High-voltage fuse characteristic data acquisition and residual life prediction device and prediction method

Abstract

The invention discloses a high-voltage fuse characteristic data acquisition and residual life prediction device and method, and relates to the technical field of high-voltage fuse monitoring, and the device comprises a housing, a power module, a detection unit, a control unit, and a cloud platform. The power supply module is connected with a to-be-detected fuse through a pluggable terminal, directly takes power from the to-be-detected fuse, and outputs 3.3 V voltage to supply power to the optical fiber sending modules in the control unit and the detection unit. The detection unit comprises voltage detection and current detection, a detected voltage signal is sent to the control unit through the optical fiber sending module, and the internal resistance value of the fuse is obtained through calculation of the control unit. And the control unit sends the internal resistance value and the current and voltage data to a cloud platform, and predicts the internal resistance value and the residual service time of the fuse through a residual life prediction method. The system has the characteristics of high efficiency, safety and intelligence, can provide real-time monitoring, fault prediction and maintenance support for maintainers, and has a good application prospect.

Description

Technical Field

[0001] The present invention relates to the field of electrical protection devices, and in particular to a device and method for collecting characteristic data and predicting the remaining life of a high-voltage fuse. Background Art

[0002] Fuses are commonly used overcurrent protection devices in power systems and electronic equipment. Their function is to cut off the circuit by melting the fuse when an overload or short circuit occurs in the circuit, thereby protecting downstream equipment. However, when a fault occurs, traditional fuses can only provide protection by melting, and lack monitoring and diagnosis of their own status.

[0003] Before a fuse blows, its internal structure may degrade or increase its internal resistance due to long-term operation and environmental influences (such as temperature rise, current shock, etc.). This performance degradation cannot be effectively monitored by traditional methods and can easily lead to unexpected shutdown of the electrical system or damage to the equipment.

[0004] Existing fuse protection devices cannot provide feedback on their working status and cannot provide fault warning. The power system lacks the means to predict and diagnose the internal resistance of the fuse, which increases the cost of maintenance and the risk of system failure.

[0005] Therefore, there is an urgent need to develop a self-diagnostic device that can predict the change in the internal resistance of a fuse and issue an early warning before the fuse is about to fail, so as to solve the above-mentioned problems. Summary of the invention

[0006] The purpose of the present invention is to provide a device and method for collecting characteristic data and predicting remaining life of a high-voltage fuse, so as to realize feedback on the working state of the fuse; and to provide a method for predicting internal resistance value, so as to realize prediction of the internal resistance value of the fuse.

[0007] In order to solve the above technical problems, the embodiments of the present application disclose the following technical solutions:

[0008] A device for collecting characteristic data and predicting remaining life of a high-voltage fuse, characterized in that it comprises: a housing, a power module, a detection unit, a control unit, and a cloud platform;

[0009] The power supply module, the control unit, and the voltage detection module of the detection unit are placed inside the housing;

[0010] A first opening is provided on the side of the housing, and a plug-in terminal is provided at the first opening for connecting the fuse to be tested;

[0011] The power module is connected to the control unit;

[0012] The detection unit is connected to the control unit;

[0013] The cloud platform is connected to the control unit via wireless communication.

[0014] The power module includes: a plug-in terminal, a fuse unit, a transformer, a first voltage conversion chip, and a second voltage conversion chip;

[0015] The pluggable terminal is electrically connected in parallel with the fuse to be tested; the pluggable terminal is electrically connected with the insurance unit; the insurance unit is electrically connected with the transformer; the transformer is electrically connected with the first voltage conversion chip; the first voltage conversion chip is electrically connected with the second voltage conversion chip; the second voltage conversion chip is electrically connected with the control unit;

[0016] The transformer amplifies the collected voltage to obtain a first voltage; the first voltage conversion chip converts the first voltage into a second voltage; and the second voltage conversion chip converts the second voltage into a third voltage.

