Gas medium platform data processing optimization method based on data analysis

Abstract

The present application relates to gas data processing technology field, specifically to a gas data processing optimization method based on data analysis, comprising the following steps: S1, data acquisition and preprocessing: real-time acquisition of gas use data; S2, data analysis: analysis of the preprocessed gas use data; S3, gas pipe network fault prediction: identify weak links and potential failure points in the pipe network; S4, user behavior analysis: provide personalized services and suggestions; S5, optimization processing: combined with the prediction results of the gas pipe network, optimize the gas supply strategy; S6, real-time monitoring and user feedback: according to the monitoring data and user feedback information, dynamically adjust the optimized gas supply strategy. The present application ensures that the system can timely warn and handle potential problems, thereby optimizing the gas supply strategy and improving the safety and operation efficiency of the system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of gas data processing, and particularly relates to a gas middle station data processing optimization method based on data analysis. BACKGROUND

[0002] With the acceleration of urbanization and the improvement of living standards, gas has become one of the indispensable energy sources in daily life and industrial production of residents. In order to ensure the stability and safety of gas supply, the data processing and optimization method of gas middle station is particularly important. Through the collection, analysis and prediction of gas use data, the efficiency of gas supply system can be improved, resource waste can be reduced, and potential safety hazards can be effectively prevented and handled.

[0003] The existing SaaS platform has provided 12 kinds of core business services including engineering construction management, customer service, billing management, mobile operation, collaborative office, digital reconciliation and the like. These services effectively improve the management efficiency and business processing capacity of enterprises, and standardize the business process. However, although certain achievements have been made, the existing system still faces some challenges in data management and use:

[0004] The existing gas data processing method mostly relies on simple data collection and basic statistical analysis, lacks deep mining and intelligent analysis of data, and the traditional method is difficult to monitor the gas use in real time, cannot timely discover and predict the fault points in the pipe network, in addition, the lack of user behavior analysis and personalized service makes the gas supply strategy unable to effectively meet the needs of different users, resulting in low gas utilization efficiency, increased safety hazards and low user satisfaction.

[0005] The present application aims to provide a gas middle station data processing optimization method based on data analysis, dynamically adjust the gas supply strategy, comprehensively improve the safety, reliability and efficiency of the gas supply system, meet the diversified needs of users and improve user satisfaction. SUMMARY

[0006] Based on the above purpose, the present application provides a gas middle station data processing optimization method based on data analysis.

[0007] The gas middle station data processing optimization method based on data analysis comprises the following steps:

[0008] S1, data collection and preprocessing: real-time collection of gas use data through a sensor network, including gas pressure, flow and temperature, and preprocessing of the collected gas use data;

[0009] S2, data analysis: analyzing the preprocessed gas use data, including trend analysis, anomaly detection and demand prediction;

[0010] S3, Gas pipeline network failure prediction: Failure prediction is performed on the gas pipeline network to identify weak links and potential failure points in the network, including:

[0011] S31, Feature extraction: Key features are extracted from the gas pipeline network data, including pressure change rate, flow fluctuation, and temperature gradient;

[0012] S32, Model construction: A gas pipeline network failure prediction model is constructed based on the extracted key features to predict failures in the gas pipeline network;

[0013] S33, Real-time prediction: Real-time gas pipeline network data is input into the gas pipeline network failure prediction model to predict failures and identify weak links and potential failure points in the network;

[0014] S34, Result verification: The accuracy of the prediction results is verified by comparing them with actual failure records and inspection results, and the gas pipeline network failure prediction model is continuously optimized and adjusted;

[0015] S4, User behavior analysis: Through big data analysis of user's gas usage behavior, personalized services and suggestions are provided, including intelligent gas usage suggestions and energy-saving scheme push;

[0016] S5, Optimization: Based on the analysis results of gas usage data and user behavior, combined with the prediction results of the gas pipeline network, the gas supply strategy is optimized, including adjusting the supply pressure and optimizing the supply route;

[0017] S6, Real-time monitoring and user feedback: A real-time monitoring mechanism is established to monitor gas usage, and user feedback information is collected. The optimized gas supply strategy is dynamically adjusted based on the monitoring data and user feedback information.

[0018] Optionally, the data collection and preprocessing in S1 includes:

[0019] S11, Data collection: Real-time collection of gas usage data, including:

[0020] Gas pressure data collection: Real-time monitoring of gas pressure in the pipeline using pressure sensors;

[0021] Flow data collection: Measurement of gas flow using flow meter sensors;

[0022] Temperature data collection: Monitoring of gas temperature changes using temperature sensors;

[0023] S12, Data preprocessing: Preprocessing of collected gas usage data, including data cleaning and data format conversion.

[0024] Optionally, the data analysis in S2 includes:

[0025] S21, trend analysis: using linear regression model to analyze the trend of the preprocessed gas usage data, represented as:

[0026] y = β0 + β1x;

[0027] where y is the predicted gas usage, x is time, β0 and β1 are regression coefficients;

[0028] S22, anomaly detection: using seasonal autoregressive integrated moving average model (SARIMA) to detect anomalies in the preprocessed gas usage data, represented as:

[0029]

[0030] where y t is the current data point, μ is the constant term, is the autoregressive coefficient, θ1 is the moving average coefficient, ∈ t is the error term, Φ1 and Θ1 are the seasonal autoregressive and moving average coefficients, s is the season length;

[0031] S23, demand forecasting: using the grey prediction model (GM) to forecast the demand of the preprocessed gas usage data, specifically including:

