An intelligent electrical box monitoring system

Abstract

The present invention relates to the field of intelligent monitoring technology, and specifically to an intelligent electrical box monitoring system, which includes a current fluctuation analysis module, a micro-variable current monitoring module, an electromagnetic field monitoring module, an abnormality positioning module, a fault warning module, and a fault diagnosis and suggestion module. In the present invention, by using wavelet transform and random forest algorithm to refine current signal processing and fault type classification, fault positioning is made more accurate, avoiding misjudgment of traditional systems, and monitoring the electromagnetic field at the same time, strengthening the comprehensive perception of the working environment of electrical equipment, providing more comprehensive data support, and providing a new dimension for fault source positioning. Real-time warning and fault diagnosis suggestions significantly reduce response time and maintenance costs, while ensuring the stability and safety of power supply, significantly improving the reliability and performance of the electrical system, and improving the accuracy and efficiency of fault diagnosis.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to an intelligent electrical box monitoring system. Background Art

[0002] The field of intelligent monitoring technology involves the use of advanced information processing technology, sensor technology, data communication technology, and artificial intelligence algorithms to monitor, analyze, and manage various objects and environments in real time. It is particularly important in electrical safety and maintenance, and can monitor the working status of the electrical system in real time to prevent electrical fires, failures, and other potential safety issues. Through intelligent monitoring, the operating efficiency of the electrical system can be optimized, and problems can be discovered and solved in a timely manner, thereby ensuring the safety of personnel and the stable operation of electrical equipment.

[0003] Among them, the intelligent electrical box monitoring system is a monitoring solution that integrates multiple intelligent technologies. It is specifically designed to perform real-time monitoring of the performance and safety of electrical boxes and their internal components. The purpose is to achieve early identification and early warning of potential safety hazards by continuously monitoring the working status and environmental conditions of the electrical box, protect electrical equipment from damage, ensure the stability and safety of power supply, and aim to reduce downtime caused by electrical failures, reduce maintenance costs, and improve the overall performance and reliability of the electrical system.

[0004] Traditional electrical box monitoring systems have obvious deficiencies in the monitoring and analysis of subtle current changes and the perception and processing of electromagnetic field anomalies, resulting in inaccurate diagnosis and positioning of potential faults. Maintenance work relies heavily on experience-based judgment, increasing the risk of misjudgment. The lack of efficient early warning and fault diagnosis suggestion mechanisms prolongs fault response time and increases maintenance costs. In severe cases, it may even affect the stability and safety of the power supply, reduce the operating efficiency of the electrical system and the long-term stability of the equipment, and increase the potential risk of electrical safety accidents. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose an intelligent electrical box monitoring system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: an intelligent electrical box monitoring system comprises:

[0007] The current fluctuation analysis module collects real-time current data of the electrical box, analyzes current fluctuations, identifies high-risk areas, and obtains current abnormality indicators;

[0008] The micro-variable current monitoring module collects subtle current changes based on the current abnormality index, uses wavelet transform to subdivide the current signal, combines the random forest algorithm to classify the fault type, and obtains the micro-variable current characteristics;

[0009] The electromagnetic field monitoring module utilizes the micro-variable current characteristics to monitor the electromagnetic field changes, collect electromagnetic field data, construct an electromagnetic field distribution map, and obtain electromagnetic field abnormality indicators;

[0010] The abnormality location module calculates the fault point based on the electromagnetic field abnormality index, uses finite element analysis and Maxwell equations to locate the fault source and generate fault location information;

[0011] The warning information issuing module sends warning information according to the fault location information and obtains warning execution feedback;

[0012] The fault diagnosis and suggestion module combines the early warning execution feedback, analyzes the cause of the fault, combines the historical maintenance data and the fault library, proposes repair or replacement suggestions, and builds maintenance and preventive measures.

