A low-power wireless communication module and wireless communication method for temperature and pressure monitoring

Abstract

This invention discloses a low-power wireless communication module and method for temperature and pressure monitoring, relating to the field of wireless communication network technology. It addresses the technical problems of high communication redundancy, excessive overall network power consumption, and short battery life of battery-powered nodes in wireless temperature and pressure monitoring of thermal pipelines. The method includes: acquiring historical operating data of the thermal pipeline; classifying historical abnormal events based on the historical operating data to obtain an abnormal sample set, which includes a subset of blockage abnormal samples and a subset of leakage abnormal samples; acquiring real-time pressure data from boundary monitoring points, and predicting abnormal risk sections in the thermal pipeline based on the real-time pressure data and the abnormal sample set, wherein the abnormal risk section is located between two adjacent boundary monitoring points; and sending a wake-up command to at least one interval monitoring point within the abnormal risk section; wherein the wake-up command instructs at least one interval monitoring point within the abnormal risk section to upload temperature and pressure data.

Description

Technical Field

[0001] This invention relates to the field of wireless communication network technology, specifically to a low-power wireless communication module and wireless communication method for temperature and pressure monitoring. Background Technology

[0002] In heating pipeline systems, temperature and pressure are key parameters reflecting the network's operational status and ensuring heating safety and efficiency. Continuous monitoring of pipeline temperature and pressure is crucial for timely detection of leaks, blockages, and other faults, and for optimizing system performance. Because heating pipelines are typically widely distributed and operate in complex environments, continuous cabling is difficult; therefore, using low-power wireless sensor networks for data acquisition and transmission has become an important technological direction.

[0003] Existing wireless temperature and pressure monitoring solutions typically rely on a fixed-period data upload strategy. All monitoring nodes actively communicate with the gateway and upload data at preset time intervals, regardless of changes in operating conditions. This periodic reporting model ensures data continuity, but in actual pipeline network operation, temperature and pressure data remain stable for most of the time, with valuable changes only occurring when anomalies occur. The fixed-frequency communication mechanism results in a large amount of repetitive, routine data filling the network. This not only strains and wastes wireless channel resources but also directly leads to frequent and unnecessary wireless transmissions by monitoring nodes, especially those powered by batteries. Consequently, the communication energy consumption of the entire monitoring system remains high, battery life is significantly reduced, and maintenance costs increase, hindering its large-scale deployment and application in scenarios requiring long-term unattended operation. Summary of the Invention

[0004] To address the technical problems of high communication redundancy, excessive overall network power consumption, and short battery life of battery-powered nodes in wireless temperature and pressure monitoring of thermal pipelines caused by the current fixed-frequency data upload strategy, this invention aims to provide a low-power wireless communication module and method for temperature and pressure monitoring. The specific technical solution adopted is as follows:

[0005] In a first aspect, the present invention provides a low-power wireless communication method for temperature and pressure monitoring, comprising: acquiring historical operating data of a thermal pipeline; wherein the historical operating data includes historical temperature data and historical pressure data of multiple monitoring points on the thermal pipeline, the types of multiple monitoring points include boundary monitoring points and interval monitoring points, and multiple interval monitoring points are set between each pair of adjacent boundary monitoring points; classifying historical abnormal events according to the historical operating data to obtain an abnormal sample set; wherein the abnormal sample set includes a subset of blockage abnormal samples and a subset of leakage abnormal samples; acquiring real-time pressure data of the boundary monitoring points, and predicting abnormal risk sections in the thermal pipeline based on the real-time pressure data and the abnormal sample set; wherein the abnormal risk section is located between two adjacent boundary monitoring points; sending a wake-up command to at least one interval monitoring point in the abnormal risk section; wherein the wake-up command is used to instruct at least one interval monitoring point in the abnormal risk section to upload temperature data and pressure data.

[0006] In one possible implementation, historical anomalies are classified based on historical operational data to obtain an anomaly sample set. Specifically, this includes: determining the severity of the anomaly at each monitoring point at each historical moment based on historical temperature data; where the severity of the anomaly characterizes the extent to which the temperature deviates from the normal level; determining the pressure fluctuation index at each monitoring point based on historical pressure data; where the pressure fluctuation index characterizes the degree of abnormal disturbance in upstream pressure fluctuations at each historical moment; determining an anomaly differentiation index for each historical anomaly based on the anomaly severity and the pressure fluctuation index; where the anomaly differentiation index characterizes the probability that the historical anomaly belongs to a blockage anomaly; and performing cluster analysis on all historical anomalies based on the anomaly differentiation index to obtain a blockage anomaly sample subset and a leakage anomaly sample subset.

[0007] In one possible implementation, the pressure variation index for each monitoring point is determined based on historical pressure data. Specifically, this includes: for each monitoring point, obtaining historical pressure data from a preset number of adjacent monitoring points upstream of the monitoring point; and determining the pressure variation index based on the historical pressure data from the preset number of adjacent monitoring points, the historical pressure data of the monitoring point, and the distance between each adjacent monitoring point and the monitoring point.

[0008] In one possible implementation, an anomaly differentiation index is determined for each historical anomaly event based on the severity of the anomaly and the pressure variation index. Specifically, this includes: analyzing the correlation characteristics between the pressure variation index and the severity of the anomaly based on the historical operational data corresponding to all historical anomalies; for each historical anomaly event, compensating and correcting the pressure variation index of the historical anomaly event based on the correlation characteristics to obtain a compensation value; and determining the anomaly differentiation index based on the compensation value and the pressure variation index of the historical anomaly event.

[0009] In one possible implementation, based on real-time pressure data and anomaly sample sets, abnormal risk sections in the thermal pipeline are predicted. Specifically, this includes: for each boundary monitoring point, determining the pressure variation index of the boundary monitoring point based on its real-time pressure data; if the pressure variation index of the boundary monitoring point is greater than a first preset threshold, matching the pressure variation index of the boundary monitoring point with a subset of blockage anomaly samples to determine the highest blockage matching degree of the boundary monitoring point; wherein, the highest blockage matching degree is the highest value among the blockage matching degrees between the boundary monitoring point and each historical blockage sample in the subset of blockage anomaly samples; if the highest blockage matching degree of the boundary monitoring point is greater than a second preset threshold, determining the anomaly type as blockage, and identifying the abnormal risk section based on the location information of the historical blockage sample records corresponding to the highest blockage matching degree.

[0010] In one possible implementation, the method further includes: when the highest blockage matching degree of the boundary monitoring point is less than or equal to a second preset threshold, matching the pressure change index of the boundary monitoring point with each historical leak sample in the abnormal leak sample subset to determine multiple leak similarities; wherein, a leak similarity is used to characterize the degree of similarity between the boundary monitoring point and a historical leak sample; determining a leak reference distance based on the multiple leak similarities and the leak point locations in the historical leak samples corresponding to the multiple leak similarities; determining the location point at which the leak reference distance is from the starting point along the upstream direction of the heat pipeline, with the boundary monitoring point as the starting point, as the predicted leak point, and predicting the pipeline section where the predicted leak point is located, defined by two adjacent boundary monitoring points, as an abnormal risk section.

[0011] In one possible implementation, a wake-up command is sent to at least one interval monitoring point in the abnormal risk section, specifically including: determining at least one target interval monitoring point located in the abnormal risk section as the core wake-up point; determining at least one monitoring point adjacent to the core wake-up point on the thermal pipeline topology as the cooperative wake-up point; and generating and sending a wake-up command containing the core wake-up point address and the cooperative wake-up point address.

[0012] In one possible implementation, before acquiring the real-time pressure data of the boundary monitoring point, the method further includes: configuring the boundary monitoring point to collect and upload temperature and pressure data at a first preset frequency.

[0013] In one possible implementation, before acquiring the real-time pressure data of the boundary monitoring point, the method further includes: configuring the interval monitoring point to collect and upload temperature and pressure data at a second preset frequency lower than the first preset frequency, and controlling the main communication module of the interval monitoring point to enter a sleep state during non-upload periods, while keeping the wake-up receiving module of the interval monitoring point in a working state.

