A perimeter alarm false alarm filtering control method and system

Abstract

The perimeter alarm false alarm filtering control method and system provided by the application relate to the technical field of perimeter security, after original sampling data corresponding to a plurality of continuous sampling points of the region perimeter are collected and acquired, twice of judgment are performed to screen out sampling data with both instantaneous impact characteristics and tension sustained jump characteristics, after the screening is completed, actual tension attenuation data sequences in a preset time period are collected, the actual tension attenuation data sequences are compared with theoretical tension attenuation data sequences, and corresponding operations are performed according to the comparison result, the application effectively distinguishes non-intrusion factors such as thermodynamic effect and real intrusion behavior, so that the application has the advantages that tension changes caused by thermodynamic effect and intrusion behavior can be effectively distinguished, and false alarms caused by environmental temperature difference and other non-intrusion factors are significantly reduced.

Description

Technical Field

[0001] This invention relates to the field of perimeter security technology, and in particular to a perimeter alarm false alarm filtering and control method and system. Background Technology

[0002] In critical infrastructure security systems, electronic tension fences are widely used as an important perimeter alarm device. This device detects intrusion by monitoring changes in the tension of steel wire ropes deployed along the boundary. When the wire ropes are climbed, sheared, or pulled, the tension sensors detect abnormal signal changes and trigger an alarm. Existing tension alarm systems typically possess some signal processing capabilities, capable of filtering out overall tension drift caused by factors such as slow changes in ambient temperature or continuous light winds, thus reducing false alarm rates under normal weather conditions. However, in certain specific applications, conventional filtering mechanisms still face challenges, especially when dealing with complex signals generated by non-intrusive human activities.

[0003] For example, at a critical infrastructure site located next to a large reservoir, the core of its security system is a highly sensitive electronic tension fence. This system stretches for several kilometers along the plant boundary and consists of multiple parallel, taut steel cables, tension sensors fixed to posts, and a central signal processing unit at the rear. Under normal alert conditions, each steel cable maintains a precise tension reference value. To ensure the long-term reliability of the system, the plant regularly uses mobile high-pressure water guns to wash and clean the fence posts and steel cables. During planned maintenance, operators pre-set the work zone to "maintenance mode," in which the system temporarily ignores specific signals to avoid generating invalid alarms. However, in a special and unforeseen situation, the system's limitations become apparent. Imagine an autumn afternoon when a tanker truck transporting chemicals experiences a minor leak within the plant, splashing a small amount of liquid near a section of the fence. For safety reasons, a maintenance worker must immediately perform emergency cleaning of the leaked area. Without time to request maintenance mode from the security center beforehand, he directly activates a nearby high-pressure water gun to flush the area. The water plant is unique in that its clean water is directly drawn from a nearby reservoir. In autumn, the surface temperature of the steel wire rope exposed to sunlight during the day can still remain above 20 degrees Celsius, while the water temperature in the reservoir has dropped to around 8 degrees Celsius, creating a significant temperature difference of over 10 degrees Celsius. When this high-pressure, low-temperature water flow impacts the steel wire rope of the fence, which is in a standard alert state, the central signal processing unit receives an extremely complex set of composite signals. First, the enormous kinetic energy of the water flow produces violent, high-frequency instantaneous vibrations, the signal characteristics of which are similar to those of a tool striking the fence. Second, and more crucially, the 8-degree Celsius water flow spraying onto the steel wire rope, which is over 20 degrees Celsius, causes a drastic local temperature drop. The "cold contraction" effect of the metal is instantly triggered, causing the impacted section of the steel wire rope to contract rapidly, resulting in a sudden and significant increase in tension, which remains high for several seconds afterward. The signal pattern is almost identical to that of an intruder using a jack or other tools to continuously pull on the steel wire rope.

[0004] Existing alarm systems, when analyzing this combination of "impact followed by sustained pressure," highly correlate it with a highly destructive violent intrusion method (such as the use of heavy shearing tools), because this intrusion also produces the signal characteristics of impact and sustained pressure. The system cannot distinguish this coincidental superposition of "kinetic impact" and "thermal impact" caused by unplanned cleaning operations from a genuine violent intrusion, ultimately leading to the highest level of false alarms, wasting security resources and causing unnecessary stress.

[0005] Therefore, providing a perimeter alarm false alarm filtering control method and system is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a perimeter alarm false alarm filtering control method, which has the advantages of effectively distinguishing tension changes caused by thermodynamic effects from intrusion behavior and significantly reducing false alarms caused by non-intrusive factors such as environmental temperature differences.

[0007] Based on the above objectives, the technical solution provided by the present invention is as follows:

[0008] A perimeter alarm false alarm filtering and control method, applied to an electronic tension alarm system for intrusion detection, includes the following steps:

[0009] Obtain the raw sampling data corresponding to multiple consecutive sampling points at the perimeter of the suspected area;

[0010] The first determination is whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold. If the first determination result is yes, the second determination is whether the original sampling data after moving average processing jumps above the high tension alarm threshold within a preset time period and remains at a high level. If the second determination result is yes, it is defined as a high-risk event to be confirmed.

[0011] Obtain the actual tension decay data sequence of the original sampled data after moving average processing within the preset time period;

[0012] The third step is to determine whether the actual tension decay data sequence matches the preset theoretical tension decay sequence.

[0013] Based on the third judgment result, execute the corresponding false alarm filtering operation or alarm operation.

[0014] The above scheme can effectively distinguish between tension changes caused by thermodynamic effects and intrusion behavior, and significantly reduce false alarms caused by non-intrusive factors such as environmental temperature differences.

[0015] Preferably, the first determination of whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold, if the first determination result is yes, includes the following steps:

[0016] Calculate the difference between the original sampled data corresponding to two adjacent consecutive sampled points;

[0017] Within a first preset window length, the first determination is whether the difference between the original sampling data corresponding to two adjacent consecutive sampling points is greater than the tension change threshold.

[0018] If the first judgment result is yes, then the existence of rapid impact characteristics is defined.

[0019] Preferably, the second determination of whether the original sampled data after moving average processing jumps above the high tension alarm threshold and remains at a high level within a preset time period, if the second determination result is yes, is defined as a high-risk event to be confirmed, including the following steps:

[0020] Based on the second preset window length, the original sampled data is processed by moving average to obtain the quasi-static tension value;

[0021] Within a preset time period, the second determination is whether the quasi-static tension value jumps above the high tension alarm threshold and remains at a high level.

[0022] If the second judgment result is yes, then the existence of a continuous tension jump characteristic is defined;

[0023] When both the rapid impact characteristic and the continuous increase in tension are present, it is defined as a high-risk event to be confirmed.

[0024] By introducing a two-step judgment mechanism, the above scheme can filter out data that meets the conditions from the original sampled data, providing a basis for subsequent signal judgment and operation.

[0025] Preferably, the third step of determining whether the actual tension decay data sequence matches the preset theoretical tension decay sequence includes the following steps:

[0026] Iterate through each data point in the actual tension decay data sequence and calculate the square of the difference between the actual tension decay data and the theoretical tension decay data for each data point.

[0027] By summing the squares of the differences at each data point, we obtain the sum of squared residuals of the tension decay data.

[0028] Third, determine whether the sum of squared residuals of the tension decay data is less than a preset threshold for the sum of squared residuals;

[0029] The theoretical tension decay data for each data point is obtained based on a preset theoretical tension decay sequence.

[0030] The above method further determines whether the actual tension decay data sequence matches the preset theoretical tension decay sequence in the thermodynamic model. Based on the matching result, it is determined whether the event conforms to the physical recovery law of thermal shock, thereby improving the accuracy of identification.

