A digital-based steam pipe network intelligent management system and method

Abstract

The application discloses a kind of based on digitization's steam pipe network wisdom management system and method, it is related to steam pipe network monitoring management technical field, including steam pipe network data monitoring module, data re-monitoring module, fouling data acquisition and analysis module and abnormal data transmission module, by steam pipe network data monitoring module to steam pipe network carries out data acquisition and monitoring, analysis pipe network data abnormal condition and obtain pipe network abnormal occurrence time information, by data re-monitoring module analysis to collect pipe network data with historical sampling frequency and judge the accuracy rate when judging abnormal condition, set the best frequency range, select whether to re-monitoring processing is carried out to pipe network data, by fouling data acquisition and analysis module predict the time that current pipe network appears fouling abnormality, by abnormal data transmission module compare pipe network abnormal occurrence time and predicted time, analyze the reason of pipe network abnormality and carry out abnormal data transmission, reduce the influence caused by steam pipe network reconstruction data not timely upload.

Description

Technical Field

[0001] This invention relates to the field of steam pipeline network monitoring and management technology, specifically a digital-based intelligent management system and method for steam pipeline networks. Background Technology

[0002] Steam pipeline network refers to the pipeline system that rationally distributes steam from steam production units such as boilers to various steam consumption points. Steam pipeline network intelligent management platform refers to a platform that uses modern information technologies, such as the Internet of Things, big data, and cloud computing, to intelligently manage and optimize the steam pipeline network. Using the steam pipeline network intelligent management platform to monitor the data of the steam pipeline network is conducive to timely detection and resolution of abnormal conditions in the pipeline network.

[0003] If a steam-using unit renovates its steam pipeline network, it needs to promptly upload the renovation information to the intelligent management platform. This helps the platform obtain updated pipeline structure data and monitor the renovated network. However, after the renovation, the intelligent management platform may fail to receive or receive the renovation information in a timely manner. While the comprehensive resistance coefficient parameter of the steam pipeline network collected by the intelligent management platform can be used to predict whether the pipeline has undergone renovation, similar fluctuations in the comprehensive resistance coefficient could also be due to scaling on the pipe walls, potentially leading to misjudgments. Current technology lacks a predictive method that uses the comprehensive resistance coefficient parameter to predict pipeline renovations and does not consider the possibility of misjudgments. Therefore, it cannot help the intelligent management platform detect potential renovations in a timely manner even when renovation information is not received promptly, nor can it improve the accuracy of the pipeline renovation prediction results. Summary of the Invention

[0004] The purpose of this invention is to provide a digital-based intelligent management system and method for steam pipeline networks to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a digital-based intelligent management system for steam pipeline networks, comprising: a steam pipeline network data monitoring module, a data re-monitoring module, a scaling data acquisition and analysis module, and an abnormal data transmission module;

[0006] The steam pipeline network data monitoring module collects and monitors data from the steam pipeline network, analyzes abnormal conditions in the pipeline network data, and obtains information on the time of occurrence of pipeline network abnormalities.

[0007] The current pipeline data sampling frequency is obtained through the data re-monitoring module. The accuracy of judgment when collecting pipeline data at the historical sampling frequency and judging abnormal conditions is analyzed. The optimal frequency range is set, and the pipeline data re-monitoring process is selected based on the comparison result between the current sampling frequency and the range.

[0008] The scaling data acquisition and analysis module collects and analyzes historical data on scaling phenomena in the pipeline network to predict the timing of current scaling anomalies in the pipeline network.

[0009] The abnormal data transmission module compares the time of the pipeline anomaly with the predicted time to analyze the cause of the anomaly and transmit the abnormal data.

[0010] Preferably, the steam pipeline network data monitoring module includes a pipeline network parameter acquisition unit, a parameter change analysis unit, and an abnormal time acquisition unit;

[0011] The pipeline parameter acquisition unit is used to obtain the comprehensive resistance coefficient of the steam pipeline network. The comprehensive resistance coefficient can be calculated by jointly monitoring basic parameters such as pressure, flow rate and temperature.

[0012] The parameter change analysis unit is used to analyze the changes in the comprehensive resistance coefficient of the steam pipeline network and to determine whether there is an abnormality in the steam pipeline network based on the changes.

[0013] The abnormal time acquisition unit is used to acquire the time when the steam pipeline network experiences an abnormality.

[0014] Preferably, the data re-monitoring module includes a sampling frequency acquisition unit, a data monitoring misjudgment analysis unit, and a data re-monitoring planning unit;

[0015] The sampling frequency acquisition unit is used to acquire the sampling frequency of the pipeline parameters required for calculating the current comprehensive resistance coefficient of the steam pipeline when an anomaly is detected in the steam pipeline network.

