Water immersion alarm method based on intelligent waterproof baffle and storage medium

Abstract

The invention relates to the technical field of information, and discloses a water immersion alarm method based on an intelligent waterproof baffle and a storage medium, and the method comprises the steps: collecting the water level data in a power distribution room and the surface humidity data of a water blocking strip in real time, and determining the current water level rising speed and the initial expansion state of the water blocking strip; the incidence relation between water level rising and water blocking strip expansion is recognized; the expansion rate distribution of the water blocking strip in different areas is identified according to the incidence relation between the water level rising and the expansion of the water blocking strip, and the expansion non-uniformity degree is obtained by comparing the expansion rate differences between the different areas; if the matching degree is lower than a preset matching threshold value, determining an alarm threshold value according to the potential position of the water leakage channel and the expansion non-uniformity degree; water level alarm key parameters are extracted according to the accurate water level early warning signal, and an execution path of intelligent response about water level abnormity of the power distribution room is determined according to the water level alarm key parameters and the early warning signal type, so that the reliability of a waterproof system of the power distribution room is improved, and the water immersion risk of the power distribution room is reduced.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a water immersion alarm method and storage medium based on an intelligent waterproof baffle. Background Technology

[0002] As a core component of urban power infrastructure, the waterproofing of power distribution rooms directly impacts the stability and security of power supply. During heavy rains, rapidly rising water levels can submerge equipment, causing short circuits or malfunctions, resulting in widespread power outages and economic losses. Especially in environments with strong water flow, traditional sealing materials struggle to adapt to dynamic changes, leading to inaccurate water level signal detection. For example, the coupling effect between water flow intensity and material deformation significantly weakens the overall protective effect under complex conditions, failing to meet the high reliability requirements of power distribution rooms. In waterproof baffle systems, water-blocking strips, as core sealing elements, should enhance the sealing of gaps due to their water-expanding properties, but they face the challenge of uneven expansion rates. This uneven expansion stems from the interaction of water flow intensity, water-blocking strip material properties, and gap geometry. For instance, in areas with rapid water flow, the water-blocking strip expands quickly, rapidly sealing local gaps, while in corners with gentler flow, the expansion process lags, leaving small openings. Water flows through these incompletely sealed areas, forming bypass channels that directly bypass water immersion alarm sensors, interfering with their real-time water level detection. For example, in a simulated rainstorm test, when the water level rose rapidly from its initial height to the warning line, the sensor delayed triggering the alarm for several minutes due to the presence of localized leakage channels. During this time, water had already seeped into the equipment base, amplifying the potential risk. This difference in expansion rate not only reduced the sensor's triggering frequency but also made it difficult to maintain the accuracy of water level monitoring, thus affecting the overall system's protective response. Therefore, accurately capturing the dynamic changes between the expansion rate of the water-blocking strip and the triggering frequency of the water immersion sensor during a rapid rise in water level has become a key issue in improving the reliability of the power distribution room's waterproofing system. Summary of the Invention

[0003] The main objective of this invention is to provide a water immersion alarm method based on an intelligent waterproof baffle, which aims to improve the reliability of the waterproof system in the power distribution room and reduce the risk of water immersion in the power distribution room.

[0004] To achieve the above objectives, the water immersion alarm method based on an intelligent waterproof baffle proposed in this invention includes the following steps: Collect water level data and surface humidity data of water-blocking strips in the power distribution room to determine the water level rise rate and the initial expansion state of the water-blocking strips, and identify the correlation between water level rise and water-blocking strip expansion. Based on the correlation between the rise in water level and the expansion of the water-blocking strip, the expansion rate distribution of different regions of the water-blocking strip is obtained, the difference in expansion rate of each region is calculated, and the degree of expansion unevenness is determined. When the degree of expansion unevenness exceeds the preset expansion unevenness threshold, the expansion rate distribution is subjected to spectral analysis to identify the expansion rate abrupt change region, determine the potential location of the leakage channel, and obtain the water flow bypass path data of the potential location of the leakage channel. Feature parameters are extracted from the expansion rate abrupt change region, and the sensor triggering frequency is determined based on the feature parameters. The sensor triggering frequency is then matched with the water level sensing accuracy to obtain the matching degree. When the matching degree is lower than the preset matching degree threshold, an alarm threshold is determined based on the potential location of the leakage channel and the degree of expansion unevenness. Based on the alarm threshold, the water level rise rate, and the water flow bypass path data, a water level warning signal is generated. The warning signal type is determined by comparing the water level warning signal with a historical database. Based on the water level warning signal, extract the water level height, expansion rate change amplitude, and potential location of the leakage channel, and determine the response path for abnormal water level in the power distribution room by combining the warning signal type.

[0005] More specifically, the steps of collecting water level data and surface humidity data of the water-blocking strip in the power distribution room, determining the rate of water level rise and the initial expansion state of the water-blocking strip, and identifying the correlation between water level rise and water-blocking strip expansion include: Collect water level values ​​at different heights in the power distribution room, calculate the ratio of the water level difference between adjacent time points to the time interval, and determine the rate of water level rise; Collect surface humidity data of the water-blocking strip, determine the water absorption based on the humidity data and the material characteristics of the water-blocking strip, and determine the initial expansion state by the ratio of the water absorption to the original volume of the water-blocking strip. Based on the water level rise rate and the initial expansion state, the volumetric deformation at different water levels is calculated. By fitting the volumetric deformation with a time series, an expansion rate curve is generated. The slope change of the expansion rate curve is extracted to determine the boundary between accelerated expansion and stability.

[0006] More specifically, the steps of calculating the volumetric deformation at different water levels based on the water level rise rate and the initial expansion state, generating an expansion rate curve by fitting the volumetric deformation to a time series, extracting the slope change of the expansion rate curve, and determining the boundary point between accelerated and stable expansion include: Based on the water level rise rate and the initial expansion state, and combined with the water absorption characteristics of the water-blocking strip material, the volume expansion per unit time is calculated. An expansion rate curve is generated by fitting the volume expansion amount with a time series. Extract the point in the expansion rate curve where the slope changes from increasing to constant, and determine the constant point as the dividing point between accelerated expansion and stable expansion.

[0007] More specifically, the step of obtaining the expansion rate distribution of different regions of the water-blocking strip based on the correlation between the water level rise and the expansion of the water-blocking strip, calculating the difference in expansion rate between each region, and determining the degree of expansion unevenness includes: Based on the correlation between the rise in water level and the expansion of the water-blocking strip, the monitoring area of ​​the water-blocking strip is divided, and the expansion rate data of each area is obtained. Calculate the ratio of the expansion rate difference between adjacent regions to the distance between regions to determine the rate gradient; generate the expansion rate distribution based on the change of the rate gradient along the length of the water-blocking strip. Extract the peak and valley values ​​of each region in the expansion rate distribution, calculate the ratio of the peak-valley difference to the average expansion rate, and determine the local coefficient of variation; The degree of expansion non-uniformity is determined by the standard deviation of the local coefficient of variation.

[0008] More specifically, the step of performing spectral analysis on the expansion rate distribution, identifying regions of abrupt changes in expansion rate, determining potential locations of leakage channels, and obtaining water flow bypass path data for potential locations of leakage channels when the degree of expansion non-uniformity exceeds a preset threshold for expansion non-uniformity includes: Spectral analysis is performed on the expansion rate distribution to obtain the amplitude spectrum and phase spectrum. Frequency components in the amplitude spectrum that exceed a preset multiple of the average amplitude are extracted to determine the abrupt expansion rate regions. Based on the boundary of the abrupt expansion rate region, the ratio of the expansion rate difference between adjacent measuring points to the distance is calculated to determine the gradient value; Based on the gradient value sampled along the direction of decreasing expansion rate, a tracking path is generated. The potential location of the leakage channel is determined by the intersection of the tracking path and the sealing surface of the water-blocking strip. Based on the potential location of the leakage channel, read the pressure sensor data, calculate the ratio of the pressure difference between adjacent measuring points to the distance, and generate a pressure gradient vector; By determining the direction change of the pressure gradient vector, the water flow inflection point is identified, and the inflection point is connected to generate an initial path; Calculate the ratio of pressure drop at adjacent points on the initial path to the path length to determine the pressure drop per unit length; Based on the ratio of the pressure drop per unit length to the density of water and the acceleration due to gravity, water flow bypass path data is generated.

