Building construction safety monitoring method and system

Abstract

The invention relates to the technical field of construction safety monitoring, in particular to a building construction safety monitoring method and system, and the method comprises the steps: collecting building construction data corresponding to each construction building in a building construction region in a current time period, carrying out the preprocessing of the building construction data, and obtaining the processed building construction data, extracting an inclination angle feature, a vibration feature, a displacement feature, a first environment feature and a second environment feature in the processed building construction data, and inputting the inclination angle feature, the vibration feature, the displacement feature, the first environment feature and the second environment feature into a trained preset construction early warning model for recognition to obtain a building recognition result, and judging whether feature alarm information exists in the building identification result, if so, generating a building alarm instruction based on the feature alarm information, and controlling the alarm equipment to output a response. According to the invention, the monitoring efficiency of building construction safety is improved.

Description

Technical Field

[0001] This application relates to the technical field of construction safety monitoring, and in particular to a method and system for monitoring construction safety. Background Technology

[0002] In the construction industry, ensuring construction safety is a core element for the smooth progress of projects. The construction process involves numerous complex stages, from foundation work and main structure erection to decoration and finishing, each accompanied by various potential safety risks. For example, the stability of a building structure may be affected by changes in geological conditions, increased construction loads, etc., leading to structural tilting, abnormal vibration, or displacement, and in severe cases, even collapse, causing significant casualties and property damage. Simultaneously, the construction environment is also complex. Environmental factors both inside and outside the building, such as temperature, humidity, wind speed, and dust concentration, adversely affect the performance of building materials and construction equipment, thus indirectly impacting construction safety.

[0003] Currently, construction safety monitoring mainly relies on manual inspections and some simple monitoring equipment. While manual inspections allow direct observation of the construction site, they suffer from drawbacks such as long inspection cycles, low efficiency, and susceptibility to human interference, making it difficult to grasp the dynamic information of construction in real time and comprehensively. Furthermore, most existing monitoring equipment is single-function, monitoring only a specific parameter, such as tilt angle or vibration, failing to comprehensively consider the interaction of multiple factors. This results in inaccurate and incomplete monitoring results, making it difficult to meet the needs of modern construction safety monitoring. Summary of the Invention

[0004] To address at least one of the aforementioned technical problems, this application provides a method and system for monitoring construction safety.

[0005] Firstly, this application provides a method for monitoring construction safety, employing the following technical solution: Collect construction data for each building under construction in the current time period. The construction data includes tilt angle monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data. The environmental monitoring data includes external environmental data and internal environmental data. The construction data are preprocessed to obtain the processed construction data. The tilt angle feature, vibration feature, displacement feature, first environmental feature and second environmental feature are extracted from the processed building construction data, and the tilt angle feature, vibration feature, displacement feature, first environmental feature and second environmental feature are input into the trained preset construction early warning model for identification to obtain building identification results; Determine whether there is feature alarm information in the building identification result. If so, generate a building alarm command based on the feature alarm information and control the alarm device to output a response.

[0006] By employing the aforementioned technical solution, construction data for each building under construction within the current timeframe is collected, encompassing multiple dimensions including tilt angle, vibration, displacement, and environmental factors (both external and internal). This comprehensively and meticulously reflects the dynamic changes of the building itself and its surrounding environment during construction. The construction data is then preprocessed, removing noise and filling in missing values, resulting in processed construction data. This ensures the accuracy and completeness of the data, providing a reliable foundation for subsequent feature extraction and model recognition. This allows the building recognition results based on this data to more accurately reflect the actual state of construction, effectively avoiding recognition biases caused by data issues. After obtaining the processed construction data, tilt angle features, vibration features, displacement features, first environmental features, and second environmental features are extracted. Features are key information carriers of the construction status; for example, tilt angle features directly reflect the degree of tilt of the building structure, and vibration features reflect the external forces acting on the building. These features are input into a pre-trained construction early warning model for recognition. A comprehensive analysis of the tilt angle features, vibration features, displacement features, first environmental features, and second environmental features yields accurate building recognition results. This significantly improves the accuracy and comprehensiveness of assessing the construction status. After obtaining the building identification results, it checks for the presence of characteristic alarm information. If characteristic alarm information is found, a building alarm command is generated based on this information, controlling the alarm equipment to output a response. This allows for the rapid issuance of alarms when abnormalities occur during construction, alerting relevant personnel to take measures to prevent further escalation of accidents, greatly improving construction safety and protecting the lives of construction workers and the quality of the building.

[0007] In one possible implementation, determining whether feature alarm information exists in the building recognition result includes: If not, determine whether there is feature warning information in the building identification result. If there is, determine the warning feature type, warning sample data and warning feature value corresponding to the warning feature type based on the feature warning information, and retrieve feature monitoring data within a preset time period according to the warning feature type. The building warning anomaly and the warning anomaly time are determined based on the feature monitoring data, the warning sample data, and the warning feature value. Based on the building warning anomaly and the time of the warning anomaly, a building warning command is generated, and the warning equipment is controlled to output a response.

[0008] In one possible implementation, the step of inputting the tilt angle feature, the vibration feature, the displacement feature, the first environmental feature, and the second environmental feature into a trained preset construction early warning model for identification to obtain the building identification result further includes: Collect building monitoring data within a historical time period. The building monitoring data includes tilt angle monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data inside and outside the building for each construction building in different construction areas. Using the construction area, the building under construction, and the environmental monitoring data as matching fields, the building monitoring data is matched with the building construction data within the current time period to obtain a set of sample data in the building monitoring data that corresponds to the building construction data. Identify the abnormal time points in the sample data set where monitoring data anomalies exist, as well as the target experimental data corresponding to the monitoring data anomalies, and collect evolution experimental data corresponding to the target experimental data according to the abnormal time points and a preset time period; The target experimental data and the evolution experimental data are spliced ​​together according to time nodes to obtain periodic sample data; The abnormal monitoring data and the periodic sample data are input into the preset construction early warning feature model according to the corresponding relationship to obtain the trained preset construction early warning feature model.

