Pipeline state early warning method and system based on data driving

Abstract

The invention provides a pipeline state early warning method and system based on data driving, and relates to the technical field of environmental engineering, and the method comprises the steps: obtaining first operation state data of a pipeline in real time through a first sensor, carrying out the preprocessing, and extracting the features of a first pipeline state, the control system triggers an emergency response mechanism to adjust the state of the pipeline so as to obtain an adjusted pipeline; after the pipeline is adjusted, a second sensor is used for monitoring second running state data of the pipeline in real time and preprocessing the second running state data, model training is conducted according to the pipeline state early warning requirement and historical pipeline state data, and the trained model is used for recognizing the preprocessed second running state data of the pipeline; and judging whether the pipeline is in an abnormal state or not according to the state recognition result and a preset threshold value, and controlling the system to start an emergency response mechanism for the pipeline state according to the early warning information. According to the method, the problem that potential safety hazards are brought to downstream equipment due to the fact that residual anomalies cannot be effectively removed is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental engineering, and in particular to a data-driven pipeline status early warning method and system. Background Art

[0002] With the development of science and technology, pipeline transportation plays an increasingly important role in the fields of energy, chemical industry, water conservancy, etc. However, pipelines may be affected by various factors during operation, such as corrosion, wear, external force damage, etc., which may cause changes in pipeline status and even cause safety accidents. Therefore, real-time monitoring and early warning of pipeline status, timely discovery and handling of potential safety hazards are of great significance to ensure the safe operation of pipelines. The existing technology uses a spark detector on the pipeline to detect and extinguish the spark once, but it is impossible to confirm whether the spark has been extinguished. Once there is a hidden danger, sparks or hot particles will enter downstream process equipment, such as dust collector collection systems, drying towers and ovens, cyclone separators, silos and mechanical processing equipment, etc., and bring the risk of explosion to these equipment. Summary of the invention

[0003] The present invention provides a data-driven pipeline status early warning method and system, which solves the problem that residual anomalies cannot be effectively removed, causing potential safety hazards to downstream equipment.

[0004] In order to solve the above technical problems, the technical solution of the present invention is as follows: In a first aspect, a data-driven pipeline status early warning method is provided, the method comprising: Using a first sensor to acquire first operating state data of the pipeline in real time and preprocessing the data to obtain preprocessed first operating state data of the pipeline, wherein the first operating state data of the pipeline includes temperature, pressure, flow, vibration, and spark presence; Extracting the features of the first pipeline state according to the preprocessed pipeline first operation state data, wherein the features of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal features; According to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain an adjusted pipeline; After the pipeline is adjusted, the second operating state data of the pipeline is monitored in real time using the second sensor and preprocessed to obtain the preprocessed second operating state data of the pipeline; Model training is performed according to pipeline status warning requirements and historical pipeline status data, and the trained model is used to identify the preprocessed pipeline second operation status data to obtain a status recognition result; According to the state recognition results and the preset threshold, determine whether the pipeline is in an abnormal state to obtain early warning information; Based on the early warning information, the control system initiates an emergency response mechanism for the pipeline status.

[0005] Further, the first operating state data of the pipeline is acquired in real time using the first sensor and preprocessed to obtain the preprocessed first operating state data of the pipeline, wherein the first operating state data of the pipeline includes temperature, pressure, flow, vibration, and spark presence, including: Using a first sensor to collect first operating status data of the pipeline in real time; removing outliers from the first operating state data to obtain cleaned data; Performing normalization processing on the cleaned data to obtain normalized data; The normalized data are sorted in chronological order to obtain an ordered data set, namely, the preprocessed first operation state data of the pipeline.

[0006] Further, according to the preprocessed pipeline first operation state data, the characteristics of the first pipeline state are extracted, and the characteristics of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics, including: According to the pre-processed first operation state data of the pipeline, a time series analysis is performed on the temperature data to calculate the rate of change of the temperature over time to obtain the temperature change trend; According to the pre-processed pipeline first operation state data, statistical analysis is performed on the pressure data to calculate the standard deviation or range of the pressure value to obtain the pressure fluctuation amplitude; According to the pre-processed pipeline first operation state data, time series analysis is performed on the flow data, and the rate of change of the flow over time is calculated to obtain the flow change rate; According to the pre-processed pipeline first operation state data, spectrum analysis is performed on the vibration data to identify the main frequency components in the vibration signal to obtain the vibration frequency characteristics; According to the pre-processed pipeline first operation state data, the pre-processed spark presence signal is analyzed to extract the spark signal characteristics, and the spark signal characteristics include the spark occurrence frequency, intensity, and duration; The extracted temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics are integrated to obtain the characteristics of the first pipeline state.

