Outgoing cable wiring pile head temperature early warning method and system based on edge calculation

Abstract

The embodiment of the invention provides an outgoing cable wiring pile head temperature early warning method and system based on edge calculation, and relates to the technical field of electric power detection. The temperature early warning method comprises the following steps: acquiring the joint temperature of a cable, the current of the cable, the temperature and humidity in a power distribution cabinet, and the tightness of a cable joint nut measured by fastening last time; preprocessing the joint temperature, the current, the temperature in the power distribution cabinet and the humidity in the power distribution cabinet to obtain data meeting requirements; substituting the data meeting the requirements into a temperature prediction model to obtain a predicted temperature; and obtaining the predicted temperature and carrying out early warning according to the predicted temperature. According to the invention, the temperature change of the cable wiring pile head can be monitored in real time, the temperature rise trend is accurately predicted, timely and effective temperature early warning is realized, and complete and stable operation of a power system is ensured.

Description

Technical Field

[0001] The present invention relates to the field of power detection technology, and in particular to an outgoing cable connection pile temperature early warning method and system based on edge computing. Background Art

[0002] In power systems, temperature changes at the terminal ends of outgoing cables are an important indicator of their operating status. Excessively high temperatures can cause cable insulation aging, short circuits, and other faults, seriously affecting the safe and stable operation of the power system. Traditional temperature monitoring relies on manually operated infrared temperature guns, which is labor-intensive and not only labor-intensive, but also particularly labor-intensive when there are many monitoring points and a high monitoring frequency. Manual monitoring cannot achieve continuous real-time monitoring, resulting in the inability to promptly detect abnormal temperature rises that may occur between measurements, thereby reducing monitoring efficiency. Furthermore, due to the lack of an effective early warning mechanism, preventive measures may not be taken in a timely manner once the temperature exceeds the safety threshold, increasing the risk of cable failure. Therefore, there is an urgent need for a temperature monitoring system that can provide accurate, real-time, and reliable early warnings. Summary of the Invention

[0003] The purpose of the embodiments of the present invention is to provide a temperature warning method and system for the outgoing cable terminal head based on edge computing. The present invention can monitor the temperature changes of the cable terminal head in real time, accurately predict the temperature rise trend, realize timely and effective temperature warning, and ensure the completely stable operation of the power system.

[0004] In order to achieve the above objectives, the present invention provides an outgoing cable terminal head temperature early warning method based on edge computing, comprising: Obtain the cable connector temperature, cable current, distribution cabinet temperature and humidity, the tightness of the cable connector nut measured during the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located; Preprocessing the joint temperature, current, temperature inside the distribution cabinet, and humidity inside the distribution cabinet to obtain data that meets the requirements; Bring the data that meets the requirements into the temperature prediction model to obtain the predicted temperature; The predicted temperature is obtained and an early warning is issued according to the predicted temperature.

[0005] Optionally, data that meets the requirements is fed into a temperature prediction model to obtain a predicted temperature, including: Obtain data that meets the requirements; The data meeting the requirements are respectively input into the random forest model and the vector autoregression model included in the temperature prediction model to obtain the corresponding first temperature and second temperature predicted at the next moment; Dynamically adjust the weights of the first temperature and the second temperature based on the connector temperature, current, and humidity in the distribution cabinet in the data that meets the requirements; The temperature prediction model obtains the predicted temperature according to the adjusted weights of the first temperature and the second temperature using formula (1): , formula (1) in, is the first temperature weight, To predict temperature, is the second temperature weight, is the first temperature, is the second temperature.

