A coal-based solid waste filling fire prevention and extinguishing intelligent early warning method, system, device and medium

Abstract

The application discloses a coal-based solid waste filling fire prevention and extinguishing intelligent early warning method, system, device and medium, the method first obtains the environmental physical quantity data and network state original data of the filling area through the wireless mesh network including the anchoring node and the drift node, then carries out signal propagation analysis on the network state original data, extracts the signal propagation abnormal characteristics caused by environmental medium disturbance, obtains the first perception quantity, at the same time, carries out combustion analysis on the environmental physical quantity data, extracts the physicochemical abnormal characteristics related to the combustion process, obtains the second perception quantity, then carries out space-time correlation calculation on the first perception quantity and the second perception quantity in the same space-time range, obtains the coupling risk index, and generates the early warning instruction when the coupling risk index exceeds the preset risk threshold. The application can improve the reliability, accuracy and early perception ability of the coal-based solid waste filling area hidden fire source early warning.

Description

Technical Field

[0001] This invention relates to the field of safety technology in coal mine backfilling mining, and in particular to an intelligent early warning method, system, equipment and medium for fire prevention and extinguishing in coal-based solid waste backfilling. Background Technology

[0002] Using coal-based solid waste (such as coal gangue) for goaf backfilling is an important technical means to dispose of solid waste, control surface subsidence, and prevent spontaneous combustion of residual coal pillars or broken surrounding rock. After pretreatment, the content of residual coal powder (or residual coal) in the backfill material is significantly reduced, making it less prone to low-temperature oxidation under normal conditions. However, the internal environment of the backfill is complex. Abnormal localized heating may occur due to heat conduction from concealed ignition sources in adjacent areas, the slow reaction of small amounts of easily oxidized components in the backfill material, or the influence of other external heat sources, forming a heat accumulation zone hidden within the backfill. These heat sources are highly concealed, and their development cannot be directly observed. If early detection and accurate warning are not provided, they can easily develop into open flames, causing fires in the backfill area and seriously threatening safe mine production.

[0003] Currently, fire prevention and extinguishing monitoring in coal mine backfill areas mainly relies on traditional monitoring methods, which have many technical shortcomings: First, the monitoring methods have physical blind spots. Traditional monitoring often uses fixed point sensors or bundled tube sampling systems deployed in accessible areas such as roadways and working faces. The monitoring probes cannot extend into the backfill body that has been buried by materials, making it impossible to obtain the true environmental conditions of the core area of ​​the backfill body and to detect early hidden heat sources. Second, the reliability of early warning is insufficient. Existing technologies mostly use single-parameter threshold alarms such as temperature and gas, resulting in a single dimension of monitoring information. This makes them susceptible to fluctuations in the underground environmental background and interference from local non-fire source heat sources, generally leading to high false alarm rates and a high risk of missed alarms. Furthermore, they cannot accurately locate the spatial location of heat risk sources. Third, they lack early detection capabilities. Existing technologies cannot detect changes in the overall physical field of the environmental medium inside the backfill body, making it difficult to capture early abnormal signs before a significant temperature increase in the heat accumulation zone and the occurrence of open combustion, resulting in a significant lag in early warning. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent early warning method, system, equipment, and medium for fire prevention and extinguishing in coal-based solid waste filling areas, thereby improving the reliability, accuracy, and early detection capability of early warning of concealed fire sources in coal-based solid waste filling areas.

[0005] To achieve the above objectives, this invention provides an intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling, comprising: Environmental physical quantity data and native network status data of the filling area are acquired through a wireless mesh network; wherein, the wireless mesh network includes anchoring nodes and drifting nodes; The network state raw data is analyzed for signal propagation, and abnormal signal propagation features caused by environmental medium disturbances are extracted to obtain the first sensing quantity. Combustion analysis is performed on the environmental physical quantity data to extract physicochemical anomaly features related to the combustion process, thereby obtaining the second sensing quantity; The first and second sensed quantities within the same spatiotemporal range are spatiotemporally correlated to obtain a coupling risk index. When the coupling risk index exceeds a preset risk threshold, an early warning instruction is generated.

[0006] Optionally, the anchoring node is fixed to the surrounding rock structure of the filling area, and the drifting node is put into the filling area along with the filling material according to a preset deployment strategy.

[0007] Optionally, the step of performing signal propagation analysis on the raw network state data to extract abnormal signal propagation features caused by environmental medium disturbances and obtain the first sensing quantity includes: From the raw network status data, obtain the link quality indicator value sequence and the received signal strength indicator value sequence corresponding to the target communication link; Trend analysis is performed on the link quality indicator sequence and the received signal strength indicator sequence to extract persistent degradation features not caused by network faults; The persistent degradation characteristics are quantified into an index characterizing the degree of thermal disturbance of the medium, thus obtaining the first sensing quantity.

[0008] Optionally, the step of performing flammability analysis on the environmental physical quantity data to extract physicochemical anomaly features related to the combustion process and obtain a second sensing quantity includes: Combustion analysis was performed on the environmental physical quantity data to calculate the temperature change rate and characteristic gas concentration ratio of the target node; A comprehensive analysis was performed on the temperature change rate and the ratio of the characteristic gas concentrations to obtain a comprehensive analysis result. When the temperature change rate is greater than a preset change rate threshold and the characteristic gas concentration ratio is greater than a preset ratio threshold, the comprehensive analysis result is quantified into an index characterizing the intensity of the combustion reaction, thus obtaining a second sensing quantity.