[0032] S231, data accumulation generation: performing one accumulation generation on the original data sequence x = {x(1), x(2),..., x(n)} to obtain a new data sequence X = {X(1), X(2),..., X(n)}, represented as:

[0033]

[0034] S232, establishing differential equation: establishing a grey differential equation based on the accumulated data sequence, represented as:

[0035]

[0036] where a and b are undetermined parameters;

[0037] S233, solving parameters: using the least squares method to solve the undetermined parameters a and b, represented as:

[0038] Define matrix B and vector Y:

[0039]

[0040] Solve the undetermined parameters using the least squares method:

[0041]

[0042] where (BT B) 1 is B T inverse matrix of B, B T is the transpose of matrix B;

[0043] S234, prediction value generation: according to the solved parameters, a prediction model is established and the prediction value of the original data is solved is expressed as:

[0044]

[0045] wherein, is the prediction value of the cumulative generation sequence at the k+1 time, X(1) is the value of the cumulative generation sequence at the initial time, e is the base of natural logarithm, and k is a time index, is the prediction value of the original data sequence at the k+1 time, is the prediction value of the cumulative generation sequence at the k time.

[0046] Optionally, the feature extraction in the S31 comprises:

[0047] S311, pressure change rate: calculate the rate of change of pressure with time, expressed as:

[0048]

[0049] wherein, ΔP is the pressure change rate, P(t) is the pressure at the current time, P(t-1) is the pressure at the previous time, and Δt is the time interval;

[0050] S312, flow fluctuation: calculate the fluctuation amplitude of flow with time, expressed as:

[0051]

[0052] wherein, ΔF is the flow fluctuation, F i is the flow value at the i time, is the average value of the flow, and n is the observation number;

[0053] S313, temperature gradient: calculate the change of temperature at different positions of the pipe network, expressed as:

[0054]

[0055] wherein, is the temperature gradient, T(x2) and T(x1) are the temperatures at positions x2 and x1 of the pipe network respectively, and x2-x1 is the distance between the two positions.

[0056] Optionally, the gas pipeline network fault prediction model in S32 adopts a graph neural network (GNN) model, and the graph neural network (GNN) model comprises:

[0057] S321, graph representation and feature extraction: the gas pipeline network is represented as a graph G=(V,E), where the nodes V represent gas equipment and sensors, and the edges E represent pipeline connections; key features are extracted from the gas pipeline network data as node attributes, and the feature vector h v including pressure rate of change ΔP, flow fluctuation ΔF and temperature gradient

[0058] S322, node feature enhancement: time feature t and location feature p are introduced as additional dimensions of node features, represented as:

[0059]

[0060] S323, message passing and updating mechanism: the message passing function is optimized, considering the edge features e uv and the weighted sum of node features (such as pipeline length, pipeline diameter), represented as:

[0061] Message passing:

[0062] wherein, is the message of node v at the k-th layer, is the neighbor node set of node v, and α uv is the attention weight between node u and node v, W e and W h is the weight matrix, and b h is the bias;

[0063]

[0064] Node update:

[0065] wherein, is the feature representation of node v at the k-th layer, W m is the message weight matrix, and b m is the message bias;

[0066] S324, multi-layer stacking and global information fusion: a global pooling layer is introduced to fuse the information of all nodes together for global feature extraction, represented as:

[0067]

[0068] wherein, h G is the global feature representation, and Pooling is average pooling;

[0069] S325, fault prediction output: using the fully connected layer to make the fault prediction, predicting whether the node has a fault, denoted as:

[0070]

[0071] wherein, is the fault prediction value of node v, σ is the Sigmoid activation function, W o is the output layer weight matrix, b o is the output layer bias.

[0072] Optionally, the real-time prediction in the S33 comprises:

[0073] S331, real-time data input: input the real-time collected gas pipeline network data as the node feature vector h v into the gas pipeline network fault prediction model;

[0074] S332, forward propagation: through the gas pipeline network fault prediction model, the node feature vector h v is calculated by forward propagation, and each layer of graph neural network is sequentially passed through, and the node feature representation is updated until the Kth layer, and the final node feature representation is obtained

[0075] S333, fault prediction: using the fully connected layer and the activation function, the final node feature representation is subjected to fault prediction, denoted as:

[0076]

[0077] wherein, is the fault prediction value of node v, σ is the Sigmoid activation function, W o is the output layer weight matrix, b o is the output layer bias.

[0078] S334, weak link and potential fault point identification: according to the fault prediction value the weak link and the potential fault point in the gas pipeline network are identified, and when the set fault threshold is exceeded, the node v is marked as a potential fault point.

[0079] Optionally, the result verification in the S34 comprises:

[0080] S341, prediction result collection: collecting the fault prediction results of the gas pipeline network fault prediction model on the gas pipeline network data

[0081] S342, actual fault record comparison: comparing the prediction results Compared with actual fault records y v Compare these metrics and calculate the accuracy of the predictions, including accuracy, precision, recall, and F1 score.

[0082] S343, Inspection Result Comparison: Compare the predicted results with the inspection results to evaluate the effectiveness of the model in practical applications;

[0083] S344, Model Optimization and Tuning: Based on the validation results, the model parameters are optimized and tuned using the backpropagation algorithm and gradient descent method, as shown below:

[0084]

[0085] Where θ is the model parameter, η is the learning rate, and L(θ) is the loss function.