[0016] The data monitoring misjudgment analysis unit is used to collect pipeline network parameters from previous sampling frequencies, calculate the comprehensive resistance coefficient of the steam pipeline network based on these parameters, analyze the comprehensive resistance coefficient, and determine if an anomaly has occurred in the steam pipeline network. After verification, the number of correct judgments is confirmed. Based on this number of judgments, the accuracy rate of judging anomalies using historical sampling frequencies is analyzed. The optimal sampling frequency range is determined based on the accuracy rate, and the current sampling frequency is compared to this range. If it is within the optimal range, the current anomaly judgment is correct; if not, the current anomaly judgment is incorrect. Insufficient sampling frequency may also lead to changes in the comprehensive resistance coefficient of the steam pipeline network and affect the accuracy of judgments when a true anomaly occurs. The changes in the overall resistance coefficient of the steam pipeline network are the same, that is, the overall resistance coefficient shows a sudden increase, but the actual overall resistance coefficient of the pipeline network does not show a sudden increase, that is, the actual pipeline network does not show any abnormality. By analyzing the accuracy of the judgment results of pipeline network abnormality by collecting data at different sampling frequencies in the past, the optimal sampling frequency range is set. When the pipeline network is judged to be abnormal, the sampling frequency range is calibrated first. If the sampling frequency is not within the optimal range, it is predicted that the abnormality judgment result may be incorrect. Choosing to adjust the sampling frequency and re-monitor the overall resistance coefficient of the steam pipeline network is beneficial to increasing the probability of verifying the judgment result of steam pipeline network abnormality, reducing unnecessary verification operations, and improving the effectiveness of abnormality verification.

[0017] The data re-monitoring planning unit is used to select any data sampling frequency from the optimal sampling frequency range as the adjusted data sampling frequency if the prediction of the current steam pipeline network anomaly is incorrect. The adjusted data sampling frequency is used to re-collect and monitor the pipeline network parameters, calculate the comprehensive resistance coefficient of the steam pipeline network based on the re-collected data, and analyze whether the current steam pipeline network has an anomaly.

[0018] Preferably, the scaling data acquisition and analysis module includes a data acquisition unit, a scaling data acquisition unit, and a scaling time prediction unit;

[0019] The data acquisition unit is used to collect the usage time data of the current steam pipeline network;

[0020] The scaling data acquisition unit is used to collect the number of times the scaling thickness of the steam pipeline network has exceeded the thickness threshold in the past, which is the same as the current steam pipeline network has been used for the same period of time. The thickness threshold is set by the system. When the scaling thickness of the steam pipeline network is detected to exceed the thickness threshold, the scaling area of ​​the pipeline network needs to be cleaned. The scaling phenomenon of the steam pipeline network can be monitored by the inspection robot, and the time required for the scaling thickness of the steam pipeline network to exceed the thickness threshold after each cleaning can be collected.

[0021] The scaling time prediction unit is used to obtain the time of the last scaling cleaning of the steam pipeline network before the current time, and to obtain the last cleaning as the xth cleaning, and to predict the time when the scaling thickness of the current steam pipeline network exceeds the thickness threshold after the xth cleaning.

[0022] Preferably, the abnormal data transmission module includes a time comparison unit, a pipeline abnormality tracing unit, and a suggested measures transmission and notification unit;

[0023] The time comparison unit is used to compare the current time with the predicted time when the current scale thickness of the current steam pipeline exceeds the thickness threshold. The current time is the time when the pipeline abnormality occurs.

[0024] The pipeline anomaly tracing unit is used to predict that if the current time is earlier than the predicted time, the cause of the current pipeline anomaly is that the pipeline has been modified; otherwise, it predicts that the cause of the current pipeline anomaly is that the pipeline has scale buildup. Scale buildup is defined as scale thickness exceeding a certain threshold. Scale buildup will also cause a similar change in the overall resistance coefficient of the steam pipeline, i.e., a sudden increase. Therefore, to reduce the probability of misjudging pipeline modifications due to scale buildup anomalies, a scale anomaly time prediction model is established by collecting time information on the recurrence of scale buildup anomalies after each previous cleaning of the pipeline. The system measures the time when the current pipeline network re-emerges with scaling abnormalities after the most recent cleaning and compares the time. If the current time is not earlier than the predicted time, it is determined that the pipeline network has already experienced scaling abnormalities, and the cause of the current abnormality is predicted to be scaling abnormalities rather than pipeline network modifications. Otherwise, the cause of the abnormality is predicted to be pipeline network modifications. When the steam pipeline network intelligent management platform does not receive or does not receive pipeline network modification phenomena in a timely manner, parameter analysis helps to promptly identify potential pipeline network modification phenomena, thereby reducing the probability of misjudgment of pipeline network modifications and improving the accuracy of pipeline network modification prediction results.

[0025] The suggested measures transmission and prompting unit is used to transmit the current abnormal data and cause prediction data of the pipeline network to the pipeline network monitoring terminal and provide action prompts: if the cause of the current pipeline network abnormality is predicted to be pipeline network modification, it prompts to verify whether the current pipeline network has been modified; if so, it prompts the steam user to upload the pipeline network modification information to the system; if the cause of the current pipeline network abnormality is predicted to be abnormal scaling in the pipeline network, it prompts to arrange an inspection robot to monitor the scaling in the pipeline network and to perform cleaning operations after detecting abnormal scaling.

[0026] After predicting and verifying that the steam-using unit has modified the steam pipe network, prompt it to upload the modification data to the system in a timely manner, that is, the intelligent management platform for the steam pipe network, which helps the platform monitor data based on the modified steam pipe network. If the platform fails to update the modification data in a timely manner and still monitors data based on the steam pipe network before modification, it is likely to result in invalid and incorrect monitoring data. The present invention helps the platform quickly receive the modification data of the steam pipe network, which is conducive to reducing the duration of invalid and incorrect monitoring data and alleviating the impact caused by the failure to upload the modification data of the steam pipe network in a timely manner.