[0006] Optionally, the data meeting the requirements are respectively fed into a random forest model and a vector autoregressive model included in the temperature prediction model to obtain corresponding predictions of the first temperature and the second temperature at the next moment, including: Obtain the required data, including the connector temperature, cable current, temperature and humidity inside the distribution cabinet, tightness of the cable connector nut measured during the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located; The connector temperature, cable current, distribution cabinet temperature and humidity, the tightness of the cable connector nut measured at the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located in the data that meet the requirements are fed into the random forest model and the first temperature at the next moment is obtained through formula (2): , formula (2) in, is the first temperature, is the random forest model, is the temperature of the cable joint at the current moment, is the current in the cable at the current moment, is the temperature inside the distribution cabinet of the cable at the current moment, is the humidity inside the distribution cabinet of the cable at the current moment, is the tightness of the nut of the cable connector measured when it was last tightened, The number of cables to be spliced ​​in the distribution cabinet for the cable; The connector temperature, cable current, distribution cabinet temperature and humidity, the tightness of the cable connector nut measured at the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located in the data that meet the requirements are fed into the vector autoregression model and the second temperature at the next moment is obtained through formula (3): , formula (3) in, is the second temperature, is a constant term, is the autoregressive coefficient of temperature, is the autoregressive coefficient of current, is the autoregressive coefficient of the temperature inside the distribution cabinet, is the autoregressive coefficient of humidity in the distribution cabinet, is the coefficient of nut tightness, is the coefficient for the number of cable splices, is the error term, is the lag order of temperature history data, is the lag order of the current history data, is the time index of the temperature variable, is the time index of the current variable, is the lag order of the distribution cabinet temperature history data, is the time index of the temperature variable of the power distribution cabinet, is the lag order of the historical data of humidity in the power distribution cabinet, The time index of the humidity variable of the power distribution cabinet.

[0007] Optionally, a temperature prediction model is constructed, including: Training the temperature prediction model according to the acquired data; Construct a minimization loss function and obtain the loss function value; Determine whether the loss difference between the current loss function value and the previous loss function value is less than a preset difference threshold; When the difference is greater than the preset threshold, adjusting the parameters of the temperature prediction model and returning to the step of training the temperature prediction model according to the acquired data; When the difference is less than the preset threshold, the temperature prediction model after adjusting the parameters is output.

[0008] Optionally, dynamically adjusting the weights of the first temperature and the second temperature based on the connector temperature, current, and humidity in the power distribution cabinet in the data that meets the requirements includes: Obtaining, as input features, a temperature mean value in a time window, a fluctuation amplitude variance in the time window, a temperature change trend slope in the time window, a current fluctuation variance of a current load change in the time window, and a humidity fluctuation variance of a humidity change in the time window; Constructing an input linear model based on the input features; Construct ReLU activation function based on hidden features; Constructing an output linear model based on the input linear model and the ReLU activation function; Determine whether the fluctuation of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window is greater than the fluctuation of the temperature change trend slope within the corresponding time window; When the fluctuation of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window is greater than the fluctuation of the slope of the temperature change trend within the corresponding time window, constructing a Softmax function according to the output linear model, obtaining the first temperature weight and the second temperature weight, and increasing the second temperature weight and decreasing the first temperature weight; When the fluctuations of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window are less than the fluctuations of the slope of the temperature change trend within the corresponding time window, a Softmax function is constructed according to the output linear model to obtain the first temperature weight and the second temperature weight, and the first temperature weight is increased and the second temperature weight is reduced.

[0009] Optionally, the first temperature weight and the second temperature weight meet the constraint conditions shown in formula (4): , formula (4).

[0010] Optionally, obtaining the predicted temperature and issuing an early warning according to the predicted temperature includes: Obtaining the predicted temperature under the current working condition and the historical average temperature under the same working condition according to the temperature prediction model; Determine whether the absolute value of the temperature difference between the predicted temperature and the historical average temperature is greater than a set alarm threshold; If the value is greater than the alarm threshold, an alarm is triggered.

[0011] Optionally, obtaining a loss function value and a judgment difference threshold includes: According to formula (5), the loss function value is obtained. , formula (5) in, is the loss function value, For the Samples in The measured temperature at the time, For the Samples in The predicted temperature at the time, is the number of training samples; According to formula (6), it is determined whether the difference between the loss function value at the current moment and the loss function value at the previous moment is less than the preset difference threshold. , formula (6) in, is the loss function value at the current moment, is the previous loss function value, is the difference threshold.

