Intelligent monitor and method for multi-parameter fusion analysis in natural gas pipeline operation

Abstract

The invention belongs to the field of natural gas pipeline operation and maintenance, and particularly discloses an intelligent monitor for multi-parameter fusion analysis in natural gas pipeline operation. The sensor module is arranged on a natural gas pipeline and is used for acquiring original data and signal intensity; the preprocessing module preprocesses the original data to obtain preprocessed data, and then adopts a self-adaptive compression algorithm to obtain compressed data; the fusion module adopts a multi-source data fusion algorithm for the compressed data to obtain fusion features; the diagnosis module inputs the fusion features into a fault diagnosis model based on a machine learning algorithm, and when it is diagnosed that a fault exists, the fault type is output; the positioning module obtains the position, geographic information data and historical operation data of the sensor module, and obtains a fault position by adopting a positioning algorithm according to the signal intensity, the position, geographic information data and historical operation data of the sensor module and a fault type; and the early warning module generates a fault report according to the fault position and the fault type, sends out an early warning signal and uploads the fault report.

Description

Technical Field

[0001] The present invention relates to the field of natural gas pipeline operation and maintenance, and in particular to an intelligent monitoring instrument for multi-parameter fusion analysis during natural gas pipeline operation. Background Art

[0002] Traditional natural gas pipeline monitoring systems primarily rely on single sensors for parameter monitoring, such as pressure, temperature, or flow. However, the performance of these sensors is often limited by environmental conditions. For example, sensor accuracy degrades due to temperature fluctuations or long-term use, resulting in inaccurate monitoring data. Furthermore, the monitoring range of a single sensor is limited, failing to fully reflect the overall operational status of the pipeline. For example, a pressure sensor can only monitor pressure changes but cannot identify corrosion or leak risks within the pipeline. These limitations make it difficult for traditional monitoring systems to provide reliable early warning and diagnostic capabilities in complex environments.

[0003] Existing natural gas pipeline monitoring systems typically rely on local data storage and periodic manual data reading, resulting in inefficient data processing and slow response times. For example, sensor data must be manually read or recorded on simple local storage devices, making real-time transmission and remote monitoring impossible. Furthermore, traditional systems have limited data processing capabilities and are unable to comprehensively analyze multi-source data, resulting in low fault diagnosis accuracy. This inefficient data processing method not only increases operational and maintenance costs but can also miss optimal troubleshooting opportunities due to data lags.

[0004] Natural gas pipelines are often laid in complex geographical environments or extreme climatic conditions, such as mountainous areas, deserts, or cold regions. These environmental factors pose significant challenges to the stability and reliability of monitoring systems. For example, extreme temperatures can degrade sensor performance, and inclement weather can affect the stability of data transmission. Furthermore, the complex terrain along pipelines complicates sensor deployment and maintenance, further limiting the applicability of traditional monitoring systems. Under these conditions, existing monitoring technologies struggle to meet the requirements for long-term, safe pipeline operation. Summary of the Invention

[0005] In view of this, an embodiment of the present invention provides an intelligent monitoring instrument for multi-parameter fusion analysis during natural gas pipeline operation to solve at least one of the above technical problems.

[0006] To achieve the above objectives, in a first aspect, an intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation is provided, comprising:

[0007] A sensor module is installed on the natural gas pipeline to obtain raw data and signal strength;

[0008] A preprocessing module, configured to preprocess the original data to obtain preprocessed data, and apply an adaptive compression algorithm to the preprocessed data to obtain compressed data;

[0009] A fusion module, configured to obtain fusion features by applying a multi-source data fusion algorithm to the compressed data;

[0010] a diagnosis module, configured to input the fusion features into a fault diagnosis model based on a machine learning algorithm, and output a fault type when a fault is diagnosed;

[0011] a positioning module, configured to obtain the position of the sensor module, geographic information data near the natural gas pipeline, and historical operation data of the natural gas pipeline, and to determine the fault location using a positioning algorithm based on the signal strength, the position of the sensor module, the geographic information data, the historical operation data, and the fault type;

[0012] The early warning module is used to generate a fault report according to the fault location and the fault type, issue a warning signal, and upload the fault report.

[0013] In a second aspect, an intelligent monitoring method for multi-parameter fusion analysis in natural gas pipeline operation is provided, comprising the following steps:

[0014] Get raw data and signal strength;

[0015] Preprocessing the original data to obtain preprocessed data, and applying an adaptive compression algorithm to the preprocessed data to obtain compressed data;

[0016] A multi-source data fusion algorithm is used to obtain fusion features on the compressed data;

[0017] Inputting the fusion features into a fault diagnosis model based on a machine learning algorithm, and outputting the fault type when a fault is diagnosed;

[0018] Obtaining the location of the sensor module, geographic information data near the natural gas pipeline, and historical operation data of the natural gas pipeline, and determining the fault location using a positioning algorithm based on the signal strength, the location of the sensor module, the geographic information data, the historical operation data, and the fault type;

[0019] Generate a fault report based on the fault location and the fault type, issue an early warning signal, and upload the fault report.

[0020] The above technical solution has the following beneficial technical effects:

[0021] The intelligent monitoring instrument for multi-parameter fusion analysis in the operation of the natural gas pipeline realizes multi-dimensional data collection through the sensor module, and combines the adaptive compression algorithm of the preprocessing module to reduce the data transmission and storage costs, while ensuring the integrity of key information; the fusion module adopts a multi-source data fusion algorithm to effectively improve the spatiotemporal correlation and feature representation ability of the monitoring data, overcoming the limitations of single parameter analysis; the fault diagnosis model constructed by the diagnosis module based on the machine learning algorithm greatly improves the accuracy of fault identification and early warning capabilities through high-dimensional analysis of fusion features; the positioning module realizes the precise spatial positioning and traceability analysis of the fault point by fusing the multi-dimensional parameters of signal strength, geographic information and historical operation data; finally, through the automatic fault report generation and warning signal linkage mechanism of the early warning module, a full-process closed-loop monitoring system from data collection, feature analysis, fault diagnosis to location tracking is constructed, which significantly improves the operational safety level of the natural gas pipeline, reduces the cost of manual inspections and the risk of leakage accidents, and provides intelligent support for pipeline maintenance decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are provided for a better understanding of the present invention and are not intended to limit the present invention.

[0023] Figure 1 This is a structural block diagram of an intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to an embodiment of the present invention;

[0024] Figure 2 This is a schematic structural diagram of an intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to an embodiment of the present invention;

[0025] Figure 3 is a structural block diagram of a preprocessing module in an embodiment of the present invention;

[0026] Figure 4 is a structural block diagram of a fusion module in an embodiment of the present invention;

[0027] Figure 5 is a structural block diagram of a diagnostic module in an embodiment of the present invention;

[0028] Figure 6 is a structural block diagram of a positioning module in an embodiment of the present invention;

[0029] Figure 7 This is a structural block diagram of an early warning module in an embodiment of the present invention;

[0030] Figure 8 This is a flow chart of an intelligent monitoring method for multi-parameter fusion analysis in natural gas pipeline operation according to an embodiment of the present invention;

[0031] Figure 9 Schematic diagram of the structure of a computer system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following description of exemplary embodiments of the present invention is made in conjunction with the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding. These details should be considered as merely exemplary. Therefore, it should be appreciated by those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0033] Example 1

[0034] like Figures 1 to 2 As shown, this embodiment provides an intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation, including:

[0035] A sensor module is installed on the natural gas pipeline to obtain raw data and signal strength;

[0036] A preprocessing module, configured to preprocess the original data to obtain preprocessed data, and apply an adaptive compression algorithm to the preprocessed data to obtain compressed data;

[0037] A fusion module, configured to obtain fusion features by applying a multi-source data fusion algorithm to the compressed data;

[0038] a diagnosis module, configured to input the fusion features into a fault diagnosis model based on a machine learning algorithm, and output a fault type when a fault is diagnosed;

[0039] a positioning module, configured to obtain the position of the sensor module, geographic information data near the natural gas pipeline, and historical operation data of the natural gas pipeline, and to determine the fault location using a positioning algorithm based on the signal strength, the position of the sensor module, the geographic information data, the historical operation data, and the fault type;

[0040] The early warning module is used to generate a fault report according to the fault location and the fault type, issue a warning signal, and upload the fault report.