[0007] Further, according to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain an adjusted pipeline, including: According to the characteristics of the first pipeline state, the first pipeline state is compared with a preset safety threshold to obtain a comparison result; According to the comparison result, the risk level of the pipeline status is evaluated to obtain an evaluation result; According to the assessment results, if the risk level does not exceed the preset threshold, the emergency response mechanism will not be triggered; Based on the assessment results, if the risk level exceeds the preset threshold, the emergency response mechanism is triggered; The emergency response mechanism adjusts the pipeline status according to the specific characteristics and risk level of the pipeline status. The pipeline status adjustment includes adjusting pipeline process parameters, closing or opening valves, and starting or stopping fans.

[0008] Further, after the pipeline is adjusted, the second operating state data of the pipeline is monitored in real time using a second sensor and preprocessed to obtain the preprocessed second operating state data of the pipeline, including: Using a second sensor to collect second operating status data of the pipeline in real time; removing abnormal values ​​from the second operating state data to obtain cleaned second data; Performing normalization processing on the cleaned second data to obtain normalized second data; The normalized second data are sorted in chronological order to obtain an ordered second data set, that is, the preprocessed second operation status data of the pipeline.

[0009] Furthermore, model training is performed according to pipeline status warning requirements and historical pipeline status data, and the trained model is used to identify the preprocessed pipeline second operation status data to obtain a status recognition result, including: According to the pipeline status warning requirements, use The prediction model is constructed, where Is an explicit decision function used to judge the input data Category, is the Lagrange multiplier, is a heterokernel function used to calculate the input data Training materials The similarity between is the bias term, N is the total data, x is the input data, i.e. the second running state data of the pipeline after preprocessing, are the data points in the training dataset, where is the width parameter of the kernel function, which controls the radial range of the kernel function. Represents input data With training data The Euclidean distance between Based on historical pipeline status data, use , The prediction model is trained to obtain a trained prediction model, wherein: is the training data point Tags, is the regularization parameter, For training data points Tags, is the Lagrange multiplier, N is the total data, is the kernel function, calculating the data points and The similarity between The trained prediction model is used to identify the preprocessed pipeline second operation state data to obtain a state identification result.

[0010] Furthermore, according to the state recognition result and the preset threshold, it is determined whether the pipeline is in an abnormal state to obtain early warning information, including: According to the state recognition result, a comparison is performed with a preset threshold to obtain a comparison result, wherein the comparison result includes an abnormal state and a normal state; If the state recognition result exceeds the preset threshold, the pipeline is judged to be in an abnormal state; If the state recognition result does not exceed the preset threshold, the pipeline is judged to be in a normal state; If the judgment result is an abnormal state, the early warning mechanism is triggered to obtain early warning information according to the preset rules.

[0011] In the second aspect, a data-driven pipeline status early warning system includes: an acquisition module, configured to acquire first operation status data of the pipeline in real time using a first sensor and perform preprocessing to obtain preprocessed first operation status data of the pipeline, wherein the first operation status data of the pipeline includes temperature, pressure, flow, vibration, and spark presence; and after the pipeline is adjusted, to monitor second operation status data of the pipeline in real time using a second sensor and perform preprocessing to obtain preprocessed second operation status data of the pipeline; The processing module is used to extract the characteristics of the first pipeline state according to the pre-processed first operation state data of the pipeline, wherein the characteristics of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics; according to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain the adjusted pipeline; according to the pipeline state warning requirements and historical pipeline state data, the model is trained, and the pre-processed second operation state data of the pipeline is identified using the trained model to obtain a state identification result; according to the state identification result and a preset threshold, it is determined whether the pipeline is in an abnormal state to obtain early warning information; according to the early warning information, the control system starts the emergency response mechanism for the pipeline state.

[0012] According to a third aspect, a computing device includes: one or more processors; The storage system is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the above method.

[0013] In a fourth aspect, a computer-readable storage medium stores a program, and the program implements the above method when executed by a processor.