[0012] Optionally, obtaining the absolute value of the temperature difference includes: According to formula (7), the absolute value of the temperature difference is obtained. , formula (7) in, is the historical average temperature under the same working conditions, is the predicted temperature of the model at the current moment, The set alarm threshold.

[0013] On the other hand, the present invention also provides an outgoing cable terminal head temperature warning system based on edge computing, the temperature warning system comprising: The data acquisition and processing module is used to obtain the cable joint temperature, cable current, temperature and humidity inside the distribution cabinet, and the tightness of the cable joint nut measured during the last tightening, and perform preprocessing; The temperature prediction module is used to bring the data that meets the requirements into the temperature prediction model to obtain the predicted temperature; The real-time monitoring and early warning module is used to obtain the predicted temperature and issue an early warning based on the predicted temperature.

[0014] Through the above technical solution, the present invention provides a method and system for warning the temperature of the outgoing cable connection pile head based on edge computing. By obtaining the cable joint temperature, the cable current, the temperature and humidity in the distribution cabinet, and the tightness of the cable joint nut measured at the last tightening, the joint temperature, current, temperature and humidity in the distribution cabinet are pre-processed to obtain data that meets the requirements, and the data that meets the requirements is brought into the temperature prediction model to obtain the predicted temperature, obtain the predicted temperature and issue an early warning based on the predicted temperature. The present invention can monitor the temperature changes of the cable connection pile head in real time, accurately predict the temperature rise trend according to the temperature prediction model, realize timely and effective temperature warning, and ensure the complete and stable operation of the power system.

[0015] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following detailed description, they are used to explain the embodiments of the present invention, but do not constitute a limitation of the embodiments of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of a temperature warning method for an outgoing cable terminal head based on edge computing according to an embodiment of the present invention; Figure 2 is a flow chart for obtaining predicted temperature according to one embodiment of the present invention; Figure 3 A flowchart of obtaining a corresponding predicted first temperature and second temperature at a next moment according to an embodiment of the present invention; Figure 4 is a flow chart of constructing a temperature prediction model according to one embodiment of the present invention; Figure 5 is a flow chart of dynamically adjusting the weights of the first temperature and the second temperature according to one embodiment of the present invention; Figure 6 This is a flowchart of issuing an early warning based on predicted temperature according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following describes the specific implementation of the embodiment of the present invention in detail with reference to the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the embodiment of the present invention and is not used to limit the embodiment of the present invention.

[0018] In the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned. They should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0019] like Figure 1 The figure shows a flow chart of a temperature warning method for the outgoing cable terminal head based on edge computing according to an embodiment of the present invention. Figure 1 In the embodiment of the present invention, the temperature early warning method may include: In step S1, the cable joint temperature, the cable current, the temperature and humidity inside the distribution cabinet, the tightness of the cable joint nut measured during the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located are obtained; In step S2, the joint temperature, current, temperature inside the distribution cabinet, and humidity inside the distribution cabinet are preprocessed to obtain data that meets the requirements; In step S3, the data that meets the requirements is brought into the temperature prediction model to obtain the predicted temperature; In step S4, the predicted temperature is obtained and an early warning is issued according to the predicted temperature.

[0020] In this Figure 1In the method shown, step S1 can be used to obtain the cable joint temperature, the cable current, the temperature and humidity inside the distribution cabinet, the tightness of the cable joint nut measured during the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located. The cable joint temperature is obtained by a temperature sensor, the cable current is obtained by an ammeter, the temperature and humidity inside the distribution cabinet are obtained by an ambient temperature and humidity detection system, and the tightness of the cable joint nut measured during the last tightening is updated after each tightening operation and remains unchanged until the next tightening operation. Step S2 can be used to pre-process the joint temperature, current, temperature inside the distribution cabinet, and humidity inside the distribution cabinet to obtain data that meets the requirements. Step S3 can be used to bring the data that meets the requirements into a temperature prediction model to obtain a predicted temperature. Specific methods for obtaining the predicted temperature in a temperature prediction model may include the following: Figure 2 The steps shown. Specifically: In step S11, data meeting the requirements is obtained; In step S12, the data that meets the requirements are respectively input into the random forest model and the vector autoregression model included in the temperature prediction model to obtain the corresponding predicted first temperature and second temperature at the next moment; In step S13, the weights of the first temperature and the second temperature are dynamically adjusted according to the joint temperature, current, and humidity in the power distribution cabinet in the data that meets the requirements; In step S14, the temperature prediction model obtains the predicted temperature according to the adjusted weights of the first temperature and the second temperature using formula (1): , formula (1) in, is the first temperature weight, To predict temperature, is the second temperature weight, is the first temperature, is the second temperature.