[0086] Optionally, the user behavior analysis in S4 includes:

[0087] S41, Data Collection: Real-time collection of user gas usage data via smart gas meters and sensor networks to form a user gas usage behavior dataset D, represented as:

[0088] D = {(u i ,t j ,x ij )};

[0089] Among them, u i For user i,t j For time j, x ij For user i, the amount of gas used at time j;

[0090] S42, Behavioral Pattern Recognition: Using the K-means clustering algorithm to identify users' gas usage behavior patterns;

[0091] S43, Personalized Service and Recommendation Generation: Based on identified gas usage patterns, including smart gas usage suggestions and energy-saving solution recommendations, personalized services and recommendations are generated, specifically including:

[0092] S431, Smart Gas Usage Suggestions: Based on user behavior patterns, smart gas usage suggestions are generated using the Apriori algorithm;

[0093] S432, Energy-saving solution push: Based on user behavior patterns and usage data, regression analysis is used to predict users' gas demand and provide energy-saving solutions.

[0094] Optionally, the optimization process in S5 includes:

[0095] S51, supply pressure adjustment: based on gas usage data and user behavior analysis results, combined with the prediction results of the gas pipeline network, the supply pressure is adjusted using an optimization algorithm, represented as:

[0096]

[0097] where P opt is the optimized supply pressure, P base is the basic supply pressure, a is the adjustment coefficient, w i is the weight of user i, d i is the actual demand of user i, is the predicted demand of user i;

[0098] S52, supply route optimization: based on the failure prediction results of the gas pipeline network, the gas supply route is optimized using a flow distribution algorithm, represented as:

[0099] min∑ (i,j)∈E c ij f ij ;

[0100] Constraints:

[0101]

[0102] where c ij represents the transmission cost of edge (i,j), f ij represents the flow on edge (i,j), d i represents the demand of node i, and E represents the edge set.

[0103] S53, comprehensive optimization: combining the results of pressure adjustment and route optimization, considering cost and benefit comprehensively, the optimal gas supply strategy is formulated using linear programming, represented as:

[0104] Objective function:

[0105] Constraints:

[0106] Flow conservation constraint:

[0107] Capacity constraint:

[0108] Supply pressure constraint: P min ≤P opt ≤P max ;

[0109] where u ij is the capacity of edge (i,j), P opt is the optimized supply pressure, P baseis the base supply pressure, α is the adjustment coefficient, w ij is the weight of user i at location j, d ij is the actual demand of user i at location j, is the predicted demand of user i at location j, β is the weight coefficient of supply pressure adjustment, P min and P max are the minimum and maximum limits of supply pressure.

[0110] Optionally, the real-time monitoring and user feedback in S6 include:

[0111] S61, Real-time monitoring mechanism establishment: through the sensor network and the intelligent gas meter, real-time collection of gas use data, including pressure, flow and temperature, and transmission of the gas use data to the monitoring center;

[0112] S62, User feedback information collection: through the user feedback platform (such as mobile application or customer service system), collection of user feedback information on gas supply, including gas experience, safety hazards and suggestions;

[0113] S63, Data fusion and analysis: fusion of real-time monitoring data D t and user feedback information U f , evaluation of the effect of the current gas supply strategy by using data analysis algorithm, and identification of potential problems, represented as:

[0114] S eval = α∑ t D t + β∑ i U f ;

[0115] Wherein, S eval is the evaluation score of the supply strategy, and α and β are weight coefficients;

[0116] S64, Dynamic adjustment: according to the evaluation score S eval and the identified potential problems, dynamic adjustment of the gas supply strategy by using the adaptive optimization algorithm, represented as:

[0117]

[0118] Wherein, θ is the parameter of the gas supply strategy, η is the learning rate, and L(θ, S eval ) is the loss function based on the evaluation score.

[0119] The beneficial effects of the present application are:

[0120] The present application, through the real-time collection of gas use data by the sensor network and the preprocessing, ensures the accuracy, synchronization and consistency of the data, uses the linear regression, SARIMA model and grey prediction model algorithm for data analysis, effectively performs trend analysis, anomaly detection and demand prediction, ensures the precision and reliability of data processing, provides a solid data foundation for subsequent fault prediction and optimization processing, and finally improves the safety and efficiency of the gas supply system.

[0121] The present application, through the feature extraction, node feature enhancement, message passing and updating mechanism, multi-layer stacking and global information fusion in the graph neural network model, combines the pressure change rate, flow fluctuation, temperature gradient, time feature and location feature, improves the capture ability of the gas pipe network dynamic characteristics, combines real-time prediction with result verification, continuously optimizes and adjusts the model parameters, improves the accuracy and real-time performance of fault prediction, ensures that the system can timely warn and handle potential problems, thereby optimizing the gas supply strategy and improving the safety and operation efficiency of the system.

[0122] The present application, through user behavior analysis, uses big data to analyze the gas use behavior of users, provides personalized services and suggestions, including intelligent gas use suggestions and energy saving scheme pushing, optimizes the analysis results of gas use data and user behavior, combines the prediction results of the gas pipe network, optimizes the gas supply strategy, including adjusting the supply pressure and optimizing the supply route, combines the real-time monitoring mechanism with user feedback, dynamically adjusts the gas supply strategy through data fusion and analysis, improves the reliability and safety of gas supply, reduces the operation cost, effectively deals with demand fluctuations and potential fault points, and enhances the overall benefit and user satisfaction of the system. BRIEF DESCRIPTION OF DRAWINGS

[0123] In order to more clearly illustrate the technical solutions in the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only illustrate the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0124] Fig. 1 The figure is the optimization method flowchart of the embodiment of the present application.