[0031] Preferably, the step of performing the corresponding false alarm filtering operation or alarm operation based on the third judgment result includes the following steps:

[0032] If the third judgment result is yes, then the actual tension decay data sequence is defined to match the preset theoretical tension decay sequence, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed.

[0033] If the third judgment result is negative, then the actual tension attenuation data sequence is defined as not matching the preset theoretical tension attenuation sequence, and the high-risk event to be confirmed will be archived as a real intrusion event, and an alarm operation will be performed.

[0034] The above approach further determines whether the event to be confirmed is an identified interference event or a real intrusion event based on the matching results, and performs corresponding operations, thereby improving the accuracy of event analysis.

[0035] Preferably, after defining the actual tension decay data sequence as not matching the preset theoretical tension decay sequence if the third judgment result is negative, the following steps are further included:

[0036] Fourth, determine whether there is a macroscopic attenuation trend in the actual tension attenuation data sequence. If the result of the fourth determination is yes, then the fifth determination is whether the macroscopic attenuation process is stable. If the result of the fifth determination is yes, then the high-risk event to be confirmed is archived as an identified interference event and a false alarm filtering operation is performed.

[0037] If either the fourth or fifth judgment result is negative, the high-risk event to be confirmed will be archived as a real intrusion event, and an alarm operation will be performed.

[0038] Preferably, the fourth determination of whether the actual tension attenuation data sequence has a macroscopic attenuation trend, if the result of the fourth determination is yes, includes the following steps:

[0039] All data points in the actual tension decay data sequence are divided into a first part of data points and a second part of data points.

[0040] The average value of the first actual tension attenuation data corresponding to the first part of data points and the average value of the second actual tension attenuation data corresponding to the second part of data points are obtained respectively.

[0041] Fourth, determine whether the difference between the average value of the first actual tension attenuation data and the average value of the second actual tension attenuation data is greater than a preset average difference threshold.

[0042] If the fourth judgment result is yes, then the actual tension decay data sequence is defined to have a macroscopic decay trend.

[0043] Preferably, the fifth determination is whether the macroscopic attenuation process is stable. If the fifth determination result is yes, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed, including the following steps:

[0044] The average value of the absolute difference between the actual tension decay data corresponding to two consecutive adjacent sampling points in all data points of the actual tension decay data sequence is obtained.

[0045] Fifth, determine whether the average of the absolute values ​​of the differences falls within the preset stable range.

[0046] If the result of the fifth judgment is yes, then the macroscopic decay process is defined as stationary;

[0047] When the actual tension decay data sequence shows a macroscopic decay trend and the macroscopic decay process is stable, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed.

[0048] By introducing the above steps, macroscopic attenuation trend and macroscopic attenuation stability, it is possible to further determine whether it is a real intrusion signal or a composite false signal, thereby effectively solving the problem of event recognition accuracy in the special scenario where the two effects of "thermal shock contraction" and "peeling and emulsification of surface adhesive rust inhibitor layer" are superimposed due to water flow impact.

[0049] Preferably, after obtaining the original sampled data after moving average processing and the actual tension decay data sequence over the preset time period, the method further includes the following steps:

[0050] By iterating through each data point of the actual tension decay data sequence, the second derivative of the actual tension decay data corresponding to each data point is calculated to obtain the tension change acceleration sequence.

[0051] Obtain the variance value of the tension change acceleration sequence;

[0052] The sixth step is to determine whether the variance of the tension change acceleration sequence is less than a preset smoothing threshold.

[0053] Based on the sixth judgment result, execute the corresponding false alarm filtering operation or alarm operation.

[0054] By introducing a smoothing threshold through the above steps, signal characteristics can be identified more precisely, and non-invasive behaviors such as crystallization processes can be distinguished from invasive behaviors.

[0055] A perimeter alarm false alarm filtering and control system includes:

[0056] The data acquisition module is used to acquire the raw sampling data corresponding to multiple consecutive sampling points at the perimeter of the suspected area;

[0057] The module for defining high-risk events to be confirmed is used to first determine whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold. If the first determination result is yes, then the module for second determination is whether the original sampling data after moving average processing jumps above the high tension alarm threshold within a preset time period and remains at a high level. If the second determination result is yes, it is defined as a high-risk event to be confirmed.

[0058] The actual tension decay data sequence acquisition module is used to acquire the actual tension decay data sequence of the original sampled data after moving average processing within the preset time period.

[0059] The sequence matching judgment module is used to make a third judgment on whether the actual tension decay data sequence matches the preset theoretical tension decay sequence;

[0060] The execution module is used to perform corresponding false alarm filtering or alarm operations based on the third judgment result.

[0061] The perimeter alarm false alarm filtering control method provided by this invention effectively distinguishes between non-invasive factors such as thermodynamic effects and actual intrusion behavior by analyzing the attenuation characteristics of composite tension signals and comparing them with theoretical models. Thus, it has the advantage of effectively distinguishing tension changes caused by thermodynamic effects from intrusion behavior and significantly reducing false alarms caused by non-invasive factors such as environmental temperature differences.

[0062] The present invention also provides a perimeter alarm false alarm filtering control system. Since it solves the same technical problem as the method and belongs to the same technical concept, it should have the same beneficial effects, and will not be described in detail here. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 A flowchart of a perimeter alarm false alarm filtering control method provided in an embodiment of the present invention;

[0065] Figure 2 This is a flowchart of the first determination in step S2 provided in an embodiment of the present invention;

[0066] Figure 3 This is a flowchart of the second determination in step S2 provided in an embodiment of the present invention;

[0067] Figure 4 A flowchart of step S4 provided in an embodiment of the present invention;

[0068] Figure 5 A flowchart for a specific scenario of rust inhibitor emulsification provided in an embodiment of the present invention;

[0069] Figure 6 A flowchart for determining the existence of a macroscopic attenuation trend provided in an embodiment of the present invention;

[0070] Figure 7 A flowchart for judging the smoothness of the macroscopic attenuation process provided in an embodiment of the present invention;

[0071] Figure 8 A flowchart provided for embodiments of the invention, illustrating the application of heat transfer mode changes in the dissolution and endothermic reaction of surface deposits and subsequent solution evaporation.

[0072] Figure 9 This is a structural diagram of a perimeter alarm false alarm filtering control system provided in an embodiment of the present invention. Detailed Implementation

[0073] 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.

[0074] The embodiments of this invention are written in a progressive manner.

[0075] Traditional electronic tension fences, when monitoring perimeter tension changes to detect intrusion, typically have the ability to filter out overall tension drift caused by slow changes in ambient temperature or continuous light winds. However, existing filtering mechanisms have limitations when dealing with complex signals generated by non-invasive human activities. Specifically, effectively distinguishing between a genuine, violent intrusion characterized by impact and continuous pressure, and a composite false signal resulting from the combined physical effects of kinetic energy impact vibration and localized thermal shock contraction caused by unplanned high-pressure, low-temperature fluid impacting a high-temperature steel wire rope, is a current technical challenge. This insufficient ability to differentiate between them affects the signal processing accuracy and false alarm control performance of perimeter alarm systems.