[0009] In one possible implementation, determining the building warning anomaly and the warning anomaly time based on the feature monitoring data, the warning sample data, and the warning feature value includes: Construct a feature monitoring coordinate system, where the X-axis represents different time nodes and the Y-axis represents different unit feature values ​​corresponding to each type of feature monitoring data. The feature monitoring data and the early warning feature value are respectively imported into the feature monitoring coordinate system, and the monitoring data points corresponding to the feature monitoring data and the early warning data points corresponding to the early warning feature value are connected in time order to obtain the feature monitoring curve; The early warning sample data is imported into the feature monitoring coordinate system, and the monitoring data points corresponding to the early warning sample data are connected in chronological order to obtain the sample monitoring curve. The characteristic monitoring curve and the sample monitoring curve are analyzed to determine the building early warning anomaly and the time of the early warning anomaly.

[0010] In one possible implementation, the analysis of the feature monitoring curve and the sample monitoring curve to determine the building early warning anomaly and the time of the early warning anomaly includes: Determine whether there is a feature value inflection point in the sample type curve corresponding to each data feature type in the sample monitoring curve. If there is, determine the initial sample feature value and the termination sample feature value corresponding to the sample type curve, and determine the sample growth trend corresponding to each data feature type based on the initial sample feature value, the inflection point corresponding to the feature value inflection point and the termination sample feature value. Determine whether there is a feature value inflection point in the monitoring type curve corresponding to each data feature type in the feature monitoring curve. If there is, determine the termination monitoring feature value corresponding to the monitoring type curve and the inflection feature value corresponding to the feature value curve. Then, determine the real-time growth trend corresponding to each data feature type based on the warning feature value, the inflection feature value and the termination monitoring feature value. Based on the sample growth trend, the real-time growth trend is predicted, and the floating change value corresponding to each data feature type in the real-time growth trend in the future time period is determined. The feature monitoring curve is virtualized and extended based on the changed floating value, and the building early warning anomaly and the early warning anomaly time are determined based on the extended feature monitoring curve.

[0011] In one possible implementation, determining the sample growth trend corresponding to each data feature type based on the initial sample feature value, the tortuous feature value corresponding to the feature value inflection point, and the termination sample feature value includes: Determine whether the tortuous feature value is a single tortuous feature value. If it is a single tortuous feature value, then determine the first feature difference and the first feature time period based on the initial sample feature value and the tortuous feature value. Determine the second feature difference and the second feature time period based on the tortuous feature value and the terminated sample feature value. Calculate the ratio of the first feature difference to the first feature time period and the ratio of the second feature difference to the second feature time period to obtain the first feature trend corresponding to the first feature difference and the second feature trend corresponding to the second feature difference. The first feature trend and the second feature trend are integrated and calculated according to the ratio between the first feature time period and the second feature time period to obtain the sample growth trend corresponding to each data feature type.

[0012] In one possible implementation, determining whether the tortuous feature value is a single tortuous feature value includes: If the tortuous feature value is not a single tortuous feature value, the tortuous feature values ​​are numbered according to the time sequence, and the first feature difference and the first feature time period between adjacent tortuous feature values ​​are determined according to the adjacency characteristics between tortuous feature values. The second feature difference and the second feature time period are determined according to the initial sample feature value and the first tortuous feature value. The third feature difference and the third feature time period are determined according to the termination sample feature value and the Nth tortuous feature value. Calculate the ratio of the first feature difference to the first feature time period, the ratio of the second feature difference to the second feature time period, and the ratio of the third feature difference to the third feature time period, respectively, to obtain the first feature trend corresponding to the first feature difference, the second feature trend corresponding to the second feature difference, and the third feature trend corresponding to the third feature difference, where N is the total number of tortuous feature values; The first characteristic trend, the second characteristic trend, and the third characteristic trend are integrated and calculated according to the proportional relationship between the first characteristic time period, the second characteristic time period, and the third characteristic time period to obtain the sample growth trend corresponding to each data characteristic type.

[0013] Secondly, this application provides a construction safety monitoring system, which adopts the following technical solution: A construction safety monitoring system, comprising: The data acquisition module is used to collect construction data for each building under construction in the construction area within the current time period. The construction data includes tilt angle monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data. The environmental monitoring data includes external environmental data and internal environmental data. The data processing module is used to preprocess the construction data to obtain processed construction data. The feature extraction module is used to extract tilt features, vibration features, displacement features, first environmental features and second environmental features from the processed building construction data, and input the tilt features, vibration features, displacement features, first environmental features and second environmental features into the trained preset construction early warning model for recognition, to obtain building recognition results; The alarm control module is used to determine whether there is feature alarm information in the building identification results. If so, it generates a building alarm command based on the feature alarm information and controls the alarm device to output a response.

[0014] In one possible implementation, when determining whether feature alarm information exists in the building identification result, the alarm control module is specifically used for: If not, determine whether there is feature warning information in the building identification result. If there is, determine the warning feature type, warning sample data and warning feature value corresponding to the warning feature type based on the feature warning information, and retrieve feature monitoring data within a preset time period according to the warning feature type. The building warning anomaly and the warning anomaly time are determined based on the feature monitoring data, the warning sample data, and the warning feature value. Based on the building warning anomaly and the time of the warning anomaly, a building warning command is generated, and the warning equipment is controlled to output a response.

[0015] In another possible implementation, the system further includes: a data acquisition module, a field matching module, an anomaly detection module, a data concatenation module, and a model training module, wherein, The data acquisition module is used to collect building monitoring data within a historical time period. The building monitoring data includes tilt angle monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data inside and outside the building for each construction building in different construction areas. The field matching module is used to use the construction area, the construction building, and the environmental monitoring data as matching fields to match the building monitoring data with the construction data within the current time period, thereby obtaining a set of sample data in the building monitoring data that corresponds to the construction data. The anomaly determination module is used to determine the abnormal time nodes in the sample data set where there are abnormal monitoring data and the target experimental data corresponding to the abnormal monitoring data, and to collect evolution experimental data corresponding to the target experimental data according to the abnormal time nodes and a preset time period. The data splicing module is used to splice the target experimental data and the evolution experimental data according to time nodes to obtain periodic sample data. The model training module is used to input the abnormal monitoring data and the periodic sample data into the preset construction early warning feature model according to the corresponding relationship for training, so as to obtain the trained preset construction early warning feature model.