[0017] The insurance unit uses a 5kV fuse, which is a spring-type high-voltage fuse, the shell material of which is a glass fiber tube, and the number is at least 2.

[0018] The detection unit includes a current detection module and a voltage detection module, both of which are electrically connected to the control unit, wherein the voltage detection module is electrically connected to the transformer.

[0019] The voltage detection module of the detection unit includes: a voltage dividing circuit, a differential ADC module, and an optical fiber transmission module;

[0020] The voltage divider circuit is electrically connected in parallel with the transformer;

[0021] The voltage divider circuit is electrically connected to the differential ADC module;

[0022] The differential ADC module is electrically connected to the optical fiber transmission module;

[0023] The voltage divider circuit divides the first voltage into a fourth voltage in proportion; the differential ADC module converts the fourth voltage into a digital quantity, and sends the digital signal to the control unit through the optical fiber sending module.

[0024] The current detection module of the detection unit includes: a Rogowski coil and an integrator module; the Rogowski coil is passed through the center of the fuse to be tested; the integrator module is electrically connected to the Rogowski coil; the integrator module is electrically connected to the control unit; the Rogowski coil converts the rate of change of the alternating current into a voltage signal proportional to it; the integrator module integrates the voltage signal to obtain a voltage signal proportional to the actual alternating current.

[0025] The control unit includes: a signal receiving module, a data processing unit, a 5G communication module and an alarm module; the signal receiving module is used to receive the output signal of the detection unit; the output signal includes the voltage value of the fuse to be tested and the current value flowing through it; the data processing unit is used to analyze and process the received voltage value and current value signals of the fuse to be tested to calculate the internal resistance of the fuse; the 5G communication module wirelessly sends the fuse internal resistance, voltage and current data to the cloud platform; the alarm module analyzes and compares the status information such as voltage, current and calculated internal resistance value with a preset threshold, so as to timely perceive the abnormal status of the equipment and remind maintenance personnel to take maintenance measures through various alarm methods.

[0026] The inside of the shell is sealed with silica gel material, which can prevent the intrusion of moisture and dust and is used for electrical insulation protection.

[0027] The cloud platform includes: a data receiving module, a data storage module, and a data analysis module; the data receiving module is used to receive data from the control unit through the 5G network; the data storage module is used to store and manage the received data; the data analysis module is used to analyze and process the stored data, and use the remaining life prediction method to predict the internal resistance and remaining service life of the fuse.

[0028] The remaining life prediction method uses a neural network prediction algorithm, specifically a gradient boosting decision tree (GBDT) model, which is used to predict the change trend of the internal resistance of the fuse to be tested. When the internal resistance of the fuse exceeds 20% of its factory rated value, it is considered that the fuse life has been reached, thereby triggering an alarm mechanism to prompt the staff to handle it in time. The specific operation steps and the remaining life prediction method include the following:

[0029] (1) Initialization

[0030] System initialization includes configuration of detection unit, control unit and cloud platform;

[0031] Set the sampling frequency and threshold parameters of the current detection module and voltage detection module in the detection unit, and ensure that each module works normally;

[0032] Load the pre-trained gradient boosted decision tree (GBDT) model and set the input features of the model, including current, voltage, and surge current;

[0033] After the system is ready, it starts to monitor the electrical status of the fuse to be tested;

[0034] (2) Data collection and internal resistance measurement

[0035] ① The current detection module uses a Rogowski coil to measure the current passing through the fuse. The Rogowski coil converts the rate of change of the AC current into a voltage signal. The voltage signal is then integrated through the integrator module to obtain the actual current value, completing the conversion and calculation of the current signal;

[0036] ②Measure the voltage signal at both ends of the fuse to be tested through the voltage divider circuit in the voltage detection module;

[0037] The voltage divider circuit divides the actual voltage in proportion to generate a voltage value suitable for processing by the differential ADC module;

[0038] The differential ADC module converts the analog voltage signal into a digital signal and sends it to the control unit through the optical fiber transmission module;