[0076] For example, suppose a water treatment plant uses a high-sensitivity electronic tension fence as the core of its security system, consisting of steel wire ropes deployed along the boundary, tension sensors, and a central signal processing unit. Under normal alert conditions, the steel wire ropes maintain a precise tension reference value. When an unplanned emergency cleaning operation is carried out in the plant area, using high-pressure, low-temperature water to impact the high-temperature fence steel wire ropes under standard alert conditions, the central signal processing unit will receive a composite signal. The kinetic energy impact of the water flow generates an instantaneous vibration signal, and the low-temperature water sprayed onto the high-temperature steel wire rope causes a violent local temperature drop, resulting in a rapid contraction of the steel wire rope and a sudden, significant increase in tension value, which remains high for several seconds. This combination of initial impact followed by continuous pressurization is highly similar to the signal characteristics generated by violent intrusion in existing systems. Existing systems cannot distinguish this composite false signal formed by the superposition of kinetic and thermal impacts from genuine violent intrusion signals.

[0077] If the aforementioned issues are not addressed, the perimeter alarm system will be unable to accurately identify the signal source, leading to the misinterpretation of composite signals generated by non-intrusive activities as actual intrusions, triggering the highest level of alarm. This will waste security resources, increase unnecessary emergency response burdens, and potentially reduce the reliability of the perimeter alarm system in the event of a real intrusion. Such false alarms can occur frequently under specific environmental conditions (such as high-pressure fluid impacts with temperature differences), severely impacting the system's practicality and reliability.

[0078] Faced with the aforementioned problems, this application initially considered distinguishing the signal source by analyzing the instantaneous impact amplitude or the peak value of sustained pressure. However, real violent intrusions and specific non-invasive operations can both produce similar impact amplitudes and sustained pressure peak values, rendering this method ineffective. Therefore, this application further considers that although the instantaneous impact and peak value of the signal are similar, the attenuation process after reaching the peak may differ. The physical mechanisms of recovery or attenuation differ between tension changes caused by thermodynamic effects and those caused by mechanical forces. The contraction recovery process caused by thermodynamic effects should follow specific physical laws. Therefore, this application proposes an approach that analyzes the attenuation characteristics of a composite tension signal after reaching its peak and compares it with theoretical models characterizing specific physical effects to determine the true source of the signal.

[0079] like Figure 1 As shown, a perimeter alarm false alarm filtering and control method is applied to an electronic tension alarm system for intrusion detection, and includes the following steps:

[0080] S1. Obtain the original sampling data corresponding to multiple consecutive sampling points at the perimeter of the suspected area;

[0081] S2. First, determine whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold. If the first determination result is yes, then the second determination is whether the original sampling data after moving average processing jumps above the high tension alarm threshold within a preset time period and remains at a high level. If the second determination result is yes, it is defined as a high-risk event to be confirmed.

[0082] S3. Obtain the actual tension decay data sequence of the original sampled data after moving average processing within a preset time period;

[0083] S4. Third, determine whether the actual tension decay data sequence matches the preset theoretical tension decay sequence;

[0084] S5. Based on the third judgment result, perform the corresponding false alarm filtering operation or alarm operation.

[0085] The technical problem to be solved by this application is how to effectively distinguish between a real violent intrusion behavior that combines the characteristics of impact and continuous pressure, and a composite false signal formed by the superposition of two physical effects, namely "kinetic energy impact vibration" and "local thermal impact contraction", caused by unplanned high-pressure low-temperature fluid impacting a high-temperature steel wire rope.

[0086] This application proposes a perimeter alarm false alarm filtering control method. After acquiring the original sampling data corresponding to multiple continuous sampling points of the perimeter of the area, two judgments are made to filter out the sampling data that has both instantaneous impact characteristics and tension continuous jump characteristics. After the filtering is completed, the actual tension decay data sequence within a preset time period is collected. The actual tension decay data sequence is compared with the theoretical tension decay data sequence, and the corresponding operation is performed according to the comparison result.

[0087] Among them, the sampling data that combines instantaneous impact characteristics and continuous tension rise characteristics refers to the monitoring of tension changes in tension cables by devices such as tension sensors. When the detected signal simultaneously contains short-term, high-amplitude fluctuations (instantaneous impact characteristics) and a phenomenon in which the tension value increases significantly and remains at a high level for a period of time (continuous tension rise characteristics), the signal is identified and captured. The purpose is to identify abnormal tension events that may be caused by intrusion or specific environmental factors. Collecting the actual tension decay data sequence within a preset time period means that after capturing the composite tension signal and its tension reaches its maximum point, continuous data points are recorded to show how the tension value changes over time, forming a data sequence that reflects the gradual decrease in tension. The purpose is to obtain the dynamic change characteristics after the signal peak, providing a basis for subsequent signal source determination. Comparing the actual tension decay data sequence with the theoretical tension decay data sequence means comparing the actual measured tension decay data curve with a theoretical tension change curve based on physical principles (such as the temperature-tension relationship of metallic materials, thermal conductivity, etc.) that describes the recovery of the tension cable to its original state after cooling and contraction as the temperature rises. The purpose is to quantify the degree of matching between the measured decay process and the thermodynamic recovery process. Performing corresponding operations based on the matching comparison results means determining the main cause of the composite tension signal based on the degree of matching between the measured data and the theoretical model. This determines whether the tension change is caused by thermodynamic effects or by other reasons (such as physical pulling). The purpose is to distinguish signals generated by different physical mechanisms. If the main cause of the composite tension signal is a tension change caused by thermodynamic effects, a conventional intrusion alarm will not be triggered, or the event will be marked as a non-intrusion event. If the tension change is caused by other reasons (such as physical pulling), an alarm will be triggered. The purpose is to avoid false alarms caused by thermodynamic effects.

[0088] The core innovation of this application lies in the fact that by collecting measured tension attenuation data after the peak of the composite tension signal and comparing it with a preset theoretical tension recovery model, the main cause of the composite tension signal can be determined based on the comparison results, thus effectively distinguishing between false alarms caused by thermodynamic effects and real intrusion signals.

[0089] In some preferred embodiments, tension signals are captured by piezoelectric sensors or strain gauge sensors mounted on the tension cable. After amplification, filtering, and digitization, the signals are analyzed by an embedded processor. When a rapid increase in signal amplitude accompanied by sustained high tension is detected, the processor records a sequence of tension values ​​after the signal reaches its peak. The processor internally stores a theoretical tension recovery model based on parameters such as the tension cable material properties, coefficient of linear expansion, and thermal conductivity. This model can be a mathematical function or a lookup table. The processor performs least-squares fitting or correlation analysis on the collected measured tension decay data points and the theoretical model curve to calculate the goodness-of-fit index. For example, if the goodness-of-fit exceeds a preset threshold (e.g., a correlation coefficient greater than 0.9), the processor determines that the signal originates from a thermodynamic effect and sends a filtering command to the alarm control unit to prevent the generation of an alarm event.

[0090] The above technical solution effectively identifies and filters out false alarms caused by composite tension signals resulting from non-invasive thermodynamic effects (such as high-pressure, low-temperature water flow impacting high-temperature tension cables). This method analyzes the attenuation characteristics after the signal peak and compares them with a theoretical model of thermodynamic recovery, providing a basis for distinguishing between genuine and false intrusion signals. This improves the perimeter alarm system's resistance to false alarms, especially in application scenarios with specific environmental interference factors, ensuring the reliability and accuracy of the alarm system.

[0091] like Figure 2 As shown, preferably, the first determination is whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold. If the first determination result is yes, the following steps are included:

[0092] A1. Calculate the difference between the original sampled data corresponding to two adjacent consecutive sampling points;

[0093] A2. Within the first preset window length, the first judgment is made as to whether the difference between the original sampling data corresponding to two adjacent consecutive sampling points is greater than the tension change threshold.

[0094] A3. If the first judgment result is yes, then the existence of rapid impact characteristics is defined.