[0009] Optionally, the step of performing spatiotemporal correlation calculation on the first and second sensed quantities within the same spatiotemporal range to obtain a coupling risk index includes: Within a preset time-space window, determine whether the first perceived quantity and the second perceived quantity are consistent in spatial location; If there is consistency, the joint anomaly strength is calculated based on the first and second perceived quantities, and the joint anomaly strength is quantified as a coupled risk index. If there is no consistency, no warning will be triggered.

[0010] Optionally, determining whether the first perceived quantity and the second perceived quantity are consistent in spatial location within a preset spatiotemporal window includes: Within a preset time-space window, the spatial path of the target communication link that causes the first sensing quantity to be abnormal is determined, and a first spatial influence range is constructed along the spatial path. Within a preset spatiotemporal window, determine the spatial coordinates of the target node for generating the second sensing quantity, and construct a second spatial influence range centered on the target node; When the first spatial influence range and the second spatial influence range intersect in space, it is determined that the first perceived quantity and the second perceived quantity are consistent in spatial location.

[0011] Optionally, calculating the joint anomaly intensity based on the first sensed quantity and the second sensed quantity includes: Calculate the spatial density of all abnormal links corresponding to the first sensing quantity; Calculate the spatial cluster centers and dispersion of all abnormal nodes corresponding to the second sensing quantity; Based on the spatial density, the spatial cluster centers, and the dispersion, a comprehensive factor characterizing the spatial focusing degree of abnormal events is calculated. The combined factor is weighted and fused with the intensities of the first and second perceived quantities to obtain the joint anomaly intensity.

[0012] To achieve the above objectives, the present invention also provides an intelligent early warning system for fire prevention and extinguishing in coal-based solid waste backfilling, comprising: The data acquisition module is used to acquire environmental physical quantity data and network status native data of the filling area through a wireless mesh network; wherein, the wireless mesh network includes anchoring nodes and drifting nodes; The first sensing module is used to perform signal propagation analysis on the raw network state data, extract abnormal signal propagation features caused by environmental medium disturbances, and obtain the first sensing quantity. The second sensing module is used to perform combustion analysis on the environmental physical quantity data, extract physicochemical anomaly features related to the combustion process, and obtain the second sensing quantity; The correlation calculation module is used to perform spatiotemporal correlation calculation on the first sensing quantity and the second sensing quantity within the same spatiotemporal range to obtain a coupling risk index. The early warning module is used to generate an early warning command when the coupled risk indicator exceeds a preset risk threshold.

[0013] To achieve the above objectives, the present invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the intelligent early warning method for fire prevention and extinguishing of coal-based solid waste backfill as described above.

[0014] To achieve the above objectives, the present invention also provides a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the intelligent early warning method for fire prevention and extinguishing of coal-based solid waste backfill as described above.

[0015] Compared with existing technologies, the present invention provides a method, system, equipment, and medium for intelligent early warning of fire prevention and extinguishing in coal-based solid waste backfilling. By simultaneously extracting the signal propagation anomaly characteristics of native network status data and the combustion physicochemical anomaly characteristics of environmental physical quantity data, multi-dimensional monitoring information can be formed. This effectively overcomes the shortcomings of single-parameter threshold alarms, such as susceptibility to environmental interference, high false alarm rate, and high risk of missed alarms, thereby improving the reliability of early warning. By performing spatiotemporal correlation calculations on the two types of sensing quantities to obtain coupled risk indicators and trigger early warnings, early abnormal signs can be captured before significant temperature rise and open combustion in the thermal accumulation zone, improving early sensing capabilities and the operability of early warning, thereby ensuring safe production in the mine. Attached Figure Description

[0016] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the embodiments 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.

[0017] Figure 1 This is a flowchart of an intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling, provided by an embodiment of the present invention. Figure 2 This is a structural block diagram of an intelligent early warning system for fire prevention and extinguishing in coal-based solid waste backfilling, provided in an embodiment of the present invention. Figure 3 This is a structural block diagram of a terminal device provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] See Figure 1 , Figure 1 This is a flowchart of an intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling, provided by an embodiment of the present invention. Figure 1 As shown, the intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling includes steps S1 to S5: Step S1: Obtain environmental physical quantity data and network status native data of the filling area through a wireless mesh network; wherein, the wireless mesh network includes anchoring nodes and drifting nodes; It should be noted that the filling area refers to the specific spatial range in which the filling mining process is carried out underground in a coal mine; the surrounding rock structure refers to the original rock mass or artificial support body around the filling area that has not been mined or needs to be supported, such as the roof of the roadway, the two sides, and the coal face; the wireless mesh network can be a Z-Wave wireless mesh network, a low-power wireless network based on the Z-Wave communication protocol that has self-organizing and multi-hop routing capabilities.

[0020] In one alternative embodiment, the anchoring node is fixed to the surrounding rock structure of the filling area, and the drifting node is put into the filling area along with the filling material according to a preset deployment strategy.

[0021] For example, an anchoring node is a wireless sensing device fixed to the surrounding rock structure; a drifting node is a wireless sensing device without a fixed installation position that can move with the filling material and eventually be buried inside the filling body.

[0022] Therefore, in practice, the first step is to deploy a Z-Wave wireless mesh network consisting of anchor nodes and drifting nodes within the filling area.

[0023] For example, based on the boundary planning of the filling area, engineers physically fix the anchoring nodes at pre-selected locations in the surrounding rock structure, for example, using bolts or special clamps. The selection of these locations must ensure that the nodes will not be impacted or buried by materials throughout the filling operation. After fixing, the anchoring nodes are connected to the existing intrinsically safe power supply network underground via mining cables, thereby providing them with a continuous and stable power supply to form fixed network backbone nodes.