[0041] Specifically, the preprocessing module uses an adaptive compression algorithm to reduce redundant data transmission while ensuring the integrity of key information, thereby lowering operator network costs. The fusion module integrates compressed data from different sensors using a multi-source data fusion algorithm, effectively improving the accuracy of subsequent fault diagnosis. The diagnosis module utilizes a machine learning algorithm for efficient and accurate fault diagnosis. The positioning module uses the signal strength, sensor module location, geographic information data, historical operational data, and fault type to locate the fault, effectively improving positioning accuracy. Geographic information data refers to geographic information data about the natural gas pipeline and its surrounding environment, specifically including the following data objects: First, pipeline geographic data, including the three-dimensional coordinates of the pipeline (including latitude and longitude, altitude), pipeline depth parameters, pipeline routing topology, pipe segment material and diameter parameters, and the coordinates of key nodes (e.g., valve stations and booster stations); second, surrounding environment geographic data, including geological structure information (e.g., soil type and rock formation distribution), surface features (e.g., rivers, mountains, roads, buildings), and sensitive area annotations (e.g., residential areas and ecological protection areas).

[0042] Specifically, the adaptive compression algorithm dynamically selects a compression method based on the characteristics of the preprocessed data. For example, Principal Component Analysis (PCA) is used for dimensionality reduction compression of steady-state temperature data (low-frequency data). Lossless compression is used for pressure spike data (high-frequency data) to ensure that key information is not lost.

[0043] Specifically, the multi-source data fusion algorithm includes: a weighted average algorithm, which assigns weights according to sensor accuracy (for example, a pressure sensor weight of 0.7, a flow sensor weight of 0.3); or a Kalman filter algorithm, which iteratively optimizes the estimated value through the state equation and the observation equation to eliminate noise interference for data fusion.

[0044] Specifically, the positioning algorithm may include a triangulation positioning method, a distance calculation method, a machine learning algorithm generation model, or a geographic information data direct annotation method, etc.

[0045] Specifically, the signal strength refers to the amplitude of the physical signal detected by each sensor in the sensor module. For example, the signal strength of an acoustic sensor is used to locate leaks. The signal strength of a gas sensor reflects changes in gas concentration within a pipeline. For example, in the event of a natural gas leak, the signal strength detected by the gas sensor will reflect the concentration of the leaked gas, helping to locate the leak. The signal strength of a fiber optic sensor reflects the attenuation of the optical signal within the optical fiber. When a pipeline is ruptured or deformed, changes in the signal strength of the fiber optic sensor can detect the location of damage. The signal strength of a gas quality sensor is related to changes in gas flow within the pipeline. The sensor can sense changes in gas flow and determine whether the pipeline is clogged or has flow anomalies based on changes in signal strength. The signal strength of a radar sensor reflects the distance and position of objects around the pipeline. Radar sensors can detect external settlement, displacement, or obstacles near the pipeline. Changes in signal strength can be used to determine whether there is external interference or potential pipeline damage. The signal strength of an infrared thermal imaging sensor reflects changes in pipeline surface temperature. When a pipeline leak or anomaly occurs, the thermal imaging sensor can locate the heat source or leak by monitoring temperature differences.

[0046] Specifically, the sensor module may include any combination of acoustic wave sensors, gas sensors, optical fiber sensors, gas quality sensors (for example, methane concentration sensors or carbon dioxide concentration sensors), radar sensors, infrared thermal imaging sensors and environmental radiation sensors.

[0047] Specifically, the acoustic wave sensor is preferably an ultrasonic sensor, installed on the outer surface or inside the pipeline. It is used to detect abnormal sounds or structural issues such as cracks and bubbles within the natural gas pipeline. It is particularly suitable for detecting acoustic fluctuations caused by microcracks or corrosion in the pipeline wall. By analyzing the frequency and propagation pattern of the acoustic waves using Fourier transforms, wavelet transforms, time-frequency decomposition, or support vector machines, it can detect potential leaks or failures.

[0048] Specifically, the gas sensor uses electronic nose technology and is installed around natural gas pipelines, particularly at pipeline outlets or in areas with a higher risk of gas leaks. The electronic nose, composed of multiple sub-gas sensors, analyzes the combined characteristics of gases and can effectively identify subtle changes in the gas, particularly those in trace amounts of harmful gases (such as hydrogen sulfide and ammonia) within natural gas. This electronic nose technology is highly sensitive in gas leak detection, capable of identifying subtle changes in the gas flow and providing more accurate monitoring data.

[0049] Specifically, the fiber optic sensor is installed on the outer surface of a natural gas pipeline. By utilizing the changes in the optical fiber's response to different environmental conditions (e.g., temperature, strain, etc.), it can monitor physical parameters such as strain and temperature outside the pipeline. It can also indirectly infer the internal state of the pipeline (e.g., temperature, pressure, etc.) through one or a combination of physical models, regression algorithms, or finite element analysis. In particular, the use of distributed fiber optic sensing technology enables real-time detection of conditions at multiple points along the pipeline with high spatial resolution and accuracy. The fiber optic sensor is particularly useful for long-distance monitoring and high-risk areas (e.g., mountainous areas and complex terrain). The fiber optic sensor is connected to a data acquisition device via optical fiber, transmitting data in real time to a central processing system for analysis. A physical model is a mathematical model based on the principles of material mechanics and thermodynamics that establishes the relationship between external physical quantities (e.g., temperature, strain, etc.) and internal states (e.g., temperature, pressure, etc.) of the pipeline. For example, changes in pressure inside the pipeline will cause changes in the strain on the pipeline surface. By measuring the strain on the pipeline surface, changes in pressure inside the pipeline can be inferred. A regression algorithm is a regression model based on historical data (e.g., linear regression, support vector regression, etc.) that can be used to infer internal states using external measurement data. Machine learning algorithms can be used to train a model based on historical data to predict internal conditions. Finite element analysis (FEA) simulates the physical behavior of a pipeline. This allows data such as surface strain and temperature acquired by fiber optic sensors to be combined with a finite element model to infer internal conditions such as temperature and pressure. FEA can accurately simulate the response of a pipeline under varying conditions, providing a theoretical basis for inferring internal conditions.

[0050] Specifically, the gas quality sensor is installed within the pipeline, preferably in a section with high flow rates and large pressure differentials. It is used to monitor gas quality (e.g., methane or oxygen concentration) in real time within the natural gas pipeline. It is highly sensitive to fluctuations in gas composition and is suitable for use in pipeline environments with high flow rates and large pressure differentials. Compared to traditional gas analyzers, gas quality sensors offer greater adaptability and accuracy.

[0051] Specifically, radar sensors are installed on the exterior of pipelines, particularly at key locations (such as bridges, tunnels, and landslide areas) to monitor pipeline deformation or displacement, providing real-time monitoring during natural disasters such as earthquakes and landslides. By emitting electromagnetic waves and receiving changes in reflected waves, these radar sensors can detect subtle changes in the ground or pipeline structure, providing early warning of potential pipeline deformation, collapse, and other issues. The radar sensors are connected to a central processing system via wireless or wired connections to transmit deformation data.

[0052] Specifically, the infrared thermal imaging sensor is installed on the surface of a natural gas pipeline, particularly at leak points, corroded areas, or sections of pipe with a higher risk of gas leakage. The infrared thermal imaging sensor is used to detect the surface temperature distribution of the pipeline and identify any abnormal temperature points (e.g., leaks) in the pipeline. By analyzing temperature changes, abnormal heat flow in the pipeline can be detected in advance, thereby identifying potential fault areas (e.g., gas leaks, corrosion, blockages, etc.). The infrared thermal imaging sensor is connected to the central processing system via a data cable, transmitting real-time temperature change data.

