A risk early warning method and system for steel cofferdam construction safety monitoring data

Abstract

The present application relates to the technical field of bridge engineering construction safety monitoring, and discloses a risk early warning method and system for steel cofferdam construction safety monitoring data, which comprises the following steps: obtaining construction state data and safety monitoring data, determining the current construction stage and identifying the stress state adjustment event; obtaining the monitoring response benchmark according to the safety monitoring data before the event and the matching reference monitoring record; comparing the safety monitoring data after the event with the monitoring response benchmark to obtain the monitoring deviation result, screening the abnormal monitoring position and determining the temporary fluctuation state or the continuous abnormal state; forming the risk characteristics under the continuous abnormal state, matching the preset risk determination condition to determine the risk type, the risk position and the early warning level, and forming the risk early warning result. The present application distinguishes the construction disturbance from the continuous abnormality through the construction event, the response benchmark and the change sequence, thereby improving the accuracy of the risk position determination.

Description

Technical Field

[0001] This invention relates to the field of bridge construction safety monitoring technology, and more specifically, to a risk early warning method and system for monitoring safety data during steel cofferdam construction. Background Technology

[0002] Steel cofferdams are widely used in the construction of bridge piers and other water-related foundation projects. During the construction of steel cofferdams, monitoring points are typically set up at structural components such as sheet piles, pipe piles, walers, and internal supports. Data such as displacement gauges, strain gauges, water level gauges, total stations, and levels are collected using equipment including displacement gauges, component stress, support deformation, and water levels inside and outside the cofferdam. Monitoring personnel adjust the monitoring frequency according to the construction progress, such as cofferdam driving, bottom sealing, dewatering, internal support installation, pier construction, and support removal. The monitoring results are compared with pre-set control values, and on-site inspections, monitoring verification, and safety measures are carried out based on the comparison results.

[0003] For example, Chinese invention patent application CN118657380A discloses an intelligent monitoring system and method for bridge steel cofferdam construction based on digital twins. This scheme calculates risk assessment scores based on the actual, satisfactory, and unacceptable values ​​of various indicators, then calculates the total safety risk score according to the indicator weights, and determines the warning level by comparing the monitored values ​​with the warning values. When using the above method, the risk result mainly depends on the numerical range of each monitored value and its weighted result. During the continuous construction operations such as pumping, internal support installation, or support removal, multiple monitoring data may change synchronously with the construction operations. Continuous anomalies at a few key measuring points may also be diluted by the stable data of other monitoring indicators. Therefore, it is difficult to determine from the total risk score whether the risk increase is due to temporary changes caused by construction operations or to the continuous deterioration of the local structural condition, and it is also difficult to identify the construction links or structural parts that need to be checked first.

[0004] Therefore, it is necessary to design a risk early warning method and system for monitoring the safety data of steel cofferdam construction in order to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a risk early warning method and system for steel cofferdam construction safety monitoring data, aiming to solve the problem that existing risk assessment methods are unable to distinguish between temporary monitoring fluctuations caused by construction operations and continuous deterioration of local structural conditions, and are unable to accurately determine the source of risk and key inspection areas.

[0006] This invention proposes a risk early warning method for safety monitoring data during steel cofferdam construction, comprising: Acquire construction status data and safety monitoring data of the steel cofferdam, determine the current construction stage based on the construction status data, and identify stress state adjustment events that cause changes in the stress state of the steel cofferdam. Based on the safety monitoring data prior to the stress state adjustment event, and the reference monitoring records that match the current construction stage and the stress state adjustment event, a monitoring response benchmark is obtained. The safety monitoring data after the stress state adjustment event occurs is compared with the monitoring response benchmark to obtain the monitoring deviation results of each monitoring location; abnormal monitoring locations are screened according to the monitoring deviation results, and temporary fluctuation state or continuous abnormal state is determined according to the monitoring deviation results of the abnormal monitoring locations during the continuous monitoring period; When the continuous abnormal state is determined, risk characteristics are formed based on the spatial distribution of the abnormal monitoring locations and the order of changes in different safety monitoring data. The risk characteristics are matched with preset risk determination conditions to determine the risk type and risk location; The warning level is determined based on the duration of the continuous abnormal state, the distribution range of the abnormal monitoring locations, and the degree of deviation, and a risk warning result is generated.

[0007] Furthermore, processing the security monitoring data includes: The water levels inside and outside the cofferdam, the horizontal displacement of the pile top, and the deflection of the internal support are aligned according to the collection time. The water level difference is determined based on the water levels inside and outside the cofferdam, forming a safety monitoring sequence for each monitoring location.

[0008] Further, determining the current construction stage and identifying the stress state adjustment event includes: The current construction stage and internal support setting status are determined based on the construction procedure records in the construction status data; when the construction procedure records indicate the start of pumping, installation of internal supports, or removal of internal supports, the stress state adjustment event is identified, and the event type and event start time are determined.

[0009] Furthermore, obtaining the monitoring response benchmark includes: Reference monitoring records that match the current construction stage, event type, and internal support setting status are selected from historical qualified construction records; the safety monitoring data before the event is adjusted according to the stress state to determine the current monitoring benchmark value; the reference change range is determined according to the monitoring changes before and after the event in the reference monitoring records; the reference change range is superimposed on the current monitoring benchmark value to obtain the monitoring response benchmark.

[0010] Furthermore, obtaining the monitoring deviation results and filtering the abnormal monitoring locations includes: The safety monitoring sequence after the start time of the event is compared with the monitoring response benchmark, and the monitoring deviation result is obtained according to whether it exceeds the monitoring response benchmark and the amount of excess; when the monitoring deviation reaches the deviation threshold and continuously reaches the confirmation cycle, the corresponding monitoring location is determined as the abnormal monitoring location.

[0011] Further, determining the temporary fluctuation state or the persistent abnormal state includes: The monitoring deviation results of the abnormal monitoring location are reviewed during the continuous monitoring period. When the monitoring deviation decreases continuously within the review period and returns to the monitoring response benchmark limit, it is determined to be the temporary fluctuation state. When the monitoring deviation does not return to the limit and continues to reach the abnormal time limit, or when the monitoring deviation of the same type of safety monitoring data of the monitoring location spatially adjacent to the abnormal monitoring location reaches the deviation threshold and continuously reaches the confirmation period, it is determined to be the continuous abnormal state.

