Gas monitoring fault analysis method, system and storage medium

Abstract

This application relates to the field of data processing technology and discloses a gas monitoring fault analysis method, system, and storage medium. The method includes: collecting multidimensional data through a sensor group and transmitting it to a processing unit via an Internet of Things protocol for pre-screening; filtering, feature calculation, and dimensionality reduction processing of the data to obtain feature vectors; establishing a fault database and generating mapping relationships; fusing time series and spatial data to calculate fault judgment values; combining sensor status data evaluation to obtain early warning results; retrieving historical data to calculate similarity and generate action instructions. This application improves the accuracy of gas pipeline network fault diagnosis through the fusion analysis of multi-source data, achieving early fault identification and early warning.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a gas monitoring fault analysis method, system and storage medium. Background Art

[0002] Currently, gas pipeline network monitoring generally uses a single sensor or a simple combination of multiple sensors for data collection, focusing primarily on basic parameters such as gas concentration and pressure. These monitoring systems typically employ fixed threshold judgment methods for fault detection, triggering alarms when monitored parameters exceed preset thresholds. In terms of data processing, most employ simple filtering and statistical analysis methods for basic processing and storage of collected data. Furthermore, existing fault analysis methods primarily rely on manual experience, comparing historical data with expert knowledge for fault diagnosis and early warning. The development of response plans also relies primarily on manual judgment.

[0003] However, the existing gas monitoring fault analysis methods have the following shortcomings: a single data collection method is difficult to fully reflect the operating status of the gas pipeline network; the fixed threshold judgment method is prone to false alarms and missed alarms; simple data processing methods cannot fully mine the fault feature information contained in the data; it is highly dependent on manual experience, making it difficult to ensure the objectivity and accuracy of the analysis results; there is a lack of predictive ability for fault development trends, making it difficult to achieve early warning; the formulation of disposal plans lacks systematicity and pertinence, affecting the efficiency and effectiveness of fault disposal. Summary of the Invention

[0004] The present application provides a gas monitoring fault analysis method, system and storage medium for improving the accuracy of gas pipeline network fault diagnosis through fusion analysis of multi-source data, and realizing early identification and warning of faults.

[0005] In a first aspect, the present application provides a gas monitoring fault analysis method, the gas monitoring fault analysis method comprising:

[0006] The gas monitoring sensor group collects gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline vibration data. The data is transmitted to the processing unit according to the Internet of Things communication protocol, and the data is pre-screened to obtain the original gas monitoring data.

[0007] The raw gas monitoring data is calculated using a median filter formula and outliers are removed. Feature parameters are calculated for the filtered data, and dimensionality reduction is performed on the calculation results to obtain a gas feature vector.

[0008] Establishing a fault type database based on the gas characteristic vector, classifying and calculating the characteristic data, comparing the calculation results with preset parameters, and generating a gas fault mapping relationship;

[0009] Extracting time series data based on the gas feature vector and gas fault mapping relationship, performing weighted fusion calculation on the time series data and spatial data, and outputting a gas fault judgment value;

[0010] Compare the gas fault judgment value with the preset threshold value, perform weighted calculation on the classification result, and perform credibility evaluation in combination with the sensor status data to obtain a gas fault warning result;

[0011] Based on the gas fault warning result, historical data is retrieved, similarity calculation and priority sorting are performed on the retrieved data, and a gas fault handling instruction is generated.

[0012] In a second aspect, the present application provides a gas monitoring fault analysis system, the gas monitoring fault analysis system comprising:

[0013] The transmission module is used to collect gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline vibration data through the gas monitoring sensor group, transmit the data to the processing unit according to the Internet of Things communication protocol, pre-screen the data, and obtain the raw gas monitoring data;

[0014] A dimensionality reduction module is used to calculate the raw gas monitoring data according to the median filter formula and remove outliers, calculate characteristic parameters for the filtered data, and perform dimensionality reduction processing on the calculation results to obtain a gas characteristic vector;

[0015] A comparison module is used to establish a fault type database based on the gas characteristic vector, perform classification calculations on the characteristic data, compare the calculation results with preset parameters, and generate a gas fault mapping relationship;

[0016] a fusion module, configured to extract time series data according to the gas feature vector and the gas fault mapping relationship, perform weighted fusion calculation on the time series data and the spatial data, and output a gas fault judgment value;

[0017] An evaluation module is used to compare and classify the gas fault judgment value with a preset threshold, perform weighted calculation on the classification results, and perform credibility evaluation in combination with the sensor status data to obtain a gas fault warning result;

[0018] The sorting module is used to retrieve historical data based on the gas fault warning result, perform similarity calculation and priority sorting on the retrieved data, and generate a gas fault handling instruction.

[0019] A third aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned gas monitoring fault analysis method.

[0020] In the technical solution provided by this application, the gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data and pipeline vibration data are collected simultaneously by the gas monitoring sensor group, thereby realizing the coordinated monitoring of multi-dimensional data and comprehensively reflecting the operating status of the gas pipeline network. In addition, the Internet of Things communication protocol is used for data transmission to ensure the reliability and real-time performance of data transmission. The pre-screening process effectively reduces the influence of data noise. The median filter formula calculation can effectively remove outliers and improve data quality. The characteristic parameter calculation and dimensionality reduction processing realize the effective compression and feature extraction of data. The establishment of a fault type database enables the fault characteristics to be systematically organized and managed, and the classification calculation of characteristic data The comparison analysis improves the accuracy of fault identification. The weighted fusion calculation of time series data and spatial data fully utilizes the spatiotemporal correlation of fault characteristics, making fault judgment more comprehensive and accurate. The hierarchical comparison and weighted calculation of preset thresholds realize the accurate division of fault levels. The credibility assessment combined with sensor status data ensures the reliability of the early warning results. By retrieving historical data and performing similarity calculation, the most similar historical cases are found, providing a reliable reference basis for fault handling. Priority sorting ensures the optimal selection of handling solutions, thus forming a complete, automated and intelligent gas monitoring fault analysis process, which significantly improves the accuracy of gas pipeline fault diagnosis and the efficiency of handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 This is a schematic diagram of an embodiment of a gas monitoring fault analysis method in an embodiment of the present application;

[0023] Figure 2 This is a schematic diagram of an embodiment of a gas monitoring fault analysis system in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The embodiments of the present application provide a gas monitoring fault analysis method, system and storage medium. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0025] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In one embodiment of the gas monitoring fault analysis method of the present application, the method includes:

[0026] Step S101: Collect gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline vibration data through a gas monitoring sensor group, transmit the data to a processing unit according to an Internet of Things communication protocol, pre-screen the data, and obtain raw gas monitoring data;

[0027] Step S102: Calculate the raw gas monitoring data using the median filter formula and remove outliers. Calculate characteristic parameters for the filtered data and perform dimensionality reduction on the calculation results to obtain a gas characteristic vector.

[0028] Step S103: Establish a fault type database based on the gas characteristic vector, perform classification calculation on the characteristic data, compare the calculation results with preset parameters, and generate a gas fault mapping relationship;

[0029] Step S104: extracting time series data based on the gas feature vector and the gas fault mapping relationship, performing weighted fusion calculation on the time series data and the spatial data, and outputting a gas fault judgment value;

[0030] Step S105: Compare the gas fault judgment value with the preset threshold value and perform classification, perform weighted calculation on the classification result, and perform credibility assessment in combination with the sensor status data to obtain a gas fault warning result;

[0031] Step S106: Retrieve historical data based on the gas fault warning result, perform similarity calculation and priority sorting on the retrieved data, and generate a gas fault handling instruction.

[0032] It is understandable that the execution subject of this application can be a gas monitoring fault analysis system, or a terminal or a server, which is not limited here. The embodiment of this application is described by taking the server as the execution subject as an example.

[0033] Specifically, the gas monitoring sensor group includes a gas concentration sensor, a pressure sensor, a temperature sensor, an acoustic sensor, and a fiber optic vibration sensor, each used to collect different types of monitoring data. The gas concentration sensor uses a metal oxide semiconductor sensor to measure gas concentration by detecting changes in electrical conductivity caused by redox reactions between gas molecules and the sensor surface. The sampling frequency is 100ms. The pressure sensor uses a piezoresistive sensor, measuring pipeline pressure by using changes in resistance under pressure. The sampling frequency is 200ms. The temperature sensor uses a thermocouple to measure gas temperature using thermoelectromotive force. The sampling frequency is 500ms. The acoustic sensor uses a piezoelectric sensor to convert acoustic vibrations into electrical signals to measure the acoustic characteristics of leaks. The sampling frequency is 50ms. The fiber optic vibration sensor uses the fiber Bragg grating principle to measure pipeline vibration characteristics. The sampling frequency is 20ms. The collected data is transmitted using the MQTT IoT communication protocol, which uses a publish / subscribe model and features low bandwidth usage and high reliability. After the data is transmitted to the processing unit, it undergoes pre-screening. We set thresholds for gas concentration data within a range of 0-100ppm, pressure data within a range of 0-1.6MPa, temperature data within a range of -20°C-60°C, acoustic data within a range of 30-120dB, and vibration data within a range of 0-5mm. We use range checks to eliminate abnormal data, time-align the data, construct data packets in 1s units, and add check bits to ensure data integrity.