[0053] Specifically, the environmental radiation sensor is installed in the environment surrounding a natural gas pipeline, particularly near radiation sources such as nuclear power plants and chemical plants. It is used to detect radiation levels in the environment surrounding the natural gas pipeline. In some special environments (e.g., near nuclear power plants), changes in environmental radiation levels can indicate safety hazards. This sensor can provide additional safety monitoring data, especially in the event of unexpected events such as external impacts or damage to the pipeline. The environmental radiation sensor is connected to a central processing system via wireless or wired connections, transmitting radiation data for real-time analysis.

[0054] Specifically, if Figure 3 As shown, the preprocessing module specifically includes:

[0055] a first preprocessing unit, configured to remove duplicate data from the original data to obtain first data;

[0056] a second preprocessing unit, configured to remove outliers from the first data using an anomaly detection algorithm to obtain second data;

[0057] a filtering unit, configured to obtain historical second data of a previous period, generate a threshold interval using a threshold generation algorithm based on the historical second data, and remove values ​​not belonging to the threshold interval from the second data to obtain preprocessed data;

[0058] A compression unit, configured to apply an adaptive compression algorithm to the pre-processed data to obtain compressed data;

[0059] The adaptive compression algorithm includes an adaptive selection algorithm based on variation amplitude, an adaptive compression algorithm based on compression error, and an adaptive compression algorithm based on machine learning.

[0060] Specifically, the first pre-processing unit traverses the raw data, marks duplicate data, and deletes excess duplicate data, retaining only one copy of each data point to ensure data accuracy and consistency. The deduplication algorithm is based on fields such as the timestamp of the data acquisition to ensure that each record is unique.

[0061] Specifically, in the second preprocessing unit, the anomaly detection algorithm includes the Z-score method and the interquartile range method. An outlier is a data point that is significantly different from the normal data pattern. An outlier is caused by misoperation or system failure. Taking the Z-score method as an example, the calculation formula is as follows:

[0062]

[0063] Where z is the Z value representing the standard score, x is the data point to be tested, μ is the mean of the data, and σ is the standard deviation of the data. If |z| > 3, the data point is considered an outlier.

[0064] Specifically, in the second preprocessing unit, deletion or filling processing is adopted for abnormal values. The filling processing includes mean filling or adjacent value filling, etc. Mean filling is to calculate the average value of other data and fill the average value to the position of the abnormal value. Missing values ​​or data with inconsistent formats are also corrected by filling or deletion processing.

[0065] Specifically, in the filtering unit, the mean or standard deviation of the historical second data is calculated, and then the threshold interval is set according to the threshold or standard deviation of the historical second data. The threshold generation algorithm is a dynamic threshold filtering method. The maximum value of the threshold interval is the sum of the mean and the preset multiplier standard deviation, and the minimum value of the threshold interval is the difference between the mean and the preset multiplier standard deviation. The preset multiplier is preferably [1, 3].

[0066] Specifically, when executing the adaptive selection algorithm based on the variation amplitude, the following steps are included:

[0067] Calculating the variation range of each data in the preprocessed data;

[0068] Compressing the pre-processed data whose variation amplitude is greater than a preset compression threshold using Huffman coding to obtain high-frequency data;

[0069] Compressing the preprocessed data whose variation amplitude is less than or equal to a preset compression threshold using a principal component analysis algorithm or a wavelet transform algorithm to obtain low-frequency data, wherein the compressed data includes the high-frequency data and the low-frequency data;

[0070] Specifically, the variation range is calculated using variance, and the adaptive selection algorithm based on the variation range is as follows:

[0071]

[0072] Where X represents the variation, Variance(X) represents the result of data feature extraction, ε represents the compression threshold, A represents the compression method A (for example, principal component analysis algorithm or wavelet transform algorithm), and B represents the compression method B (for example, Huffman coding).

[0073] When executing the adaptive compression algorithm based on compression error, the following steps are included:

[0074] Pre-compressing the pre-processed data to obtain pre-compressed data;

[0075] Calculating a compression error based on the pre-compressed data and the pre-processed data;

[0076] When the compression error is less than a preset error threshold, the compression ratio is increased to update the pre-compressed data until the compression error is equal to the preset error threshold, and the pre-compressed data at this time is output as compressed data;

[0077] When the compression error is greater than a preset error threshold, the compression rate is reduced to update the pre-compressed data until the compression error is equal to the preset error threshold, and the pre-compressed data at this time is output as compressed data.

[0078] Specifically, the pre-processed data is represented as X={X1, X2, ..., X n}, pre-compressed data X c ={X c1 , X c2 …, X cn}, the calculation formula of the compression error is as follows:

[0079]

[0080] Where, Error(X, X c ) represents the compression error.

[0081] Specifically, the adaptive compression algorithm based on compression error preserves as much information as possible during the compression process while controlling the compression error. For data such as temperature and pressure, quantization error can be used to control compression accuracy. Error is controlled by dynamically adjusting the compression ratio, achieving high compression efficiency. This algorithm is suitable for data calculations that balance compression ratio and data accuracy, and can adaptively adjust the compression strategy based on different data characteristics.

[0082] Specifically, the adaptive compression algorithm also includes adaptive compression based on machine learning, which obtains historical pre-processed data to train a machine learning model (e.g., a decision tree model, a random forest model, a support vector machine, a gradient boosting machine, etc.), and predicts the appropriate compression method based on the characteristics of different sensor data. For example, given a data set X, the machine learning model f(X) predicts which compression method should be used: Method(X) = f(X), which can adaptively adjust the compression strategy based on the training data, and is particularly suitable for complex and changing data environments. Gradient Boosting Machine (GBM) is an ensemble learning method based on decision trees that improves model performance by gradually reducing prediction errors. GBM achieves a strong classifier by weighted training of multiple weak classifiers (usually decision trees), and gradually adjusts model parameters to minimize residuals.

[0083] Specifically, the compressed data is transmitted using an adaptive data transmission strategy, which includes dynamic transmission control based on network status and layered transmission and distributed storage. The dynamic transmission control based on network status dynamically adjusts the frequency and amount of data transmission based on the signal quality, bandwidth, and latency fluctuations of the wireless network. When network quality is good, data can be transmitted at a higher frequency. When network quality is poor, the data sampling frequency is reduced or important data is selectively transmitted. The layered transmission and distributed storage utilize a layered transmission mechanism, prioritizing the transmission of monitoring data with high real-time requirements (e.g., pressure, temperature, flow rate, pipeline vibration data, and gas pressure fluctuations), while caching or delaying the transmission of more stable data (e.g., flow rate, gas composition, pipeline wall thickness or corrosion monitoring data, and gas composition within pipelines). Furthermore, some data can be pre-stored in the distributed storage system. When network conditions are poor, the system can select to read data from local storage as needed, ensuring data continuity and real-time performance.

[0084] Specifically, if Figure 4 As shown, the fusion module specifically includes:

[0085] A time series alignment unit, configured to process the compressed data using a timestamp interpolation method to obtain time series data;

[0086] a normalization unit, configured to process the time series data using a Z-value method to obtain normalized data;

[0087] An extraction unit, configured to perform feature extraction on the standardized data to obtain a number of key features;

[0088] The fusion unit is used to process the key features by adopting a Kalman filter algorithm or a weighted average algorithm to obtain a fusion feature.

[0089] Specifically, in the time series alignment unit, the timestamp interpolation method sets a time interval (for example, 1 minute) to determine a unified time series as the target for aligning compressed data from all sources. For the compressed data from different sensors, interpolation calculations are performed on the same time series based on the timestamps and data values. Taking linear interpolation as an example, given time points t1 and t2, the corresponding data values ​​are y1 and y2. The calculation formula for the interpolation result y at time point t between t1 and t2 is as follows:

[0090]

[0091] Specifically, in the normalization unit, the Z-score method converts the time series data into a standard normal distribution with a mean of 0 and a standard deviation of 1, thereby eliminating the dimensionality effect between different sensor data and making time series data from different sources comparable. For each source of time series data, the mean μ and standard deviation σ are calculated, where x represents the time series data. The calculation formula for the normalized data is as follows:

[0092]

[0093] Where Z represents the normalized data.

