A circuit breaker condition monitoring and self-diagnosis system

Abstract

This invention provides a circuit breaker condition monitoring and self-diagnosis system, relating to the field of power monitoring technology. It includes multiple heterogeneous sensor modules, such as temperature sensors, current sensors, pressure sensors, vibration sensors, and voltage sensors. These sensor modules are installed at multiple key locations on the circuit breaker to collect multi-dimensional operational data. A heterogeneous sensor collaborative sensing unit receives and processes real-time data from each sensor module. This unit performs correlation analysis on data from different types of sensors through data fusion. This circuit breaker condition monitoring and self-diagnosis system, by combining a grey prediction model GM(1,1) with real-time sensor data, can efficiently perform fault prediction and trend analysis. By accumulating key indicators and modeling with grey differential equations, this scheme can predict the potential time of equipment failure in advance and generate trend analysis results.

Description

Technical Field

[0001] This invention relates to the field of power monitoring technology, specifically to a circuit breaker condition monitoring and self-diagnosis system. Background Technology

[0002] Currently, circuit breaker equipment in power systems primarily relies on traditional monitoring methods for condition detection, including sensor monitoring, manual inspections, and periodic maintenance. While these methods ensure the normal operation of power equipment to a certain extent, they also have significant shortcomings and are insufficient to meet the demands of increasingly complex power systems and high-efficiency management. Sensor monitoring technology collects various operational data in real time by installing various sensors on circuit breakers, such as temperature, pressure, and current sensors. These sensors can detect basic operating parameters of the equipment and monitor its status in real time through a data acquisition system. However, this method typically only provides basic operational data and is limited by the number and installation location of sensors, making it difficult to comprehensively monitor all potential faults. Furthermore, sensors may malfunction or fail, leading to data loss or inaccuracy, severely impacting the accuracy of fault detection. Manual inspections rely on personnel periodically checking the appearance and operating status of circuit breaker equipment to ensure its normal operation. Manual inspections can detect external anomalies to some extent, such as aging, damage, or other apparent signs of failure, but they have significant limitations. Manual inspections typically need to be conducted periodically, but the operating environment of power equipment can be complex and unpredictable, limiting the frequency of manual inspections and making 24 / 7 monitoring impossible. Due to human negligence or operational errors, potential faults may be missed during inspections, leading to undetected equipment problems and consequently affecting the stability and security of the power system.

[0003] Traditional monitoring technologies rely primarily on manual inspections or periodic maintenance, failing to provide 24 / 7 real-time monitoring. Potential equipment malfunctions cannot be detected promptly, and fault diagnosis methods depend on human experience and routine checks, lacking intelligent support, resulting in low accuracy. Faults are difficult to detect using traditional methods, easily missing the optimal repair window. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a circuit breaker condition monitoring and self-diagnosis system. The technical problem this invention aims to solve is: how to improve the safety and reliability of power systems through fault early warning and emergency response mechanisms.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a circuit breaker condition monitoring and self-diagnosis system, comprising:

[0006] Multiple heterogeneous sensor modules, including temperature sensors, current sensors, pressure sensors, vibration sensors and voltage sensors, are installed at multiple key locations of the circuit breaker to collect multi-dimensional operating data of the equipment.

[0007] The heterogeneous sensor collaborative sensing unit is used to receive and process real-time data from various sensor modules. The heterogeneous sensor collaborative sensing unit performs correlation analysis on data from different types of sensors through data fusion to obtain standardized data in order to capture the characteristic information of various potential faults.

[0008] The fault diagnosis and analysis unit, based on the fused data from the heterogeneous sensor collaborative sensing unit, performs fault mode recognition, fault prediction and trend analysis through a gray prediction model, automatically diagnoses the equipment status and predicts potential faults.

[0009] The fault warning and response unit automatically issues a fault warning and executes corresponding emergency response measures based on the diagnostic results of the fault diagnosis and analysis unit.

[0010] The data transmission and storage unit is responsible for transmitting the data, fault diagnosis analysis results, and early warning information of the sensor module to the remote monitoring platform, and storing and managing the relevant data.

[0011] Preferably, the output data of the multiple heterogeneous sensor modules are digital signals and are wirelessly transmitted to the heterogeneous sensor collaborative sensing unit for processing.