[0016] In another possible implementation, when the alarm control module determines the building warning anomaly and the warning anomaly time based on the feature monitoring data, the warning sample data, and the warning feature value, it is specifically used for: Construct a feature monitoring coordinate system, where the X-axis represents different time nodes and the Y-axis represents different unit feature values ​​corresponding to each type of feature monitoring data. The feature monitoring data and the early warning feature value are respectively imported into the feature monitoring coordinate system, and the monitoring data points corresponding to the feature monitoring data and the early warning data points corresponding to the early warning feature value are connected in time order to obtain the feature monitoring curve; The early warning sample data is imported into the feature monitoring coordinate system, and the monitoring data points corresponding to the early warning sample data are connected in chronological order to obtain the sample monitoring curve. The characteristic monitoring curve and the sample monitoring curve are analyzed to determine the building early warning anomaly and the time of the early warning anomaly.

[0017] In another possible implementation, when the alarm control module analyzes the feature monitoring curve and the sample monitoring curve to determine the building warning anomaly and the warning anomaly time, it is specifically used for: Determine whether there is a feature value inflection point in the sample type curve corresponding to each data feature type in the sample monitoring curve. If there is, determine the initial sample feature value and the termination sample feature value corresponding to the sample type curve, and determine the sample growth trend corresponding to each data feature type based on the initial sample feature value, the inflection point corresponding to the feature value inflection point and the termination sample feature value. Determine whether there is a feature value inflection point in the monitoring type curve corresponding to each data feature type in the feature monitoring curve. If there is, determine the termination monitoring feature value corresponding to the monitoring type curve and the inflection feature value corresponding to the feature value curve. Then, determine the real-time growth trend corresponding to each data feature type based on the warning feature value, the inflection feature value and the termination monitoring feature value. Based on the sample growth trend, the real-time growth trend is predicted, and the floating change value corresponding to each data feature type in the real-time growth trend in the future time period is determined. The feature monitoring curve is virtualized and extended based on the changed floating value, and the building early warning anomaly and the early warning anomaly time are determined based on the extended feature monitoring curve.

[0018] In another possible implementation, when the alarm control module determines the sample growth trend corresponding to each data feature type based on the initial sample feature value, the tortuous feature value corresponding to the feature value inflection point, and the termination sample feature value, it is specifically used for: Determine whether the tortuous feature value is a single tortuous feature value. If it is a single tortuous feature value, then determine the first feature difference and the first feature time period based on the initial sample feature value and the tortuous feature value. Determine the second feature difference and the second feature time period based on the tortuous feature value and the terminated sample feature value. Calculate the ratio of the first feature difference to the first feature time period and the ratio of the second feature difference to the second feature time period to obtain the first feature trend corresponding to the first feature difference and the second feature trend corresponding to the second feature difference. The first feature trend and the second feature trend are integrated and calculated according to the ratio between the first feature time period and the second feature time period to obtain the sample growth trend corresponding to each data feature type.

[0019] In another possible implementation, when determining whether the tortuous feature value is a single tortuous feature value, the alarm control module is specifically used for: If the tortuous feature value is not a single tortuous feature value, the tortuous feature values ​​are numbered according to the time sequence, and the first feature difference and the first feature time period between adjacent tortuous feature values ​​are determined according to the adjacency characteristics between tortuous feature values. The second feature difference and the second feature time period are determined according to the initial sample feature value and the first tortuous feature value. The third feature difference and the third feature time period are determined according to the termination sample feature value and the Nth tortuous feature value. Calculate the ratio of the first feature difference to the first feature time period, the ratio of the second feature difference to the second feature time period, and the ratio of the third feature difference to the third feature time period, respectively, to obtain the first feature trend corresponding to the first feature difference, the second feature trend corresponding to the second feature difference, and the third feature trend corresponding to the third feature difference, where N is the total number of tortuous feature values; The first characteristic trend, the second characteristic trend, and the third characteristic trend are integrated and calculated according to the proportional relationship between the first characteristic time period, the second characteristic time period, and the third characteristic time period to obtain the sample growth trend corresponding to each data characteristic type.

[0020] Thirdly, this application provides an electronic device that adopts the following technical solution: At least one processor; Memory; At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, the at least one application being configured to: perform a construction safety monitoring method as described in any of the first aspects.

[0021] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform a construction safety monitoring method as described in any of the first aspects.

[0022] In summary, this application includes at least one of the following beneficial technical effects: By employing the aforementioned technical solution, construction data for each building under construction within the current timeframe is collected, encompassing multiple dimensions including tilt angle, vibration, displacement, and environmental factors (both external and internal). This comprehensively and meticulously reflects the dynamic changes of the building itself and its surrounding environment during construction. The construction data is then preprocessed, removing noise and filling in missing values, resulting in processed construction data. This ensures the accuracy and completeness of the data, providing a reliable foundation for subsequent feature extraction and model recognition. This allows the building recognition results based on this data to more accurately reflect the actual state of construction, effectively avoiding recognition biases caused by data issues. After obtaining the processed construction data, tilt angle features, vibration features, displacement features, first environmental features, and second environmental features are extracted. Features are key information carriers of the construction status; for example, tilt angle features directly reflect the degree of tilt of the building structure, and vibration features reflect the external forces acting on the building. These features are input into a pre-trained construction early warning model for recognition. A comprehensive analysis of the tilt angle features, vibration features, displacement features, first environmental features, and second environmental features yields accurate building recognition results. This significantly improves the accuracy and comprehensiveness of assessing the construction status. After obtaining the building identification results, it checks for the presence of characteristic alarm information. If characteristic alarm information is found, a building alarm command is generated based on this information, controlling the alarm equipment to output a response. This allows for the rapid issuance of alarms when abnormalities occur during construction, alerting relevant personnel to take measures to prevent further escalation of accidents, greatly improving construction safety and protecting the lives of construction workers and the quality of the building. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating a construction safety monitoring method provided in an embodiment of this application.