[0014] The above solution of the present invention includes at least the following beneficial effects: According to the above-mentioned scheme of the present invention, through the precise detection of the first sensor, the system can quickly identify and extinguish most of the abnormalities in the pipeline, effectively reducing the risk of the pipeline; the second sensor, as a monitoring means, further enhances the safety of the system, and can continuously monitor the abnormal conditions in the downstream dust collection system, and promptly discover and control the trend of the abnormal conditions spreading downstream, and the real-time monitoring function enables the system to respond quickly to avoid potential safety hazards; when the system detects an abnormal situation, the fire damper will be closed immediately, and the fan will be controlled to stop running, which not only cuts off the physical channel for the abnormal situation to spread downstream, but also eliminates the power source for its propagation; through the dual effects of extinguishing and monitoring, the danger of the pipeline in the industrial dust process system is significantly reduced, which not only protects the safety of production equipment and employees, but also ensures the continuity and stability of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flow chart of a data-driven pipeline status early warning method provided by an embodiment of the present invention; Figure 2 is a schematic diagram of a data-driven pipeline status early warning system provided by an embodiment of the present invention; Figure 3 It is a structural diagram of a specific embodiment provided by an embodiment of the present invention; Figure 4 It is a control circuit diagram of a specific embodiment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0017] like Figure 1 As shown, an embodiment of the present invention proposes a pipeline status early warning method based on data drive, the method comprising: 11. Using a first sensor to acquire first operation status data of a pipeline in real time and preprocessing the data to obtain preprocessed first operation status data of the pipeline, wherein the first operation status data of the pipeline includes temperature, pressure, flow, vibration, and spark presence; 12. Extracting the characteristics of the first pipeline state according to the preprocessed pipeline first operation state data, wherein the characteristics of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics; 13. According to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain an adjusted pipeline; 14. After the pipeline is adjusted, the second operating state data of the pipeline is monitored in real time using the second sensor and preprocessed to obtain the preprocessed second operating state data of the pipeline; 15. Perform model training according to pipeline status warning requirements and historical pipeline status data, and use the trained model to identify the preprocessed pipeline second operating status data to obtain a status recognition result; 16. According to the state recognition result and the preset threshold, determine whether the pipeline is in an abnormal state to obtain early warning information; 17. Based on the early warning information, the control system initiates the emergency response mechanism for the pipeline status.

[0018] In the embodiment of the present invention, a first sensor is used to obtain first operation status data of the pipeline in real time. These data include key parameters such as temperature, pressure, flow, vibration, and spark presence. The sensor, as the "eyes" and "ears" of the pipeline real-time monitoring system, can monitor the operation status of the pipeline in real time and transmit the collected data to the data center by wired or wireless means; the acquired data is preprocessed to remove noise, fill missing values, standardize data, etc. to ensure the quality and consistency of the data. The preprocessed data will be used for subsequent feature extraction and model training; based on the preprocessed first operation status data of the pipeline, the features of the first pipeline state are extracted, and the features include temperature change trend, The characteristics of the pipeline state, such as potential, pressure fluctuation amplitude, flow rate change, vibration frequency and spark signal characteristics, are extracted as an important step in machine learning. It helps to extract the most representative and distinguishing information from the raw data for subsequent analysis and early warning. According to the characteristics of the extracted first pipeline state, the control system can trigger the emergency response mechanism to adjust the pipeline state, such as adjusting the pipeline pressure, flow and other parameters to maintain the normal operation of the pipeline. The triggering of the emergency response mechanism helps to find problems in time and take measures to avoid further deterioration of the situation. Model training is carried out according to the pipeline state early warning needs and historical pipeline state data. Model training is a key step in machine learning and helps to build a pipeline that can accurately A model for identifying pipeline status. The trained model will be used to identify the pre-processed second operation status data of the pipeline. The trained model is used to identify the pre-processed second operation status data of the pipeline to obtain a status identification result. The status identification result will be used for subsequent judgment and early warning. According to the status identification result and the preset threshold, it is judged whether the pipeline is in an abnormal state. The preset threshold is set according to historical data and experience, and is used to judge whether the pipeline status exceeds the normal range. If the pipeline status exceeds the normal range, an early warning message is triggered. According to the early warning message, the control system starts an emergency response mechanism for the pipeline status. The emergency response mechanism may include shutting down the pipeline, starting the backup pipeline, and notifying maintenance personnel. By acquiring pipeline operation status data in real time and performing preprocessing and feature extraction, the method can more accurately identify pipeline status and trigger early warning information, which helps to reduce false alarms and missed alarms and improve the accuracy of early warnings. By timely triggering the emergency response mechanism to adjust the pipeline status and initiating early warning and emergency response measures when necessary, the method can reduce the risk of pipeline accidents, help protect people's lives and property safety, and reduce adverse effects on the environment. The method can realize real-time monitoring and early warning of pipeline status, help to timely discover and solve potential problems, help improve pipeline operation efficiency, and reduce losses caused by downtime due to failures.

[0019] like Figure 1As shown, 11, using a first sensor to obtain first operating state data of the pipeline in real time and preprocessing to obtain preprocessed first operating state data of the pipeline, wherein the first operating state data of the pipeline includes temperature, pressure, flow, vibration, and spark presence, including: Using a first sensor to collect first operating status data of the pipeline in real time; removing outliers from the first operating state data to obtain cleaned data; Performing normalization processing on the cleaned data to obtain normalized data; The normalized data are sorted in chronological order to obtain an ordered data set, namely, the preprocessed first operation state data of the pipeline.