[0012] Furthermore, the formation of the aforementioned risk characteristic includes: The abnormal monitoring locations that are in the continuous abnormal state are formed into abnormal monitoring areas according to their spatial adjacency, wherein isolated abnormal monitoring locations form separate abnormal monitoring areas. Based on the data types of safety monitoring that deviate in each abnormal monitoring area and the time when they first reach the deviation threshold, determine the composition and sequence of the deviations. The risk characteristics are formed based on the abnormal monitoring area, the composition of the deviation, and the sequence of changes.

[0013] Furthermore, determining the risk type and the risk location includes: When the water level difference deviates first, and the horizontal displacement of the pile top and the deflection of the internal support subsequently deviate, and the deviation trend is consistent with the direction of the force response to the change in water level difference, it is identified as a risk of water level difference change. When the deflection of the internal support deviates before the horizontal displacement of the adjacent pile top, and the water level difference does not deviate first, it is determined to be an abnormal risk of the internal support. When the horizontal displacement at the top of the pile deviates first and extends to the adjacent monitoring location, the water level difference is within the range defined by the monitoring response benchmark, and the deflection of the internal support is always within the range defined by the monitoring response benchmark, it is determined to be a risk of local deformation of the retaining structure.

[0014] When the risk characteristics do not meet the above risk determination conditions, it is determined to be an unmatched persistent abnormal risk; The location of the risk is determined based on the risk type and the safety monitoring data showing deviations within the abnormal monitoring area.

[0015] Furthermore, determining the warning level and generating the risk warning result includes: The duration of the continuous abnormal state, the number of spatially continuous abnormal monitoring locations, and the maximum monitoring deviation are compared with the corresponding classification thresholds to obtain the duration level, distribution range level, and deviation level; the highest level is determined as the warning level, and the risk warning result is formed based on the risk type, risk location, and warning level.

[0016] Compared with existing technologies, the advantages of this invention are as follows: By combining construction status data with safety monitoring data, the current construction stage is first determined and stress state adjustment events that cause changes in the stress state of the steel cofferdam, such as pumping, installation or removal of internal supports, are identified. Then, safety monitoring data before the event occurs and matching reference monitoring records are used to form a monitoring response benchmark, so that subsequent judgments no longer rely solely on fixed alarm values, but use the normal monitoring response under the current construction operation as the comparison object. By comparing the safety monitoring data after the event occurs with the monitoring response benchmark, abnormal monitoring locations are screened, and temporary fluctuation states and continuous abnormal states are distinguished by combining the monitoring deviation results within the continuous monitoring period, which can reduce misjudgments caused by short-term disturbances in construction operations. After determining the continuous abnormal state, risk characteristics are formed based on the spatial correlation between abnormal monitoring locations and the sequential relationship of changes in different safety monitoring data, and the risk type, risk location and warning level are determined, so that the warning results can point to specific risk sources and key inspection areas, facilitating the implementation of corresponding on-site measures.

[0017] On the other hand, this application also provides a risk early warning system for steel cofferdam construction safety monitoring data, used to apply the aforementioned risk early warning method for steel cofferdam construction safety monitoring data, including: The status recognition unit is used to acquire construction status data and safety monitoring data of the steel cofferdam, determine the current construction stage based on the construction status data, and identify stress state adjustment events that cause changes in the stress state of the steel cofferdam. The benchmark determination unit is used to obtain a monitoring response benchmark based on the safety monitoring data prior to the occurrence of the stress state adjustment event and the reference monitoring records that match the current construction stage and the stress state adjustment event. An anomaly determination unit is used to compare the safety monitoring data after the occurrence of the stress state adjustment event with the monitoring response benchmark to obtain the monitoring deviation results of each monitoring location; to filter abnormal monitoring locations based on the monitoring deviation results, and to determine a temporary fluctuation state or a continuous abnormal state based on the monitoring deviation results of the abnormal monitoring locations during a continuous monitoring period. The feature forming unit is used to form risk features based on the spatial correlation between the abnormal monitoring locations and the sequential relationship of changes in different safety monitoring data when the continuous abnormal state is determined. The risk determination unit is used to match the risk characteristics with preset risk judgment conditions to determine the risk type and risk location; The early warning generation unit is used to determine the early warning level based on the duration of the continuous abnormal state, the distribution range of the abnormal monitoring locations, and the degree of deviation, and to generate a risk early warning result.

[0018] It is understandable that the risk warning methods and systems for the aforementioned steel cofferdam construction safety monitoring data have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0019] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating a risk warning method for steel cofferdam construction safety monitoring data provided in this embodiment of the invention; Figure 2 A flowchart for determining abnormal monitoring states provided in an embodiment of the present invention; Figure 3 A functional block diagram of a risk warning system for steel cofferdam construction safety monitoring data provided in an embodiment of the present invention. Detailed Implementation

[0020] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey its scope to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] In some embodiments of this application, see Figures 1-2 As shown, this application proposes a risk early warning method for steel cofferdam construction safety monitoring data, including: S100: Acquire construction status data and safety monitoring data of the steel cofferdam, determine the current construction stage based on the construction status data, and identify stress state adjustment events that cause changes in the stress state of the steel cofferdam. S200: Based on the safety monitoring data prior to the occurrence of the stress state adjustment event, and the reference monitoring records that match the current construction stage and the stress state adjustment event, a monitoring response benchmark is obtained; S300: Compare the safety monitoring data after the stress state adjustment event with the monitoring response benchmark to obtain the monitoring deviation results of each monitoring location; screen abnormal monitoring locations based on the monitoring deviation results, and determine the temporary fluctuation state or the continuous abnormal state based on the monitoring deviation results of the abnormal monitoring locations during the continuous monitoring period; S400: When a continuous abnormal state is determined, risk characteristics are formed based on the spatial distribution of abnormal monitoring locations and the order of changes in different safety monitoring data. S500: Match risk characteristics with preset risk assessment conditions to determine the risk type and risk location; S600: Determine the warning level based on the duration of the continuous abnormal state, the distribution range of the abnormal monitoring locations, and the degree of deviation, and generate a risk warning result.