[0034] The pre-screened raw gas monitoring data is subjected to median filtering. A sliding time window is used for each type of data to calculate the median and interquartile range of the data within the window. Outliers are identified and removed based on an interval three times the interquartile range. Feature parameter calculations are performed on the filtered data, including: calculating the mean, standard deviation, kurtosis, and skewness for gas concentration data; calculating the maximum fluctuation amplitude and fluctuation period for pressure data; calculating the rate of change and temperature gradient for temperature data; calculating the sound pressure level and frequency distribution for acoustic data; and calculating the vibration amplitude and vibration frequency for vibration data. The calculation results are mapped to a unified scale through data normalization, and then reduced in dimension through principal component analysis. Feature dimensions with a cumulative variance contribution rate of 95% are retained. Finally, a linear combination is performed based on feature importance weights to generate a gas feature vector.

[0035] A fault type database is constructed based on the generated gas feature vectors. First, the feature vectors are segmented into 24-hour time windows, and statistical features within each window are calculated. Feature data is grouped by fault type, and the mean center and variance distribution of each group are calculated. Segmented statistics are performed on the feature data, and the probability distribution characteristics of different fault types along each feature dimension are calculated. The calculated results are compared with historical fault case data, and feature similarity coefficients are calculated to establish a mapping relationship between features and fault types. Time-series-spatial data fusion analysis is performed based on the gas feature vectors and fault mapping relationships. Feature vectors are segmented into three time scales: hourly, daily, and weekly. Trends are calculated for each time scale, and time-series features are extracted. A spatial correlation matrix is ​​constructed using sensor location information to analyze the spatial propagation characteristics of fault features. After standardizing the temporal and spatial features, weight coefficients for the two feature types are calculated, and weighted fusion is performed to output a fault judgment value in the range of 0-1.

[0036] The fault judgment value is compared with the four preset threshold nodes of 0.3, 0.5, 0.7, and 0.9 for grading. The frequency and duration of occurrence are calculated for the grading results, and weighted with weights of 0.4 and 0.6 respectively. Reliability is assessed by combining the three indicators of signal strength, response time, and data integrity rate in the sensor status data. The weighted results are combined with the reliability score to calculate the confidence level of the fault warning and form a graded warning result. Finally, historical data is retrieved based on the warning results. Features such as fault type, degree, and location are extracted to establish retrieval conditions and match them with the historical case library. The Euclidean distance between the feature parameters of the matched historical cases is calculated to evaluate the case similarity. Cases with a similarity greater than 0.8 are selected and their treatment plans are extracted. The plans are scored and ranked according to the three indicators of treatment time, resource consumption, and repair effect, and the optimal plan is selected as the treatment instruction output.

[0037] For example, a gas pipeline monitoring point detected a gas concentration increase from 5 ppm to 45 ppm within 10 minutes, a pressure drop from 1.2 MPa to 0.9 MPa, and a temperature increase of 2°C. Simultaneously, an acoustic sensor detected a continuous noise level of 50-60 dB, and a vibration sensor showed an amplitude of 0.8 mm. After pre-screening and filtering, the extracted characteristic parameters revealed typical trace leak characteristics. Time-series and spatial analysis revealed minor anomalies at adjacent monitoring points, with a fault determination value of 0.65. Combined with sensor status assessment, this resulted in a medium-level warning. The system matched three similar historical cases, and the solution with the shortest response time and lowest resource consumption was selected as the optimal action.

[0038] In the embodiment of the present application, the gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data and pipeline vibration data are collected simultaneously by the gas monitoring sensor group, thereby realizing the coordinated monitoring of multi-dimensional data and comprehensively reflecting the operating status of the gas pipeline network. In addition, the Internet of Things communication protocol is used for data transmission to ensure the reliability and real-time performance of data transmission. The pre-screening process effectively reduces the influence of data noise. The median filter formula calculation can effectively remove outliers and improve data quality. The characteristic parameter calculation and dimensionality reduction processing realize the effective compression and feature extraction of data. The establishment of a fault type database enables the fault characteristics to be systematically organized and managed, and the classification calculation and comparison of characteristic data are carried out. The analysis improves the accuracy of fault identification. The weighted fusion calculation of time series data and spatial data fully utilizes the spatiotemporal correlation of fault characteristics, making fault judgment more comprehensive and accurate. The hierarchical comparison and weighted calculation of preset thresholds realize the precise division of fault levels. The credibility assessment combined with sensor status data ensures the reliability of the early warning results. By retrieving historical data and performing similarity calculations, the most similar historical cases are found, providing a reliable reference basis for fault handling. Priority sorting ensures the optimal selection of handling solutions, thus forming a complete, automated, and intelligent gas monitoring fault analysis process, which significantly improves the accuracy of gas pipeline fault diagnosis and the efficiency of handling.

[0039] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0040] (1) Gas concentration data is collected at a sampling interval of 100ms, pipeline pressure data is collected at a sampling interval of 200ms, gas temperature data is collected at a sampling interval of 500ms, leakage acoustic data is collected at a sampling interval of 50ms, and pipeline network vibration data is collected at a sampling interval of 20ms;

[0041] (2) Align the timestamps of gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline network vibration data, construct data packets in units of 1 second, and add check bits to the data packets according to the Internet of Things communication protocol;

[0042] (3) Pre-screen the numerical range of the gas concentration data in the data packet by setting the concentration range threshold of 0-100ppm, the pipeline pressure data by setting the pressure range threshold of 0-1.6MPa, the gas temperature data by setting the temperature range threshold of -20℃-60℃, the leakage acoustic data by setting the sound pressure range threshold of 30-120dB, and the pipeline network vibration data by setting the amplitude range threshold of 0-5mm;

[0043] (4) The pre-screened gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline network vibration data are merged into data records, and each data record is marked with the acquisition time and sensor number;

[0044] (5) Perform data integrity checks on data records, remove records with missing data, mark abnormal data records, and establish data quality scores;

[0045] (6) Organize the data records that have passed the integrity check into a time series data table, add data source identification, and generate gas monitoring raw data.

[0046] Specifically, gas concentration data is collected using a metal oxide semiconductor sensor. This sensor operates by detecting changes in electrical conductivity caused by the redox reaction between gas molecules and the surface of a semiconductor material. The sampling interval is set to 100ms, enabling timely capture of even small changes in gas concentration. Pipeline pressure data is collected using a piezoresistive pressure sensor, which measures pressure by measuring the resistance change of a strain gauge under pressure. The sampling interval is 200ms, meeting the requirements for monitoring pipeline pressure changes. Gas temperature data is collected using a K-type thermocouple, which measures temperature using the thermoelectromotive force generated by temperature differences. The sampling interval is 500ms, taking into account the relatively slow nature of temperature changes. Leakage acoustic data is collected using a piezoelectric acoustic sensor, which converts acoustic vibrations into electrical signals. The sampling interval is 50ms, ensuring the integrity of high-frequency acoustic signals. Pipeline vibration data is collected using a fiber Bragg grating sensor, which measures vibration characteristics based on the periodic changes of the fiber Bragg grating. The sampling interval is 20ms, meeting the requirements for detecting even small vibrations. Time alignment is performed on the different types of collected data. Due to the different sampling intervals of the various sensors, the data needs to be aligned to the same time base. Timestamp alignment uses interpolation. For data points sampled with an interval greater than 1 second, linear interpolation is used to supplement the intermediate value. For data points sampled with an interval less than 1 second, the average value within the 1 second is taken. Data packets are constructed using the MQTT (Message Queuing Telemetry Transport) protocol, a lightweight message transmission protocol that uses a publish / subscribe model. Each packet contains all sensor data within 1 second and includes a CRC (Cyclic Redundancy Check) checksum for data transmission error detection.