[0027] A digital-based intelligent management method for a steam pipe network includes the following steps:

[0028] S1: Collect and monitor data of the steam pipe network, analyze the abnormal conditions of the pipe network data, and obtain the time information when the pipe network abnormality occurs;

[0029] S2: Obtain the current sampling frequency of the pipe network data, analyze the judgment accuracy rate when collecting pipe network data at the historical sampling frequency and judging abnormal conditions, set the optimal frequency range, and select whether to re-monitor the pipe network data based on the comparison result between the current sampling frequency and the range;

[0030] S3: Collect and analyze the historical data of the scaling phenomenon in the pipe network, and predict the time when the current pipe network has scaling abnormalities;

[0031] S4: Compare the time when the pipe network abnormality occurs with the predicted time, analyze the cause of the pipe network abnormality, and transmit the abnormal data.

[0032] Preferably, the S1 includes: collecting pipe network parameters, calculating and obtaining the comprehensive resistance coefficient of the steam pipe network based on the pipe network parameters, and obtaining that the set of comprehensive resistance coefficients of the steam pipe network calculated from the pipe network parameters collected at the current sampling frequency is K = {K1, K2,... Kn}, and a total of n data collections are made. If the coefficients in the set K show an upward trend, that is, the coefficients are getting larger, then the sampling time and the comprehensive resistance coefficient calculated at the corresponding time are used to form data points, and the data points are linearly fitted to generate a comprehensive resistance coefficient change line. The comprehensive resistance coefficient change line represents the line of the comprehensive resistance coefficient of the steam pipe network changing over time. The slope of the obtained line is b, and the slope threshold is set as B. Compare b and B: If b < B, it is judged that the steam pipe network has no abnormality; otherwise, it is judged that the steam pipe network has an abnormality, and the time when the steam pipe network is judged to have an abnormality is obtained. The comprehensive resistance coefficient of the steam pipe network refers to the total resistance encountered when steam flows in the pipeline. If there is a sudden increase in the comprehensive resistance coefficient of the steam pipe network, that is, the slope of the line is not less than the slope threshold, it may be that the pipe network has been modified, such as adding valves, etc., or it may be that the abnormal thickness of the scale in the pipeline has caused a similar abnormal change in the comprehensive resistance coefficient of the steam pipe network.

[0033] Preferably, step S2 includes: obtaining the sampling frequency of the current pipeline parameters as g, and collecting the set of sampling frequencies of previously set pipeline parameters as F = {F1, F2, ... F}. m}, where m represents the number of previously set sampling frequency items, and the result is obtained from previous sampling frequencies F at a random sampling frequency. i Pipeline network parameters are collected, and the comprehensive resistance coefficient of the steam pipeline network is calculated based on these parameters. The comprehensive resistance coefficient of the steam pipeline network is analyzed, and the number of times an anomaly occurs in the steam pipeline network is determined as D. The number of times the judgment is confirmed to be correct after verification is d. This is then used to obtain the historical sampling frequency F. i The accuracy rate of data collection and anomaly assessment when identifying pipeline network data is R. i =d / D, obtaining the set of accuracy rates for judging abnormal conditions when collecting pipeline network data at different historical sampling frequencies, as R = {R1,R2,...R}. m The m sampling frequencies are arranged in descending order of their corresponding judgment accuracy and then randomly grouped. In the first group, the judgment accuracy of all sampling frequencies is greater than the judgment accuracy of each sampling frequency in the next group. The resulting set of average judgment accuracies for each sampling frequency in a random group is p = {p1, p2, ..., p...}. r The data was divided into r groups. The group with the highest reference value was selected, where the reference value of a randomly selected group is L. The maximum value of the sampling frequency in the first group of the group with the highest reference degree is F. (1,max) The minimum value among the sampling frequencies in the first group is F. (1,min) Set the optimal sampling frequency range to [F (1,min) F (1,max) To determine if g is within the optimal sampling frequency range: if it is, the current steam pipeline network anomaly is correctly predicted; if not, the current steam pipeline network anomaly is incorrectly predicted. Select any data sampling frequency from the optimal sampling frequency range as the adjusted data sampling frequency, and re-collect and monitor the pipeline network parameters using the adjusted data sampling frequency. Calculate the comprehensive resistance coefficient of the steam pipeline network based on the re-collected data, and analyze whether the current steam pipeline network is abnormal. If the current steam pipeline network is still determined to be abnormal, update the time of the abnormality.

[0034] Preferably, step S3 includes: collecting the current usage time of the steam pipeline network as Y, and obtaining the set of times T = {T1, T2, ... T} required for the scale thickness to exceed the thickness threshold after each previous cleaning of a random steam pipeline network with a usage time of Y. w}, where w represents the number of times the corresponding steam pipeline network has been cleaned for scale. The most recent scale cleaning of the current steam pipeline network before the current time is identified as the xth cleaning, and the time of the xth cleaning is U. If x is less than or equal to w, the predicted time after the xth cleaning when the scale thickness exceeds the thickness threshold is U+T. x ', T x '=T x If x > w and w + q = x, establish a model for predicting the time of scaling anomalies: The predicted time for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the xth cleaning is U+T. x ', For smoothing coefficients, T w+q-1 Z represents the time required for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the w+q-1th cleaning. w+q-1 This represents the exponential smoothing value of the time required for the scale thickness of the current steam pipeline to exceed the thickness threshold after the w+q-1th cleaning.