[0020] In the embodiment of the present invention, the first sensor is deployed at a key position of the pipeline, and can monitor and record the operation status data of the pipeline in real time, including key parameters such as temperature, pressure, flow, vibration, and the presence of sparks, which are crucial for evaluating the health and safety of the pipeline. Real-time collection ensures the timeliness and accuracy of the data, and provides a reliable basis for subsequent analysis and early warning. In the data collection process, outliers may be generated due to sensor failure, environmental interference, or data transmission errors. If the outliers are not processed, the accuracy of subsequent analysis may be seriously affected. These outliers are identified and removed. After removing the outliers, the quality of the data set is improved, noise and interference are reduced, and more accurate input is provided for subsequent data processing and analysis. Due to the different dimensions and value ranges of different parameters, direct comparison or analysis can be It may lead to inaccurate results. Normalization is to convert these data to the same scale so that different parameters can be compared fairly. Z-score normalization is used to convert the data into a normal distribution with a mean of 0 and a standard deviation of 1. Normalization makes the comparison between different parameters fairer and more accurate, providing better input for subsequent feature extraction and model training. Time series data usually have temporal dependencies and correlations. Arranging data in chronological order helps capture these dependencies and correlations, facilitating subsequent time series analysis. The normalized data is sorted by timestamp and stored in an appropriate data structure. The sorted data set is more orderly and easy to analyze, providing better input for subsequent time series analysis, feature extraction and model training. The preprocessed pipeline first running status data.

[0021] like Figure 1 As shown in 12, according to the pre-processed pipeline first operation state data, the characteristics of the first pipeline state are extracted, and the characteristics of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics, including: According to the pre-processed first operation state data of the pipeline, a time series analysis is performed on the temperature data to calculate the rate of change of the temperature over time to obtain the temperature change trend; According to the pre-processed pipeline first operation state data, statistical analysis is performed on the pressure data to calculate the standard deviation or range of the pressure value to obtain the pressure fluctuation amplitude; According to the pre-processed pipeline first operation state data, time series analysis is performed on the flow data, and the rate of change of the flow over time is calculated to obtain the flow change rate; According to the pre-processed pipeline first operation state data, spectrum analysis is performed on the vibration data to identify the main frequency components in the vibration signal to obtain the vibration frequency characteristics; According to the pre-processed pipeline first operation state data, the pre-processed spark presence signal is analyzed to extract the spark signal characteristics, and the spark signal characteristics include the spark occurrence frequency, intensity, and duration; The extracted temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics are integrated to obtain the characteristics of the first pipeline state.

[0022] In an embodiment of the present invention, the temperature data in the first operation state data of the pipeline after preprocessing is subjected to time series analysis, the rate of change of temperature over time is calculated, and continuous temperature data points are subjected to differential operation, or the trend of temperature change over time is fitted by moving average, linear regression and other methods to obtain the temperature change trend, which reflects the dynamic change of temperature in the pipeline over time; the pressure data in the first operation state data of the pipeline after preprocessing is subjected to statistical analysis, the standard deviation or range of the pressure value is calculated, the standard deviation reflects the discrete degree of the pressure data distribution, and the range reflects the difference between the maximum value and the minimum value in the pressure data, and the pressure fluctuation amplitude is obtained, which reflects the stability and fluctuation of the pressure in the pipeline; the flow data in the first operation state data of the pipeline after preprocessing is subjected to time series analysis, the rate of change of flow over time is calculated, which is similar to the calculation of the temperature change trend, but is aimed at the flow data, and the flow change rate is obtained, which reflects the dynamic change of the flow in the pipeline; the vibration data in the first operation state data of the pipeline after preprocessing is subjected to spectrum analysis on the vibration data, and the main frequency components in the vibration signal are identified. This usually involves signal processing technologies such as Fourier transform and wavelet transform to obtain vibration frequency characteristics, which reflect the spectral characteristics of pipeline vibration, including the main vibration frequency and its corresponding energy distribution; the spark presence signal in the pre-processed pipeline first operating state data, data analysis of the pre-processed spark presence signal, extraction of spark signal characteristics, including statistics of spark occurrence frequency, intensity, duration and other parameters, to obtain spark signal characteristics, reflecting the characteristics of spark activity that may exist in the pipeline; the temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics extracted above are integrated to form a first pipeline state feature set containing multiple feature dimensions, and the obtained first pipeline state feature set provides a comprehensive information basis for subsequent pipeline state monitoring, early warning and fault diagnosis.