[0013] As a further solution of the present invention, the current abnormality indicators include current surge points, current reduction points and abnormal current duration; the micro-change current characteristics are specifically small current fluctuation patterns, short-time current jump times and current micro-change periodic characteristics; the electromagnetic field abnormality indicators include abnormal electromagnetic wave amplitude, non-periodic electromagnetic wave changes and electromagnetic interference frequency distribution; the fault location information is specifically the circuit board number where the fault occurs, the sequence of faulty electrical components and the physical location of the fault; the early warning execution feedback includes the division of early warning levels, the allocation of early warning response personnel and the efficiency of early warning information transmission; the maintenance and preventive measures include fault cause analysis records, maintenance operation steps and recommendations for future preventive measures.

[0014] As a further solution of the present invention, the micro-variable current monitoring module includes:

[0015] The current change acquisition submodule acquires subtle changes in the current based on the current abnormality index, including recording the current value every millisecond and frequency analysis of current fluctuations, to obtain a subtle current data set;

[0016] The signal subdivision processing submodule uses wavelet transform to subdivide the current signal based on the subtle current data set, decomposes the current signal into multiple frequency bands, identifies subtle changes and abnormal fluctuations in the current, and obtains the current signal subdivision result;

[0017] The fault classification analysis submodule performs classification analysis of fault types based on the current signal segmentation results and in combination with the random forest algorithm. By comparing the current change characteristics under different fault types, the potential fault type corresponding to the current current change is identified and the slight change current characteristics are generated.

[0018] As a further solution of the present invention, the random forest algorithm is according to the formula:

[0019] ;

[0020] Compute the improved feature importance score where, Score the improved feature importance, is the set of all decision trees that make up the random forest, For Tree A split node in For splitting nodes Because of the characteristics The reduction in impurity due to the use of Features The frequency of occurrence, Features Correlation with fault category, Features The coefficient of variation, , , is the weight coefficient.

[0021] As a further solution of the present invention, the electromagnetic field monitoring module includes:

[0022] The characteristic electromagnetic field acquisition submodule collects corresponding electromagnetic field change data based on the micro-variable current characteristics, reveals the real-time state of the electromagnetic field around the electrical box, and obtains a characteristic electromagnetic field data set;

[0023] The electromagnetic field data analysis submodule analyzes the change trend and abnormal pattern of the electromagnetic field based on the characteristic electromagnetic field data set, identifies abnormal changes in the electromagnetic field by comparing the electromagnetic field data at different times and locations, and obtains electromagnetic field abnormality analysis results;

[0024] The electromagnetic field distribution diagram construction submodule constructs a diagram showing the distribution of the electromagnetic field in the electrical box and its surroundings based on the electromagnetic field anomaly analysis result, and generates an electromagnetic field anomaly index.

[0025] As a further solution of the present invention, the abnormality locating module includes:

[0026] The electromagnetic field data processing submodule filters, denoises and normalizes the collected electromagnetic field data based on the electromagnetic field anomaly index to obtain the processed electromagnetic field data;

[0027] The fault point calculation submodule calculates the potential fault point location based on the processed electromagnetic field data, using the finite element analysis method, combined with the distribution and intensity changes of the electromagnetic field, and generates the fault point location estimation result;

[0028] The fault source location submodule uses Maxwell's equations to analyze the electromagnetic field around each estimated point based on the fault point location estimation results, locates the exact location of the fault source, and obtains the fault location information.

[0029] As a further solution of the present invention, the finite element analysis method is according to the formula:

[0030] ;

[0031] Calculate the electromagnetic energy density distribution, where is the improved electromagnetic energy density, is the electric field strength, is the electric displacement density, is the magnetic field strength, is the magnetic flux density, is the temperature of the material, is the conductivity change, is the material density, is the vibration speed, , , , is the weight coefficient.

[0032] As a further solution of the present invention, the fault diagnosis and suggestion module includes:

[0033] The fault cause analysis submodule performs in-depth analysis on the fault signal based on the early warning execution feedback, identifies the fault signal characteristics and fault cause, and obtains the fault cause analysis result;

[0034] The historical data comparison submodule matches and compares the fault cause analysis result with the historical maintenance data and the fault library to find similar fault cases and obtain similar fault comparison results;

[0035] The maintenance suggestion building submodule proposes targeted maintenance or replacement suggestions based on the similar fault comparison results, combined with equipment conditions and maintenance experience, and builds maintenance and preventive measures.