[0047] During the data pre-screening phase, reasonable thresholds are set for various data types. The gas concentration threshold range of 0-100 ppm is based on common combustible gas leak concentration limits. The pipeline pressure threshold range of 0-1.6 MPa takes into account the typical operating pressure of urban gas pipeline networks. The gas temperature threshold range of -20°C to 60°C covers the temperature range under different climate conditions. The leakage acoustics threshold range of 30-120 dB encompasses the sound pressure level range from minor leaks to major leaks. The pipeline vibration threshold range of 0-5 mm is based on the safety limits of the pipeline structure. Data outside the threshold range is marked as abnormal and receives special processing. The data record merging process integrates the pre-screened multidimensional data into a unified data structure. Each data record contains a complete set of sensor data, along with a timestamp and sensor identification information. The timestamp is accurate to the millisecond and records the specific moment of data acquisition. Sensor numbers use unique identification codes that include sensor type and installation location information to facilitate subsequent data traceability and analysis.

[0048] Data integrity checks ensure the quality of data records. First, data records are checked for missing values. If a certain type of sensor data is completely missing, the record is discarded. For records with partial data missing, the missing items are marked and the reasons for the missing items are documented. Abnormal data records are marked based on multiple dimensions, including numerical anomalies (exceeding the threshold range), change anomalies (sudden changes or excessive fluctuations), and association anomalies (inconsistency with other sensor data). Data quality scoring is weighted, considering data completeness, accuracy, and timeliness. The scoring results are used in subsequent data credibility analysis.

[0049] The integrity-checked data records are organized into a standardized time-series data table. Each row in the data table represents a complete monitoring record at a specific point in time, with columns including timestamps, various sensor data types, and data quality scores. The data source identifies the data collection node and transmission path for data traceability. The organized data table is the raw gas monitoring data, serving as the foundation for subsequent fault analysis.

[0050] For example, during continuous monitoring, a gas concentration sensor records data every 100ms, generating 100 data points over 10 seconds, showing a gradual increase in concentration from 3ppm to 12ppm. Over the same period, a pressure sensor records data every 200ms, generating 50 data points, showing a decrease in pressure from 1.2MPa to 1.1MPa. A temperature sensor records data every 500ms, generating 20 data points, showing a constant temperature fluctuation around 28°C. An acoustic sensor records data every 50ms, generating 200 data points, detecting a sustained sound pressure of 40-45dB. A vibration sensor records data every 20ms, generating 500 data points, showing minute vibrations of 0.2-0.3mm. These data are time-aligned and consolidated into 10 complete data records in 1-second increments. Data pre-screening shows that all data are within the set thresholds, and data integrity checks reveal no missing or abnormal records. The generated time series data table fully records the evolution of all sensor data over a 10-second period, and the data quality score indicates good data reliability. This set of data reflects a typical early leakage scenario, with a slow increase in gas concentration accompanied by a slight drop in pressure, and accompanied by weak acoustic and vibration signals.

[0051] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0052] (1) The raw gas monitoring data were grouped into 10-minute time windows, and the median and interquartile range of each group of data were calculated. Outliers were identified and eliminated based on the interval of 3 times the interquartile range to obtain filtered data.

[0053] (2) Calculate the mean, standard deviation, kurtosis, and skewness of the gas concentration data in the filtered data; calculate the maximum fluctuation amplitude and fluctuation period of the pipeline pressure data; calculate the rate of change and temperature gradient of the gas temperature data; calculate the sound pressure level and frequency distribution of the leakage acoustic data; and calculate the vibration amplitude and vibration frequency of the pipeline network vibration data to generate a feature data set;

[0054] (3) normalizing the mean, standard deviation, kurtosis, and skewness data in the feature data group to the range of 0-1, normalizing the maximum fluctuation amplitude and fluctuation period data to the range of -1-1, linearly transforming the rate of change and temperature gradient data to the standard normal distribution, logarithmically transforming the sound pressure level and frequency distribution data, and performing minimum-maximum transformation on the vibration amplitude and vibration frequency data to obtain normalized feature data;

[0055] (4) Sort the normalized feature data according to the variance contribution rate, select the feature dimensions whose cumulative variance contribution rate reaches 95% to form the main feature data set;

[0056] (5) Calculate the feature importance of each feature data in the main feature data set according to the correlation coefficient matrix, perform linear combination of the feature data based on the importance weight, and construct compressed feature data;

[0057] (6) Convert the compressed feature data into a numerical vector form, add the time index and sensor location identifier, and generate a gas feature vector.

[0058] Specifically, in the data filtering and feature extraction stage, the raw gas monitoring data is first grouped into time windows. 10 minutes is used as a fixed time window to segment the data. In each time window, the median value and interquartile range of the data are calculated. The median is calculated by sorting the data in the window by size and taking the value in the middle position. The interquartile range is obtained by calculating the difference between the upper quartile (75th percentile) and the lower quartile (25th percentile). The outlier judgment limit is set based on 3 times the interval of the interquartile range, that is, data points outside the range of (Q1-3IQR) to (Q3+3IQR) are judged as outliers and eliminated, where Q1 is the lower quartile, Q3 is the upper quartile, and IQR is the interquartile range.

[0059] Multi-dimensional feature extraction is performed on the filtered data. For gas concentration data, mathematical statistical features are calculated, including the mean (reflecting the concentration level), standard deviation (reflecting the degree of fluctuation), kurtosis (reflecting the degree of distribution sharpness), and skewness (reflecting the degree of deviation from symmetry in the distribution). For pipeline pressure data, dynamic characteristics are analyzed, with the maximum fluctuation amplitude (the difference between the highest and lowest pressures) and the fluctuation period (the time interval between pressure fluctuations) calculated. Feature extraction for gas temperature data focuses on changing trends, calculating the temperature change rate (the amount of temperature change per unit time) and the temperature gradient (the temperature change at a spatial location). Analysis of leak acoustic data includes calculations of the sound pressure level (reflecting the intensity of the sound wave) and frequency distribution (reflecting the spectral characteristics of the sound wave). For pipeline vibration data, vibration amplitude (maximum value of the vibration displacement) and vibration frequency (the inverse of the vibration period) are extracted. To ensure comparability of feature data of different dimensions and ranges, feature normalization is required. Statistical quantities such as the mean, standard deviation, kurtosis, and skewness are normalized to the 0-1 range, converting the original values ​​to the [0,1] range through linear mapping. The maximum fluctuation amplitude and fluctuation period characteristics were normalized to the [-1, 1] interval, suitable for describing physical quantities that vary in both directions. Temperature change rate and temperature gradient data were mapped to data conforming to a standard normal distribution through linear transformation, facilitating subsequent statistical analysis. Due to the large span of sound pressure level and frequency distribution data, a logarithmic transformation was used to compress the data range. Vibration amplitude and frequency data were normalized using the minimum-maximum method to maintain relative proportionality.

[0060] The normalized feature data is subjected to dimensionality reduction, mainly using the principal component analysis (PCA) method. The features are ranked by calculating their variance contribution rate, and the variance contribution rate calculation formula is:

[0061]

[0062] Among them: VCR i is the variance contribution rate of the i-th feature; i is the eigenvalue of the i-th feature; n is the total number of features; k is the feature number.

[0063] Feature dimensions with a cumulative variance contribution rate of 95% are retained to form the primary feature dataset. During the selection of the primary feature dataset, feature importance is assessed by calculating the correlation coefficient matrix between features. Each element of the correlation coefficient matrix represents the degree of correlation between two features, and the importance weight of each feature is calculated based on this matrix. The importance weight calculation takes into account the correlation between the feature and other features as well as the variance contribution of the feature itself. The obtained feature importance weights are normalized so that the sum of the weights is 1. These weights are then applied to the linear combination of the feature data to construct the compressed feature data.

[0064] The compressed feature data must ultimately be converted to a standard numeric vector format. Each vector contains the reduced and weighted feature values, along with a time index (accurate to the second) and sensor location identifier (including installation point coordinates). This vector format facilitates subsequent fault mapping and diagnostic analysis.

[0065] For example, at a gas pipeline monitoring point, 10 minutes of monitoring data is processed. The raw data consists of 600 sampling points (sampling frequency 1 Hz). First, outliers are identified by calculating the median and interquartile range. For gas concentration data, the median is calculated to be 15 ppm, the upper quartile is 20 ppm, the lower quartile is 10 ppm, and the interquartile range is 10 ppm. Based on this, the outlier identification interval is determined to be [-20 ppm, 50 ppm]. Data points outside this range are removed.