[0125] Fig. 2 The figure is the gas pipe network fault prediction schematic diagram of the embodiment of the present application. DETAILED DESCRIPTION

[0126] The application will be described in greater detail in connection with the accompanying drawings and specific embodiments. It should be noted that the embodiments described below are the best mode, preferred embodiments, and other alternative embodiments can also be implemented by those skilled in the art with some known technologies. Moreover, the accompanying drawings are only used to more specifically describe the embodiments, and are not intended to specifically limit the application.

[0127] It should be noted that in the specification, "one embodiment", "embodiment", "exemplary embodiment", "some embodiments" and the like indicate that the described embodiment can include a specific feature, structure or characteristic, but not necessarily every embodiment includes the specific feature, structure or characteristic. In addition, when a specific feature, structure or characteristic is described in connection with an embodiment, it should be within the knowledge of those skilled in the related art to implement such a feature, structure or characteristic in connection with other embodiments, whether or not explicitly described.

[0128] Generally, the terms can be understood at least in part from the context in which they are used. For example, depending on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular or can be used to describe combinations of features, structures, or characteristics, whether large or small, whether related or unrelated to each other. In addition, the term "based on" can be understood as not necessarily intended to convey a set of exclusive factors, but can instead, at least in part, depend on the context, allowing the presence of other factors not necessarily explicitly described.

[0129] As shown in Figs. 1-2 The gas data processing optimization method based on data analysis includes the following steps:

[0130] S1, data acquisition and preprocessing: real-time acquisition of gas use data through a sensor network, including gas pressure, flow rate and temperature, and preprocessing of the acquired gas use data;

[0131] S2, data analysis: analysis of preprocessed gas use data, including trend analysis, anomaly detection and demand prediction;

[0132] S3, gas pipe network fault prediction: fault prediction of the gas pipe network to identify weak links and potential fault points in the pipe network, improve prediction accuracy, including:

[0133] S31, feature extraction: extracting key features from gas pipe network data, including pressure change rate, flow fluctuation and temperature gradient;

[0134] S32, model construction: constructing a gas pipe network fault prediction model based on the extracted key features to predict faults in the gas pipe network;

[0135] S33, Real-time Prediction: Input real-time gas pipeline network data into the gas pipeline network fault prediction model to perform fault prediction and identify weak links and potential fault points in the pipeline network;

[0136] S34, Result Verification: Compare the prediction results with actual fault records and inspection results to verify the accuracy of the prediction results and continuously optimize and adjust the gas pipeline network fault prediction model;

[0137] S4, User Behavior Analysis: Analyze users' gas usage behavior through big data to provide personalized services and recommendations, including intelligent gas usage suggestions and energy-saving scheme push, improve user satisfaction and service quality, and optimize overall gas supply strategy;

[0138] S5, Optimization Processing: Based on the analysis results of gas usage data and user behavior, combined with the prediction results of the gas pipeline network, optimize the gas supply strategy, including adjusting the supply pressure and optimizing the supply route;

[0139] S6, Real-time Monitoring and User Feedback: Establish a real-time monitoring mechanism to monitor gas usage and collect user feedback information, and dynamically adjust the optimized gas supply strategy based on monitoring data and user feedback information;

[0140] Through the above steps, not only the accuracy and real-time performance of data processing are improved, but also the fault prediction and response ability are enhanced, and the user satisfaction and overall efficiency and safety of gas supply are improved.

[0141] Data collection and preprocessing in S1 includes:

[0142] S11, Data Collection: Real-time collection of gas usage data, including:

[0143] Gas pressure data collection: Use pressure sensors to monitor the gas pressure in the pipeline in real time;

[0144] Flow data collection: Use flow meter sensors to measure the flow of gas;

[0145] Temperature data collection: Use temperature sensors to monitor the temperature change of gas;

[0146] S12, Data Preprocessing: Preprocess the collected gas usage data, including data cleaning and data format conversion;

[0147] Data cleaning uses data cleaning algorithms to remove outliers and noise data, including:

[0148] Outlier detection: Use the 3σ principle for outlier detection, i.e. for each data point x i , if |x i- μ | > 3σ, it is considered that the point is an outlier, where μ is the mean of the data set, and σ is the standard deviation;

[0149] Data smoothing: using the moving average method to smooth the data, represented as:

[0150]

[0151] where S i is the smoothed data, x j is the jth data point, and k is the size of the sliding window;

[0152] Data format conversion: unify the data format collected by different sensors to a standard format for subsequent analysis and processing, including:

[0153] Timestamp alignment: unify the timestamps of different sensor data to ensure data synchronization, assuming the timestamp of sensor A is t A , and the timestamp of sensor B is t B , then the converted timestamp t' is:

[0154]

[0155] Unit unification: convert the units of different sensor data to a unified unit, unify the pressure unit to Pascal (Pa), the flow unit to cubic meters per hour (m 3 / h), and the temperature unit to Celsius (℃);

[0156] Through the above, the accuracy, synchronization and consistency of the data are ensured, the data quality is improved, and a solid data foundation is provided for fault prediction and optimization processing, ultimately improving the safety and efficiency of the gas supply system.