[0094] Specifically, in the fusion unit, the Kalman filter algorithm is used as an example. The system state is continuously estimated through two steps: prediction and update. First, a state transition equation is established based on the standardized data to predict the state at the next moment. The covariance of the state estimate is then predicted based on the state transition equation. The Kalman gain is calculated based on the predicted covariance and the observation noise covariance. The state estimate is updated based on the observed and predicted values. Finally, the covariance of the estimated state is updated to obtain the fused feature.

[0095] Specifically, the weighted average algorithm assigns a weight to each key feature according to its importance, performs weighted summation on multiple key features, and obtains a fusion feature.

[0096] The intelligent monitoring instrument for multi-parameter fusion analysis during natural gas pipeline operation, according to an embodiment of the present invention, effectively improves the processing accuracy of monitoring data and the accuracy of fault detection through the design of a fusion module. The time series alignment unit uses timestamp interpolation to process compressed data, resolving the issue of inconsistent data acquisition times and ensuring time alignment of multi-source data, enabling subsequent analysis to be based on precise time series information. The normalization unit uses the Z-score method to normalize time series data, eliminating dimensional differences between different monitoring parameters and improving data comparability and analytical consistency. The extraction unit performs feature extraction on the standardized data, effectively extracting key features from large amounts of data, reducing interference from irrelevant information and thus improving the efficiency of subsequent analysis. Finally, the fusion unit processes key features using a Kalman filter or weighted average algorithm, enabling comprehensive analysis of multi-dimensional information. This makes the resulting fused features more stable and accurate, further improving the accuracy of fault identification and prediction. Overall, the above technical solution, through multi-level data processing and feature fusion, provides more accurate data support and decision-making basis for real-time monitoring of natural gas pipelines, enhancing the safety and reliability of pipeline operations.

[0097] Specifically, if Figure 5 As shown, the diagnostic module specifically includes:

[0098] A random forest unit, configured to input the fusion feature into a pre-trained random forest model to obtain a first diagnosis result;

[0099] A support vector machine unit, configured to input the fusion feature into a pre-trained support vector machine model to obtain a second diagnosis result;

[0100] a long short-term memory network unit, configured to input the fusion feature into a pre-trained long short-term memory network model to obtain a third diagnosis result;

[0101] The diagnosis unit is configured to obtain a fault type by using a multi-model optimization algorithm according to the first diagnosis result, the second diagnosis result, and the third diagnosis result.

[0102] The advantages of this technical solution are that traditional fault diagnosis models typically rely on a single algorithm (such as a random forest or support vector machine) for decision-making, making it difficult to coordinate the analysis of both static and dynamic time series characteristics in pipeline data. This solution, however, achieves a breakthrough by leveraging a heterogeneous model collaborative decision-making mechanism. The random forest unit excels at capturing nonlinear correlations between high-dimensional fused features (e.g., spatial distribution anomalies in multi-sensor data), the support vector machine unit enhances the robustness of classification boundaries in small sample sizes (e.g., distinguishing marginal samples of rare fault types), and the long-short-term memory network unit leverages its time series modeling capabilities to analyze the dynamic evolution of fault characteristics (e.g., the propagation delay effect of pressure fluctuations). By fusing the diagnostic results of these three models through a multi-model optimization algorithm, the system not only mitigates misjudgments caused by single-model data distribution shifts or noise interference (e.g., random forest overfitting of transient anomalies or LSTM's delayed response to long-term trends), but also extracts complementary diagnostic evidence in complex operating conditions with multiple faults (e.g., pipe wall corrosion superimposed on pressure transients), thereby improving the interpretability and accuracy of fault type identification.

[0103] Specifically, the random forest unit, based on the random forest algorithm, is used to handle multi-classification tasks, helping the system determine the normal and abnormal states of pipelines. By constructing multiple decision trees and performing voting decisions, it can process complex and high-dimensional data, making it suitable for analyzing a variety of sensor data. During the training process, it learns from the input fusion features (e.g., temperature, pressure, or flow), ultimately obtaining a first diagnostic result (i.e., a classification result of the fault type, such as a pipeline leak or abnormal pressure).

[0104] Specifically, in the random forest unit, it specifically includes:

[0105] A training set establishment subunit, used for establishing a plurality of random forest training sets according to the fusion features;

[0106] A decision tree establishment subunit is used to establish a plurality of random forest decision trees using a random forest algorithm according to a plurality of random forest training sets, wherein the number of the random forest decision trees is the same as the number of the random forest training sets;

[0107] A decision subunit, configured to generate a plurality of decision results based on the plurality of random forest decision trees;

[0108] The diagnosis subunit is used to vote on a number of decision results using a voting algorithm to obtain a first diagnosis result.

[0109] Specifically, the support vector machine unit uses a support vector machine algorithm to perform binary classification tasks, helping the system determine whether a specific type of fault, such as a leak, exists. The support vector machine unit constructs an optimal classification hyperplane to segment data into different categories. During training, by finding the hyperplane that minimizes misclassification, it can effectively handle nonlinear data and accurately determine the fault type when inputting real-time data, outputting a binary classification result, such as whether a pipeline leak has occurred.

[0110] Specifically, the long short-term memory network unit is used to analyze the time series data in the fusion features, process the changes in sensor data in the pipeline over time, and identify long-term dependencies in the data. The long-term dependencies of the data are processed through built-in memory units and gating mechanisms, and during the training process, it learns how to identify fluctuations in data such as temperature, pressure, and flow. It is used to identify sudden changes or abnormal fluctuations in pipeline operation. For example, when the temperature continues to rise and the flow rate decreases, it predicts that faults such as thermal expansion or partial blockage may occur in the pipeline. The output results include the type of fault (for example, pipeline leakage, temperature abnormality, flow fluctuation, etc.) and possible causes of the fault (for example, equipment aging, pipeline leakage, or changes in the external environment, etc.). Based on this information, the system accurately diagnoses the fault, provides targeted maintenance suggestions, and provides support for subsequent fault location and repair work.

[0111] Specifically, the output results of the random forest unit, the support vector machine unit, and the long short-term memory network unit are not exactly the same. Typically, the outputs will differ, especially in complex systems where multiple models may produce different fault diagnosis results. The multi-model optimization algorithm uses a voting algorithm, confidence judgment method, rule-based post-processing method, model fusion method, or priority setting method to filter multiple output results to obtain accurate results.

[0112] Specifically, the voting algorithm is used for integrated learning, such as classification problems. Majority voting can be used, that is, the fault type that occurs the most times from the first diagnostic result, the second diagnostic result, and the third diagnostic result is selected as the fault type to be outputted. For example, if two of the three models predict "pipeline leakage" and one model predicts "temperature anomaly", then the final result is "pipeline leakage". A weighted voting method can also be used, by assigning different weights to the first diagnostic result, the second diagnostic result, and the third diagnostic result, and considering the weights of different diagnostic results when voting. For example, the diagnostic result output by the unit with a higher accuracy rate has a higher weight, thereby having a greater impact on the final result.

[0113] Specifically, the confidence determination method is used to generate the first, second, and third diagnostic results with a confidence value indicating the accuracy of the diagnostic result. A diagnostic result with a higher confidence value is more reliable. If the first, second, and third diagnostic results differ, the diagnostic result with the higher confidence value is selected as the fault type output by comparing the confidence values. For example, if the confidence value of the first diagnostic result indicating a pipeline leak is 0.9, the confidence value of the second diagnostic result indicating a leak is 0.7, and the confidence value of the third diagnostic result indicating a temperature anomaly is 0.8, the first diagnostic result is selected.

[0114] Specifically, the rule-based post-processing method introduces a rule engine or expert system to post-process the first diagnostic result, the second diagnostic result and the third diagnostic result, and makes a final judgment based on context information, historical data or other known rules. For example, when the first diagnostic result, the second diagnostic result and the third diagnostic result are inconsistent, but there is a diagnostic result that is the same as the historical data, the final decision is made based on the historical database.

