintenance rather than predictive maintenance based on the operating data of the equipment, this may lead to inaccurate judgment of the operating status of the power equipment, especially in the diagnosis and maintenance of relatively complex power equipment, and may lead to untimely disposal of the power equipment, thereby creating safety hazards for the power grid. Therefore, the embodiments of the present application provide a power equipment disposal method, device, terminal and storage medium based on a data middle station to solve practical technical problems.

[0054] In some embodiments, the device may be specifically integrated into an electronic device, and the electronic device may be a terminal, a server, or other device.

[0055] In some embodiments, the server may also be implemented in the form of a terminal.

[0056] Among them, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0057] The terminal may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal and the server may be directly or indirectly connected via wired or wireless communication, and this application does not limit this.

[0058] It should be noted that the serial numbers of the following embodiments are not intended to limit the preferred order of the embodiments.

[0059] An embodiment of the present application provides a method for handling power equipment based on a data middle platform. The method can accurately provide adaptive handling measures for the current status of the power equipment through the data middle platform and a neural network model, so as to handle the power equipment in a timely manner.

[0060] A method for handling electric power equipment based on a data middle station is applied to the data middle station, wherein the data middle station includes a data acquisition device and a server, and the data acquisition device and the server are both in the same network.

[0061] Data Middleware Platform is a type of enterprise-level data management and application architecture model that plays a core role in the modern data-driven business environment. The goal of the data middleware platform is to unify and optimize the management and use of data resources to support more efficient and flexible data analysis and business decision-making. Specifically, the data middleware platform has the following features:

[0062] Data integration: The data center integrates and standardizes data from different sources and formats to form a unified data view to eliminate information silos and ensure data consistency and accuracy;

[0063] Data governance: provides a complete set of data management processes, including data quality control, data security, data lifecycle management, etc., to ensure data compliance and availability;

[0064] Data analysis and services: The data center platform supports real-time analysis, predictive analysis and intelligent decision-making by providing pre-processed data, models and algorithms, helping enterprises to respond quickly to market changes;

[0065] Data sharing and reuse: It promotes cross-departmental and cross-system data sharing, reduces duplication of construction, increases data reuse rate, and reduces the cost of data acquisition and processing;

[0066] Flexibility and scalability: The data center adopts a microservice architecture, which can be easily expanded and adjusted to adapt to the rapid development and changes of the business;

[0067] Technology stack: Data middle platform usually involves components such as big data processing technology, data warehouse, data lake, data API, machine learning platform, etc.

[0068] In some embodiments, the data acquisition device can be set in the key part or data storage part of the power equipment in the power grid to facilitate the collection of relevant operating data of the power equipment. The server can be used to store the operating data of the power equipment and analyze and process the operating data.

[0069] like Figure 1 As shown, the specific process of the method can be as follows:

[0070] S110. Collect current operating data of the power equipment through the data acquisition device.

[0071] In some embodiments, the data acquisition equipment can be sensors and measuring devices, such as current transformers (CT), voltage transformers (PT), temperature sensors, vibration sensors, etc., which are used to monitor various parameters of power equipment. The data acquisition equipment can also be a data acquisition unit, such as a data collector, a smart meter, an RTU (remote terminal unit), a PLC (programmable logic controller), etc., which is used to collect sensor data and perform preliminary processing.

[0072] The current operating data of power equipment can be information collected in real time or near real time and used to monitor the working status and performance of the equipment. The current operating data is essential for preventive maintenance, fault diagnosis, efficiency analysis, and system optimization. The following is the data content or part of the data content of the current operating data of power equipment:

[0073] Real-time power output: The power currently being output by the power equipment, which is particularly important for generators and transformers;

[0074] Voltage and Current: Real-time monitoring of line voltage and current, which is important for detecting overload, short circuit or unbalance conditions;

[0075] Temperature: The temperature of key components, such as windings, cores, bearings, etc. High temperatures may indicate overheating or cooling problems;

[0076] Frequency: The frequency of the power supply system, used to monitor the stability of the power grid;

[0077] Load rate: The ratio of the current load of the power equipment to the rated load, used to determine whether the power equipment is overloaded;

[0078] Harmonic analysis: The harmonic content of current or voltage in the power grid, used to assess power quality;

[0079] Insulation resistance: The resistance of the insulating material of an electrical device during operation, used to monitor insulation aging or damage;

[0080] Oil quality analysis: For oil-immersed equipment (such as transformers), the quality of the oil (water content, gas content, acid value, etc.) can reflect the health of the equipment;

[0081] Vibration analysis: The vibration level of power equipment during operation. Abnormal vibration may indicate mechanical failure.