[0035] Preferably, S4 includes: if the current time is earlier than U+T x If the current pipeline anomaly is predicted to be caused by pipeline renovation, the abnormal data and the predicted cause of the anomaly will be transmitted to the pipeline monitoring terminal along with the action prompts: The prompts will ask the steam user to verify whether the pipeline has been renovated. If so, the prompts will ask the user to upload the pipeline renovation information to the system. Otherwise, if the current pipeline anomaly is predicted to be caused by scaling, the prompts will ask the user to arrange for an inspection robot to monitor the pipeline for scaling and to perform cleaning operations upon detection of scaling anomalies.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] This invention establishes a scaling anomaly time prediction model by collecting time information on the recurrence of scaling anomalies after each cleaning of the pipeline network. It predicts the time when scaling anomalies will reappear after the most recent cleaning and compares the predicted time with the current time. If the current time is not earlier than the predicted time, it is determined that scaling anomalies have already occurred in the pipeline network, and the cause of the anomaly is predicted to be scaling anomalies rather than pipeline modifications. Otherwise, the cause of the anomaly is predicted to be pipeline modifications. When the steam pipeline network intelligent management platform does not receive or does not receive pipeline modification information in a timely manner, parameter analysis helps to promptly identify potential pipeline modification phenomena, thereby reducing the probability of misjudgment of pipeline modifications and improving the accuracy of pipeline modification prediction results.

[0038] After anticipating and verifying that the steam-using unit has modified the steam pipeline network, it prompts the unit to upload the modification data to the system in a timely manner, namely the steam pipeline network intelligent management platform. This helps the platform to monitor data based on the modified steam pipeline network. If the platform fails to update the modification data in a timely manner and still monitors data based on the steam pipeline network before the modification, it is easy to cause invalid and erroneous monitoring data. This invention helps the platform to quickly receive the steam pipeline network modification data, which helps to reduce the duration of invalid and erroneous monitoring data and mitigates the impact of the failure to upload the steam pipeline network modification data in a timely manner.

[0039] Furthermore, considering that insufficient sampling frequency may lead to changes in the overall resistance coefficient of the steam pipeline network that are identical to those when the network is actually abnormal (i.e., a sudden increase in the overall resistance coefficient, even if the actual network resistance coefficient does not show a sudden increase, indicating no actual network abnormality), the optimal sampling frequency range is set by analyzing the accuracy of previous anomaly assessments based on data collected at different sampling frequencies. When an anomaly is detected, the sampling frequency is calibrated first. If the sampling frequency is not within the optimal range, it may lead to incorrect anomaly assessments. Adjusting the sampling frequency and re-monitoring the overall resistance coefficient of the steam pipeline network improves the probability of correctly identifying an anomaly, reduces unnecessary verification operations, and enhances the effectiveness of anomaly verification. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the structure of a digitally based intelligent management system for steam pipeline networks according to the present invention.

[0041] Figure 2 This is a flowchart illustrating a digital-based intelligent management method for steam pipeline networks according to the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Example 1:

[0044] like Figure 1 As shown, this embodiment provides a digital-based intelligent management system for steam pipeline networks. The system includes: a steam pipeline network data monitoring module, a data re-monitoring module, a scaling data acquisition and analysis module, and an abnormal data transmission module.

[0045] The steam pipeline network data monitoring module collects and monitors data from the steam pipeline network, analyzes abnormal conditions in the network data, and obtains information on the time when these abnormalities occur.

[0046] The current pipeline data sampling frequency is obtained through the data re-monitoring module. The accuracy of judgment when collecting pipeline data at historical sampling frequencies and judging abnormal conditions is analyzed. The optimal frequency range is set, and the pipeline data re-monitoring process is selected based on the comparison results between the current sampling frequency and the range.

[0047] The scaling data acquisition and analysis module collects and analyzes historical data on scaling phenomena in the pipeline network to predict the timing of current scaling anomalies in the pipeline network.

[0048] By comparing the time of occurrence and the predicted time of pipeline anomalies with the time of occurrence of pipeline anomalies through the abnormal data transmission module, the causes of pipeline anomalies are analyzed and abnormal data transmission is performed.

[0049] The steam pipeline network data monitoring module includes a pipeline network parameter acquisition unit, a parameter change analysis unit, and an abnormal time acquisition unit;

[0050] The pipeline parameter acquisition unit is used to obtain the comprehensive resistance coefficient of the steam pipeline network. The comprehensive resistance coefficient can be calculated by jointly monitoring basic parameters such as pressure, flow rate and temperature.

[0051] The parameter change analysis unit is used to analyze the changes in the comprehensive resistance coefficient of the steam pipeline network and to determine whether there are any abnormalities in the steam pipeline network based on the changes.

[0052] The abnormal time acquisition unit is used to acquire the time when an abnormality occurs in the steam pipeline network.

[0053] The data re-monitoring module includes a sampling frequency acquisition unit, a data monitoring misjudgment analysis unit, and a data re-monitoring planning unit;

[0054] The sampling frequency acquisition unit is used to acquire the sampling frequency of the pipeline parameters required for calculating the current comprehensive resistance coefficient of the steam pipeline when an anomaly is detected in the steam pipeline network.

