Circuit breaker control system based on cloud data

Abstract

The present invention relates to the technical field of circuit breakers, specifically to a circuit breaker control system based on cloud data. The system includes a data collection module, a data stream processing module, a fault diagnosis and prediction module, and a system management and optimization module. The data collection module collects current changes, temperature fluctuations, and vibration frequencies in real time based on installed current, temperature, and vibration sensors, and continuously transmits data to the cloud through LoRa technology. The present invention automatically performs the operation of the circuit breaker through real-time data analysis and prediction algorithms, significantly improves the real-time monitoring and fault response capabilities of the power system, monitors the sudden changes of current and temperature in real time, enables the system to quickly identify and respond to potential faults, and effectively prevents equipment damage and power outages by automatically performing the opening and closing operations of the circuit breaker. In addition, the cloud platform is used to deeply analyze historical data, optimize maintenance cycles and control logic, make power management more efficient, and ensure the continuity of power supply and the long-term stable operation of equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit breakers, and in particular to a circuit breaker control system based on cloud data. Background Art

[0002] The technical field of circuit breakers involves switchgear used in power systems. A circuit breaker is an automated electrical switching device that can interrupt the flow of current and protect the safety of electrical equipment and power systems when the system detects a short circuit, overload or other electrical fault. This field includes the design, manufacture, installation and maintenance of circuit breakers, covering low-voltage, medium-voltage and high-voltage circuit breakers. It also involves various auxiliary equipment and control systems related to circuit breakers, such as protection relays, measurement and monitoring equipment, communication modules, etc., to ensure the reliability and safety of the power system.

[0003] Among them, the circuit breaker control system is a system that realizes remote monitoring and intelligent control of circuit breakers. The system automatically performs circuit breaker operations through real-time data analysis and fault detection to ensure the stability and safety of the power system. Its main purpose is to improve the intelligence level of the power system and provide more efficient fault response and power management capabilities. It is widely used in smart grids, industrial automation and large-scale power infrastructure.

[0004] Traditional control systems lack effective prediction and real-time analysis mechanisms. Their limitations make it impossible to effectively intervene before a fault occurs, thereby increasing the risk and maintenance cost of circuit breakers. The lack of real-time data analysis capabilities also means that control operations cannot be adjusted immediately or potential faults cannot be prevented, resulting in untimely responses in emergency situations, causing the power system to face higher safety risks. In addition, traditional control systems fail to make full use of cloud computing resources, resulting in functions such as data analysis, maintenance optimization, and fault prediction failing to reach optimal levels, limiting the development of intelligent power equipment management. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a circuit breaker control system based on cloud data.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a circuit breaker control system based on cloud data, the system comprising:

[0007] The data collection module collects current changes, temperature fluctuations and vibration frequencies in real time based on the installed current, temperature and vibration sensors, continuously transmits data to the cloud through LoRa technology, calculates the current status of the circuit breaker, and generates a status monitoring snapshot;

[0008] The data stream processing module performs real-time data stream analysis based on the state monitoring snapshot, detects abnormal indicators, including sudden current increase and temperature peak, and adjusts the data processing flow according to the analysis results, automatically performs the opening and closing operations of the circuit breaker, and generates a fault alarm signal;

[0009] The fault diagnosis and prediction module applies random forest and support vector machine algorithms based on the fault alarm signal to analyze the fault mode of the sensor data, predict the fault development trend, and automatically adjust the circuit breaker protection settings to generate fault trend warning information;

[0010] Based on the fault trend warning information, the system management and optimization module uses the cloud platform to analyze historical data, identify circuit breaker performance degradation trends and operational problems, adjust and optimize the maintenance cycle and control logic of the circuit breaker, and generate a maintenance optimization record summary.

[0011] The present invention is improved in that the steps of obtaining the status monitoring snapshot are specifically as follows:

[0012] Based on the installed current, temperature and vibration sensors, the current change, temperature fluctuation and vibration frequency data are collected in real time to obtain real-time data sets;

[0013] Use the formula to calculate the average value of the real-time data set:

[0014]

[0015] Get monitoring indicator m, where a i ,t i and v i They represent the current, temperature and vibration values ​​of the ith data point, respectively, n represents the total number of data points, α, β and γ are weight coefficients used to adjust the influence of the parameters in the index;

[0016] The monitoring index m is continuously transmitted to the cloud through LoRa technology, and the data trend is analyzed using the threshold judgment method:

[0017]

[0018] By comparing the monitoring indicator m with the normal operation threshold μ, the current circuit breaker state is judged, the predicted state label is obtained, and the current state monitoring snapshot of the circuit breaker is obtained.