[0082] Noise Level: The noise level of electrical equipment when it is operating. Abnormal noise may indicate potential problems.

[0083] S120: Acquire equipment parameters of the power equipment from the server.

[0084] In some embodiments, the server is used to store the operating data of the power equipment and analyze and process the operating data. The historical operating data of the power equipment before the current moment can be stored in the server and called when needed.

[0085] Equipment parameters of power equipment are a series of technical indicators that describe the performance, specifications and working conditions of the equipment. They are crucial for the correct selection, operation and maintenance, and fault diagnosis of the equipment. Equipment parameters may vary depending on the type of equipment (such as generators, transformers, switchgear, cables, motors, etc.). The following are some common equipment parameters of power equipment:

[0086] Rated Voltage: The highest voltage level that the equipment is designed to operate at;

[0087] Rated Current: The maximum current that a device can withstand under normal working conditions;

[0088] Rated Power: The nominal power output or input capacity of a device;

[0089] Short Circuit Capacity: For equipment such as transformers, it refers to the maximum instantaneous current that can be carried in the event of a short circuit.

[0090] Insulation Class: The highest temperature level that the insulation material of the equipment can withstand;

[0091] Frequency: The frequency of alternating current that the equipment is designed to operate at, such as 50Hz or 60Hz;

[0092] Efficiency: The efficiency of a device in converting energy, usually expressed as a percentage;

[0093] Power Factor: The ratio of the active power consumed by the equipment to the apparent power, reflecting the efficiency of the equipment in using electrical energy;

[0094] Temperature Rise: The increase in internal temperature relative to the ambient temperature when the equipment is running at full load.

[0095] Cooling Method: The cooling method of the equipment, such as natural cooling, air cooling, water cooling, etc.

[0096] Protection Rating: The protection level of the device casing, such as IPXX, indicating its dust and water resistance.

[0097] S130. Generate a first probability value and a second probability value through an electric power equipment determination model according to the equipment parameters and the current operating data, wherein the first probability value is used to characterize the probability that the electric power equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to characterize the probability that the electric power equipment is in a condition that requires immediate replacement.

[0098] In some embodiments, the power equipment determination model can be stored in the data center in the form of an application. When using it, the staff can input operation instructions to the data center through a human-computer interaction device to enable the power equipment determination model and perform a determination action on the current status of the power equipment.

[0099] In some embodiments, the electric power equipment determination model includes an input layer, a hidden layer, and an output layer, and the output layer has two neurons.

[0100] The input layer is usually the first layer of a neural network model. Its main function is to receive and pass raw input data to subsequent layers of the network for processing. The number of nodes in the input layer is usually equal to the number of features of the input data, and the shape of the input layer reflects the structure of the input data. For one-dimensional data, the input layer will be a vector.

[0101] Hidden layers are neural network layers located between the input layer and the output layer. They do not interact directly with the input or output data, but are responsible for processing and transforming the input data to extract features and learn patterns in the data. Hidden layers are the key to the ability of neural networks to perform complex tasks. Through multi-level nonlinear transformations, the network can learn the deep representation and features of the data.

[0102] The output layer is usually the last component of the neural network model, responsible for converting the final calculation results of the network into interpretable outputs, which usually correspond to the specific goals of the task to be solved by the model. The design and configuration of the output layer depends on the task type of the model (such as classification, regression, sequence generation, etc.).

[0103] In some embodiments, the generating of the first probability value and the second probability value by the power equipment determination model according to the equipment parameters and the current operation data includes the following steps S121 to S123:

[0104] S121. Generate a one-dimensional feature vector corresponding to the power equipment through the input layer according to the equipment parameters and the current operating data.

[0105] In some embodiments, the step of generating a one-dimensional feature vector corresponding to the power equipment according to the equipment parameters and the current operation data through the input layer includes the following steps S1211 to S1212:

[0106] S1211. Combine the device parameters and the current operation data into an input array.

[0107] In some embodiments, the current operating data may be a data set corresponding to the currently monitored data point, such as temperature, current, voltage, etc., and the device parameters may be a data set of inherent properties of the device, such as model, maximum power, rated voltage, etc.