[0055] The data monitoring misjudgment analysis unit is used to collect pipeline network parameters from previous sampling frequencies, calculate the comprehensive resistance coefficient of the steam pipeline network based on these parameters, analyze the comprehensive resistance coefficient, and verify the number of times the judgment result is correct after determining whether an anomaly has occurred in the steam pipeline network. Based on this verification, the unit analyzes the accuracy rate of judgments made when collecting pipeline network data at historical sampling frequencies and determining anomalies. Based on the accuracy rate, it determines the optimal sampling frequency range and compares the current sampling frequency with this range: if it is within the optimal range, the current anomaly prediction is correct; otherwise, the current anomaly prediction is incorrect. The data re-monitoring planning unit is used if an incorrect prediction of the current anomaly is found. If the current anomaly is predicted, it selects a data sampling frequency from the optimal range as the adjusted sampling frequency, and re-collects and monitors the pipeline network parameters using this adjusted frequency. Based on the re-collected data, it calculates the comprehensive resistance coefficient of the steam pipeline network and analyzes whether an anomaly has occurred in the current steam pipeline network.

[0056] The scaling data acquisition and analysis module includes a usage data acquisition unit, a scaling data acquisition unit, and a scaling time prediction unit;

[0057] The data acquisition unit is used to collect the usage time data of the current steam pipeline network;

[0058] The scaling data acquisition unit is used to collect the number of times the scaling thickness of the steam pipeline network, which has been in use for the same period as the current steam pipeline network, has exceeded the thickness threshold. The thickness threshold is set by the system. When the scaling thickness of the steam pipeline network is detected to exceed the thickness threshold, the scaling area of ​​the pipeline network needs to be cleaned. The scaling phenomenon of the steam pipeline network can be monitored by the inspection robot, and the time required for the scaling thickness of the steam pipeline network to exceed the thickness threshold after each cleaning can be collected.

[0059] The scaling time prediction unit is used to obtain the time of the last scaling cleaning of the steam pipeline network before the current time. The most recent cleaning is identified as the xth cleaning, and the unit predicts the time after the xth cleaning when the scaling thickness of the current steam pipeline network exceeds the thickness threshold.

[0060] The abnormal data transmission module includes a time comparison unit, a pipeline anomaly tracing unit, and a suggested measures transmission and notification unit;

[0061] The time comparison unit is used to compare the current time with the predicted time when the current scale thickness of the steam pipeline exceeds the thickness threshold. The current time is the time when the pipeline anomaly occurs.

[0062] The pipeline anomaly tracing unit is used to determine that if the current time is earlier than the predicted time, the cause of the current pipeline anomaly is that the pipeline has been modified; otherwise, the cause of the current pipeline anomaly is that the pipeline has scale buildup.

[0063] The recommended measure transmission and prompt unit is used to transmit the abnormal data and cause prediction data of the current pipe network to the pipe network monitoring terminal and give measure prompts: If it is predicted that the reason for the abnormality of the current pipe network is that the pipe network has been renovated, it is prompted to verify whether the current pipe network has been renovated. If so, it is prompted that the steam-using unit uploads the pipe network renovation information to the system; If it is predicted that the reason for the abnormality of the current pipe network is fouling abnormality of the pipe network, it is prompted to arrange an inspection robot to monitor the fouling of the pipe network, and perform cleaning operations after detecting fouling abnormality.

[0064] Embodiment 2:

[0065] As Figure 2 shown, this embodiment provides a digital-based intelligent management method for steam pipe networks, which is implemented based on the intelligent management system in the embodiment, and specifically includes the following steps:

[0066] S1: Collect and monitor data of the steam pipe network, analyze the abnormal conditions of the pipe network data, and obtain the time information when the pipe network abnormality occurs: Collect pipe network parameters, calculate and obtain the comprehensive resistance coefficient of the steam pipe network based on the pipe network parameters, and obtain the comprehensive resistance coefficient set of the steam pipe network calculated from the pipe network parameters collected at the current sampling frequency as K = {K1, K2,... Kn}, and a total of n data collections are made. If the coefficients in the set K show an upward trend, that is, the coefficients are getting larger, then form data points with the sampling time and the comprehensive resistance coefficient calculated at the corresponding time, perform linear fitting on the data points to generate a comprehensive resistance coefficient change line. The comprehensive resistance coefficient change line represents the line of the comprehensive resistance coefficient of the steam pipe network changing over time, obtain the slope of the line as b, set the slope threshold as B, and compare b and B: If b < B, it is judged that the steam pipe network has no abnormality; otherwise, it is judged that the steam pipe network has an abnormality, and obtain the time when it is judged that the steam pipe network has an abnormality.