[0014] In one possible implementation, in a second aspect, the present invention provides a low-power wireless communication module for temperature and pressure monitoring, comprising: a data acquisition module, an anomaly sample classification module, an anomaly risk prediction module, and a wake-up command sending module; the data acquisition module is used to acquire historical operating data of a thermal pipeline; wherein the historical operating data includes historical temperature data and historical pressure data of multiple monitoring points on the thermal pipeline, the types of multiple monitoring points include boundary monitoring points and interval monitoring points, and multiple interval monitoring points are set between each pair of adjacent boundary monitoring points; the anomaly sample classification module is used to classify historical anomaly events according to the historical operating data to obtain an anomaly sample set; wherein the anomaly sample set includes a subset of blockage anomaly samples and a subset of leakage anomaly samples; the data acquisition module is also used to acquire real-time pressure data of the boundary monitoring points; the anomaly risk prediction module is used to predict anomaly risk sections in the thermal pipeline according to the real-time pressure data and the anomaly sample set; wherein the anomaly risk section is located between two adjacent boundary monitoring points; the wake-up command sending module is used to send a wake-up command to at least one interval monitoring point in the anomaly risk section; wherein the wake-up command is used to instruct at least one interval monitoring point in the anomaly risk section to upload temperature data and pressure data.

[0015] Thirdly, the present invention provides an electronic device, comprising: a processor and a memory; wherein the memory is used to store one or more programs, the one or more programs including computer-executable instructions, and when the electronic device is running, the processor executes the computer-executable instructions stored in the memory to cause the electronic device to perform a low-power wireless communication method for temperature and pressure monitoring as described in the first aspect and any possible implementation thereof.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by an electronic device of the present invention, cause the electronic device to perform a low-power wireless communication method for temperature and pressure monitoring as described in the first aspect and any possible implementation thereof.

[0017] Fifthly, the present invention provides a computer program product containing instructions that, when executed on a computer, cause the electronic device of the present invention to perform a low-power wireless communication method for temperature and pressure monitoring as described in the first aspect and any possible implementation thereof.

[0018] In a sixth aspect, the present invention provides a chip system applied to a low-power wireless communication device for temperature and pressure monitoring; the chip system includes one or more interface circuits and one or more processors. The interface circuits and the processors are interconnected via lines; the interface circuits are used to receive signals from the memory of the low-power wireless communication device for temperature and pressure monitoring and to send the signals to the processors, the signals including computer instructions stored in the memory. When the processor executes the computer instructions, the low-power wireless communication device for temperature and pressure monitoring performs the low-power wireless communication method for temperature and pressure monitoring as described in the first aspect and any possible design embodiment thereof.

[0019] The present invention has the following beneficial effects: by constructing a historical anomaly sample knowledge base and performing intelligent matching and prediction based on real-time pressure characteristics, it realizes a paradigm shift from fixed-frequency uploading of data from all nodes to precise wake-up based on anomaly risk. This significantly reduces the number of wireless communication operations and network load while ensuring the timeliness of anomaly monitoring in the thermal pipeline network, greatly reduces the energy consumption of monitoring nodes, effectively extends the service life of battery-powered nodes and the independent operation cycle of the system, and maintains the ability to quickly locate and respond to key faults such as pipeline blockage and leakage. Attached Figure Description

[0020] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the architecture of a low-power wireless communication module for temperature and pressure monitoring provided in one embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram of the architecture of a data acquisition module provided in one embodiment of the present invention;

[0023] Figure 3 This is a schematic diagram of the architecture of an abnormal risk prediction module provided in one embodiment of the present invention;

[0024] Figure 4 This is one of the flowcharts illustrating a low-power wireless communication method for temperature and pressure monitoring provided in an embodiment of the present invention;

[0025] Figure 5 This is a second schematic flowchart of a low-power wireless communication method for temperature and pressure monitoring provided in one embodiment of the present invention.

[0026] Figure 6 This is the third flowchart illustrating a low-power wireless communication method for temperature and pressure monitoring provided in one embodiment of the present invention. Detailed Implementation

[0027] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0029] The following description, in conjunction with the accompanying drawings, details the specific scheme of a low-power wireless communication module and wireless communication method for temperature and pressure monitoring provided by the present invention.

[0030] For example, such as Figure 1 The diagram shown is an architectural schematic of a low-power wireless communication module (hereinafter referred to as the wireless communication module) for temperature and pressure monitoring according to an embodiment of the present invention. The wireless communication module 10 includes: a data acquisition module 11, an abnormal sample classification module 12, an abnormal risk prediction module 13, a wake-up command issuing module 14, and a temperature and pressure monitoring terminal 15. The modules are described below in sequence:

[0031] (1) Data acquisition module 11.

[0032] The data acquisition module 11 is responsible for communicating with all monitoring nodes deployed on the heating pipeline to acquire historical and real-time operating data, providing a data foundation for subsequent intelligent analysis and decision-making.

[0033] Optionally, the data acquisition module 11 is used to acquire historical operating data of the thermal pipeline and continuously acquire real-time pressure data of the boundary monitoring points. The historical operating data includes historical temperature and pressure data from multiple monitoring points, which are of various types, including boundary monitoring points and interval monitoring points. Multiple interval monitoring points are set between each pair of adjacent boundary monitoring points.

[0034] For example, such as Figure 2 As shown, the data acquisition module 11 may include two sub-modules: a historical data acquisition sub-module 111 and a real-time data acquisition sub-module 112. These will be described in detail below:

[0035] (1.1) Historical data collection submodule 111.

[0036] Optionally, the historical data acquisition submodule 111 is used to acquire historical temperature data and historical pressure data of boundary monitoring points and interval monitoring points at a preset frequency, and to preprocess and store them in a structured manner.

[0037] Specifically, the historical data acquisition submodule 111 establishes a connection with the temperature and pressure monitoring terminal 15 via a wireless communication protocol and configures the frequency according to the "classification acquisition" rule: a first preset frequency (once every 2 minutes) is configured for boundary monitoring points to ensure the data density of core nodes; a second preset frequency (at least once every 1 hour) is configured for interval monitoring points to balance data integrity and energy consumption. After acquisition, the historical data acquisition submodule 111 stores the data according to the dimension of "monitoring point number - acquisition time - temperature value - pressure value". At the same time, missing data is filled with the average of adjacent time points, and data exceeding the normal temperature range (-40℃ to +150℃) is marked to avoid invalid data interfering with subsequent analysis.

[0038] The preprocessing results (structured historical operation data) of the historical data acquisition submodule 111 are synchronized to the abnormal sample classification module 12 as the basic data for abnormal sample classification.

[0039] (1.2) Real-time data acquisition submodule 112.

[0040] Optionally, the real-time data acquisition submodule 112 is used to acquire real-time pressure data of boundary monitoring points and receive real-time temperature and pressure data of interval monitoring points after the wake-up command is triggered.

[0041] Specifically, the real-time data acquisition submodule 112 initiates a real-time pressure data request to the boundary monitoring terminal at a first preset frequency (once every 2 minutes), receives the data through a wireless communication protocol and records the acquisition time. The priority acquisition of pressure data is because when there is an anomaly in the thermal pipeline, the pressure change precedes the temperature change, which can shorten the anomaly response time.

[0042] When no wake-up command is received from the wake-up command sending module 14, the real-time data acquisition submodule 112 controls the main communication module of the control section monitoring terminal to enter a sleep state, and only maintains the wake-up receiving module to work; when the wake-up command is received, it synchronously receives the real-time temperature data and pressure data uploaded by the control section monitoring terminal to supplement the detailed data of the abnormal risk section.

[0043] The real-time pressure data of the boundary monitoring points of the real-time data acquisition submodule 112 is transmitted to the abnormal risk prediction module 13 for abnormal risk judgment; the real-time temperature and pressure data of the interval monitoring points are used for abnormal confirmation, providing supplementary support for subsequent working condition analysis.

[0044] (2) Abnormal sample classification module 12.

