A method for intelligent identification and early warning of safety risks at construction sites

Abstract

This invention discloses an intelligent identification and early warning method for safety risks at construction sites, relating to the field of construction safety monitoring. By acquiring monitoring data from the construction site, the method identifies the rainfall process and determines the peak rainfall time, then defines a post-rain analysis period after the peak rainfall time. During this period, the method tracks the changes in monitoring data such as groundwater level, structural deformation, and support stress, extracting the response time sequence and trend of each monitored quantity relative to the rainfall process. Combining the response sequence relationship and the degree of coordination between the various monitoring data, a coupled feature reflecting changes in the construction status is constructed, and based on this, abnormal conditions at the construction site are identified and risk warning information is output. This invention, by introducing response time sequence analysis and trend judgment between multi-source monitoring data, can more accurately reflect the changing patterns of construction structures under rainfall disturbances, effectively improving the accuracy of risk identification and the reliability of early warning compared to traditional single-threshold judgment methods.

Description

Technical Field

[0001] This invention relates to the field of construction safety monitoring, and in particular to an intelligent identification and early warning method for safety risks at construction sites. Background Technology

[0002] With the continuous expansion of urban construction, deep foundation pit engineering is increasingly widely used in various building construction projects. During the construction of deep foundation pits, factors such as rainfall, changes in groundwater, and disturbance of soil structure can easily lead to safety problems such as deformation of the support structure, abnormal groundwater seepage, and settlement of the surrounding surface. In severe cases, this may result in foundation pit instability or even safety accidents. Therefore, real-time monitoring of the construction site and timely identification of potential risks are of great significance for ensuring construction safety.

[0003] In existing technologies, monitoring and early warning for the safety of deep foundation pit construction often rely on setting thresholds for single monitoring quantities (such as groundwater level, support structure displacement, or support axial force) to make judgments. When the monitored value exceeds the preset threshold, an alarm is issued. However, this type of method is mainly based on a single point or a single physical quantity, which makes it difficult to reflect the coupled changes of multiple physical quantities caused by rainfall. Especially in the post-rain stage, there are obvious response lags and transmission processes between different monitoring quantities, and relying solely on a single monitoring quantity is prone to misjudgment or omission.

[0004] In addition, although some existing solutions introduce multi-source data for analysis, they mostly remain at the level of simple data superposition or independent judgment, lacking systematic analysis of the temporal correlation between rainfall, water level changes, structural deformation and support stress, making it difficult to accurately identify abnormal states under complex working conditions, resulting in insufficient reliability and foresight of the early warning results.

[0005] Therefore, a method for intelligent identification and early warning of safety risks at construction sites is proposed. Summary of the Invention

[0006] The main objective of this invention is to provide an intelligent identification and early warning method for safety risks at construction sites, which can effectively solve the problems in the background art.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A method for intelligent identification and early warning of safety risks at construction sites includes the following steps:

[0009] S1. Obtain monitoring data at the construction site;

[0010] S2. Identify the rainfall process based on the monitoring data, determine the peak time of the rainfall process, and define the post-rain analysis period after the peak time;

[0011] S3. During the post-rain analysis period, track the changes in various monitoring data and extract their response time sequence and trend relative to the rainfall process.

[0012] S4. By combining the response sequence and the degree of coordination among the various monitoring data, a coupling characteristic reflecting the changes in the construction status is formed;

[0013] S5. Identify abnormal conditions at the construction site based on the coupling characteristics and output corresponding risk warning information.

[0014] Furthermore, the monitoring data comes from monitoring points deployed within the deep foundation pit construction area, including water level monitoring points on the outside of the retaining structure, water level monitoring points in the dewatering wells inside the pit, deformation monitoring points of the support structure, and stress monitoring points of the supporting components. The monitoring points are deployed in segments along the circumference of the foundation pit, and are densely deployed in the support corner area, support node area, and on the side adjacent to existing buildings and structures. The sampling period for different types of monitoring points is set according to their respective physical response characteristics, wherein the sampling period of the water level monitoring point is no greater than the sampling period of the displacement monitoring point.

