Early warning method and device for energy equipment and computer readable storage medium

Abstract

The invention discloses an early warning method and device for energy equipment and a computer readable storage medium. The early warning method for the energy equipment comprises the steps that monitoring information of the energy equipment is collected, and the monitoring information comprises equipment load information, equipment temperature information and equipment aging information; analyzing the monitoring information, and determining an equipment load rate, an equipment temperature abnormal value and an equipment aging degree; determining a hidden danger coefficient value through a coefficient adjustment model according to the equipment load rate, the equipment temperature abnormal value and the equipment aging degree; adjusting the initial hidden danger threshold range by using the hidden danger coefficient value, and respectively determining updated hidden danger threshold ranges corresponding to the equipment load and the equipment temperature value; and comparing the equipment load capacity and the equipment temperature value at the current moment with the corresponding updated hidden danger threshold range, and performing early warning under the condition that the equipment load capacity and the equipment temperature value exceed the updated hidden danger threshold range.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to an early warning method, device and computer-readable storage medium for energy equipment. Background Art

[0002] With the rapid development of the energy industry, massive amounts of energy data not only record the production, transmission and consumption of energy, but also contain information on equipment operating status, environmental impacts and market trends. In-depth mining and intelligent analysis of these data are essential for improving energy efficiency, predicting and avoiding equipment failures, and achieving refined energy management. However, the current energy data analysis system has significant limitations in terms of function and performance, which are mainly reflected in the following aspects:

[0003] (1) Limitations of the early warning mechanism: Traditional systems rely on threshold early warning. This static early warning method can only respond to known, explicit anomalies, but is helpless against latent, progressive operational risks. It cannot capture subtle trends in data, nor can it predict possible failure modes through correlation analysis, which greatly limits the system's early warning capabilities and failure prevention effects.

[0004] (2) Lack of multi-dimensional analysis: The complexity of energy systems requires analytical tools to be able to examine data from multiple perspectives and levels, including but not limited to time series analysis, spatial distribution analysis, equipment status monitoring, etc. However, existing systems are often limited to analysis of a single or a few dimensions, which not only limits the comprehensiveness of data interpretation, but also hinders the in-depth understanding of complex issues.

[0005] (3) Lack of intelligent diagnosis and early warning functions: In the face of increasing data volume and complexity, systems that lack intelligent diagnosis capabilities find it difficult to accurately identify potential operational risks in real time. This includes not only rapid response to real-time data, but also deep learning of historical data and predictive analysis of future trends. Without such intelligent functions, the system cannot provide effective early warnings before the problem worsens, thus missing the best time for intervention.

[0006] Currently, no effective solution has been proposed to solve the technical problems of high warning limitations caused by the static threshold warning method and single-dimensional warning in the above-mentioned prior art. Summary of the invention

[0007] The embodiments of the present application provide an early warning method, device and computer-readable storage medium for energy equipment, so as to at least solve the technical problems of high early warning limitations caused by static threshold early warning methods and single-dimensional early warnings existing in the prior art.

[0008] According to one aspect of an embodiment of the present application, there is provided an early warning method for energy equipment, comprising: collecting monitoring information of energy equipment, wherein the monitoring information comprises equipment load information, equipment temperature information and equipment aging information; analyzing the monitoring information to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree; determining the hidden danger coefficient value according to the equipment load rate, equipment temperature abnormality value and equipment aging degree through a coefficient adjustment model; adjusting the initial hidden danger threshold range using the hidden danger coefficient value to respectively determine the updated hidden danger threshold range corresponding to the equipment load and the equipment temperature value; and comparing the equipment load and the equipment temperature value at the current moment with the corresponding updated hidden danger threshold range, and issuing an early warning when the updated hidden danger threshold range is exceeded.

[0009] According to another aspect of an embodiment of the present application, there is also provided an early warning device for energy equipment, including: an information collection module, used to collect monitoring information of energy equipment, wherein the monitoring information includes equipment load information, equipment temperature information and equipment aging information; an information analysis module, used to analyze the monitoring information, and determine the equipment load rate, equipment temperature abnormality value and equipment aging degree; a coefficient determination module, used to determine the hidden danger coefficient value according to the equipment load rate, equipment temperature abnormality value and equipment aging degree through a coefficient adjustment model; a range determination module, used to adjust the initial hidden danger threshold range using the hidden danger coefficient value, and determine the updated hidden danger threshold range corresponding to the equipment load and equipment temperature value respectively; and an early warning module, used to compare the equipment load and equipment temperature value at the current moment with the corresponding updated hidden danger threshold range, and issue an early warning when the updated hidden danger threshold range is exceeded.

[0010] According to another aspect of an embodiment of the present application, there is also provided an early warning device for energy equipment, including: a processor; and a memory, connected to the processor, for providing the processor with instructions for processing the following processing steps: collecting monitoring information of the energy equipment, wherein the monitoring information includes equipment load information, equipment temperature information and equipment aging information; analyzing the monitoring information to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree; determining the hidden danger coefficient value according to the equipment load rate, equipment temperature abnormality value and equipment aging degree through a coefficient adjustment model; adjusting the initial hidden danger threshold range using the hidden danger coefficient value to determine the updated hidden danger threshold range corresponding to the equipment load and equipment temperature value respectively; and comparing the equipment load and equipment temperature value at the current moment with the corresponding updated hidden danger threshold range, and issuing an early warning when the updated hidden danger threshold range is exceeded.

[0011] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0012] According to another aspect of the embodiments of the present application, a computer program product is also provided, including a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0013] In the embodiment of the present application, the server analyzes and processes the monitoring information such as the equipment load information, the equipment temperature information and the equipment aging information in sequence to determine the corresponding equipment load rate, the equipment temperature abnormal value and the equipment aging degree. And the server also generates a hidden danger coefficient value based on the equipment load rate, the equipment temperature abnormal value and the equipment aging degree, so as to update the hidden danger threshold range in real time according to the hidden danger coefficient value. Therefore, compared with the early warning mechanism that relies on static thresholds in the prior art, this technical solution can more sensitively capture subtle changes in the equipment operation status by introducing the hidden danger coefficient value to dynamically adjust the initial hidden danger threshold range, thereby achieving more accurate and timely early warning and effectively preventing equipment failure. And this technical solution determines whether the equipment load, equipment temperature and equipment aging need early warning in sequence according to the updated hidden danger threshold range, thereby overcoming the limitations of the traditional system in multi-dimensional analysis, and realizing comprehensive monitoring of the operation status of energy equipment by monitoring various dimensions such as equipment load information, equipment temperature information and equipment aging information. The use of multi-dimensional monitoring helps to understand the health status of the equipment more deeply, making the monitoring more comprehensive and reducing limitations. In addition, this technical solution uses machine learning methods such as logistic regression models to intelligently judge and predict the status of energy equipment, so that it can respond quickly to real-time data and improve the efficiency of early warning. This solves the technical problems of high early warning limitations caused by static threshold early warning methods and single-dimensional early warnings in the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0015] Figure 1 is a hardware structure block diagram of a computing device for implementing the method according to Embodiment 1 of the present application;