[0039] ③The control unit follows Ohm's law Substitute the collected real-time voltage U and current I into the calculation formula to obtain the internal resistance value of the fuse to be tested;

[0040] ④ The control unit wirelessly sends the voltage U, current I and internal resistance data to the data storage module of the cloud platform, and the data storage module performs characterization processing on the above data;

[0041] (3) GBDT model predicts internal resistance

[0042] ① Input the characteristic data of the data storage module into the gradient boosting decision tree (GBDT) model of the data analysis module. The input features include internal resistance value, voltage U, current I, inrush current, etc. The GBDT model receives these characteristic data and processes them as model input variables. These features are used as the basis for splitting each decision tree to predict the internal resistance change trend of the fuse and save them as historical data;

[0043] ②GBDT initializes a base model by calculating the average internal resistance value of the training data. This average value is used as the starting prediction of the first tree, indicating the current preliminary estimate of the internal resistance;

[0044] ③For each sample, GBDT calculates the difference between the current predicted value and the true value. This difference is called residual. The residual reflects the error of the model in the current prediction. Subsequent decision trees will focus on fitting these residuals.

[0045] ④After calculating the residual, GBDT starts to build a new decision tree. During the construction process, a greedy algorithm is used. The model will select a feature and an optimal split threshold to minimize the mean square error in the child nodes after the split. At each node, all features and thresholds are traversed to find the feature that can maximize the information gain.

[0046] ⑤ Generate a new decision tree. GBDT adds the prediction results of this tree to the current model. This update process is affected by the learning rate. The learning rate controls the contribution of each tree to the final model. Through the learning rate, the model can gradually improve the accuracy of the prediction without overfitting.

[0047] ⑥GBDT repeats steps ③ to ⑤ until the set number of trees is reached or the residual error falls within a certain tolerance range. Each tree will correct the prediction error of the previous tree and gradually reduce the error of the internal resistance prediction. Finally, the prediction results of all decision trees will be combined to form the final prediction model.

[0048] ⑦ Input the real-time data of the current fuse into the GBDT model, calculate the remaining service life of the fuse based on the relationship between the predicted internal resistance of the fuse and the preset threshold value, and combine historical data. If the remaining service life is lower than the set value, it is recommended to replace the fuse immediately. If the remaining service life is long, but the internal resistance is close to the threshold, it is recommended to arrange regular inspections. When the internal resistance value exceeds the set threshold or the remaining service life is lower than a certain critical value, promptly notify the maintenance personnel to take maintenance measures.

[0049] The specific working process of the device and method for collecting characteristic data of high-voltage fuse and predicting remaining life is as follows:

[0050] The power module is connected to the fuse through plug-in terminals, directly drawing power from the fuse to be tested and supplying power to the entire system;

[0051] The internal series-connected fuse unit is used to protect the device from overload or short circuit;

[0052] The secondary side of the transformer converts AC power into DC power through the first voltage conversion chip, and then outputs 3.3V voltage through the second voltage conversion chip to power the control unit and the optical fiber transmission module respectively;

[0053] The Rogowski coil in the current detection part generates a voltage signal by inducing the AC current in the fuse to be tested. The voltage signal is directly proportional to the actual current.

[0054] The voltage signal is processed by the integrator module, which converts the voltage signal into a current signal that is proportional to the actual current;

[0055] The voltage detection part is measured by a differential ADC module, which is connected to the resistor of the voltage divider circuit in parallel on the secondary side of the transformer to measure the voltage;

[0056] The voltage signal acquired by the detection unit is transmitted to the control unit through the optical fiber transmission module;

[0057] The current signal acquired by the detection unit is transmitted to the control unit via the integrator module;

[0058] After receiving the voltage and current signals, the control unit performs internal calculations to obtain the internal resistance value of the fuse to be tested;

[0059] The control unit sends the internal resistance, voltage, and current data of the fuse to be tested to the cloud platform through the 5G communication module;

[0060] The cloud platform further stores, analyzes and processes this data;

[0061] The trained GBDT model is used to predict the internal resistance value and the remaining service life of the fuse.