[0045] The abnormal sample classification module 12 is responsible for quantitative analysis and clustering of historical abnormal events based on the historical operation data provided by the data acquisition module 11, generating an abnormal sample set that includes a subset of blockage abnormal samples and a subset of leakage abnormal samples, providing a reference standard for the abnormal risk prediction module 13 to determine the real-time abnormal type and locate the abnormal risk section.

[0046] Optionally, the abnormal sample classification module 12 is used to classify historical abnormal events based on historical operating data to obtain an abnormal sample set that includes a subset of blockage abnormal samples and a subset of leakage abnormal samples.

[0047] Specifically, the execution process of the abnormal sample classification module 12 includes: First, calculating the severity of the anomaly at each monitoring point at each historical moment based on historical temperature data. This severity characterizes the extent to which the temperature deviates from its historical normal level. Second, calculating the pressure variation index at different historical moments for each monitoring point based on historical pressure data. This index is derived by weighting the deviation between the current pressure of its upstream neighboring monitoring points and its own historical average pressure, and is used to characterize the degree of abnormal pressure disturbance upstream. Third, filtering out preliminarily determined historical abnormal data points from all historical data based on the severity of the anomaly (e.g., setting a threshold). Then, for each historical abnormal data point, combining its corresponding severity of the anomaly and pressure variation index, calculating an anomaly differentiation index that can effectively distinguish between blockage and leakage tendencies by analyzing the overall data correlation characteristics and making compensation corrections. Finally, based on the anomaly differentiation index of all historical abnormal data points, using a preset clustering algorithm (such as K-means) for analysis, classifying all historical abnormal data points and forming structured subsets of blockage and leakage anomalies. These two sample subsets together constitute the abnormal sample set, which serves as the core feature knowledge base and is provided to the abnormal risk prediction module 13 for real-time comparison.

[0048] (3) Abnormal risk prediction module 13.

[0049] The abnormal risk prediction module 13 is responsible for determining the abnormal type of the thermal pipeline and locating the abnormal risk section (located between two pairs of adjacent boundary monitoring points) based on the real-time pressure data of the boundary monitoring points provided by the data acquisition module 11 and the abnormal sample set provided by the abnormal sample classification module 12. Its output results provide the core basis for the wake-up target of the wake-up command issuing module 14.

[0050] Optionally, the abnormal risk prediction module 13 is used to predict the abnormal risk section in the thermal pipeline located between two adjacent boundary monitoring points based on the real-time pressure data of the boundary monitoring points and the abnormal sample set.

[0051] For example, such as Figure 3As shown, the anomaly risk prediction module 13 may include two sub-modules: anomaly matching and judgment sub-module 131 and leakage prediction and location sub-module 132. These will be described below:

[0052] (3.1) Anomaly matching and judgment submodule 131.

[0053] Optionally, the anomaly matching and judgment submodule 131 is responsible for the detection and location of suspected blockage anomalies.

[0054] Specifically, the anomaly matching and determination submodule 131 processes the pressure data from the boundary monitoring points in real time from the data acquisition module 11. For each boundary monitoring point, it first calculates its real-time pressure variation index. When the index is greater than a first preset threshold, an anomaly prediction process is triggered. The anomaly matching and determination submodule 131 matches the real-time pressure characteristics of the current boundary monitoring point with each historical sample in the blockage anomaly sample subset provided by the anomaly sample classification module 12, and calculates a series of blockage similarities. The highest blockage matching degree is selected from these; if this highest value is greater than a second preset threshold, a blockage anomaly is determined to have occurred.

[0055] Furthermore, the anomaly matching and judgment submodule 131, based on the positional relationship between the anomaly location points recorded in the historical blockage sample corresponding to the highest blockage matching degree and their associated historical boundary monitoring points, determines the corresponding anomaly risk section in the current pipeline, located between two adjacent boundary monitoring points, through a proportional mapping method. The judgment result and location information will be directly output to the wake-up command issuing module 14.

[0056] (3.2) Leakage prediction and location submodule 132.

[0057] Optionally, the leakage prediction and location submodule 132 is responsible for detecting and locating suspected leakage anomalies when the current situation does not meet the characteristics of blockage.

[0058] Specifically, when the highest blockage matching degree calculated by the anomaly matching and judgment submodule 131 is not greater than the second preset threshold, the process transfers to this submodule. The leakage prediction and location submodule 132 matches the pressure characteristics of the current boundary monitoring point with each historical sample in the leakage anomaly sample subset to obtain multiple leakage similarities. Then, based on these leakage similarities and the leakage point location information recorded in their respective historical leakage samples, a leakage reference distance calculated from the current boundary monitoring point is obtained through weighted calculation.

[0059] Furthermore, the leak prediction and location submodule 132 takes the current boundary monitoring point as the starting point and moves upstream along the pipeline. It determines the pipeline location point that is a distance from the starting point to the leak reference distance as the predicted leak point, and predicts the pipeline section where the predicted leak point is located, defined by two adjacent boundary monitoring points, as an abnormal risk section. This location result is also output to the wake-up command issuing module 14.

[0060] (4) Wake-up command issuing module 14.

[0061] The wake-up command issuing module 14 is responsible for generating and issuing wake-up commands based on the abnormal risk section information provided by the abnormal risk prediction module 13, controlling the monitoring points in the control section to upload real-time temperature and pressure data, and achieving low power consumption control through targeted wake-up, that is, only waking up necessary monitoring points to avoid redundant energy consumption.

[0062] Optionally, the wake-up command issuing module 14 is used to send a wake-up command to at least one interval monitoring point in the abnormal risk section to instruct it to upload temperature and pressure data.

[0063] Specifically, the wake-up command issuing module 14 receives abnormal risk segment information from the abnormal risk prediction module 13. It first determines one or more interval monitoring points located within the segment as core wake-up points based on the segment's range. To obtain more comprehensive data for cross-validation and precise location, the wake-up command issuing module 14 also identifies at least one other monitoring point (which may be an interval monitoring point or a boundary monitoring point) directly adjacent to the core wake-up point on the pipeline topology as a cooperative wake-up point.

[0064] Furthermore, the wake-up command issuing module 14 generates a low-power wireless wake-up frame (i.e., a wake-up command) containing the dedicated address codes of these core wake-up points and collaborative wake-up points, and sends it via broadcast or targeted transmission through the gateway. This command is recognized by the low-power wake-up receiver built into the target monitoring point, thereby triggering it to switch from sleep mode to normal working mode, complete a temperature and pressure acquisition, and upload the data via the wireless network to achieve on-demand and accurate data acquisition.

[0065] (5) Temperature and pressure monitoring terminal 15.

[0066] The temperature and pressure monitoring terminal 15 is the data acquisition and execution unit of the wireless communication module 10. It provides direct temperature and pressure data acquisition support for the data acquisition module 11. According to the monitoring point type, it is divided into boundary monitoring terminal and interval monitoring terminal. The two have the same hardware configuration but different operating strategies.

[0067] Optionally, the temperature and pressure monitoring terminal 15 is used to be deployed at boundary monitoring points or interval monitoring points. It collects temperature and pressure data through built-in sensors, uploads data at a preset frequency or by wake-up command, and supports low-power sleep mode.

[0068] The wireless communication module 10 and its constituent modules have been described above.

[0069] For example, such as Figure 4 The diagram shown is a flowchart illustrating a low-power wireless communication method for temperature and pressure monitoring according to an embodiment of the present invention, comprising the following steps:

[0070] S401. Obtain historical operating data of the heating pipeline. The historical operating data includes historical temperature and pressure data from multiple monitoring points on the heating pipeline. The types of multiple monitoring points include boundary monitoring points and interval monitoring points. Multiple interval monitoring points are set between each pair of adjacent boundary monitoring points.

[0071] For example, this step can be performed by the historical data acquisition submodule 111 in the data acquisition module 11 described above, specifically including: the historical data acquisition submodule 111 establishes a connection with the temperature and pressure monitoring terminal 15 through a wireless communication protocol, and configures the frequency according to the classification acquisition rules: configure a first preset frequency (once every 2 minutes) for the boundary monitoring points to ensure the data density of the core nodes; configure a second preset frequency (at least once every 1 hour) for the interval monitoring points to balance data integrity and energy consumption.