[0016] Figure 2 is a schematic diagram of an early warning system for energy equipment according to Example 1 of the present application;

[0017] Figure 3 is a module schematic diagram of the server according to Embodiment 1 of the present application;

[0018] Figure 4 is a flow chart of an early warning method for energy equipment according to the first aspect of Example 1 of the present application;

[0019] Figure 5 is a schematic diagram of a load monitoring model according to the first aspect of Example 1 of the present application;

[0020] Figure 6 is a schematic diagram of a feature extraction module of a load monitoring model according to the first aspect of Embodiment 1 of the present application;

[0021] Figure 7 is a schematic diagram of the information analysis architecture according to the first aspect of Embodiment 1 of the present application;

[0022] Figure 8 is a sequential flow chart of an early warning method for energy equipment according to the first aspect of Example 1 of the present application;

[0023] Fig. 9 is a schematic diagram of an early warning device for energy equipment according to Example 2 of the present application; and

[0024] Fig.10 This is a schematic diagram of an early warning device for energy equipment according to Example 3 of the present application. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the technical solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only embodiments of a part of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] Example 1

[0028] According to this embodiment, a method embodiment of an early warning method for energy equipment is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0029] The method embodiment provided in this embodiment can be executed in a mobile terminal, a computer terminal, a server or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computing device for implementing an early warning method for energy equipment. Figure 1 As shown, the computing device may include one or more processors (the processor may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory for storing data, and a transmission device for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0030] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computing device. As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0031] The memory can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the early warning method for energy equipment in the embodiment of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, the early warning method for energy equipment of the above-mentioned application is realized. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computing device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0032] The transmission device is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computing device. In one example, the transmission device includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device can be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet wirelessly.

[0033] The display may be, for example, a touch screen liquid crystal display (LCD) that may enable a user to interact with a user interface of the computing device.

[0034] It should be noted that, in some optional embodiments, the above Figure 1 The computing device shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware elements and software elements. Figure 1 This is merely one example of a particular embodiment and is intended to illustrate the types of components that may be present in the computing devices described above.

[0035] Figure 2 is a schematic diagram of an early warning system for energy equipment according to this embodiment. Figure 2 As shown, the system includes: energy equipment 100, server 200 and database 300.

[0036] refer to Figure 3 As shown, the server 200 includes an information collection module 210 , an information processing module 220 , an information analysis module 230 and an early warning module 240 .

[0037] The server 200 is connected to the energy device 100 and the database 300 respectively, so that the information collection module 210 of the server 200 collects the monitoring information of the monitored energy device 100, and stores the monitoring information in the database 300. The energy device 100 can be an electric energy device, a water energy device, a gas device, a steam device, etc.

[0038] The information processing module 220 is used to pre-process the monitoring information before information analysis.

[0039] The information analysis module 230 is used to analyze the pre-processed monitoring information to determine whether an early warning is required for the energy equipment.

[0040] The warning module 240 is used to send a warning message to the staff according to the analysis result of the information analysis module 230.

[0041] It should be noted that the energy device 100, the server 200 and the database 300 in the system can all be applicable to the hardware structure described above.

[0042] Under the above operating environment, according to the first aspect of this embodiment, a warning method for energy equipment is provided. The method comprises: Figure 2 The server 200 shown in FIG. Figure 4 A schematic diagram showing the process of the method is shown in FIG. Figure 4 As shown, the method includes:

[0043] S402: Collect monitoring information of energy equipment, where the monitoring information includes equipment load information, equipment temperature information, and equipment aging information;

[0044] S404: Analyze the monitoring information to determine the equipment load rate, equipment temperature abnormality value, and equipment aging degree;

[0045] S406: Determine the hidden danger coefficient value according to the equipment load rate, the equipment temperature abnormality value and the equipment aging degree through the coefficient adjustment model;

[0046] S408: adjusting the initial hidden danger threshold range using the hidden danger coefficient value, and determining updated hidden danger threshold ranges corresponding to the device load and the device temperature value respectively; and

[0047] S410: Compare the current device load and device temperature value with the corresponding update potential risk threshold range, and issue an early warning when the update potential risk threshold range is exceeded.

[0048] Specifically, the server 200 continuously monitors the energy device 100, collects monitoring information of the energy device 100 through the information collection module 210, and sends the monitoring information to the database 300 for storage. The monitoring information includes equipment load information, equipment temperature information, and equipment aging information. The load amount of the equipment load information can be collected by devices such as electric energy meters, water meters, and gas meters that can be used to directly read the original energy consumption data; the equipment temperature information can be collected by sensors; and the equipment aging information can be the maintenance record of the energy equipment. The information collection module 210 then sends the monitoring information to the information processing module 220.

[0049] Furthermore, after receiving the monitoring information, the information processing module 220 pre-processes the monitoring information to filter out abnormal data in the monitoring information. The pre-processing operation at least includes: cleaning, integration and formatting. The information processing module 220 then sends the pre-processed monitoring information to the information analysis module 230.

[0050] Further, the information analysis module 230 receives the pre-processed monitoring information, and analyzes the equipment load information, equipment temperature information, and equipment aging information in the monitoring information, respectively, so as to obtain the equipment load rate DL corresponding to the equipment load information, the equipment temperature abnormality value DT corresponding to the equipment temperature information, and the equipment aging degree DA corresponding to the equipment aging information. The equipment load rate, the equipment temperature abnormality value, and the equipment aging degree are values ​​in the interval (0,1].

[0051] Further, the information analysis module 230 inputs the equipment load rate DL, the equipment temperature abnormal value DT and the equipment aging degree DA into the preset coefficient adjustment model, and outputs the hidden danger coefficient value y according to the equipment load rate DL, the equipment temperature abnormal value DT and the equipment aging degree DA through the coefficient adjustment model. The hidden danger coefficient value is used to adjust the preset hidden danger threshold range. And the coefficient adjustment model is a logistic regression model:

[0052] g=k0+k1*DL+k2*DT+k3*DA (1)

[0053] y=1 / 1+e -g (2)

[0054] Wherein k0~k3 represent linear regression coefficients, which can be determined by training through machine learning methods, for example.