[0024] Then, during the process of transporting the filling material (such as crushed coal gangue slurry or solid) to the filling area via pipelines or belt conveyors, drifting nodes encased in robust explosion-proof shells are introduced into the material flow according to a pre-calculated deployment strategy. For example, one deployment strategy is based on a spatial grid, where a drifting node is deployed for every certain volume of material filled or at fixed time intervals to ensure statistically uniform distribution of nodes within the filling body. This allows the drifting nodes to move naturally with the flow, settling, and consolidation of the material, and ultimately be randomly distributed within the buried filling body, with their initial deployment positions known.

[0025] Finally, after all anchored nodes and deployed drifting nodes are powered on, their integrated Z-Wave communication modules will automatically execute the network discovery, neighbor identification, and routing table establishment processes defined in the protocol stack. Anchored nodes, with their stable power supply and fixed positions, naturally become the backbone routing nodes and gateways of the network; drifting nodes join the network as mobile end-sensing nodes. Through multi-hop relay communication, even deeply buried drifting nodes can relay data to the network gateway through neighboring drifting nodes or anchored nodes, thus spontaneously forming a wireless mesh data transmission channel that covers the surface and interior of the filling area and has dynamic repair capabilities.

[0026] It is worth noting that the embodiments of the present invention, through a combination of fixed and mobile node deployment, can extend sensing capabilities to the interior of buried filling bodies that are inaccessible by traditional methods, fundamentally solving the problem that monitoring methods cannot reach concealed areas, and providing a physical possibility for obtaining early characteristic signals of fire sources.

[0027] In an optional embodiment, step S1 includes steps S101 to S102: Step S101: Control the anchoring node and the drifting node through a wireless mesh network, and periodically collect environmental physical quantity data through the node sensing unit. It should be noted that, in specific implementations, the node sensing unit can be a combination of miniaturized sensor modules integrated inside each anchored node and drifting node; environmental physical quantity data are physical parameter values ​​that are directly measured and output by the node sensing unit, reflecting the local micro-environmental state of the node; and network status native data are technical parameters that are automatically generated and provided by the underlying layer of the communication protocol stack when nodes transmit and receive wireless data packets based on the Z-Wave protocol, used to quantitatively evaluate the current wireless link transmission quality.

[0028] For example, in step S101, a server deployed at a ground monitoring center or a controller integrated into a gateway node can be used to uniformly set the working rhythm of all anchored and drifting nodes through downlink network commands. The microcontroller inside each node wakes up its connected sensor units at regular intervals based on this time. For instance, a temperature sensor reads the current temperature value through a digital interface, and a gas sensor obtains the concentration values ​​of carbon monoxide and carbon dioxide through an analog-to-digital converter. These measurements are appended with the unique identifier of the acquisition node and a precise timestamp, and packaged into a standard format environmental data message.

[0029] Step S102: Record the native network status data generated in real time by the anchor node and the drifting node during network communication through the node sensing unit.

[0030] For example, when any node (whether a drifting node acting as a data source or an anchored node acting as a relay) needs to send the aforementioned environmental data packets or routing information, it encapsulates the data into frames according to the Z-Wave protocol and sends them through the radio frequency module. During this physical process of sending and receiving, the protocol stacks of both communicating parties monitor the link status in real time. For instance, the receiver's radio frequency chip measures the strength of the received signal, i.e., the received signal strength indicator value, while the physical layer evaluates the quality of signal decoding and generates a link quality indicator value. These values ​​are not application data but byproducts of the communication process. The protocol stack software is configured to automatically generate a network status record after each successful packet exchange, along with the node identifiers of both communicating parties and the communication timestamp. Ultimately, the environmental data packets are transmitted to the gateway via multi-hop routing, while the network status record is stored locally by each participating node and subsequently synchronously reported to the monitoring center, thus achieving parallel acquisition of both types of data.

[0031] It is worth noting that step S1 in this embodiment of the invention not only obtains direct evidence reflecting the combustion chemical process, such as temperature and gas concentration, but also indirect evidence reflecting changes in the physical properties of the environmental medium, namely the wireless link status. This can build a richer pool of sensing information than single sensor data, laying a data foundation for solving the problem of difficulty in capturing early signs.

[0032] Step S2: Perform signal propagation analysis on the original network state data, extract abnormal signal propagation features caused by environmental medium disturbances, and obtain the first sensing quantity; In an optional embodiment, step S2 includes steps S201 to S203: Step S201: Obtain the link quality indicator value sequence and the received signal strength indicator value sequence corresponding to the target communication link from the network status raw data; Step S202: Perform trend analysis on the link quality indicator value sequence and the received signal strength indicator value sequence to extract persistent degradation features not caused by network faults; Step S203: Quantify the continuous degradation characteristics into an index characterizing the degree of thermal disturbance of the medium to obtain the first sensing quantity.

[0033] It should be noted that signal propagation analysis refers to the process of analyzing the changes in the quality parameters of wireless communication signals as they propagate in a medium; environmental medium disturbance specifically refers to the comprehensive physical state changes inside the filling body, such as the increase in medium temperature, changes in gas composition and convection, caused by the thermal effect of potential fire sources; the numerical index obtained by quantifying the abnormal characteristics of wireless signal propagation, used to characterize the degree of thermal disturbance of the environmental medium, serves as the first sensing quantity.

[0034] Specifically, the time-series data sequence of the target link is first extracted from the aggregated network state raw data. For example, for each pair of nodes with a historically stable communication record, such as a communication link between an anchored node and a drifting node, all data belonging to that link are filtered from the massive network state records. These records are then sorted by timestamp, and the sequences of link quality indicators (BMIs) and received signal strength indicators (RSIs) changing over time are extracted. For instance, for the link between anchored node A and drifting node D, a sequence of link quality indicators from multiple consecutive sampling times and a corresponding sequence of RSIs can be obtained.