[0019] The present invention is improved in that the abnormal indicator detection step is specifically as follows:

[0020] Based on the state monitoring snapshot, current and temperature data are extracted using the formula:

[0021]

[0022] and

[0023]

[0024] Calculate the standardized deviation of current and temperature, where ΔI is the standard deviation of current, ΔT is the standard deviation of temperature, and I is the current value of a single data point. is the average value of the current, σ I is the standard deviation of the current data, T is the temperature value of a single data point, is the average temperature, σ T is the standard deviation of the temperature data;

[0025] Apply anomaly detection logic to current deviation and temperature deviation using the formula:

[0026]

[0027] Compare the weighted deviation of current and temperature deviation with the threshold to determine whether it is abnormal, where θ is the fault threshold, and α and β are the weight coefficients of current and temperature deviation;

[0028] According to the anomaly detection results, the abnormal state in the data is marked, and the time point corresponding to the abnormal state is selected to obtain the anomaly indicator.

[0029] The present invention is improved in that the step of obtaining the fault alarm signal is specifically as follows:

[0030] Based on the detected abnormal indicators, the fault detection logic is activated, using the formula:

[0031]

[0032] Calculate the square root of the weighted sum of the current and temperature deviations to obtain the abnormal severity D, where ΔI and ΔT are the current and temperature deviations, ω I and ω T are the weight coefficients of current and temperature;

[0033] Compare the abnormal severity D and the fault response threshold, using the formula:

[0034]

[0035] Determine whether an alarm needs to be issued, where k is a constant for adjusting the steepness of the curve, λ is the fault response threshold, and e is a natural constant;

[0036] If the alarm determination result exceeds the threshold, the fault alarm is triggered and a fault alarm signal is obtained. The judgment method is: triggering the alarm = alarm determination ≥ 0.5.

[0037] The present invention is improved in that the analysis steps of the failure mode are specifically as follows:

[0038] Based on the fault alarm signal, current and temperature readings are extracted using the formula:

[0039] X1=log(βe i +∈)

[0040] and

[0041]

[0042] The current and temperature data are transformed to obtain the original data feature set, where e i represents the current reading at a single point in time, t i represents the corresponding temperature reading, α, β, γ and ∈ are parameters for adjusting the data scale;

[0043] Apply random forest and support vector machine algorithms to learn and classify the original data feature set using the formula:

[0044] M = SVM (X1, X2, σ1, σ2)

[0045] and

[0046] N = RF(X1, X2, δ1, δ2)

[0047] Among them, SVM is support vector machine, RF is random forest, X1 and X2 are processed features, σ1, σ2, δ1 and δ2 are algorithm adjustment parameters, and M and N represent the recognized pattern results;

[0048] Based on the output of the random forest and support vector machine algorithms, a weighted average method is used to determine the failure mode using the formula:

[0049]

[0050] Among them, F represents the determined fault mode, M and N represent the identified mode results, and λ1 and λ2 are the weights assigned to the SVM and RF outputs.

[0051] The present invention is improved in that the step of obtaining the fault trend warning information is specifically as follows:

[0052] Based on the identified fault mode, the fault development rate and impact range are calculated using the formula:

[0053]

[0054] Obtain fault development data, where R is the fault development rate, F1, F2 and F3 represent the differential fault characteristic parameters obtained from the fault mode analysis, and β1, β2 and β3 are weight coefficients used to evaluate the potential speed and impact of fault development;

[0055] Compare the fault development data with historical trends and apply the formula:

[0056]

[0057] Predict the potential failure points in the future and obtain the failure probability P, where R is the failure development rate, θ is the threshold, and k is the adjustment response sensitivity constant;

[0058] To adjust the protection settings of the circuit breaker based on operating parameters and environmental conditions, use the formula:

[0059] Warning = if P>0.5 then “high risk” else “normal”

[0060] According to the value of P, the fault risk level is judged and identified, and fault trend warning information is obtained.

