Gas pipeline safety monitoring system and method based on data analysis

Abstract

The invention relates to a gas pipeline safety monitoring system and method based on data analysis, and relates to the technical field of gas pipeline safety, and the system comprises a data acquisition module which is used for collecting the pressure, temperature and flow in a gas pipeline and the gas concentration of the surrounding environment; the abnormity identification module is used for analyzing the pipeline data of the sensor in real time and identifying the abnormal condition of the gas pipeline through an abnormity detection discrimination function; the model building module is used for analyzing and predicting future data of the gas pipeline and building a safety monitoring model based on historical data and real-time data of the gas pipeline; the data adjusting module is used for adjusting the pipeline data of the gas pipeline after the gas pipeline is detected to be abnormal or the potential risk is predicted; the model training module is used for obtaining an optimized safety monitoring model; and the safety monitoring module is used for carrying out safety monitoring on the gas pipeline according to the optimized safety monitoring model. The system can improve the accuracy and efficiency of gas pipeline safety monitoring.

Description

Technical Field

[0001] The present invention relates to the field of gas pipeline safety technology, and in particular to a gas pipeline safety monitoring system, method, electronic equipment and non-transient computer-readable storage medium based on data analysis. Background Art

[0002] Nowadays, the safety monitoring of gas pipelines is an important part of ensuring public safety and stable energy supply. At present, traditional gas pipeline monitoring methods mainly rely on regular inspections, manual inspections and installation of fixed sensors. These methods can detect external damage, corrosion and leakage of pipelines, and ensure the normal operation of pipelines through manual inspections or regular maintenance. In addition, some areas have also adopted buried sensors and gas detectors to warn of potential safety hazards by real-time monitoring of changes in gas concentrations in the environment around the pipeline.

[0003] However, the frequency of regular inspections and manual checks is low, making it difficult to detect sudden leaks or damage in real time, posing a major safety hazard. Secondly, fixed sensors are expensive to install and maintain, with limited coverage, making it difficult to fully monitor all key nodes of the pipeline. Finally, traditional methods rely on manual analysis and judgment, are easily affected by human factors, and are difficult to achieve efficient and accurate safety monitoring. Summary of the invention

[0004] In view of the technical problems existing in the prior art, the present invention provides a gas pipeline safety monitoring system, method, electronic device and non-transitory computer-readable storage medium based on data analysis, which can improve the accuracy and efficiency of gas pipeline safety monitoring.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] The present invention provides a gas pipeline safety monitoring system based on data analysis, the system comprising:

[0007] A data acquisition module is used to install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration of the surrounding environment;

[0008] An abnormality identification module, used for analyzing the pipeline data of the sensor in real time and identifying the abnormality of the gas pipeline through an abnormality detection discriminant function;

[0009] A model building module, used to predict the future data of the gas pipeline through time series analysis, and to establish a safety monitoring model based on the historical data and real-time data of the gas pipeline;

[0010] A data adjustment module, used to adjust the pipeline data of the gas pipeline after detecting an abnormality of the gas pipeline or predicting a potential risk;

[0011] A model training module, used to continue monitoring the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model;

[0012] A safety monitoring module is used to perform safety monitoring on the gas pipeline according to the optimized safety monitoring model.

[0013] Optionally, the abnormality identification module is further used to:

[0014] Acquire different pipeline data of the gas pipeline at a preset time;

[0015] According to different pipeline data of the gas pipeline at the current time t, calculate the average value and standard deviation of each type of pipeline data;

[0016] Obtaining the weight of the covariance matrix between each of the pipeline data and the corresponding adaptive weighting coefficient;

[0017] Constructing the abnormality detection discriminant function according to different pipeline data of the gas pipeline at the current time t, the average value, standard deviation of each type of pipeline data, the weight of the covariance matrix and the adaptive weighting coefficient;

[0018] An abnormality detection value of the gas pipeline is determined according to the abnormality detection discriminant function, and abnormality detection is performed on the gas pipeline based on the abnormality detection value.

[0019] Optionally, the anomaly detection function is expressed as:

[0020]

[0021] Among them, X i (t), X j (t), X k (t) represents different pipeline data at time t, μ i , μ j , μ k are the average values ​​of various pipeline data, σ i , σ j , σ k is the standard deviation of various pipeline data, W ij is the weight of the covariance matrix between each pipeline data, δ k is the adaptive weighting coefficient.

[0022] Optionally, the model building module is further used to:

[0023] Get the pipeline data at the current time t and the historical pipeline data before time h;

[0024] Obtaining exogenous variables generated by external factors of the gas pipeline;

[0025] Obtain the lagged error term of the pipeline data before time s;

[0026] The safety monitoring model for predicting future data of the gas pipeline is constructed based on the pipeline data at the current time t, the historical pipeline data before time h, the exogenous variables and the lagged error term.

[0027] Optionally, the future data of the gas pipeline is expressed as:

[0028]

[0029] in, is the autoregressive coefficient, t, h, s are the time, Z v (t) is an exogenous variable, λ v is the weight coefficient of the exogenous variable, η1 is the acceleration coefficient, θ s is the moving average coefficient, ∈(ts) is the lagged error term, is the second-order derivative, γ is the velocity coefficient, is the first-order derivative, ∈ t is a random error term, D(t) is the pipeline data at the current time t, and D(th) represents the historical pipeline data at the previous h time points.

[0030] Optionally, the data adjustment module is further used for:

[0031] Obtaining the current value and set value of each of the pipeline data;

[0032] Calculate a first difference between a current value and a set value of each pipeline data;

[0033] Get the feedforward variable of each exogenous variable at a delay of τ j The variable value after time;

[0034] The plurality of first difference values ​​are summed, integrated and differentiated respectively, and combined with each of the variable values ​​to obtain the adjusted pipeline data.

