Dynamic report generation method for business data of gas system

Abstract

The invention relates to the technical field of fuel gas system data management, in particular to a fuel gas system business data dynamic report generation method, which comprises the following steps: S1, data collection and preprocessing: collecting business data in a fuel gas system in real time; s2, equipment state monitoring: identifying an abnormal condition on the surface of the equipment; s3, data analysis: extracting business indexes and abnormal data; s4, cross-platform data collaboration and synchronization: carrying out real-time collaboration and synchronization on the business data of the gas system, the energy consumption data of the energy management system and the home environment data of the smart home system; s5, report generation: generating a dynamic report of the gas system; s6, report distribution: distributing the generated dynamic report to related personnel; and S7, user feedback: collecting feedback information of the user to the report. According to the method, the comprehensive dynamic report is generated, and the safety and the operation efficiency of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of gas system data management, and in particular to a method for generating a dynamic report of gas system business data. Background Art

[0002] As the digital transformation of the global energy industry accelerates, the gas industry is also facing unprecedented opportunities for change. The widespread application of technologies such as the Internet of Things, big data, and cloud computing has provided strong technical support for the intelligent management of the gas industry. At the same time, users' demand for gas services is becoming increasingly diversified and personalized, requiring gas companies to continuously improve service quality to meet market demand.

[0003] At present, the gas system occupies an important position in urban infrastructure. Its operating efficiency and safety directly affect the quality of life of urban residents and public safety. However, due to the complexity of the gas system and the high requirements for real-time monitoring, traditional monitoring and reporting methods are often difficult to meet the needs of modern management. With the development of the Internet of Things, big data and artificial intelligence technologies, how to use these advanced technologies to efficiently and accurately monitor and manage the gas system has become an urgent problem to be solved.

[0004] Although the existing gas system monitoring technology can achieve basic data collection and processing, it has many shortcomings. First, the accuracy and consistency of the data are difficult to guarantee, especially under the interference of noise and invalid data, the data quality is difficult to meet the requirements of high-precision analysis. Secondly, equipment monitoring mainly relies on manual inspections, which are inefficient and costly, and it is difficult to detect equipment abnormalities in a timely manner. Cross-platform data collaboration and synchronization have not been fully realized, resulting in the isolation of data from each system and the inability to provide comprehensive operational data support. In addition, the existing report generation methods mostly rely on static data, lack real-time dynamic analysis and user feedback mechanisms, and are difficult to adapt to the rapidly changing operating environment.

[0005] The purpose of the present invention is to provide a method for generating dynamic reports of gas system business data, thereby improving the operating efficiency and safety of the gas system, ensuring the accuracy and consistency of data, providing comprehensive and real-time operational data support, and providing a reliable basis for managers' decision-making. Summary of the invention

[0006] Based on the above purpose, the present invention provides a method for generating a dynamic report of gas system business data.

[0007] A method for generating a dynamic report of gas system business data comprises the following steps:

[0008] S1, data collection and preprocessing: collect business data in the gas system in real time through the sensor network, including gas flow, pressure and temperature, and preprocess the collected business data;

[0009] S2, Equipment Status Monitoring: Use cameras to monitor the operating status of gas system equipment in real time and identify abnormal conditions on the equipment surface, including:

[0010] S21, equipment image acquisition: The operating status of the gas system equipment is collected in real time through the camera, including the equipment appearance, operating posture and angle, surface temperature distribution and vibration;

[0011] S22, image processing: preliminary processing of the collected running status, including denoising and enhancement;

[0012] S23, anomaly detection: using a fast regional convolutional neural network to detect the operating status of the preliminary processing and identify abnormal conditions on the surface of the equipment;

[0013] S24, status recording: recording and storing detected device status and abnormal conditions;

[0014] S3, data analysis: Combined with the abnormal conditions on the equipment surface, the pre-processed business data is analyzed in real time to extract business indicators and abnormal data;

[0015] S4, cross-platform data collaboration and synchronization: real-time collaboration and synchronization of the gas system’s business data, the energy management system’s energy consumption data, and the smart home system’s home environment data to form comprehensive data, provide operational data support and optimization suggestions;

[0016] S5, report generation: Generate a dynamic report of the gas system based on the extracted business indicators and abnormal data combined with comprehensive data. The report content includes data statistics, trend analysis and abnormal alarm;

[0017] S6, report distribution: distribute the generated dynamic reports to relevant personnel through multiple channels (such as email, SMS, Web interface, etc.);

[0018] S7, User Feedback: Collect user feedback on the report and continuously optimize the report generation method.

[0019] Optionally, the data collection and preprocessing in S1 includes:

[0020] S11, data collection: real-time collection of business data in the gas system, including gas flow, pressure and temperature, including:

[0021] Gas flow data collection: Use ultrasonic flowmeters installed on pipelines to measure gas flow through transmission and reception of ultrasonic signals;

[0022] Pressure data collection: The pressure value in the pipeline is monitored in real time through the pressure sensor, which converts the pressure signal into an electrical signal;

[0023] Temperature data collection: Temperature data is collected through thermocouples, which generate voltage signals through temperature differential electromotive force;

[0024] S12, data preprocessing: preprocessing the collected business data in the gas system, including cleaning, filtering and formatting.

