A construction safety environment monitoring system and a monitoring data analysis method

Abstract

This invention relates to the field of construction monitoring technology, and discloses a construction safety environment monitoring system and a monitoring data analysis method. The system includes data acquisition, risk assessment, adjustment execution, anomaly analysis, strategy correction, and failure early warning units. The data acquisition unit acquires environmental and structural parameters of the construction site at fixed intervals; the risk assessment unit performs threshold judgments on the parameters and generates anomaly type signals; the adjustment execution unit performs parameter adjustments according to the priority of anomaly types; the anomaly analysis unit calculates the parameter offset ratio and adjustment efficiency value based on the anomaly signals and adjustment data; the strategy correction unit adjusts the data acquisition frequency and optimizes the adjustment intensity accordingly; and the failure early warning unit counts the total number of anomalies and generates failure signals based on the construction duration. This system achieves comprehensive monitoring of the construction safety environment, anomaly handling, system optimization, and risk early warning.

Description

Technical Field

[0001] This invention relates to the field of building construction monitoring technology, specifically to a building construction safety environment monitoring system and a monitoring data analysis method. Background Technology

[0002] During construction, the environmental conditions and structural stability of the construction site directly affect the smooth progress of the project. Traditional construction safety management relies heavily on manual inspections, where staff periodically visit the site to collect data and observe conditions to determine the presence of potential safety hazards. However, manual inspections have significant limitations. Fixed inspection cycles make it difficult to capture dynamic changes in environmental and structural parameters in real time. When sudden anomalies occur, they are often not detected in time, potentially leading to safety accidents. Furthermore, the accuracy of manually collected data is easily affected by human factors, such as the operator's experience and sense of responsibility, which can easily lead to data errors or omissions, affecting the assessment of the construction safety status.

[0003] With the development of the construction industry, some construction sites have begun to introduce simple monitoring equipment to collect some environmental parameters, such as temperature, humidity, and dust concentration. However, most of these devices are single-function, only capable of data collection and simple display, and cannot perform in-depth analysis or risk assessment of the collected data. When environmental or structural parameters become abnormal, they cannot issue early warning signals in a timely manner, nor can they automatically implement corresponding adjustment measures, still requiring manual intervention. This not only increases management costs but also reduces the efficiency and timeliness of safety management.

[0004] Existing monitoring systems lack effective utilization and feedback mechanisms for historical data. They cannot optimize and correct system parameter settings based on past anomalies and adjustment effects, making it difficult to improve system adaptability and accuracy. Furthermore, there are no effective statistical and early warning methods for potential overall failure risks during construction, failing to provide construction managers with comprehensive and reliable safety decision-making references and making it difficult to meet the high standards of safety management required in modern construction. Summary of the Invention

[0005] The purpose of this invention is to provide a construction safety environment monitoring system and a monitoring data analysis method to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a construction safety environment monitoring system and a monitoring data analysis method, the system comprising:

[0007] The data acquisition unit acquires environmental and structural parameters of the construction site at fixed time intervals;

[0008] The risk assessment unit performs threshold judgments on the environmental and structural parameters and generates anomaly type signals.

[0009] The adjustment execution unit adjusts the execution parameters according to the type and priority of the abnormal signal.

[0010] The anomaly analysis unit calculates the parameter offset ratio and regulation efficiency value based on the anomaly type signal and regulation process data;

[0011] The strategy correction unit adjusts the parameter acquisition frequency of the data acquisition unit according to the parameter offset ratio, and optimizes the adjustment intensity of the adjustment execution unit based on the adjustment efficiency value;

[0012] The failure early warning unit counts the total number of abnormal signals from the anomaly analysis unit and generates a construction failure signal by combining the construction duration.

[0013] Preferably, the risk assessment unit compares the environmental parameters and structural parameters with a preset safety parameter range;

[0014] If the vibration amplitude exceeds the preset vibration threshold, an abnormal vibration signal will be generated.

[0015] If the structural settlement exceeds the preset settlement threshold, a settlement anomaly signal will be generated.

[0016] If the dust concentration exceeds the preset dust threshold, a dust abnormality signal will be generated.

[0017] When the risk determination unit generates the settlement anomaly signal, it first triggers the adjustment execution unit.

[0018] Preferably, the adjustment process of the adjustment execution unit for the vibration abnormal signal includes: selecting the median value of the vibration threshold range as the standard vibration value, calculating the vibration difference between the real-time vibration amplitude and the standard vibration value, and generating a power adjustment command for the vibration reduction equipment;

[0019] The adjustment process for the settlement anomaly signal includes: selecting the median value of the settlement threshold range as the standard settlement value, calculating the settlement difference between the real-time settlement and the standard settlement value, and generating a pressure adjustment command for the support structure.