[0021] In this Figure 2 In the method shown, step S11 can be used to obtain data that meets the requirements. Step S12 can be used to input the data that meets the requirements into the random forest model and the vector autoregressive model included in the temperature prediction model to obtain the corresponding first temperature and second temperature predicted at the next moment. The specific method for obtaining the corresponding first temperature and second temperature predicted at the next moment can be a known form known to those skilled in the art. In one example of the present invention, the method for obtaining the corresponding first temperature and second temperature predicted at the next moment may include the following: Figure 3 The steps shown. Specifically: In step S21, the connector temperature, cable current, distribution cabinet temperature and humidity, tightness of the cable connector nut measured during the last tightening, and the number of cables connected to the distribution cabinet where the cable is located are obtained from the data that meets the requirements; In step S22, the connector temperature, cable current, distribution cabinet temperature and humidity, the tightness of the cable connector nut measured at the last tightening, and the number of cables spliced ​​in the distribution cabinet that meet the requirements are fed into the random forest model and the first temperature at the next moment is obtained using formula (2): , formula (2) in, is the first temperature, is the random forest model, is the temperature of the cable joint at the current moment, is the current in the cable at the current moment, is the temperature inside the distribution cabinet of the cable at the current moment, is the humidity inside the distribution cabinet of the cable at the current moment, is the tightness of the nut of the cable connector measured when it was last tightened, The number of cables to be spliced ​​in the distribution cabinet for the cable; In step S23, the connector temperature, cable current, distribution cabinet temperature and humidity, the tightness of the cable connector nut measured at the last tightening, and the number of cables spliced ​​in the distribution cabinet that meet the requirements are fed into the vector autoregressive model and the second temperature at the next moment is obtained through formula (3): , formula (3) in, is the second temperature, is a constant term, is the autoregressive coefficient of temperature, is the autoregressive coefficient of current, is the autoregressive coefficient of the temperature inside the distribution cabinet, is the autoregressive coefficient of humidity in the distribution cabinet, is the coefficient of nut tightness, is the coefficient for the number of cable splices, is the error term, is the lag order of temperature history data, is the lag order of the current history data, is the time index of the temperature variable, is the time index of the current variable, is the lag order of the distribution cabinet temperature history data, is the time index of the temperature variable of the power distribution cabinet, is the lag order of the historical data of humidity in the power distribution cabinet, The time index of the humidity variable of the power distribution cabinet.

[0022] The specific construction method of the temperature prediction model input can be various forms known to those skilled in the art. In one embodiment of the present invention, the method of constructing the temperature prediction model may include: Figure 4 The steps shown. Specifically: In step S31, the temperature prediction model is trained based on the acquired data; In step S32, a loss minimization function is constructed and a loss function value is obtained; In step S33, it is determined whether the loss difference between the loss function value at the current moment and the loss function value at the previous moment is less than a preset difference threshold; In step S34, if the difference is greater than a preset threshold, the parameters of the temperature prediction model are adjusted, and the process returns to the step of training the temperature prediction model based on the acquired data; In step S35 , when the difference is less than a preset threshold, the temperature prediction model after adjusting the parameters is output.