[0025] Optionally, the device image acquisition in S21 includes:

[0026] S211, equipment appearance status collection: using a camera to take real-time appearance images of gas system equipment;

[0027] S212, operation posture and angle acquisition: using a multi-axis camera installed on the equipment to capture the operation posture and angle changes of the equipment in real time;

[0028] S213, surface temperature distribution collection: real-time monitoring of the temperature distribution on the surface of the equipment through an infrared thermal imaging camera;

[0029] S214, vibration condition collection: real-time monitoring of the equipment's minute vibration conditions through a high-frequency camera.

[0030] Optionally, the image processing in S22 includes:

[0031] S221, denoising: using a Gaussian filter algorithm to denoise the collected image of the device operation status;

[0032] S222, image enhancement: using histogram equalization to enhance the denoised running state image.

[0033] Optionally, the fast regional convolutional neural network in S23 includes:

[0034] S231, Feature Extraction: Use Convolutional Neural Network (CNN) to extract multi-scale features from the input image, including low-level features, mid-level features, and high-level features;

[0035] S232, feature fusion: features at different levels are fused, expressed as:

[0036]

[0037] Among them, F iFeatures extracted from different layers (including low-level features F low , middle-level features F mid , high-level features F high ), w i is the fusion weight, F fusion is the fused feature;

[0038] S233, Region Proposal Network: Add Rotated RPN to adapt to abnormal detection of different angles and postures of the device, expressed as:

[0039] score rot =softmax(W rot ·F fusion );

[0040] bbox rot =W bbox ·F fusion ;

[0041] Among them, W rot and W bbox The weight of the rotation proposal, score rot is the score of the rotation proposal, bbox rot Bounding box proposed for rotation;

[0042] S234, multi-task loss function: adding constraints on temperature and vibration features, expressed as:

[0043] L=L cls +λ1L bbox +λ2L temp +λ3L vib ;

[0044]

[0045] Among them, L cls is the classification loss, L bbox is the bounding box regression loss, L temp and L vib are the losses of temperature and vibration characteristics respectively, λ1,λ2,λ3 are the balance coefficients, T i and is the actual and predicted temperature value, V i and are the actual and predicted vibration values;

[0046] S235, feature mapping layer: introduces the attention mechanism to enhance the attention to abnormal areas, expressed as:

[0047] F att =softmax(W att ·Ffusion );

[0048] F map =F fusion ⊙F att ;

[0049] Among them, W att is the weight of the attention mechanism, F att is the attention feature, F map is the final feature map, ⊙ represents the element-by-element product;

[0050] S236, anomaly detection layer: add a feature-based classifier, expressed as:

[0051] p(c)=softmax(W cls ·F map );

[0052] t=W reg ·F map ;

[0053] Among them, W cls and W reg are the weights of the classifier and regressor, p(c) is the category probability, and t is the bounding box parameter.

[0054] Optionally, the status record in S24 includes:

[0055] S241, data structuring: storing the detected device status and abnormal conditions in a structured format, including device ID, timestamp, abnormality type, location and related characteristics;

[0056] S242, database storage: using a relational database to store structured data;

[0057] S243, Data backup: Back up the recorded data regularly.

[0058] Optionally, the data analysis in S3 includes:

[0059] S31, abnormal situation fusion: The abnormal situation on the device surface is fused with the pre-processed business data to form a comprehensive data set, which is expressed as:

[0060] D fused =αD business +βD anomaly ;

[0061] Among them, D fused is the fused data set, D business is the business data, D anomaly is abnormal data, α and β are weight coefficients;

[0062] S32, real-time analysis: The real-time analysis algorithm based on the temporal convolutional network (TCN) is used to process the fused data set in real time and extract business indicators and abnormal data, which is expressed as:

[0063]

[0064] Among them, y t is the output at time t, W i is the convolution kernel weight, x t-i is the input at time ti, k is the convolution kernel size;

[0065] S33, business indicator extraction: Use the fully connected layer and Softmax function to extract business indicators and identify abnormal data, expressed as:

[0066] Fully connected layer: y = W h t +b;

[0067] Softmax function:

[0068] Among them, y is the output of the fully connected layer, W and b are weights and biases respectively, and p(y i ) is the probability of category i.

[0069] Optionally, the cross-platform data collaboration and synchronization in S4 includes:

[0070] S41, data collection: collect energy consumption data of the energy management system and home environment data of the smart home system in real time through the sensor network. The energy consumption data includes electricity consumption, water resource consumption and energy consumption indicators. The home environment data includes indoor temperature, humidity, air quality and household electricity consumption data.

[0071] S42, data standardization: standardize the business data of the gas system, the energy consumption data of the energy management system, and the home environment data of the smart home system;

[0072] S43, data alignment: align the data from different data sources according to the timestamp, expressed as:

[0073] D aligned ={(t,x1(t),x2(t),...,x n (t))};

[0074] Among them, t is the timestamp, x i (t) is the data corresponding to time t;

[0075] S44, data fusion: Use weighted fusion algorithm to fuse data from different data sources to form a comprehensive data set, expressed as:

[0076] D compreh ensive =w1D gas +w2PD energy +w3D home ;

[0077] Among them, D gas is the business data of the gas system, D energy Energy consumption data for energy management systems,

[0078] D home is the home environment data of the smart home system, w1, w2, w3 are weighted coefficients.