[0061] The present invention is improved in that the steps of identifying the performance degradation trend and operation problem are specifically as follows:

[0062] Based on the fault trend warning information, historical data is collected and analyzed through the cloud platform, using the formula:

[0063]

[0064] Calculate the performance degradation index of the circuit breaker and get the average performance degradation score T, where age i Indicates the service life of the circuit breaker at the time of the i-th record, fail i represents the corresponding number of failures, n is the total number of records, α and β are weight coefficients used to adjust the impact of age and number of failures in the performance decay score, and γ and δ are regularization coefficients used to balance the denominator;

[0065] According to the average performance decay score T, statistical analysis methods are applied to detect outliers in the data using the formula:

[0066]

[0067] Among them, T i is the performance decay score of each record, M is the moving average of the performance score, ξ and η are normalization coefficients used to adjust the sensitivity of anomaly detection, and Q represents a quantitative indicator of the operational problem;

[0068] According to the value of Q, the operational health of the circuit breaker is evaluated. If Q>τ, there is an operational problem, otherwise the operation is normal, and τ is the risk threshold.

[0069] The present invention is improved in that the steps of obtaining the maintenance optimization record summary are specifically as follows:

[0070] Based on the identified performance degradation trends and operational issues, the formula is used:

[0071]

[0072] Determine the current maintenance requirements of the circuit breaker, where S i represents the performance decay score of the i-th record, O i represents the severity of the operational problem, w1 and w2 are weight parameters used to adjust the impact of performance degradation and operational problems in the maintenance requirement decision, n is the number of records, and R represents the average maintenance requirement score;

[0073] According to the maintenance demand data, combined with historical maintenance records and scheduled maintenance cycles, the maintenance cycle of the circuit breaker is adjusted and optimized using the formula:

[0074] M=if R>θthen“increase maintenance frequency”else“maintain current frequency”

[0075] Generate maintenance adjustment results and obtain a summary of maintenance optimization records, where θ is the threshold of the maintenance requirement score, which is used to determine whether the maintenance frequency needs to be increased, and M represents the maintenance adjustment plan.

[0076] Compared with the prior art, the advantages and positive effects of the present invention are:

[0077] In the present invention, the operation of the circuit breaker is automatically performed through real-time data analysis and prediction algorithms, which significantly improves the real-time monitoring and fault response capabilities of the power system. The sudden changes in current and temperature are monitored in real time, so that the system can quickly identify and respond to potential faults. By automatically performing the opening and closing operations of the circuit breaker, equipment damage and power outages are effectively prevented. In addition, the cloud platform is used to deeply analyze historical data, optimize maintenance cycles and control logic, make power management more efficient, and ensure the continuity of power supply and the long-term stable operation of equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 A module diagram of a circuit breaker control system based on cloud data is proposed for the present invention;

[0079] Figure 2 A flow chart for obtaining status monitoring snapshots in a circuit breaker control system based on cloud data is proposed for the present invention;

[0080] Figure 3 A flow chart for detecting abnormal indicators in a circuit breaker control system based on cloud data is proposed for the present invention;

[0081] Figure 4 A flow chart for obtaining fault alarm signals in a circuit breaker control system based on cloud data is proposed for the present invention;

[0082] Figure 5 A flow chart for analyzing failure modes in a circuit breaker control system based on cloud data is proposed for the present invention;

[0083] Figure 6 A flow chart for obtaining fault trend warning information in a circuit breaker control system based on cloud data is proposed for the present invention;

[0084] Figure 7 A flow chart for identifying performance degradation trends and operational problems in a circuit breaker control system based on cloud data is proposed for the present invention;

[0085] Figure 8 The present invention proposes a flow chart for obtaining a maintenance optimization record summary in a circuit breaker control system based on cloud data. DETAILED DESCRIPTION

[0086] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0087] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0088] Example

[0089] See also Figure 1 The present invention provides a technical solution: a circuit breaker control system based on cloud data includes:

