Fire-fighting pipe network pressure dynamic monitoring method and system combined with internet of things, and medium

Abstract

The application discloses a fire-fighting pipe network pressure dynamic monitoring method and system combined with the Internet of Things, and a medium, relates to the related technical field of pressure monitoring, and the method comprises the steps of: deploying a pressure sensor and an auxiliary sensor on a fire-fighting pipe network monitoring node, establishing a topology sensing structure by using wireless Internet of Things communication; after adaptive sampling adjustment, a node sampling data set is established; the multi-time window pressure evolution model is executed to perform sliding differential analysis, and an abnormal trend index is calculated; a neighborhood broadcast request is configured and broadcast consistency verification is performed; abnormal trend index enhancement correction is performed, and group resonance detection is performed to establish linkage and cooperative early warning; and a fire-fighting pipe network pressure abnormal early warning signal is reported. The technical problems that real-time sensing of dynamic abnormal data of the fire-fighting pipe network is difficult, cooperation is lacking, and response is lagging in the prior art, resulting in poor precision and reliability of pressure monitoring, are solved, real-time and accurate dynamic monitoring is achieved, and the technical effects of improving the accuracy and reliability of fire-fighting pipe network pressure monitoring are achieved.

Description

Technical Field

[0001] This invention relates to the field of pressure monitoring technology, specifically to a method, system, and medium for dynamic pressure monitoring of fire protection pipelines combined with the Internet of Things. Background Technology

[0002] Fire protection piping networks are the infrastructure of fire protection systems, and their stable operation is directly related to public safety. Traditional fire protection piping network pressure monitoring suffers from problems such as delayed response, limited coverage, and inability to detect dynamic anomalies in real time. Especially in large and complex piping networks, due to the numerous and widely distributed nodes, local pressure anomalies are often difficult to detect in a timely manner, which may lead to decreased system efficiency or even failure. Furthermore, fire protection piping networks are often affected by environmental factors such as temperature changes and flow fluctuations, further increasing the uncertainty of pressure status. With the development of the Internet of Things (IoT), real-time acquisition and transmission of piping network data can be achieved by deploying multiple sensors such as pressure, flow, and temperature sensors combined with wireless communication. However, in practical applications, monitoring nodes often suffer from mismatched data sampling frequencies, unbalanced communication loads, and insufficient anomaly identification accuracy. Moreover, pressure data from a single node cannot accurately reflect the overall operating status of the piping network, thus affecting the accuracy and real-time performance of fire protection piping network pressure anomaly detection.

[0003] Therefore, current technologies suffer from technical problems such as difficulty in real-time perception of dynamic abnormal data in fire protection pipe networks, lack of coordination, and delayed response, resulting in poor accuracy and reliability of pressure monitoring. Summary of the Invention

[0004] This application provides a method, system, and medium for dynamic monitoring of fire pipeline pressure that integrates the Internet of Things. This solves the technical problems in the prior art, such as the difficulty in real-time perception of dynamic abnormal data of fire pipelines, lack of coordination, and delayed response, which lead to poor accuracy and reliability of pressure monitoring. It achieves real-time and accurate dynamic monitoring and improves the accuracy and reliability of fire pipeline pressure monitoring.

[0005] This application provides a method for dynamic monitoring of fire protection pipeline network pressure using the Internet of Things (IoT). The method includes: deploying pressure sensors and auxiliary sensors, including flow sensors and temperature sensors, at each monitoring node of the fire protection pipeline network; establishing a topology sensing structure of the fire protection pipeline network using wireless IoT communication of the monitoring nodes; establishing a node sampling dataset after adaptive sampling adjustment of the monitoring nodes based on the topology sensing structure; inputting the node sampling dataset into a multi-time-window pressure evolution model mapped to the monitoring nodes, performing sliding differential analysis, and calculating abnormal trend indicators; configuring neighborhood broadcast requests based on the abnormal trend indicators, performing broadcast consistency verification based on the topology sensing structure, and generating a consistency factor; enhancing and correcting the abnormal trend indicators using the consistency factor, establishing a coordinated early warning system based on the enhancement correction results and the topology sensing structure using group resonance detection; and issuing an abnormal pressure warning signal for the fire protection pipeline network based on the enhancement correction results and the coordinated early warning system.

[0006] In one possible implementation, the method for dynamic monitoring of fire pipeline pressure combined with the Internet of Things further performs the following processing: after receiving the corresponding node sampling dataset using a multi-time-window pressure evolution model, time-weighted layering of the sampling sequence is performed, and a pressure evolution sub-model is configured according to the time-weighted layering results. The pressure evolution sub-model is configured with time window weights based on the length of the time window, the intensity of data fluctuations, and the topology-aware connectivity. The configured pressure evolution sub-model is used to perform sliding differential analysis under the corresponding time window to extract the first-order trend term and the second-order acceleration term of pressure change, and to calculate the auxiliary coupling response of the auxiliary sensor. The outputs of all pressure evolution sub-models are weighted and fused using the time window weights to calculate abnormal trend indicators.

[0007] In one possible implementation, the method for dynamic monitoring of fire pipeline pressure combined with the Internet of Things further performs the following processing: performing nonlinear cross-correlation analysis within nodes on the output of each pressure evolution sub-model to establish the perturbation coupling degree within nodes; performing dynamic correction of time window weights based on the perturbation coupling degree within nodes, and then performing weighted fusion of the outputs of all pressure evolution sub-models to extract the consistent principal component; and using the fluctuation amplitude of the consistent principal component to construct an abnormal trend index.