[0036] Compared with the prior art, the advantages and positive effects of the present invention are:

[0037] In the present invention, by utilizing wavelet transform and random forest algorithm to refine current signal processing and fault type classification, fault location is made more accurate, avoiding misjudgment of traditional systems. At the same time, the electromagnetic field is monitored to enhance the comprehensive perception of the working environment of electrical equipment, provide more comprehensive data support, and provide a new dimension for fault source location. Real-time warning and fault diagnosis suggestions significantly reduce response time and maintenance costs, while ensuring the stability and safety of power supply, significantly improving the reliability and performance of the electrical system, and enhancing the accuracy and efficiency of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a system flow chart of the present invention;

[0039] Figure 2 It is a schematic diagram of the system framework of the present invention;

[0040] Figure 3 This is a flow chart of the current fluctuation analysis module of the present invention;

[0041] Figure 4 This is a flow chart of the micro-current monitoring module of the present invention;

[0042] Figure 5 This is a flow chart of the electromagnetic field monitoring module of the present invention;

[0043] Figure 6 This is a flow chart of the abnormality location module of the present invention;

[0044] Figure 7 A flow chart of a module for issuing early warning information of the present invention;

[0045] Figure 8 This is a flow chart of the fault diagnosis and suggestion module of the present invention. DETAILED DESCRIPTION

[0046] 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.

[0047] 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" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the 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, "multiple" means two or more, unless otherwise clearly and specifically defined. Embodiment 1

[0048] See also Figure 1 and Figure 2 , an intelligent electrical box monitoring system includes a current fluctuation analysis module, a micro-variable current monitoring module, an electromagnetic field monitoring module, an abnormality positioning module, a warning information issuing module and a fault diagnosis and suggestion module;

[0049] The current fluctuation analysis module collects real-time current data of the electrical box, analyzes current fluctuations, identifies high-risk areas, and obtains current abnormality indicators;

[0050] The micro-current monitoring module collects subtle current changes based on current abnormality indicators, uses wavelet transform to subdivide current signals, and combines random forest algorithm to classify fault types and obtain micro-current characteristics;

[0051] The electromagnetic field monitoring module uses the characteristics of micro-variable current to monitor the changes in the electromagnetic field, collect electromagnetic field data, construct an electromagnetic field distribution map, and obtain electromagnetic field abnormality indicators;

[0052] The abnormality location module uses finite element analysis and Maxwell equations to calculate the fault point, locate the fault source, and generate fault location information based on the electromagnetic field abnormality index;

[0053] The warning information issuing module sends warning information according to the fault location information and obtains warning execution feedback;

[0054] The fault diagnosis and suggestion module analyzes the early warning execution feedback, analyzes the cause of the fault, combines historical maintenance data and fault library, proposes repair or replacement suggestions, and builds maintenance and preventive measures.

[0055] The current anomaly indicators include the current surge point, current decrease point and abnormal current duration. The micro-change current characteristics include the small current fluctuation pattern, the number of short-time current jumps and the periodic characteristics of current micro-changes. The electromagnetic field anomaly indicators include the abnormal electromagnetic wave amplitude, non-periodic electromagnetic wave changes and electromagnetic interference frequency distribution. The fault location information includes the circuit board number where the fault occurred, the sequence of faulty electrical components and the physical location of the fault. The early warning execution feedback includes the classification of warning levels, the allocation of early warning response personnel and the efficiency of early warning information transmission. The maintenance and preventive measures include fault cause analysis records, maintenance operation steps and recommendations for future preventive measures.

[0056] See also Figure 2 and Figure 3 ,The current fluctuation analysis module includes a current data collection submodule, a current fluctuation analysis submodule, and a high-risk area identification submodule;

[0057] The current data collection submodule collects the real-time current data of the electrical box, including the real-time value, change trend and fluctuation frequency of the current, and obtains the current data set;

[0058] In the current data collection submodule, the time series analysis method is adopted and Python's Pandas library is used to collect the real-time current data of the electrical box, including the real-time value, change trend and fluctuation frequency of the current. The DataFrame structure is used to store the current data set, setting the timestamp as the index and the current value, change trend and fluctuation frequency as the columns to ensure that the data is arranged in chronological order and retain the current data set.