[0012] Preferably, the data fusion includes the following steps:

[0013] S3.1. The real-time data is preprocessed, including denoising, standardization and normalization, in order to eliminate interference factors in the data of each sensor and unify the data scale;

[0014] S3.2. Perform a weighted average on the preprocessed data to integrate different types of data and generate a comprehensive equipment health index;

[0015] S3.3. Based on the equipment health index, fault judgment is performed through a set threshold. If the equipment health index exceeds the predetermined threshold, a fault alarm signal is triggered to indicate equipment failure.

[0016] Preferably, the grey prediction model is performed through the following steps:

[0017] S4.1. Select key indicators related to the fault from the standardized data. The key indicators are used to reflect the current health status of the equipment and are used as input data for subsequent predictive analysis.

[0018] S4.2. The key indicators are used as input to model the grey prediction model GM(1,1). The grey prediction model uses the cumulative generation method to construct a cumulative data sequence. The cumulative data sequence is used to establish a grey differential equation for trend analysis and generate trend analysis results.

[0019] S4.3. Based on the trend analysis results, generate an equipment health prediction report and perform fault diagnosis. By comparing the prediction results with the set fault threshold, if the equipment health index is lower than the set threshold or the predicted fault occurrence time is approaching, a fault alarm signal is triggered to indicate that the equipment may be in a fault state or close to a fault state, guiding maintenance personnel to intervene in a timely manner.

[0020] Preferably, the specific steps of the trend analysis are as follows:

[0021] S4.2.1 The standardized data x0(k) is accumulated to obtain the accumulated sequence x1(k);

[0022] S4.2.2 The least squares method is used to fit the grey differential equation to estimate the development coefficient a and the background value b;

[0023] S4.2.3 Based on the estimated parameters a and b, perform prediction calculations for future time points to obtain predicted values ​​of equipment operating status. This step can analyze future changes in equipment status and calculate the potential time and evolution path of failure.

[0024] Preferably, the accumulated sequence is specifically:

[0025]

[0026] Where x1(k) represents the accumulated data, x0(k) represents the original data, and n represents the length of the data sequence;

[0027] The grey differential equation is in the form of:

[0028]

[0029] Where a is the growth coefficient, representing the rate of change of the data, and b is the background value, representing the normal fluctuation of the system. The appropriate values ​​of a and b are obtained by solving the equation using the least squares method.

[0030] Preferably, the data transmission and storage unit includes: a real-time data transmission module, which uses wireless communication technology to transmit data to a remote monitoring platform in real time, ensuring that the remote platform can obtain the circuit breaker's operating status and fault information in real time.

[0031] The data encryption module is used to encrypt data from the sensor module, fault diagnosis analysis results, and early warning information to ensure data security during transmission.

[0032] The data storage module includes local storage and cloud storage. Local storage is used to temporarily store data and diagnostic results from various sensors, while cloud storage is used for long-term storage and backup of all device data, fault analysis reports, and early warning records.

[0033] Preferably, the fault warning and response unit uses an adaptive threshold adjustment algorithm to dynamically adjust the fault warning threshold of the equipment health index. The adaptive threshold adjustment algorithm is calculated based on the following parameters:

[0034] T a =T b +k1·Δ(avg)+k2·std(x)

[0035] Among them, T a T is the dynamically adjusted warning threshold. b Δ(avg) is the basic health index threshold of the equipment, Δ(avg) is the average change of the health index of the equipment in the most recent time window, std(x) is the standard deviation of the historical operating data of the equipment, and k1 and k2 are constants adjusted according to the operating characteristics of the equipment.

[0036] This invention provides a circuit breaker condition monitoring and self-diagnosis system. It has the following beneficial effects:

[0037] This circuit breaker condition monitoring and self-diagnosis system, by combining the grey prediction model GM(1,1) and real-time sensor data, can efficiently perform fault prediction and trend analysis. By accumulating key indicators and modeling with grey differential equations, this solution can predict the potential time of equipment failure in advance and generate trend analysis results, helping maintenance personnel to grasp the health status of equipment in real time, take timely maintenance measures, and effectively reduce downtime and maintenance costs caused by sudden failures.