[0072] Optionally, the first preset frequency can be set to once every 2 minutes to achieve dense sampling of the temperature and pressure status of key nodes (i.e., boundary monitoring points) of the pipeline without excessively increasing the energy consumption of the boundary monitoring points, thereby providing highly timely data input for the real-time anomaly prediction module; the second preset frequency can be set to once every 1 hour to ensure that all interval monitoring points can upload data regularly, maintain the integrity of basic data timing and system connectivity self-check, while placing them in a very low-power sleep listening state most of the time, thereby minimizing the network communication load and node energy consumption under normal conditions.

[0073] After the data collection is completed, the historical data collection submodule 111 stores the data according to four dimensions: monitoring point number, collection time, temperature value, and pressure value. At the same time, missing data is filled with the average of adjacent time points, and data that exceeds the normal temperature range (-40℃~+150℃) is marked to avoid invalid data interfering with subsequent analysis.

[0074] The historical data acquisition submodule 111 will synchronize the processed structured historical operation data to the abnormal sample classification module 12 as the basis data for subsequent abnormal sample classification.

[0075] S402. Classify historical abnormal events based on historical operational data to obtain an abnormal sample set. This abnormal sample set includes a subset of blockage abnormal samples and a subset of leakage abnormal samples.

[0076] For example, this step can be performed by the anomaly sample classification module 12 described above, specifically including: First, calculating the severity of the anomaly at each monitoring point at each historical moment based on historical temperature data, which characterizes the extent to which the temperature deviates from its historical normal level. Second, calculating the pressure variation index at different historical moments for each monitoring point based on historical pressure data. This index is derived by weighting the deviation between the current pressure of its upstream neighboring monitoring points and its own historical average pressure, and is used to characterize the degree of abnormal disturbance in upstream pressure. Third, filtering out preliminarily determined historical anomaly data points from all historical data based on the severity of the anomaly (e.g., setting a threshold). Then, for each historical anomaly data point, combining its corresponding severity of the anomaly and pressure variation index, calculating an anomaly differentiation index that can effectively distinguish between blockage and leakage tendencies by analyzing the overall data correlation characteristics and making compensation corrections. Finally, based on the anomaly differentiation index of all historical anomaly data points, using a preset clustering algorithm (such as K-means) for analysis, classifying all historical anomaly data points and forming structured blockage anomaly sample subsets and leakage anomaly sample subsets. It should be noted that the specific procedures for the aforementioned sub-steps are described in S501-S504 below, and will not be repeated here.

[0077] In another possible implementation, the abnormal sample classification module 12, when classifying historical abnormal events, can also calculate a comprehensive abnormal feature value by assigning preset fusion weights to three features: temperature deviation, upstream pressure fluctuation, and duration. Then, a preset clustering algorithm is used to divide the historical abnormal data points into two sample subsets: blockage and leakage. This method, by introducing a time dimension, is suitable for scenarios where the duration of abnormal events differs significantly.

[0078] Alternatively, the anomaly sample classification module 12 can replace the clustering algorithm with a density-based clustering algorithm. This algorithm automatically forms clusters based on the distance and density between sample points, without requiring a preset number of clusters. It can effectively handle situations where historical anomaly samples are unevenly distributed or contain isolated points, thereby improving the robustness and adaptability of the classification.

[0079] Therefore, the abnormal sample classification module 12 can extract and learn the feature patterns of different abnormal types by performing offline analysis on historical operating data, and transform the original data into a set of classified abnormal samples, providing a searchable and comparable intelligent judgment basis for subsequent real-time prediction.

[0080] S403. Obtain real-time pressure data from boundary monitoring points, and based on the real-time pressure data and the set of abnormal samples, predict abnormal risk sections in the heating pipeline. The abnormal risk section is located between two adjacent boundary monitoring points.

[0081] For example, this step can be performed by the anomaly risk prediction module 13 described above, specifically including the following steps: First, continuously acquire real-time pressure data from each boundary monitoring point and calculate its real-time pressure variation index. The calculation logic of this index is consistent with the pressure variation index in historical data, used to characterize the degree of abnormal disturbance of upstream pressure at the current moment. Second, when the real-time pressure variation index of any boundary monitoring point exceeds a preset trigger threshold, the anomaly prediction process is initiated. Then, the current pressure characteristics are matched with a subset of blocked anomaly samples in the anomaly sample set to calculate the blockage matching degree, which characterizes the degree of similarity. If the highest blockage matching degree exceeds the confidence threshold, it is determined to be a blockage anomaly, and the current anomaly risk segment is mapped and located based on the location information recorded by the historical blockage sample with the highest matching degree. If the blockage matching degree does not reach the threshold, the process switches to leakage prediction, matching the current pressure characteristics with a subset of leak anomaly samples, locating the suspected leak point by calculating the leakage reference distance, and thus determining the anomaly risk segment in which it is located. It should be noted that the specific process of the aforementioned prediction and location process is described in S601-S606 below, and will not be repeated here.

[0082] In another possible implementation, when performing anomaly type matching, the anomaly risk prediction module 13 can use different similarity calculation models (such as cosine similarity based on feature vectors or dynamic time warping algorithm based on time series morphology) to evaluate the degree of matching between real-time data and historical samples, so as to adapt to different data fluctuation characteristics.

[0083] Alternatively, when predicting leakage risk zones, the abnormal risk prediction module 13 can make joint judgments based on multiple dimensions such as pressure drop rate and upstream and downstream pressure gradient, in addition to distance mapping, thereby improving the accuracy of positioning under complex working conditions.

[0084] Therefore, the anomaly risk prediction module 13 can quickly compare and intelligently reason based on real-time monitoring data and an offline anomaly sample knowledge base to predict the precise pipeline sections where blockages or leaks may occur. The output of this step directly determines the target range of subsequent wake-up commands and is a key decision-making step in realizing the transition from continuous monitoring to precise investigation.

[0085] S404. Send a wake-up command to at least one monitoring point in the abnormal risk section. The wake-up command instructs at least one monitoring point in the abnormal risk section to upload temperature and pressure data.

[0086] For example, this step can be executed by the wake-up command issuing module 14 described above, and specifically includes the following steps:

[0087] (1) Identify at least one target interval monitoring point located in the abnormal risk zone as the core wake-up point.

[0088] Specifically, the wake-up command issuing module 14 selects at least one point as the core wake-up point from all the interval monitoring points deployed in the abnormal risk section predicted by S403. The selection rule can be to select the interval monitoring point closest to the predicted abnormal location (such as the blockage point predicted by S603 or the leakage point predicted by S606), or to select according to the rules preset by engineering experience.

[0089] (2) At least one monitoring point adjacent to the core wake-up point on the thermal pipeline topology is identified as the collaborative wake-up point.

[0090] In this step, to obtain more comprehensive field data for anomaly cross-validation and precise location, the wake-up command issuing module 14 will further determine the collaborative wake-up point. The collaborative wake-up point refers to at least one monitoring point directly adjacent to the determined core wake-up point on the pipeline topology corresponding to the heating pipeline. This monitoring point can be another interval monitoring point or a boundary monitoring point serving as the boundary of the section.

[0091] (3) Generate and send a wake-up command containing the core wake-up point address and the cooperative wake-up point address.

[0092] For example, the wake-up command issuing module 14 encodes the independent device addresses of the core wake-up point and the cooperative wake-up point as determined above, generating a specific low-power wireless wake-up signal, i.e., a wake-up command. The wake-up command is injected into the wireless channel through the system gateway in a broadcast or directional manner. The interval monitoring point located at the target location (i.e., the device address in the wake-up command) continuously listens to the channel with its built-in low-power wake-up receiver. When it receives a wake-up command that matches its own address, it immediately starts the main control and communication module of the monitoring point, controls the sensor to collect the current temperature and pressure data, and immediately uploads it through the wireless network.