[0108] Specifically, first determine the specific structure and dimensions of the two data sets, and then splice them as needed. Assuming that the device parameters and current operating data are both numerical and can be represented as one-dimensional or multi-dimensional arrays, the Python numpy library can be called to merge the two data sets into one array, namely the input array.

[0109] S1212. Input the input array to the input layer, and output the one-dimensional feature vector.

[0110] In some embodiments, the number of nodes in the input layer is equal to the total number of features, including the number of features of device parameters and current operating data. Assuming there are n features in total, the input layer will have n nodes.

[0111] S122. Generate a one-dimensional output vector through the hidden layer according to the one-dimensional feature vector.

[0112] In some embodiments, the hidden layer includes a first fully connected layer, a second fully connected layer, and a third fully connected layer, the first fully connected layer has a first number of neurons, the second fully connected layer has a second number of neurons, and the third fully connected layer has a third number of neurons, the second number is half of the first number, and the third number is half of the second number.

[0113] The fully connected layer (FC layer) is a common structure in neural networks. The characteristic of the fully connected layer is that each neuron is connected to all neurons in the previous layer, forming complete connectivity. Usually, in a fully connected layer, the input of each neuron is obtained by weighting the output of all neurons in the previous layer through the weight matrix.

[0114] Neurons are the basic units of artificial neural networks, inspired by nerve cells in biological nervous systems. In artificial neural networks, neurons (also called nodes) are responsible for receiving inputs, processing these inputs, and generating outputs.

[0115] Specifically, the step of generating a one-dimensional output vector according to the one-dimensional feature vector through the hidden layer includes steps A1 to A3 as shown below:

[0116] A1. Input the one-dimensional feature vector into the first fully connected layer, and output a first vector.

[0117] In some embodiments, the first fully connected layer may be a fully connected layer having 128 neurons, and the activation function used is a ReLU function.

[0118] ReLU function ReLU function, full name Rectified Linear Unit, is an activation function widely used in deep learning. The mathematical definition of ReLU function is: for any input value 𝑥, the output of ReLU function 𝑓(𝑥) is defined as: 𝑓(𝑥)=max(0,𝑥), which means that if the input is positive, the function returns 𝑥 itself; if the input is negative, it returns 0. The graph of ReLU function is a piecewise linear function with a slope of 1 when 𝑥>0 and 0 when 𝑥≤0.

[0119] A2. Input the first vector to the second fully connected layer, and output a second vector.

[0120] In some embodiments, the second fully connected layer may be a fully connected layer having 64 neurons, and the activation function used is a ReLU function.

[0121] A3. Input the second vector to the third fully connected layer, and output the one-dimensional output vector.

[0122] In some embodiments, the third fully connected layer may be a fully connected layer with 32 neurons, and the activation function used is the ReLU function.

[0123] In some embodiments, the hidden layer includes a fourth fully connected layer, a first random dropout layer, a fifth fully connected layer, a second random dropout layer, a sixth fully connected layer, and a batch normalization layer.

[0124] Dropout Layer is a regularization technique commonly used in training deep learning models to prevent neural networks from overfitting.

[0125] Batch Normalization Layer is a technique widely used in deep learning for regularization and accelerated training. Its function is to reduce the so-called "internal covariate shift", that is, during the network training process, the input distribution of the intermediate layer changes, causing the subsequent layers to constantly adapt to these changes, thereby reducing the learning speed.

[0126] Specifically, the step of generating a one-dimensional output vector through the hidden layer according to the one-dimensional feature vector includes steps B1 to B6 as shown below:

[0127] B1. Input the one-dimensional feature vector to the fourth fully connected layer, and output a third vector.

[0128] B2. Input the third vector to the first random inactivation layer, and output an output vector corresponding to the third vector, wherein the value of the third vector is the same as the value of the output vector corresponding to the third vector.

[0129] In some embodiments, during the training phase of the power equipment determination model, the first random deactivation layer randomly selects a certain proportion of neurons and sets their outputs to 0. This process is called "deactivation" or "discarding". The discarded neurons will not participate in the forward propagation and backpropagation processes, that is, they do not play a role in the current training iteration. This ratio is usually a hyperparameter, called the dropout rate, which is generally set between 0.2 and 0.5. In the prediction phase, the value of the output vector will remain unchanged and will be passed directly to the next neural network layer.