[0035] Optionally, the adjusted pipeline data is expressed as:

[0036]

[0037] Among them, X i (t) is the current value of the i-th pipeline data, X i,setpoint Yes Xi (t) setting value, K p,i , K i,i , K d,i are proportional, integral, and differential coefficients, τ j It's time, z j (t+τ j ) is the feedforward variable at a delay of τ j The value after j is the feedforward compensation coefficient, N is the total number of pipeline data, X i (t)-X i,setpoint is the first difference.

[0038] Optionally, the model training module is also used to:

[0039] Calculating the mean and standard deviation of the Gaussian kernel in the safety monitoring model;

[0040] Constructing a corresponding Gaussian kernel function according to the mean and the standard deviation;

[0041] Gradient descent and Bayesian optimization are performed on old parameters of the safety monitoring model according to the monitoring data, the loss function of the safety monitoring model and the Gaussian kernel function to obtain the optimized safety monitoring model with updated parameters.

[0042] Optionally, the update parameters of the safety monitoring model are expressed as:

[0043]

[0044] Where η2 is the learning rate, Q is the number of Gaussian kernels, and θ old are the old parameters of the model, θ new is the update parameter of the model, L(θ) is the loss function, σ is the uncertainty, n is the current number of iterations, and α k is the Bayesian optimization weight of the kth Gaussian kernel, μ k , σ k are the mean and standard deviation of the kth Gaussian kernel.

[0045] The present invention also provides a gas pipeline safety monitoring method based on data analysis, the method comprising:

[0046] Install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration in the surrounding environment;

[0047] Analyze the pipeline data of the sensor in real time, and identify the abnormal situation of the gas pipeline through an abnormal detection discriminant function;

[0048] Predicting future data of the gas pipeline through time series analysis, and establishing a safety monitoring model based on historical data and real-time data of the gas pipeline;

[0049] After detecting an abnormality of the gas pipeline or predicting a potential risk, adjusting pipeline data of the gas pipeline;

[0050] Continue to monitor the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model;

[0051] The gas pipeline is safety monitored according to the optimized safety monitoring model.

[0052] In addition, to achieve the above objectives, the present invention also proposes an electronic device, comprising: a memory for storing computer software programs; a processor for reading and executing the computer software programs, thereby implementing a gas pipeline safety monitoring method based on data analysis as described above.

[0053] In addition, to achieve the above-mentioned purpose, the present invention also proposes a non-transitory computer-readable storage medium, in which a computer software program is stored. When the computer software program is executed by a processor, it implements a gas pipeline safety monitoring method based on data analysis as described above.

[0054] The beneficial effects of the present invention are:

[0055] (1) The present invention can monitor the key parameters of the gas pipeline (such as pressure, temperature, flow, etc.) in real time by introducing movable sensors and multivariate data analysis. Compared with the traditional method that relies on manual inspection and fixed sensors, this solution greatly improves the real-time monitoring and effectively reduces the possibility of safety hazards.

[0056] (2) The present invention adopts a multivariate anomaly detection function, combined with high-order moments, covariance matrix and adaptive weighting, which can more accurately detect minor anomalies and reduce the probability of missed reports and false alarms. This high-precision detection method can identify potential safety risks in a timely manner and improve the overall safety of the pipeline system.

[0057] (3) The present invention reduces the possibility of human misoperation and misjudgment through automated and intelligent monitoring and analysis methods, thereby improving the overall reliability and stability of the system.

[0058] In summary, the present invention combines advanced technologies such as real-time monitoring, high-precision anomaly detection, intelligent prediction and early warning, automated response, self-learning and optimization, which significantly improves the safety monitoring capability of gas pipelines, reduces the risks and costs of system operation, and provides more effective protection for the safe operation of gas pipelines. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A scene diagram of a gas pipeline safety monitoring method based on data analysis provided by the present invention;

[0060] Figure 2 A flow chart of a gas pipeline safety monitoring method based on data analysis provided by the present invention;

[0061] Figure 3 A schematic diagram of the structure of a gas pipeline safety monitoring system based on data analysis provided by the present invention;

[0062] Figure 4 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0063] Figure 5 A schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0065] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0066] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0067] See also Figure 1 , Figure 1 A scenario diagram of a gas pipeline safety monitoring method based on data analysis provided by the present invention. Figure 1 As shown, the terminal and the server are connected via a network, such as a wired or wireless network connection. The terminal may include but is not limited to portable terminals such as mobile phones and tablets installed with various network platform applications, as well as fixed terminals such as computers, query machines, and advertising machines. The server provides users with various business services, including service push servers, user recommendation servers, etc.

[0068] It should be noted that Figure 1 The scenario diagram of a gas pipeline safety monitoring method based on data analysis shown is only an example. The terminal, server and application scenario described in the embodiment of the present invention are for more clearly illustrating the technical solution of the embodiment of the present invention, and do not generate limitations on the technical solution provided by the embodiment of the present invention. Ordinary technicians in this field can know that with the evolution of the system and the emergence of new business scenarios, the technical solution provided by the embodiment of the present invention is also applicable to similar technical problems.

[0069] Among them, the terminal can be used for:

[0070] Install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration in the surrounding environment;

[0071] Analyze the pipeline data of the sensor in real time, and identify the abnormal situation of the gas pipeline through an abnormal detection discriminant function;

[0072] Predicting future data of the gas pipeline through time series analysis, and establishing a safety monitoring model based on historical data and real-time data of the gas pipeline;

[0073] After detecting an abnormality of the gas pipeline or predicting a potential risk, adjusting pipeline data of the gas pipeline;

[0074] Continue to monitor the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model;

[0075] The gas pipeline is safety monitored according to the optimized safety monitoring model.

[0076] See also Figure 2 , Figure 2 A structural schematic diagram of a gas pipeline safety monitoring system based on data analysis provided by the present invention.