[0024] Figure 2 This is a structural schematic diagram of a building construction safety monitoring system provided in an embodiment of this application.

[0025] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] The following is in conjunction with the appendix Figure 1-3This application will be described in further detail.

[0027] This specific embodiment is merely an explanation of this application and is not intended to limit it. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they are within the scope of this application.

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0029] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0030] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0031] This application provides a method for monitoring construction safety, executed by an electronic device. This electronic device can be a standalone physical electronic device, a cluster of multiple physical electronic devices, a distributed system, or a cloud electronic device providing cloud computing services. This application does not impose any limitations on this method. Figure 1 As shown, the method includes: Step S10: Collect the construction data corresponding to each construction building in the construction area within the current time period.

[0032] The construction data includes tilt monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data, which includes external environmental data and internal environmental data.

[0033] In the embodiments of this application, the construction area refers to a specific geographical area where construction activities are underway. The construction building refers to a building being constructed or renovated within this area. Tilt monitoring data refers to data indicating changes in the tilt angle of the construction building, acquired through tilt sensors, reflecting whether the building is tilting and the degree of tilt. Vibration monitoring data refers to data recording the impact of vibration on the construction building during construction, acquired through vibration sensors, allowing for timely understanding of the building structure's response to vibration. Displacement monitoring data refers to data reflecting the positional movement of the construction building, acquired through displacement sensors, monitoring for abnormal displacement. External environmental data refers to various indicators of the external environment of the construction area, such as temperature, humidity, wind speed, and wind direction, which affect construction progress and building quality. Internal environmental data refers to various indicators of the internal environment of the construction area, such as air quality, dust concentration, and noise levels, which have a significant impact on the health and safety of construction workers.

[0034] Specifically, the time period for data collection needs to be determined, based on key construction milestones, monitoring frequency requirements, or specific construction stages. For example, frequent data collection is required during the foundation construction phase to closely monitor the stability of the building foundation. Next, the scope of the construction area is defined, which can be achieved through a site plan, boundary markers, or geographic coordinates, ensuring that the collected data covers the entire area requiring monitoring. Then, each construction building within the area is identified. For large construction sites with multiple buildings of different functions and construction progress, each building is identified and recorded individually. Afterward, for each construction building, appropriate monitoring equipment is used to collect construction data. For tilt monitoring data, tilt sensors are installed at appropriate locations on the building, such as the top or key structural parts. The sensors measure the building's tilt angle in real time and transmit the data to the data acquisition system. Vibration monitoring data is obtained by installing vibration sensors on the building structure. These sensors can detect vibrations during different construction activities (such as piling and blasting). Displacement monitoring data utilizes displacement sensors, fixed at the locations where displacement occurs, to monitor changes in building displacement. Regarding environmental monitoring data, environmental monitoring stations are set up at appropriate locations outside the building, and sensors such as temperature, humidity, and wind speed are installed to collect external environmental data. Inside the building, sensors for air quality, dust concentration, and noise levels are deployed according to the functions of different areas and the activities of people to collect internal environmental data. Finally, all collected building construction data are aggregated and stored in designated data storage devices or systems for subsequent analysis and processing.

[0035] Step S11: Preprocess the construction data to obtain the processed construction data.

[0036] Specifically, a preliminary check is performed on the collected construction data to examine for obvious missing, incorrect, or abnormal data. For example, check if any values ​​in the monitoring data exceed the sensor's range, and if there are any missing or duplicate timestamps in the temperature data. For missing data, an appropriate filling method is selected based on the data type and characteristics, such as using the average or median of adjacent data, or predictive filling based on historical data. For erroneous data, such as incorrect time recordings, corrections are performed to ensure the data's chronological order and accuracy. For abnormal data, the cause needs further analysis. If it is due to accidental factors such as sensor malfunction or external interference, it can be removed; if it is caused by special circumstances during construction, it needs to be marked and explained so that it can be correctly handled in subsequent analysis. Next, the data is standardized or normalized. If the construction data contains multiple types of indicators with significantly different dimensions and numerical ranges—for example, stress data is in Pascals (Pa), while temperature data is in degrees Celsius (°C)—direct comprehensive analysis of these data would be difficult. Standardization or normalization transforms data from different indicators into the same numerical range, eliminating the influence of units and making the data comparable. Common standardization methods include Z-score standardization, which subtracts the mean from the data and divides by the standard deviation; and Min-Max normalization, which linearly transforms the data to the [0, 1] interval. Finally, the processed data undergoes quality assessment to check whether it meets the requirements for subsequent analysis, such as whether the accuracy, completeness, and consistency of the data meet the preset standards. If the data quality does not meet the requirements, the previous steps need to be returned for further adjustments and optimizations until the processed construction data meets the requirements.

[0037] Step S12: Extract the tilt angle features, vibration features, displacement features, first environmental features, and second environmental features from the processed building construction data, and input the tilt angle features, vibration features, displacement features, first environmental features, and second environmental features into the trained preset construction early warning model for identification, and obtain the building identification results.

[0038] For this embodiment, necessary Python libraries are installed, such as Pandas for data processing, NumPy for numerical computation, Scikit-learn for machine learning model training and prediction, and PyWavelets for signal processing operations such as wavelet transform. The processed construction data is read using the Pandas library and stored as a DataFrame structure. Then, different types of features are extracted. For tilt angle features, NumPy statistical functions are used to calculate the maximum tilt angle, average tilt angle, and rate of change of tilt angle; for vibration features, the PyWavelets library is used to perform wavelet transform on the vibration signal to extract frequency components and energy distribution features; for displacement features, the cumulative displacement and displacement velocity are calculated; for the first and second environmental features, Pandas aggregation functions are used to calculate the corresponding statistical indicators. The extracted features are integrated into a feature vector, and NumPy array operations are used to concatenate the data of different features together.