[0066] Feature extraction was performed on the filtered data. The gas concentration data was calculated to have a mean of 16 ppm, a standard deviation of 5 ppm, a kurtosis of 2.1, and a skewness of 0.3. The pipeline pressure data showed a maximum fluctuation amplitude of 0.2 MPa, with a fluctuation period of approximately 120 seconds. The temperature data showed a rate of change of 0.1°C / min, with a temperature gradient of 0.05°C / m. The acoustic data had a sound pressure level of 45 dB, with frequencies primarily distributed in the 200-500 Hz range. The vibration data had an amplitude of 0.3 mm and a dominant frequency of 2 Hz. These features were normalized to a uniform numerical scale. The statistical features of gas concentration were mapped to the [0, 1] interval, the pressure fluctuation features to the [-1, 1] interval, and the temperature features were linearly adjusted to a distribution with a mean of 0 and a standard deviation of 1. The acoustic features were logarithmically transformed to compress the data range, and the vibration features were transformed using a minimum-maximum transformation to maintain relative proportions.

[0067] Variance contribution analysis revealed that, among all features, the cumulative variance contribution of four features—gas concentration mean, pressure fluctuation amplitude, sound pressure level, and vibration frequency—reached 95%. Therefore, these features were retained as primary features. Further correlation matrix analysis revealed a strong negative correlation between gas concentration and pressure fluctuation, while a weak positive correlation between sound pressure level and vibration frequency. Based on this, the importance weights of the four features were calculated to be 0.35, 0.3, 0.2, and 0.15, respectively. These features were linearly combined according to their weights and converted into a four-dimensional feature vector. The timestamp (e.g., 2024-01-01 10:00:00) and sensor location identifier (e.g., X = 100 m, Y = 50 m, Z = 2 m) were added to form the gas feature vector. This method preserves the most important information in the original data while significantly reducing the data dimensionality.

[0068] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0069] (1) The gas feature vector is divided into 24-hour data segments according to the timestamp, and the statistical features of the gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data and pipeline network vibration data in each data segment are calculated to generate a feature statistics table;

[0070] (2) Group the data in the feature statistics table according to the fault type label, calculate the mean center and variance distribution of each group of data, and form a fault feature statistics matrix;

[0071] (3) The data in the fault feature statistical matrix are segmented according to the gas concentration range of 0-100 ppm, the pressure fluctuation range of 0-1.6 MPa, the temperature change range of -20°C-60°C, the sound pressure level of 30-120 dB, and the vibration amplitude of 0-5 mm, and segmented statistical data are constructed;

[0072] (4) Calculate the probability density of the segmented statistical data to obtain the characteristic distribution curve of each fault type and generate a fault characteristic probability table;

[0073] (5) Compare the fault feature probability table with the historical fault data, calculate the feature similarity coefficient, and establish a fault type database;

[0074] (6) Based on the fault type database, the input gas feature vector is mapped, the fault type probability value is calculated, and the gas fault mapping relationship is generated.

[0075] Specifically, the gas feature vectors are organized into 24-hour data segments according to the time series. This time scale is chosen based on the diurnal variations in gas usage characteristics. Within each 24-hour data segment, the gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline network vibration data are processed separately. For the gas concentration data, the 24-hour average, maximum, minimum, and trend are calculated; for the pipeline pressure data, the pressure fluctuation range and frequency are analyzed; for the gas temperature data, the temperature variation amplitude and rate are calculated; for the leakage acoustic data, the acoustic pressure characteristics and spectral features are extracted; and for the pipeline network vibration data, the vibration intensity and frequency characteristics are recorded. These statistical features form the basis of the feature statistics table. Based on the feature statistics table, the data is grouped and labeled according to different fault types based on historical experience and expert knowledge. Fault types include pipeline leaks (micro, medium, and large), pressure anomalies (overpressure, underpressure, and pressure fluctuations), temperature anomalies (overtemperature, undertemperature), pipeline blockage, and pipeline corrosion. The mean center is calculated for each set of data to reflect the typical characteristic value of this type of fault; the variance distribution is calculated to reflect the fluctuation range of the characteristic. These statistics constitute the fault characteristic statistical matrix. The data in the fault characteristic statistical matrix needs to be segmented according to the reasonable range of physical quantities. The gas concentration range of 0-100ppm is divided into 5 intervals, corresponding to different degrees of leakage; the pressure fluctuation range of 0-1.6MPa is divided into 4 intervals, corresponding to different pressure anomalies; the temperature variation range of -20℃-60℃ is divided into 6 intervals, reflecting different temperature anomalies; the sound pressure level of 30-120dB is divided into 5 intervals, corresponding to different leakage acoustic characteristics; the vibration amplitude of 0-5mm is divided into 4 intervals, reflecting different mechanical fault conditions.

[0076] The probability density function is calculated for the segmented statistical data. The calculation formula of the probability density function is:

[0077]

[0078] Where: P(x) is the probability density of eigenvalue x; M is the number of mixed distributions; ω i is the weight coefficient of the i-th distribution; μi is the mean of the i-th distribution; σ i is the standard deviation of the i-th distribution; N is the number of constraints; is the jth constraint function.

[0079] Based on the probability density calculation results, a characteristic distribution curve is established for each fault type, forming a fault characteristic probability table. This probability table records the distribution characteristics of different fault types along various characteristic dimensions, including information such as the center location, distribution shape, and distribution range. This probability table is compared with historical fault data to calculate a similarity coefficient. The similarity calculation considers differences in characteristic values, differences in distribution patterns, and correlations between features. The similarity coefficient serves as a quantitative indicator of the degree of similarity between a new fault and historical faults. A fault type database is established based on the fault characteristic probability table and similarity analysis results. Each record in the database contains information such as the fault type identifier, characteristic parameter range, probability distribution characteristics, and similar fault cases. The database adopts a graph database structure, which facilitates the establishment of correlations between features and records feature evolution patterns. When a new gas feature vector is input, it is matched with existing records in the database through data mapping, and the probability values ​​of various fault types are calculated to generate a gas fault mapping relationship.

[0080] For example, consider analyzing data from a gas pipeline monitoring point. First, the 24-hour continuous monitoring data was organized into time series, generating data segments covering five dimensions: gas concentration, pressure, temperature, acoustics, and vibration. Statistical analysis revealed that gas concentration showed a slow upward trend from 3 ppm to 15 ppm between 8:00 and 10:00 AM. During the same period, pressure data showed a slight decrease, from 1.2 MPa to 1.1 MPa. Temperature remained stable around 28°C. Acoustic data fluctuated between 40 and 45 dB. Vibration data exhibited small amplitudes of 0.2 to 0.3 mm. These characteristic data were compared with records in a fault type database. In the fault feature statistical matrix, the characteristic distribution of this data set fell within the characteristic interval of the micro-leak fault type. Specifically, the gas concentration change amplitude and rate of change closely matched the characteristic distribution of micro-leak faults. The pressure drop was small but consistent with a leak. The acoustic and vibration characteristics also displayed typical micro-leak characteristics. Probability density calculations determined that the probability of this data set belonging to the micro-leak fault type was 0.85, significantly higher than the probability values ​​for other fault types.

[0081] When establishing the mapping relationship, the feature vector is associated with trace leak cases in the historical database, and the evolution of its features is recorded. This association includes both numerical similarities and similarities in temporal trends. Examples include a slow increase in gas concentration, a slight drop in pressure, and specific frequency characteristics of acoustic and vibration signals. This not only reflects the current fault diagnosis results but also captures the dynamic characteristics of fault development. This dynamic mapping relationship is crucial for the timely detection and resolution of various faults in the gas pipeline network. The mapping process fully considers the correlation and temporal evolution of multi-dimensional feature data, ensuring the accuracy and reliability of fault diagnosis.

[0082] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0083] (1) Divide the gas characteristic vector into data segments according to the time scales of hour, day, and week, calculate the change trend of the data at each time scale, and generate time series change data;

[0084] (2) Extract the time association rules of fault characteristics from the gas fault mapping relationship, perform time correlation analysis on the time series change data, and obtain time series feature data;

[0085] (3) Convert the sensor location information in the gas feature vector into coordinate data, calculate the distance matrix between adjacent sensors, and construct a spatial association table;

[0086] (4) Perform spatial correlation analysis on the sensor data in the spatial association table, calculate the propagation characteristics of the fault characteristics in space, and form spatial feature data;

[0087] (5) Perform numerical standardization on the temporal feature data and the spatial feature data, calculate the weight coefficients of the two types of feature data, and generate feature fusion data;

[0088] (6) Numerical calculation is performed on the feature fusion data and the gas fault mapping relationship, the fault degree is quantified according to the 0-1 interval, and the gas fault judgment value is output.