[0009] More specifically, the steps of extracting feature parameters from the abrupt expansion rate region, determining the sensor triggering frequency based on the feature parameters, and matching the sensor triggering frequency with the water level sensing accuracy to obtain the matching degree include: Peak points are extracted from the regions of abrupt changes in expansion rate, and the difference between the peak points and the average expansion rate is calculated to determine the peak amplitude. Spectral analysis is performed on the data in the region of abrupt expansion rate changes to extract the deviation between the dominant frequency and the stable expansion frequency, and to determine the frequency offset; the ratio of the difference in expansion rate between adjacent sampling points to the time interval is calculated to determine the rate of change. A comprehensive feature value is generated by weighting the normalized peak amplitude, frequency offset, and rate of change; the sensor trigger frequency is then determined using the comprehensive feature value. Based on the sensor trigger frequency, water level data is sampled, the root mean square value of water level deviation is calculated, the water level sensing accuracy is determined, and the matching degree is generated.

[0010] More specifically, the step of determining an alarm threshold based on the potential location of the leakage channel and the degree of expansion unevenness when the matching degree is lower than a preset matching degree threshold includes: Calculate the location risk coefficient based on the distance between the potential location of the leakage channel and the center of the protected area; An expansion deviation coefficient is generated based on the ratio of the degree of expansion unevenness to the standard value; A comprehensive risk value is generated based on the location risk coefficient and the expansion deviation coefficient; The risk level is determined based on the comprehensive risk value, and an alarm threshold is generated.

[0011] More specifically, the step of generating a water level warning signal based on the alarm threshold, the water level rise rate, and the water flow bypass path data, and determining the warning signal type by comparing the water level warning signal with a historical database, includes: A water level margin is generated based on the difference between the alarm threshold and the current water level. An early warning time is generated based on the water level rise rate and the water flow bypass path data; Based on the warning time, current water level, and water level rise rate, a feature vector for the warning signal is constructed. The type of warning signal is determined by comparing the feature vector of the warning signal with the historical database.

[0012] More specifically, the step of extracting water level height, expansion rate abrupt change amplitude, and potential location of leakage channels based on the water level warning signal, and determining the response path for abnormal water level in the power distribution room in combination with the warning signal type, includes: The water level warning signal was analyzed to determine the water level height, the magnitude of the sudden change in the expansion rate, and the potential location of the leakage channel. Based on the water level height, the abrupt change in the expansion rate, and the potential location of the leakage channel, a water level risk factor, an expansion risk factor, and a location risk factor are generated, respectively. A comprehensive risk index is generated based on the water level risk factor, the expansion risk factor, and the location risk factor. Based on the comprehensive risk index and the warning signal type, the response path for abnormal water levels in the power distribution room is determined.

[0013] The present invention also provides a storage medium storing a water immersion alarm program, which, when executed by a processor, implements the steps of the water immersion alarm method based on an intelligent waterproof baffle.

[0014] In this invention, addressing the problem of accurately locating leakage channels due to rising water levels and uneven expansion of water-blocking strips in power distribution rooms, the invention collects real-time data on water level and surface humidity of the water-blocking strips. It analyzes the correlation between the rate of water level rise and the expansion rate of the water-blocking strips, identifies areas of abrupt expansion rate changes, and determines potential leakage channels. This invention integrates water level, humidity, pressure, and fluid sensors, combining spectral frequency analysis and water flow path data to accurately extract abnormal expansion rate points and water pressure distribution changes. It calculates water flow velocity, direction, and curvature, assesses turbulence characteristics, and generates accurate water level warning signals. By comparing with a historical warning database, it determines high-risk emergency types and sends alarm signals to user terminals. An intelligent response path is developed based on water level height, expansion abruptness, and leakage channel location. This invention significantly improves the accuracy of leakage channel location and the real-time performance and accuracy of water level anomaly warnings, effectively reducing the risk of water flooding in power distribution rooms. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0016] Figure 1 The flowchart illustrates the steps of the water immersion alarm method based on an intelligent waterproof baffle provided by the present invention. Figure 2 This is a schematic diagram of the water immersion alarm method based on an intelligent waterproof baffle provided by the present invention. Detailed Implementation

[0017] 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 a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0019] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0020] like Figure 1 and Figure 2 The water immersion alarm method based on the intelligent waterproof baffle in this embodiment may specifically include: S101. Real-time acquisition of water level data and surface humidity data of water-blocking strips in the power distribution room, determination of the current water level rise rate and the initial expansion state of the water-blocking strips, and identification of the correlation between water level rise and water-blocking strip expansion.

[0021] A water level sensor array collects water level values ​​at different heights within the power distribution room at preset time intervals. The water level rise rate is obtained by calculating the ratio of the water level difference between adjacent time points to the time interval. Simultaneously, a humidity sensor is deployed to cover the surface of the water-blocking strip in a grid pattern to obtain the humidity percentage of each section of the strip. The water absorption of the material is determined based on the correlation between the humidity percentage and the density change of the water-blocking strip material. The initial expansion state is obtained by the ratio of the water absorption to the original volume of the water-blocking strip. For the water level rise rate and the initial expansion state, the volumetric deformation at different water levels is calculated based on the product of the water absorption and the volumetric expansion coefficient of the water-blocking strip material. An expansion rate curve is obtained by fitting the volumetric deformation to a time series. When the slope of the expansion rate curve changes from increasing to constant, this inflection point is identified as the boundary between accelerated expansion and stable expansion. If the expansion rate at the boundary point exceeds a preset rate threshold, the moment when the expansion rate reaches its peak is extracted from the expansion rate curve. The time difference between the peak moment and the moment when the water level reaches the corresponding height is calculated. The correlation between the water level rise and the expansion of the water-blocking strip is determined based on the percentage of the time difference to the total water level rise time.

[0022] Specifically, in one embodiment, a water level sensor array is arranged vertically at equal intervals within the power distribution room, with each sensor node spaced 10 centimeters apart, covering the entire area from the ground to the top of the equipment. The preset time interval is determined based on the fastest rate of water level rise in historical rainstorm data, typically set to 3 to 10 seconds. A capacitive humidity sensor is attached to the surface of the water-blocking strip with a 5 cm × 5 cm grid density to monitor local humidity changes in real time.

[0023] Specifically, the correlation between the percentage of humidity and the density change of the water-blocking strip material was obtained through pre-calibration. In a laboratory environment, water-blocking strip samples were placed in environments with different humidity gradients, and their mass increase and volume expansion were measured to establish a humidity-density change curve.

[0024] For example, when the surface humidity increases from 30% to 80%, the density of the polyurethane-based water-blocking strip decreases from 1.2 g / cm³ to 0.8 g / cm³, corresponding to a water absorption of 150% to 200% of its original mass. By consulting this calibration curve, real-time humidity data can be directly converted into the material's current water absorption state. The initial expansion state is calculated using the ratio of water absorption to the original volume of the water-blocking strip; when the water absorption reaches 50% of the original mass, the volume expansion rate is approximately 1.3 times the original volume.

[0025] It should be noted that the volumetric expansion coefficient is not a fixed value, but rather a dynamic function that changes with the amount of water absorbed. In the initial stage of water absorption, the expansion coefficient is relatively large, and it gradually decreases as the internal pores of the material become saturated. The accurate expansion coefficient at the current moment is obtained by matching the real-time water absorption with a preset expansion coefficient lookup table.

[0026] For example, the expansion rate curve is fitted using a cubic polynomial fitting method, which connects discrete volume deformation data points into a smooth curve. The instantaneous expansion rate is obtained by calculating the first derivative at each point on the curve, and the point where the sign of the second derivative changes is the boundary between accelerated expansion and stable expansion.

[0027] For example, V(t) represents the volumetric deformation at time t, a0 represents the initial volumetric deformation constant term, a1 represents the coefficient of the first term, a2 represents the coefficient of the second term, and a3 represents the coefficient of the third term. This formula uses a cubic polynomial to fit discrete volumetric deformation data points, forming a smooth expansion curve.