[0039] Specifically, historical building monitoring data is collected, including tilt angle monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data for each building under construction within different construction areas. The building construction area, the building under construction, and the environmental monitoring data are used as matching fields to match the building monitoring data with the building construction data within the current time period, resulting in a sample data set corresponding to the building construction data. Abnormal time points with monitoring data anomalies are identified in the sample data set, along with the corresponding target experimental data. Evolutionary experimental data corresponding to the target experimental data are collected based on the abnormal time points and a preset time period. The target experimental data and the evolutionary experimental data are then concatenated according to time points to obtain periodic sample data. The abnormal monitoring data and the periodic sample data are input into a preset construction early warning feature model for training, resulting in the trained preset construction early warning feature model.

[0040] In this embodiment of the application, the building recognition result includes feature alarm information and feature warning information within the current time period. The feature alarm information refers to the existing abnormal feature situation, which needs to be processed in a timely manner. The feature warning information indicates that the current feature trend is consistent with the feature trend of historical abnormal features, and abnormal feature situations may occur in the future time period.

[0041] Step S13: Determine whether there is feature alarm information in the building recognition result. If so, generate a building alarm command based on the feature alarm information and control the alarm device to output a response.

[0042] In this embodiment of the application, feature alarm information refers to warning information related to certain abnormal features triggered when certain features (such as tilt angle, vibration, displacement, environmental features, etc.) exceed preset safety thresholds or conform to specific abnormal patterns in the building identification results. This information is used to indicate potential safety hazards or abnormal conditions during building construction. Building alarm commands are control commands generated based on feature alarm information, used to instruct alarm devices to take corresponding output response actions, such as issuing audible alarms, flashing lights, sending SMS notifications, etc., to alert relevant personnel to abnormal situations during building construction.

[0043] Alarm devices are hardware devices used to output alarm information. Common types include sound alarms (such as buzzers and horns), light warning devices (such as red warning lights), electronic displays (used to display text alarm information), and devices integrated with communication systems (such as SMS sending modules and mobile APP push devices).

[0044] For example, at a construction site, if building identification results show that the tilt angle of a certain area exceeds the normal range and the vibration characteristics exhibit abnormal fluctuations, these two abnormal features will generate feature alarm information. Based on this information, the generated building alarm command will control the on-site audible alarm to emit a sharp alarm sound, while a red warning light will begin flashing, alerting construction workers to the safety hazard in the area.

[0045] This application provides a method for monitoring building construction safety. It collects construction data for each building under construction within a construction area during the current time period, covering multi-dimensional data including tilt angle, vibration, displacement, and environmental factors (both inside and outside the building). This comprehensively and meticulously reflects the dynamic changes of the building itself and its surrounding environment during construction. The construction data is then preprocessed to remove noise and fill in missing values, resulting in processed construction data. This ensures the accuracy and completeness of the data, providing a reliable foundation for subsequent feature extraction and model recognition. This allows the building recognition results based on this data to more accurately reflect the actual state of construction, effectively avoiding recognition bias caused by data issues. After obtaining the processed construction data, tilt angle features, vibration features, displacement features, first environmental features, and second environmental features are extracted. These features are key information carriers of the construction status; for example, tilt angle features directly reflect the degree of tilt of the building structure, and vibration features reflect the external forces acting on the building. These features are input into a trained, pre-set construction early warning model for recognition. A comprehensive analysis of the tilt angle features, vibration features, displacement features, first environmental features, and second environmental features yields accurate building recognition results. This significantly improves the accuracy and comprehensiveness of assessing the construction status. After obtaining the building identification results, it checks for the presence of characteristic alarm information. If characteristic alarm information is found, a building alarm command is generated based on this information, controlling the alarm equipment to output a response. This allows for the rapid issuance of alarms when abnormalities occur during construction, alerting relevant personnel to take measures to prevent further escalation of accidents, greatly improving construction safety and protecting the lives of construction workers and the quality of the building.

[0046] Furthermore, when no feature alarm information is found in the building identification results, it is determined whether feature warning information exists in the building identification results. If it does, the warning feature type, warning sample data, and warning feature value corresponding to the warning feature type are determined based on the feature warning information. Feature monitoring data within a preset time period is retrieved according to the warning feature type. Building warning anomalies and their times are determined based on the feature monitoring data, warning sample data, and warning feature values. Building warning commands are generated based on the building warning anomalies and their times, controlling the warning equipment to output a response.

[0047] Specifically, a feature monitoring coordinate system is constructed, with the X-axis representing different time nodes and the Y-axis representing different unit feature values ​​corresponding to each type of feature monitoring data. Feature monitoring data and early warning feature values ​​are imported into the feature monitoring coordinate system, and the monitoring data points corresponding to the feature monitoring data and the early warning data points corresponding to the early warning feature values ​​are connected in chronological order to obtain the feature monitoring curve. Early warning sample data is also imported into the feature monitoring coordinate system, and the monitoring data points corresponding to the early warning sample data are connected in chronological order to obtain the sample monitoring curve. The feature monitoring curve and the sample monitoring curve are analyzed to determine building early warning anomalies and their timing.

[0048] When analyzing feature monitoring curves and sample monitoring curves, it is determined whether there are characteristic value inflection points in the sample type curve corresponding to each data feature type in the sample monitoring curve. If so, the initial sample feature value and the termination sample feature value corresponding to the sample type curve are determined, and the sample growth trend corresponding to each data feature type is determined based on the initial sample feature value, the inflection point corresponding to the feature value inflection point, and the termination sample feature value. It is also determined whether there are characteristic value inflection points in the monitoring type curve corresponding to each data feature type in the feature monitoring curve. If so, the termination monitoring feature value and the inflection feature value corresponding to the feature value curve are determined, and the real-time growth trend corresponding to each data feature type is determined based on the warning feature value, the inflection feature value, and the termination monitoring feature value. Then, the real-time growth trend is predicted based on the sample growth trend, and the change floating value corresponding to each data feature type in the real-time growth trend is determined in the future time period. The feature monitoring curve is virtualized and extended based on the change floating value, and the building warning anomaly and the warning anomaly time are determined based on the extended feature monitoring curve.