[0062] Based on the predicted internal resistance value and the remaining usage time, determine whether it exceeds the preset threshold;

[0063] If the internal resistance value exceeds the threshold or the remaining usage time is lower than the preset value, the system will issue an alarm to prompt maintenance personnel to take necessary measures.

[0064] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0065] The present invention can monitor the current, voltage and internal resistance of the fuse, discover potential faults in time through data analysis, and reduce the risk of unexpected equipment shutdown; use the GBDT machine learning model to predict the internal resistance and remaining service life of the fuse, improve the accuracy and reliability of the measurement, and help maintenance personnel better evaluate the health status of the equipment; by monitoring and predicting the remaining service life, maintenance personnel can perform maintenance and replacement in time, thereby effectively extending the service life of the fuse and reducing equipment maintenance costs; combined with the data analysis capabilities of the cloud platform, maintenance personnel can obtain more comprehensive equipment status information, realize intelligent power equipment management, and optimize Operation and maintenance strategy; the use of silicone material for potting design allows the equipment to operate stably in harsh environments (such as humidity, high temperature, etc.), thereby improving the reliability and durability of the system; high-voltage safety is taken into consideration in the design (such as the use of 5kV fuses) to ensure the safe use of the equipment under high-voltage conditions and reduce potential electrical safety hazards; through intelligent prediction and monitoring functions, the need for manpower maintenance is reduced, and downtime caused by equipment failure is reduced, thereby effectively reducing overall operating costs; the above effects demonstrate the comprehensive advantages of the present invention and reflect its potential value in fault diagnosis, equipment management and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0067] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0068] Figure 1 It is a structural schematic diagram of a high-voltage fuse characteristic data collection and remaining life prediction device provided by an embodiment of the present invention;

[0069] Figure 2 is a schematic diagram of the structure of a power module provided in an embodiment of the present invention;

[0070] Figure 3 is a schematic diagram of the structure of a detection unit provided in an embodiment of the present invention;

[0071] Figure 4 is a GBDT flow chart provided by an embodiment of the present invention; DETAILED DESCRIPTION

[0072] In order to enable those skilled in the art to better understand the technical solutions in 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 part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application.

[0073] In order to better understand the present application, the implementation methods of the present application are explained in detail below with reference to the accompanying drawings.

[0074] The embodiment of the present invention provides a device and method for collecting characteristic data of a high-voltage fuse and predicting the remaining life. Figure 1 , Figure 1 The figure is a schematic diagram of the structure of a device for collecting characteristic data and predicting remaining life of a high-voltage fuse provided by an embodiment of the present invention. Figure 1 It can be known that the high-voltage fuse characteristic data collection and remaining life prediction device provided in the embodiment of the present invention mainly includes: a housing 10, a power module 11, a detection unit 12, a control unit 13, and a cloud platform 14.

[0075] The power module 11, the control unit 13, and the voltage detection module 118 of the detection unit 12 are placed inside the housing 10;

[0076] A first opening is formed on the side of the housing 10, and a plug-in terminal 111 is provided at the first opening for connecting a fuse to be tested;

[0077] The power module 11 is connected to the control unit 13;

[0078] The detection unit 12 is connected to the control unit 13;

[0079] The cloud platform 14 is connected to the control unit 13 via wireless communication.