[0059] The current fluctuation analysis submodule analyzes the current fluctuation pattern based on the current data set, including events where the current rises or falls sharply in a short period of time, and intervals where the current is continuously unstable, to obtain the current fluctuation pattern;

[0060] In the current fluctuation analysis submodule, the Fourier transform algorithm is adopted, and the fft function of the NumPy library is used to perform frequency domain analysis on the current values ​​in the current data set. The time domain signal is converted into a frequency domain signal through the fft function, and the current fluctuation pattern is analyzed, including events in which the current rises or falls sharply in a short period of time, and intervals in which the current is continuously unstable. The current fluctuation pattern is identified through the amplitude and frequency of the frequency domain signal, and the current fluctuation pattern is generated.

[0061] The high-risk area identification submodule identifies high-risk areas that may cause damage or failure of electrical equipment based on the current fluctuation pattern, including the location and duration of current surge points and current reduction points, and generates current anomaly indicators;

[0062] In the high-risk area identification submodule, a clustering algorithm, specifically the K-means algorithm, is adopted. The KMeans class of the scikit-learn library is used to perform cluster analysis on the current fluctuation pattern. The number of clusters is set to the preset number of high-risk levels. The features in the current fluctuation pattern, such as the location and duration of the current surge point and the current decrease point, are used as the input parameters of the clustering. The fit_predict method is executed for clustering. According to the clustering results, the high-risk areas that may cause damage or failure of electrical equipment are identified, and the current anomaly index is generated.

[0063] See also Figure 2 and Figure 4 ,The micro-variable current monitoring module includes a current change acquisition submodule, a signal segmentation processing submodule, and a fault classification and analysis submodule;

[0064] The current change acquisition submodule collects subtle changes in current based on current anomaly indicators, including recording the current value every millisecond and frequency analysis of current fluctuations, to obtain subtle current data sets;

[0065] In the current change acquisition submodule, based on the current anomaly index, a signal acquisition algorithm is adopted, and the analog signal is read using the Arduino programming environment. The current value is collected once every millisecond through the analog-to-digital converter (ADC) module, and the sampling rate is set to 1000 Hz. The real-time value, change trend and fluctuation frequency of the sampled current are collected, and the serial output function Serial.println is used to record and transmit the current value every millisecond to the connected computer. The frequency of current fluctuation is obtained through spectrum analysis to generate a subtle current data set.

[0066] The signal subdivision processing submodule uses wavelet transform to subdivide the current signal based on the subtle current data set, decomposes the current signal into multiple frequency bands, identifies subtle changes and abnormal fluctuations in the current, and obtains the current signal subdivision results;

[0067] In the signal segmentation processing submodule, based on the subtle current data set, the wavelet transform algorithm is adopted and the Python PyWavelets library is used for operation. The pywt.Wavelet function is called to select the Daubechies wavelet as the mother wavelet for wavelet transform. The decomposition level is set to 5 layers, and the current signal is decomposed into multiple frequency bands. Each frequency band represents a specific frequency range of the current signal. The wavelet reconstruction function pywt.waverec is used to identify subtle changes and abnormal fluctuations in the current and generate the current signal segmentation results.

[0068] The fault classification analysis submodule performs classification analysis of fault types based on the current signal segmentation results and combines the random forest algorithm. By comparing the current change characteristics under different fault types, it identifies the potential fault type corresponding to the current current change and generates the slight change current characteristics.

[0069] The random forest algorithm follows the formula:

[0070] ;

[0071] Compute the improved feature importance score where, Score the improved feature importance, is the set of all decision trees that make up the random forest, For Tree A split node in For splitting nodes Because of the characteristics The reduction in impurity due to the use of Features The frequency of occurrence of reflects the frequency with which the feature is used as a split node in all trees. Features Correlation with the fault class, which indicates the strength of the statistical association between the feature and the target fault class, Features The coefficient of variation of , which evaluates the volatility of the eigenvalues, , , is the weight coefficient, which adjusts the impact of each parameter on the final importance score.