[0008] In one possible implementation, the method for dynamic monitoring of fire pipeline pressure combined with the Internet of Things further performs the following processing: evaluating the degree of anomaly using the abnormal trend index and establishing a first broadcast influence factor; obtaining the node importance of the monitoring node corresponding to the abnormal trend index and establishing a second broadcast influence factor based on the node importance, wherein the node importance is calculated through the node key value and node complexity; configuring a neighborhood broadcast request using the first broadcast influence factor and the second broadcast influence factor, and performing neighborhood broadcast processing.

[0009] In one possible implementation, the method for dynamic monitoring of fire pipeline pressure combined with the Internet of Things further performs the following processing: identifying the neighborhood propagation path based on the neighborhood broadcast request and the topology sensing structure, and configuring a consistency verification request for the abnormal trend index along the corresponding propagation path; after performing path adaptability verification based on the propagation path on the consistency verification request, outputting a consistency factor, wherein the path adaptability verification includes path distance attenuation consistency verification and path time sequence propagation delay verification.

[0010] In one possible implementation, the method for dynamic monitoring of fire pipeline pressure combined with the Internet of Things further performs the following processing: establishing topology weight coefficients based on the topology sensing structure; locating abnormal nodes using the enhanced correction results, performing abnormal node resonance detection based on the topology weight coefficients through the swarm resonance detection channel to generate a swarm resonance index; and matching early warning signals based on the swarm resonance index and the enhanced correction results to establish a collaborative early warning system.

[0011] In one possible implementation, the IoT-integrated dynamic monitoring method for fire-fighting pipeline pressure further performs the following processing: the group resonance detection channel is as follows: ; in, Characterizing the population resonance index, The total number of monitoring nodes, For monitoring nodes The set of node pairs, i.e., with monitoring nodes The set of node pairs consisting of monitoring nodes that satisfy the adjacency relationship. Characterization monitoring nodes At any moment Pressure change Characterization monitoring nodes At any moment Pressure change Characterization by monitoring nodes propagation to monitoring nodes The propagation delay represents the weight coefficient of the node pair.

[0012] In one possible implementation, the method for dynamic monitoring of fire pipeline pressure combined with the Internet of Things further performs the following processing: configuring an early warning response strategy based on the early warning signal of abnormal fire pipeline pressure, and establishing an execution detection cycle for the early warning response strategy; using the execution detection cycle to evaluate the timeliness of the early warning response strategy and establish timeliness compensation; and issuing additional early warnings based on the timeliness compensation.

[0013] This application also provides a dynamic monitoring system for fire protection pipeline network pressure combined with the Internet of Things (IoT). The system includes: a sensing structure establishment module, used to deploy pressure sensors and auxiliary sensors at each monitoring node of the fire protection pipeline network, wherein the auxiliary sensors include flow sensors and temperature sensors, and to establish a topological sensing structure of the fire protection pipeline network using wireless IoT communication of the monitoring nodes; a sampling data acquisition module, used to establish a node sampling dataset after adaptive sampling adjustment of the monitoring nodes based on the topological sensing structure; a data analysis module, used to input the node sampling dataset into a multi-time-window pressure evolution model mapped to the monitoring nodes, perform sliding differential analysis, and calculate abnormal trend indicators; a consistency verification module, used to configure a neighborhood broadcast request according to the abnormal trend indicators, and then perform broadcast consistency verification based on the topological sensing structure to generate a consistency factor; a collaborative early warning module, used to enhance and correct the abnormal trend indicators using the consistency factor, and then establish a coordinated early warning system based on the enhancement correction result and the topological sensing structure through group resonance detection; and an early warning signal reporting module, used to issue an abnormal pressure early warning signal for the fire protection pipeline network based on the enhancement correction result and the coordinated early warning system.

[0014] This application also provides a computer-readable storage medium, including: a computer program stored thereon, which, when executed by a processor, implements a method for dynamic monitoring of fire-fighting pipeline pressure in conjunction with the Internet of Things.

[0015] This application proposes a method, system, and medium for dynamic monitoring of fire-fighting pipeline pressure using an Internet of Things (IoT) approach. Pressure sensors and auxiliary sensors are deployed at monitoring nodes in the fire-fighting pipeline network. A topology sensing structure is established using wireless IoT communication. After adaptive sampling adjustment, a node sampling dataset is created. This dataset is then input to a multi-time-window pressure evolution model for sliding differential analysis to calculate abnormal trend indicators. Neighborhood broadcast requests are configured, and broadcast consistency verification is performed. Abnormal trend indicators are enhanced and corrected, and group resonance detection is used to establish a coordinated early warning system. Finally, an abnormal pressure warning signal for the fire-fighting pipeline network is issued. This addresses the technical problems in existing technologies, such as difficulty in real-time sensing of dynamic abnormal data in the fire-fighting pipeline network, lack of coordination, and delayed response, which lead to poor accuracy and reliability of pressure monitoring. The proposed method achieves real-time and accurate dynamic monitoring, improving the accuracy and reliability of fire-fighting pipeline network pressure monitoring. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 A schematic diagram of the process for dynamic monitoring of fire protection pipeline pressure using the Internet of Things (IoT) provided in this application embodiment.

[0018] Figure 2 A schematic diagram of the structure of a fire-fighting pipeline pressure dynamic monitoring system combined with the Internet of Things provided in this application embodiment.