[0015] Furthermore, before analyzing the monitoring data, the monitoring data from different sources are processed to unify the time base, so that all types of monitoring data are mapped to a unified time series; for abnormal jump data that occur during the monitoring process, corrections or removals are made in combination with the changing trends of adjacent time periods and the historical change range of the monitoring point; and the trend of continuous sampling data is confirmed by using a sliding time window to distinguish between short-term construction disturbances and continuous structural responses.

[0016] Furthermore, when identifying the rainfall process, the rainfall change is divided into an initial stage, an intensification stage, and a decay stage, with the moment when the rainfall reaches its maximum value being taken as the peak moment. After the peak moment, the post-rain analysis period is adjusted according to the rainfall duration, rainfall intensity, and foundation pit drainage capacity to ensure that the post-rain analysis period covers the main processes of groundwater level changes and structural responses. The post-rain analysis period continues until the rate of change of each monitoring data is lower than a preset threshold for multiple consecutive sampling periods, or until the trend of change between each monitoring data tends to stabilize.

[0017] Furthermore, the response time series is characterized by the response time difference of each monitored quantity relative to the rainfall peak:

[0018]

[0019] in, This is the peak time of rainfall. The continuous change corresponds to the moment when the monitored quantity begins to change continuously, and the continuous change means that the change direction is consistent within multiple consecutive sampling periods; and the monitored quantities are sorted according to the response time difference to reflect the transmission relationship between rainfall, water level change and structural response.

[0020] Furthermore, the changing trend includes the direction, rate, and duration of the changes in the monitored quantities during the post-rain analysis period; when the deformation of the support structure maintains a unidirectional increase in multiple consecutive sampling periods, and the corresponding groundwater level change does not show a changing trend consistent with the direction of the structural deformation change, or the corresponding water level change amplitude is significantly lower than the structural deformation change amplitude, the changing process is marked as an abnormal trend; and the current changing trend is compared with the changing process under similar historical rainfall conditions to further confirm the degree of abnormality.

[0021] Furthermore, the coupling characteristics are characterized by the response relationships between different monitored quantities, including the response relationship between water level changes and structural deformation, and the response relationship between structural deformation and support stress; and quantified by the proportional relationship of the response time difference.

[0022]

[0023] in, The structural deformation response delay time. Water level response delay time A non-zero correction parameter is used to avoid a denominator of zero and is determined based on historical monitoring data or foundation pit drainage conditions; when the ratio deviates from the normal range, it is considered that there is an abnormal change in the construction status.

[0024] Furthermore, when identifying abnormal states, the response order, trend of change, and mutual corroboration relationship between multiple monitoring quantities are comprehensively considered. When at least two types of monitoring quantities in the same monitoring zone exhibit abnormal characteristics within the same time period, and the abnormal characteristics persist for more than a preset time length, it is determined to be a risk state. The graded early warning results are output according to the duration and degree of abnormality.

[0025] A construction site safety risk intelligent identification and early warning system includes:

[0026] The system includes a data acquisition module, a time series analysis module, a feature extraction module, a coupling analysis module, and an early warning output module.

[0027] The data acquisition module is used to acquire monitoring data at the construction site.

[0028] The time series analysis module is used to identify the rainfall process based on the monitoring data, determine the peak time of the rainfall process, and define the post-rain analysis period after the peak time;

[0029] The feature extraction module is used to track the changes in various monitoring data during the post-rain analysis period and extract their response time sequence and trend relative to the rainfall process.

[0030] The coupling analysis module is used to combine the response sequence and the degree of coordination of changes among various monitoring data to form coupling characteristics that reflect changes in construction status;

[0031] The early warning output module is used to identify abnormal conditions at the construction site based on the coupling characteristics and output corresponding risk warning information.