[0079] Optionally, the report generation in S5 includes:

[0080] S51, data integration: Integrate the extracted business indicators, abnormal data and comprehensive data to form a complete data set, expressed as:

[0081] D report ={D indicators , D anomalies ,D compreh ensive};

[0082] Among them, D indicators is the business indicator data, D anomalies is abnormal data, D compreh ensive For comprehensive data;

[0083] S52, data statistics: perform statistical analysis on the integrated data and calculate the statistical values ​​of various business indicators;

[0084] S53, Trend Analysis: Use time series analysis methods to perform trend analysis on business indicators and abnormal data to identify the changing trends and patterns of data;

[0085] S54, abnormal alarm: according to the statistical results of abnormal data, set the threshold value for abnormal detection and alarm;

[0086] S55, report generation: Integrate data statistics results, trend analysis results and abnormal alarm information to generate a dynamic report of the gas system.

[0087] Optionally, the user feedback in S7 includes:

[0088] S71, feedback collection: collect user feedback on the report through questionnaires, user evaluations and feedback forms;

[0089] S72, feedback data storage: storing the collected feedback information in a database to establish a user feedback database;

[0090] S73, Feedback data analysis: Use natural language processing (NLP) and sentiment analysis algorithms to analyze feedback comments and extract user opinions and sentiment tendencies, expressed as:

[0091]

[0092] Among them, Sentiment_Score is the sentiment score, N is the number of comments, Sentiment(c i ) is the sentiment analysis result of the i-th comment;

[0093] S74, generating optimization suggestions: generating report optimization suggestions based on the feedback data analysis results, and recording them in the optimization suggestion database;

[0094] S75, continuous optimization: According to the information in the optimization suggestion database, the report generation method is optimized regularly, which is expressed as:

[0095]

[0096] Among them, New_Meth od is the optimized method, Current_Meth od is the current method, M is the number of optimization suggestions, and Optimization_DB(i) is the i-th optimization suggestion.

[0097] Beneficial effects of the present invention:

[0098] The present invention effectively cleans, filters and formats business data such as gas flow, pressure and temperature through data collection and preprocessing steps, ensuring the reliability of the data. At the same time, the equipment status monitoring uses a camera to monitor the operating status of the equipment in real time, combines Gaussian filtering and histogram equalization technology for image processing, and uses a fast regional convolutional neural network for anomaly detection, which greatly improves the detection accuracy and sensitivity of equipment abnormalities and ensures real-time monitoring and maintenance of the equipment.

[0099] The present invention collects energy consumption data of the energy management system and home environment data of the smart home system in real time through a sensor network, standardizes and aligns data from different data sources, and uses a weighted fusion algorithm to form a comprehensive data set, providing comprehensive and synchronous data support, ensuring the consistency and comprehensiveness of data from different systems, and providing a reliable data basis for report generation. Report generation combines the extracted business indicators and abnormal data to perform data statistics, trend analysis and abnormal alarms, generate comprehensive dynamic reports, and improve the security and operating efficiency of the system.

[0100] The present invention collects user feedback information on reports through questionnaires, user evaluations and feedback forms, analyzes the feedback data using natural language processing and sentiment analysis algorithms, generates optimization suggestions, and improves the quality of reports and user satisfaction through continuous optimization, thereby ensuring continuous improvement of the report generation method, enhancing the practicality and accuracy of the report, providing managers with reliable decision-making support, and improving the operation and management level of the gas system. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0102] Figure 1 A schematic diagram of a generation method flow chart of an embodiment of the present invention;

[0103] Figure 2 Schematic diagram of equipment status monitoring according to an embodiment of the present invention. DETAILED DESCRIPTION

[0104] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0105] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0106] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0107] like Figure 1-Figure 2 As shown, a method for generating a dynamic report of gas system business data includes the following steps:

[0108] S1, data collection and preprocessing: collect business data in the gas system in real time through the sensor network, including gas flow, pressure and temperature, and preprocess the collected business data to remove noise and invalid data to ensure data accuracy and consistency;

[0109] S2, Equipment Status Monitoring: Use cameras to monitor the operating status of gas system equipment in real time and identify abnormal conditions on the equipment surface, including:

[0110] S21, equipment image acquisition: The operating status of the gas system equipment is collected in real time through the camera, including the equipment appearance, operating posture and angle, surface temperature distribution and vibration;

[0111] S22, image processing: preliminary processing of the collected running status, including denoising and enhancement;

[0112] S23, anomaly detection: using a fast regional convolutional neural network to detect the operating status of the preliminary processing and identify abnormal conditions on the surface of the equipment;

[0113] S24, status recording: recording and storing detected device status and abnormal conditions;

[0114] S3, data analysis: Combined with the abnormal conditions on the equipment surface, the pre-processed business data is analyzed in real time to extract business indicators and abnormal data;