[0038] This technical solution introduces a dynamic adjustment and intelligent response mechanism, enabling the system to automatically adjust the early warning threshold based on real-time changes in equipment operating status and trigger corresponding fault warnings and emergency response measures based on prediction results. The adaptive threshold adjustment algorithm dynamically optimizes the fault diagnosis threshold based on the equipment's historical operating data and current status, ensuring more accurate early warnings. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of a structure for realizing an invention;

[0040] Figure 2 This is a flowchart illustrating the process of implementing an invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Example 1

[0043] like Figure 1-2 As shown, this embodiment of the invention provides a circuit breaker condition monitoring and self-diagnosis system, including multiple heterogeneous sensor modules, which include temperature sensors, current sensors, pressure sensors, vibration sensors, and voltage sensors. The sensor modules are installed at multiple key locations of the circuit breaker to collect multi-dimensional operating data of the equipment. The output data of the multiple heterogeneous sensor modules are digital signals and are transmitted wirelessly to the heterogeneous sensor collaborative sensing unit for processing.

[0044] The heterogeneous sensor collaborative sensing unit receives and processes real-time data from various sensor modules. Through data fusion, it correlates and analyzes data from different types of sensors to obtain standardized data, thereby capturing characteristic information of various potential faults. The data fusion process includes the following steps:

[0045] S3.1. Preprocess the real-time data, including noise reduction, standardization and normalization, in order to eliminate interference factors in the data from various sensors and unify the data scale.

[0046] S3.2. Perform a weighted average on the preprocessed data to integrate different types of data and generate a comprehensive equipment health index.

[0047] S3.3. Based on the equipment health index, fault judgment is performed through the set threshold. If the equipment health index exceeds the predetermined threshold, a fault alarm signal is triggered to indicate that the equipment is faulty.

[0048] Assume the standardized data collected in real time is as follows:

[0049] Temperature sensor data: [50, 52, 55, 58, 60]

[0050] Current sensor data: [10,15,20,25,30]

[0051] Vibration sensor data: [0.02, 0.03, 0.04, 0.05, 0.06]

[0052] After denoising, standardization, and normalization, the following standardized data was obtained:

[0053] Temperature sensor: [0.0, 0.1, 0.3, 0.4, 0.5]

[0054] Current sensor: [-1.41,-0.71,0.0,0.71,1.41]

[0055] Vibration sensor: [0.0, 0.2, 0.4, 0.6, 0.8]

[0056] Assume the weights are set as follows:

[0057] Temperature sensor: weight 0.4

[0058] Current sensor: weight 0.3

[0059] Vibration sensor: weight 0.3

[0060] The equipment health index is calculated as follows:

[0061] DHI=0.4·0.3+0.3·0.0+0.3·0.4=0.12+0.12=0.24.

[0062] The set fault judgment threshold is T t =0.5. Since the equipment health index DHI = 0.24 is less than the threshold, the system triggers a fault alarm, indicating that the equipment needs maintenance.

[0063] The fault diagnosis and analysis unit, based on fused data from the heterogeneous sensor collaborative sensing unit, performs fault mode recognition, fault prediction, and trend analysis through a grey prediction model. It automatically diagnoses equipment status and predicts potential faults. The grey prediction model operates through the following steps:

[0064] S4.1. Select key indicators related to faults from standardized data. These key indicators reflect the current health status of the equipment and serve as input data for subsequent predictive analysis.

[0065] S4.2. Key indicators are used as input to model the grey prediction model GM(1,1). The grey prediction model uses the cumulative generation method to construct a cumulative data sequence. A grey differential equation is established from the cumulative data sequence for trend analysis, and the trend analysis results are generated. The specific steps of the trend analysis are as follows:

[0066] S4.2.1 The standardized data x0(k) is accumulated to obtain the accumulated sequence x1(k).

[0067] S4.2.2 The least squares method is used to fit the grey differential equation, and then the development coefficient a and the background value b are estimated. The cumulative sequence is as follows:

[0068]

[0069] Where x1(k) is the accumulated data, x0(k) is the original data, and n is the length of the data sequence.

[0070] S4.2.3 Based on the estimated parameters a and b, perform prediction calculations for future time points to obtain predicted values ​​of the equipment's operating status. This step can analyze future state changes of the equipment and calculate the potential time and evolution path of the failure. The grey differential equation form is as follows:

[0071]

[0072] Where a is the growth coefficient, representing the rate of change of the data, and b is the background value, representing the normal fluctuation of the system. The appropriate values ​​of a and b are obtained by solving the equation using the least squares method.