[0055] Furthermore, the information analysis module 230 is pre-set with initial hidden danger threshold ranges corresponding to the equipment load and the equipment temperature value of the energy equipment, respectively including: initial hidden danger threshold range of equipment load [IL min ,IL max], and the initial risk threshold range of the device temperature value [IT min ,IT max ]. The initial hidden danger threshold range is determined by taking the normal value as the center value, and is the maximum threshold range used to determine whether a failure of the energy equipment occurs.

[0056] Afterwards, the information analysis module 230 uses the threshold range adjustment formula and the hidden danger coefficient value y to adjust the initial hidden danger threshold ranges corresponding to the equipment load and the equipment temperature value, respectively, to obtain an updated hidden danger threshold range for the equipment load [RL min ,RL max ], and the updated risk threshold range of the device temperature value [RT min ,RT max ]. The threshold range adjustment formula is:

[0057] [RL min ,RL max ]=[IL min ,IL max ]*y;

[0058] [RT min ,RT max ]=[IT min ,IT max ]*y.

[0059] The updated hidden danger threshold range is a threshold range used to determine whether a failure occurs in the energy equipment at the current moment.

[0060] Furthermore, the information analysis module 230 collects the equipment load of the energy equipment at the current moment. The equipment load may be, for example, the power load of the electric energy equipment. Then, the information analysis module 230 compares the equipment load with the updated hidden danger threshold range [RL min ,RL max ] to determine whether the device load exceeds the update risk threshold range [RL min ,RL max ]. When the device load exceeds the update risk threshold range [RL min ,RL max ], it means that the equipment load exceeds the normal range, so the information analysis module 230 sends an early warning message to the early warning module 240 indicating that there is a hidden danger in the equipment load.

[0061] The information analysis module 230 collects the device temperature value of the energy device at the current moment. The device temperature value may be, for example, the temperature value of the generator. The information analysis module 230 then compares the device temperature value with the updated hidden danger threshold range [RL min ,RL max] to determine whether the device temperature value exceeds the update risk threshold range [RL min ,RL max ]. When the device temperature exceeds the update risk threshold range [RL min ,RL max ], it means that the temperature value of the device exceeds the normal range, so the information analysis module 230 sends a warning message to the warning module 240 that there is a hidden danger in the temperature value of the device.

[0062] Furthermore, after receiving the warning message sent by the information analysis module 230, the warning module 240 sends the warning message to the terminal device of the corresponding staff member.

[0063] As described in the background technology, with the rapid development of the energy industry, massive amounts of energy data not only record the process of energy production, transmission and consumption, but also contain information on equipment operating status, environmental impact and market trends. The in-depth mining and intelligent analysis of these data are crucial to improving energy efficiency, predicting and avoiding equipment failures, and realizing refined energy management. However, the current energy data analysis system has significant limitations in terms of function and performance, which are mainly reflected in the following aspects: (1) Limitations of the early warning mechanism: Traditional systems rely on threshold early warning. This static early warning method can only respond to known and explicit anomalies, but is helpless against those latent and progressive operational risks. It cannot capture subtle trends in the data, nor can it predict possible failure modes through correlation analysis, which greatly limits the system's early warning capabilities and failure prevention effects. (2) Lack of multi-dimensional analysis: The complexity of the energy system requires analytical tools to be able to examine data from multiple angles and multiple levels, including but not limited to time series analysis, spatial distribution analysis, equipment status monitoring, etc. However, existing systems are often limited to the analysis of a single or a few dimensions, which not only limits the comprehensiveness of data interpretation, but also hinders the in-depth understanding of complex problems. (3) Lack of intelligent diagnosis and early warning functions: In the face of the growing amount and complexity of data, systems that lack intelligent diagnosis capabilities find it difficult to accurately identify potential operational risks in real time. This includes not only rapid response to real-time data, but also deep learning of historical data and predictive analysis of future trends. Without such intelligent functions, the system cannot provide effective early warnings before the problem worsens, thus missing the best time for intervention.

[0064] In view of the technical problems described above, through the technical solutions of the embodiments of the present application, the server analyzes and processes the monitoring information such as the equipment load information, the equipment temperature information and the equipment aging information in sequence to determine the corresponding equipment load rate, the equipment temperature abnormal value and the equipment aging degree. And the server also generates a hidden danger coefficient value based on the equipment load rate, the equipment temperature abnormal value and the equipment aging degree, so as to update the hidden danger threshold range in real time according to the hidden danger coefficient value. Therefore, compared with the early warning mechanism that relies on static thresholds in the prior art, this technical solution can more sensitively capture the subtle changes in the equipment operation status by introducing the hidden danger coefficient value to dynamically adjust the initial hidden danger threshold range, thereby achieving more accurate and timely early warning and effectively preventing equipment failure. And this technical solution determines whether the equipment load, equipment temperature and equipment aging need early warning in sequence according to the updated hidden danger threshold range, thereby overcoming the limitations of the traditional system in multi-dimensional analysis, and realizing comprehensive monitoring of the operation status of energy equipment by monitoring various dimensions such as equipment load information, equipment temperature information and equipment aging information. The use of multi-dimensional monitoring helps to understand the health status of the equipment more deeply, making the monitoring more comprehensive and reducing limitations. In addition, this technical solution uses machine learning methods such as logistic regression models to intelligently judge and predict the status of energy equipment, so that it can respond quickly to real-time data and improve the efficiency of early warning. This solves the technical problems of high early warning limitations caused by static threshold early warning methods and single-dimensional early warnings in the existing technology.

[0065] Optionally, the monitoring information is analyzed to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree, including: analyzing the equipment load information through a load monitoring model to determine the equipment load rate of the energy equipment, wherein the equipment load information includes ambient temperature, equipment noise and equipment thermal infrared imaging; determining the equipment temperature abnormality value of the energy equipment based on the equipment temperature information through a Markov chain; and determining the equipment aging degree of the energy equipment based on the equipment aging information through an information analysis architecture.

[0066] Specifically, the information analysis module 230 is pre-set with a load monitoring model for analyzing the equipment load information. Thus, the information analysis module 230 collects the equipment load information at the current moment. The equipment load information includes the ambient temperature of the environment in which the energy equipment is located (such as temperature, humidity, wind speed and rainfall, etc.), equipment noise, and equipment thermal infrared imaging (such as thermal infrared imaging of the generator). Afterwards, the information analysis module 230 inputs the equipment load information into the load monitoring model, and the load monitoring model outputs the equipment load rate according to the equipment load information.

[0067] Furthermore, the information analysis module 230 determines the abnormal value of the device temperature of the energy device according to the device temperature information through a Markov chain.