[0023] like Figure 1 As shown, 13, according to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain an adjusted pipeline, including: According to the characteristics of the first pipeline state, the first pipeline state is compared with a preset safety threshold to obtain a comparison result; According to the comparison result, the risk level of the pipeline status is evaluated to obtain an evaluation result; According to the assessment results, if the risk level does not exceed the preset threshold, the emergency response mechanism will not be triggered; Based on the assessment results, if the risk level exceeds the preset threshold, the emergency response mechanism is triggered; The emergency response mechanism adjusts the pipeline status according to the specific characteristics and risk level of the pipeline status. The pipeline status adjustment includes adjusting pipeline process parameters, closing or opening valves, and starting or stopping fans.

[0024] In an embodiment of the present invention, the system first obtains the characteristics of the first pipeline state, which may include key parameters such as pressure, temperature, flow, vibration, etc. in the pipeline. The characteristic data is the basis for subsequent analysis and judgment; the system compares the extracted pipeline state characteristics with a preset safety threshold, which is determined based on factors such as pipeline design, operating experience, and safety specifications, and is used to determine whether the pipeline state is within a safe range; based on the comparison result, the system evaluates the risk level of the pipeline state, and the evaluation results may include different levels such as low risk, medium risk, and high risk. The specific level division is based on the difference between the actual state of the pipeline and the comparison result; if the evaluation result shows that the risk level does not exceed the preset threshold, the system does not trigger the emergency response mechanism and continues to monitor the pipeline state. If the evaluation result shows that the risk level does not exceed the preset threshold, the system does not trigger the emergency response mechanism and continues to monitor the pipeline state. If the risk level exceeds the preset threshold, the system immediately triggers the emergency response mechanism; after the emergency response mechanism is triggered, the system takes a series of measures to adjust the pipeline status according to the specific characteristics and risk level of the pipeline status. The adjustment measures include adjusting the process parameters of the pipeline, closing or opening the valve to change the fluid flow direction or flow, starting or stopping the fan to adjust the gas flow in the pipeline, etc. Through these adjustment measures, the system aims to restore the pipeline status to a safe range, or at least reduce its risk level; after adjusting the pipeline status, the system continues to monitor the pipeline to ensure that its status is stable and safe. If the adjustment measures fail to achieve the expected results, the system may trigger the emergency response mechanism again and take further adjustment measures; the system will feedback the monitoring data and adjustment results to relevant personnel for further analysis and decision-making.

[0025] like Figure 1 As shown in 14, after the pipeline is adjusted, the second operating state data of the pipeline is monitored in real time using a second sensor and preprocessed to obtain the preprocessed second operating state data of the pipeline, including: Using a second sensor to collect second operating status data of the pipeline in real time; removing abnormal values ​​from the second operating state data to obtain cleaned second data; Performing normalization processing on the cleaned second data to obtain normalized second data; The normalized second data are sorted in chronological order to obtain an ordered second data set, that is, the preprocessed second operation status data of the pipeline.

[0026] In the embodiment of the present invention, after the pipeline has been adjusted as necessary, in order to continuously monitor its operating status, the system deploys a second sensor, which collects the second operating status data of the pipeline in real time. These data include pressure, temperature, flow, vibration and other parameters. Real-time collection ensures the timeliness and accuracy of the data, and provides a reliable basis for subsequent preprocessing and analysis; the collected second operating status data may contain abnormal values, which are caused by sensor failure, data transmission errors or emergencies in pipeline operation; the system cleans the collected data and removes these abnormal values ​​to ensure the accuracy and reliability of the data; the cleaned second data provides a more accurate data basis for subsequent normalization processing ; The second data after cleaning may have different dimensions and value ranges, which will affect the subsequent data analysis and processing. The system normalizes the cleaned data and converts the data of different dimensions and value ranges into a unified scale. The normalized second data is more convenient for subsequent data analysis and processing, and improves the accuracy and efficiency of the analysis; The normalized second data may be collected at different time intervals or sequences. For subsequent data analysis and processing, these data need to be sorted in chronological order. The system organizes the normalized data in an orderly manner to obtain an orderly second data set. The ordered second data set provides a clearer and more intuitive data basis for subsequent data analysis, trend prediction, and fault diagnosis.