[0095] like Figure 3 As shown, preferably, the second determination involves whether the original sampled data after moving average processing jumps above the high-tension alarm threshold and remains at a high level within a preset time period. If the second determination result is yes, it is defined as a high-risk event to be confirmed, including the following steps:

[0096] B1. Based on the second preset window length, perform moving average processing on the original sampling data to obtain the quasi-static tension value;

[0097] B2. Within a preset time period, whether the second quasi-static tension value jumps above the high tension alarm threshold and remains at a high level;

[0098] B3. If the second judgment result is yes, then the existence of a continuous tension jump characteristic is defined;

[0099] B4. When both rapid impact characteristics and sustained tension increase characteristics are present, it is defined as a high-risk event to be confirmed.

[0100] Steps A1 to A3 and B1 to B3 of this application are specific implementation details of step S2: within a first preset window length, it is determined whether the difference between the original sampling data corresponding to two adjacent consecutive sampling points is greater than the tension change threshold, thus defining the existence of rapid impact characteristics, and initially screening the sampling data that meets the first judgment condition from the original sampling data; secondly, it is determined whether the quasi-static tension value jumps above the high tension alarm threshold and remains at a high level within a preset time period, thus defining the existence of tension continuous jump characteristics, and further screening the sampling data that meets the second judgment condition from the initially screened sampling data as sampling data that has both instantaneous impact characteristics and tension continuous jump characteristics.

[0101] In some preferred embodiments, determining the presence of a rapid impact characteristic involves calculating the rate of change of the signal by measuring the first-order difference between consecutive sampling points. If the corresponding tension change exceeds 50 Newtons within a 10-millisecond window, a rapid instantaneous impact component is identified. Determining the presence of a sustained tension increase characteristic involves applying a 100-millisecond moving average to the original sampled data to obtain a smoothed quasi-static tension value. If, within 500 milliseconds after identifying the impact component, the quasi-static tension value increases beyond the 200-Newton high-tension alarm threshold, a sustained tension increase characteristic is identified. When both the rapid impact characteristic and the sustained tension increase characteristic exist, i.e., after filtering out sampled data that simultaneously meet the above judgment conditions, the event is defined as a high-risk event to be confirmed. At this time, the microcontroller does not immediately drive the alarm relay, but instead sets an internal flag to temporarily suspend the issuance of external alarm commands and starts the "thermal characteristic attenuation analysis program".

[0102] The above technical solution can effectively filter out sampling data that has both instantaneous impact characteristics and continuous tension increase characteristics from the original sampling data, and define the event as a high-risk event to be confirmed, providing a basis for subsequent judgment of distinguishing between real intrusion signals and false signals, and further ensuring the reliability and accuracy of performing corresponding false alarm filtering operations or alarm operations.

[0103] like Figure 4 As shown, preferably, the third step of determining whether the actual tension decay data sequence matches the preset theoretical tension decay sequence includes the following steps:

[0104] C1. Traverse each data point in the actual tension decay data sequence and calculate the square of the difference between the actual tension decay data and the theoretical tension decay data for each data point;

[0105] C2. Sum the squares of the differences at each data point to obtain the residual sum of squares of the tension decay data;

[0106] C3. The third determination is whether the sum of squared residuals of the tension decay data is less than the preset threshold for the sum of squared residuals;

[0107] The theoretical tension decay data for each data point is obtained based on a preset theoretical tension decay sequence.

[0108] Steps C1 to C3 of this application are specific implementation details of step S4: by traversing each data point of the actual tension decay data sequence, calculating the square of the difference between the actual tension decay data and the theoretical tension decay data for each data point, accumulating the square of the difference for each data point, obtaining the residual sum of squares of the tension decay data, and determining whether the residual sum of squares is less than a preset residual sum of squares threshold.

[0109] The actual tension decay data sequence refers to the measured tension decay data sequence consisting of 50 data points obtained by continuously acquiring the quasi-static tension value at 100-millisecond intervals for a total of 5 seconds using a microcontroller. This sequence is then stored in the microcontroller's memory array. Theoretical tension decay data refers to a mathematical model describing the thermodynamic recovery process. This model can be specifically defined as an exponential decay function. .in, For time The theoretical tension value at time t. This represents the measured peak tension. The initial tension value before the event occurs, and the attenuation constant k is crucial. The value is a specific value pre-calibrated through experiments based on the material properties, diameter, and heat transfer coefficient between the fence wire rope and low-temperature water (e.g., 8 degrees Celsius). It directly reflects the heat exchange rate caused by significant temperature differences in the scene's specific characteristics.

[0110] In some preferred embodiments, 50 data points from the actual tension decay data sequence are iterated, and the square of the difference between each point and the theoretical model value at the same time is calculated. These 50 squared differences are then summed. If the final residual sum of squares is less than a pre-set threshold (e.g., this threshold is set after calculation based on real water jet impact experimental data), the measured data is considered to be in high agreement with the thermal shock physical model. If the residual sum of squares is greater than the threshold, for example, if the tension remains essentially constant after the peak (continuous force is applied with the crowbar), or if the tension suddenly drops to zero (the steel cable is cut), the measured data is determined to be inconsistent with the thermal shock physical model.

[0111] The above technical solution shifts the basis for judgment from the vague "moment of the event" to the clear "event recovery process," using physical laws as the final adjudication standard, thus effectively improving the accuracy of judgment.

[0112] Preferably, based on the third judgment result, the corresponding false alarm filtering operation or alarm operation is performed, including the following steps:

[0113] If the third judgment result is yes, then the actual tension decay data sequence is defined to match the preset theoretical tension decay sequence, the high-risk event to be confirmed is archived as an identified interference event, and the false alarm filtering operation is performed.

[0114] If the third judgment result is negative, then the actual tension decay data sequence is defined as not matching the preset theoretical tension decay sequence, and the high-risk event to be confirmed will be archived as a real intrusion event, and an alarm operation will be executed.

[0115] The above steps are the specific implementation details of step S5: Steps C1 to C3 determine whether the measured data matches the thermal shock physical model. When the measured data matches the thermal shock physical model, the event is determined to be a thermal shock contraction effect, a "false alarm filtering" operation is performed, the suspended alarm command is revoked, and the event is archived as an identified interference event. When the measured data does not match the thermal shock physical model, the event is determined to not conform to the physical recovery law of thermal shock, and is confirmed as a real physical intrusion. At this time, the microcontroller immediately releases the suspended state, drives the alarm relay, and sends an alarm signal to the security center.

[0116] Once sampling data exhibiting both instantaneous impact characteristics and sustained tension surge characteristics is selected, the system does not immediately identify it as an intrusion and trigger an alarm as in traditional methods. Instead, recognizing the potential ambiguity of this signal—it could originate from factoring shear or the impact of high-pressure, low-temperature water flow—it enters a brief observation and verification phase. While suspending alarm decisions, the system focuses on the evolution of the signal after reaching its peak, recording changes in tension values ​​to create an actual attenuation curve. This recorded attenuation curve is then compared to a pre-set "theoretical standard thermal shock recovery curve" calculated based on the physical laws of thermal conduction. If the actual curve closely matches the theoretical standard curve, the event is determined to be due to natural "thermal recovery" after metal contraction rather than sustained external force, and is thus classified as a false alarm and filtered out. Conversely, if the recorded tension remains high after the peak, fluctuates irregularly, or suddenly disappears, completely contradicting the standard thermal recovery curve, it is determined to be a genuine intrusion, and an alarm is immediately triggered.