[0089] Specifically, the gas feature vectors are decomposed at multiple time scales, with the data divided into three time scales: hourly (reflecting short-term changes), daily (reflecting daily patterns), and weekly (reflecting long-term trends). Trends are calculated for each time scale. These include: calculating the rate of change of gas concentration, pressure, and temperature, and the frequency changes of acoustic and vibration characteristics at the hourly scale; analyzing the intraday fluctuation patterns of each parameter at the daily scale, including peak, plateau, and trough characteristics; and analyzing long-term trends at the weekly scale to identify periodic fluctuations and cumulative effects. Temporal association rules are extracted from the gas fault mapping relationships generated for the time series data. These association rules encompass the temporal patterns of fault occurrence, the evolution of fault development, and the temporal dependencies between faults. Temporal correlation analysis of the time series data focuses on the following aspects: the order of parameter changes (e.g., whether a pressure drop precedes a concentration increase), the duration of the change (the duration of the abnormal state), the periodicity of the change (whether there are regular fluctuations), and the temporal coupling relationships between multiple parameters.

[0090] In the spatial dimension, the sensor location information in the gas feature vector is first converted into three-dimensional coordinate data. The coordinate conversion takes into account the spatial layout of the pipeline network, including factors such as the pipeline's direction, burial depth, and branching structure. Based on the coordinate data, the Euclidean distance between adjacent sensors is calculated to form a distance matrix. The distance matrix is ​​used to characterize the spatial relationship between sensor nodes and construct a spatial association table. Spatial correlation analysis is performed on the sensor data in the spatial association table. The analysis includes: the propagation direction of the fault characteristics in space (the diffusion path along the pipeline network), the propagation speed (the spatial gradient of the characteristic change), and the impact range (the spatial distribution range of the abnormal characteristics). Based on the analysis results, spatial feature data is formed to describe the distribution and evolution characteristics of the fault in the spatial dimension.

[0091] Time series and spatial feature data require standardization before fusion. This standardization process maps data of varying dimensions to a unified numerical range, ensuring data comparability. The weighting coefficients for the two types of feature data are calculated based on their reliability, representativeness, and contribution to fault diagnosis. This weighted fusion generates fused feature data that comprehensively reflects the temporal and spatial characteristics of the fault.

[0092] Finally, the feature fusion data and the gas fault mapping relationship are numerically calculated, and the fault degree quantification formula is:

[0093]

[0094] Where: F(d) is the fault severity value (0-1 interval); T is the time scale number; S is the space scale number; α t is the time feature weight; β s is the spatial feature weight; d is the feature fusion data; γts is the standard value of spatiotemporal characteristics; η ts is the characteristic fluctuation range; R is the number of constraints; θ r (d) is the constraint function.

[0095] For example, a gas pipeline network experiences a leak. Initially, on an hourly scale, the gas concentration is observed to slowly increase from 3 ppm to 15 ppm, while adjacent pressure sensors indicate a pressure drop from 1.2 MPa to 1.1 MPa. On a daily scale, this trend is more pronounced during the day and relatively gradual at night. Weekly data show that anomalies begin to accumulate. Time series analysis shows that the pressure drop precedes the concentration increase by approximately 30 minutes, consistent with the typical time series characteristics of a trace leak. Spatial analysis reveals that the anomaly first appears at sensor A, followed by sensors at points B (10 meters from point A) and C (15 meters from point A). Based on the spatial layout of the sensors and the time difference between anomalies, the propagation speed of the fault signature is calculated to be approximately 0.2 meters per minute, with the propagation direction consistent with the pipeline's direction. This spatial propagation characteristic is consistent with leaks at pipeline interfaces. The temporal and spatial features are normalized and weighted before being fused. In this example, the temporal feature weight is 0.6 (reflecting the dynamic characteristics of the leak's development), and the spatial feature weight is 0.4. The calculated fault judgment value is 0.65, indicating a moderate leak risk.

[0096] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0097] (1) The gas fault judgment value is divided into five intervals according to the four numerical nodes of 0.3, 0.5, 0.7, and 0.9. The data of each interval is numerically counted to generate a fault level distribution table;

[0098] (2) Calculate the frequency and duration of the data in the fault level distribution table, and perform numerical weighting according to the frequency weight of 0.4 and the duration weight of 0.6 to form weighted statistical data;

[0099] (3) The three indicators of signal strength, response time, and data integrity rate in the sensor status data are numerically quantified to construct a sensor reliability index table;

[0100] (4) Perform 0-1 interval normalization on the data in the sensor reliability index table, calculate the comprehensive reliability score of each sensor, and obtain the sensor score table;

[0101] (5) Numerically combine the weighted statistical data with the sensor score table, calculate the confidence of the fault judgment according to the weighted coefficients of the three indicators, and generate fault credibility data;

[0102] (6) Numerically associate the fault credibility data with the fault level distribution table, modify the fault level according to the confidence level, and output the gas fault warning result.

[0103] Specifically, the gas fault judgment values ​​are divided into five intervals based on four key nodes: 0.3, 0.5, 0.7, and 0.9, corresponding to normal status (0-0.3), mild anomaly (0.3-0.5), moderate anomaly (0.5-0.7), severe anomaly (0.7-0.9), and critical status (0.9-1.0), respectively. Data within each interval are statistically analyzed, recording information such as the number of data points, value distribution, and occurrence time period, to form a fault level distribution table. The data in the fault level distribution table is then analyzed in depth, focusing on calculating two key metrics: frequency of occurrence and duration. Frequency reflects the proportion of times the fault state occurs within the monitoring period and is calculated by dividing the cumulative number of data points within each level interval by the total number of monitoring times. Duration reflects the continuity of the fault state and is calculated by calculating the duration of each fault state and summing them up to obtain the total duration. When calculating weighted statistical data, a frequency weight of 0.4 and a duration weight of 0.6 are used. This weighting configuration places greater emphasis on the persistence of the fault state, as persistent anomalies often indicate a more serious fault risk.

[0104] Analysis of sensor status data involves three core metrics: signal strength, response time, and data integrity. Signal strength, assessed by measuring the amplitude of the sensor's output signal, reflects the sensor's operating status. Response time, which measures the sensor's responsiveness to environmental changes, is determined by measuring the signal's lag time. Data integrity is calculated by counting the percentage of valid data points relative to the total number of sampling points. These three metrics are quantified to form a sensor reliability index table. The data in the sensor reliability index table is normalized to a range of 0-1 to ensure comparability among metrics of different dimensions. This normalization process considers the physical meaning of each metric. For example, higher signal strength indicates better reliability, shorter response time indicates better performance, and higher data integrity indicates greater stability. By combining these normalized metrics, a comprehensive reliability score is calculated for each sensor, forming a sensor scoring table.

[0105] The combination of weighted statistical data and the sensor scoring table requires consideration of data relevance. First, the relationship between fault level and sensor reliability is analyzed. When sensor reliability is low, the corresponding fault determination results require reliability compensation. The confidence level of the fault determination is calculated using the weighted coefficients of three indicators (signal strength coefficient, response time coefficient, and data integrity coefficient) to generate fault credibility data. Finally, the fault credibility data is correlated with the fault level distribution table. The initial fault level is corrected based on the confidence level. For example, when the confidence level is low, the fault level determination result needs to be appropriately lowered. The corrected result is output as the gas fault warning result.

[0106] For example, in 24 hours of continuous monitoring data, the fault judgment value fluctuates at different times. From 6:00 a.m. to 8:00 a.m., the judgment value fluctuates between 0.35-0.45, which is a slight abnormality range; from 10:00 a.m. to 2:00 p.m., the judgment value rises to 0.55-0.65, entering the moderate abnormality range; the rest of the time, it remains at a normal level of 0.2-0.3. Statistics show that the frequency of slight abnormalities is 0.08 (2 hours / 24 hours) and lasts for 2 hours; the frequency of moderate abnormalities is 0.17 (4 hours / 24 hours) and lasts for 4 hours. According to the set weight calculation, the weighted value of slight abnormalities is 0.08×0.4+2 / 24×0.6=0.082, and the weighted value of moderate abnormalities is 0.17×0.4+4 / 24×0.6=0.168.

[0107] At the same time, the sensor status shows that the signal strength remains above 85% of the full scale, the response time is stable within 1.2 times the design value, and the data integrity rate reaches 96%. These indicators are normalized to obtain scores of 0.85, 0.83, and 0.96, respectively. Combining the weighted coefficients of the three indicators (0.3, 0.3, and 0.4), the overall reliability score of the sensor is calculated to be 0.886. Combining the weighted statistical data with the sensor score, considering the high sensor reliability, the preliminary fault level is confirmed. The generated early warning results show that the monitoring point is in a moderate abnormal state with a confidence level of 0.886, and personnel need to be arranged for on-site inspection.