[0073] S4.3. Based on the trend analysis results, generate an equipment health prediction report and perform fault diagnosis. By comparing the prediction results with the set fault threshold, if the equipment health index is lower than the set threshold or the predicted fault occurrence time is approaching, a fault alarm signal is triggered to indicate that the equipment may be in a fault state or close to a fault state, guiding maintenance personnel to intervene in a timely manner.

[0074] The fault early warning and response unit automatically issues fault early warnings and executes corresponding emergency response measures based on the diagnostic results of the fault diagnosis and analysis unit. The fault early warning and response unit uses an adaptive threshold adjustment algorithm to dynamically adjust the fault early warning threshold of the equipment health index. The adaptive threshold adjustment algorithm is calculated based on the following parameters:

[0075] T a =T b +k1·Δ(avg)+k2·std(x)

[0076] Among them, T a T is the dynamically adjusted warning threshold. b Δ(avg) is the basic health index threshold of the equipment, Δ(avg) is the average change of the health index of the equipment in the most recent time window, std(x) is the standard deviation of the historical operating data of the equipment, and k1 and k2 are constants adjusted according to the operating characteristics of the equipment.

[0077] The following are the detailed implementation steps of the algorithm, including how to calculate the dynamically adjusted warning threshold, and illustrate them with specific data.

[0078] Step 1: Input Data Preparation

[0079] Table 1: Standardized data collected by the sensor

[0080] Time point Temperature (°C) Current (A) Voltage (V) Health Index t1 50 15 380 75 t2 52 16 381 76 t3 55 16.5 380.5 77 t4 58 17 380.8 78 t5 60 18 381 80

[0081] The health index serves as a measure of the device's health status. Warning thresholds are adjusted based on the health index.

[0082] Step 2: Calculate the average change in the health index.

[0083] First, it is necessary to calculate the average change (Δavg) of the device's health index over a recent period. For the calculation of the health index, data from the four most recent time points are selected:

[0084] Calculate the change in health index:

[0085] ΔDHI1=76-75=1

[0086] ΔDHI2=77-76=1

[0087] ΔDHI3=78-77=1

[0088] ΔDHI4=80-78=2

[0089] Calculate the average change (Δavg):

[0090]

[0091] Step 3: Calculate the standard deviation of the equipment's historical operating data.

[0092] Calculate the standard deviation of the equipment health index. Calculate the mean (μ) of the health index. D ):

[0093]

[0094] Then, calculate the square of the difference between each data point and the mean:

[0095] (75-77.2) 2 =(-2.2) 2 =4.84

[0096] (76-77.2) 2 =(-1.2) 2 =1.44

[0097] (77-77.2) 2 =(-0.2) 2 =0.04

[0098] (78-77.2) 2 =(0.8) 2 =0.64

[0099] (80-77.2)2 =(2.8) 2 =7.84

[0100] Find the average of the differences of squares:

[0101]

[0102] Finally, calculate the standard deviation:

[0103]

[0104] Step 4: Calculate the dynamically adjusted warning threshold

[0105] The average change in health index (Δavg = 1.25) and standard deviation (s(D) = 1.6) calculated based on the above steps, as well as the baseline health index threshold (T) b =80), we can calculate the dynamically adjusted warning threshold (T) based on the adaptive threshold adjustment algorithm. a ).

[0106] The formula is:

[0107] T a =T b +k1·Δavg+k2·s(D)

[0108] Assuming constants k1 = 0.5 and k2 = 0.3, then:

[0109] T a = 80 + 0.5 * 1.25 + 0.3 * 1.6

[0110] T a =80 + 0.625 + 0.48 = 81.105

[0111] Therefore, the dynamically adjusted warning threshold is 81.105.

[0112] Step 5: Fault Warning and Alarm Triggering

[0113] If the equipment health index is lower than the dynamically adjusted threshold (T) a If the health index is predicted to be 79 (below the threshold of 81.105), the system will trigger a fault warning. For example, if the predicted health index for the next time point is 79 (below the threshold of 81.105), the system will automatically issue a fault alarm signal and prompt maintenance personnel to take necessary intervention. The steps of the adaptive threshold adjustment algorithm are demonstrated, including how to calculate the average change value and standard deviation based on changes in the equipment health index, and how to dynamically adjust the warning threshold. This method can intelligently adjust the threshold in real time according to the operating status of the equipment, thereby improving the accuracy of fault warnings and reducing false alarms or missed alarms caused by unreasonable threshold settings.