[0077] like Figure 2As shown, a gas pipeline safety monitoring system based on data analysis proposed in an embodiment of the present invention includes:

[0078] The data acquisition module 201 is used to install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration of the surrounding environment.

[0079] The present invention can monitor various parameters inside and around the pipeline, i.e., pipeline data, in real time and accurately by installing movable sensors at key nodes of the gas pipeline. In some embodiments, the pipeline data includes the pressure, temperature, flow rate inside the pipeline, and the gas concentration in the environment around the pipeline. By collecting and analyzing these data, the safety monitoring capability of the gas pipeline can be effectively improved to prevent accidents.

[0080] In some embodiments, key nodes in a gas pipeline generally refer to locations in the pipeline system that are prone to failure and require intensive monitoring, such as pipeline connections, bends, pressure change points, flow rate adjustment points, etc. The operating status of these nodes is critical to the safety of the entire pipeline system, and therefore requires special attention and monitoring.

[0081] Compared with fixed sensors, movable sensors have greater flexibility and can be moved and redeployed between different nodes according to actual needs. This flexibility enables the monitoring system to quickly respond to new situations or emergencies that occur during pipeline operation.

[0082] By moving sensors between different key nodes, the monitoring coverage can be effectively expanded to ensure that all key parts of the pipeline system can be monitored in real time. This method can reduce monitoring blind spots and improve overall monitoring efficiency.

[0083] Compared to installing fixed sensors in all possible locations, the use of movable sensors can significantly reduce the cost of sensor installation and maintenance. Only a small number of sensors are needed to achieve extensive monitoring, thus saving costs while ensuring monitoring effectiveness.

[0084] The following is a detailed explanation of several pipeline data of the present invention. Pressure: The pressure inside the gas pipeline is one of the key parameters reflecting the operating status of the pipeline. Too high or too low pressure may lead to safety hazards, such as pipeline burst or gas leakage. By monitoring the pressure changes in real time, the system can detect abnormalities in time and make adjustments.

[0085] Temperature: Temperature changes can affect the fluidity of gas and the structural stability of the pipeline. Abnormal temperature fluctuations may cause changes in gas flow rate or stress changes in pipeline materials, causing safety issues. Real-time temperature monitoring can help identify potential risks.

[0086] Flow rate: The flow rate of gas is directly related to the transmission efficiency and safety of the pipeline system. Abnormal flow rate usually indicates that there may be leakage, blockage or other failures in the system. Therefore, real-time flow monitoring is crucial to ensure the normal operation of the pipeline.

[0087] Gas concentration: Monitoring the gas concentration in the environment around the pipeline, especially the concentration of combustible gases, can detect signs of pipeline leakage early. By monitoring the gas concentration, the system can provide early warning to prevent leakage accidents.

[0088] In the above manner, the present invention installs movable sensors at each key node of the gas pipeline, and can build a flexible and efficient safety monitoring system by collecting the pressure, temperature, flow rate and gas concentration data of the pipeline. This system can not only monitor the operation status of the pipeline in real time, but also timely warn of possible safety hazards through data analysis and prediction, and ultimately improve the safety and operation efficiency of the gas pipeline.

[0089] The abnormality identification module 202 is used to analyze the pipeline data of the sensor in real time and identify the abnormal situation of the gas pipeline through an abnormality detection discriminant function.

[0090] In some embodiments, the anomaly identification module 202 is further configured to:

[0091] Acquire different pipeline data of the gas pipeline at a preset time;

[0092] According to different pipeline data of the gas pipeline at the current time t, calculate the average value and standard deviation of each type of pipeline data;

[0093] Obtaining the weight of the covariance matrix between each of the pipeline data and the corresponding adaptive weighting coefficient;

[0094] Constructing the abnormality detection discriminant function according to different pipeline data of the gas pipeline at the current time t, the average value, standard deviation of each type of pipeline data, the weight of the covariance matrix and the adaptive weighting coefficient;

[0095] An abnormality detection value of the gas pipeline is determined according to the abnormality detection discriminant function, and abnormality detection is performed on the gas pipeline based on the abnormality detection value.

[0096] Among them, the anomaly detection function can be expressed as:

[0097]

[0098] Among them, X i (t), X j (t), X k (t) represents different pipeline data at time t, μj , μ j , μ k are the average values ​​of various pipeline data, σ i , σ j , σ k is the standard deviation of various pipeline data, W ij is the weight of the covariance matrix between each pipeline data, δ k is the adaptive weighting coefficient.

[0099] In the specific implementation, F anomaly (t) is the anomaly detection value of the anomaly detection function.

[0100] X i (t), X j (t), X k (t) represents different pipeline data at time t, which can also be called pipeline parameters or monitoring data. It represents the actual observed values ​​of different monitoring parameters (such as pressure, temperature, flow, etc.) at time t. These parameters can come from multiple sensors in the pipeline, and each parameter reflects the status of a certain aspect of the pipeline system.

[0101] μ i , μ j , μ k are the average values ​​of the pipeline data, respectively, and represent the pipeline data X i (t), X j (t), X k The historical average value of (t). The historical average value is usually obtained by statistical calculation of monitoring data over a period of time, reflecting the expected value of each parameter under normal operating conditions.

[0102] σ i , σ j , σ k is the standard deviation of each pipeline data, describing the volatility and dispersion of each pipeline data. The larger the standard deviation, the more drastic the fluctuation of the data.

[0103] W ij is the weight of the covariance matrix between parameters. The covariance matrix reflects the relationship between two parameters. The weight coefficient W ij Indicates the importance of the relationship between two parameters in anomaly detection. The choice of weight will affect the sensitivity of overall anomaly detection.

[0104] δ k is an adaptive weighting coefficient used to control the influence of higher-order moments (such as skewness) in anomaly detection. The adaptive weighting coefficient can be automatically adjusted according to the actual situation to ensure that the formula can maintain high detection accuracy in various environments.