[0130] B3. Input the output vector corresponding to the third vector to the fifth fully connected layer, and output a fourth vector.

[0131] B4. Input the fourth vector to the second random inactivation layer, and output an output vector corresponding to the fourth vector, wherein the value of the fourth vector is the same as the value of the output vector corresponding to the fourth vector.

[0132] Continuing from the above embodiment, when the power equipment determination model is in the training phase, the second random deactivation layer randomly selects a certain proportion of neurons and sets the output of the selected neurons to 0. In the prediction phase, the value of the output vector remains unchanged and is directly passed to the next neural network layer.

[0133] B5. Input the output vector corresponding to the fourth vector to the sixth fully connected layer, and output a fifth vector.

[0134] B6. Normalize the fifth vector through the batch normalization layer to obtain the one-dimensional output vector.

[0135] Continuing with the above embodiment, the fifth vector may be subtracted from the moving average mean, the difference obtained is divided by the square root of the moving average variance, and finally multiplied by a scale parameter γ (scale parameter), and the obtained product is added with a shift parameter β (shift parameter) to obtain the one-dimensional output vector.

[0136] The scaling parameter γ is a learnable parameter used to scale the normalized features to restore the network's expressiveness. Specifically, for each feature, the scaling parameter γ is a vector with the same dimension as the feature, and the scaling parameter γ is multiplied element-by-element with the normalized feature values, so that the values ​​of these features can be amplified or reduced.

[0137] The offset parameter β is also a learnable parameter used to offset the normalized data after the scaling operation to adjust the mean of the output. The offset parameter β allows the model to further fine-tune the distribution of the output based on the normalized data to make it more suitable for subsequent layers of the network or the final prediction task. Specifically, the offset parameter β is added to the scaled feature values ​​in an additive form. For each feature, the offset parameter β is a vector with the same dimension as the feature. The offset parameter β is added element by element to the feature values ​​after the scaling operation, so that the values ​​of these features can be shifted up or down.

[0138] S123. Generate the first probability value and the second probability value according to the one-dimensional output vector through the two neurons of the output layer.

[0139] In some embodiments, the two neurons of the output layer correspond to the generated first probability value and the second probability value respectively, and the activation function used by the output layer is a Softmax function.

[0140] The Softmax function is an activation function of the output layer, which is mainly used in multi-classification problems. It can convert the output of the last layer of the neural network into a probability distribution so that the output can be interpreted as the probability of each category. The Softmax function converts a set of arbitrary real numbers into a probability distribution, that is, the output value is between 0 and 1, and the sum of all output values ​​is 1.

[0141] Specifically, through the output layer, the one-dimensional output vector is converted into two distribution probabilities corresponding to the numerical values, including the following two distribution probabilities:

[0142] The first probability value is used to characterize the probability that the electric equipment is in a condition that requires repair but does not need to be replaced, and the second probability value is used to characterize the probability that the electric equipment is in a condition that requires immediate replacement.

[0143] S140: Process the electric power device based on a comparison result among the first probability value, the second probability value, and a preset probability threshold.

[0144] In some embodiments, the processing of the electric power device based on the comparison result between the first probability value, the second probability value and a preset probability threshold includes the following specific implementation process:

[0145] If the first probability value is not less than the preset probability threshold and the second probability value is less than the preset probability threshold, repairing the power equipment according to the actual condition of the power equipment;

[0146] If the first probability value is less than the preset probability threshold and the second probability value is not less than the preset probability threshold, replacing the power equipment;

[0147] If the first probability value and the second probability value are both smaller than the preset probability threshold, the power equipment continues to be used.

[0148] In some embodiments, a handling decision for the power device is made according to the probability value output by the output layer:

[0149] If the distribution probability of the numerical value used to characterize that the electrical equipment is in need of repair but not replacement is high, it may indicate that the equipment needs to be repaired but not replaced; if the distribution probability of the numerical value used to characterize that the electrical equipment is in need of immediate replacement is high, it may indicate that the equipment needs to be replaced immediately.

[0150] In some embodiments, after generating the first probability value and the second probability value according to the equipment parameters and the current operating data through the electric power equipment determination model, the method further includes steps T110 to T120 as shown below:

[0151] T110. Receive a model update instruction input by a user, and merge the current operation data with the historical operation data of the power equipment into updated historical operation data.