[0093] Based on the above technical solutions, this invention, by constructing a historical anomaly sample knowledge base and performing intelligent matching and prediction based on real-time pressure characteristics, realizes a paradigm shift from fixed-frequency uploading by all nodes to precise wake-up based on anomaly risk. This significantly reduces the number of wireless communication calls and network load while ensuring the timeliness of anomaly monitoring in the heating network, greatly reduces the energy consumption of monitoring nodes, effectively extends the service life of battery-powered nodes and the independent operation cycle of the system, and maintains the ability to quickly locate and respond to critical faults such as network blockage and leakage.

[0094] For example, in combination Figure 4 ,like Figure 5The diagram shown is a flowchart illustrating another low-power wireless communication method for temperature and pressure monitoring according to an embodiment of the present invention. In this method, historical abnormal events are classified based on historical operational data to obtain an abnormal sample set, specifically including the following steps:

[0095] S501. Based on historical temperature data, determine the severity of the anomaly at each monitoring point at each historical moment. The severity of the anomaly is used to characterize the extent to which the temperature deviates from the normal level.

[0096] In this step, for a given monitoring point a, the abnormal sample classification module 12 first calculates the arithmetic mean of its temperature at all times during the entire historical operation of the heating network, and records it as the historical temperature benchmark for that monitoring point. For the temperature value collected at any historical time b at this monitoring point. Calculate its relationship with historical temperature benchmarks The difference is then processed by a preset linear rectification function to filter out values ​​indicating temperature increases (which typically do not directly indicate blockage or leakage), retaining only the differences indicating temperature decreases. Finally, the result is normalized to obtain the severity of the anomaly at monitoring point a at historical time b. For example, the severity of the anomaly. The calculation formula is:

[0097]

[0098] In the above formula, This represents the quantitative value indicating the severity of the anomaly at monitoring point a at time b. This represents the raw temperature value collected at monitoring point a at time b; This represents the average temperature at monitoring point a throughout the entire historical operation period; It is a linear rectified function, ensuring that the output is zero when the input value is negative; This is a normalization function used to map numerical values ​​to a preset standard interval [0,1]. The larger the value, the more significant the heat loss at monitoring point a at time b, and the higher the degree of anomaly. For example, the aforementioned normalization function is a maximum-minimum normalization function. The maximum and minimum values ​​are preset empirical extreme values ​​derived from a large amount of historical experimental data. If the calculation result exceeds the [0, 1] interval, it is restricted to the [0, 1] range by a truncation function (i.e., if the result is less than 0, it is taken as 0; if it is greater than 1, it is taken as 1) to eliminate the influence of outliers on the evaluation indicators.

[0099] S502. Based on historical pressure data, determine the pressure variation index for each monitoring point. The pressure variation index is used to characterize the degree of abnormal disturbance in upstream pressure fluctuations at each historical moment.

[0100] For example, when the abnormal sample classification module 12 performs this step, it specifically includes the following steps:

[0101] (1) For each monitoring point, obtain the historical pressure data of a preset number of adjacent monitoring points upstream of the monitoring point.

[0102] In this step, the abnormal sample classification module 12 obtains the pressure values ​​of a preset number E adjacent monitoring points upstream of a monitoring point a at the same time b in historical time b.

[0103] Understandably, the preset quantity E is a positive integer pre-defined based on the actual pipeline topology and monitoring point density, used to limit the number of upstream adjacent monitoring points participating in the calculation of the pressure variation index. Its typical value is usually between 3 and 7, for example, E=5. The principle for selecting the value of E is: it should be sufficient to cover the adjacent upstream area that may have a significant pressure impact on the target monitoring point, so as to fully capture the spatial transmission characteristics of pressure fluctuations; at the same time, it should also avoid including monitoring points that are too far away, have a weak influence, or are irrelevant, to prevent the introduction of irrelevant noise and ensure the accuracy of the pressure variation index and the efficiency of the calculation. In specific implementation, the value of E can be determined comprehensively based on the physical length of the pipeline, the average distance between adjacent monitoring points, and the fluid dynamics characteristics.

[0104] (2) Determine the pressure variation index based on the historical pressure data of a preset number of adjacent monitoring points, the historical pressure data of the monitoring points, and the distance between each adjacent monitoring point.

[0105] Specifically, for monitoring point a at time b, its upstream E neighboring monitoring points are first sorted in order of distance from monitoring point a from the pipe, and assigned serial numbers c=1, 2, ..., E, where c=1 represents the monitoring point closest to monitoring point a, and c=E represents the monitoring point farthest from monitoring point a.

[0106] For the c-th upstream adjacent monitoring point after sorting, the abnormal sample classification module 12 obtains its pressure value. and its own average pressure value throughout the entire historical operation period. Then, the pressure deviation at this upstream point is calculated. .

[0107] Furthermore, to reflect the impact of distance on pressure propagation, we assign a weight to each upstream monitoring point. This weight is inversely proportional to its distance from the index c; that is, the closer the distance, the greater the weight. Specifically, the initial weight value for each index c is first calculated. Then, the softmax function is used to normalize all E initial weights to obtain the final weights: .

[0108] The weights obtained in this way satisfy =1, and from arrive The values ​​decrease sequentially, thus realizing the physical logic that the closer the distance, the greater the weight.

[0109] Following this, the weighted pressure deviations of all E upstream adjacent monitoring points are summed, and the summation result is normalized to obtain the pressure change index of monitoring point a at time b. For example, the pressure variation index. The calculation formula is:

[0110]

[0111] In the above formula, E represents the pressure change index at monitoring point a at time b; E represents the preset number of upstream adjacent monitoring points, which is a positive integer. This represents the pressure value at time b of the c-th upstream adjacent monitoring point; This represents the average pressure value of the c-th upstream adjacent monitoring point during the entire historical operation period; This represents the weight calculated based on the index c, ensuring that the sum of all weights is 1 and is negatively correlated with distance; This is the normalization function. (Exponential) The abnormal disturbance degree of pressure in the upstream area of ​​the target monitoring point relative to its historical normal state was comprehensively quantified. For example, the aforementioned normalization function is a maximum-minimum value normalization function. The maximum and minimum values ​​are preset empirical extreme values ​​derived from a large amount of historical experimental data. If the calculation result exceeds the interval [0, 1], it is restricted to the range [0, 1] by a truncation function (i.e., if the result is less than 0, it is taken as 0; if it is greater than 1, it is taken as 1) to eliminate the influence of outliers on the evaluation indicators.

[0112] S503. Based on the severity of the anomaly and the pressure variation index, determine the anomaly differentiation index for each historical anomaly event. The anomaly differentiation index characterizes the probability that a historical anomaly event belongs to a blockage anomaly.

[0113] For example, when the abnormal sample classification module 12 performs this step, it specifically includes the following steps:

[0114] (1) Based on the historical operational data corresponding to all historical abnormal events, analyze the correlation characteristics between the pressure variation index and the severity of the abnormality.

[0115] Specifically, the anomaly sample classification module 12 filters historical anomaly events from historical operational data based on the severity of the anomaly, forming an initial set of historical anomaly events. The filtering rule is: for any monitoring point a at time b, if its anomaly severity is... If the value exceeds a preset anomaly threshold (e.g., 0.3), the monitoring point-time pair (a, b) is identified as a preliminary anomaly data point and included in the initial historical anomaly event set. Each event in this set uses an index ( The unique identifier is d, where d is the abnormal monitoring point number and e is the abnormal time.

[0116] It should be noted that the aforementioned anomaly detection threshold can be set based on historical data analysis or engineering experience. Its purpose is to efficiently identify potentially abnormal moments worthy of further analysis from massive amounts of normal data, without excessive underreporting. Through this step, the system transforms the continuous historical data stream into a set of discrete data points with a high probability of anomalies, laying the foundation for subsequent focused analysis.

[0117] For each event (d, e) in the set, there are two calculated characteristic values: the pressure variation index. and severity of abnormality Following this, to quantify the statistical strength and direction of the correlation between the stress variation index and the severity of anomalies across all historical anomalies, a weighted Pearson correlation coefficient G was calculated for these two feature sequences as the correlation feature. In calculating this correlation coefficient G, the weight was the stress variation index of each event itself. The correlation coefficient G quantifies the strength and direction of the statistical association between the pressure variation index and the severity of the anomaly across all historical anomalies. This association characteristic reflects the positive correlation pattern that the two should exhibit in typical blockage anomalies.