[0068] Furthermore, the information analysis module 230 is pre-set with an information analysis framework for analyzing the device aging information. Thus, the information analysis module 230 inputs the device aging information into the information analysis framework, and the information analysis framework analyzes the device aging information, thereby outputting the device aging degree.

[0069] Therefore, this technical solution analyzes the monitoring information through a model and method that is compatible with various aspects of monitoring information (i.e., equipment load information, equipment temperature information, and equipment aging information), and can accurately determine the corresponding indicator data (i.e., equipment load rate, equipment temperature abnormality, and equipment aging degree), thereby improving the accuracy.

[0070] Optionally, the operation of analyzing the equipment load information through the load monitoring model to determine the equipment load rate of the energy equipment includes: extracting features from the thermal infrared imaging of the equipment through a feature extraction module to generate first feature information; fusing the first feature information, the second feature information corresponding to the ambient temperature, and the third feature information corresponding to the equipment noise through a first feature fusion module to generate first fused feature information; processing the first fused feature information through a first fully connected layer to generate fourth feature information; and classifying the fourth feature information through a first binary classification model to determine the equipment load rate of the energy equipment.

[0071] Specifically, the load monitoring model includes a feature extraction module 410, a first feature fusion module 420 and a first fully connected layer 430. The information analysis module 230 is also pre-set with a first binary classification model 440. The load monitoring model and the first binary classification model 440 are trained according to the sample device load information.

[0072] refer to Figure 5 As shown, the information analysis module 230 processes the equipment thermal infrared imaging, the ambient temperature and the equipment noise into a feature vector, and inputs the feature vector into the load monitoring model.

[0073] The load monitoring model extracts features from the feature vector corresponding to the thermal infrared imaging of the device through the feature extraction module 410 to generate the first feature information. The load monitoring model then fuses the first feature information, the feature vector corresponding to the ambient temperature (i.e., the second feature information), and the feature vector corresponding to the ambient noise (i.e., the third feature information) through the first feature fusion module 420 to generate the first fused feature information. The load monitoring model then processes the first fused feature information through the first fully connected layer 430 to generate the fourth feature information. The load monitoring model then inputs the fourth feature information into the first binary classification model 440. The first binary classification model 440 then classifies the fourth feature information and outputs the device load rate.

[0074] Therefore, this technical solution extracts and integrates the features of equipment thermal infrared imaging, ambient temperature and equipment noise through a load monitoring model, so as to generate equipment load rate in multiple dimensions.

[0075] Optionally, the operation of performing feature extraction on the thermal infrared imaging of the device through the feature extraction module to generate the first feature information includes: performing feature extraction on the thermal infrared imaging of the device through a convolutional neural network to generate a first feature map; performing feature extraction on the thermal infrared imaging of the device through a residual network to generate a second feature map; performing fusion processing on the first feature map and the second feature map through a second feature fusion module to generate second fused feature information; and performing pooling processing on the second fused feature information through a global average pooling layer to generate the first feature information.

[0076] Specifically, the feature extraction module 410 includes a convolutional neural network 411, a residual network 412, a second feature fusion module 413 and a global average pooling layer 414.

[0077] The feature extraction module 410 inputs the device thermal infrared imaging into the convolutional neural network 411 and the residual network 412 respectively. The convolutional neural network 411 performs convolution operations on the device thermal infrared imaging through multiple convolutional layers to extract feature information of the device thermal infrared imaging, thereby generating a first feature map. The residual network 421 extracts features of the device thermal infrared imaging through multiple residual blocks to generate a second feature map. The feature extraction module 410 then inputs the first feature map and the second feature map into the second feature fusion module 413, and the second feature fusion module 413 performs fusion processing on the first feature map and the second feature map to generate second feature fusion information. The feature extraction module 410 then inputs the second feature fusion information into the global average pooling layer 414, and the global average pooling layer 414 performs pooling processing on the second feature fusion information to generate a corresponding feature vector (i.e., the first feature information).

[0078] Therefore, this technical solution uses convolutional neural networks and residual networks to jointly extract features from thermal infrared imaging of the device, thereby significantly improving the efficiency and accuracy of feature extraction of thermal infrared imaging of the device through the synergy of convolutional neural networks and residual networks. Convolutional neural networks (CNNs) can effectively extract basic features such as edges and textures in imaging with their powerful local feature capture capabilities, while residual networks (ResNet) solve the gradient vanishing problem in deep network training by introducing shortcut connections, allowing the network to learn deeper and more abstract feature representations. The combination of the two not only enhances the model's ability to capture complex imaging features, but also effectively alleviates the difficulty of deep network training and improves the stability and generalization ability of the model.

[0079] Optionally, the operation of determining the abnormal temperature value of the energy device according to the device temperature information through the Markov chain includes: constructing the state space and transfer matrix of the Markov chain according to the temperature state of the energy device; collecting sample temperature information of the energy device in a preset period, and determining the parameter value of the transfer matrix according to the sample temperature information; determining the parameter value of the state space according to the transfer matrix when the Markov chain converges, and determining the probability of the temperature rise of the energy device according to the parameter value of the state space; and determining the abnormal temperature value of the energy device according to the device temperature information collected at the current moment and the probability of temperature rise through the temperature monitoring model.

[0080] Specifically, the information analysis module predetermines the temperature state of the energy device as temperature rise, temperature drop, and temperature maintenance, so that the information analysis module constructs the state space and transfer matrix of the Markov chain according to the temperature state.

[0081] More specifically, the information analysis module 230 pre-constructs a transfer matrix A based on a Markov chain:

[0082]

[0083] Among them, parameter a 1,1 represents the probability of transitioning from the state "temperature rises" to the state "temperature rises"; parameter a 1,2 represents the probability of transitioning from the state "temperature rise" to the state "temperature maintenance"; parameter a 1,3 represents the probability of transitioning from the state "temperature rises" to the state "temperature drops"; parameter a 2,1 Indicates the probability of transitioning from the state "temperature maintained" to the state "temperature rising"; parameter a 2,2 represents the probability of transitioning from the state "temperature maintained" to the state "temperature maintained"; parameter a 2,3 Indicates the probability of transitioning from the state "temperature maintained" to the state "temperature dropped"; parameter a 3,1 represents the probability of transitioning from the state "temperature drops" to the state "temperature rises"; parameter a 3,2 represents the probability of transitioning from the state "temperature drops" to the state "temperature remains"; and parameter a 3,3 It represents the probability of transitioning from the state “temperature drops” to the state “temperature drops”.

[0084] Furthermore, the information analysis module 230 constructs a state space W based on a Markov chain:

[0085]

[0086] Among them, w1 represents the probability that the temperature of the energy device rises, w2 represents the probability that the temperature of the energy device remains the same, and w3 represents the probability that the temperature of the energy device drops.