[0028] In one possible implementation, the strength of the correlation is quantified by the proportion of time difference. A strong correlation is defined as a time difference less than 10% of the total water level rise time; a moderate correlation is defined as a time difference between 10% and 30%; and a weak correlation is defined as a time difference exceeding 30%. This quantitative indicator directly reflects the synchronicity between the expansion response of the water-blocking strip and the water level change.

[0029] The water level in the power distribution room is recorded by a water level sensor at different time points to obtain the water level rise rate per unit time. At the same time, a humidity sensor is used to monitor the moisture content on the surface of the water-blocking strip. The volume expansion rate and water absorption performance of the water-blocking strip material in the initial state are analyzed. By comparing the expansion degree of the water-blocking strip under different water level rise rates, the response time and amplitude of the expansion rate with the water level change are recorded to identify the correlation between water level rise and water-blocking strip expansion.

[0030] Water level sensors collect water level data at different heights within the power distribution room at preset intervals. The instantaneous water level rise rate is obtained by calculating the ratio of the water level difference between adjacent collection points to the time interval. Simultaneously, a humidity sensor array acquires the relative humidity values ​​at various measuring points on the surface of the water-blocking strip. The water penetration depth is determined based on the correlation between relative humidity and material porosity. The water absorption saturation is obtained by calculating the ratio of penetration depth to the thickness of the water-blocking strip. For the water absorption saturation change sequence, the mass increase per unit volume of the water-blocking strip is extracted. The water absorption volume is obtained by dividing the mass increase by the density of water. The volume expansion is obtained by subtracting the original volume from the sum of the water absorption volume and the original volume of the water-blocking strip. The ratio of the volume expansion to the original volume is calculated as the volume expansion rate. The time interval from the start of water level rise to the occurrence of measurable deformation of the water-blocking strip is recorded as the response delay time. Based on the response delay time and volume expansion rate sequence, the expansion rate difference between adjacent time points is extracted to obtain the expansion rate. The time when the expansion rate reaches its peak and the corresponding water level height are calculated. The expansion lag time is obtained by the time difference between the peak time and the time when the water level reaches that height. If the expansion lag time is less than a preset threshold, the expansion amplitude change data within that time period is obtained. The turning point of the change trend is identified by the difference between adjacent data points in the expansion amplitude change data. The time span between the turning points and the corresponding water level change height are calculated. The correlation coefficient is determined based on the ratio of the average expansion rate within the time span to the water level rise rate during the same period. When the correlation coefficient is greater than a preset correlation threshold, it is determined that there is a strong correlation between the water level rise and the expansion of the water-blocking strip.

[0031] Specifically, in one embodiment, the water level sensor is a piezoresistive sensor, arranged vertically along the wall of the power distribution room, with a measuring point set every 15 centimeters from the ground to form a vertical sensor array. The preset time interval is determined based on historical rainfall intensity data of the area where the power distribution room is located. When the rainfall intensity exceeds 50 mm / hour, the sampling interval is shortened to 2 seconds, and under normal rainfall conditions, a 5-second interval is maintained.

[0032] Specifically, determining the water penetration depth requires considering the microporous structure of the water-blocking strip material. Water-blocking strips are typically made of polyurethane or rubber-based composite materials, which contain numerous micropores. When water molecules enter these micropores, they diffuse inward along the pore channels. The relative humidity value detected by the humidity sensor has an exponential relationship with the penetration depth. When the surface humidity reaches 60%, the penetration depth is approximately 20% of the water-blocking strip thickness; when the humidity reaches 90%, the penetration depth can reach 65% of the thickness. Through a pre-established humidity-penetration depth calibration curve, real-time humidity data can be converted into an accurate penetration depth value. The water saturation is determined by the ratio of the penetration depth to the total thickness of the water-blocking strip; this saturation directly reflects the moisture distribution within the material.

[0033] It should be noted that the extraction of the mass increase is based on the water absorption characteristic curve of the water-blocking strip material. In the initial dry state, the mass per unit volume of the water-blocking strip is the baseline value; as the water absorption process proceeds, the material mass gradually increases. The mass increase is obtained by real-time monitoring of local mass changes using an embedded micro-weighing sensor. The water absorption volume is calculated by dividing the mass increase by the standard density of water, 1.0 g / cm³. The calculation of the volume expansion takes into account the nonlinear expansion characteristics of the material; that is, the volume increase is not simply equal to the water absorption volume, but rather the sum of the water absorption volume and the additional volume increase caused by the reorganization of the material's internal structure.

[0034] For example, the calculation of the volumetric expansion rate involves several intermediate parameters. When the water-blocking strip absorbs water, its original volume V0 changes, and the new volume V1 includes the original solid skeleton volume, the absorbed water volume, and the void volume created by the material expansion. The volumetric expansion rate is defined as (V1-V0) / V0, which shows a trend of rapid change followed by slower change during the water absorption process. In the first 30 seconds of water absorption, the expansion rate can reach 15% to 20%, and then gradually stabilizes. The response delay time is obtained by comparing the moment when the water level sensor first detects a rise in water level with the moment when the water-blocking strip begins to produce measurable deformation (deformation exceeding 1% of the original size). This delay time is typically in the range of 5 to 15 seconds, depending on the material type and ambient temperature.

[0035] Preferably, the peak expansion rate is identified using a difference method combined with moving average filtering. The time series of volumetric expansion rates is first-order differencing to obtain the instantaneous expansion rate series. Due to measurement noise, the original rate data fluctuates; after smoothing using a 5-point moving average filter, the local maximum point is identified as the peak value. The peak time and the corresponding water level height are determined by timestamp matching; the expansion lag time reflects the sensitivity of the water-blocking strip to changes in water level.

[0036] In one possible implementation, the expansion amplitude change data is acquired through high-frequency sampling. When the expansion lag time is less than a preset threshold of 8 seconds, a fast response mode is established, and the sampling frequency is increased to 10 times per second to record the expansion amplitude change within that time period in detail. The expansion amplitude is defined as the difference between the current volume expansion and the initial state, and this data sequence contains dynamic characteristic information of the expansion process. Furthermore, the identification of trend inflection points is based on curvature analysis. The expansion amplitude change data is subjected to second-order differencing, and when the sign of the second-order differencing value changes, it is marked as a potential inflection point. By setting a minimum interval threshold, overly dense inflection points are filtered out, retaining the main trend change nodes. The time span between inflection points reflects the stage characteristics of the expansion process, typically including three stages: a rapid expansion period, a transition period, and a stable period, with the time span ratio of each stage being approximately 2:1:3.

[0037] Understandably, the correlation coefficient calculation comprehensively considers both time synchronization and amplitude correlation. Time synchronization is assessed by calculating the time difference between the peak expansion rate and the peak water level rise rate; the smaller the time difference, the higher the synchronization. Amplitude correlation is obtained by calculating the Pearson correlation coefficient between the normalized expansion rate and the water level rise rate. The correlation coefficient is a weighted sum of a time synchronization weight of 0.4 and an amplitude correlation weight of 0.6. When the correlation coefficient exceeds 0.75, it is considered a strong correlation, indicating a high degree of consistency between the expansion of the water-blocking strip and the rise in water level; a coefficient between 0.5 and 0.75 indicates a moderate correlation; and a coefficient below 0.5 indicates a weak correlation, in which case the water-blocking strip material needs to be adjusted or the sensor arrangement optimized.

[0038] For example, in actual waterproofing tests of power distribution rooms, when heavy rain causes the water level to rise from 0 to 30 centimeters, the water-blocking strip under strong correlation can start to expand within 3 seconds after the water level rises. The peak expansion rate occurs when the water level reaches 15 centimeters. The entire expansion process maintains good synchronization with the water level change, realizing real-time tracking of water level changes and dynamic adjustment of sealing performance.

[0039] S102. Identify the expansion rate distribution of the water barrier in different regions based on the correlation between water level rise and expansion of the water barrier, and obtain the degree of expansion unevenness by comparing the expansion rate differences between different regions.