[0022] In one embodiment, this method is executed by a monitoring processor located at the steel cofferdam construction site. The monitoring processor can be a construction site server, industrial controller, or edge computing device, and is communicatively connected to a construction management terminal, safety monitoring equipment, and an early warning output terminal. The construction management terminal records the status of the steel cofferdam construction process; the safety monitoring equipment collects data on water levels inside and outside the cofferdam, horizontal displacement of the pile tops, and deflection of the internal supports; the early warning output terminal outputs risk warning results. The safety monitoring equipment can be a total station, level, water level gauge, displacement gauge, or deflection monitoring equipment. Its installation, calibration, sampling, communication, and conventional filtering methods can utilize existing technologies, and this embodiment is not limited to these methods.

[0023] This implementation method is applicable to steel sheet pile cofferdams, interlocking steel pipe pile cofferdams, or other steel cofferdams employing vertical retaining structures and horizontal internal supports. Before construction, monitoring position markers are set according to the plan layout and internal support arrangement of the steel cofferdam, and position association records are established. The pile top horizontal displacement monitoring positions can be set at the corners of the cofferdam, the middle of each sidewall, and designated positions along the length of the sidewall; the internal support deflection monitoring positions can be set at the mid-span or quarter-point of the internal support; the internal and external water level monitoring positions are set at locations that reflect the internal and external water levels of the cofferdam, respectively. The position association records are used to record the spatial relationship between adjacent pile top monitoring positions, as well as the spatial relationship between the internal support monitoring position and the pile top monitoring position near the end of the internal support.

[0024] Construction status data is generated by the construction management terminal based on on-site process confirmation information. When construction personnel begin pumping water, installing internal supports, or dismantling internal supports, they submit a process start record in the construction management terminal; upon completion of the corresponding construction operation, they submit a process completion record. The monitoring processor determines the current construction stage based on the construction process record and determines the internal support setting status based on the internal supports that have been installed but not yet dismantled. For example, when the bottom sealing concrete reaches the construction requirements and the water level inside the cofferdam begins to decrease, the current construction stage is the pumping stage; when the first layer of internal supports is installed and pumping continues, the internal support setting status is "first layer of internal supports installed"; when the pier cap construction is completed and the dismantling of internal supports begins, the current construction stage is the internal support dismantling stage.

[0025] A stress state adjustment event is an event that changes the load state or constraint state of a steel cofferdam due to construction operations. When the construction procedure record indicates the start of dewatering, installation of internal supports, or removal of internal supports, the monitoring processor identifies the stress state adjustment event, determines the construction operation type as the event type, and determines the start time of the procedure as the event start time. Dewatering changes the water level difference inside and outside the cofferdam; the installation of internal supports increases the horizontal constraint of the retaining structure; the removal of internal supports decreases the horizontal constraint of the retaining structure. All of these construction operations may cause changes in the horizontal displacement of the pile top or the deflection of the internal supports; therefore, changes in monitoring data after the construction operation are not directly classified as risks. When one stress state adjustment event is not yet completed and another stress state adjustment event begins, the start time of the latter stress state adjustment event is used as the dividing time. Monitoring data before the dividing time is assigned to the former stress state adjustment event, and monitoring data after the dividing time is assigned to the latter stress state adjustment event.

[0026] After the safety monitoring data is collected, the monitoring processor aligns it according to the collection time. Specifically, a unified monitoring cycle is set according to the collection cycle of the safety monitoring equipment; when the collection time of the safety monitoring data is consistent with the unified monitoring time, the safety monitoring data is used directly; when the collection times are inconsistent, the most recent valid collection value before and after the unified monitoring time is selected for interpolation, or the most recent collection value with a time difference not exceeding the allowable time difference is selected. If there is no collection value that meets the allowable time difference requirement, the corresponding data of that monitoring cycle is marked as invalid, and invalid data is not used for monitoring deviation judgment. At the same monitoring time, the difference between the water level outside the cofferdam and the water level inside the cofferdam is processed to obtain the water level difference, and the water level difference, pile top horizontal displacement, and internal support deflection are arranged according to the collection time to form a safety monitoring sequence. The pile top horizontal displacement is recorded according to the pre-set positive direction, and the internal support deflection is recorded according to the deflection direction and deflection value relative to the initial straight position.

[0027] When obtaining the monitoring response baseline, reference monitoring records are first selected from historical qualified construction records. These records can originate from stress state adjustment events already completed for the current steel cofferdam, or from other steel cofferdams with matching structural form, construction conditions, and monitoring layout. Construction records that did not generate structural safety warnings during historical construction, passed cofferdam acceptance upon completion, and have no continuous data gaps are considered historical qualified construction records. When selecting reference monitoring records, the construction stage, event type, and internal support installation status are compared. Historical qualified construction records matching the current stress state adjustment event in all three aspects are selected as reference monitoring records. When the current project enters a construction stage for the first time and no historical qualified construction records meet the selection criteria, initial reference monitoring records can be generated through low-speed pumping, small-scale water level adjustments, or stability observation after single-layer internal support installation. These records are then updated as similar qualified construction records are subsequently obtained.

[0028] The current monitoring baseline value is determined based on the safety monitoring data prior to the event, adjusted according to the current stress state. Specifically, a baseline acquisition period is set before the event's start time. This baseline acquisition period can be selected as six consecutive valid monitoring cycles before the event's start time, or as 30 or 60 minutes prior. For the same type of safety monitoring data, the difference between the maximum and minimum values ​​within the baseline acquisition period is calculated. When this difference does not exceed the stability fluctuation limit, the median or arithmetic mean of the valid data within that period is determined as the current monitoring baseline value. When this difference exceeds the stability fluctuation limit, the baseline acquisition period is shifted forward by one monitoring cycle until continuous data meeting stability requirements are obtained. The stability fluctuation limit is obtained based on the measurement accuracy of the monitoring equipment and data fluctuations within historical stable construction periods. The differences between adjacent monitoring cycles within historical stable construction periods can be sorted from smallest to largest, and the difference at the 95th percentile after sorting is taken as the historical stability fluctuation value. The larger of the historical stability fluctuation value and twice the allowable error of the monitoring equipment is taken as the stability fluctuation limit.