[0035] Then, refined trend analysis and anomaly detection are performed on the extracted time series to eliminate transient interference and capture persistent degradation patterns. Specifically, a sliding time window method is used to analyze each series. Within a time window, for example, covering data points over several hours, the algorithm calculates the statistical trend of the series. For example, the Mann-Kendall trend test is used to determine whether there is a statistically significant downward trend in the link quality indicator series within the window. The same analysis is performed on the received signal strength indicator series. The key to this analysis is to distinguish between transient signal fluctuations caused by occasional node movement or temporary obstruction and systematic degradation of signal propagation characteristics caused by slow and continuous changes in the physical properties of the environmental medium. Only links that are determined to have a persistent downward trend are considered to be affected by environmental medium disturbances.

[0036] Finally, the identified persistent degradation features are quantified into a calculable metric, namely the first sensing quantity. Specifically, for each target communication link determined to have persistent degradation, a quantification formula is used for calculation. The principle behind this formula is to measure the deviation of the current signal quality from a baseline level. The formula can be expressed as follows: the first sensing quantity equals a weighting coefficient multiplied by the relative attenuation of the average link quality indicator value within the current window compared to the baseline link quality indicator value, plus another weighting coefficient multiplied by the relative change in the average received signal strength indicator value within the current window compared to the baseline received signal strength indicator value. For example, the formula for calculating the first sensing quantity is as follows: ; in, This is the weighting coefficient for the link quality indicator value; Indicates the baseline value of the link quality indicator; This indicates the average link quality indicator value within the current window; The weighting coefficient for the received signal strength indication value; Indicates the baseline value of the received signal strength indication; This represents the average received signal strength indication value within the current window. It should be noted that the baseline value of the link quality indication value and the baseline value of the received signal strength indication value are reference values ​​calculated for the link during the initial stable and undisturbed phase. The sum of the two weighting coefficients is one, which is used to balance the different dimensions and importance of the two parameters.

[0037] This formula can transform an abstract concept of signal degradation into a concrete, numerical first sense quantity. The larger the value, the stronger the thermal disturbance in the medium space traversed by the link.

[0038] It is worth noting that, in this embodiment of the invention, the core of step S3 is to convert the state data of the communication network itself into a sensitive sensing signal for the physical changes of the internal environment medium of the filling body. It creatively utilizes the network communication state information that will inevitably be generated, and can achieve sensitive perception of the overall thermophysical state changes of the medium inside a large area of ​​filling without additional sensors. This enables non-contact, large-scale monitoring of hidden thermal disturbances, effectively addressing the dilemma that sensors cannot be densely distributed inside the burial area.

[0039] Step S3: Perform combustion analysis on the environmental physical quantity data, extract physicochemical anomaly features related to the combustion process, and obtain the second sensing quantity; In an optional embodiment, step S3 includes steps S301 to S303: Step S301: Perform combustion analysis on the environmental physical quantity data to calculate the temperature change rate and characteristic gas concentration ratio of the target node; Step S302: Perform a comprehensive analysis on the temperature change rate and the ratio of the characteristic gas concentrations to obtain a comprehensive analysis result; Step S303: When the temperature change rate is greater than a preset change rate threshold and the characteristic gas concentration ratio is greater than a preset ratio threshold, the comprehensive analysis result is quantified into an index characterizing the intensity of the combustion reaction to obtain a second sensing quantity.

[0040] It should be noted that combustion analysis refers to the method of analyzing and judging relevant parameters based on the basic physicochemical laws followed by the combustion process of substances; the characteristic gas concentration ratio refers to the proportional relationship between the concentrations of two gases that have indicative significance during the low-temperature oxidation and combustion of coal; and the numerical index obtained by comprehensively analyzing and quantifying the physicochemical characteristics related to combustion, used to characterize the intensity of local combustion reaction, serves as the second sensing quantity.

[0041] Specifically, key characteristic parameters are first calculated based on the time-series data of environmental physical quantities at each node. For example, for each anchored or drifting node (i.e., the target node) that reported environmental data, temperature and gas concentration data sequences are extracted from its reported time-series data. For calculating the rate of temperature change, a difference method is used; for instance, the current temperature value is subtracted from the temperature value a short time ago, and then divided by the time interval, resulting in an instantaneous or average rate of change reflecting the speed of local temperature rise at that node. For calculating the characteristic gas concentration ratio, the carbon monoxide concentration value is directly taken from the gas concentration data at the same moment or within the same time window, and divided by the carbon dioxide concentration value to obtain a dimensionless ratio. This ratio is one of the key indicators for judging the development stage of spontaneous combustion of coal.

[0042] Then, a comprehensive analysis is performed on the calculated temperature change rate and characteristic gas concentration ratio. The core of this analysis is the application of a pre-set dual criterion for logical judgment. For example, two independent combustion characteristic condition thresholds are preset. The first condition is that the temperature change rate must be greater than a positive threshold, which is determined through laboratory coal sample heating experiments and on-site safety standards. This threshold is used to screen out abnormal rapid heating points. The second condition is that the characteristic gas concentration ratio must be greater than another positive threshold, which is determined based on the spontaneous combustion characteristic gas generation law of coal. This threshold is used to identify gas composition patterns that meet the early signs of combustion. The comprehensive analysis requires that both conditions must be met simultaneously, forming an "AND" logic. This dual criterion design greatly eliminates false alarms of single parameters caused by environmental background fluctuations or occasional sensor errors. For example, a temperature increase alone may be due to heat dissipation from electromechanical equipment, and an abnormal gas ratio alone may be due to the slow oxidation of a small amount of organic matter. However, an abnormality in both conditions strongly points to an active combustion reaction.