[0040] Based on the correlation between water level rise and expansion of the water-blocking strip, monitoring areas are divided at preset intervals along the length of the water-blocking strip. Local expansion rate data within each area are acquired. The difference in expansion rates between adjacent areas is calculated and divided by the area interval to obtain the rate gradient value. The distribution characteristics of the expansion rate are identified based on the change of the rate gradient value along the length of the water-blocking strip. For each area's data in the expansion rate distribution characteristics, the maximum value of the expansion rate within the monitoring period is extracted as the peak value, and the minimum value is extracted as the valley value. The ratio of the peak-valley difference to the average expansion rate of that area is calculated to obtain the local coefficient of variation. The spatial distribution dispersion is determined by the standard deviation of the coefficients of variation of all areas. If the spatial distribution dispersion exceeds a preset dispersion threshold, the deviation of the expansion rate of each area from the average rate of all areas is calculated. The degree of expansion non-uniformity is determined by the ratio of the root mean square of the deviation to the average expansion rate. The overall uniformity of the expansion of the water-blocking strip is judged by this degree of expansion non-uniformity.

[0041] Specifically, in one implementation, the water-blocking strip is divided into zones based on its physical structural characteristics. The water-blocking strip of the power distribution room waterproof baffle is typically 2 to 3 meters long, and is divided into monitoring zones of 20 centimeters each. An independent expansion monitoring sensor is deployed in each zone to collect local expansion rate data in real time.

[0042] Specifically, the rate gradient value reflects the degree of spatial variation in the expansion of the water-blocking strip. When water flows in from one side, the area closer to the inlet will first come into contact with the water and begin to expand, while areas farther from the inlet will experience a time delay. The rate gradient value, obtained by calculating the difference in expansion rates between adjacent areas and dividing by the distance between the areas, quantifies the swiftness of this spatial variation. A larger gradient value indicates a more significant difference in expansion between adjacent areas, which often occurs near the point of water impact or at defects in the water-blocking strip material. The expansion rate distribution characteristics are obtained by plotting the rate gradient along the length of the water-blocking strip, with the peak of the curve corresponding to the area of ​​most dramatic expansion.

[0043] It should be noted that the local coefficient of variation is used to measure the stability of the expansion process within a single region. During the monitoring period, the expansion rate of each region will fluctuate, with the peak value representing the maximum expansion rate and the trough value representing the minimum expansion rate. The difference between the peak and trough values ​​reflects the amplitude of the expansion process fluctuation. The coefficient of variation obtained by comparing it with the average expansion rate can dimensionlessly characterize the relative degree of fluctuation in the expansion of that region.

[0044] For example, the spatial distribution dispersion is obtained by calculating the standard deviation of the coefficient of variation of all regions. When the dispersion exceeds a preset threshold of 0.3, it indicates that there are significant differences in the expansion stability of each region, with some regions expanding smoothly while others fluctuate violently.

[0045] Preferably, the root mean square (RMS) method is used to calculate the degree of expansion non-uniformity. The deviation of the expansion rate of each region from the average rate of all regions reflects the difference between the local and the overall situation, and the root mean square of the deviation provides a comprehensive measure of this difference. The ratio obtained by comparing the RMS value with the average expansion rate is the degree of expansion non-uniformity. This value is usually between 0.1 and 0.5, and a value exceeding 0.3 is considered significantly non-uniform.

[0046] S103. If the degree of expansion unevenness exceeds the preset expansion threshold, the expansion rate distribution is analyzed by spectrum frequency band analysis to identify the expansion rate change region, the potential location of the leakage channel is identified by the expansion rate change region, and the water flow bypass path data of the potential location of the leakage channel is obtained.

[0047] If the degree of expansion unevenness exceeds a preset expansion threshold, a Fast Fourier Transform is performed on the expansion rate distribution data to obtain the amplitude and phase spectra in the frequency domain. Abnormal frequency bands are identified by frequency components in the amplitude spectrum that exceed a preset multiple of the average amplitude. The spatial location of the abnormal frequency band determines the expansion rate abrupt change region. For the boundary of the expansion rate abrupt change region, the difference in expansion rate between adjacent measuring points at the boundary is extracted and divided by the distance between the measuring points to obtain the gradient value. When the gradient value exceeds the preset abrupt change threshold, continuous sampling is performed along the direction of the fastest decrease in expansion rate to form a tracking path. The intersection of the tracking path and the sealing surface of the water-blocking strip determines the potential location of the leakage channel. Based on the potential location of the leakage channel, the pressure sensor data pre-deployed at that location is read. The pressure gradient vector is calculated by the ratio of the pressure difference between adjacent sensors to the distance between them. The direction change of the pressure gradient vector identifies the water flow turning point, and the turning point is connected to form the initial path for water flow to bypass the barrier. By using the pressure data at each point on the initial path, the pressure drop per unit length is calculated as the ratio of the pressure drop between adjacent points to the path length. The path resistance coefficient is determined based on the ratio of the pressure drop per unit length to the product of the water density and the gravitational acceleration. When the resistance coefficient is less than a preset flow threshold, the water flow bypass path data of the potential location of the leakage channel is obtained.

[0048] Specifically, in one implementation, when the degree of expansion unevenness exceeds a preset threshold of 0.35, it indicates an increased possibility of local sealing defects in the water-blocking strip. At this point, a frequency domain analysis program is initiated to perform in-depth processing on the collected spatiotemporal distribution data of the expansion rate.

[0049] Specifically, the Fast Fourier Transform (FFT) in this application transforms the spatial distribution of the expansion rate into the frequency domain for analysis. The expansion rate distribution along the length of the water-blocking strip can be considered a one-dimensional spatial signal containing components of different spatial frequencies. Normal uniform expansion corresponds to low-frequency components, while local abrupt changes manifest as high-frequency components. The amplitude spectrum obtained through the FFT reflects the intensity of each frequency component, while the phase spectrum reflects the relative position of each frequency component. When the amplitude of a certain frequency component exceeds 2.5 times the average amplitude of all frequency components, that frequency is marked as an abnormal frequency band. The abnormal frequency band is mapped back to the spatial domain through the inverse Fourier transform, and its corresponding spatial location is the region of abrupt expansion rate changes. This method can effectively filter out measurement noise and accurately locate the true location of expansion anomalies.

[0050] It should be noted that significant changes in physical characteristics often exist at the boundaries of regions where expansion rates abruptly change. The difference in expansion rates on both sides of the boundary forms a gradient field, and the magnitude of the gradient value reflects the drastic spatial variation in the expansion rate.

[0051] For example, the gradient tracing path is formed using the steepest descent method. Starting from the high expansion rate point at the boundary of the abrupt change region, continuous sampling is performed along the direction of the fastest decrease in expansion rate. The position of each sampling point is determined by the gradient direction of the current point. In practice, expansion rate values ​​are calculated in eight directions around the current sampling point, and the direction with the largest decrease is selected as the position of the next sampling point. The path formed by continuous sampling gradually approaches the local minimum region of expansion rate. When the tracing path intersects with the sealing surface of the water-blocking strip, the intersection point usually corresponds to a weak point in water seepage. This is because water flow preferentially passes through the path of least resistance, causing the expansion of the water-blocking strip to be hindered at that point, forming a local trough in expansion rate. The accuracy of identifying potential leakage channels using this method can reach over 85%.

[0052] Preferably, the pressure sensors are pre-arranged in a grid pattern within the waterproof baffle system of the power distribution room, with a sensor spacing typically of 10 centimeters. Once a potential leak location is identified, real-time pressure data from the sensors at that location and in its surrounding area is read.

[0053] In one possible implementation, the pressure gradient vector is calculated based on the principles of hydrostatics. The pressure difference detected by adjacent sensors mainly consists of the water level difference and the kinetic pressure. The magnitude of the pressure gradient is obtained by dividing the pressure difference by the sensor spacing. The gradient direction points in the direction of the fastest pressure decrease, which is usually consistent with the water flow direction. When the water flow encounters an obstacle and changes direction, the direction of the pressure gradient vector changes significantly; locations with a change angle exceeding 30 degrees are identified as water flow inflection points. Furthermore, the formation process of the water flow bypass path reflects the adaptive behavior of the water flow after encountering the expansion of the water barrier. Water flow always tends to flow along the path of least resistance; when the main channel is blocked by the expanding water barrier, the water flow will seek an alternative path. By connecting the identified inflection points, the initial path outlines the actual flow trajectory of the water. This path typically exhibits an S-shaped or Z-shaped feature, with a path length 30% to 50% longer than the straight-line distance.