[0067] S2: Obtain the current pipe network data sampling frequency, analyze the judgment accuracy rate when collecting pipe network data at the historical sampling frequency and judging abnormal conditions, set the optimal frequency range, and select whether to perform re-monitoring processing on the pipe network data according to the comparison result between the current sampling frequency and the range: Obtain the sampling frequency of the current pipe network parameters as g, and obtain the sampling frequency set of the pipe network parameters set in the past as F = {F1, F2,... F m}, m represents the number of sampling frequency items set in the past, obtain that the steam pipe network comprehensive resistance coefficient is calculated based on the pipe network parameters collected at a random sampling frequency F i in the past, analyze the steam pipe network comprehensive resistance coefficient and judge that the number of times the steam pipe network has an abnormality is D, and the number of times the judgment result is confirmed to be correct after verification is d, and obtain the judgment accuracy rate R when collecting pipe network data at the historical sampling frequency F i and judging abnormal conditionsi =d / D, obtaining the set of accuracy rates for judging abnormal conditions when collecting pipeline network data at different historical sampling frequencies, as R = {R1,R2,...R}. m The m sampling frequencies are arranged in descending order of their corresponding judgment accuracy and then randomly grouped. In the first group, the judgment accuracy of all sampling frequencies is greater than the judgment accuracy of each sampling frequency in the next group. The resulting set of average judgment accuracies for each sampling frequency in a random group is p = {p1, p2, ..., p...}. r The data was divided into r groups. The group with the highest reference value was selected, where the reference value of a randomly selected group is L. The maximum value of the sampling frequency in the first group of the group with the highest reference degree is F. (1,max) The minimum value among the sampling frequencies in the first group is F. (1,min) Set the optimal sampling frequency range to [F (1,min) F (1,max) The process involves determining whether g is within the optimal sampling frequency range. If it is, the prediction of an anomaly in the current steam pipeline network is correct. If not, the prediction of an anomaly in the current steam pipeline network is incorrect. A data sampling frequency is then randomly selected from the optimal sampling frequency range as the adjusted data sampling frequency. The pipeline network parameters are then re-acquired and monitored using the adjusted data sampling frequency. Based on the re-acquired data, the comprehensive resistance coefficient of the steam pipeline network is calculated, and the current anomaly in the steam pipeline network is analyzed. If the current anomaly is still determined, the time of the anomaly in the steam pipeline network is updated.

[0068] S3: Collect and analyze historical data on scaling phenomena in the pipeline network, and predict the time when scaling anomalies will occur in the current pipeline network: Collect the usage time of the current steam pipeline network as Y, and obtain the set of times required for the scale thickness to exceed the thickness threshold after each previous cleaning of a random steam pipeline network with usage time of Y, which is T = {T1, T2, ... T}. w}, where w represents the number of times the corresponding steam pipeline network has been cleaned for scale. The most recent scale cleaning of the current steam pipeline network before the current time is identified as the xth cleaning, and the time of the xth cleaning is U. If x is less than or equal to w, the predicted time after the xth cleaning when the scale thickness exceeds the thickness threshold is U+T. x ', T x '=T x If x > w and w + q = x, establish a model for predicting the time of scaling anomalies: The predicted time for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the xth cleaning is U+T. x ', For smoothing coefficients, Tw+q-1 Z represents the time required for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the w+q-1th cleaning. w+q-1 This represents the exponential smoothing value of the time required for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the w+q-1th cleaning. The calculated exponential smoothing value Z1 is the time required for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the first cleaning. The exponential smoothing value Z2, representing the time required for the current steam pipeline network to exceed the thickness threshold after the second cleaning, is obtained. The exponential smoothing value Z3, representing the time required for the current steam pipeline network to exceed the thickness threshold after the third cleaning, is obtained. By analogy, we can obtain Z. w+q-1 ;

[0069] S4: Compare the time of the pipeline anomaly occurrence with the predicted time, analyze the cause of the anomaly, and transmit the anomaly data: If the current time is earlier than U+T... x If the current pipeline anomaly is predicted to be caused by pipeline renovation, the abnormal data and the predicted cause of the anomaly will be transmitted to the pipeline monitoring terminal along with the suggested measures: The terminal will prompt the user to verify whether the pipeline has been renovated. If so, the user will be prompted to upload the pipeline renovation information to the system. Otherwise, if the current pipeline anomaly is predicted to be caused by scaling, the terminal will transmit the abnormal data and the predicted cause of the anomaly to the pipeline monitoring terminal along with the suggested measures: The terminal will prompt the user to arrange for an inspection robot to monitor the pipeline for scaling and to perform cleaning operations upon detection of scaling anomalies.

[0070] For example: It was predicted that the scale thickness of the current steam pipeline network would exceed the thickness threshold after the xth cleaning on March 21st, and the current anomaly in the steam pipeline network was detected on March 18th. The current time is earlier than U+T. x The system predicts that the current pipeline network anomaly is due to pipeline network modifications. It then transmits the current pipeline network anomaly data and the predicted cause data to the pipeline network monitoring terminal and provides action prompts: prompting the user to verify whether the current pipeline network has been modified; if so, prompting the steam user to upload the pipeline network modification information to the system.