[0023] In this Figure 4 In the method shown, step S31 can be used to train the temperature prediction model based on the acquired data. Step S32 can be used to construct a minimization loss function and obtain the loss function value. The specific method for obtaining the loss function value can be various forms known to those skilled in the art. In one example of the present invention, the specific method for obtaining the loss function value can be obtained according to formula (5), , formula (5) in, is the loss function value, For the Samples in The measured temperature at the time, For the Samples in The predicted temperature at the time, is the number of training samples; Step S33 can be used to determine whether the loss difference between the loss function value at the current moment and the loss function value at the previous moment is less than a preset difference threshold. The specific method for determining the difference threshold can be various forms known to those skilled in the art. In one example of the present invention, the specific method for determining the difference threshold can be to determine whether the difference between the loss function value at the current moment and the loss function value at the previous moment is less than a preset difference threshold according to formula (6). , formula (6) in, is the loss function value at the current moment, is the previous loss function value, is the difference threshold.

[0024] Step S34 can be used to adjust the parameters of the temperature prediction model when the difference is greater than a preset threshold value, and return to the step of training the temperature prediction model based on the acquired data. Step S35 can be used to output the temperature prediction model after adjusting the parameters when the difference is less than a preset threshold value. According to step S32, a minimization loss function is constructed, and then step S33 is used to determine whether the loss difference between the loss function value at the current moment and the loss function value at the previous moment is less than a preset threshold value. Step S34 is used to iterate based on the judgment result of the difference threshold value. After reaching the convergence result of step S61, the temperature prediction model after adjusting the parameters is output according to step S35.

[0025] Step S13 can be used to dynamically adjust the weights of the first temperature and the second temperature according to the joint temperature, current and humidity in the distribution cabinet in the data that meets the requirements. The specific method for dynamically adjusting the weights of the first temperature and the second temperature can be various forms known to those skilled in the art. In one example of the present invention, the specific method for dynamic adjustment may include: Figure 5 The steps shown. Specifically: In step S41, the temperature mean value in a time window, the fluctuation amplitude variance in the time window, the temperature change trend slope in the time window, the current fluctuation variance of the current load change in the time window, and the humidity fluctuation variance of the humidity change in the time window are obtained as input features; In step S42, an input linear model is constructed based on the input features; In step S43, a ReLU activation function is constructed based on the hidden features; In step S44, an output linear model is constructed based on the input linear model and the ReLU activation function; In step S45, it is determined whether the fluctuation of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window is greater than the fluctuation of the temperature change trend slope within the corresponding time window; In step S46, when the fluctuations of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window are greater than the fluctuations of the temperature change trend slope within the corresponding time window, a Softmax function is constructed according to the output linear model to obtain the first temperature weight and the second temperature weight, and the second temperature weight is increased and the first temperature weight is decreased; In step S47, when the fluctuations of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window are less than the fluctuations of the slope of the temperature change trend within the corresponding time window, a Softmax function is constructed according to the output linear model to obtain the first temperature weight and the second temperature weight, and the first temperature weight is increased and the second temperature weight is reduced.

[0026] In this Figure 5 In the method shown, step S41 can be used to obtain the temperature mean in a time window, the fluctuation amplitude variance in the time window, the temperature change trend slope in the time window, the current fluctuation variance of the current load change in the time window, and the humidity fluctuation variance of the humidity change in the time window as input features.

[0027] The slope of the temperature change trend within the time window can be expressed by formula (8): , formula (8) in, is the temperature change trend slope within the time window. A positive value indicates an upward trend, and a negative value indicates a downward trend. For time point temperature, For time point and temperature The covariance of For time point The covariance of is the time window.

[0028] Step S42 can be used to construct an input linear model based on the input features. Step S43 can be used to construct a ReLU activation function based on the hidden features. The input features obtained in step S41 are used as the 5 nodes of the input layer, and the 3 nodes of the hidden layer are determined according to step S43. The input features are determined by formula (9), , formula (9) in, is the input feature, is the mean temperature of the time window, is the fluctuation amplitude variance within the time window, is the current fluctuation variance of the current load change within the time window, is the humidity fluctuation variance of the humidity change within the time window.

[0029] According to formula (9), the input linear model of formula (10) is constructed. , formula (10) in, is the weight matrix of dimension 5×3 in formula (11), which represents the connection weight from the input layer to the hidden layer. , formula (11) is the bias vector of dimension 3×1 in formula (12), which represents the bias of each hidden node. , formula (12) is the input vector of the hidden layer, according to formula (13) and formula (14), , formula (13) , formula (14).