[0090] The data collection module collects current changes, temperature fluctuations and vibration frequencies in real time based on the installed current, temperature and vibration sensors, continuously transmits data to the cloud through LoRa technology, calculates the current status of the circuit breaker, and generates a status monitoring snapshot;

[0091] The data stream processing module performs real-time data stream analysis based on the status monitoring snapshot, detects abnormal indicators, including sudden current and temperature peaks, and adjusts the data processing flow according to the analysis results, automatically performs the opening and closing operations of the circuit breaker, and generates a fault alarm signal;

[0092] The fault diagnosis and prediction module uses random forest and support vector machine algorithms based on fault alarm signals to analyze fault patterns of sensor data, predict fault development trends, and automatically adjust circuit breaker protection settings to generate fault trend warning information;

[0093] Based on fault trend warning information, the system management and optimization module uses the cloud platform to analyze historical data, identify circuit breaker performance degradation trends and operational problems, adjust and optimize the circuit breaker maintenance cycle and control logic, and generate a maintenance optimization record summary.

[0094] The status monitoring snapshots include real-time current readings, measured temperature values, and recorded vibration data. The fault alarm signals include identified current jumps and monitored temperature peaks. The fault trend warning information includes risk level assessment results, failure probability analysis results, and trend prediction results. The maintenance optimization record summary includes the adjusted maintenance schedule, updated operating guidelines, and performance degradation factors.

[0095] See also Figure 2 , the specific steps for obtaining the status monitoring snapshot are:

[0096] Based on the installed current, temperature and vibration sensors, the current change, temperature fluctuation and vibration frequency data are collected in real time to obtain real-time data sets;

[0097] Use the formula to calculate the average value of the real-time data set:

[0098]

[0099] Get monitoring indicator m, where a i ,t i and v i They represent the current, temperature and vibration values ​​of the ith data point, respectively, n represents the total number of data points, α, β and γ are weight coefficients used to adjust the influence of the parameters in the index;

[0100] The monitoring index m is continuously transmitted to the cloud through LoRa technology, and the data trend is analyzed using the threshold judgment method:

[0101]

[0102] By comparing the monitoring indicator m with the normal operation threshold μ, the current circuit breaker state is judged, the predicted state label is obtained, and the current state monitoring snapshot of the circuit breaker is obtained.

[0103] Assume the following data:

[0104] α = 0.5;

[0105] β = 0.3;

[0106] γ=0.2.

[0107] Assume there are 3 data points with the following values:

[0108] a1=10A, a2=12A, a3=11A;

[0109] t1=30℃, t2=35℃, t3=33℃;

[0110] v1=50Hz, v2=55Hz, v3=53Hz,

[0111] Calculate the contribution value of each data:

[0112] Current term = 0.5×(10 2 +12 2 +11 2 )=0.5×(100+144+121)=182.5;

[0113]

[0114] Vibration term = 0.2×(50+55+53) = 0.2×158 = 31.6.

[0115] Calculate the comprehensive index m:

[0116]

[0117] Assume the following data:

[0118] The status is assumed to be 1 for normal and 0 for abnormal;

[0119] m obtained monitoring indicators;

[0120] The threshold for normal operation of μ is set to 70.

[0121] Determine the relationship between m and μ:

[0122] Since m=73.07 and μ=70, the state is 1, which is normal.

[0123] See also Figure 3 ,The specific steps for detecting abnormal indicators are:

[0124] Based on the status monitoring snapshot, extract the current and temperature data using the formula:

[0125]

[0126] and

[0127]

[0128] Calculate the standardized deviation of current and temperature, where ΔI is the standard deviation of current, ΔT is the standard deviation of temperature, and I is the current value of a single data point. is the average value of the current, σ I is the standard deviation of the current data, T is the temperature value of a single data point, is the average temperature, σ T is the standard deviation of the temperature data;

[0129] Apply anomaly detection logic to current deviation and temperature deviation using the formula:

[0130]

[0131] Compare the weighted deviation of current and temperature deviation with the threshold to determine whether it is abnormal, where θ is the fault threshold, and α and β are the weight coefficients of current and temperature deviation;

[0132] According to the anomaly detection results, the abnormal state in the data is marked, and the time point corresponding to the abnormal state is selected to obtain the anomaly indicator.