[0087] Furthermore, the information analysis module 230 collects sample temperature information of the energy device in a preset period. Table 1 shows a temperature information table of the sample temperature information.

[0088] Table 1

[0089] sample <![CDATA[S0]]> <![CDATA[S1]]> <![CDATA[S2]]> <![CDATA[S3]]> ... <![CDATA[S n ]]> <![CDATA[S n+1 ]]> Temperature value <![CDATA[t0]]> <![CDATA[t1]]> <![CDATA[t2]]> <![CDATA[t3]]> ... <![CDATA[t n ]]> <![CDATA[t n+1 ]]>

[0090] The information analysis module 230 collects a total of n+2 samples, namely S0~S n+1 , where each sample corresponds to a temperature value, t0~t n+1 That is, the temperature value of sample S0 is t0, the temperature value of sample S1 is t1, the temperature value of sample S2 is t2, ..., the temperature value of sample S n+1 The temperature value is t n+1 .

[0091] Furthermore, the information analysis module 230 determines the temperature state corresponding to each sample according to the temperature values ​​of n+2 samples. i (i=1~n), according to the sample S i Compared with the previous sample S i-1 Temperature value t i and t i-1 Determine the sample S i The corresponding state before transfer. For example, when the temperature t i Greater than t i-1 When the sample S i The state before the transfer is that the temperature rises; when the temperature t i Equal to t i-1 When the sample S i The state before the transfer is temperature maintenance; when the temperature t i Less than t i-1 When the sample S i The state before transfer is that the temperature drops.

[0092] For example, for each sample S i (i=1~n), according to the sample S i With the next sample S i+1 Temperature value t i and t i+1 Determine the sample S i The corresponding state after transfer. For example, when the temperature t i Greater than t i+1 When the sample S i The state after the transfer is that the temperature drops; when the temperature t i Equal to t i+1 When the sample S iThe state after the transfer is temperature maintenance; when the temperature t i Less than t i+1 When the sample S i The state after the transfer is that the temperature rises.

[0093] Furthermore, the information analysis module 230 counts the number of different pre-transfer states and post-transfer states of the samples according to the pre-transfer states and post-transfer states of each sample. For example, the information analysis module 230 counts the following numbers: the number of samples v1 whose pre-transfer state is "temperature rise"; the number of samples va1 whose pre-transfer state is "temperature rise" and whose post-transfer state is "temperature rise"; the number of samples va2 whose pre-transfer state is "temperature rise" and whose post-transfer state is "temperature maintenance"; the number of samples va3 whose pre-transfer state is "temperature rise" and whose post-transfer state is "temperature drop".

[0094] Thus, the information analysis module 230 calculates the parameter value of the transfer matrix according to the statistical sample numbers of the state before and after the transfer:

[0095] a 1,1 =va1 / v1;

[0096] a 1,2 =va2 / v1;

[0097] a 1,3 =va3 / v1.

[0098] In addition, the information analysis module 230 also counts the following quantities: the number v2 of samples whose state before transfer is "temperature maintenance"; the number vb1 of samples whose state before transfer is "temperature maintenance" and whose state after transfer is "temperature rise"; the number vb2 of samples whose state before transfer is "temperature maintenance" and whose state after transfer is "temperature maintenance"; the number vb3 of samples whose state before transfer is "temperature maintenance" and whose state after transfer is "temperature drop".

[0099] Thus, the information analysis module 230 calculates the parameter value of the transfer matrix according to the statistical sample numbers of the state before and after the transfer:

[0100] a 2,1 =vb1 / v2;

[0101] a 2,2 =vb2 / v2;

[0102] a 2,3 =vb3 / v2.

[0103] In addition, the information analysis module 230 also counts the following quantities: the number v3 of samples whose state before transfer is "temperature drop"; the number vc1 of samples whose state before transfer is "temperature drop" and whose state after transfer is "temperature rise"; the number vc2 of samples whose state before transfer is "temperature drop" and whose state after transfer is "temperature maintain"; the number vc3 of samples whose state before transfer is "temperature drop" and whose state after transfer is "temperature drop".

[0104] Thus, the information analysis module 230 calculates the parameter value of the transfer matrix according to the statistical sample numbers of the state before and after the transfer:

[0105] a 3,1 =vc1 / v3;

[0106] a 3,2 =vc2 / v3;

[0107] a 3,3 =vc3 / v3.

[0108] Furthermore, the information analysis module 230 brings the calculated parameter values ​​into the transfer matrix A, and then determines the parameter values ​​of the state space when the Markov chain converges based on the transfer matrix, and determines the probability of temperature rise of the energy device based on the parameter values ​​of the state space.

[0109] For example, the value of the converged state space W' can be obtained according to the following formula:

[0110] (AI)×W'=0,

[0111] Right now:

[0112]

[0113] Wherein, w'1 represents the probability of temperature rising under the state of Markov chain convergence, w'2 represents the probability of temperature maintaining under the state of Markov chain convergence, and w'3 represents the probability of temperature decreasing under the state of Markov chain convergence. I is the unit matrix.

[0114] The information analysis module 230 then determines the probability p of temperature rise according to the parameter w′1 of the state space.

[0115] Furthermore, the information analysis module 230 collects the temperature value nt (i.e., the device temperature information) of the energy device at the current moment, and then inputs the temperature value nt at the current moment and the probability p of temperature rise into a preset device temperature abnormal value determination model, and the device temperature abnormal value determination model outputs the device temperature abnormal value of the energy device. The device temperature abnormal value is used to indicate the abnormal degree of the temperature of the energy device. The device temperature abnormal value determination model is a logistic regression model:

[0116] d=b0+b1*nt+b2*p (3)

[0117] DT=1 / 1+e -d (4)

[0118] b0-b2 represent linear regression coefficients, which can be determined by training through machine learning methods, for example.

[0119] Therefore, the present technology can accurately predict the future change trend of the temperature of the energy equipment based on the constructed Markov chain, so as to accurately determine the abnormal value of the equipment temperature.

[0120] Optionally, the operation of determining the degree of equipment aging of energy equipment based on equipment aging information through an information analysis architecture includes: segmenting the equipment aging information, and generating word vectors based on the segmentation using a word embedding model; semantically processing the word vectors through BERT to generate an information sequence; performing a convolution operation on the information sequence through a convolution layer to generate a third feature map; processing the third feature map through a second fully connected layer to generate fifth feature information; and classifying the fifth feature information through a second binary classification model to determine the degree of equipment aging of the energy equipment.