[0117] The innovation of the above steps in this solution lies in its in-depth exploration of information dimensions and strategic shift in the focus of judgment. Instead of acquiring additional information by adding new hardware such as temperature sensors, it interprets a completely new and implicit physical process dimension from the existing, single tension sensor signal. Conventional alarm logic treats the signal as direct evidence of intrusion, while this method treats the signal as a complete record of a physical process. It innovatively proposes that the "second half" of the signal, the recovery phase after the disturbance, contains information equal to or even more important than the "first half" of the disturbance. It does not attempt to solve the vague identification problem of "whether the impact signal resembles an intrusion," but cleverly bypasses it to verify the clear physical problem of "whether the signal recovery process follows thermodynamic laws." This shift in thinking from "instantaneous morphological matching" to "process law verification" redefines the nature of the problem, fundamentally changes the focus of signal processing, and possesses significant non-obviousness.

[0118] This method can effectively filter out composite false signals generated under the specific scenario of "high pressure and low temperature fluid impact", solving the problem that existing technologies cannot distinguish such signals from real violent intrusion signals, and greatly reducing the false alarm rate under such special operating conditions.

[0119] like Figure 5 As shown, preferably, after defining that the actual tension decay data sequence does not match the preset theoretical tension decay sequence if the third judgment result is negative, the following steps are also included:

[0120] D1. Fourth, determine whether there is a macroscopic decay trend in the actual tension decay data sequence. If the result of the fourth judgment is yes, then the fifth judgment is whether the macroscopic decay process is stable. If the result of the fifth judgment is yes, then the high-risk event to be confirmed is archived as an identified interference event and a false alarm filtering operation is performed.

[0121] D2. If either the fourth or fifth judgment result is negative, the high-risk event to be confirmed will be archived as a real intrusion event, and an alarm operation will be executed.

[0122] Steps D1 to D2 in this application are applied in a more complex deployment environment: the electronic tension fence is not installed at an inland water plant, but rather for perimeter security of a nuclear power plant built in a subtropical coastal region. This area experiences high humidity year-round and high salt concentrations in the air, significantly corroding metal components. To ensure the long-term structural stability of the fence, its routine maintenance procedures include a special maintenance operation: after regular water cleaning, maintenance personnel use specialized equipment to spray a thin, highly adhesive oil-based rust inhibitor onto the surface of all wire ropes. This rust inhibitor solidifies on the wire rope surface, forming a dense protective film that isolates direct contact between salt spray and the metal. Imagine that after a summer typhoon, some sections of the fence are covered with a large amount of wind-blown plant debris and silt. A security patrol officer discovers this and, believing the contaminant may affect the sensitivity of the tension sensors, decides to clean it immediately. Instead of following the standard procedure to request maintenance mode, he directly activates the high-pressure water gun on a nearby fire hydrant. At this time, the surface temperature of the steel wire rope, exposed to the afternoon sun, reached 50 degrees Celsius, while the temperature of the emergency water in the fire hydrant was only 15 degrees Celsius, creating a huge temperature difference of 35 degrees Celsius. When this high-pressure, low-temperature water flow impacted the steel wire rope coated with rust inhibitor, as expected, the steel wire rope "contracted" due to the drastic temperature difference, and the tension value increased sharply, triggering the "high-risk event to be confirmed" judgment logic and starting to collect tension decay data after the peak. However, in the subsequent recovery phase, the kinetic energy and scouring effect of the high-pressure water flow began to destroy the oil-based rust inhibitor protective film on the surface of the steel wire rope. This viscous oil film was not washed away instantly, but was gradually emulsified and peeled off under the action of the water flow. This process resulted in the formation of a physically unstable emulsion layer on the surface of the steel wire rope, composed of a mixture of water, oil, and air microbubbles. This emulsion layer played an unexpected "insulating" role, greatly altering the heat exchange efficiency between the steel wire rope and the surrounding air. The process by which the steel wire rope absorbs ambient heat to recover its temperature no longer follows the smooth, predictable exponential decay pattern of a clean metal surface in air. Its recovery process becomes slow and irregular due to the dynamic changes in the emulsion layer (partially washed away, partially still attached). The collected slow, fluctuating, non-exponential decay data was compared with the internally stored theoretical recovery model calibrated based on the physical conditions of a "clean steel wire rope." Because the two curves differed significantly in shape, the sum of squared residuals far exceeded the judgment threshold. Based on this, it was determined that the signal's recovery process did not conform to the preset physical laws of thermal shock, thus ruling out the possibility of thermal shock and incorrectly classifying it as a real intrusion caused by continuous, irregular external forces, ultimately triggering a false alarm.

[0123] The purpose of steps D1 to D2 is to solve the technical problem of "how to effectively distinguish between two situations after encountering unplanned high-pressure low-temperature water flow impact: one is the real high tension maintained by continuous mechanical external force; the other is the composite false signal formed by the superposition of two effects of "thermal shock contraction" and "the peeling and emulsification of the surface adhesive rust inhibitor layer" caused by the water flow impact, which leads to the actual recovery process of the tension signal deviating from the preset theoretical recovery model".

[0124] Steps D1 to D2 involve determining whether there is a macroscopic attenuation trend in the overall actual tension attenuation data sequence after the measured data does not match the thermal shock physical model. If a macroscopic attenuation trend exists, the fifth step is to determine whether the macroscopic attenuation process is stable. If the macroscopic attenuation process is stable, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed. If there is no macroscopic attenuation trend or the macroscopic attenuation process is unstable, the high-risk event to be confirmed is archived as a real intrusion event, and an alarm operation is performed.

[0125] For the specific scenario described above, a "fault tolerance and feature re-identification" judgment step is added to the original single-layer decision-making logic. When the tension decay curve fails to pass an accurate comparison with the "clean steel wire rope thermal recovery model," this is not considered the final conclusion. It "takes a step back": does this mismatched curve represent "refusal to recover" (e.g., continuous external force) or "abnormal but still ongoing recovery" (e.g., insulation layer interference in this scenario)? To this end, a two-dimensional "qualitative" check is performed on this "unqualified" curve: first, from an overall perspective, is the tension developing in a downward direction (i.e., determining whether there is a macroscopic decay trend); second, how "turbulent" is this downward development process (i.e., determining whether the macroscopic decay process is stable). Through these two checks, the unique signal characteristics of the specific physical process of rust inhibitor emulsification are captured and identified.

[0126] like Figure 6 As shown, preferably, the fourth step is to determine whether the actual tension attenuation data sequence has a macroscopic attenuation trend. If the result of the fourth determination is yes, the following steps are included:

[0127] E1. Divide all data points of the actual tension decay data sequence into a first part of data points and a second part of data points;

[0128] E2. Obtain the average value of the first actual tension attenuation data corresponding to the first part of data points and the average value of the second actual tension attenuation data corresponding to the second part of data points respectively;

[0129] E3. Fourth, determine whether the difference between the average value of the first actual tension attenuation data and the average value of the second actual tension attenuation data is greater than the preset average value difference threshold;

[0130] E4. If the result of the fourth judgment is yes, then the actual tension decay data sequence is defined to have a macroscopic decay trend.

[0131] Steps E1 to E4 of this application are specific implementation details of the fourth judgment process in step D1: All data points in the measured data are divided into a first half and a second half; the average tension decay data corresponding to the first half and the average tension decay data corresponding to the second half are calculated; it is determined whether the difference between the average tension decay data corresponding to the second half and the average tension decay data corresponding to the first half is greater than a preset average difference threshold; if the difference between the average tension decay data corresponding to the second half and the average tension decay data corresponding to the first half is greater than the preset average difference threshold, then the actual tension decay data sequence is defined as having a macroscopic decay trend.