[0133] Assume the following data:

[0134] Assume there are three current data points: 10A, 12A, 11A;

[0135] Temperature data points: 30°C, 35°C, 33°C;

[0136]

[0137] Calculate the standard deviation:

[0138]

[0139] Calculation process:

[0140]

[0141] Assume the following data:

[0142] Abnormal, mark whether it is abnormal, 1 is abnormal, 0 is normal;

[0143] ΔI and ΔT are the standard deviations obtained above;

[0144] α and β are used to balance the weights of current and temperature in abnormality judgment, and are assumed to be 1;

[0145] θ is the fault judgment threshold, which is assumed to be 1.5.

[0146] Calculation process:

[0147] Combined with the given ΔI=-1.22 and ΔT=-1.28, as well as α=1 and β=1, calculate the abnormal state judgment:

[0148]

[0149] Since 1.77 is greater than the threshold value θ=1.5, anomaly=1 is obtained.

[0150] See also Figure 4 , the specific steps for obtaining the fault alarm signal are:

[0151] Based on the detected abnormal indicators, the fault detection logic is activated, using the formula:

[0152] Calculate the square root of the weighted sum of the current and temperature deviations to obtain the abnormal severity D, where ΔI and ΔT are the current and temperature deviations, ω I and ω T are the weight coefficients of current and temperature;

[0153] Compare the abnormal severity D and the fault response threshold, using the formula:

[0154]

[0155] Determine whether an alarm needs to be issued, where k is a constant for adjusting the steepness of the curve, λ is the fault response threshold, and e is a natural constant;

[0156] If the alarm determination result exceeds the threshold, the fault alarm is triggered and a fault alarm signal is obtained. The judgment method is: triggering the alarm = alarm determination ≥ 0.5.

[0157] Assume the following data:

[0158] I = 10A, Therefore, ΔI = 10-8 = 2A;

[0159] T=40℃, Therefore, ΔT = 40-35 = 5°C;

[0160] Setting up I =0.5 and ω T =0.3.

[0161] Calculate D:

[0162]

[0163] Assume the following data:

[0164] λ = 15;

[0165] k = 1;

[0166] D = 17.17 was obtained.

[0167] Calculation alerts:

[0168]

[0169] Trigger alarm = 0.898 ≥ 0.5. Since the alarm determination result is greater than 0.5, it is determined that a failure alarm should be issued.

[0170] See also Figure 5 , the specific steps of failure mode analysis are:

[0171] Based on the fault alarm signal, the current and temperature readings are extracted using the formula:

[0172] X1=log(βe i +∈)

[0173] and

[0174]

[0175] Transform the current and temperature data to obtain the original data feature set and optimize the accuracy of the model analysis, where e i represents the current reading at a single point in time, t i represents the corresponding temperature reading, α, β, γ and ∈ are parameters for adjusting the data scale;

[0176] Apply random forest and support vector machine algorithms to learn and classify the original data feature set using the formula:

[0177] M = SVM (X1, X2, σ1, σ2)

[0178] and

[0179] N = RF(X1, X2, δ1, δ2)

[0180] Among them, SVM is support vector machine, RF is random forest, X1 and X2 are processed features, σ1, σ2, δ1 and δ2 are algorithm adjustment parameters, and M and N represent the recognized pattern results;

[0181] Based on the output of the random forest and support vector machine algorithms, a weighted average method is used to determine the failure mode using the formula:

[0182]

[0183] Among them, F represents the determined fault mode, M and N represent the identified mode results, and λ1 and λ2 are the weights assigned to the SVM and RF outputs.

[0184] Assume the following data:

[0185] Assume β=1,∈=1,α=1,γ=1;

[0186] Assuming the current reading is e i =10A, temperature reading t i =50℃.

[0187] Calculation process:

[0188] X1=log1×10+1=log11≈2.40

[0189]

[0190] Assume the following data:

[0191] Parameter setting example:

[0192] Assume that σ1=0.5, σ2=0.5, δ1=0.3, δ2=0.3.

[0193] SVM prediction:

[0194] Assume that the SVM model is configured with a linear kernel and the model is adjusted using parameters σ1 and σ2. The features and parameters are trained on the training data and a prediction M=0.75 is given for the new input (X1 and X2).