[0020] Preferably, the anomaly analysis unit identifies the real-time vibration amplitude that triggers the vibration anomaly signal and records it as the current vibration parameter;

[0021] The two endpoints of the vibration threshold range are respectively denoted as the lower vibration threshold and the upper vibration threshold.

[0022] Calculate the first vibration difference between the current vibration parameter and the lower vibration threshold, and the second vibration difference between the current vibration parameter and the upper vibration threshold;

[0023] The minimum value between the first vibration difference and the second vibration difference is denoted as the vibration boundary difference;

[0024] The vibration offset ratio is generated by the ratio of the vibration boundary difference to the mean vibration parameter.

[0025] Preferably, the anomaly analysis unit records the time difference between the start of adjustment and the completion of adjustment by the adjustment execution unit, and compares it with a preset adjustment time benchmark to generate an adjustment efficiency value;

[0026] When the vibration offset ratio exceeds a preset offset threshold, the strategy correction unit shortens the fixed time interval of the data acquisition unit proportionally.

[0027] And when the adjustment efficiency value is lower than the efficiency benchmark, the adjustment intensity of the power adjustment command of the vibration reduction equipment is increased.

[0028] Preferably, the system further includes a model prediction unit and an optimization decision unit;

[0029] The model prediction unit receives the historical parameter sequence from the data acquisition unit and outputs structural deformation prediction signal and collapse risk prediction signal through the time series prediction model.

[0030] The optimization decision-making unit integrates the structural deformation prediction signal and real-time structural parameters to generate an optimized adjustment scheme, which is then sent to the adjustment execution unit.

[0031] Preferably, the model prediction unit includes a deformation feature extraction module and a risk fusion module;

[0032] The deformation feature extraction module performs frequency domain feature decomposition on the historical parameter sequence to extract structural deformation feature bands.

[0033] The risk fusion module weights and processes the structural deformation characteristic bands and real-time vibration amplitude data to output the quantified risk value of the collapse risk prediction signal.

[0034] Preferably, the system further includes a cloud-based policy engine and a permission verification unit; the cloud-based policy engine generates a hierarchical verification policy based on the anomaly type signal; the permission verification unit loads the hierarchical verification policy to perform device security level verification, and assigns operation permission levels to the construction equipment according to the verification results.

[0035] Preferably, the permission verification unit grants basic operation permissions when the construction equipment passes the first-level hierarchical verification strategy; grants high-risk operation permissions when it passes the second-level hierarchical verification strategy; and adds the construction equipment to the safety equipment list when it passes the third-level hierarchical verification strategy.

[0036] Preferably, the present invention also includes a method for analyzing construction safety and environmental monitoring data, the method comprising all the modules and process flow of the construction safety and environmental monitoring system described above.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] The data acquisition unit acquires environmental and structural parameters at fixed time intervals, enabling continuous monitoring of the construction environment and structural status, and allowing for timely understanding of dynamic changes in various parameters. The risk assessment unit performs threshold judgments on the acquired parameters and generates anomaly type signals, enabling clear identification of different types of anomalies and facilitating targeted subsequent handling.

[0039] The adjustment execution unit performs parameter adjustments based on the priority of the anomaly type signals, ensuring that when multiple anomalies occur simultaneously, they can be handled in an orderly manner according to their importance, avoiding the escalation of risks due to improper handling sequence. The anomaly analysis unit calculates the parameter offset ratio and adjustment efficiency value based on the anomaly type signals and adjustment process data, providing a data foundation for system optimization. By analyzing the parameter offset ratio, the severity and trend of the anomalies can be understood, while the adjustment efficiency value reflects the actual effect of the adjustment measures.

[0040] The strategy correction unit adjusts the parameter acquisition frequency of the data acquisition unit based on the parameter offset ratio. This allows for increased acquisition frequency during periods of significant parameter fluctuations, enabling more accurate detection of abnormal changes, while reducing the frequency during periods of parameter stability to minimize unnecessary resource consumption. Simultaneously, optimizing the adjustment intensity of the adjustment execution unit based on the adjustment efficiency value ensures more rational and effective adjustment measures, avoiding under- or over-adjustment.