[0115] S4, cross-platform data collaboration and synchronization: real-time collaboration and synchronization of the gas system’s business data, the energy management system’s energy consumption data, and the smart home system’s home environment data to form comprehensive data, provide operational data support and optimization suggestions;

[0116] S5, report generation: Generate a dynamic report of the gas system based on the extracted business indicators and abnormal data combined with comprehensive data. The report content includes data statistics, trend analysis and abnormal alarm;

[0117] S6, report distribution: distribute the generated dynamic reports to relevant personnel through multiple channels (such as email, SMS, Web interface, etc.) to timely grasp the operating status of the gas system;

[0118] S7, user feedback: collect user feedback on the report and continuously optimize the report generation method;

[0119] Through the above steps, the accuracy, consistency and comprehensive analysis of the data are achieved, and timely and detailed dynamic reports are provided, which help to improve the system's operating efficiency, equipment maintenance and management level, timely discover and solve potential problems, and ensure the safe and reliable operation of the gas system.

[0120] Data collection and preprocessing in S1 include:

[0121] S11, data collection: real-time collection of business data in the gas system, including gas flow, pressure and temperature, including:

[0122] Gas flow data collection: Using an ultrasonic flow meter installed on the pipeline, the gas flow is measured by transmitting and receiving ultrasonic signals, expressed as:

[0123]

[0124] Where Q is the flow rate, K is the calibration coefficient, T1 and T2 are the propagation time of the ultrasonic signal respectively;

[0125] Pressure data collection: The pressure value in the pipeline is monitored in real time through the pressure sensor. The pressure sensor converts the pressure signal into an electrical signal, which is expressed as:

[0126] P=V×K p ;

[0127] Among them, P is the pressure value, V is the voltage signal output by the sensor, K p is the sensitivity coefficient of the sensor;

[0128] Temperature data collection: Temperature data is collected through thermocouples. Thermocouples generate voltage signals through temperature difference electromotive force, which can be expressed as:

[0129]

[0130] Where, T is the temperature value, V is the voltage signal output by the sensor, K t is the temperature coefficient of the sensor;

[0131] S12, data preprocessing: preprocessing the collected business data in the gas system, including cleaning, filtering and formatting;

[0132] Data cleaning: Use median filtering to remove noise from the data, expressed as:

[0133] x i =Median(x i-(n-1) / 2 ,...,x i+(n-1) / 2 );

[0134] Among them, x iis the current data point, n is the size of the filter window;

[0135] Data filtering: Filter out abnormal data by setting a threshold range. The threshold range can be set based on historical data and actual needs, expressed as:

[0136]

[0137] Among them, L and U are the lower and upper thresholds of the data respectively;

[0138] Data formatting: Convert the cleaned and filtered data into a unified format for subsequent processing and analysis, expressed as:

[0139]

[0140] Among them, x i ′ is the standardized data, x i is the original data, μ is the mean of the data, and σ is the standard deviation of the data;

[0141] The above contents ensure the accuracy and consistency of the gas system business data, which helps to improve the overall operational efficiency and management level of the system.

[0142] Device image acquisition in S21 includes:

[0143] S211, equipment appearance status collection: using a camera to take real-time appearance images of gas system equipment;

[0144] S212, operation posture and angle acquisition: using a multi-axis camera installed on the equipment to capture the operation posture and angle changes of the equipment in real time;

[0145] S213, surface temperature distribution collection: real-time monitoring of the temperature distribution on the surface of the equipment through an infrared thermal imaging camera;

[0146] S214, vibration condition collection: real-time monitoring of the equipment's minute vibration condition through a high-frequency camera;

[0147] Through the above content, it is possible to provide comprehensive equipment operation status data, quickly and accurately identify equipment abnormalities, reduce the workload of manual inspections, improve the accuracy and timeliness of monitoring, ensure the safe operation and timely maintenance of equipment, thereby reducing the risk of system failures and improving the overall operational efficiency of the gas system.

[0148] Image processing in S22 includes:

[0149] S221, denoising: using a Gaussian filtering algorithm to perform denoising on the collected image of the device operation status to reduce the noise in the image, which is expressed as:

[0150]

[0151] Where I′(x,y) is the pixel value of the denoised image, I(x,y) is the pixel value of the original image, σ is the standard deviation of the Gaussian function, and k is the window radius;

[0152] S222, image enhancement: the image of the running state after denoising is enhanced by using histogram equalization, which is expressed as:

[0153]

[0154] Where I″(x, y) is the pixel value of the enhanced image, I′(x, y) is the pixel value of the denoised image, min(I′) and max(I′) are the minimum and maximum pixel values ​​of the denoised image, respectively;

[0155] Through the above steps, the preliminary processing of the collected images of the equipment operating status is achieved, including denoising and enhancement, to ensure the clarity and accuracy of the image data, provide high-quality data support for subsequent anomaly detection and status recording, and thus improve the reliability and effectiveness of gas system equipment monitoring.