[0152] T120. Based on the updated historical operation data and the equipment parameters, update the electric power equipment determination model to obtain an updated electric power equipment determination model.

[0153] In some embodiments, the features most relevant to the state of the power equipment are selected from the equipment parameters and the updated historical operating data. This may include equipment type, age, temperature, voltage, current, etc., removing or filling missing values, handling outliers, and scaling all features to ensure that they are on a similar scale, which helps the learning process of the neural network. The data set consisting of the equipment parameters and the updated historical operating data is divided into a training set, a validation set, and a test set. The training set data is used for model training, and the performance on the validation set is monitored to prevent overfitting. Hyperparameters such as learning rate, batch size, number of layers, number of neurons, etc. are adjusted according to the performance feedback on the validation set. Finally, the test set is used to evaluate the final performance of the model to ensure the generalization ability of the model. The updated power equipment judgment model is obtained through the above operations.

[0154] From the above, it can be seen that the embodiment of the present application can first collect the current operating data of the power equipment through the data acquisition device, and obtain the equipment parameters of the power equipment from the server; then, through the power equipment judgment model, according to the equipment parameters and the current operating data, accurately generate a first probability value and a second probability value, and use the probability values ​​to reflect the adaptive disposal measures of the current status of the power equipment, wherein the first probability value is used to characterize the probability that the power equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to characterize the probability that the power equipment is in a condition that requires immediate replacement; finally, based on the comparison result between the first probability value, the second probability value and the preset probability threshold, the power equipment is disposed of.

[0155] In summary, compared with the existing disposal methods of power equipment, the embodiments of the present application can accurately provide adaptive disposal measures for the current status of the power equipment through the data middle platform and the neural network model, so as to dispose of the power equipment in a timely manner.

[0156] In order to better implement the above method, the embodiment of the present application also provides a power equipment disposal device based on a data middle station, which can be integrated in an electronic device, which can be a terminal, a server, etc. Among them, the terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, a personal computer, etc.; the server can be a single server or a server cluster composed of multiple servers.

[0157] For example, in this embodiment, the method of the embodiment of the present application will be described in detail by taking the specific integration of an electric power equipment disposal device based on a data middle station in a terminal as an example.

[0158] For example, Figure 2 As shown, the power equipment handling device based on the data middle station may include a first unit 201, a second unit 202, a third unit 203 and a fourth unit 204, which are applied to the data middle station, and the data middle station includes a data acquisition device and a server, and the data acquisition device and the server are both in the same network;

[0159] The device comprises:

[0160] The first unit 201 is used to collect current operation data of the power equipment through the data collection device;

[0161] The second unit 202 is used to obtain the equipment parameters of the power equipment from the server;

[0162] The third unit 203 is used to generate a first probability value and a second probability value according to the equipment parameters and the current operation data through the power equipment determination model, wherein the first probability value is used to represent the probability that the power equipment is in a state where maintenance is required but replacement is not required, and the second probability value is used to represent the probability that the power equipment is in a state where immediate replacement is required;

[0163] The fourth unit 204 is configured to process the electric power device based on a comparison result among the first probability value, the second probability value, and a preset probability threshold.

[0164] In specific implementation, the above units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units can refer to the previous method embodiments, which will not be repeated here.

[0165] From the above, it can be seen that the embodiments of the present application can accurately provide adaptive disposal measures for the current status of the power equipment through the data middle platform and the neural network model, so as to dispose of the power equipment in a timely manner.

[0166] The embodiment of the present application also provides an electronic device, which can be a terminal, a server, etc. The terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, a personal computer, etc. The server can be a single server or a server cluster composed of multiple servers, etc.

[0167] In some embodiments, the power equipment handling device based on the data middle station can also be integrated into multiple electronic devices. For example, the power equipment handling device based on the data middle station can be integrated into multiple servers, and the power equipment handling method based on the data middle station of the present application can be implemented by multiple servers.

[0168] In this embodiment, the electronic device of this embodiment is a terminal as an example for detailed description, for example, Figure 3 As shown, it shows a schematic diagram of the structure of the terminal involved in the embodiment of the present application, specifically:

[0169] The terminal may include one or more processors 301 of processing cores, one or more storage media 302, a power supply 303, an input module 304, and a communication module 305. Those skilled in the art will appreciate that Figure 3 The terminal structure shown in the figure does not constitute a limitation on the terminal, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0170] The processor 301 is the control center of the terminal. It uses various interfaces and lines to connect various parts of the entire terminal. It executes various functions of the terminal and processes data by running or executing software programs and / or modules stored in the memory 302, and calling data stored in the memory 302, so as to monitor the terminal as a whole. In some embodiments, the processor 301 may include one or more processing cores; in some embodiments, the processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 301.