[0118] (2) For each historical abnormal event, the pressure change index of the historical abnormal event is compensated and corrected according to the correlation characteristics to obtain the compensation value.

[0119] In this step, the abnormal sample classification module 12 targets each historical abnormal event in the initial set of historical abnormal events ( ), perform the following operations: First, record historical abnormal events ( When calculating the globally weighted Pearson correlation coefficient G, the weights are replaced with 1, while the weights of other events in the set remain unchanged, resulting in a temporary weighted Pearson correlation coefficient. . This demonstrates the correlation between stress and anomaly severity while maintaining the influence of the current event's own stress variation index on global relevance. Subsequently, the current event's own stress variation index is used... Global correlation coefficient G and temporary correlation coefficient Calculate this historical anomalous event ( compensation value Optionally, the compensation value Calculated using the following formula:

[0120]

[0121] In the above formula, Indicates historical anomalies ( The compensation value; Indicates historical anomalies ( The pressure variation index; G represents the global correlation coefficient; Indicates historical anomalies ( The temporary correlation coefficient of ) (the calculation method is described in the previous paragraph); Represents the normalization function; This represents the parameter tuning coefficient, which is a very small positive number, such as 10 to the power of negative 10, to avoid division by zero errors. The significance of the formula is that the greater the difference between the characteristics of the current historical anomaly and the typical congestion correlation pattern (i.e., The larger the value, the greater the compensation value it receives. The smaller the value, the better. For example, the aforementioned normalization function is a maximum-minimum value normalization function. The maximum and minimum values ​​are preset empirical extreme values ​​derived from a large amount of historical experimental data. If the calculation result exceeds the interval [0, 1], it is restricted to the range [0, 1] by a truncation function (i.e., if the result is less than 0, it is taken as 0, and if it is greater than 1, it is taken as 1) to eliminate the influence of outliers on the evaluation index.

[0122] (3) Determine the abnormality differentiation index based on the compensation value and the pressure change index of historical abnormal events.

[0123] Optionally, the anomaly sample classification module 12 will classify each historical anomaly event ( Pressure variation index Its corresponding compensation value Add them together to obtain the anomaly distinguishing index for the event. .

[0124] Understandably, anomaly differentiation indicators It integrates historical anomalies ( The intensity of its own pressure disturbance and the degree of agreement between its pressure-temperature correlation pattern and typical blockage patterns. The higher the value, the greater the probability that the historical anomaly is a blockage anomaly.

[0125] S504. Based on the anomaly differentiation index, perform cluster analysis on all historical anomaly events to obtain a subset of blockage anomaly samples and a subset of leakage anomaly samples.

[0126] For example, the anomaly sample classification module 12 classifies each historical anomaly event ( Anomaly Differentiation Indicators Using the sample features as the basis, a preset clustering algorithm is adopted and the number of clusters k=2 is set to determine the clustering between samples. The absolute value of the difference in values ​​is used as a distance metric to divide all historical anomalous events. Optionally, the preset clustering algorithm can be the k-means algorithm.

[0127] Therefore, through cluster analysis, the anomaly sample classification module 12 divides historical anomaly events into two category clusters. The clusters of all samples within one category are then... Clusters with higher average values ​​are identified and labeled as a subset of blocked anomalous samples; all samples within another category cluster are... Clusters with lower average values ​​are identified and marked as a subset of leakage anomalies. These two subsets together constitute the set of anomalies used for subsequent real-time comparisons.

[0128] It should be noted that each historical anomaly sample categorized into the blockage anomaly sample subset or the leakage anomaly sample subset retains all the feature data calculated in steps S501 to S503, mainly including:

[0129] (1) Calculation characteristics: the severity of the anomaly corresponding to the event. With pressure change index ;

[0130] (2) Identification information: The inherent location and time identifier of the event, namely the monitoring point number d and the abnormal time e.

[0131] Therefore, the resulting set of abnormal samples is a structured historical knowledge base, which not only contains the classification labels of events (blockage / leakage), but more importantly, stores the typical characteristic patterns of each type of event. , The data includes features such as time and space information of their occurrence, thus providing a direct data foundation for real-time feature matching and risk segment prediction in subsequent steps.

[0132] Based on the above technical solution, this embodiment of the invention transforms raw historical monitoring data into a structured anomaly feature knowledge base through an offline data processing workflow. Its technical effects are as follows: First, by independently quantifying temperature deviations and upstream pressure disturbances, the pipeline physical state is converted into calculable feature parameters; second, through correlation analysis and compensation correction, temperature and pressure features are creatively integrated to generate a single quantitative indicator that can significantly distinguish between blockage and leakage tendencies; finally, based on this indicator, cluster analysis is performed to automatically and accurately classify historical anomaly events. The entire process achieves the goal of extracting and solidifying the features of two typical fault modes from mixed data, providing a reliable and directly accessible data foundation for rapid pattern matching and intelligent on-demand wake-up decisions in subsequent real-time monitoring. This is the core prerequisite for achieving low-power intelligent monitoring in this solution.

[0133] For example, in combination Figure 4 ,like Figure 6 The diagram shown is a flowchart illustrating another low-power wireless communication method for temperature and pressure monitoring according to an embodiment of the present invention. In this method, based on real-time pressure data and anomaly sample sets, abnormal risk sections in thermal pipelines are predicted, specifically including the following steps:

[0134] S601. For each boundary monitoring point, determine the pressure variation index of the boundary monitoring point based on the real-time pressure data of the boundary monitoring point.

[0135] Specifically, the anomaly matching and judgment submodule 131 calculates the real-time pressure variation index for a boundary monitoring point f, which serves as the current prediction starting point, using the same logic as in S502. However, its data source and calculation object are adapted for the real-time scenario. That is, the data source for calculating the real-time pressure variation index here is real-time pressure data, specifically referring to the pressure readings at the current moment for boundary monitoring points f that continuously collect and upload data at a higher frequency (i.e., the first preset frequency); and the calculation object changes from each monitoring point to the boundary monitoring point f. The calculation method remains unchanged from the formula in S502, and the resulting real-time pressure variation index for the boundary monitoring point f is denoted as... . It characterizes in real time the degree of pressure anomaly disturbance in the upstream region of the boundary monitoring point f at the current moment.

[0136] S602. When the pressure fluctuation index of the boundary monitoring point is greater than a first preset threshold, the pressure fluctuation index of the boundary monitoring point is matched with the blockage anomaly sample subset to determine the highest blockage matching degree of the boundary monitoring point. The highest blockage matching degree is the highest value among the blockage matching degrees between the boundary monitoring point and each historical blockage sample in the blockage anomaly sample subset.

[0137] In this step, the anomaly matching and judgment submodule 131 first determines the real-time pressure change index of the boundary monitoring point f. Is it greater than the first preset threshold? The first preset threshold is used to determine whether the abnormal prediction process is triggered. Its value can be determined based on the statistical distribution of the pressure variation index under historical normal operating conditions. For example, it can be set to 0.2.

[0138] Real-time pressure change index at boundary monitoring point f If the value is greater than the first preset threshold, By matching the historical pressure variation index corresponding to each historical blockage sample g in the abnormal blockage sample subset, the blockage matching degree between the boundary monitoring point f and the historical blockage sample g is calculated. During the calculation, The pressure change index recorded in the historical blockage sample g at the time of its anomaly. A comparison is performed, and the absolute value of the difference between the two is used as the basis for further analysis. Calculate the blocking matching degree The degree of congestion matching The closer the value is to 1, the more similar the current pressure disturbance characteristics are to the historical blockage sample.

[0139] Furthermore, after traversing all historical congestion samples, from all congestion matching degrees The maximum value among them is recorded as the highest blocking match degree. .

[0140] S603. If the highest blockage matching degree at the boundary monitoring point is greater than the second preset threshold, the anomaly type is determined to be blockage, and the abnormal risk section is determined based on the location information of the historical blockage sample record corresponding to the highest blockage matching degree.

[0141] For example, when the anomaly matching and determination submodule 131 performs this step, it specifically includes the following steps:

[0142] (1) Determining the type of abnormality.