[0157] The data analysis in S2 includes:

[0158] S21, trend analysis: using a linear regression model to analyze the trend of the preprocessed gas usage data, represented as:

[0159] y = β0 + β1x;

[0160] where y is the predicted gas usage, x is the time, and β0 and β1 are the regression coefficients;

[0161] The least squares method is used to estimate the regression coefficients β0 and β1, including:

[0162] Objective function: assuming a linear regression model, the objective of the least squares method is to minimize the sum of squared errors, represented as:

[0163]

[0164] where y i is the actual value of the i-th observation, x i is the independent variable value of the i-th observation, β0 and β1 are regression coefficients, and n is the total number of observations;

[0165] Solving the regression coefficients: by taking the partial derivative of S with respect to β0 and β1, and setting it to zero, denoted as:

[0166]

[0167] where is the mean value of x, is the mean value of y;

[0168] Calculating the mean value: calculate the mean value of x and y, denoted as:

[0169]

[0170] S22, anomaly detection: using the seasonal autoregressive integrated moving average model (SARIMA) to detect anomalies in the preprocessed gas usage data, denoted as:

[0171]

[0172] where y t is the current data point, μ is a constant term, is the autoregressive coefficient, θ1 is the moving average coefficient, ∈ t is the error term, Φ1 and Θ1 are the seasonal autoregressive and moving average coefficients, and s is the season length;

[0173] S23, demand prediction: using the grey prediction model (GM) to predict the demand of the preprocessed gas usage data, which includes:

[0174] S231, data accumulation generation: the original data sequence x = {x(1), x(2),..., x(n)} is accumulated once to generate a new data sequence X = {X(1), X(2),..., X(n)}, denoted as:

[0175]

[0176] S232, establishing differential equation: according to the accumulated data sequence, a grey differential equation is established, denoted as:

[0177]

[0178] where a and b are undetermined parameters;

[0179] S233, solving parameters: solve the undetermined parameters a and b using the least squares method, denoted as:

[0180] Define matrix B and vector Y:

[0181]

[0182] Solve the undetermined parameters using the least squares method:

[0183]

[0184] Where (B T B) 1 is the inverse matrix of B T B, B T is the transpose of matrix B;

[0185] S234, prediction value generation: according to the solved parameters, establish a prediction model and solve the prediction value of the original data Denoted as:

[0186]

[0187] Where, is the prediction value of the cumulative generation sequence at the k+1 time, X(1) is the value of the cumulative generation sequence at the initial time, e is the base of natural logarithm, k is the time index, is the prediction value of the original data sequence at the k+1 time, is the prediction value of the cumulative generation sequence at the k time;

[0188] Through the above content, the accuracy and reliability of data analysis are ensured, thereby providing solid data support and scientific basis for optimizing gas supply strategy and improving system safety and efficiency.

[0189] The feature extraction in S31 includes:

[0190] S311, pressure change rate: calculate the rate of change of pressure with time, denoted as:

[0191]

[0192] Where ΔP is the pressure change rate, P(t) is the pressure at the current time, P(t-1) is the pressure at the previous time, and Δt is the time interval;

[0193] S312, flow fluctuation: calculate the fluctuation amplitude of flow with time, denoted as:

[0194]

[0195] Where ΔF is the flow fluctuation, F iQi is the flow value at the i-th time point, Qi is the average value of the flow, and n is the number of observations;

[0196] S313, temperature gradient: calculate the change of temperature at different positions of the pipe network, represented as:

[0197]

[0198] wherein, is the temperature gradient, T(x2) and T(x1) are the temperatures at positions x2 and x1 of the pipe network, respectively, and x2-x1 is the distance between the two positions;

[0199] Through the above, the prediction accuracy and reliability of the model can be improved, helping to identify weak links and potential fault points in the pipe network in a timely manner, thereby optimizing the gas supply strategy and improving the safety and efficiency of the system.

[0200] The gas pipe network fault prediction model in S32 adopts a graph neural network (GNN) model, which includes:

[0201] S321, graph representation and feature extraction: the gas pipe network is represented as a graph G=(V,E), where the nodes V represent gas equipment and sensors, and the edges E represent pipe connections. Key features are extracted from the gas pipe network data as node attributes, and the feature vector h v including pressure change rate ΔP, flow fluctuation ΔF and temperature gradient

[0202] S322, node feature enhancement: in order to better capture the dynamic characteristics of the nodes, time features t and position features p are introduced as additional dimensions of the node features, represented as:

[0203]

[0204] S323, message passing and updating mechanism: optimize the message passing function, considering the weighted sum of edge features e uv (such as pipe length, pipe diameter) and node features, represented as:

[0205] Message passing:

[0206] wherein, is the message of node v at the k-th layer, is the neighbor node set of node v, and α uv is the attention weight between node u and node v, W e and W h is the weight matrix, and b h is the bias;

[0207]

[0208] Node update:

[0209] where, is the feature representation of node v at the k-th layer, W m is the message weight matrix, b m is the message bias;

[0210] S324, multi-layer stacking and global information fusion: In order to capture global information, a global pooling layer is introduced to fuse the information of all nodes together for global feature extraction, denoted as:

[0211]

[0212] where, G is the global feature representation, Pooling is average pooling;

[0213] Average pooling Pooling is denoted as:

[0214]

[0215] where, G is the global feature representation, V is the node set, is the feature representation of node v at the k-th layer, |V| is the size of the node set, i.e., the number of nodes;

[0216] S325, fault prediction output: Use a fully connected layer for fault prediction to predict whether a node has a fault, denoted as:

[0217]

[0218] where, is the fault prediction value of node v, σ is the Sigmoid activation function, W o is the output layer weight matrix, b o is the output layer bias;

[0219] Through the above, combined with the rate of pressure change, flow fluctuation, temperature gradient, time characteristics and location characteristics, the ability to capture the dynamic characteristics of the gas pipeline network is improved, the message passing and node update mechanism more accurately reflects the dependence between nodes, and the attention mechanism enhances the extraction of important information, the global average pooling further fuses the global information, ensuring the comprehensiveness and accuracy of fault prediction, the overall model improves the reliability and real-time performance of gas pipeline network fault prediction, which helps to identify weak links and potential fault points in the pipeline network in advance, thereby optimizing the gas supply strategy and improving the safety and operation efficiency of the system.