[0029] When determining the reference change range, a reference baseline value before the event is determined for each reference monitoring record. Using the start time of the reference event as the starting point, the corresponding reference baseline value is subtracted from each monitoring value after the event to obtain the reference change amount. This is then categorized into multiple response intervals according to the relative time after the event. For each response interval and each type of safety monitoring data, the reference change amounts are sorted from smallest to largest. The reference change amount at the 5th percentile after sorting is taken as the lower limit of the reference change, and the reference change amount at the 95th percentile after sorting is taken as the upper limit of the reference change. If the number of reference monitoring records is insufficient to form a stable sorting boundary, monitoring equipment failure data can be removed, and the minimum and maximum values ​​of the remaining reference change amounts can be taken as the lower and upper limits of the reference change, respectively. The lower and upper limits of the reference change are then superimposed on the current monitoring baseline value to obtain the monitoring response baseline for the current stress state adjustment event within the corresponding response interval. The monitoring response baseline includes a lower and upper limit that change with the duration of the event.

[0030] When obtaining monitoring deviation results, the response interval of the current safety monitoring data is determined based on the duration of the stress state adjustment event, and the safety monitoring data is compared with the monitoring response benchmark for that interval. If the safety monitoring data is between the lower and upper limits of the benchmark, the monitoring deviation result is no deviation; if the safety monitoring data is above the upper limit of the benchmark, the monitoring deviation result is a positive deviation, and the difference between the safety monitoring data and the upper limit of the benchmark is determined as the monitoring deviation amount; if the safety monitoring data is below the lower limit of the benchmark, the monitoring deviation result is a negative deviation, and the difference between the lower limit of the benchmark and the safety monitoring data is determined as the monitoring deviation amount.

[0031] Deviation thresholds are set for water level difference, pile top horizontal displacement, and internal support deflection. These thresholds are used to exclude monitoring equipment errors and normal short-term fluctuations. To determine the deviation threshold, the maximum normal exceedance of each monitoring value relative to the monitoring response baseline boundary can be extracted from historical qualified construction records. All maximum normal exceedances are sorted from smallest to largest, and the maximum normal exceedance at the 95th percentile after sorting is taken as the historical deviation limit. Then, the equipment deviation limit is determined based on the monitoring equipment's accuracy, and the larger of the historical deviation limit and the equipment deviation limit is determined as the deviation threshold. The confirmation cycle is used to avoid directly determining the abnormal monitoring location when data fluctuations occur in a single monitoring cycle; this can be obtained through historical data playback. The minimum number of cycles where the false alarm rate does not exceed the set false alarm rate and the abnormal detection time meets the on-site handling requirements is selected as the confirmation cycle.

[0032] When the deviation of a monitoring location reaches the corresponding deviation threshold and continues for a certain period of time, the monitoring location is identified as an abnormal monitoring location. The monitoring processor also records the time when the monitoring location first reaches the deviation threshold; this time is used to determine the order of changes in different types of safety monitoring data.

[0033] If the safety monitoring data is within the monitoring response baseline limit, or if it exceeds the monitoring response baseline but the monitoring deviation does not reach the corresponding deviation threshold, or if the monitoring deviation reaches the corresponding deviation threshold but does not continuously reach the confirmation cycle, then the corresponding monitoring location will not be determined as an abnormal monitoring location at the current judgment time. The above situations do not mean that the safety monitoring data cannot continue to change, nor do they mean that abnormal monitoring locations will not be formed in subsequent monitoring cycles; the monitoring processor can continue to obtain monitoring deviation results and make judgments in subsequent monitoring cycles in the above manner.

[0034] After identifying abnormal monitoring locations, the monitoring deviation results for these locations are reviewed within a continuous monitoring period. The review and abnormality time limits are derived from the normal recovery process in the reference monitoring records. Specifically, for the same construction stage, event type, and internal support setup status, the time required for the monitoring data in each reference monitoring record to first exceed the monitoring response benchmark and then re-enter the benchmark's defined range is calculated to obtain the normal recovery time. The normal recovery times are then sorted from smallest to largest, and the normal recovery time at the 75th percentile after sorting is taken as the review time limit, while the normal recovery time at the 95th percentile after sorting is taken as the abnormality time limit. When the number of reference monitoring records is insufficient to form a stable sorting boundary, the duration of three consecutive monitoring cycles can be used as the review time limit, and the duration of six consecutive monitoring cycles can be used as the abnormality time limit.

[0035] If, within the review period, the monitoring deviation at an abnormal monitoring location continuously decreases and re-enters the monitoring response baseline limit, it is determined to be a temporary fluctuation state. Continuous decrease means that the monitoring deviation in the subsequent monitoring cycle is less than that in the previous monitoring cycle. When the monitoring deviation does not return to the monitoring response baseline limit and continues to reach the abnormal time limit, it is determined to be a persistent abnormal state. If the abnormal monitoring location has not yet continuously reached the abnormal time limit, but the safety monitoring data of the same type at a spatially adjacent monitoring location reaches the corresponding deviation threshold and continuously reaches the confirmation cycle, it is also determined to be a persistent abnormal state. This judgment indicates that the same type of monitoring deviation has expanded from the original abnormal monitoring location to spatially adjacent monitoring locations and is no longer treated as a single-point temporary fluctuation.

[0036] If the monitoring deviation at an abnormal monitoring location does not continuously decrease and return to the monitoring response baseline range within the review period, and does not continuously reach the abnormal time limit, and the safety monitoring data of the same type at spatially adjacent monitoring locations does not reach the deviation threshold and continuously reach the confirmation cycle, then at the current judgment time, it is uncertain whether it is a temporary fluctuation state or a continuous abnormal state. This abnormal monitoring location can remain a record pending review, and the monitoring deviation result will be updated as new monitoring data is added during the continuous monitoring period; if the judgment conditions for a temporary fluctuation state or a continuous abnormal state are met later, the corresponding state will be determined. This record pending review is not considered an independent risk type, nor does it participate in risk characteristic matching.

[0037] When establishing risk characteristics, abnormal monitoring locations in a continuously abnormal state are grouped into abnormal monitoring areas based on their spatial adjacency. Spatial adjacency is determined using location association records. Two pile top horizontal displacement monitoring locations sequentially set along the same cofferdam sidewall are spatially adjacent; a monitoring location at the cofferdam corner and the nearest monitoring location on the intersecting sidewalls are also spatially adjacent; an internal support monitoring location and a pile top monitoring location near the end of that internal support have a spatial association relationship. Abnormal monitoring locations not adjacent to other abnormal monitoring locations form separate abnormal monitoring areas.