[0114] The data transmission and storage unit is responsible for transmitting data from the sensor module, fault diagnosis analysis results, and early warning information to the remote monitoring platform, and storing and managing the relevant data. The data transmission and storage unit includes: a real-time data transmission module, which uses wireless communication technology to transmit data to the remote monitoring platform in real time, ensuring that the remote platform can obtain the circuit breaker's operating status and fault information in real time.

[0115] The data encryption module is used to encrypt data from the sensor module, fault diagnosis analysis results, and early warning information to ensure data security during transmission.

[0116] The data storage module includes local storage and cloud storage. Local storage is used to temporarily store data and diagnostic results from various sensors, while cloud storage is used for long-term storage and backup of all device data, fault analysis reports, and early warning records.

[0117] Example 2

[0118] Unlike Example 1, this example details how to perform fault mode identification, fault prediction, and trend analysis based on the grey prediction model GM(1,1). The following is a detailed description of the steps, including how to select key indicators, establish the grey prediction model, perform trend analysis, and generate health reports and trigger fault alarms.

[0119] S4.1 Screening Key Indicators Related to Faults

[0120] Before conducting fault mode identification and trend analysis, it is necessary to first screen out key indicators related to faults from the standardized data. These key indicators are usually high-impact factors in sensor data and can reflect the operational health status of the equipment. Common key indicators include temperature, current, voltage, vibration, and pressure, which are closely related to the equipment's failure risk and operating efficiency.

[0121] Table 2: Data collected by the sensor

[0122] Time point Temperature (°C) Current (A) Voltage (V) Vibration (g) t1 50 15 380 0.02 t2 52 16 381 0.03 t3 55 16.5 380.5 0.04 t4 58 17 380.8 0.05 t5 60 18 381 0.06

[0123] S4.2 Input key indicators into the grey prediction model

[0124] S4.2.1 Cumulative Generation Method:

[0125] In this step, the selected key indicators are first accumulated using a cumulative generation method. For each key indicator, the accumulated sequence is constructed as follows:

[0126] The temperature data is accumulated to generate the following:

[0127] x1(1)=x0(1)=50

[0128] x1(2)=x0(1)+x0(2)=50+52=102

[0129] x1(3)=x0(1)+x0(2)+x0(3)=50+52+55=157

[0130] x1(4)=x0(1)+x0(2)+x0(3)+x0(4)=50+52+55+58=215

[0131] x1(5)=x0(1)+x0(2)+x0(3)+x0(4)+x0(5)=50+52+55+58+60=275

[0132] By accumulating the key indicators of current, voltage, and vibration in the same way, the corresponding cumulative data sequence is obtained.

[0133] S4.2.2 Least squares method for fitting the grey differential equation:

[0134] The grey differential equation is fitted using the least squares method, and then the development coefficient a and background value b are estimated.

[0135] Assuming we already have the accumulated temperature sequence of the device:

[0136] x1 = [50, 102, 157, 215, 275]

[0137] Using grey differential equations:

[0138]

[0139] The least squares method is used to calculate a and b. The least squares formula can be obtained by solving the following system of linear equations:

[0140] A·B=Y

[0141] Here, matrix A and vector Y are values ​​constructed by accumulating data. After solving for the values ​​of a and b, the fault development trend of the equipment can be obtained.

[0142] Least squares parameter calculation based on the accumulated sequence:

[0143] 1. Definition of Cumulative Amount

[0144] For N groups of samples {(x k ,y k )}, summed sequentially, yield:

[0145]

[0146] 2. Construct the normal equation

[0147] Enter the above cumulative amount directly:

[0148]

[0149] We obtain the system of linear equations:

[0150]

[0151] 3. Solution Method

[0152] The coefficients a and b can be obtained using any standard algorithm, such as Gaussian elimination, matrix inversion, or Cramer's rule.

[0153]

[0154] S4.2.3 Making future predictions

[0155] Using parameters a and b of the grey differential equation, a prediction calculation for future time points is performed. In practical applications, the development coefficient a = 0.3 and the background value b = 4.0 are used to predict the equipment temperature at the next time point. The prediction formula is:

[0156]

[0157] The above calculations yield prediction results, which can then be used to predict the timing and trend of equipment failures. For example, if the equipment temperature exceeds a set threshold (e.g., 75°C) at some point in the future, the system will trigger a fault warning, alerting maintenance personnel that the equipment may malfunction.