[0108] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0109] (1) Extract the fault type, fault severity, and fault location information from the gas fault warning results, classify them according to the fault characteristics, and construct fault retrieval conditions;

[0110] (2) Establish a retrieval index for historical data based on the three dimensions of fault type, location, and fault severity, match the fault retrieval conditions with the retrieval index, and obtain historical case data;

[0111] (3) Compare the fault characteristic parameters in the historical case data with the gas fault warning results, calculate the Euclidean distance between the characteristic parameters, and generate similarity data;

[0112] (4) Arrange the similarity data in descending order, select historical cases with a similarity greater than 0.8, extract their disposal plans and disposal effect information, and form candidate disposal plans;

[0113] (5) Quantify the candidate disposal plans according to the three indicators of disposal time, resource consumption, and restoration effect, and generate a plan scoring table;

[0114] (6) Prioritize the candidate disposal solutions based on the solution scoring table, select the solution with the highest score, and output the gas fault disposal instructions.

[0115] Specifically, fault type information is extracted, including the specific fault category (e.g., leak, pressure anomaly, temperature anomaly), and the stage of development (early, mid-term, late stage). Fault severity information includes the severity level (minor, moderate, severe) and the impact area. Fault location information includes specific pipeline network coordinates, surrounding environmental conditions, and geological conditions. This extracted information is categorized and organized according to fault characteristics, and a fault retrieval condition is established that encompasses multiple key characteristic dimensions. The retrieval index for historical data is based on three core dimensions: the fault type dimension captures the characteristic descriptions, judgment criteria, and handling requirements for different fault types; the location dimension encompasses spatial information such as pipeline network layout, geographic environment, and population density; and the fault severity dimension reflects the severity, development speed, and impact area of ​​the fault. The fault retrieval condition is then matched against the retrieval index based on these three dimensions. A multi-level screening strategy is employed: first, a rough screening based on fault type, then a regional screening based on location information, and finally, a precise match based on fault severity, resulting in historical case data with similar characteristics to the current fault.

[0116] The comparison of historical case data with current gas fault warning results involves calculating the similarity of multiple features. First, the feature parameters are standardized to make parameters of different dimensions comparable. Euclidean distances are then calculated for each feature dimension, including gas concentration, pressure, temperature, acoustic, and vibration distances. The calculated similarity data reflects both the overall degree of similarity and the differences in each feature dimension. After sorting the similarity data in descending order, historical cases with a similarity greater than 0.8 are selected as reference cases. This threshold ensures that the selected cases have a sufficiently high similarity with the current fault. From these highly similar cases, information on the treatment plan is extracted, including the specific treatment steps, required equipment, and staffing. Information on the treatment effect is also extracted, including repair time, resource investment, and the degree of fault elimination, to form a set of candidate treatment plans.

[0117] The scoring calculation for candidate resolution plans considers three key metrics: The resolution time metric reflects the time efficiency of the resolution execution, including response time, repair time, and recovery time; the resource consumption metric reflects the cost-effectiveness of the resolution, including manpower input, equipment usage, and material consumption; and the repair effect metric evaluates the effectiveness of the resolution, including the degree of fault elimination, system recovery status, and subsequent stability. By quantifying these metrics, a scoring table is generated for each candidate resolution. A comprehensive evaluation approach is used to prioritize the resolution plans based on the resolution scoring table. First, a weight coefficient is calculated for each metric, reflecting its importance in the decision-making process. The scores for each metric are then multiplied by the weight coefficients and summed to obtain a comprehensive resolution score. The resolution plans are prioritized based on their comprehensive scores, and the highest-scoring solution is selected as the resolution plan, generating detailed fault resolution instructions.

[0118] For example, a gas pipeline monitoring point detected a fault warning indicating a moderate leak. The data extraction process yielded the following characteristic information: the fault type was a trace leak, characterized by a slow increase in gas concentration (from 3 ppm to 15 ppm) and a slight decrease in pressure (from 1.2 MPa to 1.1 MPa); the fault location was at a junction in the main pipeline, surrounded by a non-dense residential area; the fault severity was moderate, having persisted for four hours without showing any signs of rapid deterioration. The historical database was searched, first filtering by trace leak type, yielding 300 historical records. Further filtering by pipeline junction location yielded 80 records. Finally, filtering by moderate fault severity yielded 15 similar cases. These cases were compared for characteristic parameters, with similarity calculation focusing on key parameters such as the gas concentration change rate, pressure trend, and fault duration. After similarity calculation and sorting, five cases with a similarity greater than 0.8 were selected for further analysis.

[0119] The disposal plans for these five cases have their own characteristics: Plan A uses temporary sealing and then replacing the interface, with a disposal time of 8 hours and the need to interrupt the gas supply; Plan B uses pressure repair technology, with a disposal time of 12 hours and no need to interrupt the gas supply; Plan C uses an interface modification plan, with a disposal time of 24 hours and the need for local pipe network modification; Plan D uses glue injection sealing technology, with a disposal time of 6 hours and a temporary repair; Plan E uses an overall replacement plan, with a disposal time of 16 hours and the most thorough treatment effect. Through scoring calculation, taking into account the different requirements for disposal time, resource consumption and repair effect under the current circumstances, Plan B received the highest score. The generated disposal instructions clearly stipulate the use of pressure repair technology for disposal, and list in detail the required special equipment, technical personnel configuration, operating procedures and safety precautions.

[0120] In a specific embodiment, the process of performing numerical comparison between the fault characteristic parameters in the historical case data and the gas fault warning results may specifically include the following steps:

[0121] (1) Decompose historical case data and gas fault warning results according to five dimensions: gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline network vibration data to obtain characteristic component data;

[0122] (2) Perform zero-mean normalization on the feature component data, map each dimension data to a unified numerical interval, and generate standardized feature data;

[0123] (3) Assigning weight coefficients to each dimension feature in the standardized feature data according to the numerical importance, weighting the data to form weighted feature data;

[0124] (4) Construct a distance calculation matrix for the weighted feature data, calculate the Euclidean distance value between each pair of historical cases and the current fault, and generate a distance matrix;

[0125] (5) Convert the values ​​in the distance matrix into similarity coefficients according to the inverse relationship, normalize the similarity coefficients, and obtain similarity degree data;

[0126] (6) Group and count the similarity data, calculate the similarity contribution rate of each feature dimension, and output the similarity data.

[0127] Specifically, a feature decomposition was performed on historical case data and current gas fault warning results, extracting features from five dimensions: gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline network vibration data. Gas concentration data reflects changes in the leakage level, including concentration value, change rate, and fluctuation range; pipeline pressure data reflects the operating status of the pipeline network, including pressure value, pressure drop rate, and fluctuation amplitude; gas temperature data records temperature variation characteristics, including temperature value, temperature gradient, and fluctuation pattern; leakage acoustic data describes acoustic characteristics, including sound pressure level, spectral distribution, and duration; and pipeline network vibration data characterizes mechanical properties, including amplitude, frequency, and vibration mode.

[0128] The extracted feature component data undergoes zero-mean normalization, unifying data of varying dimensions and ranges onto a comparable scale. The normalization process first calculates the mean of each dimension, then subtracts the mean from the original data and divides it by the standard deviation, resulting in the processed data having zero mean and unit standard deviation. This process eliminates the influence of dimension while preserving the data's distribution. Normalized data facilitates subsequent similarity calculations. Weights are assigned to the normalized feature data based on the importance of each dimension in fault diagnosis. Gas concentration data typically receives a higher weight because it directly reflects leaks; pressure data is second, reflecting the operational status of the pipeline network; and the weights of temperature, acoustics, and vibration data are dynamically adjusted based on the fault type. The weight coefficients are determined by considering the feature's stability, response speed, and anti-interference capability. This weighting emphasizes the role of key features in similarity calculations.

[0129] The weighted feature data is used to construct a distance calculation matrix. Each element of the matrix represents the Euclidean distance between a pair of historical cases and the current fault along a specific feature dimension. Distance calculations take into account both the absolute differences in feature values ​​and the differences in their changing trends. Smaller distance values ​​indicate greater similarity. A complete distance matrix is ​​formed by calculating distances across all feature dimensions. The distance matrix is ​​converted to similarity coefficients using an inverse relationship: smaller distances indicate higher similarity. Nonlinear relationships are considered during this conversion process, ensuring that the similarity coefficients better reflect the actual degree of similarity. The similarity coefficients are normalized, mapping the values ​​to the range of 0–1 for intuitive understanding and comparison. The normalized similarity data clearly demonstrates the similarity relationships between different cases.