[0080] Further, the present invention embodiment has understood the structure of the power module 11, see Figure 2 , Figure 2 Schematic diagram of the power module structure provided by the embodiment of the present invention. Figure 2 As shown, the power module 11 includes: a plug-in terminal 111, a fuse unit 114, a transformer 115, a first voltage conversion chip 116, and a second voltage conversion chip 117;

[0081] The pluggable terminal 111 is electrically connected in parallel with the fuse to be tested; the pluggable terminal 111 is electrically connected to the insurance unit 114, the insurance unit 114 is electrically connected to the transformer 115, the transformer 115 is electrically connected to the first voltage conversion chip 116, the first voltage conversion chip 116 is electrically connected to the second voltage conversion chip 117; the second voltage conversion chip 117 is electrically connected to the control unit 13; the transformer 115 amplifies the collected voltage to obtain a first voltage; the first voltage conversion chip 116 converts the first voltage into a second voltage; the second voltage conversion chip 117 converts the second voltage into a third voltage. Among them, the plug-in terminal 111 is connected to the 0.1V-0.5V AC voltage VDD on both sides of the fuse to be tested, the transformer 115 converts the 0.1V-0.5V AC voltage VDD on both sides of the fuse into a 1.25V-6.25V AC voltage VDD1, the first voltage conversion chip 116 converts the 1.25V-6.25V AC voltage VDD1 into a 1.25V-6.25V DC voltage VDD2, and the second voltage conversion chip 117 converts the 1.25V-6.25V DC voltage VDD2 into a 3.3V DC voltage VDD3.

[0082] The fuse unit 114 uses a 5kV fuse, which is a spring-type high-voltage fuse, and the outer shell material is a glass fiber tube, and the number is at least 2.

[0083] Further, the structure of the detection unit 12 is understood in the embodiment of the present invention, see Figure 3 , Figure 3 Schematic diagram of the structure of the detection unit provided in the embodiment of the present invention. Figure 3As shown, the detection unit 12 includes a current detection module 113 and a voltage detection module 118, both of which are electrically connected to the control unit 13, wherein the voltage detection module 118 is electrically connected to the transformer 115, and the voltage detection module 118 of the detection unit 12 includes: a voltage divider circuit 119, a differential ADC module 120, and an optical fiber transmission module 121; the voltage divider circuit 119 is electrically connected to the transformer 115 in parallel, the voltage divider circuit 119 is electrically connected to the differential ADC module 120, the differential ADC module 120 is electrically connected to the optical fiber transmission module 121, the voltage divider circuit 119 divides the first voltage VDD1 into a fourth voltage VDD4 in proportion, the differential ADC module 120 converts the fourth voltage VDD4 into a digital quantity, and sends the digital signal to the control unit 13 through the optical fiber transmission module 121. The current detection module 113 of the detection unit 12 includes: a Rogowski coil 112 and an integrator module 122. The Rogowski coil 112 is passed through the center of the fuse to be tested; the integrator module 122 is electrically connected to the Rogowski coil 112; the integrator module 122 is electrically connected to the control unit 13; the Rogowski coil 112 converts the rate of change of the AC current into a voltage signal proportional to it, and the integrator module 122 integrates the voltage signal to obtain a voltage signal proportional to the actual AC current. After receiving the voltage and current signals from the detection unit 12, the control unit 13 analyzes and processes the received signals, calculates the internal resistance of the fuse, and wirelessly sends the fuse internal resistance, voltage, and current data to the cloud platform 14.

[0084] Further, see Figure 1 The interior of the housing 10 is sealed with silicone material.

[0085] Specifically, the embodiment of the present invention has studied the remaining life prediction method, using a neural network prediction algorithm, specifically a gradient boosting decision tree (GBDT) model, which is used to predict the change trend of the internal resistance of the fuse to be tested. When the internal resistance of the fuse exceeds 20% of its factory rated value, it is considered that the fuse life has been reached, thereby triggering an alarm mechanism to prompt the staff to handle it in time. The specific operation steps and life prediction method include the following:

[0086] (1) Initialization

[0087] System initialization includes configuration of detection unit, control unit and cloud platform;

[0088] Set the sampling frequency and threshold parameters of the current detection module and voltage detection module in the detection unit, and ensure that each module works normally;

[0089] Load the pre-trained gradient boosted decision tree (GBDT) model and set the input features of the model, including current, voltage, and surge current;