[0115] Specifically, the model fusion method is to perform weighted summing or averaging of the first diagnostic result, the second diagnostic result and the third diagnostic result through model fusion technology, so as to obtain a comprehensive prediction result as the fault type output. For example, using a weighted average method, the first diagnostic result, the second diagnostic result and the third diagnostic result are weighted and calculated according to the weights to obtain the final diagnostic result. It can reduce the deviation of a single model and improve the robustness of the system. For example, the first diagnostic result is a pipeline leak, the second diagnostic result is a temperature anomaly, and the third diagnostic result is a pressure anomaly. The comprehensive output result is based on the performance weight of each diagnostic result (for example, accuracy or training result).

[0116] Specifically, the setting priority is used to detect a specific type of fault (for example, a long short-term memory network unit is used for timing anomaly detection). At this time, the specific type of fault only refers to the corresponding unit, and the diagnosis results of other units are not considered.

[0117] Furthermore, in an alternative embodiment, the multi-model optimization algorithm can also use a stacking method. The stacking method is to input the outputs of multiple models as new features into another learning model for training, thereby obtaining the final prediction results. The outputs of multiple basic models are first input as features into a secondary model (such as logistic regression, neural network, etc.) for comprehensive judgment. The secondary model makes the final decision based on these output features. The stacking method can better utilize the advantages of different models, especially showing good results in complex problems.

[0118] Furthermore, in other alternative embodiments, the diagnostic module may specifically include: an extreme learning machine unit, used to input the fused features into a pre-trained extreme learning machine model to obtain a first diagnostic result; a graph convolutional network unit, used to input the fused features into a pre-trained graph convolutional network model to obtain a second diagnostic result; a Transformer network unit, used to input the fused features into a pre-trained Transformer network model to obtain a third diagnostic result; and a diagnostic unit, used to obtain a fault type based on the first diagnostic result, the second diagnostic result and the third diagnostic result using a multi-model optimization algorithm.

[0119] The Extreme Learning Machine (ELM) is a new type of single-hidden-layer feedforward neural network (SLFN). Unlike traditional neural networks, the ELM training process does not require a backpropagation algorithm. Instead, it randomly initializes the hidden layer weights and uses the least squares method to train the output layer weights. ELM boasts high computational efficiency and strong generalization capabilities, excelling in processing large amounts of data. ELM can efficiently classify and diagnose fault signatures, making it particularly well-suited for real-time monitoring systems with large amounts of data.

[0120] Graph Convolutional Network (GCN) is a deep learning model based on graph-structured data that can process non-Euclidean data and is particularly suitable for learning graph-structured data. For complex network systems such as natural gas pipelines, GCN can express the relationships between pipeline nodes (such as sensor nodes) through graph structures, and extract local features of nodes through convolution operations to perform fault diagnosis. By performing convolution operations on graph data, GCN can adaptively extract more structural features and improve the accuracy of fault diagnosis. It has strong scalability and can handle large-scale node and connection data. In the fault diagnosis of natural gas pipelines, GCN can jointly analyze the data of different sensor nodes based on the topological structure of the pipeline, thereby improving the accuracy and stability of fault detection.

[0121] The Transformer network (self-attention mechanism) is a model based on the self-attention mechanism. Compared with traditional RNNs or LSTMs, the Transformer can efficiently capture long-term dependencies in sequence data through global self-attention and can process data in parallel, significantly improving computational efficiency. The Transformer can process multiple input sequences simultaneously, making it suitable for parallel processing of multidimensional sensor data in natural gas pipeline systems. Through the self-attention mechanism, the Transformer can efficiently process multidimensional sensor time series data, making it suitable for analyzing long-term pipeline monitoring data and improving the accuracy and speed of fault prediction.

[0122] Specifically, if Figure 6 As shown, the positioning module specifically includes:

[0123] A data acquisition unit, configured to acquire the position, geographic information data, and historical operation data of the sensor module;

[0124] a first positioning unit, configured to obtain a first position by using a distance calculation method according to the signal strength and the position of the sensor module;

[0125] a second positioning unit, configured to obtain a second position by using a triangulation method according to the position of the sensor module and the fault type;

[0126] a third positioning unit, configured to train a fault location model according to the historical operation data, input the fault type into the fault location model, and obtain a third location;

[0127] a fourth positioning unit, configured to obtain a fourth position by using a map marking method according to the geographic information data, the position of the sensor module and the fault type;

[0128] The position output unit is configured to output the fault position according to any one or more combinations of the first position, the second position, the third position or the fourth position.

[0129] Specifically, the geographic information data is obtained from a geographic information system, the position of the sensor module is obtained according to a design drawing, and the historical operation data is obtained from a cloud storage.

[0130] Specifically, in the first positioning unit, a distance calculation method is used to calculate the distance between the first position (i.e., the fault position) and the position of the sensor module, thereby obtaining the first position. The calculation formula of the distance d between the first position and the position of the sensor module is as follows:

[0131] d=vt;

[0132] Wherein, v is the signal propagation speed, which is obtained according to the signal strength; t is the signal propagation time, which is obtained by calculating the delay.

[0133] Specifically, in the second positioning unit, the default arrangement of sensors is uniform, the nature of the fault (for example, leakage or temperature fluctuation) is known, and the formula of the triangulation positioning method is as follows:

[0134]

[0135] Specifically, x1, y1 and x2, y2 are the known coordinates of the two sensors, d1 and d2 are the distances between the fault and the sensors, and Position is the location where the fault occurs, that is, the second location.

[0136] Specifically, in the third positioning unit, the specific working process is as follows:

[0137] Historical operating data is one of the key inputs used by the third positioning unit to predict fault locations. This data includes, but is not limited to, real-time monitoring data such as the pipeline's gas flow, pressure, and temperature, which is continuously collected and stored by the sensor module. In addition to real-time data, historical operating data also includes records of past faults, such as the type of fault, the time and location of occurrence, and the repair measures taken. Furthermore, geographic information and environmental factors (such as geology and climate change) in the area where the pipeline is located are also important historical data. By acquiring and storing this historical data, the system can provide rich context for fault location.

[0138] The construction of the fault location model relies on the fault cases and corresponding pipeline locations in historical data. First, the system cleans and preprocesses the historical data to remove noise or irrelevant data and standardizes the useful data. The system then uses machine learning algorithms (such as support vector machines, random forests, and neural networks) to train the fault location model. The model inputs include sensor data (such as gas flow, temperature, and pressure), pipeline location data, historical fault types, and their corresponding locations. Through training with machine learning algorithms, the model can identify the relationship between fault type and pipeline location, providing a foundation for future fault prediction.

[0139] Once the system detects a possible fault event, the third location unit will acquire real-time fault type data (such as gas leaks, pipeline ruptures, etc.) and input it into the trained fault location model. By passing the correlation information between the fault type and historical data to the model, the system can calculate the possible fault location (i.e., the third location) through an algorithm. For example, if the fault type is a gas leak, the model will predict the most likely leak area based on the specific location of the pipeline in historical leak events and the signal characteristics of the sensor.

[0140] To improve fault location accuracy, the third location unit not only relies on a single type of historical data but also integrates multiple data sources. This data includes not only sensor-measured information such as temperature, pressure, and flow, but also environmental factors such as the geology and climate of the pipeline's operating area. By combining this multi-dimensional data, the model can more comprehensively simulate the pipeline's behavior under specific conditions, improving the accuracy of fault location prediction.

[0141] After the fault type is input and calculated using the location model, the third location unit outputs the predicted fault location, known as the third location. This location is expressed as geographic coordinates (such as longitude and latitude) or pipeline distance. This location output provides maintenance personnel with the specific location of the fault, allowing them to quickly locate the fault area, reduce maintenance response time, and ensure efficient system operation.

[0142] Specifically, the fourth positioning unit directly marks the area where the fault occurs on the map through the geographic information system, GC = f (Sensor Position, Data Points), where Sensor Position is the marking on the map based on the position of the sensor module, Data Points is the operating status data recorded by the sensor, and GC is the location of the fault after calculation and fusion, that is, the fourth position.