[0027] like Figure 1 As shown in 15, model training is performed according to pipeline status warning requirements and historical pipeline status data, and the trained model is used to identify the preprocessed pipeline second operation status data to obtain a status recognition result, including: According to the pipeline status warning requirements, use The prediction model is constructed, where Is an explicit decision function used to judge the input data Category, is the Lagrange multiplier, is a heterokernel function used to calculate the input data Training materials The similarity between is the bias term, N is the total data, x is the input data, i.e. the second running state data of the pipeline after preprocessing, are the data points in the training dataset, where is the width parameter of the kernel function, which controls the radial range of the kernel function. Represents input data With training data The Euclidean distance between Based on historical pipeline status data, use , The prediction model is trained to obtain a trained prediction model, wherein: is the training data point Tags, is the regularization parameter, For training data points Tags, is the Lagrange multiplier, N is the total data, is the kernel function, calculating the data points and The similarity between The trained prediction model is used to identify the preprocessed pipeline second operation state data to obtain a state identification result.

[0028] In an embodiment of the present invention, a prediction model is constructed according to the pipeline status warning demand. The model is based on the support vector machine algorithm and involves an explicit decision function, a Lagrange multiplier and a bias term. The explicit decision function is used to determine the category of the input data, that is, the type of pipeline status. The Lagrange multiplier is a key parameter in the SVM algorithm and is used to solve the optimization problem. The kernel function is used to calculate the similarity between the input data and the training data. The width parameter of the kernel function controls the radial range of the kernel function, thereby affecting the classification performance of the model. The bias term is another important parameter in the model, which determines the offset of the decision boundary. The prediction model is trained using historical pipeline status data. The historical data contains the operation data of the pipeline in different states, as well as the corresponding labels. That is, the type of pipeline state; during the training process, the model adjusts the Lagrange multiplier, kernel function parameters, etc. to minimize the classification error and find the optimal decision boundary. The regularization parameter is used to control the complexity of the model to prevent overfitting; the trained model can accurately identify the category of the input data, that is, the preprocessed pipeline second operating state data; the trained prediction model is used to identify the preprocessed pipeline second operating state data. The preprocessed data has removed outliers, normalized, and sorted in chronological order; the model classifies the input data according to the characteristics of the input data, using the decision boundary and classification rules learned during the training process, and finally obtains the state recognition result, that is, the current state type of the pipeline.

[0029] like Figure 1 As shown in 16, according to the state recognition result and the preset threshold, it is determined whether the pipeline is in an abnormal state to obtain early warning information, including: According to the state recognition result, a comparison is performed with a preset threshold to obtain a comparison result, wherein the comparison result includes an abnormal state and a normal state; If the state recognition result exceeds the preset threshold, the pipeline is judged to be in an abnormal state; If the state recognition result does not exceed the preset threshold, the pipeline is judged to be in a normal state; If the judgment result is an abnormal state, the early warning mechanism is triggered to obtain early warning information according to the preset rules.

[0030] In the embodiment of the present invention, the system obtains the current state recognition result of the pipeline according to the previous state recognition process, and the system compares the state recognition result with the preset threshold value. The preset threshold value is determined according to the normal operating range of the pipeline, historical data, safety specifications and other factors, and is used to determine whether the pipeline state is abnormal; the comparison result is divided into two types: abnormal state and normal state. If the state recognition result exceeds the preset threshold value, it is judged as an abnormal state; if it does not exceed, it is judged as a normal state; according to the comparison result, the system judges the state of the pipeline. If the judgment result is an abnormal state, it means that the operating parameters or state of the pipeline have exceeded the normal range. If the result is normal, it means that the operating parameters or status of the pipeline are within the normal range, and the system can continue to monitor. If the result is abnormal, the system will immediately trigger the early warning mechanism, which may include sound and light alarms, SMS or email notifications, and display of alarm information on the monitoring interface. The system will generate early warning information according to preset rules, including pipeline status identification results, abnormality type, abnormality degree, recommended measures, etc. The generation and sending of early warning information helps relevant personnel to understand the operating status of the pipeline in a timely manner and take corresponding measures to eliminate safety hazards or failures.

[0031] like Figure 2 As shown, an embodiment of the present invention further provides a data-driven pipeline status early warning system 20, comprising: The acquisition module 21 is used to use the first sensor to acquire the first operation status data of the pipeline in real time and perform preprocessing to obtain the preprocessed first operation status data of the pipeline, wherein the first operation status data of the pipeline includes temperature, pressure, flow, vibration and spark presence; after the pipeline is adjusted, the second sensor is used to monitor the second operation status data of the pipeline in real time and perform preprocessing to obtain the preprocessed second operation status data of the pipeline; The processing module 22 is used to extract the characteristics of the first pipeline state based on the pre-processed first pipeline operation state data, and the characteristics of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics; according to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain an adjusted pipeline; according to the pipeline state warning requirements and historical pipeline state data, the model is trained, and the pre-processed second pipeline operation state data is identified using the trained model to obtain a state identification result; according to the state identification result and the preset threshold, it is determined whether the pipeline is in an abnormal state to obtain early warning information; according to the early warning information, the control system starts the emergency response mechanism for the pipeline state.