[0019] Explanation of reference numerals in the attached diagram: Perception structure establishment module 10, sampling data acquisition module 20, data analysis module 30, consistency verification module 40, collaborative early warning module 50, and early warning signal reporting module 60. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] This application provides a method for dynamic monitoring of fire protection pipeline pressure combined with the Internet of Things, such as... Figure 1 As shown, the method includes: Step S100: In each monitoring node of the fire protection pipeline network, pressure sensors and auxiliary sensors are deployed respectively. The auxiliary sensors include flow sensors and temperature sensors. The topology sensing structure of the fire protection pipeline network is established by using the wireless Internet of Things communication of the monitoring nodes.

[0022] Preferably, multiple key physical locations in the fire protection network, such as fire hydrants, pump connections, and main pipeline branch points, are identified. Sensing devices are deployed at each key physical location as a monitoring node, including pressure sensors and auxiliary sensors. The auxiliary sensors include flow sensors and temperature sensors. Specifically, pressure sensors directly measure the pressure of the water flow at the monitoring node to determine if the network pressure is normal. Auxiliary sensors provide environmental information to aid in the analysis and interpretation of pressure data. Flow sensors measure the rate or volume of water flow at the node. When pressure changes, flow data can be used to determine whether the pressure drop is due to increased flow from fire sprinkler activation or abnormal flow accompanied by pressure anomalies caused by pipeline leaks or blockages. Temperature sensors measure the ambient temperature at the node to provide more accurate compensation and calibration for the pressure sensors. Connecting the sensors at the monitoring nodes using wireless IoT communication allows for the transmission and sharing of collected pressure, flow, and temperature data via a wireless network. This also clarifies the location and logical connections between each monitoring node and network components, thereby constructing a topological sensing structure for the fire protection network to accurately reflect the actual connection relationships and spatial structure of the physical fire protection network.

[0023] Step S200: After adaptive sampling adjustment of the monitoring nodes based on the topology sensing structure, a node sampling dataset is established.

[0024] Preferably, the topology sensing structure is analyzed to determine the location, status, and neighboring node behavior of each monitoring node in the pipeline topology, and the data sampling rate is dynamically adjusted. Specifically, adjustments may be based on the importance of the monitoring node; high-importance nodes located on the pipeline backbone, critical branches, or near potential risk sources are assigned a higher base sampling frequency, while peripheral or non-critical nodes use a lower sampling frequency to save resources. Adjustments may be based on event triggers; when an abnormal trend is detected in a certain area through threshold judgment or a broadcast request received from a neighboring node, the sampling frequency of that node and related nodes in its topology neighborhood is temporarily and significantly increased. Adjustments may be based on flow status; when the flow sensor detects zero or extremely low flow, it automatically switches to a low-frequency monitoring mode; when the flow starts or increases, indicating fire-fighting action or leakage, it triggers a switch to a high-frequency sampling mode. Finally, data is collected according to the dynamically set sampling frequency, recording the pressure, flow, and temperature data of each monitoring node at the same or related timestamps, and identifying its specific node location in the topology sensing structure, thus combining them to form a high-quality node sampling dataset.

[0025] Step S300: Input the node sampling dataset into the multi-time-window pressure evolution model mapped to the monitoring node, perform sliding differential analysis, and calculate the abnormal trend index.

[0026] Step S300 further includes step S310, after receiving the corresponding node sampling dataset using the multi-time-window pressure evolution model, performing time-weighted layering of the sampling sequence, and configuring a pressure evolution sub-model based on the time-weighted layering results. The pressure evolution sub-model is configured with time window weights based on the length of the time window, the intensity of data fluctuations, and the topology-aware connectivity. Step S320, using the configured pressure evolution sub-model, performing sliding differential analysis under the corresponding time window, extracting the first-order trend term and the second-order acceleration term of pressure change, and calculating the auxiliary coupling response of the auxiliary sensor. Step S330, using the time window weights, weightedly fusing the outputs of all pressure evolution sub-models to calculate abnormal trend indicators.

[0027] Preferably, the node sampling datasets are input into a multi-time-window pressure evolution model mapped to the monitoring nodes for anomaly detection. This model uses multiple windows of different time lengths in parallel to analyze the same pressure data segment and fuses the analysis results from different windows to form a comprehensive anomaly detection. Specifically, the pressure, flow, and temperature data sequences of the corresponding monitoring nodes are input into the multi-time-window pressure evolution model, and time-weighted stratification of the sampling sequences is performed. This involves defining multiple time windows of different lengths at the current time point. For example, short windows are used to capture instantaneous and drastic changes, medium windows to capture short-term trends, and long windows to capture long-term and slow drifts, thus stratifying the time series. Within each time window, the closer to the current time... The higher the weight of the data points from the previous time step, the stronger their indicative power of the current state. Then, based on the time-weighted stratification results, an analysis model is created for each time window to obtain a stress evolution sub-model. Each stress evolution sub-model is configured with time window weights representing importance based on the length of the time window, the intensity of data fluctuations, and the topology-aware connectivity. The length of the time window is the basic weight. The intensity of data fluctuations refers to the data variance within the time window, which indicates that events occur frequently during that period. Therefore, the weight of the stress evolution sub-model for that window is increased. If the topology-aware connectivity is high, that is, the node is in a critical position in the topology or its neighboring nodes report anomalies, the weight of the stress evolution sub-model for its short time window is increased to more sensitively capture anomalies that may spread rapidly.