[0156] The fast regional convolutional neural network in S23 includes:

[0157] S231, Feature Extraction: Use Convolutional Neural Network (CNN) to extract multi-scale features from the input image, including low-level features, mid-level features, and high-level features;

[0158] Low-level feature extraction is used to capture detailed information of the image (such as edges and textures), expressed as:

[0159] F low =σ(W low *I+b low );

[0160] Among them, F low is the low-level feature, W low is the convolution kernel, I is the input image, b low is the bias, σ is the ReLU activation function;

[0161] The middle-level feature extraction is used to capture the shape of the image and the local structure of the object, which is expressed as:

[0162] F mid =σ(W mid *F low +b mid );

[0163] Among them, F mid is the middle-level feature, W mid is the convolution kernel, Flow is the low-level feature, b mid is bias;

[0164] High-level feature extraction is used to capture the semantic information of the image (such as category and global context), expressed as:

[0165] F high =σ(w high *F mid +b high );

[0166] Among them, F high is a high-level feature, W high is the convolution kernel, F mid is the middle-level feature, b high is bias;

[0167] S232, feature fusion: fuse features at different levels to improve the detection capability of anomalies of different sizes and levels, expressed as:

[0168]

[0169] Among them, F i Features extracted from different layers (including low-level features F low , middle-level features F mid , high-level features F high ), w i is the fusion weight, F fusion is the fused feature;

[0170] S233, Region Proposal Network: Add Rotated RPN to adapt to abnormal detection of different angles and postures of the device, expressed as:

[0171] score rot =softmax(W rot ·F fusion );

[0172] bbox rot =W bbox ·F fusion ;

[0173] Among them, W rot and W bbox The weight of the rotation proposal, score rot is the score of the rotation proposal, bbox rot Bounding box proposed for rotation;

[0174] S234, multi-task loss function: add constraints on temperature and vibration features to increase sensitivity to these features, expressed as:

[0175] L=L cls +λ1L bbox +λ2L temp +λ3L vib ;

[0176]

[0177]

[0178] Among them, L cls is the classification loss, L bbox is the bounding box regression loss, L temp and L vib are the losses of temperature and vibration characteristics respectively, λ1,λ2,λ3 are the balance coefficients, T i and is the actual and predicted temperature value, V i and are the actual and predicted vibration values;

[0179] S235, feature mapping layer: introduces the attention mechanism to enhance the attention to abnormal areas, expressed as:

[0180] F att =softmax(W att ·F fusion );

[0181] F map =F fusion ⊙F att ;

[0182] Among them, W att is the weight of the attention mechanism, F att is the attention feature, F map is the final feature map, ⊙ represents the element-by-element product;

[0183] S236, anomaly detection layer: add feature-based classifiers to improve the accuracy of anomaly detection, expressed as:

[0184] p(c)=softmax(W cls ·F map );

[0185] t=W reg ·F map ;

[0186] Among them, W cls and W reg are the weights of the classifier and regressor, p(c) is the category probability, and t is the bounding box parameter;

[0187] Through the above contents, the ability to detect abnormalities of different sizes, angles and levels of gas system equipment is enhanced, and the detection accuracy and sensitivity of abnormal conditions on the equipment surface (such as cracks, corrosion, temperature abnormalities and vibration abnormalities) are improved, ensuring real-time monitoring and reliability maintenance of equipment, thereby effectively improving the safety and operation efficiency of the gas system.

[0188] The status records in S24 include:

[0189] S241, data structuring: storing the detected device status and abnormal conditions in a structured format, including device ID, timestamp, abnormality type, location and related characteristics;

[0190] S242, Database Storage: Use relational databases to store structured data to ensure efficient retrieval and management of data;

[0191] S243, Data backup: Back up recorded data regularly to ensure data security and recoverability;

[0192] Through the above steps, the detected equipment status and abnormal conditions can be efficiently recorded and stored, providing reliable data support for subsequent monitoring, analysis and maintenance, ensuring the safe operation and timely maintenance of gas system equipment.

[0193] Data analysis in S3 includes:

[0194] S31, abnormal situation fusion: The abnormal situation on the device surface is fused with the pre-processed business data to form a comprehensive data set, which is expressed as:

[0195] D fused =αD business +βD anomaly ;

[0196] Among them, D fused is the fused data set, D business is the business data, D anomaly is abnormal data, α and β are weight coefficients;

[0197] S32, real-time analysis: The real-time analysis algorithm based on the temporal convolutional network (TCN) is used to process the fused data set in real time and extract business indicators and abnormal data, which is expressed as:

[0198]

[0199] Among them, y t is the output at time t, W i is the convolution kernel weight, x t-i is the input at time ti, k is the convolution kernel size;

[0200] S33, business indicator extraction: Use the fully connected layer and Softmax function to extract business indicators and identify abnormal data, expressed as:

[0201] Fully connected layer: y = W ht + b;

[0202] Softmax function:

[0203] Among them, y is the output of the fully connected layer, W and b are weights and biases respectively, and p(y i ) is the probability of category i;

[0204] Through the above steps, the accuracy and real-time performance of data analysis are improved, ensuring the data integrity and reliability in the process of generating dynamic reports on gas system business data.