[0132] In some preferred embodiments, the data points of the actual tension decay data sequence are divided into two parts: the first 25 points (0-2.5 seconds) and the last 25 points (2.5-5 seconds). The arithmetic mean of the tension decay data points corresponding to these two parts is calculated. Let the average value of the first part be denoted as... The average value of the latter part is Then, determine Is it significantly smaller than For example, set a threshold for the difference in average values ​​(e.g., 20 Newtons), if ( If the value is greater than 20 Newtons, then it is believed that although the curve shape is not standard, a macroscopic and continuous decay trend objectively exists.

[0133] like Figure 7 As shown, preferably, the fifth judgment determines whether the macroscopic attenuation process is stable. If the result of the fifth judgment is yes, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed, including the following steps:

[0134] F1. Obtain the average of the absolute values ​​of the differences between the actual tension decay data corresponding to two consecutive adjacent sampling points in all data points of the actual tension decay data sequence;

[0135] F2. Fifth, determine whether the average of the absolute values ​​of the differences falls within the preset stable range;

[0136] F3. If the result of the fifth judgment is yes, then the macroscopic decay process is defined as stationary;

[0137] F4. When the actual tension decay data sequence shows a macroscopic decay trend and the macroscopic decay process is stable, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed.

[0138] Steps F1 to F4 of this application are specific implementation details of the fifth judgment process in step D1: by obtaining the average of the absolute values ​​of the difference between the actual tension attenuation data corresponding to two adjacent consecutive sampling points in all data points of the actual tension attenuation data sequence, it is determined whether the average of the absolute values ​​of the difference falls within a preset stable range. If it falls within the stable range, the macroscopic attenuation process is defined as stable. Only when the actual tension attenuation data sequence has a macroscopic attenuation trend and the macroscopic attenuation process is stable, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed.

[0139] In some preferred embodiments, analyzing the smoothness of this decay process involves calculating the average of the absolute values ​​of the tension changes between two adjacent sampling points (with a time interval of 100 milliseconds) throughout the entire 50-point data sequence. This average value reflects the degree of local "bumps" or "jitter" in the curve and is denoted as [value missing]. A reasonable, albeit irregular, range is pre-defined, for example, [1.0 Newtons, 5.0 Newtons]. The upper and lower limits of this range are determined through experimental data collection and calibration of the "rust inhibitor stripping and emulsification" process. If the calculated... If the curve falls within this range, it indicates that the degree of fluctuation in the curve matches the physical characteristics of a slow, uneven peeling of the rust inhibitor layer (i.e., a stable macroscopic decay process). If A value less than 1.0 Newton indicates that the curve is too smooth, resembling a stable, slow release of mechanical force rather than a physicochemical process. If... A value greater than 5.0 Newtons indicates that the curve jitters too violently, more like artificial, irregular shaking or tool scraping.

[0140] The innovation of the above steps in this solution lies in the following: When faced with a signal that does not conform to the preset "clean steel wire rope" thermal recovery model, it does not hastily classify it as an intrusion, but instead provides a second verification opportunity. For genuine intrusion behavior, such as an intruder applying continuous force with a crowbar, the tension signal cannot be verified through "macroscopic confirmation of attenuation trend" because the tension will remain high, and the difference between the average values ​​of the two segments is very small. Therefore, an alarm will be correctly triggered. For the false signal unique to this scenario, namely the impact of high-pressure low-temperature water on a steel wire rope coated with rust inhibitor, although its tension signal cannot be compared with the accurate model in the first round, it can pass the second verification: its tension is generally decreasing slowly, which can be verified through "macroscopic confirmation of attenuation trend"; at the same time, this decreasing process is accompanied by slight vibrations caused by oil film emulsification, the amplitude of which falls exactly within the calibration range of "irregularity in the recovery process". Therefore, it can be identified as a benign, composite interference caused by a specific maintenance scenario, and filtered out as a false alarm.

[0141] Through this logical progression from "precise morphological matching" to "macroscopic feature verification," this solution improves the logical discrimination capability in complex and special operating environments without increasing any hardware costs, and accurately solves the false alarm problem caused by the additional physical variables introduced by the protective coating.

[0142] like Figure 8 As shown, preferably, after obtaining the actual tension decay data sequence of the original sampled data after moving average processing over a preset time period, the following steps are also included:

[0143] G1. Traverse each data point of the actual tension decay data sequence, calculate the second derivative of the actual tension decay data corresponding to each data point, and obtain the tension change acceleration sequence.

[0144] G2. Obtain the variance of the tension change acceleration sequence;

[0145] G3. The sixth step is to determine whether the variance of the tension change acceleration sequence is less than the preset smoothing threshold.

[0146] G4. Based on the result of the sixth judgment, perform the corresponding false alarm filtering operation or alarm operation.

[0147] Steps G1 to G4 in this application are applied in a more complex deployment environment: the production process at a fertilizer production base generates a large amount of highly hygroscopic urea dust. This dust inevitably settles and adheres to the surface of the wire ropes of the electronic tension fence surrounding the plant, forming a thin layer of solid powder. In dry weather, this powder layer has minimal impact on the tension reference value of the wire ropes. However, due to higher humidity at night, the urea dust adhering to the wire ropes absorbs moisture from the air but does not completely dissolve. Instead, it forms a sticky, highly concentrated urea-saturated solution film on the wire rope surface. Subsequently, as sunlight intensifies, the surface temperature of the wire ropes rapidly rises to 45 degrees Celsius, causing some moisture to evaporate, but the sticky solution film still coats the wire ropes. At this time, a transport vehicle had an accident while traveling on an internal road within the factory area, causing some raw materials to leak. To prevent the situation from escalating, a site worker, without prior notification, immediately activated the fire hydrant and used fire water at a pressure of 0.6 MPa and a temperature of 20 degrees Celsius to urgently wash the ground and fences near the leak area. When this high-pressure water flow impacted the high-temperature steel wire rope wrapped in a film of high-concentration urea solution, the impact of the water flow and the sudden change in temperature also triggered violent vibration and "cold contraction" of the steel wire rope, causing a significant jump in tension. The alarm judgment program was suspended, and subsequent tension decay data collection began. However, in the subsequent recovery phase, the physical evolution of the steel wire rope deviated from the preset trajectory. The high-pressure water flow did not simply act as a heat exchange medium as it would when cleaning the steel wire rope. It first diluted the high-concentration urea solution on the surface of the steel wire rope. This dilution and dissolution process is a significant endothermic process. In addition to the "cold contraction" effect, it rapidly absorbed a large amount of heat from the steel wire rope itself, causing the initial slope of the tension recovery curve to be much steeper than the standard process. Subsequently, the diluted urea solution covered the entire impacted section of the steel wire rope. Due to the presence of the solution, its evaporation boiling point is higher than that of pure water, and this liquid film remains on the surface of the steel wire rope for a longer period. The heat dissipation process from the steel wire rope to the environment changes from the originally planned "metal-air" convection heat transfer to a complex "metal-urea solution-air" heat transfer process that includes the latent heat of vaporization of the liquid. This process causes the tension recovery curve to exhibit a flattening, non-exponential tailing characteristic in the middle and later stages. This measured attenuation curve, characterized by a "rapid descent in the initial stage and a flattening in the later stage," is compared with the internally stored standard exponential attenuation process calibrated based on the physical conditions of a "clean steel wire rope." Because the two curves differ fundamentally in their initial slope and overall shape, the calculated sum of squared residuals far exceeds the judgment threshold. Based on this, it is determined that the recovery process of the signal does not conform to the preset physical laws of thermal shock, thus negating the possibility of thermal shock and incorrectly attributing it to an intrusion caused by an unknown, continuous external force, ultimately triggering a false alarm.