[0090] After the system is ready, it starts to monitor the electrical status of the fuse to be tested;

[0091] (2) Data collection and internal resistance measurement

[0092] ① The current detection module uses a Rogowski coil to measure the current passing through the fuse. The Rogowski coil converts the rate of change of the AC current into a voltage signal. The voltage signal is then integrated through the integrator module to obtain the actual current value, completing the conversion and calculation of the current signal;

[0093] ②Measure the voltage signal at both ends of the fuse to be tested through the voltage divider circuit in the voltage detection module;

[0094] The voltage divider circuit divides the actual voltage in proportion to generate a voltage value suitable for processing by the differential ADC module;

[0095] The differential ADC module converts the analog voltage signal into a digital signal and sends it to the control unit through the optical fiber transmission module;

[0096] ③The control unit follows Ohm's law Substitute the collected real-time voltage U and current I into the calculation formula to obtain the internal resistance value of the fuse;

[0097] ④ The control unit wirelessly sends the voltage U, current I and internal resistance data to the data storage module of the cloud platform, and the data storage module performs characterization processing on the above data;

[0098] (3) GBDT model predicts internal resistance

[0099] ① Input the characteristic data of the data storage module into the gradient boosting decision tree (GBDT) model of the data analysis module. The input features include internal resistance value, voltage U, current I, inrush current, etc. The GBDT model receives these characteristic data and processes them as model input variables. These features are used as the basis for splitting each decision tree to predict the internal resistance change trend of the fuse and save them as historical data;

[0100] ②GBDT initializes a base model by calculating the average internal resistance value of the training data. This average value is used as the starting prediction of the first tree, indicating the current preliminary estimate of the internal resistance;

[0101] ③For each sample, GBDT calculates the difference between the current predicted value and the true value. This difference is called residual. The residual reflects the error of the model in the current prediction. Subsequent decision trees will focus on fitting these residuals.

[0102] ④After calculating the residual, GBDT starts to build a new decision tree. During the construction process, a greedy algorithm is used. The model will select a feature and an optimal split threshold to minimize the mean square error in the child nodes after the split. At each node, all features and thresholds are traversed to find the feature that can maximize the information gain.

[0103] ⑤ Generate a new decision tree. GBDT adds the prediction results of this tree to the current model. This update process is affected by the learning rate. The learning rate controls the contribution of each tree to the final model. Through the learning rate, the model can gradually improve the accuracy of the prediction without overfitting.

[0104] ⑥GBDT repeats steps ③ to ⑤ until the set number of trees is reached or the residual error falls within a certain tolerance range. Each tree will correct the prediction error of the previous tree and gradually reduce the error of the internal resistance prediction. Finally, the prediction results of all decision trees will be combined to form the final prediction model.

[0105] ⑦ Input the real-time data of the current fuse into the GBDT model, calculate the remaining service life of the fuse based on the relationship between the predicted internal resistance of the fuse and the preset threshold value, and combine historical data. If the remaining service life is lower than the set value, it is recommended to replace the fuse immediately. If the remaining service life is long, but the internal resistance is close to the threshold, it is recommended to arrange regular inspections. When the internal resistance value exceeds the set threshold or the remaining service life is lower than a certain critical value, promptly notify the maintenance personnel to take maintenance measures.

[0106] For the parts not described in detail in this embodiment, please refer to Figure 1-Figure 4 The embodiments shown in this embodiment and the above embodiments can be referenced to each other and will not be described in detail here.

[0107] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features disclosed herein.

Claims

1. A device for collecting characteristic data and predicting remaining life of a high-voltage fuse, characterized in that: include: Housing, power module, detection unit, control unit, cloud platform; The power supply module, the control unit, and the voltage detection module of the detection unit are placed inside the housing; A first opening is provided on the side of the housing, and a plug-in terminal is provided at the first opening for connecting the fuse to be tested; The power module is connected to the control unit; The detection unit is connected to the control unit; The cloud platform is connected to the control unit via wireless communication.