[0121] Specifically, the information analysis architecture includes a one-hot encoder 510, a word embedding model 520, a BERT 530, a convolutional layer 540, a second fully connected layer 550, and a second binary classification model 560. The information analysis architecture is pre-trained based on sample device aging information.

[0122] The information analysis module 230 performs word segmentation processing on the equipment aging information (such as the maintenance record of the energy equipment) through the information analysis framework to obtain m word segments Ha1~Ha m Then the information analysis module 230 uses the one-hot encoder 510 of the information analysis architecture to segment the words Ha1~Ha m Perform one-hot encoding processing to generate one-hot encoding Oa1~Oa corresponding to each word segmentation m Then the information analysis module 230 will one-hot encode Oa1~Oa m Input to the word embedding model 520, use the word embedding model to generate the corresponding one-hot encoding Oa1~Oa m The corresponding word vectors Ea1~Ea m. Then the information analysis module 230 inputs the word vector into BERT 530, uses BERT 530 to perform semantic processing on the word vector, and outputs an information sequence. Then the information analysis module 230 inputs the information sequence into the convolution layer 540, uses the convolution layer 540 to perform a convolution operation on the information sequence, and outputs a corresponding feature map (i.e., the third feature map). Then the information analysis module 230 inputs the third feature map into the second fully connected layer 550, uses the second fully connected layer 550 to process the third feature map, and generates corresponding feature information (i.e., the fifth feature information).

[0123] Furthermore, the information analysis module 230 inputs the fifth characteristic information into the second binary classification model 560, classifies the fifth characteristic information through the second binary classification model 560, and outputs the aging probability of the energy equipment (ie, the equipment aging degree DA).

[0124] Therefore, the technical solution performs natural language processing on the equipment aging information, thereby accurately identifying and semantically understanding the equipment aging information, thereby accurately determining the degree of equipment aging.

[0125] In summary, reference Figure 8 As shown, the sequential steps for early warning of energy equipment are as follows:

[0126] S801: The information collection module 210 of the server 200 collects monitoring information such as equipment load information, equipment temperature information, and equipment aging information of energy equipment;

[0127] S802: The information analysis module 230 of the server 200 analyzes the equipment load information through a load monitoring model to determine the equipment load rate of the energy equipment;

[0128] S803: The information analysis module 230 of the server 200 determines the abnormal value of the device temperature of the energy device according to the device temperature information through a Markov chain;

[0129] S804: The information analysis module 230 of the server 200 determines the degree of aging of the energy equipment according to the equipment aging information through the information analysis architecture;

[0130] S805: The information analysis module 230 of the server 200 determines the hidden danger coefficient value according to the equipment load rate, the equipment temperature abnormality value and the equipment aging degree through the coefficient adjustment model;

[0131] S806: The information analysis module 230 of the server 200 adjusts the initial hidden danger threshold range using the hidden danger coefficient value, and determines updated hidden danger threshold ranges corresponding to the device load and the device temperature value respectively;

[0132] S807: The information analysis module 230 of the server 200 compares the current device load and device temperature value with the corresponding update risk threshold range;

[0133] S808: The warning module 240 of the server 200 issues a warning when the update risk threshold range is exceeded.

[0134] In addition, according to a second aspect of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0135] In addition, according to a third aspect of this embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0136] Thus, according to this embodiment, the server determines the corresponding equipment load rate, equipment temperature abnormal value and equipment aging degree by analyzing and processing the monitoring information such as equipment load information, equipment temperature information and equipment aging information in sequence. And the server also generates a hidden danger coefficient value based on the equipment load rate, equipment temperature abnormal value and equipment aging degree, so as to update the hidden danger threshold range in real time according to the hidden danger coefficient value. Thus, compared with the early warning mechanism that relies on static thresholds in the prior art, this technical solution can more sensitively capture subtle changes in the equipment operation status by introducing the hidden danger coefficient value to dynamically adjust the initial hidden danger threshold range, thereby achieving more accurate and timely early warning and effectively preventing equipment failure. And this technical solution determines whether the equipment load, equipment temperature and equipment aging need early warning in sequence according to the updated hidden danger threshold range, thereby overcoming the limitations of the traditional system in multi-dimensional analysis, and realizing comprehensive monitoring of the operation status of energy equipment by monitoring various dimensions such as equipment load information, equipment temperature information and equipment aging information. The use of multi-dimensional monitoring helps to understand the health status of the equipment more deeply, making the monitoring more comprehensive and reducing limitations. In addition, this technical solution uses machine learning methods such as logistic regression models to intelligently judge and predict the status of energy equipment, so that it can respond quickly to real-time data and improve the efficiency of early warning. This solves the technical problems of high early warning limitations caused by static threshold early warning methods and single-dimensional early warnings in the existing technology.

[0137] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0138] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0139] Example 2

[0140] Fig. 9 FIG. 9 shows an early warning device 900 for energy equipment according to this embodiment, and the device 900 corresponds to the method according to the first aspect of embodiment 1. Fig. 9 As shown, the device 900 includes: an information acquisition module 910, which is used to collect monitoring information of energy equipment, wherein the monitoring information includes equipment load information, equipment temperature information and equipment aging information; an information analysis module 920, which is used to analyze the monitoring information and determine the equipment load rate, equipment temperature abnormality value and equipment aging degree; a coefficient determination module 930, which is used to determine the hidden danger coefficient value according to the equipment load rate, equipment temperature abnormality value and equipment aging degree through a coefficient adjustment model; a range determination module 940, which is used to adjust the initial hidden danger threshold range using the hidden danger coefficient value, and respectively determine the updated hidden danger threshold range corresponding to the equipment load and the equipment temperature value; and an early warning module 950, which is used to compare the equipment load and the equipment temperature value at the current moment with the corresponding updated hidden danger threshold range, and issue an early warning when the updated hidden danger threshold range is exceeded.

[0141] Optionally, the information analysis module 920 includes: a first determination submodule, used to analyze the equipment load information through a load monitoring model to determine the equipment load rate of the energy equipment, wherein the equipment load information includes ambient temperature, equipment noise, and equipment thermal infrared imaging; a second determination submodule, used to determine the equipment temperature abnormality value of the energy equipment according to the equipment temperature information through a Markov chain; and a third determination submodule, used to determine the equipment aging degree of the energy equipment according to the equipment aging information through an information analysis architecture.

[0142] Optionally, the first determination submodule includes: a first generation unit, used to extract features of the thermal infrared imaging of the equipment through a feature extraction module to generate first feature information; a second generation unit, used to fuse the first feature information, the second feature information corresponding to the ambient temperature, and the third feature information corresponding to the equipment noise through a first feature fusion module to generate first fused feature information; a third generation unit, used to process the first fused feature information through a first fully connected layer to generate fourth feature information; and a fourth generation unit, used to classify the fourth feature information through a first binary classification model to determine the equipment load rate of the energy equipment.