[0158] S4.3 generates a device health prediction report and triggers a fault alarm.

[0159] Based on the prediction results and trend analysis, the system generates an equipment health prediction report and compares it with the set fault threshold. Assuming the set equipment temperature fault threshold is 70℃, if the prediction results indicate that the temperature will reach or exceed this threshold at some point in the future, the system will automatically trigger a fault alarm.

[0160] For example, if the system predicts that the device temperature will reach 75°C at the next time point, exceeding the safety threshold, it will perform the following actions:

[0161] Generate a health prediction report, which will include the current status of the equipment, predicted failure values, warning times, and recommended maintenance or downtime operations.

[0162] When a fault alarm signal is triggered, it is sent to maintenance personnel via the communication module, possibly through SMS, email, or mobile app notifications.

[0163] Intervene promptly, providing maintenance personnel with a detailed fault diagnosis report and recommended measures, such as adjusting equipment parameters or arranging repairs.

[0164] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A circuit breaker condition monitoring and self-diagnostic system, characterized by, The application relates to a device state monitoring system, comprising: a plurality of heterogeneous sensor modules, including temperature sensors, current sensors, pressure sensors, vibration sensors and voltage sensors; a heterogeneous sensor cooperative perception unit for receiving and processing real-time data from each sensor module, correlating and analyzing data of different types of sensors through data fusion to obtain standardized data; a fault diagnosis analysis unit for automatically diagnosing device states and predicting potential faults through grey prediction model based on fusion data from the heterogeneous sensor cooperative perception unit for fault mode recognition, fault prediction and trend analysis; a fault warning and response unit for automatically issuing fault warnings and executing corresponding emergency response measures according to the diagnosis results of the fault diagnosis analysis unit; a data transmission and storage unit responsible for transmitting data of the sensor modules, fault diagnosis analysis results and warning information to a remote monitoring platform and storing and managing relevant data; the grey prediction model is obtained through the following steps: S4.

1. screening out key indicators related to faults from the standardized data; S4.

2. taking the key indicators as input to build a grey prediction model GM (1, 1), the grey prediction model adopts an accumulation generation method to construct a cumulative data sequence, the cumulative data sequence establishes a grey differential equation for trend analysis, and a trend analysis result is generated; S4.

3. generating a device health prediction report according to the trend analysis result, and performing fault diagnosis; the trend analysis comprises the following steps: S4.2.1 On the normalized data An accumulation process is performed to obtain an accumulation sequence ; S4.2.2 Fitting the grey differential equation using least squares, thereby estimating the development coefficient and background values ; S4.2.2 Making a prediction calculation for a future time point based on the estimated parameters and to obtain a predicted value of the device operating state. the cumulative sequence is as follows: wherein, is cumulative data, is raw data, is the length of the data sequence; the form of the grey differential equation is as follows: where, is a development coefficient, indicating the rate of change of the data, is a background value, indicating the normal fluctuation of the system.

2. The circuit breaker condition monitoring and self-diagnostic system of claim 1, wherein: output data of the plurality of heterogeneous sensor modules are digital signals and are transmitted to the heterogeneous sensor cooperative perception unit through a wireless mode for processing.

3. The circuit breaker condition monitoring and self-diagnostic system of claim 1, wherein: The data fusion comprises the following steps: S3.

1. pre-processing the real-time data, the pre-processing comprising denoising, standardization and normalization processing; S3.

2. performing weighted average on the pre-processed data to generate a comprehensive device health index; S3.

3. performing fault judgment according to the device health index through a set threshold value, if the device health index exceeds a predetermined threshold value, a fault alarm signal is triggered to prompt device faults.

4. The circuit breaker condition monitoring and self-diagnostic system of claim 1, wherein: The data transmission and storage unit comprises a real-time data transmission module, a data encryption module and a data storage module.

5. The circuit breaker condition monitoring and self-diagnostic system of claim 3, wherein: The fault warning and response unit adopts an adaptive threshold adjustment algorithm to dynamically adjust the fault warning threshold value of the device health index.

Images