[0028] Preferably, each pressure evolution sub-model performs sliding differential analysis within its corresponding time window. This involves sliding a smaller window across the data sequence and calculating its derivative to extract the first-order trend term of pressure change (the first derivative of pressure over time), representing the upward / downward trend and rate of increase / decrease. It also extracts the second-order acceleration term (the second derivative of pressure), representing the trend of pressure change as acceleration or deceleration. Simultaneously, the auxiliary coupling response of the auxiliary sensor is calculated within the same time window, analyzing the correlation between the auxiliary sensor data and the pressure data. For example, does the flow rate increase significantly and synchronously when the pressure decreases? Or does the flow rate remain unchanged when the pressure decreases? The system detects whether the pressure changes or even decreases, or whether the temperature drops sharply, and then quantifies the auxiliary coupling response index to determine whether the current pressure change pattern can be represented by normal operating conditions or known disturbances. Finally, the outputs of all pressure evolution sub-models are weighted and fused using time window weights, and the output comprehensive value is corrected in conjunction with the auxiliary coupling response. If the trend and acceleration of the pressure change pattern are significant and cannot be explained by the data from the auxiliary sensors, that is, the coupling response is very low, it indicates that it is not a normal water use or other operating conditions, and it is judged as a highly abnormal trend. Finally, it is converted into a quantified abnormal trend index. The higher the index, the greater the possibility that the node is in an abnormal state.

[0029] Furthermore, step S330 also includes step S331, performing nonlinear cross-correlation analysis within nodes on the output of each pressure evolution sub-model to establish the perturbation coupling degree within nodes; step S332, performing dynamic correction of time window weights based on the perturbation coupling degree within nodes, and then performing weighted fusion of the outputs of all pressure evolution sub-models to extract the consistency principal component; step S333, using the fluctuation amplitude of the consistency principal component to construct an abnormal trend index.

[0030] Preferably, nonlinear cross-correlation analysis is performed by calculating the output of each pressure evolution sub-model using mutual information or nonlinear correlation coefficients. This quantifies the dynamic temporal correlation strength between pressure trends and flow trends, and pressure trends and temperature changes, within the time window corresponding to the current pressure evolution sub-model. For example, a small leak may cause a slow decrease in pressure while the flow rate changes little. This establishes the intra-node perturbation coupling degree, i.e., the degree of matching between the abnormal trend of pressure data and the perturbation of auxiliary sensors. A high intra-node perturbation coupling degree indicates that the abnormal pressure change is supported by a coordinated change in flow rate or temperature. For example, a decrease in pressure accompanied by an increase in flow rate is consistent with water usage characteristics, indicating that an interpretable change in operating conditions has been detected. A low intra-node perturbation coupling degree indicates that the pressure has changed abnormally, but the flow rate and temperature have not shown the expected coordinated change. For example, a continuous decrease in pressure but no change in flow rate indicates that anomalies such as pipeline leaks, blockages, or sensor malfunctions have been detected.

[0031] Preferably, the reliability of the model analysis results is inversely proportional to its perturbation coupling degree. The perturbation coupling degree is used as input to correct the time window weights configured for short, medium, and long-term time window sub-models in real time. For example, a low perturbation coupling degree indicates the capture of suspicious signals, and the time window weight of that pressure evolution sub-model should be increased; conversely, a high perturbation coupling degree indicates that the captured changes are normal operating conditions, and the time window weight of that pressure evolution sub-model should be decreased. Then, the corrected time window weights are used to perform weighted fusion of the trend terms, acceleration terms, etc., output by all pressure evolution sub-models to obtain a comprehensive analysis. From the mixed signal output by the pressure evolution sub-model, a common trend that is consistent and stable across all time scales is extracted as the consistency principal component, representing the core signal of pressure anomalies. Finally, the fluctuation amplitude of the consistency principal component is mapped to anomaly trend indicators, such as amplitude, variance, or energy. If the amplitude is low and stable, it indicates that the pipeline network is in normal condition or the change is explainable. If the amplitude increases significantly, it indicates the existence of an abnormal trend that is consistent across multiple scales and is not in normal operating conditions. This greatly enhances the ability to identify leakage faults while effectively suppressing false alarms about changes in normal operating conditions.

[0032] Step S400: After configuring the neighborhood broadcast request according to the abnormal trend index, broadcast consistency verification is performed according to the topology-aware structure to generate a consistency factor.

[0033] Step S400 further includes step S410, using the abnormal trend index to evaluate the degree of abnormality and establish a first broadcast influence factor; step S420, obtaining the node importance of the monitoring node corresponding to the abnormal trend index, and establishing a second broadcast influence factor based on the node importance, wherein the node importance is calculated through the node key value and node complexity; step S430, using the first broadcast influence factor and the second broadcast influence factor to configure a neighborhood broadcast request and perform neighborhood broadcast processing.

[0034] Preferably, an anomaly degree evaluation is performed on the abnormal trend indicators, that is, the continuous abnormal trend indicator values ​​are mapped to discrete anomaly degree levels. For example, 0~0.3 is normal fluctuation and can be ignored; 0.3~0.7 is slightly abnormal and needs attention; 0.7~1.0 is severely abnormal and needs to be dealt with immediately. Then, the anomaly degree evaluation results are converted into a first broadcast impact factor to quantify the severity of the anomaly perceived by the current node. The higher the value of the first broadcast impact factor, the more severe the anomaly, and the stronger the necessity to initiate a broadcast request and seek verification from neighbors. In fire protection pipe networks, nodes at different locations have varying degrees of criticality within the entire network. The importance of monitoring nodes corresponding to abnormal trend indicators is calculated. Specifically, network centrality indicators are calculated based on the topology-aware structure to determine node critical values. Simultaneously, node complexity, describing the monitoring difficulty or fault risk of the node's environment, is determined. The importance of monitoring nodes is then calculated using both the node critical value and node complexity; higher critical values ​​and complexity indicate higher node importance. A second broadcast influence factor is then established through standardization to quantify the importance of the current node's location. A higher second broadcast influence factor value indicates a more severe consequence of a real fault at that node, thus prioritizing its anomaly verification. Finally, the first and second broadcast influence factors are multiplied to generate a broadcast decision score. A broadcast is triggered if either the first or second broadcast influence factor is low. The broadcast decision score, along with the current node ID, abnormal trend indicators, and timestamp, is encapsulated into a neighborhood broadcast request. This request is sent via a wireless IoT communication module and responded to by rain nodes in the topology-aware structure, significantly reducing communication overhead and false alarm interference while ensuring rapid response to real high-risk events.