[0032] In a specific embodiment of the present invention, Figure 3 to Figure 4 As shown, the primary spark detector includes a first primary spark detector 1 and a second primary spark detector 2. The working principle is that once sparks or high-temperature hot particles are found in the dust medium passing through the primary spark detector, such particles will be detected by the primary spark detector, and then the signal will be transmitted to the control system through the infrared detector. Then the control system controls the high-pressure water solenoid valve on the extinguishing device through the output. When the solenoid valve is opened, the high-pressure fine water mist is sprayed onto the cross section of the pipe through the nozzle, so that the moving sparks or hot particles are extinguished; due to the high-speed movement of the particles, the high-pressure fine water mist of the extinguishing device cannot guarantee the complete extinguishing of such high-temperature particles, and some large-particle sparks or hot particles are not completely extinguished until the dust explosion burns. point or flash point, so there is still a possibility of igniting the dust medium and causing fire or explosion during the downstream movement, so secondary spark monitoring is carried out. The secondary spark monitoring includes the first and second spark detectors 3 and the second spark detector 4, which mainly performs secondary monitoring. At this time, the high-temperature sparks or hot particles that have not been completely extinguished are detected by the secondary spark monitoring, and the signal will be transmitted to the control system. The control system outputs a signal to the fast fire damper, and the fast fire damper quickly performs a closing action through the pneumatic principle. At this time, the sparks or hot particles that have not been completely extinguished are completely blocked on the upstream side of the fire damper. At the same time, another output signal controls the fan to stop to cut off the power source of dust transmission, effectively preventing the possibility of this risk spreading downstream.

[0033] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0034] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0035] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0036] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0037] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0038] In the embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or units, which can be electrical, mechanical or other forms.

[0039] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0040] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0041] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0042] In addition, it should be noted that in the system and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. Moreover, the steps of performing the above series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order, and some steps can be performed in parallel or independently of each other. For those of ordinary skill in the art, it is understandable that all or any steps or components of the method and system of the present invention can be implemented in any computing system (including processors, storage media, etc.) or a network of computing systems in hardware, firmware, software or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0043] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing system. The computing system can be a well-known general system. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code that implements the method or system. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be pointed out that in the system and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.

[0044] The above is 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 principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A data-driven pipeline status early warning method, characterized in that: The method comprises: Using a first sensor to acquire first operating state data of the pipeline in real time and preprocessing the data to obtain preprocessed first operating state data of the pipeline, wherein the first operating state data of the pipeline includes temperature, pressure, flow, vibration, and spark presence; Extracting the features of the first pipeline state according to the preprocessed pipeline first operation state data, wherein the features of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal features; According to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain an adjusted pipeline; After the pipeline is adjusted, the second operating state data of the pipeline is monitored in real time using the second sensor and preprocessed to obtain the preprocessed second operating state data of the pipeline; Model training is performed according to pipeline status warning requirements and historical pipeline status data, and the trained model is used to identify the preprocessed pipeline second operation status data to obtain a status recognition result; According to the state recognition results and the preset threshold, determine whether the pipeline is in an abnormal state to obtain early warning information; Based on the early warning information, the control system initiates an emergency response mechanism for the pipeline status.

2. The data-driven pipeline status early warning method according to claim 1 is characterized in that: The first operating state data of the pipeline is acquired in real time using the first sensor and preprocessed to obtain the preprocessed first operating state data of the pipeline, wherein the first operating state data of the pipeline includes temperature, pressure, flow, vibration, and spark presence, including: Using a first sensor to collect first operating status data of the pipeline in real time; removing outliers from the first operating state data to obtain cleaned data; Performing normalization processing on the cleaned data to obtain normalized data; The normalized data are sorted in chronological order to obtain an ordered data set, namely, the preprocessed first operation state data of the pipeline.

3. The data-driven pipeline status early warning method according to claim 2 is characterized in that: According to the pre-processed pipeline first operation state data, the characteristics of the first pipeline state are extracted, and the characteristics of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics, including: According to the pre-processed first operation state data of the pipeline, a time series analysis is performed on the temperature data to calculate the rate of change of the temperature over time to obtain the temperature change trend; According to the pre-processed pipeline first operation state data, statistical analysis is performed on the pressure data to calculate the standard deviation or range of the pressure value to obtain the pressure fluctuation amplitude; According to the pre-processed pipeline first operation state data, time series analysis is performed on the flow data, and the rate of change of the flow over time is calculated to obtain the flow change rate; According to the pre-processed pipeline first operation state data, spectrum analysis is performed on the vibration data to identify the main frequency components in the vibration signal to obtain the vibration frequency characteristics; According to the pre-processed pipeline first operation state data, the pre-processed spark presence signal is analyzed to extract the spark signal characteristics, and the spark signal characteristics include the spark occurrence frequency, intensity, and duration; The extracted temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics are integrated to obtain the characteristics of the first pipeline state.