[0148] The purpose of steps G1 to G4 is to solve the technical problem of "how to design a false alarm identification and control method that can effectively distinguish between two situations after encountering unplanned high-pressure water flow impact: one is a real high tension maintained by continuous and irregular external mechanical force; the other is a composite false signal caused by the superposition of two physical effects, namely "thermal shock contraction" and "endothermic dissolution of surface deposits and subsequent evaporation of solution changing the heat transfer mode", which are caused by the water flow impact.

[0149] Steps G1 to G4 involve iterating through each data point after obtaining the actual tension decay data sequence, calculating the second reciprocal of the actual tension decay data corresponding to each data point, and obtaining the tension change acceleration sequence; calculating the variance value of the obtained tension change acceleration sequence, determining whether the variance value is less than a preset smoothing threshold, and performing corresponding false alarm filtering or alarm operations based on the determination result.

[0150] For the specific scenarios described above, the basis for judgment shifts from the specific shape of the tension signal recovery curve (i.e., the first-order rate of change of the signal) to the inherent smoothness of the curve's change process (i.e., the stability of the second-order rate of change of the signal). The basic judgment logic is: any state change dominated by natural physical or chemical laws, no matter how complex the process or how unusual the curve shape, must be continuous and smooth on a macroscopic level; while the application and change of force by human-induced mechanical intrusion are inherently discontinuous and uneven, filled with abrupt starts, stops, and adjustments. Therefore, by verifying the "smoothness" of the tension change process, the physical source of the event can be fundamentally distinguished. Based on this, in the existing control logic, when a high-risk complex event is captured and the alarm is suspended, no comparison with any preset curve shape is performed; instead, a dedicated "process smoothness verification procedure" is initiated.

[0151] In some preferred embodiments, when a high-risk event is detected, the microcontroller suspends the alarm and, after the tension reaches its peak, continuously collects data for 5 seconds at 100-millisecond intervals to obtain a measured tension decay data sequence consisting of 50 data points, denoted as . ;

[0152] The microcontroller iterates through the acquired tension data sequence and calculates its second derivative. For each internal point in the sequence... ( From 1 to 48), its second derivative value It can be approximated using the central difference formula: This calculation process generates a "tension change acceleration" sequence A consisting of 48 points; the microcontroller calculates the dispersion of this "tension change acceleration" sequence A, specifically by calculating the variance of the sequence. First, the average value of sequence A is calculated. Then calculate the variance. This variance value directly reflects the degree of "bumping" or "shaking" in the tension change process; it internally stores a "physical process smoothness threshold" (i.e., a preset smoothing threshold), such as a specific variance value. This threshold is a critical value that can effectively distinguish between two types of events, determined through experiments on various real physical processes (such as simple thermal shock, coated thermal shock, etc.) and various simulated intrusion behaviors (such as pulling, shaking).

[0153] If the calculated variance If the value is below this threshold, it indicates that the tension recovery process is smooth and continuous. The control unit determines that the event originates from a natural physical or chemical process (whether it is standard exponential recovery or the complex urea dissolution recovery in this scenario), and therefore performs a "false alarm filtering" operation, canceling the suspended alarm command; if the calculated variance... If the value is greater than or equal to this threshold, it indicates that the tension recovery process is filled with irregular and drastic acceleration and deceleration changes. The control unit determines that this process does not possess the smooth characteristics of a natural physical process, and its source is most likely a discontinuous artificial mechanical force. Therefore, it is confirmed as a real intrusion and immediately activates the alarm relay.

[0154] The innovation of the above steps in this solution lies in the fact that upon capturing a suspicious signal, it no longer focuses on what the recovery curve of that signal "looks like," but rather examines whether the recovery process itself is "smooth." It quantifies this "smoothness" by calculating the variance of the second derivative of the tension signal. In this scenario, the complex recovery process caused by the combined endothermic reaction of urea dissolution and solution evaporation, although its curve shape (first derivative) is non-standard, the process itself is continuous, and its second derivative variance will be small. However, when an intruder exerts continuous force, its fine-tuning of force will cause irregular fluctuations in tension, resulting in a large second derivative variance. By setting a reasonable variance threshold, it is possible to accurately distinguish all smooth "natural processes" from non-smooth "artificial processes," thereby solving the problem.

[0155] In this way, the present invention is no longer limited to the understanding of specific curve shapes, but by verifying a more essential physical process property—"continuous smoothness"—it successfully distinguishes irregular and discontinuous interference caused by human intervention from all continuous processes, no matter how complex, dominated by natural laws. This accurately solves the problem of false alarms caused by complex physicochemical effects in specific chemical environments.

[0156] like Figure 9 As shown, a perimeter alarm false alarm filtering and control system includes:

[0157] The data acquisition module is used to acquire the raw sampling data corresponding to multiple consecutive sampling points at the perimeter of the suspected area;

[0158] The module for defining high-risk events to be confirmed is used to first determine whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold. If the first determination result is yes, then the module for the second determination is whether the original sampling data after moving average processing jumps above the high tension alarm threshold within a preset time period and remains at a high level. If the second determination result is yes, it is defined as a high-risk event to be confirmed.

[0159] The actual tension decay data sequence acquisition module is used to acquire the actual tension decay data sequence of the raw sampled data after moving average processing within a preset time period;

[0160] The sequence matching judgment module is used to make a third judgment on whether the actual tension decay data sequence matches the preset theoretical tension decay sequence.

[0161] The execution module is used to perform corresponding false alarm filtering or alarm operations based on the third judgment result.

[0162] The above scheme provides a perimeter alarm false alarm filtering and control system. Through modular design, this system assigns data acquisition, definition of high-risk times to be confirmed, acquisition of actual tension decay sequences, sequence matching judgment, and operation execution functions to different units, enabling the perimeter alarm false alarm filtering and control method to be executed effectively and reliably. This system effectively distinguishes between non-intrusive factors such as thermodynamic effects and actual intrusion behavior, thus possessing the advantage of effectively differentiating tension changes caused by thermodynamic effects from intrusion behavior, and significantly reducing false alarms caused by non-intrusive factors such as environmental temperature differences.

[0163] In the embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple modules or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, and can be electrical, mechanical, or other forms.

[0164] Furthermore, in the various embodiments of the present invention, each functional module can be fully integrated into a processor, or each module can be a separate device, or two or more modules can be integrated into a device; each functional module in the various embodiments of the present invention can be implemented in hardware or in the form of hardware plus software functional units.

[0165] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by program instructions and related hardware. The aforementioned program instructions can be stored in a computer-readable storage medium. When the program instructions are executed, they perform the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0166] It should be understood that the use of terms such as "system," "device," "unit," and / or "module" in this application is merely one method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0167] As illustrated in this application, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements. An element defined by the phrase "comprising an..." does not exclude the presence of other identical elements in the process, method, product, or apparatus that includes the element.