[0035] Furthermore, step S400 also includes step S440, identifying the neighborhood propagation path based on the neighborhood broadcast request and the topology-aware structure, and configuring a consistency verification request for the abnormal trend index along the corresponding propagation path; step S450, after performing path adaptability verification based on the propagation path on the consistency verification request, outputting a consistency factor, wherein the path adaptability verification includes path distance attenuation consistency verification and propagation delay verification of path timing.

[0036] Preferably, neighborhood propagation path identification is performed based on neighborhood broadcast requests and topology-aware structures. Specifically, for each node, the physical pipeline path from its origin to each direct neighbor node is identified. Then, a consistency verification request for anomaly indicators is configured for each identified propagation path. This consistency verification request includes the node's anomaly information and the expected verification parameters for that propagation path. Next, path adaptability verification is performed on the consistency verification requests based on the propagation path, mainly including path distance attenuation consistency verification and path timing propagation delay verification. Specifically, in a real fire protection pipe network, when a pressure disturbance propagates from the source node to a neighbor node, its intensity will vary with the propagation path. Pressure waves attenuate due to propagation distance and pipe friction resistance. An attenuation model is then built based on known pipe material, diameter, and length to predict pressure attenuation. When neighboring nodes report observed pressure changes, it is checked whether these changes fall within the predicted reasonable attenuation range. Since pressure waves propagate at a speed in the fluid, there is a time delay between the source node and neighboring nodes sensing anomalies. The theoretical propagation time from the source node to the neighboring node is calculated based on pipe material and structure. When a neighboring node reports anomalies, the timestamps of the two node data are precisely compared to check if the anomaly start time of the neighboring node is approximately the sum of the anomaly start time and propagation delay of the source node. Finally, by combining the consistency verification results of path distance attenuation and path time sequence propagation delay, a consistency factor is output to quantify the degree of conformity between the neighboring node's response and physical expectations. This improves the ability to distinguish between real pipeline network faults, sensor faults, and data noise, achieving high-reliability monitoring and early warning for fire protection pipeline networks.

[0037] Step S500: After using the consistency factor to enhance and correct the abnormal trend indicators, establish a coordinated early warning system based on the enhancement correction results and the topological sensing structure to perform group resonance detection.

[0038] Preferably, the consistency factor is used as a confidence coefficient to enhance the correction of the abnormal trend index. A high consistency factor indicates that multiple neighboring nodes have verified the anomaly, thus significantly increasing the abnormal trend index value of that node. Conversely, a low consistency factor indicates that neighboring nodes have not perceived the anomaly, thus significantly reducing the abnormal trend index value of that node. This results in a more reliable node-level anomaly score and an enhanced correction result. Then, the enhanced correction result and the topology sensing structure are combined for swarm resonance detection. This involves detecting whether there are correlated and coordinated abnormal behaviors among multiple nodes, analyzing the differences in pressure changes and propagation relationships between adjacent node pairs as a swarm resonance index. For example, if three consecutive nodes on a pipe path report high-level corrected anomalies, and the anomalies are continuous in time and space, then the enhanced correction result of a single node and the swarm resonance index are combined for judgment. If only one or a few nodes have high corrected indicators but a low swarm resonance index, a local maintenance warning may be generated. If multiple nodes simultaneously have high corrected indicators and a high swarm resonance index, a higher-level coordinated warning is generated to ensure accurate and efficient emergency response.

[0039] Furthermore, step S500 also includes step S510, establishing topological weight coefficients based on the topological sensing structure; step S520, after locating abnormal nodes using the enhanced correction results, performing abnormal node resonance detection based on the topological weight coefficients through the swarm resonance detection channel to generate a swarm resonance index; step S530, matching early warning signals based on the swarm resonance index and the enhanced correction results to establish a collaborative early warning.

[0040] Preferably, a topology weight coefficient is assigned to each pair of adjacent nodes or connecting paths in the fire protection pipe network topology sensing structure based on factors such as pipe capacity and network centrality. The larger the pipe diameter connecting two nodes, the higher the weight coefficient; the path connecting key backbone nodes has an even higher weight coefficient. The topology weight coefficient is used to amplify abnormal propagation signals on important paths while suppressing signals on secondary paths. Next, based on the enhancement correction results, all nodes exceeding a predetermined threshold are filtered and marked as abnormal nodes. Then, all located abnormal nodes are input into the group resonance detection channel to perform resonance detection based on the topology weight coefficient, i.e., traversing all adjacent abnormal node pairs and calculating their propagation characteristics. The pressure change differences after the broadcast delay are weighted and normalized using topological weighting coefficients. Then, the mean of all abnormal nodes in the fire protection network is calculated to obtain the group resonance index, which is used to quantify the synergy of the anomalies. The higher the group resonance index, the worse the synergy. Finally, the early warning signal is matched based on the group resonance index and the enhancement correction result. For example, if a few nodes have high enhancement correction results and a high group resonance index, a local leakage / blockage early warning is triggered and the node location is marked. If multiple nodes have high enhancement correction results and a low group resonance index, a coordinated early warning is triggered, which includes the core abnormal area, the potential impact range, and the fault type, thereby ensuring predictive maintenance management.