[0072] The execution process is as follows: First, calculate the reduction in impurity caused by the feature splitting at the decision tree node ;

[0073] Then, add the features Frequency of occurrence Multiply by the weight factor , to account for the prevalence of features in random forests;

[0074] Next, add the features Correlation with fault category Multiply by the weight factor , consider the correlation between features and fault categories;

[0075] Finally, subtract the feature from the sum The coefficient of variation Multiply by the weight factor , to penalize features with high volatility.

[0076] Among them, the weight coefficient , , The determination of can be adjusted based on the cross-validated performance score through optimization algorithms such as grid search or genetic algorithms to optimize the accuracy of the model in classifying fault types. By improving the formula, the key current change characteristics that affect fault classification can be more accurately identified and evaluated, improving the accuracy and efficiency of fault diagnosis.

[0077] See also Figure 2 and Figure 5 ,The electromagnetic field monitoring module includes a characteristic electromagnetic field acquisition submodule, an electromagnetic field data analysis submodule, and an electromagnetic field distribution map construction submodule;

[0078] The characteristic electromagnetic field acquisition submodule collects the corresponding electromagnetic field change data based on the micro-variable current characteristics, reveals the real-time state of the electromagnetic field around the electrical box, and obtains the characteristic electromagnetic field data set;

[0079] In the characteristic electromagnetic field acquisition submodule, based on the characteristics of slightly variable current, an electromagnetic field detector is used. Arduino is used in conjunction with the electromagnetic field detection sensor. The sampling frequency is set to 1000 Hz, corresponding to the characteristic frequency of slightly variable current. The electromagnetic field change data around the electrical box is collected in real time. The sensor output is read using the analogRead function, and the electromagnetic field strength data is transmitted to the computer through the serial communication interface Serial.println. The real-time status of the electromagnetic field is recorded to generate a characteristic electromagnetic field data set.

[0080] The electromagnetic field data analysis submodule analyzes the changing trend and abnormal pattern of the electromagnetic field based on the characteristic electromagnetic field data set, identifies the abnormal changes of the electromagnetic field by comparing the electromagnetic field data at different times and locations, and obtains the abnormal analysis results of the electromagnetic field;

[0081] In the electromagnetic field data analysis submodule, based on the characteristic electromagnetic field data set, the time series analysis method is adopted, the Python Statsmodels library is used to perform operations, and the tsa.seasonal_decompose method is called to perform time series decomposition, analyze the trend component and seasonal component of the electromagnetic field data, and the irregular fluctuation of the residual. By comparing the electromagnetic field data at different times and locations, the residual fluctuation is used to identify the abnormal changes of the electromagnetic field, and obtain the electromagnetic field anomaly analysis results.

[0082] The electromagnetic field distribution diagram construction submodule constructs a diagram showing the electromagnetic field distribution of the electrical box and its surroundings based on the electromagnetic field anomaly analysis results, and generates electromagnetic field anomaly indicators;

[0083] In the electromagnetic field distribution map construction submodule, based on the electromagnetic field anomaly analysis results, geographic information system (GIS) technology is used, Python's Matplotlib library and GeoPandas library are used for operation, the GeoDataFrame structure is called to store the electromagnetic field data and the corresponding geographic location information, and the plot method is used to draw the electromagnetic field distribution map of the electrical box and its surroundings. Different colors are set to represent different electromagnetic field intensity intervals. The electromagnetic field distribution is intuitively displayed through charts, and electromagnetic field anomaly indicators are generated.

[0084] See also Figure 2 and Figure 6 ,The abnormal location module includes an electromagnetic field data processing submodule, a fault point calculation submodule and a fault source location submodule;

[0085] The electromagnetic field data processing submodule filters, denoises and normalizes the collected electromagnetic field data based on the electromagnetic field anomaly index to obtain the processed electromagnetic field data;

[0086] In the electromagnetic field data processing submodule, based on the electromagnetic field anomaly index, a low-pass filter algorithm is used, and the Python SciPy library is used to perform operations. The scipy.signal.butter function is called to create a filter, and the cutoff frequency is set to 50 Hz to remove high-frequency noise. The scipy.signal.filtfilt function is used to filter the electromagnetic field data, and then the signal.wiener function of SciPy is used for denoising. Finally, the numpy.linalg.norm function of NumPy is used to normalize the data to ensure that the electromagnetic field intensity value is between 0 and 1, so as to obtain the processed electromagnetic field data.