[0054] Understandably, pressure drop per unit length is a key parameter for assessing flow resistance in a path. A larger pressure drop indicates greater flow resistance in that section of the path, potentially indicating narrow passages or complex bends. The path resistance coefficient, obtained by normalizing pressure drop per unit length to fundamental physical parameters of water, reflects the degree to which the path geometry impedes water flow.

[0055] For example, in actual waterproofing tests of power distribution rooms, when the path resistance coefficient is less than 0.2, it indicates that there is a relatively unobstructed bypass channel, and the water can pass through at a high velocity. At this time, the complete water bypass path data obtained includes the three-dimensional coordinate sequence of the path, the estimated velocity of each segment, and the predicted total flow rate. These data provide accurate positioning basis for subsequent waterproofing reinforcement.

[0056] Anomalies in the expansion rate of the target area on the surface of the water-blocking strip are extracted, and changes in water pressure distribution near these anomalies are detected by a pressure sensor. The spatial overlap between the anomalies and the abrupt changes in expansion rate is analyzed to determine the coordinates of potential leakage channels. Simultaneously, fluid sensors are deployed around potential leakage channels to collect real-time data on water flow velocity and direction. The streamline trajectory and bypass distance of the water flow near the abrupt change area are recorded. The curvature and splitting ratio of the bypass path are analyzed, and the velocity attenuation and turbulence characteristics of the water flow in the path are evaluated to determine the location of the leakage channel and the bypass characteristics of the water flow.

[0057] Anomalies are identified by extracting data on the expansion rate of the water-blocking strip surface that exceed a preset multiple of the average value. The spatial coordinates of these anomalies are obtained, and water pressure data around them is read using pressure sensors. The ratio of the pressure difference between adjacent measuring points to their distance is calculated to obtain the water pressure gradient distribution. The spatial overlap is determined based on the distance between the extreme points of the water pressure gradient and the anomalies in the expansion rate. For areas where the spatial overlap exceeds a preset threshold, the geometric center coordinates of these areas are extracted as potential leakage channels. Velocity vector data from fluid sensors is read around these locations. The change in the angle between the velocity vectors of adjacent measuring points is used to identify flow inflection points. These inflection points are connected to form streamline trajectories, and the detour coefficient is calculated by the ratio of the trajectory length to the straight-line distance. Based on the streamline trajectory and the detour coefficient, the local radius of curvature of each segment of the trajectory is calculated. When the radius of curvature is less than a preset value, it is marked as a sharp bend. The proportion of sharp bends to the total length is statistically analyzed. Simultaneously, when velocity vectors show dispersion, branching points are identified, and the flow rate percentage of each branch is calculated as the branching ratio. The main leakage path is determined by the difference between the main flow and the tributaries. The velocity attenuation rate is obtained by dividing the velocity difference between the upstream and downstream of the main leakage path by the path length. The turbulence intensity is calculated based on the standard deviation of the velocity direction change angle of adjacent measuring points. If the turbulence intensity exceeds a preset threshold, it is determined to be the core area of ​​the leakage channel. The coordinates of the core area and the water flow bypass characteristic data are obtained.

[0058] Specifically, in one implementation, anomalies in expansion rate are identified using a statistical threshold method. By calculating the average and standard deviation of the expansion rate at all measuring points on the surface of the water-blocking strip, points exceeding the average plus twice the standard deviation are marked as anomalies. These anomalies are typically located at defects in the water-blocking strip material or in areas of concentrated water flow impact.

[0059] Specifically, determining the degree of spatial overlap involves spatial correlation analysis of two independent data sources. Extreme points of the water pressure gradient reflect abrupt changes in the water flow pressure field, while anomalies in the expansion rate reflect abnormal regions in the material response. A strong correlation is considered to exist when the spatial distance between the two is less than half the sensor grid spacing. The physical mechanism of this spatial overlap phenomenon lies in the formation of local high-pressure zones when water flows through weak points, while the expansion of the water-blocking strip at these points is restricted due to structural defects. By calculating the Euclidean distance between the two types of anomalies and normalizing it to the 0-1 interval, a spatial overlap index is obtained. When this index exceeds 0.7, it indicates that a leakage channel is highly likely to exist in the area. The geometric center coordinates are determined using the centroid method, by weighting the coordinates of all anomalies within the overlapping area, with the weight determined by the degree of anomaly at each point.

[0060] It should be noted that the velocity vector data from the fluid sensor contains two key pieces of information: the magnitude of the flow velocity and the direction of the flow. Changes in the velocity vector angle directly reflect the degree of change in the direction of the water flow.

[0061] For example, the formation of streamline trajectories is based on the principle of fluid continuity. Starting from the potential location of the leakage channel, the water flow path is traced along the velocity vector direction. When the velocity vector angle between adjacent measuring points exceeds 15 degrees, it is marked as a flow inflection point. The appearance of these inflection points is usually related to the local expansion morphology of the water-blocking strip; the uneven expansion and the resulting uneven surface force the water flow to change direction. The trajectory line formed by connecting all inflection points is the actual path of the water flow. The physical meaning of the detour coefficient is to quantify the degree to which the water flow bypasses obstacles; the larger the coefficient, the more complex the water flow detour. In the waterproofing system of the power distribution room, the detour coefficient should normally be less than 1.2. When the coefficient exceeds 1.5, it indicates the presence of severe flow channel blockage or complex flow detour.

[0062] Preferably, the local radius of curvature is calculated using a three-point circular arc fitting method. Three consecutive points on the streamline trajectory are selected, and an arc is defined by these three points; the radius of this arc is the local radius of curvature. A smaller radius of curvature indicates a more abrupt change in water flow direction.

[0063] In one possible implementation, the identification of flow splitting is based on the divergence characteristics of velocity vectors. A flow splitting point is identified when the velocity vector at a given measuring point differs in direction from the velocity vectors of multiple surrounding measuring points by more than 45 degrees. The splitting ratio is calculated by multiplying the velocity of each branch by its cross-sectional area; the main flow typically carries more than 60% of the flow. Furthermore, the velocity attenuation rate reflects the energy loss of water flow when passing through narrow or tortuous channels. The velocity attenuation per unit length is obtained by dividing the velocity difference between the beginning and end of the path by the path length. The attenuation rate for normal straight flow is approximately 0.1 m / s per meter, while it can reach over 0.3 m / s per meter when passing through complex flow paths.

[0064] Understandably, the calculation of turbulence intensity is based on the statistical characteristics of velocity direction. In laminar flow, the velocity directions of adjacent measuring points are basically consistent, with a standard deviation of less than 5 degrees for the angle of change. However, in turbulent flow, due to the presence of vortices and backflow, the velocity direction changes drastically, with a standard deviation exceeding 20 degrees. When the turbulence intensity exceeds a preset threshold of 0.3, it indicates the presence of strong flow disturbance in that region, typically corresponding to the core area of ​​a leakage channel.

[0065] For example, in a waterproofing test of a power distribution room, the above method successfully located a hidden water leakage channel. The channel was located at the joint of the water-blocking strip, and the water flow bypassed in an S-shape with a bypass coefficient of 1.8. The velocity attenuation rate of the main leakage path was 0.35 m / s per meter, and the turbulence intensity in the core area reached 0.45. The obtained water flow bypass characteristic data provided precise guidance for subsequent targeted repairs.

[0066] S104. Extract the expansion rate mutation feature parameters from the expansion rate mutation region, determine the sensor trigger frequency based on the expansion rate mutation feature parameters, and perform matching analysis between the sensor trigger frequency and the real-time water level sensing accuracy to obtain the matching degree between the sensor trigger frequency and the real-time water level sensing accuracy.