[0205] Cross-platform data collaboration and synchronization in S4 include:

[0206] S41, data collection: collect energy consumption data of the energy management system and home environment data of the smart home system in real time through the sensor network. The energy consumption data includes electricity consumption, water resource consumption and energy consumption indicators. The home environment data includes indoor temperature, humidity, air quality and household electricity consumption data.

[0207] S42, data standardization: standardize the business data of the gas system, the energy consumption data of the energy management system, and the home environment data of the smart home system so that the data from different data sources have the same scale and unit, expressed as:

[0208]

[0209] Among them, x is the original data, μ is the data mean, σ is the data standard deviation, and x′ is the standardized data;

[0210] S43, data alignment: align data from different data sources according to timestamps to ensure data synchronization, expressed as:

[0211] D aligned ={(t,x1(t),x2(t),...,x n (t))};

[0212] Among them, t is the timestamp, x i (t) is the data corresponding to time t;

[0213] S44, data fusion: Use weighted fusion algorithm to fuse data from different data sources to form a comprehensive data set, expressed as:

[0214] D compreh ensive =w1D gas+w2D energy +w3D h ome ;

[0215] Among them, D gas is the business data of the gas system, D energy is the energy consumption data of the energy management system, D h ome is the home environment data of the smart home system, w1, w2, w3 are weighted coefficients;

[0216] Through the above steps, cross-platform data collaboration and synchronization can coordinate and synchronize the business data of the gas system, the energy consumption data of the energy management system, and the home environment data of the smart home system in real time, ensuring the comprehensiveness of the data of each system. Data standardization and alignment ensure the consistency and synchronization of data from different data sources. The weighted fusion algorithm combines information from different data sources to form a comprehensive data set, providing a reliable data basis for subsequent report generation.

[0217] Report generation in S5 includes:

[0218] S51, data integration: Integrate the extracted business indicators, abnormal data and comprehensive data to form a complete data set, expressed as:

[0219] D report ={D indicators , D anomalies ,D compreh ensive};

[0220] Among them, D indicators is the business indicator data, D anomalies is abnormal data, D compreh ensive For comprehensive data;

[0221] S52, data statistics: perform statistical analysis on the integrated data and calculate the statistical values ​​of various business indicators, expressed as:

[0222]

[0223] Among them, μ is the mean, σ is the standard deviation, and x i is the i-th data point, N is the number of data points;

[0224] S53, Trend Analysis: Use time series analysis methods to perform trend analysis on business indicators and abnormal data, identify the changing trends and patterns of data, expressed as:

[0225]

[0226] Among them, MA t is the moving average of time t, k is the window size;

[0227] S54, abnormal alarm: according to the statistical results of abnormal data, set the threshold for abnormal detection and alarm, which is expressed as:

[0228]

[0229] Among them, x is the abnormal data point, and th resh old is the set threshold;

[0230] Threshold settings include:

[0231] Historical data analysis: collect a large amount of historical data, including normal and abnormal operation data, and calculate the statistical characteristics of historical data, such as mean and standard deviation;

[0232] Determine threshold range: Set thresholds within one or more standard deviations;

[0233] Industry standards and experience: Refer to industry standards and specifications to set reasonable thresholds;

[0234] Dynamic adjustment: After setting the initial threshold, continuously monitor the system's operation and dynamically adjust the threshold based on actual operation data and feedback to ensure that it adapts to system changes;

[0235] S55, report generation: integrating data statistics results, trend analysis results and abnormal alarm information to generate a dynamic report of the gas system;

[0236] Through the above content, the integrity and accuracy of the report data are ensured, key business indicators are provided through detailed data statistics, data change patterns are identified through trend analysis, and abnormal situations are discovered and reported in time through abnormal alarms. The final generated dynamic report content includes data statistics, trend analysis and abnormal alarms, providing managers with comprehensive and timely decision-making support and improving the security and efficiency of system operation.

[0237] User feedback in S7 includes:

[0238] S71, Feedback Collection: Collect user feedback on the report through questionnaires, user evaluations and feedback forms, including:

[0239] Questionnaire survey: Questionnaires are sent to users regularly to collect opinions on report content, format, timeliness, etc.;

[0240] User evaluation: Set up a user evaluation function on the report platform, and users can rate and comment on each report;

[0241] Feedback form: Provide an online feedback form where users can fill in specific feedback and suggestions;

[0242] S72, feedback data storage: storing the collected feedback information in a database to establish a user feedback database;

[0243] S73, Feedback data analysis: Use natural language processing (NLP) and sentiment analysis algorithms to analyze feedback comments and extract user opinions and sentiment tendencies, expressed as:

[0244]

[0245] Among them, Sentiment_Score is the sentiment score, N is the number of comments, Sentiment(c i ) is the sentiment analysis result of the i-th comment;

[0246] S74, generating optimization suggestions: generating report optimization suggestions based on the feedback data analysis results, and recording them in the optimization suggestion database;

[0247] S75, continuous optimization: According to the information in the optimization suggestion database, the report generation method is optimized regularly, which is expressed as:

[0248]

[0249] Among them, New_Method is the optimized method, Current_Method is the current method, M is the number of optimization suggestions, and Optimization_DB(i) is the i-th optimization suggestion;

[0250] Through the above content, the quality of the report is ensured to be continuously improved, which can better meet user needs and improve user satisfaction. At the same time, the practicality and accuracy of the report are enhanced, providing managers with more reliable decision-making support.