[0087] The fault point calculation submodule calculates the potential fault point location based on the processed electromagnetic field data, using the finite element analysis method, combined with the distribution and intensity changes of the electromagnetic field, and generates the fault point location estimation result;

[0088] Finite element analysis method, according to the formula:

[0089] ;

[0090] Calculate the electromagnetic energy density distribution, where is the improved electromagnetic energy density, is the electric field strength, is the electric displacement density, is the magnetic field strength, is the magnetic flux density, is the temperature of the material, reflecting the influence of ambient temperature on the distribution of electromagnetic field. is the conductivity change, reflecting the non-uniformity of the material conductivity. is the material density, reflecting the physical density of the material, is the vibration velocity, which reflects the possible mechanical vibration of materials or equipment in the electromagnetic field. , , , is the weight coefficient, which is used to adjust the contribution of various parameters to the calculation of electromagnetic energy density.

[0091] The execution process is as follows:

[0092] First, calculate the electric field strength and electric displacement density The product of , to consider the influence of electric field distribution;

[0093] Then calculate the magnetic field strength and magnetic flux density The product of , to consider the influence of magnetic field distribution;

[0094] Next, introduce new parameters to calculate the temperature of the material and conductivity changes The product of , to consider the effects of temperature changes and material conductivity inhomogeneity on the electromagnetic field distribution;

[0095] Finally, calculate the material density and vibration speed The product of the squares of , in order to consider the influence of mechanical vibration on the electromagnetic energy distribution.

[0096] Among them, determine the weight coefficient , , , The method steps include regression analysis based on historical data to optimize the influence of various parameters on the final electromagnetic energy density calculation results. Through this improved formula, the fault point location can be more accurately estimated, more factors affecting the electromagnetic field distribution are considered, and the accuracy of the calculation is improved.

[0097] The fault source location submodule uses Maxwell's equations to analyze the electromagnetic field around each estimated point based on the estimated fault point location, locates the exact location of the fault source, and obtains the fault location information;

[0098] In the fault source location submodule, based on the estimated results of the fault point location, Maxwell's equations are used, numerical methods are adopted, Python's NumPy and SciPy libraries are used for operation, and combined with electromagnetic field theory, numpy.meshgrid is called to create a two-dimensional grid to simulate the electrical box and its surrounding space. The integral of the electromagnetic field around each estimated point is calculated through the scipy.integrate.quad function, so as to analyze the changes in the electromagnetic field at different positions, determine the exact location of the fault source, and obtain the fault location information.

[0099] See also Figure 2 and Figure 7 ,The warning information issuing module includes a warning information formulation submodule, a warning information sending submodule, and a warning feedback collection submodule;

[0100] The early warning information formulation submodule formulates targeted early warning information content based on the fault location information, including the specific location of the fault, potential causes and preliminary response measures, and obtains the early warning information draft;

[0101] In the warning information formulation submodule, based on the fault location information, the template engine method is adopted, and the Python Jinja2 library is used to perform operations to define the warning information template, including the specific location of the fault, potential causes, and preliminary response measures. The Template.render method is called to fill the fault location information into the predetermined template, and the specific location of the fault is set as the location variable, the potential cause is set as the cause variable, and the preliminary response measures are set as the measures variable to ensure that the content of each warning information is specific and accurate, and obtain the draft of the warning information.

[0102] The warning information sending submodule sends warning information to maintenance personnel and management personnel through email or SMS based on the warning information draft, and generates warning information sending records;

[0103] In the early warning information sending submodule, based on the early warning information draft, the email sending method is adopted, the Python smtplib library is used to perform operations, the smtplib.SMTP function is called to connect to the mail server, the server address is set to smtp.example.com, the port is 465, the smtp.login method is used for identity authentication, the user name is user@example.com, the password is password, and the email content is created using email.mime.text.MIMEText. The Subject is set to the warning subject, From is the sender's email address, and To is the recipient's email address list. The smtp.sendmail method is called to send the email and generate a warning information sending record.