[0049] Specifically, it is determined whether the tortuous feature value is a single tortuous feature value. If it is a single tortuous feature value, then the first feature difference and the first feature time period are determined based on the initial sample feature value and the tortuous feature value. The second feature difference and the second feature time period are determined based on the tortuous feature value and the final sample feature value. The ratio of the first feature difference to the first feature time period and the ratio of the second feature difference to the second feature time period are calculated respectively to obtain the first feature trend corresponding to the first feature difference and the second feature trend corresponding to the second feature difference. The first feature trend and the second feature trend are integrated and calculated according to the ratio relationship between the first feature time period and the second feature time period to obtain the sample growth trend corresponding to each data feature type.

[0050] If the tortuous feature value is not a single tortuous feature value, it is numbered according to time sequence. The first feature difference and first feature time period between adjacent tortuous feature values ​​are determined based on their adjacency characteristics. The second feature difference and second feature time period are determined based on the initial sample feature value and the first tortuous feature value. The third feature difference and third feature time period are determined based on the terminating sample feature value and the Nth tortuous feature value. The ratios of the first feature difference to the first feature time period, the second feature difference to the second feature time period, and the third feature difference to the third feature time period are calculated respectively to obtain the first feature trend corresponding to the first feature difference, the second feature trend corresponding to the second feature difference, and the third feature trend corresponding to the third feature difference. Here, N is the total number of tortuous feature values. The first feature trend, the second feature trend, and the third feature trend are integrated and calculated according to the proportional relationship between the first feature time period, the second feature time period, and the third feature time period to obtain the sample growth trend corresponding to each data feature type.

[0051] In this embodiment, the method for determining the real-time growth trend is the same as that for determining the sample growth trend, and will not be repeated here. It is worth noting that the virtualized extended feature monitoring curve and the actual monitored feature monitoring curve are two different curves. This extended curve cannot directly replace the actual monitored feature curve, and the extended feature monitoring curve will be updated in real time based on the actual monitored feature monitoring curve. When the actual monitored feature monitoring curve shows a downward trend, the corresponding extended feature monitoring curve updates the existing curve trend in real time and changes the subsequent floating value. However, when the actual monitored feature monitoring curve changes from an upward trend to a downward trend, the corresponding extended feature monitoring curve has no reference value. Simultaneously, the actual behavior analysis of the feature monitoring curve is performed to determine the feature change details and actual triggers that cause the upward trend. These feature change details and actual triggers are stored in the first database. When an extended feature monitoring curve appears subsequently, it is first matched with the feature change details in the first database to determine whether the curve increase is caused by other normal factors (excessively high ambient temperature, temperature drop in the evening).

[0052] The following describes a construction safety monitoring system provided in an embodiment of this application. The construction safety monitoring system described below can be referred to in conjunction with the construction safety monitoring method described above. Figure 2 , Figure 2 This is a structural schematic diagram of a construction safety monitoring system 20 provided in an embodiment of this application, including: Data acquisition module 21 is used to collect construction data for each building in the construction area during the current time period. The construction data includes tilt angle monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data. The environmental monitoring data includes external environmental data and internal environmental data. Data processing module 22 is used to preprocess the construction data to obtain the processed construction data. The feature extraction module 23 is used to extract the tilt angle features, vibration features, displacement features, first environmental features and second environmental features from the processed building construction data, and input the tilt angle features, vibration features, displacement features, first environmental features and second environmental features into the trained preset construction early warning model for recognition, and obtain the building recognition result; The alarm control module 24 is used to determine whether there is feature alarm information in the building identification results. If so, it generates a building alarm command based on the feature alarm information and controls the alarm device to output a response.

[0053] In one possible implementation of this application embodiment, when the alarm control module 24 determines whether feature alarm information exists in the building identification result, it is specifically used for: If not, determine whether there is feature warning information in the building recognition results. If there is, determine the warning feature type, warning sample data and warning feature value corresponding to the warning feature type based on the feature warning information, and retrieve the feature monitoring data within the preset time period according to the warning feature type. The building warning anomalies and their timing are determined based on feature monitoring data, early warning sample data, and early warning feature values. Based on the building warning anomaly and the time of the anomaly, a building warning command is generated, and the warning equipment is controlled to output a response.

[0054] In another possible implementation of this application embodiment, system 20 further includes: a data acquisition module, a field matching module, an anomaly determination module, a data splicing module, and a model training module, wherein, The data acquisition module is used to collect building monitoring data within a historical time period. The building monitoring data includes tilt angle monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data inside and outside the building for each construction building in different construction areas. The field matching module is used to match building construction area, construction building, and environmental monitoring data as matching fields, and to match building monitoring data with building construction data within the current time period to obtain a set of sample data in the building monitoring data that corresponds to the building construction data. The anomaly determination module is used to determine the abnormal time points in the sample data set where there are abnormal monitoring data and the target experimental data corresponding to the abnormal monitoring data, and to collect the evolution experimental data corresponding to the target experimental data according to the abnormal time points and the preset time period. The data stitching module is used to stitch together the target experimental data and the evolution experimental data according to time nodes to obtain periodic sample data. The model training module is used to input abnormal monitoring data and periodic sample data into the preset construction early warning feature model according to the corresponding relationship, so as to obtain the trained preset construction early warning feature model.

[0055] In another possible implementation of this application embodiment, when the alarm control module 24 determines the building warning anomaly and the warning anomaly time based on the feature monitoring data, warning sample data, and warning feature values, it is specifically used for: Construct a feature monitoring coordinate system, where the X-axis represents different time nodes and the Y-axis represents different unit feature values ​​corresponding to each type of feature monitoring data. The feature monitoring data and early warning feature values ​​are imported into the feature monitoring coordinate system, and the monitoring data points corresponding to the feature monitoring data and the early warning data points corresponding to the early warning feature values ​​are connected in time sequence to obtain the feature monitoring curve. Import the early warning sample data into the feature monitoring coordinate system, and connect the monitoring data points corresponding to the early warning sample data according to the time sequence to obtain the sample monitoring curve; The characteristic monitoring curves and sample monitoring curves are analyzed to determine the building early warning anomalies and the timing of the anomalies.