[0195] RF Prediction:

[0196] RF uses δ1 and δ2 as tree construction parameters (involving feature selection or segmentation threshold adjustment). For the same input (X1 and X2), RF gives a prediction of N=0.85.

[0197] Assume the following data:

[0198] Assume λ1=0.6,λ2=0.4;

[0199] M=0.75, N=0.85.

[0200] Calculation process:

[0201]

[0202] The final failure mode was determined to have a probability of 0.79, which combined the prediction results of the two models and increased the stability and credibility of the prediction.

[0203] See also Figure 6 ,The specific steps for obtaining fault trend warning information are as follows:

[0204] Based on the identified failure modes, calculate the failure development rate and impact range using the formula:

[0205]

[0206] Obtain fault development data, where R is the fault development rate, F1, F2 and F3 represent the differential fault characteristic parameters obtained from the fault mode analysis, and β1, β2 and β3 are weight coefficients used to evaluate the potential speed and impact of fault development;

[0207] Compare the fault development data with historical trends and apply the formula:

[0208]

[0209] Predict the potential failure points in the future and obtain the failure probability P, where R is the failure development rate, θ is the threshold, and k is the adjustment response sensitivity constant;

[0210] To adjust the protection settings of the circuit breaker based on operating parameters and environmental conditions, use the formula:

[0211] Warning = if P > 0.5 then "high risk" else "normal"

[0212] According to the value of P, the fault risk level is judged and identified, and fault trend warning information is obtained.

[0213] Assume the following data:

[0214] Assume F1 = 2.0, F2 = 16.0, F3 = -4.0;

[0215] The weights are assumed to be β1=0.5, β2=0.3, and β3=0.2.

[0216] Calculation process:

[0217]

[0218] |F3|=|-4.0|=4.0

[0219]

[0220] Assume the following data:

[0221] θ = 3.5;

[0222] k=1.0.

[0223] Calculation process:

[0224] R-θ=4.0-3.5=0.5

[0225]

[0226] This indicates that the predicted probability of failure occurrence is 0.6225, or about 62%, indicating a relatively high risk.

[0227] See also Figure 7 , the steps to identify performance degradation trends and operational problems are as follows:

[0228] Based on the fault trend warning information, historical data is collected and analyzed through the cloud platform, using the formula:

[0229]

[0230] Calculate the performance degradation index of the circuit breaker and get the average performance degradation score T, where age i Indicates the service life of the circuit breaker at the time of the i-th record, fail i represents the corresponding number of failures, n is the total number of records, α and β are weight coefficients used to adjust the impact of age and number of failures in the performance decay score, and γ and δ are regularization coefficients used to balance the denominator;

[0231] According to the average performance decay score T, statistical analysis methods are applied to detect outliers in the data using the formula:

[0232]

[0233] Among them, T i is the performance decay score of each record, M is the moving average of the performance score, which is used to smooth short-term fluctuations, ξ and η are normalization coefficients used to adjust the sensitivity of anomaly detection, and Q represents a quantitative indicator of the operational problem;

[0234] According to the value of Q, the operational health of the circuit breaker is evaluated. If Q>τ, there is an operational problem, otherwise the operation is normal, and τ is the risk threshold.

[0235] Assume the following data:

[0236] Assume there are 3 record points, age i =[10, 15, 20] years, fail i =[2, 5, 3] times;

[0237] The parameters are set to α=0.5, β=0.3, γ=0.1, δ=1.

[0238] Calculation process:

[0239] Calculate the contribution of each recording point:

[0240]

[0241] Summing and averaging:

[0242]

[0243] Assume the following data:

[0244] The obtained T = [50.424, 113.171, 200.519];

[0245]

[0246] The parameters are set to ξ=1, η=3.