[0143] In some embodiments, the specific process of obtaining the fourth position is as follows:

[0144] The operation of the fourth positioning unit is based on three types of basic data input: first, geographic information data, which contains electronic map information of the area where the target natural gas pipeline is located, specifically covering spatial reference data such as topography, road distribution, pipeline direction, coordinate grid system, etc., providing a geographic coordinate framework for subsequent location marking; second, the location information of the sensor module. Each sensor module is recorded with precise geographic coordinates (longitude, latitude, and altitude if necessary) during installation. The coordinates are obtained through calibration of the global satellite navigation system (such as GPS, Beidou) or the ground coordinate system to form a sensor location data set; finally, fault type data, which is identified and generated by the sensor module or the front-end data analysis unit, such as classification identification of different fault modes such as pressure anomaly, flow mutation, and temperature exceeding the limit.

[0145] The fourth positioning unit incorporates a built-in Geographic Information System (GIS) module. This module first loads preprocessed geographic information data and constructs an electronic map interface that includes pipeline routes and sensor distribution points. Based on spatial database technology, the system maps pipeline segment codes to geographic coordinates and annotates the actual location of each sensor module with a dedicated symbol on the map, creating a visual representation of the sensor network distribution. Upon receiving fault type data, the GIS module automatically activates the annotation process and enters fault area location mode.

[0146] For the sensor module location annotation (Sensor Position), the specific implementation steps are as follows: First, the physical installation coordinates (latitude and longitude) of each sensor are converted into map projection coordinates supported by the GIS system (such as Gauss-Krüger projection or UTM projection) to ensure the consistency of the coordinate system; second, the location of the fault-related sensor is dynamically marked on the electronic map with a specific icon (such as a red dot). If multiple sensors trigger a fault signal, the locations of all related sensors are marked simultaneously to form a sensor location set; finally, the system automatically generates a sensor location annotation layer, which is superimposed on the basic geographic information layer and supports interactive operations such as zooming and panning to facilitate subsequent data fusion analysis.

[0147] The operating status data (Data Points) recorded by the sensor include time series multi-parameter monitoring values ​​(such as pressure, flow, temperature, vibration frequency, etc.), as well as abnormal data points when the fault occurs. The fourth positioning unit first performs spatiotemporal registration on the original data, attaching the corresponding acquisition timestamp and sensor coordinate information to each data point; secondly, based on the fault type, it extracts characteristic data segments (such as waveform mutation points before and after the fault occurs, the duration of parameter exceeding the limit, etc.) to form a fault feature data set with geographic coordinates; finally, through data interpolation or spatial kriging algorithms, the discrete sensor data points are expanded into a continuous spatial distribution field, such as generating a pressure anomaly distribution cloud map, providing a data basis for the spatial positioning of the fault location.

[0148] The fourth positioning unit uses a preset fusion function GC = f (Sensor Position, Data Points) to solve the fault location. This function is constructed based on a spatial analysis algorithm, and the specific implementation logic is as follows: First, the corresponding analysis model is selected according to the fault type. For example, the pipeline fluid dynamics model is used for pressure mutation faults, and the mechanical vibration propagation model is used for vibration abnormality faults; secondly, the sensor position set is used as a spatial constraint condition, combined with the spatiotemporal distribution law of the fault feature data, and the most likely area where the fault occurred is inferred through the least squares method or Kalman filtering algorithm; finally, the geographic coordinates (GC) of the fault location are calculated. The coordinates contain a confidence interval for the position accuracy, which is used to characterize the reliability of the positioning result.

[0149] In the geographic information system interface, the fourth positioning unit marks the calculated fourth position (GC) on the electronic map with a special symbol (such as a flashing yellow triangle). The specific steps are as follows: First, a position error ellipse or buffer area is generated according to the coordinate accuracy confidence interval, and the possible range of the fault location is displayed with different levels of transparency; second, attribute information such as the fault type, occurrence time, and related sensor data are displayed in association to form an interactive fault annotation pop-up window; finally, the system automatically records the annotation history, supports the location overlay display and comparative analysis of multiple fault events, and provides an export function for location annotations (such as KML format files) to facilitate docking with external geographic information systems or emergency command systems.

[0150] Specifically, the methods for obtaining the first, second, third, and fourth positions can be used in conjunction with each other. For example, based on the signal strength and the location of the sensor module, this is applicable when signal strength (e.g., pressure, temperature, or flow) is a clue to the fault. Calibration of the signal strength can provide a preliminary range of the fault location. Triangulation can then be used based on the location of the sensor module to further precisely determine the fault area based on the locations of multiple sensors and the distance of the fault detection. Signal strength provides a rough estimate, while triangulation provides a refined spatial estimate. For another example, positioning based on historical data can be combined with a geographic information system. The historical data can help the system identify common operating patterns and changing trends, providing context for real-time data. By integrating with a geographic information system, historical data and the estimated real-time fault location can be combined on a map to provide a more intuitive and accurate fault location. In some unique areas, historical data can reveal long-unnoticed patterns, while a geographic information system can provide specific geographic coordinates to enhance positioning accuracy. Combining multiple methods, such as first determining the approximate location of the fault area through signal strength-based positioning, then further refining the location using historical data and a geographic information system, and finally confirming the specific fault point through triangulation of sensor deployment locations. This combination of methods allows for the selection of the optimal positioning method based on the specific situation, providing the most accurate positioning results in complex environments. Therefore, combining multiple methods can complement each other's strengths and weaknesses. For example, signal strength-based positioning can quickly provide a preliminary fault range, while triangulation of sensor deployment locations and historical data analysis can further improve accuracy. By combining multiple methods, the system can more accurately and reliably locate the fault area, providing effective support for subsequent maintenance work.

[0151] Specifically, if Figure 7 As shown, the early warning module specifically includes:

[0152] a label setting unit, configured to classify the fault location and the fault type to obtain a fault label;

[0153] The assignment unit is used to assign values ​​and weights to each fault label based on the historical fault library to obtain the fault value and fault weight;

[0154] a scoring unit, configured to perform weighted calculation according to the fault value and the fault weight to obtain a fault score;

[0155] A rating unit, configured to obtain a fault level according to the fault score and a preset risk level;

[0156] a report generating unit, configured to generate a fault report according to the fault level, the fault type and the fault location;

[0157] The early warning unit is used to send an early warning signal and a fault report to the maintenance personnel according to the fault level.

[0158] Specifically, the function of the label setting unit is to classify the fault location and fault type in the natural gas pipeline system and then set them. The function of the label setting unit is to perform structured classification of the fault location and fault type to generate a uniquely identified fault label. The specific process is: first, the original fault information from the detection module is received, which includes data such as the fault location (for example, pipeline section A-3 weld) and the fault type (for example, corrosion); then, according to the preset classification rule base, the fault location and fault type are standardized and mapped respectively. The location classification will classify the specific location into major categories such as pipeline section, valve, compressor station and corresponding subcategories (for example, pipeline section A) according to the hierarchy. The fault type classification will classify the fault type into major categories such as corrosion, leakage, and pressure anomaly, and associate it with subcategories (for example, electrochemical corrosion); finally, the standardized location and type are combined to generate a unique label in the format of "location major category-location subcategory_type major category-type subcategory", for example, "pipeline section-A3 weld_corrosion-electrochemical". In one example, if a gas leak is detected at the flange connection at the C outlet of the compressor station due to seal aging, the label setting unit will parse it into the location category of compressor station, the subcategory of C outlet flange, the type category of leakage, and the subcategory of seal aging, and finally generate the label "Compressor Station-C Outlet Flange_Leakage-Seal Aging".

[0159] Specifically, the function of the assignment unit is to assign values ​​and weights to different types of fault labels based on the historical fault library. The historical fault library was established through years of accumulation and statistical analysis of fault cases during the operation of natural gas pipelines. The assignment and weighting process quantifies the degree of impact of each fault label on pipeline operation under different circumstances. For example, for a fault label indicating too low pressure, the impact value of the pipeline pressure drop that may be caused by the fault can be calculated based on historical data and assigned a weight. This embodiment dynamically assigns values ​​to each fault label, and the weight is adaptively adjusted based on historical data and real-time feedback, thereby improving the accuracy of fault assessment.