[0056] In another possible implementation of this application embodiment, when the alarm control module 24 analyzes the feature monitoring curve and the sample monitoring curve to determine the building early warning anomaly and the early warning anomaly time, it is specifically used for: Determine whether there are characteristic value inflection points in the sample type curve corresponding to each data feature type in the sample monitoring curve. If so, determine the initial sample feature value and the termination sample feature value corresponding to the sample type curve, and determine the sample growth trend corresponding to each data feature type based on the initial sample feature value, the inflection point of the feature value, and the termination sample feature value. Determine whether there is a characteristic value inflection point in the monitoring type curve corresponding to each data feature type in the feature monitoring curve. If so, determine the termination monitoring characteristic value corresponding to the monitoring type curve and the inflection characteristic value corresponding to the characteristic value curve. Based on the warning characteristic value, the inflection characteristic value and the termination monitoring characteristic value, determine the real-time growth trend corresponding to each data feature type. Based on the sample growth trend, predict the real-time growth trend and determine the floating value of the change in the real-time growth trend for each data feature type in the future time period; The feature monitoring curve is virtualized and extended based on the changing floating value, and the building early warning anomaly and the early warning anomaly time are determined based on the extended feature monitoring curve.

[0057] In another possible implementation of this application embodiment, when the alarm control module 24 determines the sample growth trend corresponding to each data feature type based on the initial sample feature value, the tortuous feature value corresponding to the feature value inflection point, and the termination sample feature value, it is specifically used for: Determine whether the tortuous feature value is a single tortuous feature value. If it is a single tortuous feature value, determine the first feature difference and the first feature time period based on the initial sample feature value and the tortuous feature value. Determine the second feature difference and the second feature time period based on the tortuous feature value and the final sample feature value. Calculate the ratio of the first feature difference to the first feature time period and the ratio of the second feature difference to the second feature time period, respectively, to obtain the first feature trend corresponding to the first feature difference and the second feature trend corresponding to the second feature difference. The first and second feature trends are integrated and calculated according to the ratio between the first and second feature time periods to obtain the sample growth trend corresponding to each data feature type.

[0058] In another possible implementation of this application embodiment, when determining whether the tortuous feature value is a single tortuous feature value, the alarm control module 24 is specifically used for: If the tortuous feature value is not a single tortuous feature value, the tortuous feature values ​​are numbered according to the time sequence, and the first feature difference and the first feature time period between adjacent tortuous feature values ​​are determined according to the adjacency characteristics between tortuous feature values. The second feature difference and the second feature time period are determined according to the initial sample feature value and the first tortuous feature value. The third feature difference and the third feature time period are determined according to the terminating sample feature value and the Nth tortuous feature value. Calculate the ratio of the first feature difference to the first feature time period, the ratio of the second feature difference to the second feature time period, and the ratio of the third feature difference to the third feature time period, respectively, to obtain the first feature trend corresponding to the first feature difference, the second feature trend corresponding to the second feature difference, and the third feature trend corresponding to the third feature difference. N is the total number of tortuous feature values. The first characteristic trend, the second characteristic trend, and the third characteristic trend are integrated and calculated according to the proportional relationship between the first characteristic time period, the second characteristic time period, and the third characteristic time period to obtain the sample growth trend corresponding to each data characteristic type.

[0059] This application provides an electronic device, such as... Figure 3 As shown, Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 3 The illustrated electronic device 300 includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, via a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one type, and the structure of this electronic device 300 does not constitute a limitation on the embodiments of this application.

[0060] Processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in connection with the embodiments of this application. Processor 301 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0061] Bus 302 may include a pathway for transmitting information between the aforementioned components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 302 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0062] The memory 303 may be a ROM (Read-Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or it may be an EEPROM (Electrically Erasable Programmable Read-Only Memory), a CD-ROM (Compact Disc Read-Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0063] The memory 303 is used to store application code that executes the scheme of the embodiments of this application, and its execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0064] Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0065] The following describes a computer-readable storage medium provided by an embodiment of this application. The computer-readable storage medium described below can be referred to in correspondence with the method described above.

[0066] This application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-mentioned construction safety monitoring system.

[0067] Since the embodiments of the computer-readable storage medium portion correspond to the embodiments of the method portion, please refer to the description of the embodiments of the method portion for the embodiments of the computer-readable storage medium portion.

[0068] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0069] The above are only some embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method of construction safety monitoring, characterized by, The method comprises the following steps: collecting building construction data corresponding to each construction building in a construction area in a current time period, the building construction data comprising inclination monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data, the environmental monitoring data comprising external building environment data and internal building environment data; respectively pre-processing the building construction data to obtain processed building construction data; extracting inclination features, vibration features, displacement features, first environmental features, and second environmental features from the processed building construction data, and inputting the inclination features, the vibration features, the displacement features, the first environmental features, and the second environmental features into a trained preset construction warning model for identification to obtain building identification results; determining whether there is feature warning information in the building identification results, if there is, generating a building warning instruction based on the feature warning information to control the output response of the warning device; if there is not, determining whether there is feature warning information in the building identification results, if there is, determining a warning feature type, warning sample data, and a warning feature value corresponding to the warning feature type based on the feature warning information, and calling feature monitoring data in a preset time period according to the warning feature type; determining a building warning anomaly and a warning anomaly time according to the feature monitoring data, the warning sample data, and the warning feature value; generating a building warning instruction according to the building warning anomaly and the warning anomaly time to control the output response of the warning device.

2. The method of claim 1, wherein, The method further comprises the following steps before inputting the inclination features, the vibration features, the displacement features, the first environmental features, and the second environmental features into the trained preset construction warning model for identification to obtain the building identification results: collecting building monitoring data in a historical time period, the building monitoring data being inclination monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data inside and outside different construction buildings in different construction areas; matching the building monitoring data and the building construction data in the current time period by taking the construction area, the construction building, and the environmental monitoring data as matching fields to obtain a sample data set corresponding to the building construction data in the building monitoring data; determining an abnormal time node with monitoring data anomaly and target experimental data corresponding to the monitoring data anomaly in the sample data set, and collecting evolution experimental data corresponding to the target experimental data according to the abnormal time node and a preset time period; respectively splicing the target experimental data and the evolution experimental data according to time nodes to obtain periodic sample data; inputting the monitoring data anomaly and the periodic sample data into a preset construction warning feature model according to a corresponding relationship for training to obtain a trained preset construction warning feature model.