[0043] Finally, the comprehensive analysis results satisfying both criteria are quantified into a unified second sensing quantity. Specifically, when the data of a target node is determined to simultaneously satisfy both conditions, the normalized temperature change rate and the normalized characteristic gas concentration ratio are weighted and summed using a quantification formula. Specifically, the second sensing quantity equals a weighting coefficient multiplied by the quotient of the node's actual temperature change rate and a temperature change rate threshold, plus another weighting coefficient multiplied by the quotient of the node's actual characteristic gas concentration ratio and a characteristic gas concentration ratio threshold. The sum of the two weighting coefficients is one, used to adjust the contribution ratio of the two features in the final judgment. For example, the calculation formula for the second sensing quantity is as follows: ; in, This represents the weighting coefficient for the rate of temperature change. This represents the actual rate of temperature change at the target node; The threshold representing the rate of temperature change of the target node; Indicates the weighting coefficient of the characteristic gas concentration ratio; This represents the ratio of the actual characteristic gas concentrations at the target node; The threshold value representing the characteristic gas concentration ratio of the target node.

[0044] This quantitative formula can transform the qualitative conclusion of "meets combustion characteristics" into a quantitative value that can characterize the intensity of the reaction, namely the second sensing value. The larger the value, the stronger the combustion reaction detected near the node.

[0045] It is worth noting that, in this embodiment of the invention, the core of step S4 lies in performing in-depth analysis of the raw environmental data directly collected by the nodes based on the combustion principle, thereby extracting direct and reliable evidence characterizing the local combustion reaction. By intelligently identifying and quantifying the point data collected by the drifting nodes, the accurate location and intensity assessment of local hot spots or early fire sources inside the filling body can be achieved, effectively solving the problems of missed alarms, false alarms, or slow response caused by relying on manual experience judgment or single threshold alarms.

[0046] Step S4: Perform spatiotemporal correlation calculation on the first and second sensing quantities within the same spatiotemporal range to obtain the coupling risk index; In an optional embodiment, step S4 includes steps S401 to S403: Step S401: Within a preset time-space window, determine whether the first perceived quantity and the second perceived quantity are consistent in spatial location; It should be noted that the preset spatiotemporal window refers to a time length and spatial range constraint set in advance for data association; spatial location consistency means that the abnormal area reflected by the first sensing quantity and the abnormal point reflected by the second sensing quantity have an overlapping or adjacent relationship in three-dimensional space.

[0047] In an optional embodiment, step S401 includes steps S4011 to S4013: Step S4011: Within a preset time-space window, determine the spatial path of the target communication link that causes the first sensing quantity to be abnormal, and construct a first spatial influence range along the spatial path; Step S4012: Within a preset spatiotemporal window, determine the spatial coordinates of the target node for generating the second sensing quantity, and construct a second spatial influence range centered on the target node; Step S4013: When the first spatial influence range and the second spatial influence range intersect in space, it is determined that the first perceived quantity and the second perceived quantity are consistent in spatial position.

[0048] For example, based on the coordinates of the nodes at both ends of the target communication link that generate the first abnormal sensing quantity, such as the coordinates of an anchor node and a drifting node, a spatial line segment path is defined. Along this path, with a radius determined by the typical effective distance of radio waves affected by the medium as an offset, a cylindrical spatial region centered on the line segment is constructed. This is the first spatial influence range.

[0049] At the same time, based on the known or estimated coordinates of the target node (usually a drifting node) of the reported second sensing quantity, such as the coordinates estimated by the initial placement location and the material flow model, a spherical spatial region is constructed with that point as the center and a distance determined according to the effective monitoring radius of its sensor as the radius. This is the second spatial influence range.

[0050] Then, a spatial computational geometry algorithm is invoked to determine whether the spherical region representing direct evidence (nodes) and the cylindrical region representing indirect evidence (link influence domain) intersect in three-dimensional space. If they intersect, it proves that the direct combustion characteristics detected by the node and the physical disturbance of the medium sensed by the link may originate from the same event in space, thus determining that they have spatial consistency. This judgment method based on geometric spatial relationships transforms the abstract "positional correlation" into a rigorously calculable operation, which can improve the objectivity and accuracy of the judgment.

[0051] Step S402: If there is consistency, calculate the joint anomaly strength based on the first sensing quantity and the second sensing quantity, and quantify the joint anomaly strength into a coupling risk index. In one optional embodiment, calculating the joint anomaly intensity based on the first sensing value and the second sensing value includes: Calculate the spatial density of all abnormal links corresponding to the first sensing quantity; Calculate the spatial cluster centers and dispersion of all abnormal nodes corresponding to the second sensing quantity; Based on the spatial density, the spatial cluster centers, and the dispersion, a comprehensive factor characterizing the spatial focusing degree of abnormal events is calculated. The combined factor is weighted and fused with the intensities of the first and second perceived quantities to obtain the joint anomaly intensity.

[0052] For example, firstly, all communication links with high first sensing values ​​that are within the spatiotemporal window and spatially intersect with the second sensing value anomaly region are extracted. Then, the volume coverage of the second spatial influence range (cylinder) corresponding to these links in the suspicious space is calculated, or the number of abnormal links per unit volume is directly calculated as the spatial density. The higher the density, the more widespread and intense the overall thermal disturbance of the medium in the region.

[0053] Simultaneously, all nodes with high second sensing values ​​that are within the spatiotemporal window and spatially intersect with the first sensing value anomaly region are extracted. The arithmetic mean of the coordinates of these nodes is calculated to obtain the spatial cluster center, which may indicate the core area of ​​a potential fire source. At the same time, the standard deviation or radius of dispersion of the coordinates of these nodes relative to the cluster center is calculated as the dispersion. The smaller the dispersion, the more concentrated the evidence of direct combustion is in space.