[0168] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0169] If a flowchart is used in this application, it is used to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0170] The foregoing has provided a detailed description of a perimeter alarm false alarm filtering control method and system provided by the present invention. The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A perimeter alarm false alarm filtering and control method, applied to an electronic tension alarm system for detecting intrusion scenarios, characterized in that, Includes the following steps: Obtain the raw sampling data corresponding to multiple consecutive sampling points at the perimeter of the suspected area; The first determination is whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold. If the first determination result is yes, the second determination is whether the original sampling data after moving average processing jumps above the high tension alarm threshold within a preset time period and remains at a high level. If the second determination result is yes, it is defined as a high-risk event to be confirmed. Obtain the actual tension decay data sequence of the original sampled data after moving average processing within the preset time period; The third step is to determine whether the actual tension decay data sequence matches the preset theoretical tension decay sequence. Based on the third judgment result, execute the corresponding false alarm filtering operation or alarm operation; The third step of determining whether the actual tension attenuation data sequence matches the preset theoretical tension attenuation sequence includes the following steps: Iterate through each data point in the actual tension decay data sequence and calculate the square of the difference between the actual tension decay data and the theoretical tension decay data for each data point. By summing the squares of the differences at each data point, we obtain the sum of squared residuals of the tension decay data. Third, determine whether the sum of squared residuals of the tension decay data is less than a preset threshold for the sum of squared residuals; The theoretical tension decay data for each data point is obtained based on a preset theoretical tension decay sequence. The step of performing the corresponding false alarm filtering or alarm operation based on the third judgment result includes the following steps: If the third judgment result is yes, then the actual tension decay data sequence is defined to match the preset theoretical tension decay sequence, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed. If the third judgment result is negative, then the actual tension decay data sequence is defined as not matching the preset theoretical tension decay sequence, and the high-risk event to be confirmed will be archived as a real intrusion event, and an alarm operation will be performed. After defining the actual tension decay data sequence as not matching the preset theoretical tension decay sequence if the third judgment result is negative, the following steps are also included: Fourth, determine whether there is a macroscopic attenuation trend in the actual tension attenuation data sequence. If the result of the fourth determination is yes, then the fifth determination is whether the macroscopic attenuation process is stable. If the result of the fifth determination is yes, then the high-risk event to be confirmed is archived as an identified interference event and a false alarm filtering operation is performed. If either the fourth or fifth judgment result is negative, the high-risk event to be confirmed will be archived as a real intrusion event, and an alarm operation will be performed.

2. The perimeter alarm false alarm filtering control method as described in claim 1, characterized in that, The first determination involves whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold. If the first determination result is yes, the following steps are included: Calculate the difference between the original sampled data corresponding to two adjacent consecutive sampled points; Within a first preset window length, the first determination is whether the difference between the original sampling data corresponding to two adjacent consecutive sampling points is greater than the tension change threshold. If the first judgment result is yes, then the existence of rapid impact characteristics is defined.

3. The perimeter alarm false alarm filtering control method as described in claim 2, characterized in that, The second determination determines whether the original sampled data after moving average processing jumps above the high-tension alarm threshold within a preset time period and remains at a high level. If the second determination result is yes, it is defined as a high-risk event to be confirmed, including the following steps: Based on the second preset window length, the original sampled data is processed by moving average to obtain the quasi-static tension value; Within a preset time period, the second determination is whether the quasi-static tension value jumps above the high tension alarm threshold and remains at a high level. If the second judgment result is yes, then the existence of a continuous tension jump characteristic is defined; When both the rapid impact characteristic and the continuous increase in tension are present, it is defined as a high-risk event to be confirmed.

4. The perimeter alarm false alarm filtering control method as described in claim 1, characterized in that, The fourth determination involves whether the actual tension attenuation data sequence exhibits a macroscopic attenuation trend. If the result of the fourth determination is yes, the following steps are included: All data points in the actual tension decay data sequence are divided into a first part of data points and a second part of data points. The average value of the first actual tension attenuation data corresponding to the first part of data points and the average value of the second actual tension attenuation data corresponding to the second part of data points are obtained respectively. Fourth, determine whether the difference between the average value of the first actual tension attenuation data and the average value of the second actual tension attenuation data is greater than a preset average difference threshold. If the fourth judgment result is yes, then the actual tension decay data sequence is defined to have a macroscopic decay trend.

5. The perimeter alarm false alarm filtering control method as described in claim 4, characterized in that, The fifth judgment determines whether the macroscopic attenuation process is stable. If the result of the fifth judgment is yes, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed, including the following steps: The average value of the absolute difference between the actual tension decay data corresponding to two consecutive adjacent sampling points in all data points of the actual tension decay data sequence is obtained. Fifth, determine whether the average of the absolute values ​​of the differences falls within the preset stable range. If the result of the fifth judgment is yes, then the macroscopic decay process is defined as stationary; When the actual tension decay data sequence shows a macroscopic decay trend and the macroscopic decay process is stable, the high-risk event to be confirmed is archived as an identified interference event, and a false alarm filtering operation is performed.

6. The perimeter alarm false alarm filtering control method as described in claim 1, characterized in that, After obtaining the original sampled data after moving average processing, and then the actual tension decay data sequence within the preset time period, the following steps are also included: By iterating through each data point of the actual tension decay data sequence, the second derivative of the actual tension decay data corresponding to each data point is calculated to obtain the tension change acceleration sequence. Obtain the variance value of the tension change acceleration sequence; The sixth step is to determine whether the variance of the tension change acceleration sequence is less than a preset smoothing threshold. Based on the sixth judgment result, execute the corresponding false alarm filtering operation or alarm operation.

7. A perimeter alarm false alarm filtering and control system, characterized in that, include: The data acquisition module is used to acquire the raw sampling data corresponding to multiple consecutive sampling points at the perimeter of the suspected area; The module for defining high-risk events to be confirmed is used to first determine whether the tension change value between two adjacent consecutive sampling points is greater than the tension change threshold. If the first determination result is yes, then the module for second determination is whether the original sampling data after moving average processing jumps above the high tension alarm threshold within a preset time period and remains at a high level. If the second determination result is yes, it is defined as a high-risk event to be confirmed. The actual tension decay data sequence acquisition module is used to acquire the actual tension decay data sequence of the original sampled data after moving average processing within the preset time period. The sequence matching judgment module is used to make a third judgment on whether the actual tension decay data sequence matches the preset theoretical tension decay sequence; The execution module is used to perform corresponding false alarm filtering or alarm operations based on the third judgment result; The sequence matching determination module includes: The residual sum of squares calculation unit is used to traverse each data point of the actual tension decay data sequence, calculate the square of the difference between the actual tension decay data and the theoretical tension decay data corresponding to each data point, and accumulate the square of the difference of each data point to obtain the residual sum of squares of the tension decay data. A threshold comparison unit is used to determine whether the sum of squared residuals of the tension decay data is less than a preset residual sum of squared threshold. The theoretical tension decay data for each data point is obtained based on a preset theoretical tension decay sequence. The execution module includes: The first execution unit is used to define the actual tension decay data sequence as matching the preset theoretical tension decay sequence if the third judgment result is yes, to archive the high-risk event to be confirmed as an identified interference event, and to perform a false alarm filtering operation. The second execution unit is used to define the actual tension attenuation data sequence as not matching the preset theoretical tension attenuation sequence if the third judgment result is negative, to archive the high-risk event to be confirmed as a real intrusion event, and to perform an alarm operation. The macroscopic attenuation judgment module is used to make a fourth judgment after the third judgment result is negative to determine whether there is a macroscopic attenuation trend in the actual tension attenuation data sequence. The stability judgment module is used to determine whether the macroscopic decay process is stable if the fourth judgment result is yes. The execution module further includes: The third execution unit is used to archive the high-risk event to be confirmed as an identified interference event and perform a false alarm filtering operation if the fourth judgment result is yes and the fifth judgment result is yes. The fourth execution unit is used to archive the high-risk event to be confirmed as a real intrusion event and perform an alarm operation if either the fourth judgment result or the fifth judgment result is negative.

Images