[0041] Furthermore, step S520 also includes the following: the swarm resonance detection channel is as follows: ; in, Characterizing the population resonance index, The total number of monitoring nodes, For monitoring nodes The set of node pairs, i.e., with monitoring nodes The set of node pairs consisting of monitoring nodes that satisfy the adjacency relationship. Characterization monitoring nodes At any moment Pressure change Characterization monitoring nodes At any moment Pressure change Characterization by monitoring nodes propagation to monitoring nodes The propagation delay represents the weight coefficient of the node pair.

[0042] Step S600: Based on the enhanced correction results, a fire protection network pressure abnormality warning signal is issued through the linkage and collaborative early warning system.

[0043] Preferably, based on the enhanced correction results and the coordinated early warning system, a fire protection network pressure anomaly warning signal is issued. Specifically, a high enhanced correction result and a low group resonance index jointly trigger an emergency warning, indicating a serious fault affecting the entire network or a large area, such as a main pipe rupture or a core pump station failure; a high enhanced correction result and a high group resonance index trigger an important warning, indicating a serious anomaly at a local node or in a small area, such as a branch pipe leak or a local blockage; anomalies with low corrected indicators trigger a attention warning, prompting maintenance personnel to pay attention but not requiring immediate emergency response; at the same time, the pressure anomaly warning signal clearly indicates the anomaly type and precise location information, such as a leak warning, a blockage warning, or a pressure deficiency warning, and marks the coordinates of the anomaly node, the core area of ​​the anomaly, and the scope of impact, etc., and provides audible and visual alarms through the monitoring center's large screen, thereby shortening the time from anomaly detection to taking countermeasures and improving the safety and reliability of the entire fire protection network.

[0044] Furthermore, step S600 also includes step S610, configuring an early warning response strategy based on the fire pipeline network pressure abnormality early warning signal, and establishing an execution detection cycle for the early warning response strategy; step S620, using the execution detection cycle to evaluate the timeliness of the early warning response strategy and establish timeliness compensation; step S630, executing additional early warnings based on the timeliness compensation.

[0045] Preferably, the warning response strategy is configured based on the warning level, anomaly type, and location of the fire protection network pressure anomaly warning signal. This means matching a corresponding warning response strategy to different pressure anomaly warning signals, setting timers and checkpoints for key nodes of each strategy, and determining the execution detection cycle. Then, at the end of each execution detection cycle, the timeliness of the warning response strategy is evaluated, including checking whether the predetermined response actions have been completed. If the evaluation result indicates that the response was not completed on time, a quantifiable urgency increment is generated as timeliness compensation to quantify the potential risks added due to response delays. Finally, additional warnings are issued, that is, the timeliness compensation is superimposed on the pressure anomaly warning signal to upgrade the warning level, expand the warning scope, or change the notification channel, ensuring that each warning signal is responded to in a timely manner, thereby improving the accuracy of fire protection network pressure monitoring and the safety and reliability of its operation.

[0046] In the above text, refer to Figure 1 A method for dynamic monitoring of fire protection pipeline pressure incorporating the Internet of Things (IoT) according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2 A dynamic monitoring system for fire protection pipeline pressure incorporating the Internet of Things (IoT) is described according to an embodiment of the present invention.

[0047] The fire protection pipeline pressure dynamic monitoring system based on the Internet of Things (IoT) according to embodiments of the present invention addresses the technical problems in existing technologies, such as difficulty in real-time sensing of dynamic anomalies in fire protection pipelines, lack of coordination, and delayed response, leading to poor accuracy and reliability of pressure monitoring. It achieves real-time and accurate dynamic monitoring, thus improving the accuracy and reliability of fire protection pipeline pressure monitoring. Figure 2 As shown, the fire protection pipeline pressure dynamic monitoring system combined with the Internet of Things includes: a sensing structure establishment module 10, a sampling data acquisition module 20, a data analysis module 30, a consistency verification module 40, a collaborative early warning module 50, and an early warning signal reporting module 60.

[0048] The sensing structure establishment module 10 is used to deploy pressure sensors and auxiliary sensors in each monitoring node of the fire protection pipeline network, wherein the auxiliary sensors include flow sensors and temperature sensors, and establish a topological sensing structure of the fire protection pipeline network using wireless IoT communication of the monitoring nodes; the sampling data acquisition module 20 is used to establish a node sampling dataset after adaptive sampling adjustment of the monitoring nodes based on the topological sensing structure; the data analysis module 30 is used to input the node sampling dataset into the multi-time-window pressure evolution model mapped to the monitoring nodes, perform sliding differential analysis, and calculate abnormal trend indicators; the consistency verification module 40 is used to configure neighborhood broadcast requests according to the abnormal trend indicators, perform broadcast consistency verification according to the topological sensing structure, and generate a consistency factor; the collaborative early warning module 50 is used to establish a linkage collaborative early warning based on the enhanced correction results and the topological sensing structure after enhancing and correcting the abnormal trend indicators; the early warning signal reporting module 60 is used to issue an abnormal pressure early warning signal of the fire protection pipeline network according to the enhanced correction results and the linkage collaborative early warning.