[0105] Specifically, (Xi (t)-μ i ) and (X j (t)-μ j ) represents the parameter X i (t), X j (t) Deviations from their respective means. The degree to which two parameters deviate from their means simultaneously is calculated by multiplying them, revealing their coordinated changes.

[0106] σ i , σ j is a normalization factor used to eliminate the dimensional effects of different parameters so that the covariances between different parameters can be directly compared.

[0107] W ij is to give certain parameters a greater influence on anomaly detection. For example, if the correlation between certain parameters is considered very important in historical data, then W ij will be higher, thus enhancing their anomaly detection value F anomaly (t) contribution.

[0108] E[(X k )t)-μ k ) 3 ] indicates parameter X k The third moment of (t) is the skewness. The third moment can reflect the asymmetry of the data. If the skewness is large, it means that the data distribution deviates far from the mean and there may be anomalies. is a normalization factor used to make the third-order moment dimensionless for comparison with other parameters. k Used to control the impact of skewness on overall anomaly detection. For some sensitive monitoring parameters, δ k Can be set higher to enhance their role in anomaly detection.

[0109] This formula combines the covariance weighted part and the high-order moment weighted part to capture the linear correlation and asymmetry between parameters, making anomaly detection more comprehensive and accurate. The square root is taken to make the anomaly detection value F anomaly The result of (t) is consistent with the dimension of the original data, and it is also helpful to define a reasonable threshold to judge anomalies.

[0110] In summary, the multivariate analysis and high-order moment analysis of the present invention provide an efficient anomaly detection method. Its advantage is that it can simultaneously consider the correlation between multiple parameters and the deviation of data distribution, so as to more accurately identify anomalies that may occur during the gas pipeline monitoring process. This method is particularly suitable for real-time monitoring in a multivariate and big data environment, and can significantly improve the safety and stability of the system.

[0111] The model building module 203 is used to predict the future data of the gas pipeline through time series analysis and to establish a safety monitoring model based on the historical data and real-time data of the gas pipeline.

[0112] In some embodiments, the model building module 203 is further configured to:

[0113] Get the pipeline data at the current time t and the historical pipeline data before time h;

[0114] Obtaining exogenous variables generated by external factors of the gas pipeline;

[0115] Obtain the lagged error term of the pipeline data before time s;

[0116] The safety monitoring model for predicting future data of the gas pipeline is constructed based on the pipeline data at the current time t, the historical pipeline data before time h, the exogenous variables and the lagged error term.

[0117] In some embodiments, the future data of the gas pipeline is represented as:

[0118]

[0119] in, is the autoregressive coefficient, t, h, s are the time, Z v (t) is an exogenous variable, λ v is the weight coefficient of the exogenous variable, η1 is the acceleration coefficient, θ s is the moving average coefficient, ∈(ts) is the lagged error term, is the second-order derivative, γ is the velocity coefficient, is the first-order derivative, ∈ t is a random error term, D(t) is the pipeline data at the current time t, and D(th) represents the historical pipeline data at the previous h time points.

[0120] In the specific implementation, is the autoregressive coefficient, which is used to describe the impact of historical data on the current forecast. The autoregressive coefficient determines the weight of the data at the previous h moments on the forecast value. Specifically, the size of the coefficient reflects the contribution of historical data to the future. t, h, and s are all moments.

[0121] Z v (t) is an exogenous variable, which represents the influence of external factors (such as ambient temperature, pipeline load changes, etc.) at time t. These external factors may have a significant impact on the state of the gas pipeline and therefore need to be included in the prediction model.

[0122] λ vis the weight coefficient of the exogenous variable, which is used to control the exogenous variable Z v (t) The degree of influence on the forecast results. Different exogenous variables may have different influences on the forecast results under different circumstances. The weight coefficient λ v The size of these effects can be adjusted.

[0123] η1 is the acceleration coefficient, which is used to measure the contribution of the acceleration of data change (i.e., the second-order derivative) to the prediction.

[0124] θ s is the moving average coefficient, which is used to smooth the impact of the lagged error term. It quantifies the contribution of the lagged error term ∈(ts) to the forecast result. The moving average coefficient can capture the impact of historical errors on the current forecast.

[0125] ∈(ts) is the lagged error term, which represents the forecast error at the previous s time points. The lagged error term is used to adjust the forecast model to take into account the impact of historical forecast errors, thereby improving the robustness of the model.

[0126] It is the second-order derivative, which indicates the accelerating trend of data change. It reflects how the speed of data change changes over time and is a key factor in predicting rapidly changing scenarios.

[0127] γ is the speed coefficient, which is used to measure the contribution of the speed of data change (i.e., the first-order derivative) to the prediction. The speed coefficient γ determines the impact of the first-order derivative on the predicted value.

[0128] It is the first-order derivative, which indicates how fast the data changes over time. This term is used to capture the linear trend of data and can predict how the data changes steadily over time.

[0129] ∈ is a random error term that captures random fluctuations or unmodeled factors in the forecast. This term reflects the uncertainty in the model and is usually unavoidable, and random errors in the forecast results must be taken into account.

[0130] D(th) represents the actual observation data at the previous h moments. By analyzing historical data over a period of time (i.e., multiple measurements in the past), the state at the current or future moment can be estimated.

[0131] Specifically, The autoregressive part considers the observed data D(th) at the historical moment th and uses the autoregressive coefficient To contribute to the prediction of the current time t+Δt. The role of this part is to make predictions based on historical trends. For example, if a parameter has a clear trend in the past few times, the autoregressive part can accurately reflect this trend and continue it to future predictions.

[0132] The moving average part considers the lagged error term ∈(ts) at the past time s and uses the moving average coefficient θ s These errors are smoothed and the current forecast value is adjusted. This part corrects the forecast deviation by introducing historical errors, making the forecast model more robust.