[0038] For water level differences, the group of water level monitoring points inside and outside the cofferdam that forms the water level difference can be used as the water level difference monitoring location unit. When the water levels inside and outside the cofferdam are monitored using a shared monitoring point for the entire cofferdam, the water level difference serves as common safety monitoring data for all abnormal monitoring areas, participating in the judgment of the deviation composition and the order of change in each abnormal monitoring area. When water level monitoring points are set up separately along different sidewalls or in different flow directions on the outer side of the cofferdam, the outer water level monitoring point on the same side as the abnormal monitoring area or the one closest to it is selected, and a water level difference is formed between it and the water level inside the cofferdam.

[0039] When determining the composition and sequence of deviations, based on the monitoring deviation results, the data type of safety monitoring data that deviated within the abnormal monitoring area is identified, and the time when each type of safety monitoring data first reached the deviation threshold is read. The composition of the deviation indicates the data type of safety monitoring data that deviated, which may include one, two, or three of the following: water level difference, pile top horizontal displacement, and internal support deflection. The sequence of changes indicates the order in which the different types of safety monitoring data deviated. If the time when one type of safety monitoring data first reached the deviation threshold is at least one monitoring cycle earlier than another type, the former is determined to have deviated first. If the time difference between the first reaching the deviation threshold of two types of safety monitoring data is less than one monitoring cycle, the time when they first reached twice the deviation threshold is compared, and the data that reached twice the deviation threshold first is determined to have deviated first. When only one type of safety monitoring data deviated, the sequence of changes indicates that only that type of safety monitoring data continued to deviate.

[0040] Risk characteristics are formed based on the abnormal monitoring area, the composition of the deviation, and the sequence of changes. The risk characteristics record the monitoring locations covered by the abnormal monitoring area, the type of safety monitoring data that deviated, the sequence of changes in each type of safety monitoring data, the deviation trend of each type of safety monitoring data, and whether the horizontal displacement at the pile top extends to adjacent monitoring locations. The deviation trend indicates the direction of change of the safety monitoring data relative to the monitoring response benchmark within a continuous monitoring period. When the water level difference continuously increases, the deviation trend of the water level difference is increasing; when the water level difference continuously decreases, the deviation trend of the water level difference is decreasing. The deviation trends of the horizontal displacement at the pile top and the deflection of the internal support are determined according to the pre-set positive directions of displacement and deflection.

[0041] The direction of the stress response to changes in water level difference is determined based on reference monitoring records that match the current construction stage, event type, and internal support configuration. Specifically, the subsequent changes in the horizontal displacement of the pile top and the deflection of the internal support after a change in water level difference are extracted from the reference monitoring records. The direction of change that repeats in most reference monitoring records is determined as the stress response direction corresponding to the change in water level difference. When the reference monitoring records indicate that the response direction differs under different cofferdam sections or different stress state adjustment events, the corresponding stress response directions are established for each construction stage, event type, internal support configuration, and cofferdam section.

[0042] The preset risk assessment conditions are established based on the deviations in water level difference, pile top horizontal displacement, and internal support deflection, including their composition, sequence of changes, deviation trends, and spatial expansion. In this embodiment, risk types include water level difference change risk, internal support anomaly risk, local deformation risk of the retaining structure, and mismatched continuous anomaly risk.

[0043] When the water level difference deviates first, followed by deviations in the pile top horizontal displacement and internal support deflection, and the deviation trends of both the pile top horizontal displacement and internal support deflection are consistent with the direction of the force response corresponding to the water level difference change, the risk type is determined to be water level difference change risk. The risk location of water level difference change risk is determined based on the abnormal monitoring area that generates the structural response. If the water level difference deviates first, but only the pile top horizontal displacement or only the internal support deflection subsequently deviates, or if both types of structural monitoring data deviate but at least one type of deviation trend is inconsistent with the direction of the force response, then it is uncertain whether it is water level difference change risk.

[0044] When the deflection of the internal support deviates before the horizontal displacement of the pile top associated with the internal support space, and the water level difference does not deviate first, the risk type is determined to be an internal support anomaly risk. "Water level difference not deviating first" includes the water level difference being within the monitoring response benchmark range, or the water level difference deviating but first reaching the deviation threshold later than the internal support deflection. The risk location for internal support anomaly risk is the monitoring location of the internal support where the deviation occurs first; when an internal support has multiple anomaly monitoring locations, the monitoring location with the largest deviation and its associated internal support can be jointly determined as the risk location.

[0045] When determining the risk of local deformation of the retaining structure, the starting point of the judgment period is the time when the horizontal displacement of the pile top first reaches the deviation threshold, and the ending point is the time when the current risk characteristic is formed. Within this judgment period, valid monitoring data of water level difference and internal support deflection are collected. When the horizontal displacement of the pile top deviates first and extends along the cofferdam sidewall to adjacent monitoring positions, the water level difference remains within the monitoring response benchmark limit range during the judgment period, and the internal support deflection remains within the monitoring response benchmark limit range during the judgment period, the risk type is determined to be local deformation risk of the retaining structure. The extension of the horizontal displacement of the pile top to adjacent monitoring positions means that the horizontal displacement monitoring deviation at at least one adjacent pile top monitoring position reaches the corresponding deviation threshold and continuously reaches the confirmation cycle. The risk locations of the local deformation risk of the retaining structure are the abnormal pile top monitoring positions continuously distributed along the cofferdam sidewall.

[0046] When the risk characteristics do not meet any of the complete judgment conditions for water level difference change risk, internal support anomaly risk, and local deformation risk of retaining structure, the risk type is determined as unmatched continuous anomaly risk. Unmatched continuous anomaly risk includes situations where only one type of safety monitoring data forms a continuous anomaly, as well as situations where two or three types of safety monitoring data deviate, but the order of change, deviation trend, or spatial expansion does not meet the judgment conditions of the first three types of risks. The risk location of unmatched continuous anomaly risk is determined based on the safety monitoring data that deviated within the anomaly monitoring area; when the deviation includes pile top horizontal displacement or internal support deflection, the corresponding anomaly monitoring location is determined as the risk location; when both pile top horizontal displacement and internal support deflection deviate, the anomaly monitoring locations of both are jointly determined as the risk location; when the deviation only includes water level difference, the cofferdam area associated with that water level difference is determined as the risk location.