[0143] In this sub-step, the anomaly matching and judgment submodule 131 will obtain the highest blocking matching degree. The result is compared with a second preset threshold. The second preset threshold is used to further confirm the anomaly type as congestion after triggering the prediction. Its value can be set based on the distribution of the matching degree of congestion samples in historical data and the classification accuracy requirements. For example, it can be set to 0.5.

[0144] like If the value exceeds this threshold, the anomaly occurring near the boundary monitoring point f is determined to be a blockage.

[0145] (2) Location mapping and segment prediction.

[0146] The anomaly matching and judgment submodule 131 calculates the highest blocking matching degree. The corresponding historical congestion sample g max Spatial mapping is performed on the location information recorded in (denoted as the best matching sample):

[0147] First, from the best matching sample g max To obtain the recorded historical anomaly locations, and for location mapping, it is necessary to determine a pair of upstream and downstream reference monitoring points associated with each historical anomaly location. The method for determining these points is as follows: among all monitoring points downstream of the historical anomaly location, select the pressure change index at that time of the anomaly and the real-time pressure change index of the current boundary monitoring point f. The closest monitoring point is selected as the downstream reference monitoring point. Similarly, among all monitoring points upstream of the historical anomaly location, the monitoring point whose pressure change index at the time of the anomaly is closest to the real-time pressure change index of the upstream adjacent boundary monitoring point of the current boundary monitoring point f is selected as the upstream reference monitoring point. Then, the actual pipeline distance between boundary monitoring point f and its upstream adjacent boundary monitoring point is obtained. .

[0148] Following this, the anomaly matching and determination submodule 131 determines the best matching sample g. max The proportion of historical anomalies recorded in the data within the relative positions of their upstream and downstream reference monitoring points. This ratio Multiply by the actual pipeline distance This allows for the calculation of the specific location of the predicted blockage point within the interval defined by the boundary monitoring point f and its upstream adjacent boundary monitoring point. The pipeline section where the predicted blockage point is located, jointly defined by the boundary monitoring point f and its upstream adjacent boundary monitoring point, is thus identified as the abnormal risk section for this instance.

[0149] S604. When the highest blockage matching degree of the boundary monitoring point is less than or equal to the second preset threshold, the pressure change index of the boundary monitoring point is matched with each historical leak sample in the abnormal leak sample subset to determine multiple leak similarities. Among them, a leak similarity is used to characterize the degree of similarity between the boundary monitoring point and a historical leak sample.

[0150] In this step, if the highest blockage matching degree is determined in step S603... If the value is not greater than the second preset threshold, the system switches to the leakage anomaly prediction process:

[0151] Leakage prediction and location submodule 132 calculates the leakage similarity between the boundary monitoring point f and the historical leakage sample h for each historical leakage sample h in the subset of leakage anomaly samples. In the specific calculation, for a historical leakage sample h, among all the downstream monitoring points recorded, the pressure change index and the real-time pressure change index of the current boundary monitoring point f are selected. The nearest downstream monitoring point is used as the downstream reference point for the historical leak sample h. Then, the real-time pressure change index of the boundary monitoring point f is used. Compare with the pressure variation index at the downstream reference point, and through Calculate the leakage similarity ,in This represents the pressure variation index at the downstream reference point of historical sample h. Leakage similarity. The closer the value is to 1, the more similar the current pressure disturbance characteristics are to the historical leak sample.

[0152] Furthermore, after traversing all historical leak samples, a set of leak similarity scores is obtained. , where P is the total number of abnormal leakage samples.

[0153] S605. Determine the leak reference distance based on multiple leak similarities and the location of leak points in historical leak samples corresponding to multiple leak similarities.

[0154] In this step, based on the set of leakage similarities obtained in the previous step... Based on the leak location information recorded in the corresponding historical leak samples h, a comprehensive leak reference distance is calculated. This includes the following steps:

[0155] (1) For each historical leakage sample h (h=1, 2, 3…, P, where P represents the total number of historical leakage samples in the subset of leakage anomaly samples), calculate its reference score. :

[0156]

[0157] in, This indicates the leakage similarity between a boundary monitoring point f and a downstream reference point of its historical leakage sample h; This represents the absolute value of the real-time pressure variation index difference between boundary monitoring point f and its upstream adjacent boundary monitoring point; This represents the parameter tuning coefficient, which is a very small positive number, such as 10 to the power of negative 10, to avoid division by zero errors.

[0158] It should be pointed out that, This represents a reference pressure change difference, calculated based on the record of a historical leak sample h: First, combining the downstream reference point of the historical leak sample h determined in S604, upstream of this downstream reference point, find a point at a distance equal to... The monitoring point (the actual pipeline distance between boundary monitoring point f and its upstream adjacent point) is denoted as the corresponding upstream location point. It is the absolute value of the difference between the pressure change index recorded at the downstream reference point and the corresponding upstream location point at the abnormal moment of the historical leakage sample h.

[0159] (2) Reference scores for all P historical leakage samples h Perform softmax normalization to obtain the weights for each sample. :

[0160]

[0161] Among them, this weight satisfy This reflects the degree of matching between the current leakage characteristics and historical sample h.

[0162] (3) For example, the leakage prediction and location submodule 132 calculates the leakage reference distance according to the following formula. :

[0163]

[0164] In the above formula, The reference distance to the leak is represented by P, which represents the total number of historical leak samples in the subset of leak anomaly samples. This represents the weight of each historical leaked sample. It should be noted that... This represents the actual pipeline distance between the recorded leak point and the corresponding point defined above in the historical leak sample h.

[0165] It is understandable that the formula in step (1) above compares the characteristics of current real-time pressure changes. Characteristics of each historical leak sample The degree of proximity is used to assign a weight to each historical leak sample. The smaller the value, the more similar the current leakage characteristics are to sample h, and the greater its corresponding weight. This indicates the distance between the leakage point recorded in the sample and the sample h. The greater its contribution to the weighted summation, the better. In this equation, the numerator represents the point state similarity, i.e., how close the pressure disturbance at the boundary monitoring point f is to the pressure state at a certain downstream location in a historical leak event h; the denominator represents the gradient pattern similarity, i.e., how close the pressure drop gradient characteristics caused by the leak point are to the boundary monitoring point f and the historical leak sample h. This fractional term unifies the two key dimensions of downstream point pressure matching and upstream-downstream pressure change pattern matching into a single quantitative index. The score comprehensively reflects the overall reference value of a historical leak sample h. A higher score means that the historical sample simultaneously satisfies both downstream point pressure state similarity and pressure gradient change pattern similarity, and therefore it should be given greater weight when predicting the current leak location. The Softmax function converts this score into normalized probability weights, ensuring a reasonable allocation of weights.

[0166] The final result It is a predicted distance based on a weighted average of historical similar cases, used to characterize how far upstream the leak point may be from the current boundary monitoring point f.

[0167] S606. The location point at which the leakage reference distance is located upstream of the heat pipeline, starting from the boundary monitoring point, shall be determined as the predicted leakage point. The pipeline section where the predicted leakage point is located, defined by two adjacent boundary monitoring points, shall be predicted as an abnormal risk section.

[0168] In this step, the leak prediction and location submodule 132 first performs single-point location: taking the boundary monitoring point f as the spatial starting point, along the upstream direction of the heat pipe flow, a length equal to the leak reference distance is measured. The pipeline path is determined, and the endpoint of this path is identified as the predicted leak point. Then, a segment determination is performed: within the entire pipeline network topology corresponding to the heating pipeline, the two adjacent boundary monitoring points between which the predicted leak point is located are identified. This pipeline segment jointly defined by these two boundary monitoring points is predicted as the abnormal risk segment caused by the leak anomaly.