[0104] The early warning feedback collection submodule collects the confirmation reply, response time and preliminary processing opinions of the recipient based on the early warning information sending record, and integrates them to obtain early warning execution feedback;

[0105] In the early warning feedback collection submodule, based on the early warning information sending records, the data aggregation method is adopted, and the Python Pandas library is used to perform operations. The pandas.read_csv or pandas.read_excel function is called to read the feedback data according to the feedback form, including the recipient's confirmation reply, response time and preliminary processing opinions. The collected data is integrated using the DataFrame structure, and response is set as the confirmation reply, time as the response time, and comments as the preliminary processing opinions to ensure the completeness and accuracy of the early warning execution feedback and obtain the early warning execution feedback.

[0106] See also Figure 2 and Figure 8 ,The fault diagnosis and suggestion module includes a fault cause analysis submodule, a historical data comparison submodule, and a maintenance suggestion building submodule;

[0107] The fault cause analysis submodule conducts in-depth analysis of fault signals based on early warning execution feedback, identifies fault signal characteristics and fault causes, and obtains fault cause analysis results;

[0108] In the fault cause analysis submodule, based on the early warning execution feedback, a deep learning algorithm is adopted, and Python's TensorFlow and Keras libraries are used to perform operations. A convolutional neural network (CNN) model is built to extract and classify the features of fault signals. The model structure is defined, including adding multiple Conv1D and MaxPooling1D layers to adapt to time series data processing, and the Dense layer is used to classify the fault causes. The optimizer is set to Adam, and the loss function selects categorical_crossentropy. The model is trained to identify the characteristics of fault signals and fault causes, and the fault cause analysis results are obtained.

[0109] The historical data comparison submodule matches and compares the fault cause analysis results with the historical maintenance data and the fault database to find similar fault cases and obtain similar fault comparison results;

[0110] In the historical data comparison submodule, based on the fault cause analysis results, the similarity calculation method is adopted, and the Python SciPy library is used to perform operations. The scipy.spatial.distance.cosine function is called to calculate the cosine similarity between the fault feature vector and the historical maintenance data and the data in the fault library. Several cases with the highest similarity are selected as comparison objects, and the similarity threshold is set to 0.8 through screening to ensure that similar fault cases closest to the current fault are found and similar fault comparison results are obtained.

[0111] The maintenance suggestion building submodule proposes targeted maintenance or replacement suggestions based on similar fault comparison results, combined with equipment conditions and maintenance experience, and builds maintenance and preventive measures;

[0112] In the maintenance suggestion construction submodule, based on the results of similar fault comparison, the rule engine method is adopted, and Python is used to execute operations to define a set of fault handling rules, including matching maintenance or replacement strategies based on parameters such as fault type, equipment age and fault frequency. Combined with equipment conditions and maintenance experience, targeted maintenance or replacement suggestions are put forward, and a series of maintenance and preventive measures are constructed, including regular inspections and enhanced maintenance of specific components, to ensure that the suggestions are specific and practical, and to build maintenance and preventive measures.