[0047] When matching risk types, it is determined whether the risk characteristics fully meet the criteria for risk of water level difference change, abnormal internal support, or local deformation of the retaining structure. If one of the criteria is fully met, it is identified as the corresponding risk type; if none are met, it is identified as an unmatched, persistent abnormal risk. If the timing of changes cannot be distinguished by the collection time of monitoring data, subsequent monitoring data can be collected, and the timing of changes can be determined based on the time when each type of safety monitoring data first reaches twice the deviation threshold. If the complete criteria required for the first three types of risks cannot be formed within the set judgment time limit, it is identified as an unmatched, persistent abnormal risk.

[0048] When determining the warning level, the duration, distribution range of abnormal monitoring locations, and degree of deviation are classified separately. The duration of a continuous abnormal state is calculated from the time when the first abnormal monitoring location in the abnormal monitoring area first reaches the deviation threshold until the current monitoring time. The duration classification thresholds are obtained based on the normal recovery time in historical qualified construction records and the time required for on-site safety handling; the normal recovery times can be sorted from smallest to largest, and the normal recovery time at the 75th percentile after sorting can be used as the first duration classification threshold, the normal recovery time at the 95th percentile after sorting can be used as the second duration classification threshold, and the latest handling time in the construction safety management requirements can be used as the third duration classification threshold.

[0049] The distribution range level is determined based on the number of spatially continuous anomaly monitoring locations. The distribution range grading threshold is obtained based on the density of monitoring location deployment to distinguish between single-point anomalies, localized expansions, and continuous segment anomalies. For example, when the distance between adjacent monitoring locations is basically the same, one anomaly monitoring location is classified as Level 1; two to three consecutive anomaly monitoring locations are classified as Level 2; and four or more consecutive anomaly monitoring locations are classified as Level 3. In cases where only the water level difference deviates, the water level monitoring location unit that forms that difference is counted as one anomaly monitoring location.

[0050] The deviation level is determined based on the maximum monitored deviation. Since the dimensions of water level difference, pile top horizontal displacement, and internal support deflection are different, corresponding grading thresholds are set for each type of safety monitoring data. In this embodiment, the maximum monitored deviation refers to the maximum deviation of the same type of safety monitoring data from the monitoring response benchmark boundary within a continuous abnormal state; it does not mean directly comparing the values ​​of water level difference, pile top horizontal displacement, and internal support deflection. For each type of safety monitoring data, the difference between the monitoring response benchmark boundary and the design allowable control value for that type of data is determined as the remaining allowable amount. The maximum monitored deviation of that type of data is compared with the remaining allowable amount to determine the deviation level. When the maximum monitored deviation does not exceed 30% of the remaining allowable amount, it is determined as Level 1 deviation; when the maximum monitored deviation exceeds 30% but does not exceed 60% of the remaining allowable amount, it is determined as Level 2 deviation; when the maximum monitored deviation exceeds 60% of the remaining allowable amount or reaches the design allowable control value, it is determined as Level 3 deviation. When multiple types of safety monitoring data deviations occur, the deviation level for each type of data is determined separately, and the highest level is taken as the deviation level. The duration level, distribution range level, and deviation level are then compared, and the highest level is determined as the warning level. A risk warning result is generated based on the risk type, risk location, and warning level, and then output through the warning output terminal.

[0051] In one specific embodiment, the steel cofferdam is in the continued pumping stage after the first layer of internal supports has been installed, and the internal support setting status is "first layer of internal supports installed". The construction management terminal records that the pumping operation started at 09:00, and the monitoring processor identifies this operation as a force state adjustment event of the pumping type, with the event starting at 09:00. Safety monitoring data is collected according to a 10-minute monitoring cycle. Continuous monitoring data from 08:30 to 09:00 is selected to determine the current monitoring benchmark value. During this period, the water level difference benchmark value is 0.62m, the horizontal displacement benchmark value at pile top monitoring position P3 is 7.8mm, and the deflection benchmark value at internal support monitoring position Z2 is 2.6mm.

[0052] Based on matching historical qualified construction records, the monitoring response benchmarks for the period 30 to 60 minutes after the start of pumping were determined as follows: water level difference 0.92m to 1.07m, horizontal displacement of the pile top at position P3 8.6mm to 9.4mm, and internal support deflection at position Z2 3.1mm to 3.7mm; the monitoring response benchmarks for the period 60 to 90 minutes after the start of pumping were determined as follows: water level difference 0.96m to 1.12m, horizontal displacement of the pile top at position P3 9.0mm to 9.6mm, and internal support deflection at position Z2 3.3mm to 3.9mm. The water level difference deviation threshold was set at 0.05m, the pile top horizontal displacement deviation threshold at 0.7mm, and the internal support deflection deviation threshold at 0.5mm. The confirmation period was three consecutive monitoring cycles, and the anomaly time limit was 60 minutes.

[0053] At 09:30, 09:40, and 09:50, the deflection of the inner support at position Z2 was 4.3 mm, 4.5 mm, and 4.7 mm, respectively. The deviations from the upper limit of the monitoring response benchmark of 3.7 mm within 30 to 60 minutes were 0.6 mm, 0.8 mm, and 1.0 mm, respectively. All of these deviations reached the inner support deflection threshold and continued for three consecutive monitoring cycles. Therefore, Z2 was identified as an abnormal monitoring position. The first time the deviation threshold was reached was recorded at 09:30.

[0054] At 09:50, the horizontal displacement of the pile top at location P3 was 10.4 mm, a deviation of 1.0 mm from the upper limit of the monitoring response benchmark of 9.4 mm within the 30-60 minute monitoring period. At 10:00 and 10:10, the horizontal displacements of the pile top at location P3 were 10.7 mm and 10.9 mm, respectively, a deviation of 1.1 mm and 1.3 mm from the upper limit of the monitoring response benchmark of 9.6 mm within the 60-90 minute monitoring period. All three monitoring cycles reached the pile top horizontal displacement deviation threshold; therefore, P3 was identified as an abnormal monitoring location. The first time the deviation threshold was reached was recorded as 09:50. Throughout the above process, the water level difference remained within the limits of the corresponding monitoring response benchmark. Since Z2 and P3 have a spatial relationship, and the internal support deflection deviated before the horizontal displacement of the adjacent pile top, while the water level difference did not deviate first, an internal support anomaly risk was identified, and the first layer of internal support where Z2 is located was identified as the risk location.