[0067] Peak points are extracted from the time-series data of regions with abrupt changes in expansion rate. The difference between the peak value and the average expansion rate of that region is calculated as the peak amplitude. Spectral analysis is performed on the data from the abrupt expansion region to obtain the dominant frequency component. The deviation between the dominant frequency and the frequency during stable expansion is calculated to obtain the frequency offset. The rate of change is determined based on the ratio of the difference in expansion rate between adjacent sampling points to the time interval. Based on the peak amplitude, frequency offset, and rate of change, these three are normalized and then added together according to preset weights to obtain a comprehensive characteristic value. The base trigger frequency is determined by the linear relationship between the comprehensive characteristic value and the sensor trigger frequency. When the rate of change exceeds a preset threshold, the base trigger frequency is increased proportionally to obtain the actual trigger frequency. The water level sensor is sampled using the actual trigger frequency, and the water level height sequence acquired at this frequency is recorded. The deviation between the difference between adjacent water level values ​​and the actual water level change is calculated. The root mean square value of the deviation is used to evaluate the water level sensing accuracy. Simultaneously, the response delay of the sensor from receiving the trigger signal to outputting water level data is recorded. The degree to which the trigger frequency meets the water level change sampling requirements is calculated based on the water level sensing accuracy and response delay, and the accuracy compliance is determined according to the ratio of the sensing accuracy to the preset accuracy standard. The product of the frequency compliance and the accuracy compliance is used as the matching degree between the sensor trigger frequency and the real-time water level sensing accuracy.

[0068] Specifically, in one implementation, the identification of abrupt expansion rate regions is based on the statistical characteristics of time-series data. Continuous expansion rate data segments are extracted using a sliding time window, and when the rate standard deviation within a certain time window exceeds 1.5 times the global standard deviation, the region corresponding to that window is marked as an abrupt expansion region.

[0069] Specifically, the calculation of peak amplitude involves two steps: determining the baseline value and calculating the difference. The baseline value is the average expansion rate of 10 sampling points before and after the abrupt change region. This method can eliminate the influence of the overall trend and highlight the characteristics of local abrupt changes. The peak point is identified using a local extremum search algorithm, that is, finding the position where the midpoint of a series of 5 consecutive sampling points is greater than all points on either side. The difference between the peak value and the baseline value reflects the severity of the abrupt change; the larger the difference, the more obvious the expansion anomaly of the water-blocking strip at that point. Spectrum analysis uses Discrete Fourier Transform to convert the time-domain signal to the frequency domain. The dominant frequency component is obtained by finding the frequency corresponding to the maximum value of the amplitude spectrum. The frequency during the stable expansion period is obtained by analyzing the periodicity of the expansion rate in the non-abrupt region, typically within the range of 0.1 to 0.5 Hz. The frequency offset reflects the degree of disorder in the expansion rhythm of the abrupt change region; the larger the offset, the more irregular the expansion process in that region.

[0070] It should be noted that the rate of change is calculated using the central difference method, which involves dividing the difference in expansion rates between two consecutive sampling points by twice the time interval. This method is more accurate than the one-sided difference method.

[0071] For example, the calculation process of the comprehensive feature value embodies the idea of ​​multi-parameter fusion. The three feature parameters are first normalized, mapping them to a unified dimension range of 0 to 1. Normalization uses a minimum-maximum standardization method, i.e., subtracting the minimum value from the current value and then dividing by the range. The preset weights are determined based on correlation analysis of historical data; the peak amplitude weight is typically set to 0.4, the frequency offset weight to 0.3, and the rate of change weight to 0.3. This weight allocation reflects the dominant influence of peak amplitude on the trigger frequency. The linear relationship between the comprehensive feature value and the sensor trigger frequency is obtained through least squares fitting, with the slope coefficient typically between 5 and 10 Hz and the intercept term around 2 Hz. When the rate of change exceeds a preset threshold, the dynamic adjustment of the trigger frequency adopts a piecewise linear strategy; for every 0.1 unit increase in the excess, the trigger frequency increases by 1 Hz. This adjustment mechanism provides denser sampling during rapid water level changes.

[0072] Preferably, the root mean square error (RMSE) index is used to evaluate the accuracy of water level sensing. The RMSE value is obtained by comparing the point-by-point differences between the sensor measurement sequence and the reference value sequence, calculating the square root of the average of the sum of squares of all differences.

[0073] In one possible implementation, response latency measurement involves timestamp recording and difference calculation. The moment the sensor receives the trigger signal is recorded via an interrupt service routine, and the moment the water level data is output is obtained via the timestamp in the packet header; the difference between the two is the response latency. Typical response latency is in the range of 10 to 50 milliseconds. Furthermore, frequency fit is evaluated based on the sampling theorem. The minimum required sampling frequency is calculated based on the highest frequency component of the water level change; the ratio of the actual trigger frequency to the minimum sampling frequency reflects the sufficiency of sampling. A ratio greater than 2 is considered sufficient sampling, between 1 and 2 is considered basically satisfactory, and less than 1 indicates insufficient sampling.

[0074] Understandably, the accuracy compliance is calculated by comparing the actual perceived accuracy with the preset accuracy standard. The preset accuracy standard is determined based on the waterproofing requirements of the power distribution room, and is usually set at 2% of the water level. When the actual accuracy is better than the standard, the compliance is greater than 1; when the actual accuracy is worse than the standard, the compliance is less than 1.

[0075] For example, in a test of a waterproofing system in a power distribution room, the comprehensive characteristic value of the expansion rate mutation area reached 0.75, the corresponding trigger frequency was adjusted to 8Hz, the water level sensing accuracy reached 1.5%, the response delay was 25 milliseconds, and the final calculated matching degree was 0.85, indicating that the sensor triggering strategy and the water level monitoring requirements have achieved a good fit.

[0076] S105. If the matching degree is lower than the preset matching threshold, the alarm threshold shall be determined according to the potential location of the leakage channel and the degree of uneven expansion.

[0077] If the matching degree is lower than the preset matching threshold, the location risk coefficient is calculated based on the Euclidean distance between the coordinates of the potential location of the leakage channel and the center point of the critical protection area. The closer the distance, the higher the risk coefficient. The expansion deviation coefficient is obtained by the ratio of the degree of expansion unevenness to the preset standard value. The location risk coefficient and the expansion deviation coefficient are added together according to preset weights to obtain the comprehensive risk value. The risk level is determined based on the position of the comprehensive risk value within the preset risk range, and the basic alarm threshold corresponding to the risk level is obtained. The preliminary alarm threshold is calculated by multiplying the basic alarm threshold and the comprehensive risk value. The preliminary alarm threshold is compared with the average alarm water level under the same historical risk level. If the deviation exceeds the preset range, a weighted average of the preliminary alarm threshold and the historical average value is used for correction. The alarm threshold is determined based on the corrected value.

[0078] Specifically, in one implementation, when the matching degree between the sensor trigger frequency and the water level sensing accuracy is less than 0.6, it indicates that the current monitoring strategy cannot meet the waterproofing requirements, and it is necessary to adjust the alarm threshold to compensate for the lack of monitoring accuracy.

[0079] Specifically, the location risk coefficient is calculated based on the inverse relationship between spatial distance and risk. Critical protection areas typically refer to the locations of core electrical equipment such as transformers, switchgear, and control cabinets within a power distribution room. The Euclidean distance between the potential location of the leakage path and the center point of these devices is calculated using three-dimensional coordinates, and the distance value is converted into a risk coefficient using an exponential decay function. When the distance is less than 1 meter, the risk coefficient is close to 1.0; when the distance is 3 meters, the risk coefficient is approximately 0.5; and when the distance exceeds 5 meters, the risk coefficient drops below 0.2. The expansion deviation coefficient is calculated as the ratio of the degree of expansion unevenness to the standard value of 0.15. A ratio greater than 1 indicates severe expansion unevenness, requiring a higher alert level. The comprehensive risk value uses weights of 0.6 and 0.4 for the location risk coefficient and expansion deviation coefficient, respectively, reflecting the dominant role of location factors.

[0080] It should be noted that the risk level classification adopts a three-tier system: a comprehensive risk value of 0 to 0.3 is considered low risk, 0.3 to 0.7 is medium risk, and 0.7 to 1.0 is high risk. The basic alarm thresholds for each risk level are 80%, 60%, and 40% of the water level height, respectively.