[0030] In step S43, the ReLU activation function can be constructed by formula (15) to formula (17) to determine the three nodes of the hidden layer. , formula (15) , formula (16) , formula (17) in, is the hidden layer vector, According to the input vector Constructed ReLU activation function.

[0031] Step S44 may be used to construct an output linear model based on the input linear model and the ReLU activation function.

[0032] According to formula (18), the output linear model is constructed. , formula (18) in, is the weight matrix of dimension 3×2 in formula (19), which represents the connection weight from the hidden layer to the output layer. , formula (19) is the bias vector of dimension 2×1 in formula (20), which represents the bias of each output node. , formula (20) The input vector for the output layer is expressed according to formula (21) and formula (22).

[0033] , formula (21) , formula (22).

[0034] Among them, the weight matrix initialization method adopts the He initialization strategy, and the weight is generated according to the normal distribution of formula (23). , formula (23) in, is the number of input neurons, of =5 (the number of input features is 5), of =3 (number of hidden layer nodes is 3), bias vector , initialized to 0.

[0035] Step S45 may be used to determine whether the fluctuations of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window are greater than the fluctuations of the temperature change trend slope within the corresponding time window. Step S46 may be used to construct a Softmax function based on the output linear model, obtain a first temperature weight and a second temperature weight, increase the second temperature weight, and decrease the first temperature weight if the fluctuations of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window are greater than the fluctuations of the temperature change trend slope within the corresponding time window.

[0036] Construct the Softmax function according to formula (24) and formula (25), , formula (24) , formula (25) in, , is the first temperature weight, is the second temperature weight, and the first temperature weight and the second temperature weight meet the constraints shown in formula (4): , formula (4).

[0037] Step S47 can be used to construct a Softmax function according to the output linear model when the fluctuation of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window is less than the fluctuation of the slope of the temperature change trend within the corresponding time window, obtain the first temperature weight and the second temperature weight, increase the first temperature weight, and reduce the second temperature weight.

[0038] Step S14 can be used for the temperature prediction model to obtain the predicted temperature according to the adjusted weights of the first temperature and the second temperature using formula (1): , formula (1) in, is the first temperature weight, To predict temperature, is the second temperature weight, is the first temperature, is the second temperature.

[0039] Step S4 can be used to obtain the predicted temperature and issue an early warning based on the predicted temperature. The specific method for issuing an early warning based on the predicted temperature can be a known form known to those skilled in the art. In one example of the present invention, the method for issuing an early warning based on the predicted temperature may include: Figure 6 The steps shown. Specifically: In step S51, the predicted temperature under the current working condition and the historical average temperature under the same working condition are obtained according to the temperature prediction model; In step S52, it is determined whether the absolute value of the temperature difference between the predicted temperature and the historical average temperature is greater than the set alarm threshold; In step S53, if it is determined that the value is greater than the alarm threshold, an alarm is triggered.

[0040] In this Figure 6 In the method shown, step S51 can be used to obtain the predicted temperature under the current working conditions and the historical average temperature under the same working conditions based on the temperature prediction model. Step S52 can be used to determine whether the absolute value of the temperature difference between the predicted temperature and the historical average temperature is greater than the set alarm threshold. The specific method for obtaining the absolute value of the temperature difference can be various forms known to those skilled in the art. In one example of the present invention, the absolute value of the temperature difference can be obtained according to formula (7), , formula (7) in, is the historical average temperature under the same working conditions, is the predicted temperature of the model at the current moment, The set alarm threshold.

[0041] The historical average temperature under the same working conditions can be calculated according to formula (26): , formula (26) in, is the number of historical data points under the same working conditions, For the The connector temperature of the cable for 3 historical data points.

[0042] Step S53 can be used to trigger an alarm when it is determined that the value is greater than the alarm threshold.