2. The device for collecting characteristic data and predicting remaining life of a high-voltage fuse according to claim 1, characterized in that: The power module includes: a plug-in terminal, a fuse unit, a transformer, a first voltage conversion chip, and a second voltage conversion chip; The pluggable terminal is electrically connected in parallel with the fuse to be tested; the pluggable terminal is electrically connected with the insurance unit; the insurance unit is electrically connected with the transformer; the transformer is electrically connected with the first voltage conversion chip; the first voltage conversion chip is electrically connected with the second voltage conversion chip; the second voltage conversion chip is electrically connected with the control unit; The transformer amplifies the collected voltage to obtain a first voltage; the first voltage conversion chip converts the first voltage into a second voltage; and the second voltage conversion chip converts the second voltage into a third voltage.

3. The device for collecting characteristic data and predicting remaining life of a high-voltage fuse according to claim 2, characterized in that: The insurance unit uses a 5kV fuse, which is a spring-type high-voltage fuse, the shell material of which is a glass fiber tube, and the number is at least 2.

4. The device for collecting characteristic data and predicting remaining life of a high-voltage fuse according to claim 1, characterized in that: The detection unit includes a current detection module and a voltage detection module, both of which are electrically connected to the control unit, wherein the voltage detection module is electrically connected to the transformer.

5. The device for collecting characteristic data and predicting remaining life of a high-voltage fuse according to claim 4, characterized in that: The voltage detection module of the detection unit includes: a voltage dividing circuit, a differential ADC module, and an optical fiber transmission module; The voltage divider circuit is electrically connected in parallel with the transformer; The voltage divider circuit is electrically connected to the differential ADC module; The differential ADC module is electrically connected to the optical fiber transmission module; The voltage divider circuit divides the first voltage into a fourth voltage in proportion; the differential ADC module converts the fourth voltage into a digital quantity, and sends the digital signal to the control unit through the optical fiber sending module.

6. The device for collecting characteristic data and predicting remaining life of a high-voltage fuse according to claim 4, characterized in that: The current detection module of the detection unit includes: a Rogowski coil and an integrator module; the Rogowski coil is passed through the center of the fuse to be tested; the integrator module is electrically connected to the Rogowski coil; the integrator module is electrically connected to the control unit; the Rogowski coil converts the rate of change of the alternating current into a voltage signal proportional to it; the integrator module integrates the voltage signal to obtain a voltage signal proportional to the actual alternating current.

7. The device for collecting characteristic data and predicting remaining life of a high-voltage fuse according to claim 1, characterized in that: The control unit includes: a signal receiving module, a data processing unit, a 5G communication module and an alarm module; the signal receiving module is used to receive the output signal of the detection unit; the output signal includes the voltage value of the fuse to be tested and the current value flowing through it; the data processing unit is used to analyze and process the received voltage value and current value signals of the fuse to be tested to calculate the internal resistance of the fuse; the 5G communication module wirelessly sends the internal resistance value, voltage and current data of the fuse to be tested to the cloud platform; the alarm module analyzes and compares the status information such as voltage, current and calculated internal resistance value with a preset threshold value, so as to timely perceive the abnormal status of the equipment and remind maintenance personnel to take maintenance measures through various alarm methods.

8. The device for collecting characteristic data and predicting remaining life of a high-voltage fuse according to claim 1, characterized in that: The inside of the shell is sealed with silica gel material, which can prevent the intrusion of moisture and dust and is used for electrical insulation protection.

9. The device for collecting characteristic data and predicting remaining life of a high-voltage fuse according to claim 1, characterized in that: The cloud platform includes: a data receiving module, a data storage module, and a data analysis module; the data receiving module is used to receive data from the control unit through the 5G network; the data storage module is used to store and manage the received data; the data analysis module is used to analyze and process the stored data, and use the remaining life prediction method to predict the internal resistance and remaining service life of the fuse.