[0054] Then, using the spatial density and dispersion calculated above, a comprehensive factor is constructed. For example, the comprehensive factor can be proportional to the spatial density, inversely proportional to the dispersion, and a constant. This comprehensive factor quantifies the spatial characteristics of a high-reliability risk scenario where "multiple direct evidence points appear in a concentrated manner and are simultaneously surrounded by a large range of indirect signal anomalies."

[0055] Finally, the calculated average first perception intensity, average second perception intensity, and comprehensive factor are substituted into a weighted summation formula, which states that the joint anomaly intensity equals the first weight coefficient multiplied by the average first perception intensity, plus the second weight coefficient multiplied by the average second perception intensity, plus the third weight coefficient multiplied by the comprehensive factor, with the sum of the three weight coefficients being one. For example, the formula for calculating the joint anomaly intensity is as follows: ; in, This represents the average first perceived quantity. The average second perception quantity; As a comprehensive factor; , , , The weighting coefficient is determined based on historical data or expert experience to balance the contributions of the strength of direct evidence, the strength of indirect evidence, and the consistency of spatial evidence.

[0056] The final joint anomaly strength calculated using this weighted fusion formula is a high-dimensional feature scalar that integrates strength, breadth, and consistency.

[0057] Step S403: If there is no consistency, the warning will not be triggered.

[0058] It should be noted that if there is no consistency, it indicates that the current anomaly is a single signal interference or a non-fire source disturbance. In this case, the current coupled risk calculation will be terminated, the abnormal data will be stored, and the continuous monitoring step will be returned without triggering the subsequent early warning process.

[0059] It is worth noting that, through spatiotemporal correlation and multi-level fusion, the embodiments of the present invention can effectively distinguish and integrate isolated abnormal signals that may be caused by different interference factors with the characteristics of real hidden fire sources. This can reduce the false alarm rate of a single chain of evidence, while improving the confidence in identifying early risks of real fire sources, the accuracy of spatial positioning, and the reliability of early warning.

[0060] Step S5: When the coupling risk index exceeds the preset risk threshold, an early warning instruction is generated.

[0061] In an optional embodiment, step S5 includes steps S501 to S503: Step S501: When the coupling risk index exceeds the preset risk threshold, the risk level is determined according to the value range of the coupling risk index and the preset risk level mapping relationship. Step S502: Based on the physical locations of the nodes that generate the first and second sensing quantities, predict the risk area through spatial aggregation calculation; Step S503: Generate an early warning instruction based on the risk level and the risk area.

[0062] It should be noted that the preset risk threshold can be a coupled risk indicator critical value pre-set based on historical accident data, experimental verification, and safety procedures to determine whether to activate the early warning; the risk level is a qualitative classification based on the level of the coupled risk indicator, used to characterize the urgency and severity of the fire hazard. The risk area prediction result can be a geometric description of the possible distribution range of potential fire sources obtained by analyzing the spatial location of the abnormal signal source. The early warning instruction is a structured alarm message that includes the above-mentioned risk level and risk area prediction results, used to drive the monitoring system and personnel to respond.

[0063] Specifically, the system first continuously monitors coupled risk indicators. When the value of an indicator exceeds a preset risk threshold for the first time, an early warning generation process is triggered. A pre-stored risk level mapping table maps the value range of the coupled risk indicator to different risk levels. For example, this mapping relationship can be designed such that when the indicator value exceeds the threshold but does not reach twice the threshold, it corresponds to a "low-level (blue) warning"; when the indicator value reaches more than twice the threshold but not more than four times the threshold, it corresponds to a "medium-level (yellow) warning"; and when the indicator value reaches more than four times the threshold, it corresponds to a "high-level (red) warning". By looking up the table and comparing the current value of the coupled risk indicator, the risk level of this warning can be quickly and objectively determined.

[0064] Then, based on the physical location of the node that generated the anomaly detection, the risk area is estimated. All relevant data that caused the coupling risk index to exceed the limit is traced back, specifically the coordinates of the nodes at both ends of the abnormal communication link that contributed to the effective first sensing value, and the coordinates of the abnormal nodes that contributed to the effective second sensing value. The module uses these spatial point sets as input to perform spatial aggregation calculations. For example, a density-based spatial clustering algorithm is used to find one or more clusters with the highest aggregation degree among these spatial points. Then, the outer boundary of the spatial distribution of each point cluster is calculated, for example, by calculating the two-dimensional convex hull or the three-dimensional minimum bounding box of the point cluster. The spatial range defined by this calculated geometry is estimated as the risk area. This process transforms discrete sensor alarm points into a continuous, visualized danger area.

[0065] Finally, the above results are packaged into a standard format warning instruction. According to the predetermined data protocol, the determined risk level code and the estimated geometric coordinates of the risk area (e.g., in the form of a polygon vertex sequence or center point plus radius) are packaged into a data packet. At the same time, the instruction packet can also be supplemented with auxiliary information such as timestamps and trigger indicator values. After generation, the instruction is sent to the mine's centralized monitoring platform to light up the corresponding alarm icon, highlight the risk area in the digital twin model, and push it to relevant safety management personnel through sound and light alarms, SMS platforms, etc.

[0066] It is worth noting that by providing clear risk levels and spatial locations, the embodiments of the present invention enable targeted emergency responses, thereby improving the practical value of early warning and the efficiency of disaster relief.