[0251] Improve decision-making efficiency:

[0252] Quickly and accurately obtain key business indicators to help management make more informed strategic decisions. Dynamic reports can provide immediate feedback, allowing managers to respond quickly to market changes or internal issues.

[0253] Refined management:

[0254] Dynamic reports can reveal bottlenecks in business processes and help companies optimize resource allocation. Through in-depth analysis of data, opportunities for improvement can be found, such as reducing energy waste and improving equipment utilization.

[0255] Strengthen risk management:

[0256] Identify potential risks and threats in a timely manner and take preventive measures to reduce losses. Through predictive analysis, make emergency preparations in advance to reduce the possibility of security incidents.

[0257] Promote business innovation:

[0258] Data-driven decision making helps explore new business models and services. Use data analysis to discover market trends and develop new products or services to meet emerging needs.

[0259] Improve compliance and security:

[0260] Ensure all operations comply with laws and regulations to avoid fines and other legal consequences. Strengthen network security measures to protect sensitive data from unauthorized access.

[0261] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0262] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for generating a dynamic report of gas system business data, characterized in that: The following steps are involved: S1, data collection and preprocessing: collect business data in the gas system in real time through the sensor network, including gas flow, pressure and temperature, and preprocess the collected business data; S2, Equipment Status Monitoring: Use cameras to monitor the operating status of gas system equipment in real time and identify abnormal conditions on the equipment surface, including: S21, equipment image acquisition: The operating status of the gas system equipment is collected in real time through the camera, including the equipment appearance, operating posture and angle, surface temperature distribution and vibration; S22, image processing: preliminary processing of the collected running status, including denoising and enhancement; S23, anomaly detection: using a fast regional convolutional neural network to detect the operating status of the preliminary processing and identify abnormal conditions on the surface of the equipment; S24, status recording: recording and storing detected device status and abnormal conditions; S3, data analysis: Combined with the abnormal conditions on the equipment surface, the pre-processed business data is analyzed in real time to extract business indicators and abnormal data; S4, cross-platform data collaboration and synchronization: real-time collaboration and synchronization of the gas system’s business data, the energy management system’s energy consumption data, and the smart home system’s home environment data to form comprehensive data, provide operational data support and optimization suggestions; S5, report generation: Generate a dynamic report of the gas system based on the extracted business indicators and abnormal data combined with comprehensive data. The report content includes data statistics, trend analysis and abnormal alarm; S6, report distribution: distribute the generated dynamic reports to relevant personnel through multiple channels; S7, User Feedback: Collect user feedback on the report and continuously optimize the report generation method.

2. A method for generating a dynamic report of gas system business data according to claim 1, characterized in that: The data collection and preprocessing in S1 include: S11, data collection: real-time collection of business data in the gas system, including gas flow, pressure and temperature, including: Gas flow data collection: Use ultrasonic flowmeters installed on pipelines to measure gas flow through transmission and reception of ultrasonic signals; Pressure data collection: The pressure value in the pipeline is monitored in real time through the pressure sensor, which converts the pressure signal into an electrical signal; Temperature data collection: Temperature data is collected through thermocouples, which generate voltage signals through temperature differential electromotive force; S12, data preprocessing: preprocessing the collected business data in the gas system, including cleaning, filtering and formatting.

3. A method for generating a dynamic report of gas system business data according to claim 1, characterized in that: The device image acquisition in S21 includes: S211, equipment appearance status collection: using a camera to take real-time appearance images of gas system equipment; S212, operation posture and angle acquisition: using a multi-axis camera installed on the equipment to capture the operation posture and angle changes of the equipment in real time; S213, surface temperature distribution collection: real-time monitoring of the temperature distribution on the surface of the equipment through an infrared thermal imaging camera; S214, vibration condition collection: real-time monitoring of the equipment's minute vibration conditions through a high-frequency camera.

4. A method for generating a dynamic report of gas system business data according to claim 3, characterized in that: The image processing in S22 includes: S221, denoising: using a Gaussian filter algorithm to denoise the collected image of the device operation status; S222, image enhancement: using histogram equalization to enhance the denoised running state image.