[0113] 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. An intelligent electrical box monitoring system, characterized in that: The intelligent electrical box monitoring system comprises: The current fluctuation analysis module collects real-time current data of the electrical box, analyzes current fluctuations, identifies high-risk areas, and obtains current abnormality indicators; The micro-variable current monitoring module collects subtle current changes based on the current abnormality index, uses wavelet transform to subdivide the current signal, combines the random forest algorithm to classify the fault type, and obtains the micro-variable current characteristics; The electromagnetic field monitoring module utilizes the micro-variable current characteristics to monitor the electromagnetic field changes, collect electromagnetic field data, construct an electromagnetic field distribution map, and obtain electromagnetic field abnormality indicators; The abnormality location module calculates the fault point based on the electromagnetic field abnormality index, uses finite element analysis and Maxwell equations to locate the fault source and generate fault location information; The warning information issuing module sends warning information according to the fault location information and obtains warning execution feedback; The fault diagnosis and suggestion module combines the early warning execution feedback, analyzes the cause of the fault, combines historical maintenance data and the fault library, proposes repair or replacement suggestions, and builds maintenance and preventive measures; The electromagnetic field monitoring module comprises: The characteristic electromagnetic field acquisition submodule collects corresponding electromagnetic field change data based on the micro-variable current characteristics, reveals the real-time state of the electromagnetic field around the electrical box, and obtains a characteristic electromagnetic field data set; The electromagnetic field data analysis submodule analyzes the change trend and abnormal pattern of the electromagnetic field based on the characteristic electromagnetic field data set, identifies abnormal changes in the electromagnetic field by comparing the electromagnetic field data at different times and locations, and obtains electromagnetic field abnormality analysis results; The electromagnetic field distribution diagram construction submodule constructs a diagram showing the distribution of the electromagnetic field in the electrical box and its surroundings based on the electromagnetic field anomaly analysis result, and generates an electromagnetic field anomaly index; The abnormality positioning module includes: The electromagnetic field data processing submodule filters, denoises and normalizes the collected electromagnetic field data based on the electromagnetic field anomaly index to obtain the processed electromagnetic field data; The fault point calculation submodule calculates the potential fault point location based on the processed electromagnetic field data, using the finite element analysis method, combined with the distribution and intensity changes of the electromagnetic field, and generates the fault point location estimation result; The fault source location submodule uses Maxwell's equations to analyze the electromagnetic field around each estimated point based on the estimated fault point location, locates the exact location of the fault source, and obtains the fault location information; The finite element analysis method is according to the formula: ； Calculate the electromagnetic energy density distribution, where is the improved electromagnetic energy density, is the electric field strength, is the electric displacement density, is the magnetic field strength, is the magnetic flux density, is the temperature of the material, is the conductivity change, is the material density, is the vibration speed, , , , is the weight coefficient.

2. The intelligent electrical box monitoring system according to claim 1, characterized in that: The current anomaly indicators include current surge points, current reduction points and abnormal current duration; the micro-change current characteristics include small current fluctuation patterns, short-time current jump times and current micro-change periodic characteristics; the electromagnetic field anomaly indicators include abnormal electromagnetic wave amplitude, non-periodic electromagnetic wave changes and electromagnetic interference frequency distribution; the fault location information includes the circuit board number where the fault occurred, the sequence of faulty electrical components and the physical location of the fault; the early warning execution feedback includes the division of early warning levels, the allocation of early warning response personnel and the efficiency of early warning information transmission; the maintenance and preventive measures include fault cause analysis records, maintenance operation steps and recommendations for future preventive measures.

3. The intelligent electrical box monitoring system according to claim 1, characterized in that: The micro-variable current monitoring module comprises: The current change acquisition submodule acquires subtle changes in the current based on the current abnormality index, including recording the current value every millisecond and frequency analysis of current fluctuations, to obtain a subtle current data set; The signal subdivision processing submodule uses wavelet transform to subdivide the current signal based on the subtle current data set, decomposes the current signal into multiple frequency bands, identifies subtle changes and abnormal fluctuations in the current, and obtains the current signal subdivision result; The fault classification analysis submodule performs classification analysis of fault types based on the current signal segmentation results and in combination with the random forest algorithm. By comparing the current change characteristics under different fault types, the potential fault type corresponding to the current current change is identified and the slight change current characteristics are generated.

4. The intelligent electrical box monitoring system according to claim 3 is characterized in that: The random forest algorithm is based on the formula: ； Compute the improved feature importance score where, Score the improved feature importance, is the set of all decision trees that make up the random forest, For Trees A split node in For splitting nodes Because of the characteristics The reduction in impurity due to the use of Features The frequency of occurrence, Features Correlation with fault category, Features The coefficient of variation, , , is the weight coefficient.

5. The intelligent electrical box monitoring system according to claim 1, characterized in that: The fault diagnosis and suggestion module includes: The fault cause analysis submodule performs in-depth analysis on the fault signal based on the early warning execution feedback, identifies the fault signal characteristics and fault cause, and obtains the fault cause analysis result; The historical data comparison submodule matches and compares the fault cause analysis result with the historical maintenance data and the fault library to find similar fault cases and obtain similar fault comparison results; The maintenance suggestion building submodule proposes targeted maintenance or replacement suggestions based on the similar fault comparison results, combined with equipment conditions and maintenance experience, and builds maintenance and preventive measures.

Images