[0133] The exogenous variable part introduces external factors Z that may affect the prediction results. v (t), through the weight coefficient λ v , the impact of these external factors on the predicted value can be adjusted. This part is very important because the operation of the gas pipeline is not only affected by internal conditions, but may also be disturbed by changes in the external environment. The introduction of exogenous variables makes the prediction more comprehensive and accurate.

[0134] This part combines the first derivative of the data and the second-order derivative Separately capture the speed and acceleration changes of the data. Through the speed coefficient γ and the acceleration coefficient η1, the impact of these change rates on the predicted value can be flexibly adjusted, which is especially suitable for dealing with nonlinear changes. This part can better capture the fast-changing dynamic characteristics in the pipeline system and improve the accuracy of the prediction.

[0135] Through the above methods, the present invention provides a complex and flexible time series prediction model through multiple combinations of autoregression, moving average, exogenous variables and derivative analysis. This model can fully consider the trend of historical data, changes in the external environment and the dynamic change characteristics of the data itself when predicting the future state of the gas pipeline, thereby improving the accuracy and reliability of the prediction. This prediction method is particularly suitable for complex environments such as gas pipeline systems that are affected by multiple factors, and helps to timely warn of potential risks and ensure the safe operation of the pipeline.

[0136] The data adjustment module 204 is used to adjust the pipeline data of the gas pipeline after detecting an abnormality of the gas pipeline or predicting a potential risk.

[0137] In some embodiments, the data adjustment module 204 is further configured to:

[0138] Obtaining the current value and set value of each of the pipeline data;

[0139] Calculate a first difference between a current value and a set value of each pipeline data;

[0140] Get the feedforward variable of each exogenous variable at a delay of τ j The variable value after time;

[0141] The plurality of first difference values ​​are summed, integrated and differentiated respectively, and combined with each of the variable values ​​to obtain the adjusted pipeline data.

[0142] In some embodiments, the adjusted pipeline data is represented as:

[0143]

[0144] Among them, X i (t) is the current value of the i-th pipeline data, X i,setpoint Yes X i (t) setting value, K p,i , K i,i , K d,i are proportional, integral, and differential coefficients, τ j It's time, Z j (t+τ j ) is the feedforward variable at a delay of τ j The value after j is the feedforward compensation coefficient, N is the total number of pipeline data, X i (t)-X i,setpoint is the first difference.

[0145] In the specific implementation, X i (t) is the current value of the i-th control variable. The control variable is the pipeline data, such as pressure, temperature, flow, etc. These variables directly reflect the real-time operating status of the pipeline system.

[0146] X i,setpoint Yes X i The set value of (t) is the target set value of the i-th control variable. This value represents the ideal state or goal that the system expects to achieve, and is usually determined based on safety, efficiency or other key performance indicators.

[0147] K p,i , K i,i , K d,i They are proportional, integral, and differential coefficients, respectively, where K p,i is the proportional coefficient of the i-th control variable. The proportional control part is used to adjust the current deviation, that is, (X i (t)-X i,setpoint ) linearly adjusts the response of the system. Proportional coefficient K p,i Determines the direct impact of the current deviation on the adjustment amount.

[0148] K i,i is the integral coefficient of the i-th control variable. The integral control part is through the cumulative deviation, that is, (X i (t)-X i,setpoint) to adjust the system response. Especially suitable for eliminating long-term deviations. Integration coefficient K p,i Determines the impact of cumulative deviation on the adjustment amount.

[0149] K d,i is the differential coefficient of the ith control variable. The differential control part adjusts the system response by the rate of change of the deviation (i.e., the time derivative of the deviation) to help suppress rapidly changing disturbances or oscillations. The differential coefficient K p,i Determines the impact of the deviation change rate on the adjustment amount.

[0150] τ j It's time. j (t+τ j ) is the feedforward variable at a delay of τ j The variable value after is the jth feedforward variable at time delay τ j Feedforward control is used to pre-adjust the system to respond to expected changes or disturbances. j Exogenous variables that may represent external factors, such as ambient temperature, external pressure, etc.

[0151] f j is the feedforward compensation coefficient, which is used to adjust the contribution of the feedforward variable to the system response. Feedforward control combines known external influences with appropriate compensation coefficients f j , which can effectively reduce the system's sensitivity to these interferences.

[0152] N is the total number of control variables, which indicates the number of control variables involved in the formula. m is the total number of feedforward variables, which indicates the number of feedforward compensation variables involved in the formula.

[0153] K p,i ·(X i (t)-X i,setpoint ) The proportional control part is the basis of PID control. i (t) and the set value X i,setpoint The system is adjusted based on the deviation between the two. The adjustment method of this part is linear, that is, the greater the deviation, the greater the adjustment force. Proportional coefficient K p,i The size of determines the direct impact of this part on the adjustment amount.

[0154] The integral control part adjusts the system by accumulating deviations, which is particularly suitable for eliminating static errors (i.e. deviations that exist for a long time). Integral control can ensure that the system tends to the set value over a long period of time. The integral coefficient k i,i The size of determines the contribution of the accumulated deviation to the adjustment amount. It can help the system eliminate persistent small deviations and ensure more precise control.

[0155] The differential control part adjusts the system by the rate of change of the deviation. It can predict the trend of the deviation change, so as to make corrections in advance and suppress rapid changes or shocks in the system. The differential coefficient k d,i The size of determines the impact of this trend on the adjustment amount, which helps to improve the response speed and stability of the system, especially in a rapidly changing environment.

[0156] The feedforward compensation part takes into account the influence of external factors by delaying τ j The feedforward variable Z j (t+τ j ) to pre-adjust the system. This part can deal with known external disturbances in advance, thereby reducing their negative impact on the system. Feedforward compensation coefficient f j Determines the intensity of the influence of each feedforward variable on the system response.