[0143] Optionally, the first generation unit includes: extracting features of the thermal infrared imaging of the device through a convolutional neural network to generate a first feature map; extracting features of the thermal infrared imaging of the device through a residual network to generate a second feature map; fusing the first feature map and the second feature map through a second feature fusion module to generate second fused feature information; and pooling the second fused feature information through a global average pooling layer to generate first feature information.

[0144] Optionally, the second determination submodule includes: a construction unit, used to construct the state space and transfer matrix of the Markov chain according to the temperature state of the energy device; a first determination unit, used to collect sample temperature information of the energy device in a preset period, and determine the parameter value of the transfer matrix according to the sample temperature information; a second determination unit, used to determine the parameter value of the state space according to the transfer matrix when the Markov chain converges, and determine the probability of temperature rise of the energy device according to the parameter value of the state space; and a third determination unit, used to determine the device temperature abnormality value of the energy device through a temperature monitoring model according to the device temperature information collected at the current moment and the probability of temperature rise.

[0145] Optionally, the third determination submodule includes: a fifth generation unit, used to segment the equipment aging information, and generate word vectors based on the segmentation using a word embedding model; a sixth generation unit, used to perform semantic processing on the word vectors through BERT to generate an information sequence; a seventh generation unit, used to perform a convolution operation on the information sequence through a convolution layer to generate a third feature map; an eighth generation unit, used to process the third feature map through a second fully connected layer to generate fifth feature information; and a fourth determination unit, used to classify the fifth feature information through a second binary classification model to determine the degree of equipment aging of the energy equipment.

[0146] Thus, according to this embodiment, the server determines the corresponding equipment load rate, equipment temperature abnormal value and equipment aging degree by analyzing and processing the monitoring information such as equipment load information, equipment temperature information and equipment aging information in sequence. And the server also generates a hidden danger coefficient value based on the equipment load rate, equipment temperature abnormal value and equipment aging degree, so as to update the hidden danger threshold range in real time according to the hidden danger coefficient value. Thus, compared with the early warning mechanism that relies on static thresholds in the prior art, this technical solution can more sensitively capture subtle changes in the equipment operation status by introducing the hidden danger coefficient value to dynamically adjust the initial hidden danger threshold range, thereby achieving more accurate and timely early warning and effectively preventing equipment failure. And this technical solution determines whether the equipment load, equipment temperature and equipment aging need early warning in sequence according to the updated hidden danger threshold range, thereby overcoming the limitations of the traditional system in multi-dimensional analysis, and realizing comprehensive monitoring of the operation status of energy equipment by monitoring various dimensions such as equipment load information, equipment temperature information and equipment aging information. The use of multi-dimensional monitoring helps to understand the health status of the equipment more deeply, making the monitoring more comprehensive and reducing limitations. In addition, this technical solution uses machine learning methods such as logistic regression models to intelligently judge and predict the status of energy equipment, so that it can respond quickly to real-time data and improve the efficiency of early warning. This solves the technical problems of high early warning limitations caused by static threshold early warning methods and single-dimensional early warnings in the existing technology.

[0147] Example 3

[0148] Fig.10 The early warning device 1000 for energy equipment according to the first aspect of this embodiment is shown, and the device 1000 corresponds to the method according to the first aspect of embodiment 1. Fig.10As shown, the device 1000 includes: a processor 1010; and a memory 1020, which is connected to the processor 1010 and is used to provide the processor 1010 with instructions for processing the following processing steps: collecting monitoring information of energy equipment, wherein the monitoring information includes equipment load information, equipment temperature information and equipment aging information; analyzing the monitoring information to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree; determining the hidden danger coefficient value according to the equipment load rate, equipment temperature abnormality value and equipment aging degree through a coefficient adjustment model; adjusting the initial hidden danger threshold range using the hidden danger coefficient value to determine the updated hidden danger threshold range corresponding to the equipment load and equipment temperature value respectively; and comparing the equipment load and equipment temperature value at the current moment with the corresponding updated hidden danger threshold range, and issuing an early warning when the updated hidden danger threshold range is exceeded.

[0149] Optionally, the monitoring information is analyzed to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree, including: analyzing the equipment load information through a load monitoring model to determine the equipment load rate of the energy equipment, wherein the equipment load information includes ambient temperature, equipment noise and equipment thermal infrared imaging; determining the equipment temperature abnormality value of the energy equipment based on the equipment temperature information through a Markov chain; and determining the equipment aging degree of the energy equipment based on the equipment aging information through an information analysis architecture.

[0150] Optionally, the operation of analyzing the equipment load information through the load monitoring model to determine the equipment load rate of the energy equipment includes: extracting features from the thermal infrared imaging of the equipment through a feature extraction module to generate first feature information; fusing the first feature information, the second feature information corresponding to the ambient temperature, and the third feature information corresponding to the equipment noise through a first feature fusion module to generate first fused feature information; processing the first fused feature information through a first fully connected layer to generate fourth feature information; and classifying the fourth feature information through a first binary classification model to determine the equipment load rate of the energy equipment.

[0151] Optionally, the operation of performing feature extraction on the thermal infrared imaging of the device through the feature extraction module to generate the first feature information includes: performing feature extraction on the thermal infrared imaging of the device through a convolutional neural network to generate a first feature map; performing feature extraction on the thermal infrared imaging of the device through a residual network to generate a second feature map; performing fusion processing on the first feature map and the second feature map through a second feature fusion module to generate second fused feature information; and performing pooling processing on the second fused feature information through a global average pooling layer to generate the first feature information.

[0152] Optionally, the operation of determining the abnormal temperature value of the energy device according to the device temperature information through the Markov chain includes: constructing the state space and transfer matrix of the Markov chain according to the temperature state of the energy device; collecting sample temperature information of the energy device in a preset period, and determining the parameter value of the transfer matrix according to the sample temperature information; determining the parameter value of the state space according to the transfer matrix when the Markov chain converges, and determining the probability of the temperature rise of the energy device according to the parameter value of the state space; and determining the abnormal temperature value of the energy device according to the device temperature information collected at the current moment and the probability of temperature rise through the temperature monitoring model.

[0153] Optionally, the operation of determining the degree of equipment aging of energy equipment based on equipment aging information through an information analysis architecture includes: segmenting the equipment aging information, and generating word vectors based on the segmentation using a word embedding model; semantically processing the word vectors through BERT to generate an information sequence; performing a convolution operation on the information sequence through a convolution layer to generate a third feature map; processing the third feature map through a second fully connected layer to generate fifth feature information; and classifying the fifth feature information through a second binary classification model to determine the degree of equipment aging of the energy equipment.