[0032] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method steps.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] This method incorporates a post-rainfall analysis period into the rainfall process, jointly analyzing the response time series and trends of multiple monitoring data. It constructs a coupled feature reflecting the relationship between rainfall, water level changes, and structural response, thereby identifying the sequential relationship of responses and the degree of coordination between different monitoring quantities. Compared to traditional methods relying solely on a single monitoring threshold, this invention more comprehensively reflects the actual stress and deformation state of deep foundation pits under rainfall disturbance, improving the accuracy of identifying seepage anomalies and structural deformation problems.

[0035] Furthermore, by performing time-unified processing of monitoring data, confirming continuous change trends, and cross-verification analysis among multiple monitoring quantities, this invention enables risk identification to be based on continuous change processes rather than instantaneous data changes. At the same time, it quantifies abnormal states by combining the response time difference ratio, thereby effectively reducing false alarms caused by short-term disturbances or accidental fluctuations, improving the stability and reliability of early warning results, and having better engineering applicability. Attached Figure Description

[0036] Figure 1 This is a flowchart of the method of the present invention;

[0037] Figure 2 This is a system module diagram of the present invention;

[0038] Figure 3 This is a diagram illustrating the coupling mechanism of the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0040] Example 1

[0041] This embodiment provides an intelligent identification and early warning method for safety risks at construction sites, applicable to deep foundation pit construction scenarios, and especially suitable for engineering environments that are greatly affected by rainfall. By jointly analyzing the rainfall process and the changes in multi-source monitoring data it causes, the method can identify and warn of construction safety risks.

[0042] In this embodiment, a monitoring system is first deployed at the construction site. This system includes groundwater level monitoring points outside the retaining structure, water level monitoring points in the dewatering wells within the pit, deformation monitoring points for the support structure, and stress monitoring points for the supporting components. These monitoring points are deployed in sections along the circumference of the foundation pit, with increased density at support corners, support nodes, and areas adjacent to existing buildings to improve monitoring accuracy at critical locations. Sampling cycles for different types of monitoring points are set according to their physical response characteristics. For example, groundwater level monitoring points can be set to sample every 5-10 minutes, support structure displacement monitoring points every 10-30 minutes, and support axial force monitoring points every 5-15 minutes depending on stress changes.

[0043] During data acquisition, each monitoring point continuously outputs monitoring data, which is then uniformly transmitted to the data processing unit. Due to differences in sampling frequencies among different monitoring devices, time-base unification processing of monitoring data from different sources is necessary before subsequent analysis. This involves resampling or interpolating the data at uniform time intervals to map all monitoring data onto the same time series. Simultaneously, for any abrupt data jumps that may occur during monitoring, such as sudden changes caused by equipment interference or short-term construction operations, corrections or removals are made by combining the trends of adjacent time periods with the historical range of changes at that monitoring point. Furthermore, a sliding time window is used to confirm the trend of continuously sampled data, distinguishing between short-term construction disturbances and continuous structural responses.

[0044] In the rainfall identification phase, the rainfall process is divided into an initiation phase, an intensification phase, and a decay phase by analyzing the rainfall time series. When consecutive rainfall records occur and the cumulative rainfall reaches a preset level, it is determined to be a valid rainfall event, and the moment when the rainfall reaches its maximum value is taken as the rainfall peak moment. Following the peak rainfall, a post-rainfall analysis period is defined based on the duration and intensity of rainfall and the drainage capacity of the foundation pit. This period covers the main processes of groundwater level changes and structural responses. The post-rainfall analysis period can continue until the rate of change of each monitoring data is below a preset threshold for several consecutive sampling periods, or until the changing trends of each monitoring quantity tend to stabilize.

[0045] During the post-rainfall analysis period, response time series were extracted from various monitoring data. Specifically, the response time difference of each monitoring quantity relative to the rainfall peak was calculated:

[0046]

[0047] in, To determine the moment when a monitored quantity begins to change continuously, this moment is determined as follows: when the monitored quantity changes in the same direction over multiple consecutive sampling periods, and the magnitude of the change exceeds a certain proportion of its initial stable value, the monitored quantity is considered to have entered a state of continuous change, and this moment is taken as the response start time. By comparing the response time differences of different monitored quantities, the transmission process between rainfall, water level changes, and structural responses can be reflected. For example, under normal circumstances, rainfall should first cause changes in the groundwater level, and then manifest as deformation of the support structure and changes in the stress on the supports.