[0157] This formula combines three classic control strategies: proportional, integral, and differential, as well as advanced control methods such as feedforward compensation to achieve precise and automated regulation of the gas pipeline system. Proportional control: Rapidly respond to current deviations and make direct adjustments. Integral control: Accumulate historical deviations to eliminate system errors over a long period of time. Differential control: Predict deviation trends and prevent system shocks. Feedforward control: Pre-adjust for external factors to reduce interference.

[0158] Through the comprehensive application of these control strategies, the system can accurately adjust the key parameters in the gas pipeline to ensure the safety and stability of the pipeline under various conditions. Feedforward compensation is particularly suitable for dealing with interference factors that may be delayed or already predicted, making the system respond faster and smarter.

[0159] The model training module 205 is used to continue monitoring the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model.

[0160] In some embodiments, the model training module 205 is further used to:

[0161] Calculating the mean and standard deviation of the Gaussian kernel in the safety monitoring model;

[0162] Constructing a corresponding Gaussian kernel function according to the mean and the standard deviation;

[0163] Gradient descent and Bayesian optimization are performed on old parameters of the safety monitoring model according to the monitoring data, the loss function of the safety monitoring model and the Gaussian kernel function to obtain the optimized safety monitoring model with updated parameters.

[0164] In some embodiments, the update parameters of the safety monitoring model are expressed as:

[0165]

[0166] Where η2 is the learning rate, Q is the number of Gaussian kernels, and θ old are the old parameters of the model, θ new is the update parameter of the model, L(θ) is the loss function, σ is the uncertainty, n is the current number of iterations, and α k is the Bayesian optimization weight of the kth Gaussian kernel, μ k , σ k are the mean and standard deviation of the kth Gaussian kernel.

[0167] In the specific implementation, η2 is the learning rate, which is a hyperparameter that controls the step size of parameter updates. The learning rate determines the speed at which the model parameters move toward the optimal value at each update. If the learning rate is too large, it may cause the parameters to fluctuate drastically and deviate from the optimal value; if it is too small, it may lead to slow convergence.

[0168] Q is the number of Gaussian kernels, which indicates the number of Gaussian kernels used in Bayesian optimization. Gaussian kernel is a kernel function used to capture local information in parameter space during Bayesian optimization.

[0169] θ old It is the old parameter of the model, which is calculated through historical data and previous iterations, and represents the parameter setting of the model in the previous state.

[0170] θ new is the update parameter of the model, and the update parameter is the old parameter θ old On this basis, the adjustment results are combined with gradient descent and Bayesian optimization.

[0171] L(θ) is the loss function, which indicates the deviation between the model prediction result and the true value. The smaller the loss function, the higher the accuracy of the model prediction. By calculating the gradient of the loss function, we can determine the direction in which the model parameters need to be adjusted.

[0172] σ is the uncertainty, which is used to characterize the prediction uncertainty or noise level of the model in its current state. The higher the uncertainty, the less adaptable the current model is to the data, and a larger adjustment step may be needed.

[0173] n is the current iteration number. As the iteration number increases, the step size of parameter adjustment usually decreases gradually to ensure convergence to the optimal value.

[0174] α k is the Bayesian optimization weight of the kth Gaussian kernel, indicating the contribution of the kth Gaussian kernel in adjusting parameters. kDetermines the impact of each Gaussian kernel on the final parameter update.

[0175] μ k , σ k is the mean and standard deviation of the kth Gaussian kernel, μ k Indicates the center point position in the parameter space where the Gaussian kernel focuses. Mean μ k Indicates the possible optimal area of ​​the parameters. The standard deviation of the kth Gaussian kernel indicates the expansion range or influence range of the Gaussian kernel in the parameter space. Standard deviation σ k The larger it is, the wider the parameter space covered by the Gaussian kernel.

[0176] Calculate the current old parameter θ old The gradient of the loss function, the sign and magnitude of the gradient determine the direction and strength of the adjustment. Taking into account the current number of iterations n and uncertainty σ, this adjustment step size gradually decreases as the number of iterations increases, thereby avoiding parameter oscillation and promoting convergence.

[0177] Expressed in μ k is the center and the standard deviation is σ k The Gaussian distribution of the old parameter θ old The influence of θ old The closer to μ k , the larger the value of the Gaussian kernel, the greater its contribution to parameter adjustment. The Bayesian optimization part combines the contributions of multiple Gaussian kernels, can identify the optimal parameter area in the parameter space, and guide the parameters to adjust in a more optimal direction.

[0178] This formula combines the global optimization capability of the gradient descent method and the local fine-tuning capability of Bayesian optimization to form an efficient parameter update mechanism. Gradient descent part: adjust the parameters through the global loss function gradient to gradually reduce the loss function value. Bayesian optimization part: fine-tune the parameters through the Gaussian kernel function to ensure effective exploration in the local optimal area. This combination can maintain global stability and quickly adapt to local changes during model training, thereby improving the prediction accuracy and adaptability of the model. Especially in complex gas pipeline monitoring environments, this parameter update mechanism can effectively deal with a variety of uncertain factors and ensure the reliability and practicality of the model.

[0179] The safety monitoring module 206 is used to perform safety monitoring on the gas pipeline according to the optimized safety monitoring model.

[0180] It can be understood that the safety monitoring of the gas pipeline is carried out according to the optimized safety monitoring model. On the basis of the previous data collection, analysis and model optimization, the optimized monitoring model is used to continuously monitor the real-time operating status of the gas pipeline to ensure the safety, stability and efficient operation of the pipeline system.

[0181] The optimized safety monitoring model is formed through multiple iterations and continuous adjustments. It is based on a large amount of historical data and real-time monitoring data, and is optimized and calibrated by multiple algorithms (such as autoregression, moving average, Bayesian optimization, PID control, etc.). The model can more accurately reflect the operating status of the gas pipeline, predict potential risks, and provide targeted early warning and automated response solutions.