[0169] Based on the above technical solution, this invention achieves real-time type identification and preliminary spatial location of two types of anomalies—blockage and leakage—in thermal pipelines by performing online and rapid intelligent comparison and reasoning between the real-time pressure disturbance characteristics of boundary monitoring points and an offline anomaly sample knowledge base. Its core lies in: first, intelligently triggering the prediction process based on whether the real-time pressure change index exceeds a threshold, avoiding invalid calculations when there are no anomalies; second, efficiently determining blockage anomalies by calculating the matching degree with historical blockage samples and mapping precise risk sections using the positional relationships in the samples; finally, when the characteristics do not match blockage, automatically switching to the leakage prediction process, and weightedly estimating the reference distance of the leakage point and locating the risk section based on the pressure change patterns and distance information of historical leakage samples. The entire process uses pressure, a rapidly responding parameter, as the basis for judgment, ensuring the timeliness of early anomaly detection and transforming continuous monitoring of the entire pipeline network into precise focusing on a few suspected risk sections, thus providing a direct and clear decision-making basis for subsequent on-demand wake-up strategies, fundamentally supporting the unity of low power consumption and high timeliness in the system.

[0170] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0171] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A low-power wireless communication method for temperature and pressure monitoring, characterized in that, The method includes: Acquire historical operating data of the heating pipeline; wherein, the historical operating data includes historical temperature data and historical pressure data of multiple monitoring points on the heating pipeline, the types of the multiple monitoring points include boundary monitoring points and interval monitoring points, and multiple interval monitoring points are set between each pair of adjacent boundary monitoring points; Based on the historical operational data, historical abnormal events are classified to obtain an abnormal sample set; wherein, the abnormal sample set includes a subset of blockage abnormal samples and a subset of leakage abnormal samples; The real-time pressure data of the boundary monitoring points is obtained, and based on the real-time pressure data and the abnormal sample set, the abnormal risk section in the thermal pipeline is predicted; wherein the abnormal risk section is located between two adjacent boundary monitoring points. A wake-up command is sent to at least one of the interval monitoring points in the abnormal risk zone; wherein the wake-up command is used to instruct at least one of the interval monitoring points in the abnormal risk zone to upload temperature data and pressure data.

2. The low-power wireless communication method for temperature and pressure monitoring according to claim 1, characterized in that, Based on the historical operational data, historical abnormal events are classified to obtain an abnormal sample set, specifically including: Based on the historical temperature data, the severity of the anomaly at each monitoring point at each historical moment is determined; wherein, the severity of the anomaly is used to characterize the extent to which the temperature deviates from the normal level; Based on the historical pressure data, a pressure variation index is determined for each monitoring point; wherein, the pressure variation index is used to characterize the degree of abnormal disturbance of upstream pressure fluctuation at each historical moment; Based on the severity of the anomaly and the pressure variation index, an anomaly differentiation index is determined for each historical anomaly event; wherein, the anomaly differentiation index is used to characterize the probability that the historical anomaly event belongs to a blockage anomaly. Based on the anomaly differentiation index, cluster analysis is performed on all historical anomaly events to obtain the blockage anomaly sample subset and the leakage anomaly sample subset.

3. The low-power wireless communication method for temperature and pressure monitoring according to claim 2, characterized in that, Based on the historical pressure data, the pressure variation index for each monitoring point is determined, specifically including: For each monitoring point, acquire historical pressure data from a predetermined number of adjacent monitoring points upstream of the monitoring point; The pressure variation index is determined based on the historical pressure data of the preset number of adjacent monitoring points, the historical pressure data of the monitoring points, and the distance between each adjacent monitoring point and the monitoring point.

4. The low-power wireless communication method for temperature and pressure monitoring according to claim 2, characterized in that, Based on the severity of the anomaly and the pressure fluctuation index, anomaly differentiation indicators are determined for each historical anomaly event, specifically including: Based on the historical operational data corresponding to all historical abnormal events, analyze the correlation characteristics between the pressure variation index and the severity of the abnormality; For each historical abnormal event, the pressure variation index of the historical abnormal event is compensated and corrected according to the correlation characteristics to obtain a compensation value; The anomaly differentiation index is determined based on the compensation value and the pressure variation index of the historical anomaly events.

5. The low-power wireless communication method for temperature and pressure monitoring according to claim 1, characterized in that, Based on the real-time pressure data and the set of abnormal samples, predict abnormal risk sections in the thermal pipeline, specifically including: For each boundary monitoring point, the pressure variation index of the boundary monitoring point is determined based on the real-time pressure data of the boundary monitoring point. If the pressure fluctuation index of the boundary monitoring point is greater than a first preset threshold, the pressure fluctuation index of the boundary monitoring point is matched with the blockage anomaly sample subset to determine the highest blockage matching degree of the boundary monitoring point; wherein, the highest blockage matching degree is the highest value of the blockage matching degree between the boundary monitoring point and each historical blockage sample in the blockage anomaly sample subset. If the highest congestion matching degree at the boundary monitoring point is greater than the second preset threshold, the anomaly type is determined to be congestion, and the anomaly risk segment is determined based on the location information of the historical congestion sample record corresponding to the highest congestion matching degree.

6. The low-power wireless communication method for temperature and pressure monitoring according to claim 5, characterized in that, The method further includes: If the highest blockage matching degree of the boundary monitoring point is less than or equal to the second preset threshold, the pressure change index of the boundary monitoring point is matched with each historical leak sample in the subset of leak abnormal samples to determine multiple leak similarities; wherein, a leak similarity is used to characterize the degree of similarity between the boundary monitoring point and a historical leak sample; Based on the multiple leak similarities and the leak point locations in the historical leak samples corresponding to the multiple leak similarities, a leak reference distance is determined; Starting from the boundary monitoring point, the location point along the upstream direction of the heating pipeline that is a distance from the reference leakage distance from the starting point is determined as the predicted leakage point. The pipeline section where the predicted leakage point is located, defined by two adjacent boundary monitoring points, is predicted as the abnormal risk section.

7. The low-power wireless communication method for temperature and pressure monitoring according to claim 1, characterized in that, Sending a wake-up command to at least one of the interval monitoring points in the abnormal risk segment specifically includes: At least one target interval monitoring point located within the abnormal risk zone is identified as the core wake-up point; At least one monitoring point adjacent to the core wake-up point on the thermal pipeline topology is identified as a collaborative wake-up point; Generate and send the wake-up command containing the core wake-up point address and the cooperative wake-up point address.

8. The low-power wireless communication method for temperature and pressure monitoring according to any one of claims 1-7, characterized in that, Before acquiring real-time pressure data from the boundary monitoring points, the method further includes: The boundary monitoring points are configured to collect and upload temperature and pressure data at a first preset frequency.

9. The low-power wireless communication method for temperature and pressure monitoring according to claim 8, characterized in that, Before acquiring real-time pressure data from the boundary monitoring points, the method further includes: The interval monitoring point is configured to collect and upload temperature and pressure data at a second preset frequency lower than the first preset frequency. During non-upload periods, the main communication module of the interval monitoring point is controlled to enter a sleep state, while the wake-up receiving module of the interval monitoring point is kept in working state.

10. A low-power wireless communication module for temperature and pressure monitoring, characterized in that, The low-power wireless communication module for temperature and pressure monitoring includes: a data acquisition module, an abnormal sample classification module, an abnormal risk prediction module, and a wake-up command sending module. The data acquisition module is used to acquire historical operating data of the heating pipeline; wherein, the historical operating data includes historical temperature data and historical pressure data of multiple monitoring points on the heating pipeline, and the types of the multiple monitoring points include boundary monitoring points and interval monitoring points, with multiple interval monitoring points set between each pair of adjacent boundary monitoring points; The abnormal sample classification module is used to classify historical abnormal events according to the historical operation data to obtain an abnormal sample set; wherein, the abnormal sample set includes a blockage abnormal sample subset and a leakage abnormal sample subset; The data acquisition module is also used to acquire real-time pressure data of the boundary monitoring point; The abnormal risk prediction module is used to predict abnormal risk sections in the thermal pipeline based on the real-time pressure data and the abnormal sample set; wherein, the abnormal risk section is located between two adjacent boundary monitoring points; The wake-up command sending module is used to send a wake-up command to at least one of the interval monitoring points in the abnormal risk section; wherein, the wake-up command is used to instruct at least one of the interval monitoring points in the abnormal risk section to upload temperature data and pressure data.

Images