[0048] Based on this, the changing trends of each monitored quantity are further analyzed. These trends include the direction, rate, and duration of change. When the displacement of the support structure maintains a unidirectional increase over multiple consecutive sampling periods, while the corresponding groundwater level does not show a change direction consistent with the structural deformation, or the water level change is significantly lower than the structural deformation change, an abnormal trend is identified in the structural deformation process. Simultaneously, monitoring data from historical rainfall conditions can be retrieved for comparative analysis of the current trend. For example, if, under the same rainfall intensity, historical data shows relatively small changes in structural deformation, while the current deformation is significantly aggravated, the degree of anomaly is further confirmed.

[0049] In the coupling feature construction phase, coupling features between multi-source data are formed by analyzing the response relationships between different monitoring quantities. Specifically, this includes the response relationship between groundwater level changes and support structure deformation, and the response relationship between structural deformation and support axial force. To quantitatively describe this relationship, it can be calculated using the response time difference ratio.

[0050]

[0051] in, The structural deformation response delay time. For groundwater level response delay time, To avoid non-zero correction parameters where the denominator is zero, their values ​​can be set based on historical monitoring data or on-site drainage conditions. When the proportional relationship deviates significantly from the normal range, it indicates an abnormality in the response relationship between different monitoring quantities, which may indicate problems such as changes in the seepage path or abnormal structural stress.

[0052] During the risk identification phase, the response sequence, trends, and corroborating relationships among various monitoring parameters are comprehensively considered. For example, if groundwater level changes are delayed or absent, but structural deformation occurs prematurely, or if the support axial force fluctuates before structural deformation, an abnormal coupling state can be identified. Simultaneously, within the same monitoring zone, if at least two types of monitoring parameters exhibit abnormal characteristics within the same time period, and these abnormal characteristics persist for more than a preset time duration, the area is deemed to pose a safety risk.

[0053] During the early warning output phase, risks are classified according to the duration and severity of the anomaly, such as general warning, higher warning, and severe warning, and corresponding risk alert information is output. Warning information may include the location of the risk, the type of risk, and recommended remedial measures, such as strengthening drainage, increasing support, suspending construction, or conducting on-site verification.

[0054] Through the above-described embodiments, the present invention can more accurately and timely identify safety risks at construction sites based on the response time relationship and change trend between multi-source monitoring data under rainfall disturbance conditions, thereby improving the reliability of early warning and engineering applicability.

[0055] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent identification and early warning of safety risks at construction sites, characterized in that, Includes the following steps: S1. Obtain monitoring data at the construction site; S2. Identify the rainfall process based on the monitoring data, determine the peak time of the rainfall process, and define the post-rain analysis period after the peak time; S3. During the post-rain analysis period, track the changes in various monitoring data and extract their response time sequence and trend relative to the rainfall process. S4. By combining the response sequence and the degree of coordination among the various monitoring data, a coupling characteristic reflecting the changes in the construction status is formed; S5. Identify abnormal conditions at the construction site based on the coupling characteristics and output corresponding risk warning information.

2. The intelligent identification and early warning method for safety risks at construction sites according to claim 1, characterized in that, The monitoring data comes from monitoring points deployed within the deep foundation pit construction area, including water level monitoring points on the outside of the retaining structure, water level monitoring points in the dewatering wells inside the pit, deformation monitoring points of the support structure, and stress monitoring points of the supporting components. The monitoring points are deployed in sections along the circumference of the foundation pit, and are densely deployed in the corner areas of the support, the support node areas, and on the side adjacent to existing buildings and structures. The sampling period for different types of monitoring points is set according to their respective physical response characteristics, wherein the sampling period of the water level monitoring points is no greater than the sampling period of the displacement monitoring points.