[0182] The optimization process of the security monitoring model is described in detail below:

[0183] Data collection and preprocessing: Various data including pressure, temperature, flow rate and gas concentration are collected from key nodes of the gas pipeline. After noise filtering and normalization, these data are used as inputs of the model to help the model monitor and predict more accurately.

[0184] Model training and adjustment: Train the model through historical data and adjust the model parameters (such as autoregressive coefficient, moving average coefficient, PID control coefficient, etc.) so that the model can capture the key features and abnormal patterns in the gas pipeline system.

[0185] Bayesian optimization: During the model training process, the Bayesian optimization method is used to further fine-tune the model parameters based on the Gaussian kernel function. This optimization method can find the optimal parameter combination in the parameter space, so that the model can maintain high accuracy and high robustness in various operating environments.

[0186] Model evaluation and verification: The model is evaluated through cross-validation and actual monitoring data to verify its prediction accuracy and reliability. Through continuous feedback adjustment and iteration, it is ensured that the model has sufficient adaptability when facing practical applications.

[0187] The specific safety monitoring process is as follows:

[0188] Real-time data input: Once the model optimization is completed, the system will continue to obtain real-time data from various key nodes of the gas pipeline and input this data into the optimized monitoring model.

[0189] Data analysis and anomaly detection: The model analyzes the input data and identifies possible anomalies. Because the model has been optimized, it can detect anomalies more accurately and reduce the possibility of false positives or negatives.

[0190] Risk prediction and early warning: Through time series analysis and prediction algorithms, the model can predict the possible risks in the pipeline system in advance, such as excessive pressure, abnormal temperature, sudden flow changes or gas leakage, etc. The system will issue early warning signals based on the prediction results to prompt relevant personnel to take preventive measures.

[0191] Automated response and adjustment: When an anomaly is detected or a potential risk is predicted, the model will automatically generate a corresponding adjustment plan and automatically adjust the pipeline system through mechanisms such as PID control, such as adjusting the pressure in the pipeline, controlling valve switching, etc., to ensure the safe operation of the system.

[0192] It can be understood that the advantages of the optimized safety monitoring model are as follows:

[0193] High accuracy: The optimized model can more accurately identify anomalies in the gas pipeline system, reducing the possibility of false positives and false negatives.

[0194] Quick response: Through automated adjustment mechanisms, the system can respond quickly to potential risks and prevent accidents.

[0195] Strong predictive power: The model combines historical data with real-time data to predict risks in advance and issue early warnings, giving relevant personnel enough time to intervene.

[0196] Adaptability: The model can continuously adjust and optimize itself according to the actual operating conditions, improving its adaptability in different environments and conditions.

[0197] Through the above methods, the present invention can monitor the safety of gas pipelines according to the optimized safety monitoring model, and can realize efficient and accurate monitoring of the pipeline system. Through real-time data input, anomaly detection, risk prediction and automatic response, the system can significantly improve the operational safety of gas pipelines, reduce the probability of accidents, and take effective measures in time when risks occur to ensure the stable operation of the pipeline system.

[0198] See also Figure 3 , provides a flow chart of a gas pipeline safety monitoring method based on data analysis of the present invention, comprising the following steps:

[0199] Step 301: Install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration of the surrounding environment;

[0200] Step 302: Analyze the pipeline data of the sensor in real time, and identify the abnormal situation of the gas pipeline through an abnormal detection discriminant function;

[0201] Step 303: predicting the future data of the gas pipeline through time series analysis, and establishing a safety monitoring model based on the historical data and real-time data of the gas pipeline;

[0202] Step 304: after detecting the abnormality of the gas pipeline or predicting a potential risk, adjusting the pipeline data of the gas pipeline;

[0203] Step 305: Continue to monitor the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model;

[0204] Step 306: Perform safety monitoring on the gas pipeline according to the optimized safety monitoring model.

[0205] It should be noted that the relevant contents of the above steps have been described in the aforementioned part about the gas pipeline safety monitoring system based on data analysis, and will not be repeated here.

[0206] See also Figure 4 , Figure 4 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 4 As shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, the following steps are implemented:

[0207] Install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration in the surrounding environment;

[0208] Analyze the pipeline data of the sensor in real time, and identify the abnormal situation of the gas pipeline through an abnormal detection discriminant function;

[0209] Predicting future data of the gas pipeline through time series analysis, and establishing a safety monitoring model based on historical data and real-time data of the gas pipeline;

[0210] After detecting an abnormality of the gas pipeline or predicting a potential risk, adjusting pipeline data of the gas pipeline;

[0211] Continue to monitor the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model;

[0212] The gas pipeline is safety monitored according to the optimized safety monitoring model.

[0213] See also Figure 5 , Figure 5 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 5 As shown, this embodiment provides a computer-readable storage medium 500, on which a computer program 411 is stored. When the computer program 411 is executed by a processor, the following steps are implemented:

[0214] Install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration in the surrounding environment;

[0215] Analyze the pipeline data of the sensor in real time, and identify the abnormal situation of the gas pipeline through an abnormal detection discriminant function;

[0216] Predicting future data of the gas pipeline through time series analysis, and establishing a safety monitoring model based on historical data and real-time data of the gas pipeline;

[0217] After detecting an abnormality of the gas pipeline or predicting a potential risk, adjusting pipeline data of the gas pipeline;

[0218] Continue to monitor the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model;

[0219] The gas pipeline is safety monitored according to the optimized safety monitoring model.

[0220] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0221] Those skilled in the art will appreciate that embodiments of the present invention may be provided as systems, methods, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0222] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or multiple boxes.