[0055] If the continuous abnormal state reaches 60 minutes, the duration level is classified as Level 2. Z2 and P3 constitute two spatially correlated abnormal monitoring locations, with a distribution range level of Level 2. The deviation level is determined according to the data type: the maximum monitored deviation of the internal support deflection at location Z2 is 1.0 mm, and the deviation level is determined according to the corresponding grading threshold for internal support deflection; the maximum monitored deviation of the pile top horizontal displacement at location P3 is 1.3 mm, and the deviation level is determined according to the corresponding grading threshold for pile top horizontal displacement. The highest deviation level among the various types of safety monitoring data that show deviation is taken as the deviation level of this risk; then, the highest level among the duration level, distribution range level, and deviation level is determined as the warning level.

[0056] In another specific embodiment, the horizontal displacement of the pile top at monitoring location P5 first deviates and then extends to the adjacent pile top monitoring location P6. The deviations in the horizontal displacement monitoring at both P5 and P6 reach their respective deviation thresholds and are continuously confirmed over the period. From the time P5 first reaches the deviation threshold until the determination of the risk characteristic, the water level difference remains within the monitoring response benchmark range, and the effective monitoring values ​​of the internal support deflection also remain within the monitoring response benchmark range. Therefore, the risk type is determined to be local deformation risk of the retaining structure, and P5 and P6 are identified as risk locations.

[0057] In another specific embodiment, the water level difference deviates first, followed by a deviation in the horizontal displacement of the pile top at monitoring location P7. However, the deflection of the internal support remains within the monitoring response benchmark range within the set judgment time limit. Since the internal support deflection does not subsequently deviate, this risk characteristic does not meet the judgment criteria for water level difference change risk; nor does it meet the judgment criteria for local deformation risk of the retaining structure when the horizontal displacement of the pile top at P7 does not extend to adjacent monitoring locations. Therefore, the risk type is determined to be unmatched continuous abnormal risk, and P7 and the cofferdam area associated with the water level difference are identified as risk locations. Subsequently, the warning level is determined based on the duration of the continuous abnormal state, the number of abnormal monitoring locations, and the maximum monitoring deviation corresponding to each type of safety monitoring data, and a risk warning result is generated.

[0058] In summary, by combining construction status data with safety monitoring data, the current construction stage is first determined, and stress state adjustment events that may cause changes in the stress state of the steel cofferdam, such as pumping, installation or removal of internal supports, are identified. Then, safety monitoring data before the event and matching reference monitoring records are used to form a monitoring response benchmark, so that subsequent judgments no longer rely solely on fixed alarm values, but use the normal monitoring response under the current construction operation as the comparison object. By comparing the safety monitoring data after the event with the monitoring response benchmark, abnormal monitoring locations are screened, and the monitoring deviation results within the continuous monitoring period are combined to distinguish between temporary fluctuations and continuous abnormal states, which can reduce misjudgments caused by short-term disturbances in construction operations. After determining the continuous abnormal state, risk characteristics are formed based on the spatial correlation between abnormal monitoring locations and the chronological relationship of changes in different safety monitoring data, and the risk type, risk location, and warning level are determined, so that the warning results can point to specific risk sources and key inspection areas, facilitating the implementation of corresponding on-site measures.

[0059] Based on another preferred embodiment described above, see [link to preferred embodiment]. Figure 3 As shown, this embodiment provides a risk early warning system for steel cofferdam construction safety monitoring data, used to apply the aforementioned risk early warning method for steel cofferdam construction safety monitoring data, including: The status identification unit is used to acquire construction status data and safety monitoring data of the steel cofferdam, determine the current construction stage based on the construction status data, and identify stress state adjustment events that cause changes in the stress state of the steel cofferdam. The benchmark determination unit is used to obtain the monitoring response benchmark based on the safety monitoring data before the stress state adjustment event and the reference monitoring records that match the current construction stage and stress state adjustment event. The anomaly determination unit is used to compare the safety monitoring data after the occurrence of the stress state adjustment event with the monitoring response benchmark to obtain the monitoring deviation results of each monitoring location; to screen the abnormal monitoring locations based on the monitoring deviation results, and to determine the temporary fluctuation state or the continuous abnormal state based on the monitoring deviation results of the abnormal monitoring locations during the continuous monitoring period. The feature formation unit is used to form risk features based on the spatial distribution of abnormal monitoring locations and the deviation composition and sequence of changes in safety monitoring data when a continuous abnormal state is determined. The risk determination unit is used to match risk characteristics with preset risk judgment conditions to determine the risk type and risk location. The early warning generation unit is used to determine the early warning level based on the duration of the continuous abnormal state, the distribution range of the abnormal monitoring locations, and the degree of deviation, and to generate a risk early warning result.

[0060] It is understandable that the risk warning methods and systems for the aforementioned steel cofferdam construction safety monitoring data have the same beneficial effects, and will not be elaborated further here.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A risk early warning method for safety monitoring data during steel cofferdam construction, characterized in that, include: Acquire construction status data and safety monitoring data of the steel cofferdam, determine the current construction stage based on the construction status data, and identify stress state adjustment events that cause changes in the stress state of the steel cofferdam. Based on the safety monitoring data prior to the stress state adjustment event, and the reference monitoring records that match the current construction stage and the stress state adjustment event, a monitoring response benchmark is obtained. The safety monitoring data after the stress state adjustment event occurs is compared with the monitoring response benchmark to obtain the monitoring deviation results at each monitoring location; Based on the monitoring deviation results, abnormal monitoring locations are screened, and based on the monitoring deviation results of the abnormal monitoring locations within a continuous monitoring period, a temporary fluctuation state or a continuous abnormal state is determined. When the continuous abnormal state is determined, risk characteristics are formed based on the spatial distribution of the abnormal monitoring locations and the order of changes in different safety monitoring data. The risk characteristics are matched with preset risk determination conditions to determine the risk type and risk location; The warning level is determined based on the duration of the continuous abnormal state, the distribution range of the abnormal monitoring locations, and the degree of deviation, and a risk warning result is generated.