[0130] Similarity data is grouped and statistically analyzed to determine the contribution of each feature dimension to the overall similarity. The statistical process calculates the similarity contribution rate for each dimension, reflecting the importance of each feature in fault identification. This contribution rate information helps optimize feature weighting and improve similarity calculation methods. The resulting similarity data comprehensively reflects the degree of similarity between the historical case and the current fault.

[0131] For example, in analyzing a gas pipeline leak, characteristic data from the current fault showed a slow increase in gas concentration from 3ppm to 15ppm, a decrease in pressure from 1.2MPa to 1.1MPa, and a temperature maintained at approximately 28°C. Acoustic data showed a sustained sound pressure of 40-45dB, and vibration data showed an amplitude of 0.2-0.3mm. These features were decomposed and normalized when retrieved from the historical database. Considering the characteristics of the leak, weights for the gas concentration feature were set at 0.4, the pressure feature at 0.3, the acoustic feature at 0.2, and the temperature and vibration features at 0.05 each. Through distance calculation and similarity conversion, the three most similar historical cases were identified. These cases all exhibited similar gas concentration rising trends and pressure drop characteristics, providing valuable reference for fault diagnosis and resolution. Similarity analysis results showed that the similarity contribution rates for gas concentration and pressure features reached 45% and 35%, respectively, validating the rationality of the feature weighting settings.

[0132] The above describes the gas monitoring fault analysis method in the embodiment of the present application. The following describes the gas monitoring fault analysis system in the embodiment of the present application. Figure 2 In one embodiment of the present application, a gas monitoring fault analysis system includes:

[0133] The transmission module is used to collect gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline vibration data through the gas monitoring sensor group, transmit the data to the processing unit according to the Internet of Things communication protocol, pre-screen the data, and obtain the raw gas monitoring data;

[0134] A dimensionality reduction module is used to calculate the raw gas monitoring data according to the median filter formula and remove outliers, calculate characteristic parameters for the filtered data, and perform dimensionality reduction processing on the calculation results to obtain a gas characteristic vector;

[0135] A comparison module is used to establish a fault type database based on the gas characteristic vector, perform classification calculations on the characteristic data, compare the calculation results with preset parameters, and generate a gas fault mapping relationship;

[0136] a fusion module, configured to extract time series data according to the gas feature vector and the gas fault mapping relationship, perform weighted fusion calculation on the time series data and the spatial data, and output a gas fault judgment value;

[0137] An evaluation module is used to compare and classify the gas fault judgment value with a preset threshold, perform weighted calculation on the classification results, and perform credibility evaluation in combination with the sensor status data to obtain a gas fault warning result;

[0138] The sorting module is used to retrieve historical data based on the gas fault warning result, perform similarity calculation and priority sorting on the retrieved data, and generate a gas fault handling instruction.

[0139] Through the coordinated cooperation of the above components, the gas monitoring sensor group simultaneously collects gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data and pipeline vibration data, realizing the coordinated monitoring of multi-dimensional data, comprehensively reflecting the operating status of the gas pipeline network, and using the Internet of Things communication protocol for data transmission to ensure the reliability and real-time performance of data transmission. The pre-screening process effectively reduces the impact of data noise, and the median filter formula calculation can effectively remove outliers and improve data quality. The characteristic parameter calculation and dimensionality reduction processing realize the effective compression and feature extraction of data. The establishment of a fault type database enables the systematic organization and management of fault characteristics, and the classification of characteristic data. Calculation and comparative analysis improve the accuracy of fault identification. The weighted fusion calculation of time series data and spatial data fully utilizes the spatiotemporal correlation of fault characteristics, making fault judgment more comprehensive and accurate. The hierarchical comparison and weighted calculation of preset thresholds realize the precise division of fault levels. The credibility assessment combined with sensor status data ensures the reliability of the early warning results. By retrieving historical data and performing similarity calculations, the most similar historical cases are found, providing a reliable reference basis for fault handling. Priority sorting ensures the optimal selection of handling solutions, thus forming a complete, automated, and intelligent gas monitoring fault analysis process, which significantly improves the accuracy of gas pipeline fault diagnosis and the efficiency of handling.

[0140] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, which, when executed on a computer, cause the computer to execute the steps of the gas monitoring fault analysis method.

[0141] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0142] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0143] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A gas monitoring fault analysis method, characterized in that: The gas monitoring fault analysis method comprises: The gas monitoring sensor group collects gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline vibration data. The data is transmitted to the processing unit according to the Internet of Things communication protocol, and the data is pre-screened to obtain the original gas monitoring data. The raw gas monitoring data is calculated using a median filter formula and outliers are removed. Feature parameters are calculated for the filtered data, and dimensionality reduction is performed on the calculation results to obtain a gas feature vector. A fault type database is established based on the gas characteristic vector, the characteristic data is classified and calculated, the calculation results are compared with the preset parameters, and a gas fault mapping relationship is generated, including: dividing the gas characteristic vector into 24-hour data segments according to the timestamp, calculating the statistical characteristics of the gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data and pipeline vibration data in each data segment, and generating a feature statistical table; grouping the data in the feature statistical table according to the fault type label, calculating the mean center and variance distribution of each group of data, and forming a fault feature statistical matrix; dividing the data in the fault feature statistical matrix into the gas concentration range of 0-100ppm Construct segmented statistical data by dividing the data into 5 intervals, the pressure fluctuation range of 0-1.6MPa into 4 intervals, the temperature change range of -20℃-60℃ into 6 intervals, the sound pressure level of 30-120dB into 5 intervals, and the vibration amplitude of 0-5mm into 4 intervals; perform probability density calculation on the segmented statistical data to obtain the characteristic distribution curve of each fault type and generate a fault characteristic probability table; perform numerical comparison on the fault characteristic probability table with historical fault data, calculate the characteristic similarity coefficient, and establish a fault type database; perform data mapping on the input gas characteristic vector based on the fault type database, calculate the fault type probability value, and generate the gas fault mapping relationship; Extracting time series data based on the mapping relationship between the gas feature vector and the gas fault, converting the sensor position information in the gas feature vector into coordinate data, calculating the distance matrix between adjacent sensors, and constructing a spatial association table; performing spatial correlation analysis on the sensor data in the spatial association table, calculating the propagation characteristics of the fault characteristics in space, forming spatial feature data, performing weighted fusion calculation on the time series data and the spatial feature data, and outputting a gas fault determination value; Compare the gas fault judgment value with the preset threshold value, perform weighted calculation on the classification result, and perform credibility evaluation in combination with the sensor status data to obtain a gas fault warning result; Based on the gas fault warning result, historical data is retrieved, similarity calculation and priority sorting are performed on the retrieved data, and a gas fault handling instruction is generated.

2. The gas monitoring fault analysis method according to claim 1, characterized in that: The gas monitoring sensor group collects gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipe network vibration data, transmits the data to the processing unit according to the Internet of Things communication protocol, pre-screens the data, and obtains the gas monitoring raw data, including: The gas concentration data is collected at a sampling interval of 100ms, the pipeline pressure data is collected at a sampling interval of 200ms, the gas temperature data is collected at a sampling interval of 500ms, the leakage acoustic data is collected at a sampling interval of 50ms, and the pipeline network vibration data is collected at a sampling interval of 20ms; Perform timestamp alignment on the gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline network vibration data, construct data packets in 1s units, and add check bits to the data packets according to the Internet of Things communication protocol; The gas concentration data in the data packet is pre-screened by setting a concentration range threshold of 0-100ppm, a pressure range threshold of 0-1.6MPa for the pipeline pressure data, a temperature range threshold of -20°C-60°C for the gas temperature data, a sound pressure range threshold of 30-120dB for the leakage acoustic data, and an amplitude range threshold of 0-5mm for the pipe network vibration data; Merge the pre-screened gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline network vibration data into corresponding data records, and mark each data record with the acquisition time and sensor number; Performing data integrity checks on the data records, removing records with missing data, marking abnormal data records, and establishing data quality scores; The data records that have passed the integrity check are organized into a time series data table, and a data source identifier is added to generate the gas monitoring raw data.