[0154] Thus, according to this embodiment, the server determines the corresponding equipment load rate, equipment temperature abnormal value and equipment aging degree by analyzing and processing the monitoring information such as equipment load information, equipment temperature information and equipment aging information in sequence. And the server also generates a hidden danger coefficient value based on the equipment load rate, equipment temperature abnormal value and equipment aging degree, so as to update the hidden danger threshold range in real time according to the hidden danger coefficient value. Thus, compared with the early warning mechanism that relies on static thresholds in the prior art, this technical solution can more sensitively capture subtle changes in the equipment operation status by introducing the hidden danger coefficient value to dynamically adjust the initial hidden danger threshold range, thereby achieving more accurate and timely early warning and effectively preventing equipment failure. And this technical solution determines whether the equipment load, equipment temperature and equipment aging need early warning in sequence according to the updated hidden danger threshold range, thereby overcoming the limitations of the traditional system in multi-dimensional analysis, and realizing comprehensive monitoring of the operation status of energy equipment by monitoring various dimensions such as equipment load information, equipment temperature information and equipment aging information. The use of multi-dimensional monitoring helps to understand the health status of the equipment more deeply, making the monitoring more comprehensive and reducing limitations. In addition, this technical solution uses machine learning methods such as logistic regression models to intelligently judge and predict the status of energy equipment, so that it can respond quickly to real-time data and improve the efficiency of early warning. This solves the technical problems of high early warning limitations caused by static threshold early warning methods and single-dimensional early warnings in the existing technology.

[0155] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0156] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0157] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0158] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0159] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0160] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0161] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An early warning method for energy equipment, characterized in that: include: Collecting monitoring information of energy equipment, wherein the monitoring information includes equipment load information, equipment temperature information and equipment aging information; Analyze the monitoring information to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree; Determining a hidden danger coefficient value according to the equipment load rate, the equipment temperature abnormality value and the equipment aging degree through a coefficient adjustment model; Using the hidden danger coefficient value to adjust the initial hidden danger threshold range, and respectively determine updated hidden danger threshold ranges corresponding to the equipment load and the equipment temperature value; and The current device load and the device temperature value are compared with the corresponding update potential risk threshold range, and an early warning is issued when the update potential risk threshold range is exceeded.

2. The method according to claim 1, characterized in that: The operation of analyzing the monitoring information to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree includes: Analyzing the equipment load information through a load monitoring model to determine the equipment load rate of the energy equipment, wherein the equipment load information includes ambient temperature, equipment noise, and equipment thermal infrared imaging; Determining an abnormal device temperature value of the energy device according to the device temperature information through a Markov chain; and The equipment aging degree of the energy equipment is determined according to the equipment aging information through an information analysis architecture.

3. The method according to claim 2, characterized in that The operation of analyzing the equipment load information through a load monitoring model to determine the equipment load rate of the energy equipment includes: Extracting features from the thermal infrared imaging of the device by a feature extraction module to generate first feature information; The first feature information, the second feature information corresponding to the ambient temperature, and the third feature information corresponding to the device noise are fused by a first feature fusion module to generate first fused feature information; Processing the first fused feature information through a first fully connected layer to generate fourth feature information; and The fourth characteristic information is classified and processed by using a first binary classification model to determine the equipment load rate of the energy equipment.

4. The method according to claim 3, characterized in that The operation of extracting features from the thermal infrared imaging of the device by a feature extraction module to generate first feature information includes: Extracting features from the thermal infrared imaging of the device using a convolutional neural network to generate a first feature map; Extracting features from the thermal infrared imaging of the device through a residual network to generate a second feature map; fusing the first feature map and the second feature map through a second feature fusion module to generate second fused feature information; and The second fused feature information is pooled through a global average pooling layer to generate the first feature information.

5. The method according to claim 2, characterized in that: The operation of determining the abnormal value of the device temperature of the energy device according to the device temperature information through a Markov chain includes: According to the temperature state of the energy device, construct the state space and transfer matrix of the Markov chain; Collecting sample temperature information of the energy device in a preset period, and determining parameter values ​​of the transfer matrix according to the sample temperature information; Determining, according to the transfer matrix, a parameter value of the state space when the Markov chain converges, and determining a probability of a temperature rise of the energy device according to the parameter value of the state space; and The device temperature abnormality value of the energy device is determined by a temperature monitoring model according to the device temperature information collected at the current moment and the probability of the temperature rising.

6. The method according to claim 2, characterized in that The operation of determining the degree of aging of the energy equipment according to the equipment aging information through the information analysis architecture includes: Segmenting the device aging information, and generating word vectors according to the segmented words using a word embedding model; Perform semantic processing on the word vectors through BERT to generate information sequences; Performing a convolution operation on the information sequence through a convolution layer to generate a third feature map; Processing the third feature map through a second fully connected layer to generate fifth feature information; and The fifth characteristic information is classified by a second binary classification model to determine the degree of aging of the energy equipment.

7. An early warning device for energy equipment, characterized in that: include: An information collection module is used to collect monitoring information of energy equipment, wherein the monitoring information includes equipment load information, equipment temperature information and equipment aging information; An information analysis module is used to analyze the monitoring information to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree; A coefficient determination module, used to determine the hidden danger coefficient value according to the equipment load rate, the equipment temperature abnormality value and the equipment aging degree through a coefficient adjustment model; A range determination module, used to adjust the initial hidden danger threshold range by using the hidden danger coefficient value, and respectively determine the updated hidden danger threshold range corresponding to the equipment load and the equipment temperature value; as well as The early warning module is used to compare the current device load and the device temperature value with the corresponding update potential risk threshold range, and issue an early warning when the update potential risk threshold range is exceeded.

8. An early warning device for energy equipment, characterized in that: include: processor; as well as A memory, connected to the processor, configured to provide the processor with instructions for processing the following processing steps: Collecting monitoring information of energy equipment, wherein the monitoring information includes equipment load information, equipment temperature information and equipment aging information; Analyze the monitoring information to determine the equipment load rate, equipment temperature abnormality value and equipment aging degree; Determining a hidden danger coefficient value according to the equipment load rate, the equipment temperature abnormality value and the equipment aging degree through a coefficient adjustment model; Using the hidden danger coefficient value to adjust the initial hidden danger threshold range, and respectively determine updated hidden danger threshold ranges corresponding to the equipment load and the equipment temperature value; and The current device load and the device temperature value are compared with the corresponding update potential risk threshold range, and an early warning is issued when the update potential risk threshold range is exceeded.

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

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

Images