[0171] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and data processing by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0172] The terminal also includes a power supply 303 for supplying power to various components. In some embodiments, the power supply 303 can be logically connected to the processor 301 through a power management system, so as to manage charging, discharging, and power consumption through the power management system. The power supply 303 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0173] The terminal may further include an input module 304, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.

[0174] The terminal may also include a communication module 305. In some embodiments, the communication module 305 may include a wireless module. The terminal may perform short-range wireless transmission through the wireless module of the communication module 305, thereby providing the user with wireless broadband Internet access. For example, the communication module 305 may be used to help the user send and receive emails, browse web pages, and access streaming media.

[0175] Although not shown, the terminal may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 301 in the terminal will load the executable files corresponding to the processes of one or more applications into the memory 302 according to the following instructions, and the processor 301 will run the applications stored in the memory 302, thereby realizing various functions, as follows:

[0176] Collecting current operation data of the power equipment through the data acquisition device;

[0177] Acquiring equipment parameters of the electric equipment from the server;

[0178] Generate a first probability value and a second probability value according to the equipment parameters and the current operation data through the electric equipment determination model, wherein the first probability value is used to characterize the probability that the electric equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to characterize the probability that the electric equipment is in a condition that requires immediate replacement;

[0179] The electric power device is processed based on a comparison result among the first probability value, the second probability value and a preset probability threshold.

[0180] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0181] From the above, it can be seen that the embodiments of the present application can accurately provide adaptive disposal measures for the current status of the power equipment through the data middle platform and the neural network model, so as to dispose of the power equipment in a timely manner.

[0182] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a storage medium and loaded and executed by a processor.

[0183] To this end, an embodiment of the present application provides a storage medium in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any one of the power equipment disposal methods based on a data middle station provided in the embodiment of the present application. For example, the instructions can execute the following steps:

[0184] Collecting current operation data of the power equipment through the data acquisition device;

[0185] Acquiring equipment parameters of the electric equipment from the server;

[0186] Generate a first probability value and a second probability value according to the equipment parameters and the current operation data through the electric equipment determination model, wherein the first probability value is used to characterize the probability that the electric equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to characterize the probability that the electric equipment is in a condition that requires immediate replacement;

[0187] The electric power device is processed based on a comparison result among the first probability value, the second probability value and a preset probability threshold.

[0188] The storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0189] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a storage medium. A processor of a computer device reads the computer instructions from the storage medium, and the processor executes the computer instructions, so that the computer device executes the method provided in various optional implementations provided in the above embodiments.

[0190] Since the instructions stored in the storage medium can execute the steps in any one of the methods for handling power equipment based on the data middle station provided in the embodiments of the present application, the beneficial effects that can be achieved by any one of the methods for handling power equipment based on the data middle station provided in the embodiments of the present application can be achieved. Please see the previous embodiments for details and will not be repeated here.

[0191] The above is a detailed introduction to the power equipment disposal method, device, terminal and storage medium based on the data middle station provided in the embodiment of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for handling electric power equipment based on a data center, characterized in that: Applied to a data middle station, the data middle station includes a data acquisition device and a server, and the data acquisition device and the server are both in the same network; The method comprises: Collecting current operation data of the power equipment through the data acquisition device; Acquiring equipment parameters of the electric equipment from the server; Generate a first probability value and a second probability value according to the equipment parameters and the current operation data through the electric equipment determination model, wherein the first probability value is used to characterize the probability that the electric equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to characterize the probability that the electric equipment is in a condition that requires immediate replacement; Based on a comparison result between the first probability value, the second probability value and a preset probability threshold, processing the electric power device; The electric power equipment determination model comprises an input layer, a hidden layer and an output layer, wherein the output layer has two neurons; The generating of the first probability value and the second probability value according to the equipment parameters and the current operation data by the electric power equipment determination model includes: Generate a one-dimensional feature vector corresponding to the power equipment according to the equipment parameters and the current operation data through the input layer; Generate a one-dimensional output vector according to the one-dimensional feature vector through the hidden layer; Generate the first probability value and the second probability value according to the one-dimensional output vector through two neurons of the output layer; The processing of the electric power device based on the comparison result between the first probability value, the second probability value and a preset probability threshold comprises: If the first probability value is not less than the preset probability threshold and the second probability value is less than the preset probability threshold, repairing the power equipment according to the actual condition of the power equipment; If the first probability value is less than the preset probability threshold and the second probability value is not less than the preset probability threshold, replacing the power equipment; If the first probability value and the second probability value are both smaller than the preset probability threshold, the power equipment continues to be used.