[0041] The failure early warning unit counts the total number of abnormal signals from the anomaly analysis unit and generates construction failure signals in conjunction with the construction duration. This enables an overall assessment of the safety status of the construction process, early detection of potential systemic risks, and provides construction managers with comprehensive risk warning information. This helps to take preventive measures and ensure the smooth progress of the construction process. Attached Figure Description

[0042] Figure 1 This is a timing diagram of the construction safety and environmental monitoring system described in this invention;

[0043] Figure 2 The flowchart for threshold determination of the risk assessment unit;

[0044] Figure 3 A flowchart for calculating the vibration offset ratio of the anomaly analysis unit;

[0045] Figure 4 This is a flowchart of model prediction and optimization decision-making. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Please see Figure 1 This invention provides a construction safety environment monitoring system and a monitoring data analysis method, the system comprising:

[0048] The data acquisition unit collects environmental and structural parameters from the construction site at fixed time intervals. Environmental parameters include dust concentration, temperature, and humidity, while structural parameters include vibration amplitude and structural settlement. Upon receiving these parameters, the risk assessment unit performs threshold judgment to generate anomaly type signals. These signals include vibration anomalies, settlement anomalies, and dust anomalies. The adjustment execution unit performs parameter adjustment operations based on the priority of the anomaly type signals, prioritizing settlement anomalies to ensure structural safety. The anomaly analysis unit calculates the parameter offset ratio and adjustment efficiency value based on the anomaly type signals and the adjustment process data from the adjustment execution unit. The parameter offset ratio reflects the degree to which parameters deviate from the threshold, while the adjustment efficiency value measures the timeliness of the adjustment process. The strategy correction unit dynamically adjusts the fixed time interval of the data acquisition unit based on the parameter offset ratio, shortening the interval to respond to high-risk offsets. Simultaneously, it optimizes the adjustment intensity of the adjustment execution unit based on the adjustment efficiency value to improve response speed. The failure early warning unit counts the total number of anomaly signals recorded by the anomaly analysis unit and generates a construction failure signal based on the construction duration. The construction duration refers to the cumulative time from the start of construction to the present. When the total number of anomaly signals exceeds a preset threshold, a failure signal is generated to warn of overall risk. The system achieves closed-loop control through modular design: the output of the data acquisition unit serves as the input of the risk assessment unit, the adjustment execution unit responds to the assessment results, and the anomaly analysis unit evaluates and feeds back to the strategy correction unit to optimize system behavior.

[0049] Example 1: See Figure 2In the construction safety and environmental monitoring system, the risk assessment unit continuously receives environmental and structural parameters transmitted by the data acquisition unit. Environmental parameters include monitoring indicators such as dust concentration, temperature, and humidity, while structural parameters cover physical quantities such as vibration amplitude and structural settlement. The real-time values ​​of these parameters are automatically input into the threshold comparison module of the risk assessment unit. This module pre-stores safety parameter ranges corresponding to different construction stages; for example, the vibration threshold range is set to 0.5 mm to 2.0 mm, the settlement threshold range is set to 1 mm to 3 mm, and the dust concentration threshold is set to 50 micrograms per cubic meter. When the system performs parameter comparison, it uses a parallel processing mechanism to simultaneously detect whether each parameter exceeds the threshold boundaries. Once the vibration amplitude value exceeds the 0.5–2.0 mm range, the system immediately generates a signal identified as "abnormal vibration"; when the structural settlement data exceeds the 1–3 mm threshold limit, the system simultaneously generates a signal identified as "abnormal settlement"; and when the dust concentration data exceeds 50 micrograms per cubic meter, a signal identified as "abnormal dust" is generated. All abnormal signal types are accompanied by a timestamp, parameter exceeding the limit value, and specific threshold range information to ensure the integrity of the record.

[0050] Upon the generation of an abnormal signal, the risk assessment unit initiates a signal priority sorting mechanism. The sorting logic is based on the structural stability risk level, with settlement anomaly signals assigned the highest processing level due to their involvement in building foundation safety. The system automatically places such signals at the head of the task queue of the adjustment execution unit, ensuring they trigger subsequent adjustment operations first. Vibration and dust anomaly signals enter the secondary queue in chronological order of their generation. This mechanism is implemented through an internal interrupt control module; when a settlement anomaly signal occurs, the system immediately suspends the processing threads of other signals until the settlement adjustment command is completed and output.

[0051] The adjustment execution unit responds to abnormal type signals transmitted by the risk assessment unit and executes differentiated adjustment logic based on the signal type. For abnormal vibration signals, the system extracts the corresponding threshold range (e.g., 0.5–2.0 mm) from a preset database and calculates the median of this range as the standard vibration value. For example, within the threshold range of 0.5 mm lower limit and 2.0 mm upper limit, the system determines the standard vibration value to be 1.25 mm. The adjustment execution unit reads the vibration amplitude data (e.g., 2.5 mm) fed back in real time by the data acquisition unit and calculates the absolute difference between it and the standard vibration value (2.5 mm minus 1.25 mm equals 1.25 mm). Based on this vibration difference, the system generates a power adjustment command for the vibration damping equipment. The command includes the power adjustment amount and direction: a positive vibration difference requires an increase in power, and a negative difference requires a decrease in power. The command is transmitted to the vibration damping equipment controller via the industrial bus. The control algorithm converts the vibration difference into a power increment ratio; for example, every 0.1 mm difference corresponds to a 2% power adjustment, and a 1.25 mm difference triggers a 25% power increase command. During the execution of the adjustment command, the system continuously monitors the changes in vibration amplitude until the data falls back to the threshold range.