[0223] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction system, which is implemented in the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0224] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0225] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0226] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A gas pipeline safety monitoring system based on data analysis, characterized in that: The system comprises: A data acquisition module is used to install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration of the surrounding environment; An abnormality identification module, used for analyzing the pipeline data of the sensor in real time and identifying the abnormality of the gas pipeline through an abnormality detection discriminant function; A model building module, used to predict future data of the gas pipeline through time series analysis, and to establish a safety monitoring model based on historical data and real-time data of the gas pipeline; A data adjustment module, used to adjust the pipeline data of the gas pipeline after detecting an abnormality of the gas pipeline or predicting a potential risk; A model training module, used to continue monitoring the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model; A safety monitoring module is used to perform safety monitoring on the gas pipeline according to the optimized safety monitoring model.

2. The gas pipeline safety monitoring system based on data analysis according to claim 1 is characterized in that: The abnormality identification module is also used for: Acquire different pipeline data of the gas pipeline at a preset time; According to different pipeline data of the gas pipeline at the current time t, calculate the average value and standard deviation of each type of pipeline data; Obtaining the weight of the covariance matrix between each of the pipeline data and the corresponding adaptive weighting coefficient; Constructing the abnormality detection discriminant function according to different pipeline data of the gas pipeline at the current time t, the average value, standard deviation of each type of pipeline data, the weight of the covariance matrix and the adaptive weighting coefficient; An abnormality detection value of the gas pipeline is determined according to the abnormality detection discriminant function, and abnormality detection is performed on the gas pipeline based on the abnormality detection value.

3. The gas pipeline safety monitoring system based on data analysis according to claim 2 is characterized in that: The anomaly detection function is expressed as: Among them, X i (t), X j (t), X k (t) represents different pipeline data at time t, μ i , μ j , μ k are the average values ​​of various pipeline data, σ i , σ j , σ k is the standard deviation of various pipeline data, W ij is the weight of the covariance matrix between each pipeline data, δ k is the adaptive weighting coefficient.

4. The gas pipeline safety monitoring system based on data analysis according to claim 3 is characterized in that: The model building module is also used to: Get the pipeline data at the current time t and the historical pipeline data before time h; Obtaining exogenous variables generated by external factors of the gas pipeline; Obtain the lagged error term of the pipeline data before time s; The safety monitoring model for predicting future data of the gas pipeline is constructed based on the pipeline data at the current time t, the historical pipeline data before time h, the exogenous variables and the lagged error term.

5. The gas pipeline safety monitoring system based on data analysis according to claim 4 is characterized in that: The future data of the gas pipeline is expressed as: in, is the autoregressive coefficient, t, h, s are the time, Z v (t) is an exogenous variable, λ v is the weight coefficient of the exogenous variable, η1 is the acceleration coefficient, θ s is the moving average coefficient, ∈(ts) is the lagged error term, is the second-order derivative, γ is the velocity coefficient, is the first-order derivative, ∈ t is a random error term, D(t) is the pipeline data at the current time t, and D(th) represents the historical pipeline data at the previous h time points.

6. The gas pipeline safety monitoring system based on data analysis according to claim 5 is characterized in that: The data adjustment module is also used for: Obtaining the current value and set value of each of the pipeline data; Calculate a first difference between a current value and a set value of each pipeline data; Get the feedforward variable of each exogenous variable at a delay of τ j The variable value after time; The plurality of first difference values ​​are summed, integrated and differentiated respectively, and combined with each of the variable values ​​to obtain the adjusted pipeline data.

7. The gas pipeline safety monitoring system based on data analysis according to claim 6 is characterized in that: The adjusted pipeline data is expressed as: Among them, X i (t) is the current value of the i-th pipeline data, X i，setpoint Yes X i (t) setting value, K p，i , K i，i , K d，i are proportional, integral, and differential coefficients, τ j It's time, Z j (t+τ j ) is the feedforward variable at a delay of τ j The value after j is the feedforward compensation coefficient, N is the total number of pipeline data, X i (t)-X i，setpoint is the first difference.

8. The gas pipeline safety monitoring system based on data analysis according to claim 7 is characterized in that: The model training module is also used to: Calculating the mean and standard deviation of the Gaussian kernel in the safety monitoring model; Constructing a corresponding Gaussian kernel function according to the mean and the standard deviation; Gradient descent and Bayesian optimization are performed on old parameters of the safety monitoring model according to the monitoring data, the loss function of the safety monitoring model and the Gaussian kernel function to obtain the optimized safety monitoring model with updated parameters.

9. The gas pipeline safety monitoring system based on data analysis according to claim 7 is characterized in that: The update parameters of the safety monitoring model are expressed as: Where η2 is the learning rate, Q is the number of Gaussian kernels, and θ old are the old parameters of the model, θ new is the update parameter of the model, L(θ) is the loss function, σ is the uncertainty, n is the current number of iterations, and α k is the Bayesian optimization weight of the kth Gaussian kernel, μ k , σ k are the mean and standard deviation of the kth Gaussian kernel.

10. A gas pipeline safety monitoring method based on data analysis, characterized in that: The method comprises: Install movable sensors at key nodes of the gas pipeline to collect the pressure, temperature, flow rate inside the gas pipeline and the gas concentration in the surrounding environment; Analyze the pipeline data of the sensor in real time, and identify the abnormal situation of the gas pipeline through an abnormal detection discriminant function; Predicting future data of the gas pipeline through time series analysis, and establishing a safety monitoring model based on historical data and real-time data of the gas pipeline; After detecting an abnormality of the gas pipeline or predicting a potential risk, adjusting pipeline data of the gas pipeline; Continue to monitor the gas pipeline to obtain monitoring data, and adjust and optimize the parameters of the monitoring model based on the monitoring data to obtain an optimized safety monitoring model; The gas pipeline is safety monitored according to the optimized safety monitoring model.

Images