[0247] Calculation process:

[0248] Compute the absolute value of the difference between each score and the mean:

[0249] |T1-M|=|50.424-121.371|=70.947

[0250] |T2-M|=|113.171-121.371|=8.2

[0251] |T3-M|=|200.519-121.371|=79.148

[0252] Sum and Normalize:

[0253]

[0254] See also Figure 8 , the specific steps for obtaining the maintenance optimization record summary are:

[0255] Based on the identified performance degradation trends and operational issues, the formula is used:

[0256]

[0257] Determine the current maintenance requirements of the circuit breaker, where S i represents the performance decay score of the i-th record, O i represents the severity of the operational problem, w1 and w2 are weight parameters used to adjust the impact of performance degradation and operational problems in the maintenance requirement decision, n is the number of records, and R represents the average maintenance requirement score;

[0258] According to the maintenance demand data, combined with historical maintenance records and scheduled maintenance cycles, the maintenance cycle of the circuit breaker is adjusted and optimized using the formula:

[0259] M=if R>θthen“increase maintenance frequency”else“maintain current frequency”

[0260] Generate maintenance adjustment results and obtain a summary of maintenance optimization records, where θ is the threshold of the maintenance requirement score, which is used to determine whether the maintenance frequency needs to be increased, and M represents the maintenance adjustment plan.

[0261] Assume the following data:

[0262] Assume there are 3 records, S i =[20,30,40],O i = [10, 15, 20];

[0263] The weights are set to w1=0.6, w2=0.4, and the total number of records n=3.

[0264] Calculation process:

[0265] Calculate the contribution of each record:

[0266] w1·S1+w2·O1=0.6·20+0.4·10=12+4=16

[0267] w1·S2+w2·O2=0.6·30+0.4·15=18+6=24

[0268] w1·S3+w2·O3=0.6·40+0.4·20=24+8=32

[0269] Calculate the average score:

[0270]

[0271] Assume that the threshold θ=25. Since R=24, which is less than 25, the result is: maintain the current maintenance frequency.

[0272] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A circuit breaker control system based on cloud data, characterized in that: The system comprises: The data collection module collects current changes, temperature fluctuations and vibration frequencies in real time based on the installed current, temperature and vibration sensors, continuously transmits data to the cloud through LoRa technology, calculates the current status of the circuit breaker, and generates a status monitoring snapshot; The data stream processing module performs real-time data stream analysis based on the state monitoring snapshot, detects abnormal indicators, including sudden current increase and temperature peak, and adjusts the data processing flow according to the analysis results, automatically performs the opening and closing operations of the circuit breaker, and generates a fault alarm signal; The fault diagnosis and prediction module applies random forest and support vector machine algorithms based on the fault alarm signal to analyze the fault mode of the sensor data, predict the fault development trend, and automatically adjust the circuit breaker protection settings to generate fault trend warning information; Based on the fault trend warning information, the system management and optimization module uses the cloud platform to analyze historical data, identify circuit breaker performance degradation trends and operational problems, adjust and optimize the maintenance cycle and control logic of the circuit breaker, and generate a maintenance optimization record summary; The steps for obtaining the status monitoring snapshot are specifically as follows: Based on the installed current, temperature and vibration sensors, the current change, temperature fluctuation and vibration frequency data are collected in real time to obtain real-time data sets; Use the formula to calculate the average value of the real-time data set: Get monitoring indicators ,in, , and Respectively represent Current, temperature and vibration values ​​for each data point, represents the total number of data points, , and is the weight coefficient, which is used to adjust the influence of the parameter in the indicator; Through LoRa technology, the monitoring indicators Continuously transmit to the cloud and analyze data trends using the threshold judgment method: state ; According to monitoring indicators Normal operation threshold Compare and determine the current circuit breaker status, obtain the predicted status label, and obtain the current status monitoring snapshot of the circuit breaker; The abnormal indicator detection steps are specifically as follows: Based on the state monitoring snapshot, current and temperature data are extracted using the formula: and Calculate the normalized deviation of current and temperature, where is the standard deviation of the current, is the standard deviation of temperature, is the current value of a single data point, is the average value of the current, is the standard deviation of the current data, is the temperature value of a single data point, is the average temperature, is the standard deviation of the temperature data; Apply anomaly detection logic to current deviation and temperature deviation using the formula: abnormal ; The weighted deviation of current and temperature deviation is compared with the threshold to determine whether it is abnormal. is the fault threshold, and is the weight coefficient of current and temperature deviation; According to the anomaly detection results, mark the abnormal state in the data, and select the time point corresponding to the abnormal state to obtain the abnormal index; The analysis steps of the failure mode are specifically as follows: Based on the fault alarm signal, current and temperature readings are extracted using the formula: and The current and temperature data are transformed to obtain the original data feature set, where: Represents the current reading at a single point in time, Indicates the corresponding temperature reading, , , and is a parameter to adjust the data scale; Apply random forest and support vector machine algorithms to learn and classify the original data feature set using the formula: and Among them, SVM is support vector machine, RF is random forest, and is the processed feature, , , and is the algorithm tuning parameter, and Indicates the recognized pattern result; Based on the output of the random forest and support vector machine algorithms, a weighted average method is used to determine the failure mode using the formula: in, Indicates a certain failure mode, and Indicates the recognized pattern result, and are the weights assigned to the SVM and RF outputs.