2. The risk early warning method for steel cofferdam construction safety monitoring data according to claim 1, characterized in that, Processing the security monitoring data includes: The water levels inside and outside the cofferdam, the horizontal displacement of the pile top, and the deflection of the internal support are aligned according to the collection time. The water level difference is determined based on the water levels inside and outside the cofferdam, forming a safety monitoring sequence for each monitoring location.

3. The risk early warning method for steel cofferdam construction safety monitoring data according to claim 2, characterized in that, When determining the current construction stage and identifying the stress state adjustment event, the following are included: The current construction stage and internal support setting status are determined based on the construction procedure records in the construction status data; when the construction procedure records indicate the start of pumping, installation of internal supports, or removal of internal supports, the stress state adjustment event is identified, and the event type and event start time are determined.

4. The risk early warning method for steel cofferdam construction safety monitoring data according to claim 3, characterized in that, When obtaining the monitoring response benchmark, the following are included: Reference monitoring records that match the current construction stage, event type, and internal support setting status are selected from historical qualified construction records; the safety monitoring data before the event is adjusted according to the stress state to determine the current monitoring benchmark value; the reference change range is determined according to the monitoring changes before and after the event in the reference monitoring records; the reference change range is superimposed on the current monitoring benchmark value to obtain the monitoring response benchmark.

5. The risk early warning method for steel cofferdam construction safety monitoring data according to claim 4, characterized in that, When obtaining the monitoring deviation results and filtering the abnormal monitoring locations, the process includes: The safety monitoring sequence after the start time of the event is compared with the monitoring response benchmark, and the monitoring deviation result is obtained according to whether it exceeds the monitoring response benchmark and the amount of excess; when the monitoring deviation reaches the deviation threshold and continuously reaches the confirmation cycle, the corresponding monitoring location is determined as the abnormal monitoring location.

6. The risk early warning method for steel cofferdam construction safety monitoring data according to claim 5, characterized in that, Determining the temporary fluctuation state or the persistent abnormal state includes: The monitoring deviation results of the abnormal monitoring location are reviewed during the continuous monitoring period. When the monitoring deviation decreases continuously within the review period and returns to the monitoring response benchmark limit, it is determined to be the temporary fluctuation state. When the monitoring deviation does not return to the limit and continues to reach the abnormal time limit, or when the monitoring deviation of the same type of safety monitoring data of the monitoring location spatially adjacent to the abnormal monitoring location reaches the deviation threshold and continuously reaches the confirmation period, it is determined to be the continuous abnormal state.

7. The risk early warning method for steel cofferdam construction safety monitoring data according to claim 6, characterized in that, When the risk characteristics are formed, the following are included: The abnormal monitoring locations that are in the continuous abnormal state are formed into abnormal monitoring areas according to their spatial adjacency, wherein isolated abnormal monitoring locations form separate abnormal monitoring areas. Based on the data types of safety monitoring that deviate in each abnormal monitoring area and the time when they first reach the deviation threshold, determine the composition and sequence of the deviations. The risk characteristics are formed based on the abnormal monitoring area, the composition of the deviation, and the sequence of changes.

8. The risk early warning method for steel cofferdam construction safety monitoring data according to claim 7, characterized in that, Determining the risk type and the risk location includes: When the water level difference deviates first, and the horizontal displacement of the pile top and the deflection of the internal support subsequently deviate, and the deviation trend is consistent with the direction of the force response to the change in water level difference, it is identified as a risk of water level difference change. When the deflection of the internal support deviates before the horizontal displacement of the adjacent pile top, and the water level difference does not deviate first, it is determined to be an abnormal risk of the internal support. When the horizontal displacement at the top of the pile deviates first and extends to the adjacent monitoring location, the water level difference is within the range defined by the monitoring response benchmark, and the deflection of the internal support is always within the range defined by the monitoring response benchmark, it is determined to be a risk of local deformation of the retaining structure. When the risk characteristics do not meet the above risk determination conditions, it is determined to be an unmatched persistent abnormal risk; The location of the risk is determined based on the risk type and the safety monitoring data showing deviations within the abnormal monitoring area.

9. The risk early warning method for steel cofferdam construction safety monitoring data according to claim 8, characterized in that, Determining the warning level and generating the risk warning result includes: The duration of the continuous abnormal state, the number of spatially continuous abnormal monitoring locations, and the maximum monitoring deviation are compared with the corresponding classification thresholds to obtain the duration level, distribution range level, and deviation level; the highest level is determined as the warning level, and the risk warning result is formed based on the risk type, risk location, and warning level.

10. A risk early warning system for steel cofferdam construction safety monitoring data, used for applying the risk early warning method for steel cofferdam construction safety monitoring data as described in any one of claims 1-9, characterized in that, include: The status recognition unit is used to acquire construction status data and safety monitoring data of the steel cofferdam, determine the current construction stage based on the construction status data, and identify stress state adjustment events that cause changes in the stress state of the steel cofferdam. The benchmark determination unit is used to obtain a monitoring response benchmark based on the safety monitoring data prior to the occurrence of the stress state adjustment event and the reference monitoring records that match the current construction stage and the stress state adjustment event. An anomaly determination unit is used to compare the safety monitoring data after the occurrence of the force state adjustment event with the monitoring response benchmark to obtain the monitoring deviation results of each monitoring location; Based on the monitoring deviation results, abnormal monitoring locations are screened, and based on the monitoring deviation results of the abnormal monitoring locations within a continuous monitoring period, a temporary fluctuation state or a continuous abnormal state is determined. The feature forming unit is used to form risk features based on the spatial distribution of the abnormal monitoring locations and the order of changes in different safety monitoring data when the continuous abnormal state is determined. The risk determination unit is used to match the risk characteristics with preset risk judgment conditions to determine the risk type and risk location; The early warning generation unit is used to determine the early warning level based on the duration of the continuous abnormal state, the distribution range of the abnormal monitoring locations, and the degree of deviation, and to generate a risk early warning result.

Images