[0071] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A digital-based intelligent management method for steam pipeline networks, characterized in that: It includes the following steps: S1: Collect and monitor data of the steam pipeline network, analyze the abnormal conditions of the pipeline network data, and obtain the time information of the occurrence of pipeline network abnormalities; S2: Obtain the current data sampling frequency of the pipeline network, analyze the judgment accuracy rate when collecting pipeline network data at the historical sampling frequency and judging abnormal conditions, set the optimal frequency range, and select whether to re-monitor the pipeline network data according to the comparison result between the current sampling frequency and the range; S3: Collect and analyze the historical data of scale formation in the pipeline network, and predict the time of scale formation abnormality in the current pipeline network; S4: Compare the time of occurrence of pipeline network abnormalities with the predicted time, analyze the causes of pipeline network abnormalities, and transmit abnormal data; S2 includes: obtaining the sampling frequency of the current pipeline parameters as g, and collecting the set of sampling frequencies of previously set pipeline parameters as F={F1,F2,...F... m }, where m represents the number of previously set sampling frequency items, and the result is obtained from previous sampling frequencies F at a random sampling frequency. i Pipeline network parameters are collected, and the comprehensive resistance coefficient of the steam pipeline network is calculated based on these parameters. The comprehensive resistance coefficient of the steam pipeline network is analyzed, and the number of times an anomaly occurs in the steam pipeline network is determined as D. The number of times the judgment is confirmed to be correct after verification is d. This is then used to obtain the historical sampling frequency F. i The accuracy rate of data collection and anomaly assessment when identifying pipeline network data is R. i =d / D, and the set of judgment accuracy rates when collecting pipeline network data at different historical sampling frequencies and judging abnormal conditions is obtained as R={R1,R2,...R m Arrange the m sampling frequencies in descending order of their corresponding judgment accuracy and randomly group them. The resulting set of average judgment accuracy values ​​for each sampling frequency group is p = {p1, p2, ..., p...}. r The data was divided into r groups. The group with the highest reference value was selected, where the reference value of a randomly selected group is L. The maximum value of the sampling frequency in the first group of the group with the highest reference degree is F. (1,max) The minimum value among the sampling frequencies in the first group is F. (1,min) Set the optimal sampling frequency range to [F (1,min) F (1,max) ]; S3 includes: collecting the current usage time of the steam pipeline network as Y, and obtaining the set of times T={T1,T2,...T} required for the scale thickness to exceed the thickness threshold after each previous scale cleaning of another random steam pipeline network with a usage time of Y. w }, where w represents the number of times the corresponding steam pipeline network has been cleaned for scale. The most recent scale cleaning of the current steam pipeline network before the current time is identified as the xth cleaning, and the time of the xth cleaning is U. If x is less than or equal to w, the predicted time after the xth cleaning when the scale thickness exceeds the thickness threshold is U+T. x ’ T x ’ =T x If x > w and w + q = x, establish a model for predicting the time of scaling anomalies: The predicted time for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the xth cleaning is U+T. x ’ , For smoothing coefficients, , This represents the time required for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the w+q-1th cleaning cycle. This represents the exponential smoothing value of the time required for the scale thickness of the current steam pipeline network to exceed the thickness threshold after the w+q-1th cleaning; S4 includes: if the current time is earlier than U+T x ’ If the current pipeline network is abnormal, it is predicted that the network has been modified; otherwise, it is predicted that the current pipeline network is abnormal due to scaling.

2. The intelligent management method for steam pipeline networks based on digitalization according to claim 1, characterized in that: The S1 includes: Collect pipeline network parameters, calculate and obtain the comprehensive resistance coefficient of the steam pipeline network based on the pipeline network parameters, and obtain the set of comprehensive resistance coefficients of the steam pipeline network calculated from the pipeline network parameters collected at the current sampling frequency as K = {K1, K2,... Kn}, and a total of n data collections are made. If the coefficients in the set K show an upward trend, then form data points with the sampling time and the comprehensive resistance coefficient calculated at the corresponding time, perform linear fitting on the data points to generate a comprehensive resistance coefficient change line. The comprehensive resistance coefficient change line represents the line of the comprehensive resistance coefficient of the steam pipeline network changing over time. Obtain the slope of the line as b, set the slope threshold as B, and compare b and B: If b < B, judge that the steam pipeline network has no abnormality; otherwise, judge that the steam pipeline network has an abnormality and obtain the time when it is judged that the steam pipeline network has an abnormality.

3. The intelligent management method for steam pipeline networks based on digitalization according to claim 2, characterized in that: Judge whether g is within the optimal sampling frequency range: If it is, predict that the current abnormal judgment of the steam pipeline network is correct; if not, predict that the current abnormal judgment of the steam pipeline network is incorrect. Select any data sampling frequency from the optimal sampling frequency range as the adjusted data sampling frequency, re-collect and monitor the pipeline network parameters at the adjusted data sampling frequency, calculate the comprehensive resistance coefficient of the steam pipeline network based on the re-collected data, analyze whether the current steam pipeline network has an abnormality. If it is still judged that the current steam pipeline network has an abnormality, update the time when it is judged that the steam pipeline network has an abnormality.

4. The intelligent management method for steam pipeline networks based on digitalization according to claim 3, characterized in that: If it is pre-judged that the reason for the current pipeline network abnormality is that the pipeline network has been renovated, transmit the abnormal data of the current pipeline network and the pre-judged data of the reason to the pipeline network monitoring terminal and give a measure prompt: Prompt to verify whether the current pipeline network has been renovated. If so, prompt the steam-using unit to upload the pipeline network renovation information to the system; If it is pre-judged that the reason for the current pipeline network abnormality is that the pipeline network has scale formation abnormality, transmit the abnormal data of the current pipeline network and the pre-judged data of the reason to the pipeline network monitoring terminal and give a measure prompt: Prompt to arrange an inspection robot to monitor the pipeline network for scale formation, and perform cleaning operations after detecting scale formation abnormalities.