10. A method for predicting remaining life, characterized in that: The remaining life prediction method uses a neural network prediction algorithm, specifically a gradient boosting decision tree (GBDT) model, which is used to predict the change trend of the internal resistance of the fuse to be tested. When the internal resistance of the fuse exceeds 20% of its factory rated value, it is considered that the fuse life has been reached, thereby triggering an alarm mechanism to prompt the staff to handle it in time. The specific operation steps and the remaining life prediction method include the following: (1) Initialization System initialization includes configuration of detection unit, control unit and cloud platform; Set the sampling frequency and threshold parameters of the current detection module and voltage detection module in the detection unit, and ensure that each module works normally; Load the pre-trained gradient boosted decision tree (GBDT) model and set the input features of the model, including current, voltage, and surge current; After the system is ready, it starts to monitor the electrical status of the fuse to be tested; (2) Data collection and internal resistance measurement ① The current detection module uses a Rogowski coil to measure the current passing through the fuse. The Rogowski coil converts the rate of change of the AC current into a voltage signal. The voltage signal is then integrated through the integrator module to obtain the actual current value, completing the conversion and calculation of the current signal; ②Measure the voltage signal at both ends of the fuse to be tested through the voltage divider circuit in the voltage detection module; The voltage divider circuit divides the actual voltage in proportion to generate a voltage value suitable for processing by the differential ADC module; The differential ADC module converts the analog voltage signal into a digital signal and sends it to the control unit through the optical fiber transmission module; ③The control unit follows Ohm's law Substitute the collected real-time voltage U and current I into the calculation formula to obtain the internal resistance value of the fuse to be tested; ④ The control unit wirelessly sends the voltage U, current I and internal resistance data to the data storage module of the cloud platform, and the data storage module performs characterization processing on the above data; (3) GBDT model predicts internal resistance ① Input the characteristic data of the data storage module into the gradient boosting decision tree (GBDT) model of the data analysis module. The input features include internal resistance value, voltage U, current I, inrush current, etc. The GBDT model receives these characteristic data and processes them as model input variables. These features are used as the basis for splitting each decision tree to predict the internal resistance change trend of the fuse and save them as historical data; ②GBDT initializes a base model by calculating the average internal resistance value of the training data. This average value is used as the starting prediction of the first tree, indicating the current preliminary estimate of the internal resistance; ③For each sample, GBDT calculates the difference between the current predicted value and the true value. This difference is called residual. The residual reflects the error of the model in the current prediction. Subsequent decision trees will focus on fitting these residuals. ④After calculating the residual, GBDT starts to build a new decision tree. During the construction process, a greedy algorithm is used. The model will select a feature and an optimal split threshold to minimize the mean square error in the child nodes after the split. At each node, all features and thresholds are traversed to find the feature that can maximize the information gain. ⑤ Generate a new decision tree. GBDT adds the prediction results of this tree to the current model. This update process is affected by the learning rate. The learning rate controls the contribution of each tree to the final model. Through the learning rate, the model can gradually improve the accuracy of the prediction without overfitting. ⑥GBDT repeats steps ③ to ⑤ until the set number of trees is reached or the residual error falls within a certain tolerance range. Each tree will correct the prediction error of the previous tree and gradually reduce the error of the internal resistance prediction. Finally, the prediction results of all decision trees will be combined to form the final prediction model. ⑦ Input the real-time data of the current fuse into the GBDT model, calculate the remaining service life of the fuse based on the relationship between the predicted internal resistance of the fuse and the preset threshold value, and combine historical data. If the remaining service life is lower than the set value, it is recommended to replace the fuse immediately. If the remaining service life is long, but the internal resistance is close to the threshold, it is recommended to arrange regular inspections. When the internal resistance value exceeds the set threshold or the remaining service life is lower than a certain critical value, promptly notify the maintenance personnel to take maintenance measures.

Images