[0067] In summary, the intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling provided by this invention can form multi-dimensional monitoring information by simultaneously extracting the signal propagation anomaly characteristics of the original network status data and the combustion physicochemical anomaly characteristics of the environmental physical quantity data. This effectively overcomes the shortcomings of single-parameter threshold alarms, such as susceptibility to environmental interference, high false alarm rate, and high risk of missed alarms, thereby improving the reliability of early warning. By performing spatiotemporal correlation calculations on the two types of sensing quantities to obtain coupled risk indicators and trigger early warnings, it can capture early abnormal signs before significant temperature rise in the thermal accumulation zone and the formation of open combustion, thereby improving early sensing capabilities and the operability of early warning, and thus ensuring safe production in the mine.

[0068] Based on the above method items, the present invention provides corresponding system items embodiments.

[0069] See Figure 2 , Figure 2 This is a structural block diagram of an intelligent early warning system for fire prevention and extinguishing in coal-based solid waste backfilling, provided in an embodiment of the present invention. The intelligent early warning system for fire prevention and extinguishing in coal-based solid waste backfilling includes: The data acquisition module 21 is used to acquire environmental physical quantity data and network status native data of the filling area through a wireless mesh network; wherein, the wireless mesh network includes anchoring nodes and drifting nodes; The first sensing module 22 is used to perform signal propagation analysis on the original network state data, extract abnormal signal propagation features caused by environmental medium disturbances, and obtain the first sensing quantity. The second sensing module 23 is used to perform combustion analysis on the environmental physical quantity data, extract physicochemical anomaly features related to the combustion process, and obtain the second sensing quantity; The correlation calculation module 24 is used to perform spatiotemporal correlation calculation on the first sensing quantity and the second sensing quantity within the same spatiotemporal range to obtain a coupling risk index; The early warning module 25 is used to generate an early warning command when the coupled risk indicator exceeds a preset risk threshold.

[0070] In one alternative embodiment, the first sensing module 22 is configured to: From the raw network status data, obtain the link quality indicator value sequence and the received signal strength indicator value sequence corresponding to the target communication link; Trend analysis is performed on the link quality indicator sequence and the received signal strength indicator sequence to extract persistent degradation features not caused by network faults; The persistent degradation characteristics are quantified into an index characterizing the degree of thermal disturbance of the medium, thus obtaining the first sensing quantity.

[0071] In one alternative embodiment, the second sensing module 23 is configured to: Combustion analysis was performed on the environmental physical quantity data to calculate the temperature change rate and characteristic gas concentration ratio of the target node; A comprehensive analysis was performed on the temperature change rate and the ratio of the characteristic gas concentrations to obtain a comprehensive analysis result. When the temperature change rate is greater than a preset change rate threshold and the characteristic gas concentration ratio is greater than a preset ratio threshold, the comprehensive analysis result is quantified into an index characterizing the intensity of the combustion reaction, thus obtaining a second sensing quantity.

[0072] In one optional embodiment, the association calculation module 24 includes: The consistency judgment unit is used to determine whether the first perceived quantity and the second perceived quantity are consistent in spatial location within a preset time and space window; The correlation calculation unit is used to calculate the joint anomaly strength based on the first sensing quantity and the second sensing quantity if there is consistency, and quantify the joint anomaly strength into a coupling risk index. The early warning monitoring unit is used to prevent early warning from being triggered if there is inconsistency.

[0073] In one optional embodiment, the consistency determination unit is configured to: Within a preset time-space window, the spatial path of the target communication link that causes the first sensing quantity to be abnormal is determined, and a first spatial influence range is constructed along the spatial path. Within a preset spatiotemporal window, determine the spatial coordinates of the target node for generating the second sensing quantity, and construct a second spatial influence range centered on the target node; When the first spatial influence range and the second spatial influence range intersect in space, it is determined that the first perceived quantity and the second perceived quantity are consistent in spatial location.

[0074] In one alternative embodiment, the associated calculation unit is configured to: Calculate the spatial density of all abnormal links corresponding to the first sensing quantity; Calculate the spatial cluster centers and dispersion of all abnormal nodes corresponding to the second sensing quantity; Based on the spatial density, the spatial cluster centers, and the dispersion, a comprehensive factor characterizing the spatial focusing degree of abnormal events is calculated. The combined factor is weighted and fused with the intensities of the first and second perceived quantities to obtain the joint anomaly intensity.

[0075] It should be noted that the intelligent early warning system for fire prevention and extinguishing of coal-based solid waste filling provided in this embodiment of the invention is used to execute all the process steps of the intelligent early warning method for fire prevention and extinguishing of coal-based solid waste filling in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0076] This invention also provides a terminal device, such as... Figure 3 The diagram shown is a structural block diagram of a preferred embodiment of a terminal device provided by the present invention. The terminal device includes a processor 31, a memory 32, and a computer program stored in the memory 32 and configured to be executed by the processor 31. When the processor 31 executes the computer program, it implements the intelligent early warning method for fire prevention and extinguishing of coal-based solid waste backfilling as described in any of the above embodiments.

[0077] In addition, embodiments of the present invention also provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the intelligent early warning method for fire prevention and extinguishing of coal-based solid waste backfill as described in any of the above embodiments.

[0078] When the processor 31 executes the computer program, it implements the steps in the above-described embodiment of the intelligent early warning method for fire prevention and extinguishing of coal-based solid waste backfilling, for example... Figure 1 The method illustrates all steps of the intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling. Alternatively, when the processor 31 executes the computer program, it implements the functions of each module in the above-described intelligent early warning system embodiment for fire prevention and extinguishing in coal-based solid waste backfilling, for example... Figure 2 The diagram shows the functions of each module in the intelligent early warning system for fire prevention and extinguishing in coal-based solid waste backfilling.

[0079] Preferably, the computer program can be divided into one or more modules / units, which are stored in the memory 32 and executed by the processor 31 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device.