3. A method of construction safety monitoring according to claim 2, wherein, The method further comprises the following steps of determining a building warning anomaly and a warning anomaly time according to feature monitoring data, warning sample data, and a warning feature value: A feature monitoring coordinate system is constructed, an X axis of the feature monitoring coordinate system is different time nodes, and a Y axis of the feature monitoring coordinate system is different unit feature values corresponding to each feature monitoring data type; The feature monitoring data and the early warning feature value are respectively imported into the feature monitoring coordinate system, and monitoring data points corresponding to the feature monitoring data and early warning data points corresponding to the early warning feature value are connected in time sequence to obtain a feature monitoring curve; The early warning sample data is imported into the feature monitoring coordinate system, and monitoring data points corresponding to the early warning sample data are connected in time sequence to obtain a sample monitoring curve; The feature monitoring curve and the sample monitoring curve are analyzed to determine a building early warning anomaly and an early warning anomaly time.

4. The method of claim 3, wherein, The analysis of the feature monitoring curve and the sample monitoring curve to determine a building early warning anomaly and an early warning anomaly time comprises: It is determined whether a sample type curve corresponding to each data feature type in the sample monitoring curve has a feature value inflection point, if so, initial sample feature values and terminal sample feature values corresponding to the sample type curve are determined, and a sample growth trend corresponding to each data feature type is determined according to the initial sample feature values, an inflection feature value corresponding to the feature value inflection point, and the terminal sample feature values; It is determined whether a monitoring type curve corresponding to each data feature type in the feature monitoring curve has a feature value inflection point, if so, a terminal monitoring feature value corresponding to the monitoring type curve and an inflection feature value corresponding to the feature value curve are determined, and a real-time growth trend corresponding to each data feature type is determined according to the early warning feature value, the inflection feature value, and the terminal monitoring feature value; The real-time growth trend is predicted according to the sample growth trend to determine a change floating value corresponding to each data feature type in a future time period; The feature monitoring curve is virtually expanded according to the change floating value, and a building early warning anomaly and an early warning anomaly time are determined based on the expanded feature monitoring curve.

5. A method of construction safety monitoring according to claim 4, wherein, The determination of the sample growth trend according to the initial sample feature values, the inflection feature value corresponding to the feature value inflection point, and the terminal sample feature values comprises: It is determined whether the inflection feature value is a single inflection feature value, if so, a first feature difference and a first feature period are determined according to the initial sample feature values and the inflection feature value, a second feature difference and a second feature period are determined according to the inflection feature value and the terminal sample feature values, and a ratio of the first feature difference to the first feature period and a ratio of the second feature difference to the second feature period are respectively calculated to obtain a first feature trend corresponding to the first feature difference and a second feature trend corresponding to the second feature difference; The first feature trend and the second feature trend are integrated and calculated according to a proportional relationship of the first feature period and the second feature period, to obtain a sample growth trend corresponding to each data feature type.

6. A method of construction safety monitoring according to claim 5, wherein, The determination of whether the tortuous feature value is a single tortuous feature value comprises: If the tortuous feature value is not a single tortuous feature value, the tortuous feature values are sequentially marked according to time sequence, and a first feature difference and a first feature period between adjacent tortuous feature values are determined according to adjacent characteristics between the tortuous feature values, a second feature difference and a second feature period are determined according to the initial sample feature value and the first tortuous feature value, and a third feature difference and a third feature period are determined according to the terminal sample feature value and the Nth tortuous feature value; The ratios of the first feature difference to the first feature period, the second feature difference to the second feature period, and the third feature difference to the third feature period are calculated respectively, to obtain a first feature trend corresponding to the first feature difference, a second feature trend corresponding to the second feature difference, and a third feature trend corresponding to the third feature difference, and N is the total number of the tortuous feature values; The first feature trend, the second feature trend, and the third feature trend are integrated and calculated according to a proportional relationship of the first feature period, the second feature period, and the third feature period, to obtain a sample growth trend corresponding to each data feature type.

7. A construction safety monitoring system characterized by, Comprise: The data acquisition module is used for acquiring building construction data corresponding to each construction building in a building construction area in a current time period, the building construction data comprising inclination monitoring data, vibration monitoring data, displacement monitoring data, and environmental monitoring data, the environmental monitoring data comprising building external environment data and building internal environment data; The data processing module is used for pre-processing the building construction data respectively, to obtain processed building construction data; The feature extraction module is used for extracting inclination features, vibration features, displacement features, first environmental features, and second environmental features from the processed building construction data, and inputting the inclination features, the vibration features, the displacement features, the first environmental features, and the second environmental features into a trained preset construction early warning model for identification, to obtain a building identification result; The alarm control module is used for judging whether there is feature alarm information in the building identification result, and if so, generating a building alarm instruction based on the feature alarm information, and controlling an alarm device to output a response; When judging whether there is feature alarm information in the building identification result, the alarm control module is specifically used for: If not, judging whether there is feature early warning information in the building identification result, and if so, determining an early warning feature type, early warning sample data, and an early warning feature value corresponding to the early warning feature type according to the feature monitoring data in a preset time period, determining a building early warning anomaly and an early warning anomaly time according to the early warning sample data and the early warning feature value; ​ Generate a building warning instruction according to the building warning anomaly and the warning anomaly time, and control a warning device to output a response.

8. An electronic device, comprising: The electronic device includes: at least one processor; a memory; at least one application program, wherein the at least one application program is stored in the memory and is configured to be executed by the at least one processor, and the at least one application program is configured to execute the building construction safety monitoring method in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, including: a computer program stored with the processor capable of being loaded and executed as claimed in any one of claims 1-6. a computer program stored with the processor capable of being loaded and executed as claimed in any one of claims 1-6.

Images