[0043] On the other hand, the present invention also provides an outgoing cable terminal head temperature warning system based on edge computing, the temperature warning system comprising: The data acquisition and processing module is used to obtain the cable joint temperature, cable current, temperature and humidity inside the distribution cabinet, and the tightness of the cable joint nut measured during the last tightening, and perform preprocessing; The temperature prediction module is used to bring the data that meets the requirements into the temperature prediction model to obtain the predicted temperature; The real-time monitoring and early warning module is used to obtain the predicted temperature and issue early warnings based on the predicted temperature.

[0044] Through the above technical solution, the present invention provides a method and system for warning the temperature of the outgoing cable connection pile head based on edge computing. By obtaining the cable joint temperature, the cable current, the temperature and humidity in the distribution cabinet, and the tightness of the cable joint nut measured at the last tightening, the joint temperature, current, temperature and humidity in the distribution cabinet are pre-processed to obtain data that meets the requirements, and the data that meets the requirements is brought into the temperature prediction model to obtain the predicted temperature, obtain the predicted temperature and issue an early warning based on the predicted temperature. The present invention can monitor the temperature changes of the cable connection pile head in real time, accurately predict the temperature rise trend according to the temperature prediction model, realize timely and effective temperature warning, and ensure the complete and stable operation of the power system.

[0045] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0046] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0047] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0048] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0049] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0050] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0051] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0052] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0053] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A temperature warning method for outgoing cable connection piles based on edge computing, characterized in that: The temperature early warning method comprises: Obtain the cable connector temperature, cable current, distribution cabinet temperature and humidity, the tightness of the cable connector nut measured during the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located; Preprocessing the joint temperature, current, temperature inside the distribution cabinet, and humidity inside the distribution cabinet to obtain data that meets the requirements; Bring the data that meets the requirements into the temperature prediction model to obtain the predicted temperature; The predicted temperature is obtained and an early warning is issued according to the predicted temperature.

2. The temperature early warning method according to claim 1, characterized in that: Submit the required data into the temperature prediction model to obtain the predicted temperature, including: Obtain data that meets the requirements; The data meeting the requirements are respectively input into the random forest model and the vector autoregression model included in the temperature prediction model to obtain the corresponding first temperature and second temperature predicted at the next moment; Dynamically adjust the weights of the first temperature and the second temperature based on the connector temperature, current, and humidity in the distribution cabinet in the data that meets the requirements; The temperature prediction model obtains the predicted temperature according to the adjusted weights of the first temperature and the second temperature using formula (1): , Formula (1) in, is the first temperature weight, To predict temperature, is the second temperature weight, is the first temperature, is the second temperature.

3. The temperature early warning method according to claim 2, characterized in that: The data meeting the requirements are respectively fed into the random forest model and the vector autoregressive model included in the temperature prediction model to obtain the corresponding predicted first temperature and second temperature at the next moment, including: Obtain the required data, including the connector temperature, cable current, temperature and humidity inside the distribution cabinet, tightness of the cable connector nut measured during the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located; The connector temperature, cable current, distribution cabinet temperature and humidity, the tightness of the cable connector nut measured at the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located in the data that meet the requirements are fed into the random forest model and the first temperature at the next moment is obtained through formula (2): , Formula (2) in, is the first temperature, is the random forest model, is the temperature of the cable joint at the current moment, is the current in the cable at the current moment, is the temperature inside the distribution cabinet of the cable at the current moment, is the humidity inside the distribution cabinet of the cable at the current moment, is the tightness of the nut of the cable connector measured when it was last tightened, The number of cables to be spliced ​​in the distribution cabinet for the cable; The connector temperature, cable current, distribution cabinet temperature and humidity, the tightness of the cable connector nut measured at the last tightening, and the number of cables spliced ​​in the distribution cabinet where the cable is located in the data that meet the requirements are fed into the vector autoregression model and the second temperature at the next moment is obtained through formula (3): , Formula (3) in, is the second temperature, is a constant term, is the autoregressive coefficient of temperature, is the autoregressive coefficient of current, is the autoregressive coefficient of the temperature inside the distribution cabinet, is the autoregressive coefficient of humidity in the distribution cabinet, is the coefficient of nut tightness, is the coefficient for the number of cable splices, is the error term, is the lag order of temperature history data, is the lag order of the current history data, is the time index of the temperature variable, is the time index of the current variable, is the lag order of the distribution cabinet temperature history data, is the time index of the temperature variable of the power distribution cabinet, is the lag order of the historical data of humidity in the power distribution cabinet, The time index of the humidity variable of the power distribution cabinet.