[0080] The processor 31 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor 31 can be any conventional processor. The processor 31 is the control center of the terminal device, connecting various parts of the terminal device through various interfaces and lines.

[0081] The memory 32 mainly includes a program storage area and a data storage area. The program storage area can store the operating system, applications required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory 32 can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, and a flash card, etc., or the memory 32 can also be other volatile solid-state storage devices.

[0082] It should be noted that the aforementioned terminal devices may include, but are not limited to, processors and memory, as will be understood by those skilled in the art. Figure 3 The structural block diagram shown is merely a structural example of the terminal device described above and does not constitute a limitation on the structure of the terminal device. The terminal device may include more or fewer components than shown, or combine certain components, or use different components.

[0083] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A smart early warning method for fire prevention and extinguishing in coal-based solid waste backfilling, characterized in that, include: Environmental physical quantity data and native network status data of the filling area are acquired through a wireless mesh network; wherein, the wireless mesh network includes anchoring nodes and drifting nodes; The network state raw data is analyzed for signal propagation, and abnormal signal propagation features caused by environmental medium disturbances are extracted to obtain the first sensing quantity. Combustion analysis is performed on the environmental physical quantity data to extract physicochemical anomaly features related to the combustion process, thereby obtaining the second sensing quantity; The first and second sensed quantities within the same spatiotemporal range are spatiotemporally correlated to obtain a coupling risk index. When the coupling risk index exceeds a preset risk threshold, an early warning instruction is generated.

2. The intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling as described in claim 1, characterized in that, The anchoring node is fixed to the surrounding rock structure of the filling area, and the drifting node is put into the filling area along with the filling material according to a preset deployment strategy.

3. The intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling as described in claim 1, characterized in that, The step of performing signal propagation analysis on the raw network state data to extract abnormal signal propagation features caused by environmental medium disturbances and obtain the first sensing quantity includes: From the raw network status data, obtain the link quality indicator value sequence and the received signal strength indicator value sequence corresponding to the target communication link; Trend analysis is performed on the link quality indicator sequence and the received signal strength indicator sequence to extract persistent degradation features not caused by network faults; The persistent degradation characteristics are quantified into an index characterizing the degree of thermal disturbance of the medium, thus obtaining the first sensing quantity.

4. The intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling as described in claim 1, characterized in that, The combustion analysis of the environmental physical quantity data, extracting physicochemical anomaly features related to the combustion process, yields a second sensing quantity, including: Combustion analysis was performed on the environmental physical quantity data to calculate the temperature change rate and characteristic gas concentration ratio of the target node; A comprehensive analysis was performed on the temperature change rate and the ratio of the characteristic gas concentrations to obtain a comprehensive analysis result. When the temperature change rate is greater than a preset change rate threshold and the characteristic gas concentration ratio is greater than a preset ratio threshold, the comprehensive analysis result is quantified into an index characterizing the intensity of the combustion reaction, thus obtaining a second sensing quantity.

5. The intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling as described in claim 1, characterized in that, The step of performing spatiotemporal correlation calculations on the first and second sensed quantities within the same spatiotemporal range to obtain a coupling risk index includes: Within a preset time-space window, determine whether the first perceived quantity and the second perceived quantity are consistent in spatial location; If there is consistency, the joint anomaly strength is calculated based on the first and second perceived quantities, and the joint anomaly strength is quantified as a coupled risk index. If there is no consistency, no warning will be triggered.

6. The intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling as described in claim 5, characterized in that, The step of determining whether the first perceived quantity and the second perceived quantity are consistent in spatial location within a preset spatiotemporal window includes: Within a preset time-space window, the spatial path of the target communication link that causes the first sensing quantity to be abnormal is determined, and a first spatial influence range is constructed along the spatial path. Within a preset spatiotemporal window, determine the spatial coordinates of the target node for generating the second sensing quantity, and construct a second spatial influence range centered on the target node; When the first spatial influence range and the second spatial influence range intersect in space, it is determined that the first perceived quantity and the second perceived quantity are consistent in spatial location.

7. The intelligent early warning method for fire prevention and extinguishing in coal-based solid waste backfilling as described in claim 5, characterized in that, The calculation of the joint anomaly intensity based on the first and second sensing values ​​includes: Calculate the spatial density of all abnormal links corresponding to the first sensing quantity; Calculate the spatial cluster centers and dispersion of all abnormal nodes corresponding to the second sensing quantity; Based on the spatial density, the spatial cluster centers, and the dispersion, a comprehensive factor characterizing the spatial focusing degree of abnormal events is calculated. The combined factor is weighted and fused with the intensities of the first and second perceived quantities to obtain the joint anomaly intensity.

8. A smart early warning system for fire prevention and extinguishing in coal-based solid waste backfilling, characterized in that, include: The data acquisition module is used to acquire environmental physical quantity data and network status native data of the filling area through a wireless mesh network; wherein, the wireless mesh network includes anchoring nodes and drifting nodes; The first sensing module is used to perform signal propagation analysis on the raw network state data, extract abnormal signal propagation features caused by environmental medium disturbances, and obtain the first sensing quantity. The second sensing module is used to perform combustion analysis on the environmental physical quantity data, extract physicochemical anomaly features related to the combustion process, and obtain the second sensing quantity; The correlation calculation module is used to perform spatiotemporal correlation calculation on the first sensing quantity and the second sensing quantity within the same spatiotemporal range to obtain a coupling risk index. The early warning module is used to generate an early warning command when the coupled risk indicator exceeds a preset risk threshold.

9. A terminal device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the intelligent early warning method for fire prevention and extinguishing of coal-based solid waste backfill as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the intelligent early warning method for fire prevention and extinguishing of coal-based solid waste backfill as described in any one of claims 1 to 7.

Images