5. A method for generating a dynamic report of gas system business data according to claim 4, characterized in that: The fast regional convolutional neural network in S23 includes: S231, Feature Extraction: Use convolutional neural networks to extract multi-scale features from the input image, including low-level features, mid-level features, and high-level features; S232, feature fusion: features at different levels are fused, expressed as: Among them, F i are the features extracted from different layers, w i is the fusion weight, F fusion is the fused feature; S233, Region Proposal Network: Add rotation region proposal to adapt to abnormal detection of different angles and postures of the device, expressed as: score rot =softmax(W rot ·F fusion ); bbox rot =W bbox ·F fusion ; Among them, W rot and W bbox The weight of the rotation proposal, score rot is the score of the rotation proposal, bbox rot Bounding box proposed for rotation; S234, multi-task loss function: adding constraints on temperature and vibration features, expressed as: L=L cls +λ1L bbox +λ2L temp +λ3L vib ; Among them, L cls is the classification loss, L bbox is the bounding box regression loss, L temp and L vib are the losses of temperature and vibration characteristics respectively, λ1,λ2,λ3 are the balance coefficients, T i and is the actual and predicted temperature value, V i and are the actual and predicted vibration values; S235, feature mapping layer: introduces the attention mechanism to enhance the attention to abnormal areas, expressed as: F att =softmax(W att ·F fusion ); F map =F fusion ⊙F att ; Among them, W att is the weight of the attention mechanism, F att is the attention feature, F map is the final feature map, ⊙ represents the element-by-element product; S236, anomaly detection layer: add a feature-based classifier, expressed as: p(c)=softmax(W cls ·F map ); t=W reg ·F map ; Among them, W cls and W reg are the weights of the classifier and regressor, p(c) is the category probability, and t is the bounding box parameter.

6. A method for generating a dynamic report of gas system business data according to claim 5, characterized in that: The status record in S24 includes: S241, data structuring: storing the detected device status and abnormal conditions in a structured format, including device ID, timestamp, abnormality type, location and related characteristics; S242, database storage: using a relational database to store structured data; S243, Data backup: Back up the recorded data regularly.

7. A method for generating a dynamic report of gas system business data according to claim 6, characterized in that: The data analysis in S3 includes: S31, abnormal situation fusion: The abnormal situation on the device surface is fused with the pre-processed business data to form a comprehensive data set, which is expressed as: D fused =αD business +βD anomaly ; Among them, D fused is the fused data set, D business is the business data, D anomaly is abnormal data, α and β are weight coefficients; S32, real-time analysis: using the real-time analysis algorithm based on the time series convolutional network, the fused data set is processed in real time to extract business indicators and abnormal data, expressed as: Among them, y t is the output at time t, W i is the convolution kernel weight, x t-i is the input at time ti, k is the convolution kernel size; S33, business indicator extraction: Use the fully connected layer and Softmax function to extract business indicators and identify abnormal data, expressed as: Fully connected layer: y = W h t +b; Softmax function: Among them, y is the output of the fully connected layer, W and b are weights and biases respectively, and p(y i ) is the probability of category i.

8. A method for generating a dynamic report of gas system business data according to claim 1, characterized in that: The cross-platform data collaboration and synchronization in S4 includes: S41, data collection: collect energy consumption data of the energy management system and home environment data of the smart home system in real time through the sensor network. The energy consumption data includes electricity consumption, water resource consumption and energy consumption indicators. The home environment data includes indoor temperature, humidity, air quality and household electricity consumption data. S42, data standardization: standardize the business data of the gas system, the energy consumption data of the energy management system, and the home environment data of the smart home system; S43, data alignment: align the data from different data sources according to the timestamp, expressed as: D aligned ={(t,x1(t),x2(t),...,x n (t))}; Among them, t is the timestamp, x i (t) is the data corresponding to time t; S44, data fusion: Use weighted fusion algorithm to fuse data from different data sources to form a comprehensive data set, expressed as: D comprehensive =w1D gas +w2D energy +w3D home ; Among them, D gas is the business data of the gas system, D energy is the energy consumption data of the energy management system, D home is the home environment data of the smart home system, w1, w2, w3 are weighted coefficients.

9. A method for generating dynamic reports of gas system business data according to claim 1, characterized in that: The report generation in S5 includes: S51, data integration: Integrate the extracted business indicators, abnormal data and comprehensive data to form a complete data set, expressed as: D report ={D indicators ,D anomalie ,D comprehensive }; Among them, D indicators is the business indicator data, D anomalies is abnormal data, D comprehensive For comprehensive data; S52, data statistics: perform statistical analysis on the integrated data and calculate the statistical values ​​of various business indicators; S53, Trend Analysis: Use time series analysis methods to perform trend analysis on business indicators and abnormal data to identify the changing trends and patterns of data; S54, abnormal alarm: according to the statistical results of abnormal data, set the threshold value for abnormal detection and alarm; S55, report generation: Integrate data statistics results, trend analysis results and abnormal alarm information to generate a dynamic report of the gas system.

10. A method for generating dynamic reports of gas system business data according to claim 1, characterized in that ,The user feedback in S7 includes: S71, feedback collection: collect user feedback on the report through questionnaires, user evaluations and feedback forms; S72, feedback data storage: storing the collected feedback information in a database to establish a user feedback database; S73, Feedback data analysis: Use natural language processing and sentiment analysis algorithms to analyze feedback comments and extract user opinions and sentiment tendencies, expressed as: Among them, Sentiment_Score is the sentiment score, N is the number of comments, Sentiment(c i ) is the sentiment analysis result of the i-th comment; S74, generating optimization suggestions: generating report optimization suggestions based on the feedback data analysis results, and recording them in the optimization suggestion database; S75, continuous optimization: According to the information in the optimization suggestion database, the report generation method is optimized regularly, which is expressed as: Among them, Mew_Method is the optimized method, Current_Method is the current method, M is the number of optimization suggestions, and Optimization_DB(i) is the i-th optimization suggestion.

Images