3. The gas monitoring fault analysis method according to claim 1, characterized in that: The gas monitoring raw data is calculated according to the median filter formula and outliers are removed. Feature parameters are calculated for the filtered data. Dimensionality reduction is performed on the calculation results to obtain a gas feature vector, including: The raw gas monitoring data is grouped according to 10-minute time windows, the median value and interquartile range are calculated for each group of data, and outliers are determined and eliminated based on an interval of three times the interquartile range to obtain filtered data; Calculating the mean, standard deviation, kurtosis, and skewness of the gas concentration data in the filtered data, calculating the maximum fluctuation amplitude and fluctuation period of the pipeline pressure data, calculating the rate of change and temperature gradient of the gas temperature data, calculating the sound pressure level and frequency distribution of the leakage acoustic data, and calculating the vibration amplitude and vibration frequency of the pipeline network vibration data to generate a feature data set; Normalizing the mean, standard deviation, kurtosis, and skewness data in the feature data group to the range of 0-1, normalizing the maximum fluctuation amplitude and fluctuation period data to the range of -1-1, linearly transforming the rate of change and temperature gradient data to a standard normal distribution, logarithmically transforming the sound pressure level and frequency distribution data, and performing a minimum-maximum transformation on the vibration amplitude and vibration frequency data to obtain normalized feature data; The normalized feature data are sorted according to the variance contribution rate, and the feature dimensions with a cumulative variance contribution rate of 95% are selected to form a main feature data set; The feature importance of each feature data in the main feature data set is calculated according to the correlation coefficient matrix, and the feature data is linearly combined based on the importance weight to construct compressed feature data; the compressed feature data is converted into a numerical vector form, and a time index and sensor location identifier are added to generate the gas feature vector.

4. The gas monitoring fault analysis method according to claim 1, characterized in that: The method extracts time series data based on the mapping relationship between the gas feature vector and the gas fault, converts sensor position information in the gas feature vector into coordinate data, calculates a distance matrix between adjacent sensors, and constructs a spatial association table; performs spatial correlation analysis on the sensor data in the spatial association table, calculates the propagation characteristics of the fault feature in space to form spatial feature data, performs weighted fusion calculation on the time series data and the spatial feature data, and outputs a gas fault determination value, including: Divide the gas characteristic vector into data segments according to three time scales: hour, day, and week, calculate the change trend of the data at each time scale, and generate time series change data; Extracting the time association rules of the fault characteristics from the gas fault mapping relationship, performing time association analysis on the time series change data, and obtaining time series feature data; Converting the sensor position information in the gas feature vector into coordinate data, calculating the distance matrix between adjacent sensors, and constructing a spatial association table; Performing spatial correlation analysis on the sensor data in the spatial association table, calculating the propagation characteristics of the fault features in space, and forming spatial feature data; Normalizing the temporal feature data and the spatial feature data, calculating weight coefficients of the two types of feature data, and generating feature fusion data; The feature fusion data and the gas fault mapping relationship are numerically calculated, the fault degree is quantified according to the interval 0-1, and the gas fault judgment value is output.

5. The gas monitoring fault analysis method according to claim 1, characterized in that: The gas fault judgment value is compared with a preset threshold value for grading, weighted calculation is performed on the grading result, and credibility evaluation is performed in combination with the sensor status data to obtain a gas fault warning result, including: The gas fault judgment value is divided into five intervals according to the four numerical nodes of 0.3, 0.5, 0.7 and 0.9, and the data of each interval is statistically analyzed to generate a fault level distribution table; Calculate the frequency and duration of occurrence of the data in the fault level distribution table, perform numerical weighting according to a frequency weight of 0.4 and a duration weight of 0.6, and form weighted statistical data; Numerical quantification of the three indicators of signal strength, response time, and data integrity rate in the sensor status data is performed to construct a sensor reliability index table; Normalizing the data in the sensor reliability index table to a range of 0-1, calculating the comprehensive reliability score of each sensor, and obtaining a sensor score table; Numerically combining the weighted statistical data with the sensor scoring table, calculating the confidence of the fault judgment according to the weighted coefficients of the three indicators, and generating fault credibility data; The fault credibility data is numerically associated with the fault level distribution table, the fault level is corrected according to the confidence level, and the gas fault warning result is output.

6. The gas monitoring fault analysis method according to claim 1, characterized in that: The method of retrieving historical data based on the gas fault warning result, performing similarity calculation and priority sorting on the retrieved data, and generating a gas fault handling instruction includes: Extract the fault type, fault severity, and fault location information from the gas fault warning results, classify them according to the fault characteristics, and construct fault retrieval conditions; Establishing a retrieval index for the historical data according to three dimensions: fault type, location, and fault severity, matching the fault retrieval condition with the retrieval index to obtain historical case data; Comparing the fault characteristic parameters in the historical case data with the gas fault warning results, calculating the Euclidean distance between the characteristic parameters, and generating similarity data; Arrange the similarity data in descending order, select historical cases with a similarity greater than 0.8, extract their disposal plans and disposal effect information, and form candidate disposal plans; Quantitatively calculate the candidate disposal plans based on three indicators: disposal time, resource consumption, and repair effect, and generate a plan scoring table; The candidate disposal solutions are prioritized based on the solution scoring table, the disposal solution with the highest score is selected, and the gas fault disposal instruction is output.

7. The gas monitoring fault analysis method according to claim 6, characterized in that: The method of comparing the fault characteristic parameters in the historical case data with the gas fault warning results, calculating the Euclidean distance between the characteristic parameters, and generating similarity data includes: Decomposing the historical case data and the gas fault warning results according to five dimensions: gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline network vibration data, to obtain characteristic component data; Performing zero-mean normalization processing on the feature component data, mapping each dimension data to a unified numerical interval, and generating standardized feature data; Assigning weight coefficients to each dimensional feature in the standardized feature data according to the numerical importance, weighting the data to form weighted feature data; Constructing a distance calculation matrix for the weighted feature data, calculating the Euclidean distance value between each pair of historical cases and the current fault, and generating a distance matrix; The values ​​in the distance matrix are converted into similarity coefficients according to an inverse relationship, and the similarity coefficients are normalized to obtain similarity data; the similarity data are grouped and counted, and the similarity contribution rate of each feature dimension is calculated, and the similarity data is output.

8. A gas monitoring fault analysis system, used to implement the gas monitoring fault analysis method according to any one of claims 1 to 7, characterized in that: The gas monitoring fault analysis system includes: The transmission module is used to collect gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data, and pipeline vibration data through the gas monitoring sensor group, transmit the data to the processing unit according to the Internet of Things communication protocol, pre-screen the data, and obtain the raw gas monitoring data; A dimensionality reduction module is used to calculate the raw gas monitoring data according to the median filter formula and remove outliers, calculate characteristic parameters for the filtered data, and perform dimensionality reduction processing on the calculation results to obtain a gas characteristic vector; A comparison module is used to establish a fault type database based on the gas characteristic vector, classify and calculate the characteristic data, compare the calculation results with the preset parameters, and generate a gas fault mapping relationship, including: dividing the gas characteristic vector into 24-hour data segments according to the timestamp, calculating the statistical characteristics of the gas concentration data, pipeline pressure data, gas temperature data, leakage acoustic data and pipeline vibration data in each data segment, and generating a feature statistical table; grouping the data in the feature statistical table according to the fault type label, calculating the mean center and variance distribution of each group of data, and forming a fault feature statistical matrix; grouping the data in the fault feature statistical matrix according to the gas concentration range of 0-100ppm The range is divided into 5 intervals, the pressure fluctuation range of 0-1.6MPa is divided into 4 intervals, the temperature change range of -20℃-60℃ is divided into 6 intervals, the sound pressure level of 30-120dB is divided into 5 intervals, and the vibration amplitude of 0-5mm is divided into 4 intervals, and segmented statistical data is constructed; probability density calculation is performed on the segmented statistical data to obtain the characteristic distribution curve of each fault type and generate a fault feature probability table; the fault feature probability table is numerically compared with historical fault data, the feature similarity coefficient is calculated, and a fault type database is established; data mapping is performed on the input gas feature vector based on the fault type database, the fault type probability value is calculated, and the gas fault mapping relationship is generated; A fusion module is configured to extract time series data based on the mapping relationship between the gas feature vector and the gas fault, convert the sensor position information in the gas feature vector into coordinate data, calculate the distance matrix between adjacent sensors, and construct a spatial association table; perform spatial correlation analysis on the sensor data in the spatial association table, calculate the propagation characteristics of the fault characteristics in space, form spatial feature data, perform weighted fusion calculation on the time series data and the spatial feature data, and output a gas fault judgment value; An evaluation module is used to compare and classify the gas fault judgment value with a preset threshold, perform weighted calculation on the classification results, and perform credibility evaluation in combination with the sensor status data to obtain a gas fault warning result; The sorting module is used to retrieve historical data based on the gas fault warning result, perform similarity calculation and priority sorting on the retrieved data, and generate a gas fault handling instruction.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the gas monitoring fault analysis method according to any one of claims 1 to 7 is implemented.

Images