[0049] The specific configuration of the data analysis module 30 will be described in detail below. The data analysis module 30 further includes: receiving the corresponding node sampling dataset using a multi-time-window pressure evolution model; performing time-weighted stratification of the sampling sequence; configuring a pressure evolution sub-model based on the time-weighted stratification results; the pressure evolution sub-model being configured with time window weights based on the time window length, data fluctuation intensity, and topology-aware connectivity; performing sliding differential analysis under the corresponding time window using the configured pressure evolution sub-model to extract the first-order trend term and second-order acceleration term of pressure change, and calculating the auxiliary coupling response of the auxiliary sensor; and weighting and fusing the outputs of all pressure evolution sub-models using the time window weights to calculate abnormal trend indicators.

[0050] The specific configuration of the data analysis module 30 will be described in detail below. The data analysis module 30 further includes: performing intra-node nonlinear cross-correlation analysis on the output of each pressure evolution sub-model to establish intra-node perturbation coupling degree; dynamically correcting the time window weights based on the intra-node perturbation coupling degree, then performing weighted fusion of the outputs of all pressure evolution sub-models to extract the consistency principal component; and constructing anomaly trend indicators using the fluctuation amplitude of the consistency principal component.

[0051] The specific configuration of the consistency verification module 40 will be described in detail below. The consistency verification module 40 further includes: evaluating the degree of anomaly using the anomaly trend index and establishing a first broadcast influence factor; obtaining the node importance of the monitoring node corresponding to the anomaly trend index and establishing a second broadcast influence factor based on the node importance, wherein the node importance is calculated through the node key value and the node complexity; configuring a neighborhood broadcast request using the first broadcast influence factor and the second broadcast influence factor, and performing neighborhood broadcast processing.

[0052] The specific configuration of the consistency verification module 40 will be described in detail below. The consistency verification module 40 further includes: identifying the neighborhood propagation path based on the neighborhood broadcast request and the topology-aware structure, and configuring a consistency verification request for anomaly trend indicators along the corresponding propagation path; after performing path adaptability verification based on the propagation path on the consistency verification request, outputting a consistency factor, wherein the path adaptability verification includes path distance attenuation consistency verification and propagation delay verification of path timing.

[0053] The specific configuration of the collaborative early warning module 50 will be described in detail below. The collaborative early warning module 50 further includes: establishing topological weight coefficients based on the topological sensing structure; locating abnormal nodes using the enhanced correction results, and then performing abnormal node resonance detection based on the topological weight coefficients through a swarm resonance detection channel to generate a swarm resonance index; and matching early warning signals based on the swarm resonance index and the enhanced correction results to establish a collaborative early warning system.

[0054] The specific configuration of the collaborative early warning module 50 will be described in detail below. The swarm resonance detection channel of the collaborative early warning module 50 is as follows: ; in, Characterizing the population resonance index, The total number of monitoring nodes, For monitoring nodes The set of node pairs, i.e., with monitoring nodes The set of node pairs consisting of monitoring nodes that satisfy the adjacency relationship. Characterization monitoring nodes At any moment Pressure change Characterization monitoring nodes At any moment Pressure change Characterization by monitoring nodes propagation to monitoring nodes The propagation delay represents the weight coefficient of the node pair.

[0055] The specific configuration of the early warning signal reporting module 60 will be described in detail below. The early warning signal reporting module 60 further includes: configuring an early warning response strategy based on the early warning signal of abnormal pressure in the fire protection pipeline network, and establishing an execution detection cycle for the early warning response strategy; using the execution detection cycle to evaluate the timeliness of the early warning response strategy and establish timeliness compensation; and issuing additional early warnings based on the timeliness compensation.

[0056] The fire protection pipeline pressure dynamic monitoring system combined with the Internet of Things provided in this embodiment of the invention can execute the fire protection pipeline pressure dynamic monitoring method combined with the Internet of Things provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0057] Based on the foregoing embodiments, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the dynamic monitoring method for fire protection pipeline pressure combined with the Internet of Things as described in any of the preceding embodiments.

[0058] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for dynamic monitoring of fire protection pipeline pressure using the Internet of Things, characterized in that, The method includes: Pressure sensors and auxiliary sensors are deployed at each monitoring node of the fire protection pipeline network. The auxiliary sensors include flow sensors and temperature sensors. The topological sensing structure of the fire protection pipeline network is established by using the wireless Internet of Things communication of the monitoring nodes. After adaptive sampling adjustment of the monitoring nodes based on the aforementioned topology sensing structure, a node sampling dataset is established. The node sampling datasets are input into the multi-time-window pressure evolution model mapped to the monitoring nodes, and sliding differential analysis is performed to calculate abnormal trend indicators. After configuring the neighborhood broadcast request according to the abnormal trend indicator, broadcast consistency verification is performed according to the topology-aware structure to generate a consistency factor. After using the aforementioned consistency factor to enhance and correct abnormal trend indicators, a coordinated early warning system is established based on the enhanced correction results and the topological sensing structure to perform group resonance detection. Based on the enhanced correction results, a coordinated early warning signal for abnormal pressure in the fire protection pipeline network is issued. Based on the enhanced correction results and topology sensing structure, a coordinated early warning system is established for swarm resonance detection, including: Establish topology weight coefficients based on the topology-aware structure; After locating abnormal nodes using the enhanced correction results, abnormal node resonance detection based on topological weight coefficients is performed through the swarm resonance detection channel to generate a swarm resonance index. Based on the population resonance index and the enhancement correction results, early warning signals are matched to establish a collaborative early warning system. The group resonance detection channel is as follows: ； in, Characterizing the population resonance index, The total number of monitoring nodes, For monitoring nodes The set of node pairs, i.e., with monitoring nodes The set of node pairs consisting of monitoring nodes that satisfy the adjacency relationship. Characterization monitoring nodes At any moment Pressure change Characterization monitoring nodes At any moment Pressure change Characterization by monitoring nodes propagation to monitoring nodes The propagation delay represents the weight coefficient of the node pair.