5. A digital-based intelligent management system for steam pipeline networks, applied to the digital-based intelligent management method for steam pipeline networks as described in claim 1, characterized in that: The system includes a steam pipeline network data monitoring module, a data re-monitoring module, a scale formation data collection and analysis module, and an abnormal data transmission module; Through the steam pipeline network data monitoring module, collect and monitor data of the steam pipeline network, analyze the abnormal conditions of the pipeline network data, and obtain the time information of the occurrence of pipeline network abnormalities; The current pipeline data sampling frequency is obtained through the data re-monitoring module. The accuracy of judgment when collecting pipeline data at the historical sampling frequency and judging abnormal conditions is analyzed. The optimal frequency range is set, and the pipeline data re-monitoring process is selected based on the comparison result between the current sampling frequency and the range. The scaling data acquisition and analysis module collects and analyzes historical data on scaling phenomena in the pipeline network to predict the timing of current scaling anomalies in the pipeline network. The abnormal data transmission module compares the time of the pipeline anomaly with the predicted time to analyze the cause of the anomaly and transmit the abnormal data.

6. The intelligent management system for a digitalized steam pipeline network according to claim 5, characterized in that: The steam pipeline network data monitoring module includes a pipeline network parameter acquisition unit, a parameter change analysis unit, and an abnormal time acquisition unit. The pipeline parameter acquisition unit is used to obtain the comprehensive resistance coefficient of the steam pipeline network; The parameter change analysis unit is used to analyze the changes in the comprehensive resistance coefficient of the steam pipeline network and determine whether there is an abnormality in the steam pipeline network based on the changes. The abnormal time acquisition unit is used to acquire the time when the steam pipeline network experiences an abnormality.

7. The intelligent management system for a digital steam pipeline network according to claim 6, characterized in that: The data re-monitoring module includes a sampling frequency acquisition unit, a data monitoring misjudgment analysis unit, and a data re-monitoring planning unit. The sampling frequency acquisition unit is used to acquire the sampling frequency of the pipeline parameters required for calculating the current comprehensive resistance coefficient of the steam pipeline when an anomaly is detected in the steam pipeline network. The data monitoring misjudgment analysis unit is used to collect pipeline parameters collected at different sampling frequencies in the past, calculate the comprehensive resistance coefficient of the steam pipeline based on the pipeline parameters, analyze the comprehensive resistance coefficient of the steam pipeline and determine the abnormality of the steam pipeline. After verification, it confirms the number of times the judgment result is correct, analyzes the judgment accuracy when collecting pipeline data at historical sampling frequencies and judging abnormal conditions based on the number of times, determines the optimal sampling frequency range based on the judgment accuracy, and compares whether the current sampling frequency is within the optimal sampling frequency range: if it is, the judgment of the current steam pipeline abnormality is correct. If not, the prediction of an anomaly in the current steam pipeline network is incorrect; The data re-monitoring planning unit is used to select any data sampling frequency from the optimal sampling frequency range as the adjusted data sampling frequency if the prediction of the current steam pipeline network anomaly is incorrect. The adjusted data sampling frequency is used to re-collect and monitor the pipeline network parameters, calculate the comprehensive resistance coefficient of the steam pipeline network based on the re-collected data, and analyze whether the current steam pipeline network has an anomaly.

8. The intelligent management system for a digitalized steam pipeline network according to claim 7, characterized in that: The scaling data acquisition and analysis module includes a data acquisition unit, a scaling data acquisition unit, and a scaling time prediction unit. The data acquisition unit is used to collect the usage time data of the current steam pipeline network; The scaling data acquisition unit is used to collect the number of times the scaling thickness of the steam pipeline network has exceeded the thickness threshold in the past, which is the same as the current steam pipeline network has been used for the same period of time. The thickness threshold is set by the system. When the scaling thickness of the steam pipeline network is detected to exceed the thickness threshold, the scaling area of ​​the pipeline network needs to be cleaned. The unit collects the time required for the scaling thickness of the steam pipeline network to exceed the thickness threshold after each cleaning. The scaling time prediction unit is used to obtain the time of the last scaling cleaning of the steam pipeline network before the current time, and to obtain the last cleaning as the xth cleaning, and to predict the time when the scaling thickness of the current steam pipeline network exceeds the thickness threshold after the xth cleaning.

9. A digital-based intelligent management system for steam pipelines according to claim 8, characterized in that: The abnormal data transmission module includes a time comparison unit, a pipeline abnormality tracing unit, and a suggested measures transmission and notification unit. The time comparison unit is used to compare the current time with the predicted time when the current scale thickness of the current steam pipeline exceeds the thickness threshold. The pipeline anomaly tracing unit is used to predict that if the current time is earlier than the predicted time, the cause of the current pipeline anomaly is that the pipeline has been modified; otherwise, it is predicted that the cause of the current pipeline anomaly is that the pipeline has scale buildup. The suggested measures transmission and prompting unit is used to transmit the current pipeline network's abnormal data and predicted causes to the pipeline network monitoring terminal and provide prompts for measures: if the predicted cause of the current pipeline network's abnormality is that the pipeline network has been modified, it prompts the user to verify whether the pipeline network has been modified; if so, it prompts the steam user to upload the pipeline network modification information to the system; if the predicted cause of the current pipeline network's abnormality is that the pipeline network has scale buildup, it prompts the user to arrange an inspection robot to monitor the pipeline network for scale buildup and to perform cleaning operations after detecting scale buildup abnormalities.

Images