3. The intelligent identification and early warning method for safety risks at construction sites according to claim 1, characterized in that, Before analyzing the monitoring data, the monitoring data from different sources are processed to unify the time base, so that all types of monitoring data are mapped to a unified time series. For abnormal jump data that occur during the monitoring process, corrections or removals are made in combination with the changing trends of adjacent time periods and the historical change range of the monitoring point. The trend of continuous sampling data is confirmed by using a sliding time window to distinguish between short-term construction disturbances and continuous structural responses.

4. The intelligent identification and early warning method for safety risks at construction sites according to claim 1, characterized in that, When identifying rainfall processes, rainfall changes are divided into an initial stage, an intensification stage, and a decay stage, with the moment when the rainfall reaches its maximum value being designated as the peak moment. After the peak moment, the post-rain analysis period is adjusted based on rainfall duration, rainfall intensity, and foundation pit drainage capacity to ensure that the post-rain analysis period covers the main processes of groundwater level changes and structural responses. The post-rain analysis period continues until the rate of change of each monitoring data is below a preset threshold for multiple consecutive sampling periods, or until the trend of change among the monitoring data tends to stabilize.

5. The intelligent identification and early warning method for safety risks at construction sites according to claim 1, characterized in that, The response time series is characterized by the time difference of each monitored quantity relative to the rainfall peak: in, This is the peak time of rainfall. The continuous change corresponds to the moment when the monitored quantity begins to change continuously, and the continuous change means that the change direction is consistent within multiple consecutive sampling periods; and the monitored quantities are sorted according to the response time difference to reflect the transmission relationship between rainfall, water level change and structural response.

6. The intelligent identification and early warning method for safety risks at construction sites according to claim 1, characterized in that, The changing trend includes the direction, rate and duration of the changes in the monitored quantities during the post-rain analysis period; When the deformation of the support structure continues to increase in one direction over multiple consecutive sampling periods, and the corresponding groundwater level change does not show a trend consistent with the direction of the structural deformation change, or the corresponding water level change is significantly lower than the structural deformation change, the change process is marked as an abnormal trend. The current trend is compared with the changes under similar historical rainfall conditions to further confirm the degree of anomaly.

7. The intelligent identification and early warning method for safety risks at construction sites according to claim 1, characterized in that, The coupling characteristics are characterized by the response relationship between different monitoring quantities, including the response relationship between water level change and structural deformation, and the response relationship between structural deformation and support force. And quantified by the proportional relationship of the response time difference: in, The structural deformation response delay time. Water level response delay time The non-zero correction parameter is used to avoid a denominator of zero and is determined based on historical monitoring data or foundation pit drainage conditions; when the ratio deviates from the normal range, it is considered that there is an abnormal change in the construction status.

8. The intelligent identification and early warning method for safety risks at construction sites according to claim 1, characterized in that, When identifying abnormal states, the response order, trend of change, and mutual corroboration relationship between multiple monitoring quantities are comprehensively considered. When at least two types of monitoring quantities in the same monitoring zone show abnormal characteristics in the same time period, and the abnormal characteristics last for more than the preset time length, it is determined to be a risk state. The graded early warning results are output according to the duration and degree of abnormality.

9. A construction site safety risk intelligent identification and early warning system, characterized in that, include: The system includes a data acquisition module, a time series analysis module, a feature extraction module, a coupling analysis module, and an early warning output module. The data acquisition module is used to acquire monitoring data at the construction site. The time series analysis module is used to identify the rainfall process based on the monitoring data, determine the peak time of the rainfall process, and define the post-rain analysis period after the peak time; The feature extraction module is used to track the changes in various monitoring data during the post-rain analysis period and extract their response time sequence and trend relative to the rainfall process. The coupling analysis module is used to combine the response sequence and the degree of coordination of changes among various monitoring data to form coupling characteristics that reflect changes in construction status; The early warning output module is used to identify abnormal conditions at the construction site based on the coupling characteristics and output corresponding risk warning information.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1 to 8.

Images