[0220] The real-time prediction in S33 includes:

[0221] S331, Real-time data input: Collect real-time gas pipeline network data as node feature vector h v Input into the gas pipeline network fault prediction model;

[0222] S332, Forward propagation: Through the gas pipeline network fault prediction model, the node feature vector h v is calculated by forward propagation, passing through each layer of graph neural network in turn, updating the node feature representation until the Kth layer, obtaining the final node feature representation

[0223] S333, Fault prediction: Using the fully connected layer and the activation function, the final node feature representation is predicted as:

[0224]

[0225] where, is the fault prediction value of node v, σ is the Sigmoid activation function, W o is the output layer weight matrix, b o is the output layer bias;

[0226] S334, Weak link and potential fault point identification: According to the fault prediction value identify the weak link and potential fault point in the gas pipeline network, when exceeds the set fault threshold, node v is marked as a potential fault point;

[0227] Through the above content, according to the prediction result, the weak link and potential fault point in the gas pipeline network are identified, to ensure that the system can timely alarm and handle potential problems, improve the safety and reliability of gas supply.

[0228] The fault threshold is set through historical data analysis, represented as:

[0229] Collect historical data: Collect a large amount of historical data, including node feature data h v and corresponding fault label y v under normal and fault conditions;

[0230] Calculate the distribution of fault prediction value: Use the gas pipeline network fault prediction model to predict the historical data, and get the distribution of fault prediction value ;

[0231] Determine the threshold: According to the distribution of fault prediction value, select the 95% percentile as the fault threshold, represented as:

[0232]

[0233] The result verification in S34 includes:

[0234] S341, prediction result collection: collect the fault prediction results of the gas pipeline network fault prediction model on the gas pipeline network data

[0235] S342, actual fault record comparison: compare the prediction results with the actual fault records y v , calculate the accuracy indicators of the prediction, including accuracy, precision, recall and F1 score, specifically including:

[0236]

[0237] Where TP is the true positive, i.e. the number of instances correctly predicted as fault, TN is the true negative, i.e. the number of instances correctly predicted as non-fault, FP is the false positive, i.e. the number of instances incorrectly predicted as fault, and FN is the false negative, i.e. the number of instances incorrectly predicted as non-fault;

[0238] S343, inspection result comparison: compare the prediction results with the inspection results, evaluate the effect of the model in actual application, and verify the accuracy of the model;

[0239] S344, model optimization and adjustment: according to the verification results, use back propagation algorithm and gradient descent method to optimize and adjust the model parameters, represented as:

[0240]

[0241] Where θ is the model parameter, η is the learning rate, and L(θ) is the loss function;

[0242] The loss function L(θ) uses the binary cross-entropy loss function, represented as:

[0243]

[0244] Through the above, the prediction accuracy and reliability of the model are continuously improved, ensuring that the gas supply system can accurately and timely identify and handle potential faults, ensuring the safety and stable operation of the system.

[0245] The user behavior analysis in S4 includes:

[0246] S41, data collection: collect the user's gas use data in real time through intelligent gas meters and sensor networks, form the user's gas use behavior data set D, represented as:

[0247] D = {(ui t j ,x ij )};

[0248] where u i is the user i, t j is the time j, x ij is the gas usage of user i at time j;

[0249] S42, behavior pattern recognition: using K-means clustering algorithm to identify the user's gas usage behavior pattern, represented as:

[0250]

[0251] where C i is the i-th cluster, μ i is the centroid of the i-th cluster, x is the data point, and k is the number of clusters;

[0252] S43, personalized service and suggestion generation: based on the identified gas usage behavior pattern, including intelligent gas usage suggestions and energy-saving scheme pushing, generating personalized services and suggestions, including:

[0253] S431, intelligent gas usage suggestions: based on the user's behavior pattern, using Apriori algorithm to generate intelligent gas usage suggestions, represented as:

[0254]

[0255] where A and B are two different behaviors or events, Count(A) is the number of times event A occurs, Count(A∪B) is the number of times events A and B occur simultaneously, and N is the total number of records;

[0256] S432, energy-saving scheme pushing: based on the user's behavior pattern and usage data, using regression analysis to predict the user's gas demand and provide energy-saving schemes, represented as:

[0257] y = β0 + β1x1 + β2x2 + … + β n x n ;

[0258] where y is the predicted gas demand, x1, x2, …, x n are the features that affect gas demand, and β0, β1, …, β n are the regression coefficients;

[0259] Through the above, not only the user satisfaction and service quality are improved, but also the overall gas supply strategy is optimized, the gas utilization efficiency is improved, the energy-saving and emission-reduction goals are achieved, and the economic and environmental performance of the system is enhanced.

[0260] The intelligent gas usage suggestions specifically include:

[0261] Gas usage time optimization: Suggest users to use gas during off-peak hours to reduce pressure and possible price fluctuations during peak usage periods, and provide users with daily, weekly, and monthly optimal gas usage period analysis;

[0262] Device usage suggestions: Prompt users to use gas devices reasonably, such as regularly cleaning gas stoves, checking pipe connections, ensuring high device operation efficiency, and suggesting users to replace high-efficiency energy-saving gas devices, and provide corresponding product recommendations;

[0263] Safety reminders: According to user gas usage behavior, prompt possible safety hazards such as long-term non-closed gas devices, abnormal gas usage, etc., and provide regular safety check suggestions and emergency contact information when necessary;

[0264] Gas usage mode suggestions: Based on user's gas usage history data, suggest more reasonable gas usage modes, such as reducing the use frequency of gas water heaters, setting reasonable temperature, and providing similar user gas usage behavior comparison analysis to help users adjust their gas usage habits;

[0265] Energy-saving tips: Provide gas energy-saving tips, such as using pot covers reasonably, controlling flame size, etc., and suggest users to arrange cooking time and methods reasonably to achieve the effect of saving gas.