2. The method for dynamic monitoring of fire protection pipeline pressure combined with the Internet of Things as described in claim 1, characterized in that, The node sampling datasets are input into a multi-time-window pressure evolution model mapped to the monitoring nodes, and sliding differential analysis is performed to calculate abnormal trend indicators, including: After receiving the corresponding node sampling dataset using the multi-time-window stress evolution model, the sampling sequence is subjected to time-weighted stratification. The stress evolution sub-model is configured according to the time-weighted stratification result. The stress evolution sub-model is configured with time window weights based on the length of the time window, the intensity of data fluctuations, and the topology-aware connectivity. Using the configured pressure evolution sub-model, a sliding differential analysis was performed under the corresponding time window to extract the first-order trend term and the second-order acceleration term of the pressure change, and the auxiliary coupling response of the auxiliary sensor was calculated. The outputs of all pressure evolution sub-models are weighted and fused using the time window weights to calculate abnormal trend indicators.

3. The method for dynamic monitoring of fire protection pipeline pressure combined with the Internet of Things as described in claim 2, characterized in that, The outputs of all pressure evolution sub-models are weighted and fused using the aforementioned time window weights to calculate abnormal trend indicators, including: Nonlinear cross-correlation analysis is performed on the output of each pressure evolution sub-model within the node to establish the perturbation coupling degree within the node; After dynamically correcting the time window weights based on the perturbation coupling degree within the node, the weighted fusion of the outputs of all pressure evolution sub-models is performed to extract the consistent principal component. An abnormal trend indicator is constructed using the fluctuation amplitude of the consistent principal component.

4. The method for dynamic monitoring of fire protection pipeline pressure combined with the Internet of Things as described in claim 1, characterized in that, After configuring the neighborhood broadcast request according to the abnormal trend indicator, broadcast consistency verification is performed according to the topology-aware structure to generate a consistency factor, including: Anomaly degree evaluation is performed using the aforementioned abnormal trend indicators to establish a first broadcast impact factor; Obtain the node importance of the monitoring node corresponding to the abnormal trend indicator, and establish a second broadcast influence factor based on the node importance. The node importance is calculated by the node key value and the node complexity. Configure neighborhood broadcast requests using the first broadcast impact factor and the second broadcast impact factor, and perform neighborhood broadcast processing.

5. The method for dynamic monitoring of fire protection pipeline pressure combined with the Internet of Things as described in claim 4, characterized in that, Configure neighborhood broadcast requests using the first broadcast impact factor and the second broadcast impact factor, and perform neighborhood broadcast processing, including: Based on the neighborhood broadcast request and topology-aware structure, the neighborhood propagation path is identified, and a consistency verification request for the abnormal trend indicator is configured along the corresponding propagation path. After performing path-adaptive verification based on the propagation path on the consistency verification request, a consistency factor is output. The path-adaptive verification includes path distance attenuation consistency verification and path timing propagation delay verification.

6. The method for dynamic monitoring of fire protection pipeline pressure combined with the Internet of Things as described in claim 1, characterized in that, Based on the enhanced correction results, the coordinated early warning system issues a warning signal for abnormal pressure in the fire protection pipeline network, including: Configure early warning response strategies based on abnormal pressure warning signals from the fire protection pipeline network, and establish an execution detection cycle for the early warning response strategies; The timeliness of the early warning response strategy is evaluated using the aforementioned execution detection cycle, and timeliness compensation is established. Additional early warnings will be issued based on the aforementioned timeliness compensation.

7. A dynamic monitoring system for fire-fighting pipeline pressure combined with the Internet of Things, characterized in that, The system is used to implement the dynamic monitoring method for fire protection pipeline pressure combined with the Internet of Things as described in any one of claims 1 to 6, and the system includes: The sensing structure establishment module is used to deploy pressure sensors and auxiliary sensors in each monitoring node of the fire protection pipeline network, wherein the auxiliary sensors include flow sensors and temperature sensors, and to establish the topological sensing structure of the fire protection pipeline network using the wireless Internet of Things communication of the monitoring nodes. The sampling data acquisition module is used to establish a node sampling dataset after adaptively adjusting the sampling of the monitoring nodes based on the topology sensing structure. The data analysis module is used to input the node sampling datasets into the multi-time-window pressure evolution model mapped to the monitoring nodes, perform sliding differential analysis, and calculate abnormal trend indicators. The consistency verification module is used to configure the neighborhood broadcast request according to the abnormal trend index, and then perform broadcast consistency verification according to the topology-aware structure to generate a consistency factor. The collaborative early warning module is used to establish a linkage collaborative early warning based on the enhanced correction result and the topological sensing structure after using the consistency factor to enhance and correct the abnormal trend indicators and perform group resonance detection. The early warning signal reporting module is used to issue an early warning signal for abnormal pressure in the fire protection pipeline network based on the enhanced correction results and the linkage and collaborative early warning system.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method for dynamic monitoring of fire protection pipeline pressure in conjunction with the Internet of Things as described in any one of claims 1-6.

Images