[0081] For example, the initial alarm threshold is calculated by multiplying the base alarm threshold by the comprehensive risk value. This method enables continuous adjustment of the alarm threshold. When the comprehensive risk value is 0.8, the base alarm threshold for the high-risk level is 40%, and the calculated initial alarm threshold is 32% of the water level height.

[0082] Preferably, the alarm threshold correction adopts a historical data verification mechanism. By querying the historical alarm record database, the average alarm level of the most recent 30 days under the same risk level is extracted. When the deviation between the initial alarm threshold and the historical average exceeds 15%, the initial value and the historical value are weighted and averaged with weights of 0.7 and 0.3 to obtain the final alarm threshold.

[0083] For example, if a water leakage channel in a power distribution room is 2 meters away from the transformer, the location risk coefficient is 0.75, the degree of uneven expansion is 0.2, and the calculated comprehensive risk value is 0.53, corresponding to a medium risk level. The final alarm threshold is determined to be 35% of the water level.

[0084] S106. Based on the alarm threshold, comprehensively analyze the water level rise rate and water flow bypass path data to determine the accurate water level warning signal. Based on the accurate water level warning signal, compare it with the historical water level warning signal database to identify the warning signal type. If the warning signal type is a high-risk emergency type, immediately send an alarm signal to the user terminal.

[0085] The water level margin is calculated based on the difference between the alarm threshold and the current water level. The estimated time to reach the alarm threshold is obtained by dividing the water level rise rate by the water level margin. The water flow influence coefficient is calculated by multiplying the ratio of the bypass path length to the straight-line distance and the average curvature of the path. The corrected warning time is obtained by dividing the estimated time by the water flow influence coefficient. A warning signal feature vector is constructed by sequentially arranging the normalized value of the corrected warning time, the ratio of the current water level to the equipment's safe water level, and the ratio of the water level rise rate to the historical average rise rate. The cosine similarity of this feature vector with the feature vectors in the historical water level warning signal database is calculated, and the warning type corresponding to the historical record with the highest similarity is selected as the current warning signal type. If the warning signal type is a high-risk emergency type, the preset user terminal communication address is obtained, an alarm data packet containing the current water level, estimated arrival time, and preset emergency response procedure number is constructed, and the alarm signal is sent to the user terminal through a preset communication protocol.

[0086] Specifically, in one implementation, real-time calculation of the water level margin provides the basic data for estimating the warning time. The difference between the alarm threshold and the current water level directly reflects the safety margin, and dividing this difference by the rate of water level rise yields the theoretical warning time.

[0087] Specifically, determining the flow influence coefficient involves a comprehensive assessment of the path's geometric characteristics. The ratio of the bypass path length to the ideal straight-line distance reflects the path's tortuousness; a larger ratio indicates that the flow needs to traverse a longer path to reach the monitoring point, thus extending the actual impact time. The average curvature of the path is obtained by calculating the arithmetic mean of the curvatures at various points along the path; a larger curvature indicates a more abrupt path turn and a greater loss of flow velocity. The flow influence coefficient, obtained by multiplying the path length ratio by the average curvature, is typically between 1.2 and 2.5. Dividing the predicted time by this coefficient yields the corrected warning time, which more accurately reflects the actual time it takes for the flow to reach the danger level. This correction method considers the delay effect of complex flow channels on flow propagation, improving the accuracy of warnings compared to simple linear predictions.

[0088] It should be noted that the three components of the early warning signal's feature vector represent the urgency of the situation, the danger of the water level, and the upward trend, respectively. Normalization makes the parameters with different dimensions comparable.

[0089] For example, cosine similarity is calculated by dividing the vector inner product by the product of the vector magnitudes. The historical water level warning signal database stores the feature vectors of all warning events in the past year and their corresponding actual risk levels. Historical records with a similarity greater than 0.85 are considered to have reference value.

[0090] Preferably, the criteria for determining a high-risk emergency include: a revised warning time of less than 30 minutes, a water level exceeding 70% of the equipment's safe water level, and a rise rate exceeding twice the historical average. Meeting any two of these conditions constitutes a high-risk situation.

[0091] In one possible implementation, the alarm data packet adopts a standardized format, including four essential fields: timestamp, water level value, estimated arrival time, and emergency response procedure number. The communication protocol supports three methods: SMS, application push, and voice call, ensuring reliable delivery of alarm information.

[0092] S107. Extract key parameters for water level alarm based on the accurate water level early warning signal, and determine the execution path for intelligent response to abnormal water level in the power distribution room based on the key parameters for water level alarm and the type of early warning signal.

[0093] Based on the numerical fields in the precise water level early warning signal, three key parameters are extracted: water level height, peak value of the expansion rate mutation amplitude, and coordinates of the potential location of the leakage channel. The water level risk factor is obtained by multiplying the ratio of the water level height to the preset safe water level by a weighting coefficient; the expansion risk factor is obtained by multiplying the ratio of the mutation amplitude to the historical average by a weighting coefficient; and the location risk factor is obtained by multiplying the reciprocal of the Euclidean distance between the location coordinates and the center of the key equipment by a weighting coefficient. These three risk factors are then added together according to preset weights to obtain a comprehensive risk index. The corresponding basic response level value is retrieved based on the early warning signal type, and the comprehensive risk index is multiplied by the basic response level to obtain the final response intensity value. Based on the final response intensity value, the corresponding interval is determined from multiple preset response intensity intervals. The execution instruction sequence and resource configuration parameters for that interval are obtained, and the intelligent response execution path for the power distribution room water level anomaly is determined by prioritizing the instructions in the instruction sequence.

[0094] Specifically, in one implementation, the precise water level early warning signal includes multiple numerical fields, which are parsed using field identifiers. The water level height field records the current measured water level value, the expansion rate mutation amplitude peak field stores the maximum mutation amount, and the potential location coordinate field of the leakage channel contains three-dimensional spatial coordinates.

[0095] Specifically, the calculation of the three risk factors reflects risk assessment from different dimensions. The water level risk factor reflects the degree of flooding risk by the ratio of the current water level to the preset safe water level; the closer the ratio is to 1, the closer it is to a dangerous state. This ratio is multiplied by a weighting coefficient of 0.4 to obtain the water level risk factor. The expansion risk factor is based on a comparison of the magnitude of the sudden change with the historical 30-day average. When the magnitude of the sudden change exceeds twice the average, the risk factor is close to its upper limit, and its weighting coefficient is set to 0.3 to reflect its contribution to the overall risk. The location risk factor is calculated using the reciprocal distance method. The closer the leakage channel is to critical equipment such as transformers and switchgear, the larger the reciprocal value, and the higher the risk. A weighting coefficient of 0.3 ensures that location factors are appropriately considered. The sum of the three weighting coefficients is 1, ensuring the normalization of the comprehensive risk index.

[0096] It should be noted that the basic response level is determined according to the type of warning signal: low-risk type corresponds to level 1, medium-risk type corresponds to level 2, and high-risk emergency type corresponds to level 3.

[0097] For example, the response intensity range is divided into four levels: 0-2 for routine monitoring, 2-4 for enhanced inspection, 4-6 for emergency preparedness, and greater than 6 for immediate response. Each range corresponds to a different sequence of execution instructions.

[0098] Preferably, the execution instruction sequence includes equipment power-off instructions, drainage pump start instructions, waterproof gate closing instructions, and personnel evacuation instructions. Priority is ordered according to the degree of safety impact, with equipment protection instructions having the highest priority, followed by waterproofing measures instructions, and finally early warning notification instructions.

[0099] This aspect also proposes a storage medium storing a water immersion alarm program, which, when executed by a processor, implements the steps of the water immersion alarm method based on an intelligent waterproof baffle.