2. The method according to claim 1, characterized in that The step of generating a one-dimensional feature vector corresponding to the power equipment according to the equipment parameters and the current operation data through the input layer includes: Combining the device parameters and the current operation data into an input array; The input array is input to the input layer, and the one-dimensional feature vector is obtained as output.

3. The method according to claim 1, characterized in that The hidden layer includes a first fully connected layer, a second fully connected layer, and a third fully connected layer, the first fully connected layer has a first number of neurons, the second fully connected layer has a second number of neurons, and the third fully connected layer has a third number of neurons, the second number is half of the first number, and the third number is half of the second number; The step of generating a one-dimensional output vector according to the one-dimensional feature vector through the hidden layer includes: Input the one-dimensional feature vector to the first fully connected layer, and output a first vector; Input the first vector to the second fully connected layer, and output a second vector; The second vector is input into the third fully connected layer, and output to obtain the one-dimensional output vector.

4. The method according to claim 1, characterized in that The hidden layer includes a fourth fully connected layer, a first random dropout layer, a fifth fully connected layer, a second random dropout layer, a sixth fully connected layer and a batch normalization layer; The step of generating a one-dimensional output vector according to the one-dimensional feature vector through the hidden layer includes: Inputting the one-dimensional feature vector into the fourth fully connected layer, and outputting a third vector; Inputting the third vector to the first random dropout layer, and outputting an output vector corresponding to the third vector, wherein a value of the third vector is the same as a value of the output vector corresponding to the third vector; Inputting the output vector corresponding to the third vector into the fifth fully connected layer, and outputting a fourth vector; Inputting the fourth vector to the second random dropout layer, and outputting an output vector corresponding to the fourth vector, wherein a value of the fourth vector is the same as a value of the output vector corresponding to the fourth vector; Inputting the output vector corresponding to the fourth vector into the sixth fully connected layer, and outputting a fifth vector; The fifth vector is normalized by the batch normalization layer to obtain the one-dimensional output vector.

5. The method according to claim 1, characterized in that After generating the first probability value and the second probability value according to the equipment parameters and the current operation data through the electric power equipment determination model, the method further includes: receiving a model update instruction input by a user, and merging the current operation data with the historical operation data of the electric power equipment into updated historical operation data; Based on the updated historical operation data and the equipment parameters, the electric power equipment determination model is updated to obtain an updated electric power equipment determination model.

6. A power equipment handling device based on a data middle station, using the power equipment handling method based on a data middle station as claimed in claim 1, characterized in that: Applied to a data middle station, the data middle station includes a data acquisition device and a server, and the data acquisition device and the server are both in the same network; The device comprises: The first unit is used to collect current operation data of the power equipment through the data collection device; A second unit is used to obtain equipment parameters of the power equipment from the server; A third unit is used to generate a first probability value and a second probability value according to the equipment parameters and the current operation data through an electric equipment determination model, wherein the first probability value is used to represent the probability that the electric equipment is in a condition that requires maintenance but does not need to be replaced, and the second probability value is used to represent the probability that the electric equipment is in a condition that requires immediate replacement; The fourth unit is used to process the power device based on a comparison result among the first probability value, the second probability value and a preset probability threshold.

7. A terminal, characterized in that: It comprises a processor and a memory, wherein the memory stores a plurality of instructions; the processor loads instructions from the memory to execute the steps in the method as claimed in any one of claims 1 to 5.

8. A storage medium, characterized in that: The storage medium stores a plurality of instructions, and the instructions are suitable for loading by a processor to execute the steps in the method according to any one of claims 1 to 5.

Citation Information

Patent Citations

  • Equipment management method and device based on neural network

    CN110232482A

  • Server dormancy control method and device, storage medium and electronic equipment

    CN118193072A