4. The data-driven pipeline status early warning method according to claim 3 is characterized in that: According to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain an adjusted pipeline, including: According to the characteristics of the first pipeline state, the first pipeline state is compared with a preset safety threshold to obtain a comparison result; According to the comparison result, the risk level of the pipeline status is evaluated to obtain an evaluation result; According to the assessment results, if the risk level does not exceed the preset threshold, the emergency response mechanism will not be triggered; Based on the assessment results, if the risk level exceeds the preset threshold, the emergency response mechanism is triggered; The emergency response mechanism adjusts the pipeline status according to the specific characteristics and risk level of the pipeline status. The pipeline status adjustment includes adjusting pipeline process parameters, closing or opening valves, and starting or stopping fans.

5. The data-driven pipeline status early warning method according to claim 4 is characterized in that: After the pipeline is adjusted, the second operating state data of the pipeline is monitored in real time using the second sensor and preprocessed to obtain the preprocessed second operating state data of the pipeline, including: Using a second sensor to collect second operating status data of the pipeline in real time; removing abnormal values ​​from the second operating state data to obtain cleaned second data; Performing normalization processing on the cleaned second data to obtain normalized second data; The normalized second data are sorted in chronological order to obtain an ordered second data set, that is, the preprocessed second operation status data of the pipeline.

6. The data-driven pipeline status early warning method according to claim 5 is characterized in that: Model training is performed based on pipeline status warning requirements and historical pipeline status data, and the trained model is used to identify the preprocessed pipeline second operation status data to obtain status recognition results, including: According to the pipeline status warning requirements, use The prediction model is constructed, where Is an explicit decision function used to judge the input data Category, is the Lagrange multiplier, is a heterokernel function used to calculate the input data Training materials The similarity between is the bias term, N is the total data, x is the input data, i.e. the second running state data of the pipeline after preprocessing, x i are the data points in the training dataset, where is the width parameter of the kernel function, which controls the radial range of the kernel function. Represents input data With training data The Euclidean distance between Based on historical pipeline status data, use , The prediction model is trained to obtain a trained prediction model, wherein: is the training data point Tags, is the regularization parameter, For training data points Tags, is the Lagrange multiplier, N is the total data, is the kernel function, calculating the data points and The similarity between The trained prediction model is used to identify the preprocessed pipeline second operation state data to obtain a state identification result.

7. The data-driven pipeline status early warning method according to claim 6 is characterized in that: Based on the status recognition results and the preset threshold, determine whether the pipeline is in an abnormal state to obtain early warning information, including: According to the state recognition result, a comparison is performed with a preset threshold to obtain a comparison result, wherein the comparison result includes an abnormal state and a normal state; If the state recognition result exceeds the preset threshold, the pipeline is judged to be in an abnormal state; If the state recognition result does not exceed the preset threshold, the pipeline is judged to be in a normal state; If the judgment result is an abnormal state, the early warning mechanism is triggered to obtain early warning information according to the preset rules.

8. A data-driven pipeline status early warning system, characterized in that: include: an acquisition module, configured to acquire first operation status data of the pipeline in real time using a first sensor and perform preprocessing to obtain preprocessed first operation status data of the pipeline, wherein the first operation status data of the pipeline includes temperature, pressure, flow, vibration, and spark presence; After the pipeline is adjusted, the second operating state data of the pipeline is monitored in real time using the second sensor and preprocessed to obtain the preprocessed second operating state data of the pipeline; A processing module is used to extract the characteristics of the first pipeline state according to the pre-processed first pipeline operation state data, wherein the characteristics of the first pipeline state include temperature change trend, pressure fluctuation amplitude, flow rate change rate, vibration frequency and spark signal characteristics; according to the characteristics of the first pipeline state, the control system triggers the emergency response mechanism to adjust the pipeline state to obtain an adjusted pipeline; according to the pipeline state warning requirements and historical pipeline state data, a model is trained, and the pre-processed second pipeline operation state data is identified using the trained model to obtain a state identification result; according to the state identification result and a preset threshold, whether the pipeline is in an abnormal state is judged to obtain early warning information; Based on the early warning information, the control system initiates an emergency response mechanism for the pipeline status.

9. A computing device, characterized in that include: one or more processors; A storage system for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images