4. The temperature early warning method according to claim 3, characterized in that: Build a temperature prediction model, including: Training the temperature prediction model according to the acquired data; Construct a minimization loss function and obtain the loss function value; Determine whether the loss difference between the current loss function value and the previous loss function value is less than a preset difference threshold; When the difference is greater than the preset threshold, adjusting the parameters of the temperature prediction model and returning to the step of training the temperature prediction model according to the acquired data; When the difference is less than the preset threshold, the temperature prediction model after adjusting the parameters is output.

5. The temperature early warning method according to claim 2, characterized in that: The weights of the first and second temperatures are dynamically adjusted based on the connector temperature, current, and humidity in the distribution cabinet in the data that meets the requirements, including: Obtaining, as input features, a temperature mean value in a time window, a fluctuation amplitude variance in the time window, a temperature change trend slope in the time window, a current fluctuation variance of a current load change in the time window, and a humidity fluctuation variance of a humidity change in the time window; Constructing an input linear model based on the input features; Construct ReLU activation function based on hidden features; Constructing an output linear model based on the input linear model and the ReLU activation function; Determine whether the fluctuation of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window is greater than the fluctuation of the temperature change trend slope within the corresponding time window; When the fluctuation of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window is greater than the fluctuation of the slope of the temperature change trend within the corresponding time window, constructing a Softmax function according to the output linear model, obtaining the first temperature weight and the second temperature weight, and increasing the second temperature weight and decreasing the first temperature weight; When the fluctuations of the current fluctuation variance and the humidity fluctuation variance of the current load change within the time window are less than the fluctuations of the slope of the temperature change trend within the corresponding time window, a Softmax function is constructed according to the output linear model to obtain the first temperature weight and the second temperature weight, and the first temperature weight is increased and the second temperature weight is reduced.

6. The temperature early warning method according to claim 5, characterized in that: The first temperature weight and the second temperature weight meet the constraints shown in formula (4): , formula (4).

7. The temperature early warning method according to claim 1, characterized in that: Obtaining the predicted temperature and issuing an early warning based on the predicted temperature includes: Obtaining the predicted temperature under the current working condition and the historical average temperature under the same working condition according to the temperature prediction model; Determine whether the absolute value of the temperature difference between the predicted temperature and the historical average temperature is greater than a set alarm threshold; If the value is greater than the alarm threshold, an alarm is triggered.

8. The temperature early warning method according to claim 4, characterized in that: Obtain the loss function value and judgment difference threshold, including; According to formula (5), the loss function value is obtained. , Formula (5) in, is the loss function value, For the Samples in The measured temperature at the time, For the Samples in The predicted temperature at the time, is the number of training samples; According to formula (6), it is determined whether the difference between the loss function value at the current moment and the loss function value at the previous moment is less than the preset difference threshold. , Formula (6) in, is the loss function value at the current moment, is the previous loss function value, is the difference threshold.

9. The temperature early warning method according to claim 7, characterized in that: Obtaining the absolute value of the temperature difference includes: According to formula (7), the absolute value of the temperature difference is obtained. , Formula (7) in, is the historical average temperature under the same working conditions, is the predicted temperature of the model at the current moment, The set alarm threshold.

10. A temperature warning system for outgoing cable connection piles based on edge computing, characterized in that: The temperature early warning system comprises: The data acquisition and processing module is used to obtain the cable joint temperature, cable current, temperature and humidity inside the distribution cabinet, and the tightness of the cable joint nut measured during the last tightening, and perform preprocessing; The temperature prediction module is used to bring the data that meets the requirements into the temperature prediction model to obtain the predicted temperature; The real-time monitoring and early warning module is used to obtain the predicted temperature and issue an early warning based on the predicted temperature.

Images