[0160] In some embodiments, the value assignment unit assigns a fault value and weight to each fault tag by quantifying historical fault data. The process is as follows: First, all records related to the current fault tag are retrieved from the historical fault database to obtain parameters such as occurrence frequency, repair cost, and downtime duration. Then, based on the fault occurrence frequency and combined with the time decay factor, the dynamic fault value V is calculated using the formula:

[0161]

[0162] Where α is the time decay factor, T i is the time when the historical fault occurred, n is the total number of times the record corresponding to this fault label appears in the historical fault database, and Tc is the current time;

[0163] Finally, according to the degree of fault impact, the analytic hierarchy process (AHP) is used to calculate the comprehensive impact score of each label from the historical data, and the fault weight value is obtained after normalization.

[0164] As an alternative, the entropy weighting method is used to dynamically adjust fault weights. This dynamic weighting method objectively quantifies the importance of each evaluation metric based on the data's distribution characteristics, thereby automatically assigning fault weights. In practice, the system first extracts multi-dimensional metric data associated with the fault label (such as occurrence frequency, repair cost, downtime duration, and safety risk level) from the historical fault database and normalizes this data to eliminate dimensional differences. Subsequently, the information entropy value of each metric is calculated by analyzing the distribution dispersion of the data. If a metric exhibits significant data variability (e.g., significant fluctuations in the repair cost for certain faults), its entropy value is low, indicating that it contains more valid information and should be assigned a higher weight. Conversely, if the metric data converges (e.g., similar downtime duration for all faults), its entropy value is high, and the weight is reduced accordingly. This entire process relies entirely on the objectivity of the data and requires no manual intervention. When new fault data is added, the system automatically reassesses the entropy value and weighting of each metric, reflecting the latest data characteristics in real time. This dynamic mechanism can adapt to changes in the pipeline operating environment, such as increased seasonal corrosion or failure mode migration caused by equipment aging, ensuring that the weight distribution always matches the actual situation, thereby improving the accuracy and timeliness of the early warning model.

[0165] Specifically, the scoring unit performs a weighted calculation based on the assigned fault values ​​and fault weights to generate a fault score. Using a weighted algorithm (such as a weighted average or weighted sum), the fault values ​​and weights of each fault feature are combined to produce a comprehensive score. The scoring unit assigns different weights to different features and performs a weighted calculation based on the combined effect of these weights. This algorithm can employ a weighted average method or an improved weighted model to quantitatively assess fault risk.

[0166] Specifically, the rating unit compares the fault score with a preset risk level threshold to determine the fault's severity. The risk level is determined by multiple criteria, such as the severity of the fault type, the likelihood of occurrence, and the scope of impact. Preset risk levels are categorized into several levels, such as minor, major, critical, and extreme. Based on the fault score calculated by the rating unit, the rating unit automatically determines the severity of the fault and assigns a corresponding fault level.

[0167] Specifically, the report generation unit generates detailed fault reports based on fault severity, fault type, and location. These reports include, but are not limited to, fault type, location, impact, and emergency response recommendations, helping maintenance personnel understand the specific fault situation and take appropriate action. Detailed reports are automatically generated based on fault severity, enabling maintenance personnel to quickly locate and address the fault.

[0168] Specifically, the early warning unit sends a warning signal and fault report to maintenance personnel based on the generated fault level. Warning signals can be sent via audio and visual alarms, text messages, emails, or app notifications, ensuring maintenance personnel are notified immediately and can take corrective action.

[0169] The advantages of the above technical solution are that, whereas traditional pipeline fault early warning systems rely on fixed thresholds or linear scoring rules, they struggle to accurately distinguish risk differences in complex fault scenarios. This solution, by coupling dynamic weighting with multidimensional scoring, assigns differentiated weights and calculates scores for fault labels based on a historical fault database. This enables the system to adaptively identify the potential correlations between high-frequency, low-risk faults and low-frequency, high-risk faults. Furthermore, through a nonlinear mapping algorithm between fault scores and risk levels, the system avoids the limitations of traditional discrete grading and achieves dynamic assessment of continuous risk probabilities. This design, particularly in complex scenarios with multiple concurrent faults or sudden changes in pipeline operating parameters, can accurately capture transition states between different risk levels (e.g., the progressive deterioration characteristics of a sudden pressure change and localized corrosion), reducing the risk of misjudgment. Furthermore, the collaborative feedback mechanism between the fault report generation unit and the early warning signal intelligently associates fault levels with pipeline section operating parameters and maintenance priorities, achieving closed-loop optimization from fault diagnosis to emergency response. This significantly improves the efficiency of maintenance resource scheduling and demonstrates synergy that cannot be achieved with traditional single-threshold early warning mechanisms.

[0170] Example 2

[0171] like Figure 8 As shown, this embodiment provides an intelligent monitoring method for multi-parameter fusion analysis in natural gas pipeline operation, which specifically includes the following steps:

[0172] S10: obtaining raw data and signal strength;

[0173] S20: Preprocessing the original data to obtain preprocessed data, and applying an adaptive compression algorithm to the preprocessed data to obtain compressed data;

[0174] S30: applying a multi-source data fusion algorithm to the compressed data to obtain fusion features;

[0175] S40: Inputting the fusion feature into a fault diagnosis model based on a machine learning algorithm, and outputting the fault type when a fault is diagnosed;

[0176] S50: Acquire the position of the sensor module, geographic information data near the natural gas pipeline, and historical operation data of the natural gas pipeline, and determine the fault location using a positioning algorithm based on the signal strength, the position of the sensor module, the geographic information data, the historical operation data, and the fault type;

[0177] S60: Generate a fault report according to the fault location and the fault type, issue a warning signal, and upload the fault report.

[0178] Specifically, the step S20 includes the following steps:

[0179] S21: removing duplicate data from the original data to obtain first data;

[0180] S22: using an anomaly detection algorithm to remove outliers in the first data to obtain second data;

[0181] S23: Acquire historical second data of the previous period, generate a threshold interval using a threshold generation algorithm based on the historical second data, and remove values ​​that do not belong to the threshold interval from the second data to obtain preprocessed data;

[0182] S24: applying an adaptive compression algorithm to the pre-processed data to obtain compressed data;

[0183] The adaptive compression algorithm includes an adaptive selection algorithm based on variation amplitude, an adaptive compression algorithm based on compression error, and an adaptive compression algorithm based on machine learning.

[0184] Specifically, step S30 includes the following steps:

[0185] S31: Processing the compressed data using a timestamp interpolation method to obtain time series data;

[0186] S32: Processing the time series data using the Z-value method to obtain standardized data;

[0187] S33: performing feature extraction on the standardized data to obtain several key features;

[0188] S34: Using a Kalman filter algorithm or a weighted average algorithm to process the key features to obtain a fusion feature.

[0189] Specifically, the step S40 includes the following steps:

[0190] S41: Inputting the fusion feature into a pre-trained random forest model to obtain a first diagnosis result;

[0191] A support vector machine unit, configured to input the fusion feature into a pre-trained support vector machine model to obtain a second diagnosis result;

[0192] S42: Inputting the fusion feature into a pre-trained long short-term memory network model to obtain a third diagnosis result;

[0193] S43: Obtain a fault type using a multi-model optimization algorithm according to the first diagnostic result, the second diagnostic result, and the third diagnostic result.

[0194] Specifically, the step S50 includes the following steps:

[0195] S51: Acquire the position, geographic information data and historical operation data of the sensor module;

[0196] S52: Obtain a first position using a distance calculation method according to the signal strength and the position of the sensor module;

[0197] S53: Obtain a second position using a triangulation method according to the position of the sensor module and the fault type;

[0198] S54: training a fault location model according to the historical operation data, inputting the fault type into the fault location model, and obtaining a third location;

[0199] S55: Obtaining a fourth position using a map marking method according to the geographic information data, the position of the sensor module, and the fault type;

[0200] S56: Output the fault location according to any one or more combinations of the first position, the second position, the third position or the fourth position.