[0266] Energy-saving scheme push specifically includes:

[0267] Gas usage prediction: Based on user's historical gas usage data and current climate conditions, predict future gas usage for a period of time to help users plan gas usage reasonably;

[0268] Energy-saving device recommendation: According to user's gas usage habits and needs, recommend more efficient and energy-saving gas devices and appliances, such as high-efficiency gas stoves, intelligent gas water heaters, etc., and provide performance comparison and use cost analysis of energy-saving devices to help users make wise choices;

[0269] Energy-saving behavior suggestions: Suggest users to take specific energy-saving behaviors, such as reducing unnecessary gas usage, optimizing cooking methods, etc., and provide user-customized energy-saving plans and steps;

[0270] Gas usage optimization suggestions: Through data analysis, propose optimization schemes for user's gas usage mode, such as reducing peak period gas usage, concentrating gas device usage, etc., and provide energy-saving gas strategies for different seasons and time periods;

[0271] Feedback and adjustment: According to user's actual gas usage effect and feedback, continuously adjust and optimize energy-saving schemes, and provide regular energy-saving reports to help users understand energy-saving effects and further improvement space.

[0272] The optimization process in S5 includes:

[0273] S51, supply pressure adjustment: based on gas usage data and user behavior analysis results, combined with the prediction results of the gas pipeline network, the supply pressure is adjusted using an optimization algorithm, represented as:

[0274]

[0275] where P opt is the optimized supply pressure, P base is the basic supply pressure, α is the adjustment coefficient, w i is the weight of user i, d i is the actual demand of user i, is the predicted demand of user i;

[0276] S52, supply route optimization: based on the fault prediction results of the gas pipeline network, the gas supply route is optimized using a flow distribution algorithm to ensure the efficiency and safety of gas delivery, represented as:

[0277] min∑ (i,j)∈E c ij f ij ;

[0278] Constraints:

[0279]

[0280] where c ij represents the transmission cost of edge (i,j), f ij represents the flow on edge (i,j), d i represents the demand of node i, and E represents the edge set.

[0281] S53, comprehensive optimization: combining the results of pressure adjustment and route optimization, considering cost and benefit comprehensively, using linear programming to develop the optimal gas supply strategy, represented as:

[0282] Objective function:

[0283] Constraints:

[0284] Flow conservation constraint:

[0285] Capacity constraint:

[0286] Supply pressure constraint: P min ≤ P opt ≤ P max ;

[0287] where u ijis the capacity of edge (i,j), P opt is the optimized supply pressure, P base is the base supply pressure, α is the adjustment coefficient, w ij is the weight of user i at location j, d ij is the actual demand of user i at location j, is the predicted demand of user i at location j, β is the weight coefficient of supply pressure adjustment, P min and P max are the minimum and maximum limits of supply pressure;

[0288] Through the above, the reliability and safety of gas supply are improved, the operating cost is reduced, and the demand fluctuation and potential failure points are effectively dealt with, thereby improving the overall benefit and user satisfaction of the system.

[0289] Real-time monitoring and user feedback in S6 include:

[0290] S61, Real-time monitoring mechanism establishment: through the sensor network and intelligent gas meter, real-time collection of gas use data, including pressure, flow and temperature, and transmission of gas use data to the monitoring center;

[0291] S62, User feedback information collection: through the user feedback platform (such as mobile application or customer service system), collect user feedback information on gas supply, including gas experience, safety hazards and suggestions;

[0292] S63, Data fusion and analysis: fuse real-time monitoring data D t and user feedback information U f , use data analysis algorithm to evaluate the effect of current gas supply strategy, and identify potential problems, represented as:

[0293] S eval = α∑ t D t + β∑ i U f ;

[0294] Wherein, S eval is the evaluation score of supply strategy, α and β are weight coefficients;

[0295] S64, Dynamic adjustment: according to the evaluation score S eval and the identified potential problems, use adaptive optimization algorithm to dynamically adjust the gas supply strategy to ensure the safety and efficiency of supply, represented as:

[0296]

[0297] Wherein, θ is the parameter of gas supply strategy, η is the learning rate, L(θ,Seval ) is a loss function based on the evaluation score;

[0298] Through the above, not only the real-time monitoring ability of gas supply is improved, but also the user participation is enhanced, the user demand and feedback can be responded in time, and therefore the overall performance and user satisfaction of the gas supply system are optimized.

[0299] The present application encompasses any alternatives, modifications, equivalent methods and solutions made to the essence and scope of the present application. In order to make the public have a thorough understanding of the present application, specific details are described in the following preferred embodiments of the present application, and the present application can also be fully understood without the description of these details for those skilled in the art. In addition, in order to avoid unnecessary confusion to the essence of the present application, well-known methods, processes, procedures, elements and circuits are not described in detail.

[0300] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can also be made, and these improvements and refinements should be considered as the protection scope of the present application.