[0100] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0101] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A water immersion alarm method based on an intelligent waterproof baffle, characterized in that, The steps of the water immersion alarm method based on an intelligent waterproof baffle include: Collect water level data and surface humidity data of water-blocking strips in the power distribution room to determine the water level rise rate and the initial expansion state of the water-blocking strips, and identify the correlation between water level rise and water-blocking strip expansion. Based on the correlation between the rise in water level and the expansion of the water-blocking strip, the expansion rate distribution of different regions of the water-blocking strip is obtained, the difference in expansion rate of each region is calculated, and the degree of expansion unevenness is determined. When the degree of expansion unevenness exceeds the preset expansion unevenness threshold, the expansion rate distribution is subjected to spectral analysis to identify the expansion rate abrupt change region, determine the potential location of the leakage channel, and obtain the water flow bypass path data of the potential location of the leakage channel. Feature parameters are extracted from the expansion rate abrupt change region, and the sensor triggering frequency is determined based on the feature parameters. The sensor triggering frequency is then matched with the water level sensing accuracy to obtain the matching degree. When the matching degree is lower than the preset matching degree threshold, an alarm threshold is determined based on the potential location of the leakage channel and the degree of expansion unevenness. Based on the alarm threshold, the water level rise rate, and the water flow bypass path data, a water level warning signal is generated. The warning signal type is determined by comparing the water level warning signal with a historical database. Based on the water level warning signal, extract the water level height, expansion rate change amplitude, and potential location of the leakage channel, and determine the response path for abnormal water level in the power distribution room by combining the warning signal type.

2. The water immersion alarm method based on an intelligent waterproof baffle as described in claim 1, characterized in that, The steps of collecting water level data and surface humidity data of the water-blocking strip in the power distribution room, determining the rate of water level rise and the initial expansion state of the water-blocking strip, and identifying the correlation between water level rise and water-blocking strip expansion include: Collect water level values ​​at different heights in the power distribution room, calculate the ratio of the water level difference between adjacent time points to the time interval, and determine the rate of water level rise; Collect surface humidity data of the water-blocking strip, determine the water absorption based on the humidity data and the material characteristics of the water-blocking strip, and determine the initial expansion state by the ratio of the water absorption to the original volume of the water-blocking strip. Based on the water level rise rate and the initial expansion state, the volumetric deformation at different water levels is calculated. By fitting the volumetric deformation with a time series, an expansion rate curve is generated. The slope change of the expansion rate curve is extracted to determine the boundary between accelerated expansion and stability.

3. The water immersion alarm method based on an intelligent waterproof baffle as described in claim 2, characterized in that, The steps of calculating the volumetric deformation at different water levels based on the water level rise rate and the initial expansion state, generating an expansion rate curve by fitting the volumetric deformation to a time series, extracting the slope change of the expansion rate curve, and determining the boundary point between accelerated and stable expansion include: Based on the water level rise rate and the initial expansion state, and combined with the water absorption characteristics of the water-blocking strip material, the volume expansion per unit time is calculated. An expansion rate curve is generated by fitting the volume expansion amount with a time series. Extract the point in the expansion rate curve where the slope changes from increasing to constant, and determine the constant point as the dividing point between accelerated expansion and stable expansion.

4. The water immersion alarm method based on an intelligent waterproof baffle as described in claim 1, characterized in that, The steps of obtaining the expansion rate distribution of different regions of the water-blocking strip based on the correlation between the water level rise and the expansion of the water-blocking strip, calculating the difference in expansion rate between each region, and determining the degree of expansion unevenness include: Based on the correlation between the rise in water level and the expansion of the water-blocking strip, the monitoring area of ​​the water-blocking strip is divided, and the expansion rate data of each area is obtained. Calculate the ratio of the expansion rate difference between adjacent regions to the distance between regions to determine the rate gradient; generate the expansion rate distribution based on the change of the rate gradient along the length of the water-blocking strip. Extract the peak and valley values ​​of each region in the expansion rate distribution, calculate the ratio of the peak-valley difference to the average expansion rate, and determine the local coefficient of variation; The degree of expansion non-uniformity is determined by the standard deviation of the local coefficient of variation.

5. The water immersion alarm method based on an intelligent waterproof baffle as described in claim 1, characterized in that, The step of performing spectral analysis on the expansion rate distribution, identifying regions of abrupt changes in expansion rate, determining potential locations of leakage channels, and obtaining water flow bypass path data for potential locations of leakage channels when the degree of expansion unevenness exceeds a preset threshold for expansion unevenness includes: Spectral analysis is performed on the expansion rate distribution to obtain the amplitude spectrum and phase spectrum. Frequency components in the amplitude spectrum that exceed a preset multiple of the average amplitude are extracted to determine the abrupt expansion rate regions. Based on the boundary of the abrupt expansion rate region, the ratio of the expansion rate difference between adjacent measuring points to the distance is calculated to determine the gradient value; Based on the gradient value sampled along the direction of decreasing expansion rate, a tracking path is generated. The potential location of the leakage channel is determined by the intersection of the tracking path and the sealing surface of the water-blocking strip. Based on the potential location of the leakage channel, read the pressure sensor data, calculate the ratio of the pressure difference between adjacent measuring points to the distance, and generate a pressure gradient vector; By determining the direction change of the pressure gradient vector, the water flow inflection point is identified, and the inflection point is connected to generate an initial path; Calculate the ratio of pressure drop at adjacent points on the initial path to the path length to determine the pressure drop per unit length; Based on the ratio of the pressure drop per unit length to the density of water and the acceleration due to gravity, water flow bypass path data is generated.

6. The water immersion alarm method based on an intelligent waterproof baffle as described in claim 1, characterized in that, The steps of extracting feature parameters from the expansion rate abrupt change region, determining the sensor trigger frequency based on the feature parameters, and matching the sensor trigger frequency with the water level sensing accuracy to obtain the matching degree include: Peak points are extracted from the regions of abrupt changes in expansion rate, and the difference between the peak points and the average expansion rate is calculated to determine the peak amplitude. Spectral analysis is performed on the data in the region of abrupt expansion rate changes to extract the deviation between the dominant frequency and the stable expansion frequency, and to determine the frequency offset; the ratio of the difference in expansion rate between adjacent sampling points to the time interval is calculated to determine the rate of change. A comprehensive feature value is generated by weighting the normalized peak amplitude, frequency offset, and rate of change; the sensor trigger frequency is then determined using the comprehensive feature value. Based on the sensor trigger frequency, water level data is sampled, the root mean square value of water level deviation is calculated, the water level sensing accuracy is determined, and the matching degree is generated.

7. The water immersion alarm method based on an intelligent waterproof baffle as described in claim 1, characterized in that, The step of determining the alarm threshold based on the potential location of the leakage channel and the degree of expansion unevenness when the matching degree is lower than the preset matching degree threshold includes: Calculate the location risk coefficient based on the distance between the potential location of the leakage channel and the center of the protected area; An expansion deviation coefficient is generated based on the ratio of the degree of expansion unevenness to the standard value; A comprehensive risk value is generated based on the location risk coefficient and the expansion deviation coefficient; The risk level is determined based on the comprehensive risk value, and an alarm threshold is generated.

8. The water immersion alarm method based on an intelligent waterproof baffle as described in claim 1, characterized in that, The step of generating a water level warning signal based on the alarm threshold, the water level rise rate, and the water flow bypass path data, and determining the warning signal type by comparing the water level warning signal with a historical database, includes: A water level margin is generated based on the difference between the alarm threshold and the current water level. An early warning time is generated based on the water level rise rate and the water flow bypass path data; Based on the warning time, current water level, and water level rise rate, a feature vector for the warning signal is constructed. The type of warning signal is determined by comparing the feature vector of the warning signal with the historical database.

9. The water immersion alarm method based on an intelligent waterproof baffle as described in claim 1, characterized in that, The step of extracting water level height, expansion rate abrupt change amplitude, and potential location of leakage channels based on the water level warning signal, and determining the response path for abnormal water level in the power distribution room in combination with the warning signal type, includes: The water level warning signal was analyzed to determine the water level height, the magnitude of the sudden change in the expansion rate, and the potential location of the leakage channel. Based on the water level height, the abrupt change in the expansion rate, and the potential location of the leakage channel, a water level risk factor, an expansion risk factor, and a location risk factor are generated, respectively. A comprehensive risk index is generated based on the water level risk factor, the expansion risk factor, and the location risk factor. Based on the comprehensive risk index and the warning signal type, the response path for abnormal water levels in the power distribution room is determined.

10. A storage medium, characterized in that, The storage medium stores a water immersion alarm program, which, when executed by a processor, implements the steps of the water immersion alarm method based on an intelligent waterproof baffle as described in any one of claims 1 to 9.

Images