[0201] Specifically, step S60 includes the following steps:

[0202] S61: Classify the fault location and the fault type to obtain a fault label;

[0203] S62: Assign values ​​and weights to each fault label based on the historical fault database to obtain a fault value and a fault weight;

[0204] S63: Perform weighted calculation according to the fault value and the fault weight to obtain a fault score;

[0205] S64: Obtaining a fault level according to the fault score and a preset risk level;

[0206] S65: Generate a fault report according to the fault level, the fault type, and the fault location;

[0207] S66: Sending a warning signal and a fault report to the maintenance personnel according to the fault level.

[0208] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0209] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the program implements the intelligent monitoring method for multi-parameter fusion analysis in the operation of a natural gas pipeline as described above.

[0210] If the integrated module / 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 present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program, when executed by the processor, can implement the steps of the above-mentioned various method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. Of course, there are other ways of readable storage media, such as quantum memory, graphene memory, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0211] The present invention also provides an electronic device. The electronic device in an embodiment of the present invention includes: one or more processors; and a storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the intelligent monitoring method for multi-parameter fusion analysis in natural gas pipeline operation provided by the present invention.

[0212] Reference below Figure 9 , which shows a schematic structural diagram of a computer system 800 of an electronic device suitable for implementing an embodiment of the present invention. Figure 9 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0213] like Figure 9 As shown, the computer system 800 includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage unit 808 into a random access memory (RAM) 803. Various programs and data required for the operation of the computer system 800 are also stored in the RAM 803. The CPU 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0214] The following components are connected to the I / O interface 805: an input section 806 including a keyboard, a mouse, and the like; an output section 807 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 808 including a hard disk; and a communication section 809 including a network interface card such as a LAN card or a modem. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 810 as needed, so that computer programs read therefrom can be installed in the storage section 808 as needed.

[0215] In particular, according to embodiments disclosed herein, the processes described in the main step diagrams above can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for executing the methods shown in the main step diagrams. In the above embodiments, the computer program can be downloaded and installed from a network via the communication section 809 and / or installed from removable media 811. When the computer program is executed by the central processing unit 801, the above-described functions defined in the system of the present invention are performed.

[0216] It should be noted that the computer-readable medium described in the present invention may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present invention, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical cable, RF, or any suitable combination thereof.

[0217] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0218] The units involved in the embodiments of the present invention may be implemented in software or hardware. The units described may also be provided in a processor. For example, a processor may include a pre-response unit, a receiving unit, and a requesting unit.

[0219] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. An intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation, characterized in that: include: A sensor module is installed on the natural gas pipeline to obtain raw data and signal strength; A preprocessing module, configured to preprocess the original data to obtain preprocessed data, and apply an adaptive compression algorithm to the preprocessed data to obtain compressed data; A fusion module, configured to obtain fusion features by applying a multi-source data fusion algorithm to the compressed data; a diagnosis module, configured to input the fusion features into a fault diagnosis model based on a machine learning algorithm, and output a fault type when a fault is diagnosed; a positioning module, configured to obtain the position of the sensor module, geographic information data near the natural gas pipeline, and historical operation data of the natural gas pipeline, and to determine the fault location using a positioning algorithm based on the signal strength, the position of the sensor module, the geographic information data, the historical operation data, and the fault type; The early warning module is used to generate a fault report according to the fault location and the fault type, issue an early warning signal, and upload the fault report.

2. The intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to claim 1 is characterized in that: The preprocessing module specifically includes: a first preprocessing unit, configured to remove duplicate data from the original data to obtain first data; a second preprocessing unit, configured to remove outliers from the first data using an anomaly detection algorithm to obtain second data; a filtering unit, configured to obtain historical second data of a previous period, generate a threshold interval using a threshold generation algorithm based on the historical second data, and remove values ​​not belonging to the threshold interval from the second data to obtain preprocessed data; The compression unit is used to apply an adaptive compression algorithm to the pre-processed data to obtain compressed data.

3. The intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to claim 2 is characterized in that: The adaptive compression algorithm includes an adaptive selection algorithm based on variation amplitude, an adaptive compression algorithm based on compression error, and an adaptive compression algorithm based on machine learning.

4. The intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to claim 1 is characterized in that: The fusion module specifically includes: A time series alignment unit, configured to process the compressed data using a timestamp interpolation method to obtain time series data; a normalization unit, configured to process the time series data using a Z-value method to obtain normalized data; An extraction unit, configured to perform feature extraction on the standardized data to obtain a number of key features; The fusion unit is used to process the key features by adopting a Kalman filter algorithm or a weighted average algorithm to obtain a fusion feature.

5. The intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to claim 1 is characterized in that: The diagnostic module specifically includes: A random forest unit, configured to input the fusion feature into a pre-trained random forest model to obtain a first diagnosis result; A support vector machine unit, configured to input the fusion feature into a pre-trained support vector machine model to obtain a second diagnosis result; a long short-term memory network unit, configured to input the fusion feature into a pre-trained long short-term memory network model to obtain a third diagnosis result; The diagnosis unit is configured to obtain a fault type by using a multi-model optimization algorithm according to the first diagnosis result, the second diagnosis result, and the third diagnosis result.

6. The intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to claim 1 is characterized in that: The positioning module specifically includes: A data acquisition unit, configured to acquire the position, geographic information data, and historical operation data of the sensor module; a first positioning unit, configured to obtain a first position by using a distance calculation method according to the signal strength and the position of the sensor module; a second positioning unit, configured to obtain a second position by using a triangulation method according to the position of the sensor module and the fault type; a third positioning unit, configured to train a fault location model according to the historical operation data, input the fault type into the fault location model, and obtain a third location; a fourth positioning unit, configured to obtain a fourth position by using a map marking method according to the geographic information data, the position of the sensor module and the fault type; The position output unit is configured to output a fault position according to any combination of the first position, the second position, the third position, or the fourth position.

7. The intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to claim 1 is characterized in that: The early warning module specifically includes: a label setting unit, configured to classify the fault location and the fault type to obtain a fault label; The assignment unit is used to assign values ​​and weights to each fault label based on the historical fault library to obtain the fault value and fault weight; a scoring unit, configured to perform weighted calculation according to the fault value and the fault weight to obtain a fault score; A rating unit, configured to obtain a fault level according to the fault score and a preset risk level; a report generating unit, configured to generate a fault report according to the fault level, the fault type and the fault location; The early warning unit is used to send an early warning signal and a fault report to the maintenance personnel according to the fault level.

8. The intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to claim 5 is characterized in that: The multi-model optimization algorithm includes any one of a voting algorithm, a confidence judgment method, a rule-based post-processing method, a model fusion method or a priority setting method; The model fusion method obtains a final diagnosis result by weighted summing or averaging the first diagnosis result, the second diagnosis result, and the third diagnosis result, and obtains the fault type according to the final diagnosis result.

9. The intelligent monitoring instrument for multi-parameter fusion analysis in natural gas pipeline operation according to claim 1 is characterized in that: The sensor module includes any combination of acoustic wave sensors, gas sensors, optical fiber sensors, gas quality sensors, radar sensors, infrared thermal imaging sensors and environmental radiation sensors.

10. An intelligent monitoring method for multi-parameter fusion analysis in natural gas pipeline operation, characterized in that: The following steps are involved: S10: obtaining raw data and signal strength; S20: Preprocessing the original data to obtain preprocessed data, and applying an adaptive compression algorithm to the preprocessed data to obtain compressed data; S30: applying a multi-source data fusion algorithm to the compressed data to obtain fusion features; S40: Inputting the fusion feature into a fault diagnosis model based on a machine learning algorithm, and outputting the fault type when a fault is diagnosed; S50: Acquire the location of the sensor module, geographic information data near the natural gas pipeline, and historical operation data of the natural gas pipeline, and determine the fault location using a positioning algorithm based on the signal strength, the location of the sensor module, the geographic information data, the historical operation data, and the fault type; S60: Generate a fault report according to the fault location and the fault type, issue a warning signal, and upload the fault report.

Images