Claims

1. A data analysis-based data processing optimization method for a gas intermediate platform, characterized in that, The method comprises the following steps: S1, data collection and preprocessing: real-time collection of gas use data through a sensor network, including gas pressure, flow rate, and temperature, and preprocessing of the collected gas use data; S2, data analysis: analysis of the preprocessed gas use data, including trend analysis, anomaly detection, and demand prediction; S3, gas pipeline network fault prediction: fault prediction of the gas pipeline network to identify weak links and potential fault points in the pipeline network, specifically including: S31, feature extraction: extraction of key features from gas pipeline network data, including pressure rate of change, flow fluctuation, and temperature gradient; the feature extraction includes: S311, pressure rate of change: calculation of the rate of change of pressure over time, represented as: ； wherein, is the rate of pressure change, is the current pressure, is the pressure at the previous time, is the time interval; S312, flow fluctuation: calculation of the fluctuation amplitude of flow over time, represented as: ； in, For traffic fluctuations, For the first The flow rate at any given moment. This represents the average flow rate. The number of observations; S313, temperature gradient: calculation of the temperature change at different positions in the pipeline network, represented as: ； wherein is the temperature gradient, and are the temperatures at the pipe network locations and respectively, is the distance between the two locations; S32, model construction: construction of a gas pipeline network fault prediction model based on the extracted key features for fault prediction of the gas pipeline network; the gas pipeline network fault prediction model uses a graph neural network model, which includes: S321, graph representation and feature extraction: the gas network is represented as a graph where nodes represent gas equipment and sensors, and edges represent pipe connections. Key features are extracted from the gas network data as attributes of the nodes, each node has a feature vector including rate of pressure change , flow fluctuation and temperature gradient ; S322, node feature enhancement: introduce time feature and location feature As an additional dimension of node features, represented as: ； S323, Message passing and update mechanism: optimize message passing function, consider edge features and the weighted sum of node features, denoted as: Messaging: ; wherein, is a node In a first layer of messages, is a node set of neighbor nodes of the node is a node and a node attention weight between the node and is a weight matrix, is a bias; ； Node update: ; wherein, is a node In a first characteristic representation of a layer, is a message weight matrix, is a message bias; S324, multi-layer stacking and global information fusion: introduction of a global pooling layer to fuse the information of all nodes together for global feature extraction, represented as: ； wherein, is a global feature representation, is an average pooling; S325, fault prediction output: use of a fully connected layer for fault prediction to predict whether a node has a fault, represented as: ； wherein, is a failure prediction value for a node , is a Sigmoid activation function, is an output layer weight matrix, is an output layer bias; S33, real-time prediction: input of real-time gas pipeline network data into the gas pipeline network fault prediction model for fault prediction to identify weak links and potential fault points in the pipeline network; S34, result verification: comparison with actual fault records and inspection results to verify the accuracy of the prediction results, and continuous optimization and adjustment of the gas pipeline network fault prediction model; S4, user behavior analysis: analysis of user gas use behavior through big data to provide personalized services and suggestions, including intelligent gas use suggestions and energy-saving scheme pushing; S5, optimization processing: optimization of gas supply strategies based on the analysis results of gas use data and user behavior, combined with the prediction results of the gas pipeline network, including adjustment of supply pressure and optimization of supply routes; S6, real-time monitoring and user feedback: establishment of a real-time monitoring mechanism to monitor gas use and collect user feedback information, and dynamic adjustment of the optimized gas supply strategy based on monitoring data and user feedback information.

2. The data analysis based gas platform data processing optimization method according to claim 1, characterized in that, The data collection and preprocessing in S1 includes: S11, data collection: real-time collection of gas use data, specifically including: gas pressure data collection: real-time monitoring of gas pressure in the pipeline using a pressure sensor; flow data collection: measurement of gas flow using a flow meter sensor; temperature data collection: monitoring of gas temperature changes using a temperature sensor; S12, data preprocessing: preprocessing of the collected gas use data, including data cleaning and data format conversion. 3.The data analysis based gas platform data processing optimization method according to claim 1, characterized in that, The real-time prediction in S33 includes: S331, real-time data input: the real-time collected gas pipeline network data as the node feature vector is input into the gas pipeline network fault prediction model; S332, forward propagation: performing forward propagation calculation on the node feature vector through the gas pipeline fault prediction model , sequentially passing through each layer of the graph neural network, updating the node feature representation , until the layer, obtaining the final node feature representation ; S333, failure prediction: using fully connected layers and activation functions on the final node feature representation Failure prediction is performed, denoted as: ； wherein, is a failure prediction value for a node , is a sigmoid activation function, is an output layer weight matrix, is an output layer bias; S334, Weak link and potential failure point identification: based on failure prediction value Identify weak links and potential failure points in the gas network, when a node is marked as a potential failure point when exceeding a set failure threshold.

4. The data analysis based gas panel data processing optimization method of claim 1, wherein, The user behavior analysis in S4 includes: S41, data collection: real-time collection of user's gas use data through intelligent gas meter and sensor network, forming user's gas use behavior data set , is represented as: ； wherein, for the user for the time for the user at the time gas usage; S42, behavior pattern recognition: identification of user gas use behavior patterns using a K-means clustering algorithm; S43, personalized service and suggestion generation: based on the identified gas usage behavior patterns, including intelligent gas usage suggestions and energy-saving scheme pushing, personalized services and suggestions are generated, including: S431, intelligent gas usage suggestions: based on the user's behavior patterns, intelligent gas usage suggestions are generated using the Apriori algorithm; S432, energy-saving scheme pushing: based on the user's behavior patterns and usage data, the user's gas demand is predicted using regression analysis, and energy-saving schemes are provided.

Images