2. The circuit breaker control system based on cloud data according to claim 1, characterized in that: The steps of obtaining the fault alarm signal are specifically as follows: Based on the detected abnormal indicators, the fault detection logic is activated, using the formula: Calculate the square root of the weighted sum of the current and temperature deviations to get the severity of the anomaly ,in, and is the deviation between current and temperature, and are the weight coefficients of current and temperature; Comparing abnormality severity and the fault response threshold, using the formula: Determine whether an alarm needs to be issued, where: is a constant that adjusts the steepness of the curve, is the fault response threshold, is a natural constant; If the alarm judgment result exceeds the threshold, the fault alarm is triggered and a fault alarm signal is obtained. The judgment method is: trigger alarm = ≥0.

5.

3. The circuit breaker control system based on cloud data according to claim 1, characterized in that: The steps for obtaining the fault trend warning information are specifically as follows: Based on the identified fault mode, the fault development rate and impact range are calculated using the formula: Obtain fault development data, where is the failure development rate, , and represents the differential fault characteristic parameters obtained from the failure mode analysis, , and is a weighting factor used to assess the potential speed and impact of fault development; Compare the fault development data with historical trends and apply the formula: Predict potential future failure points and obtain failure probability ,in, is the failure growth rate, is the threshold value, is the adjustment response sensitivity constant; To adjust the protection settings of the circuit breaker based on operating parameters and environmental conditions, use the formula: Warning = if >0.5 then "high risk" else "normal" according to value, judge and identify the fault risk level, and obtain fault trend warning information.

4. The circuit breaker control system based on cloud data according to claim 1, characterized in that: The steps for identifying the performance degradation trend and operational problems are as follows: Based on the fault trend warning information, historical data is collected and analyzed through the cloud platform, using the formula: Calculate the performance degradation index of the circuit breaker and get the average performance degradation score ,in, Indicates that the circuit breaker is Years of use at the time of the first record, Indicates the corresponding number of failures, is the total number of records, and is the weight coefficient used to adjust the impact of age and number of failures in the performance degradation score, and is the regularization coefficient, used to balance the denominator; According to the average performance decay score , apply statistical analysis methods to detect outliers in the data, using the formula: in, is the performance decay score for each record, is the moving average of the performance scores, and is the normalization coefficient used to adjust the sensitivity of anomaly detection, Quantitative indicators that indicate operational problems; according to The value of , evaluates the operational health of the circuit breaker. , there is an operation problem, otherwise the operation is normal. is the risk threshold.

5. The circuit breaker control system based on cloud data according to claim 1, characterized in that: The steps for obtaining the maintenance optimization record summary are specifically as follows: Based on the identified performance degradation trends and operational issues, the formula is used: Determine the current maintenance needs of the circuit breaker, where: Indicates Record performance decay scores, Indicates the severity of the operational problem, and is a weight parameter used to adjust the impact of performance degradation and operational problems in maintenance requirement decisions, is the number of records, represents the average maintenance need score; According to the maintenance demand data, combined with historical maintenance records and scheduled maintenance cycles, the maintenance cycle of the circuit breaker is adjusted and optimized using the formula: = if > then"Increase maintenance frequency"else"Maintain current frequency" Generate maintenance adjustment results and obtain a summary of maintenance optimization records, including: is the threshold of the maintenance requirement score, which is used to decide whether to increase the maintenance frequency. Indicates maintenance adjustment plan.

Images