[0052] In response to abnormal settlement signals, the adjustment unit executes an independent control process. The system first retrieves settlement threshold range data (e.g., 1–3 mm), taking the median as the standard settlement value (2 mm). Real-time settlement data (e.g., 4 mm) is input into the calculation module, and the difference between this and the standard settlement value (4 mm minus 2 mm equals 2 mm) serves as the core adjustment basis. Based on the settlement difference, the system generates a pressure adjustment command for the support structure, specifying the pressure adjustment magnitude and direction: a positive settlement difference increases the support pressure, while a negative difference decreases it. Upon receiving this command, the hydraulic control system converts the settlement difference into a pressure valve opening adjustment; for example, a 0.5 mm difference corresponds to a 5 MPa pressure increment, and a 2 mm difference generates a 20 MPa pressure increase command. Pressure adjustment employs a closed-loop control mode. The system collects real-time deformation data of the support structure using displacement sensors, compares the feedback data with the target pressure value, and dynamically corrects the valve control parameters. When the real-time settlement recovers to the threshold range, the system marks the adjustment process as complete and generates an operation log.

[0053] The operation logs of the adjustment execution unit are fully saved, including the adjustment start time, completion time, initial parameter values, target parameter values, and actual adjustment results. The recorded data is transmitted to the anomaly analysis unit in a structured format for subsequent analysis. An operation timeout detection mechanism is implemented within the unit. When the duration of a single adjustment exceeds the preset maximum allowable duration (e.g., 300 seconds), the system automatically terminates the current adjustment process, escalates the alarm level, and sends a timeout fault code to the failure warning unit. Throughout the implementation process, the risk assessment unit and the adjustment execution unit synchronize data through shared memory to avoid signal transmission delays. Priority processing mechanisms and parameter differentiation adjustment logic together form the basis of the system's real-time response, ensuring rapid correction capabilities after deviations in safety parameters at the construction site.

[0054] Example 2: See Figure 3 In the construction safety and environmental monitoring system, the anomaly analysis unit is responsible for quantitatively analyzing abnormal vibration signals and calculating the vibration offset ratio to assess the degree to which parameters deviate from thresholds. This unit receives abnormal vibration signals transmitted from the risk assessment unit. These signals contain real-time vibration amplitude data that triggers the anomaly, recorded as the current vibration parameters. The system extracts the vibration threshold range data corresponding to the abnormal signal from the local configuration library. This range is defined by two boundary values: a lower vibration threshold and an upper vibration threshold. For example, if the lower vibration threshold is set to 0.5 mm and the upper vibration threshold to 2.0 mm for a certain construction stage, these two values ​​serve as baseline parameters in subsequent calculations.

[0055] When the anomaly analysis unit initiates the calculation process, it first determines the distance between the current vibration parameter and the threshold boundary. The system calculates the first vibration difference between the current vibration parameter and the lower vibration threshold, and the second vibration difference between the current vibration parameter and the upper vibration threshold. These two differences reflect the deviation of the current vibration amplitude from the threshold boundaries on both sides. For example, when the current vibration parameter is 2.5 mm, the first vibration difference is 2.5 mm minus 0.5 mm, which equals 2.0 mm, and the second vibration difference is 2.5 mm minus 2.0 mm, which equals 0.5 mm. The system automatically compares these two differences and selects the smaller value as the vibration boundary difference; in this example, 0.5 mm is recorded as the effective boundary difference. This selection logic ensures that the system always focuses on the closest threshold boundary, avoiding overestimation of the offset.

[0056] The calculation of the vibration offset ratio requires the mean vibration parameter as a benchmark. The system retrieves the arithmetic mean of all vibration monitoring data from the historical database within the last 24 hours; this value reflects the typical vibration level of the construction environment. The anomaly analysis unit uses a specific formula to calculate the vibration offset ratio:

[0057]

[0058] in: Indicates the vibration offset ratio. Represents the vibration boundary difference. This represents the mean vibration parameter. For example, when the vibration boundary difference is 0.5 mm and the mean vibration parameter is 1.2 mm, the calculated vibration offset ratio is approximately 0.42. The larger this ratio, the more significant the deviation of the current vibration parameters from the normal range. During the calculation process, the system implements a dynamic update mechanism for the mean vibration parameter, recalculating the mean every 100 sets of new vibration data to ensure that the reference benchmark reflects the latest operating conditions.

[0059] The anomaly analysis unit establishes multiple data verification mechanisms during calculations. After acquiring the current vibration parameters, the system automatically verifies the data validity: checking whether the values ​​are within the sensor's range (e.g., 0-10 mm), and excluding abnormally abrupt data (marked as suspicious when the change between adjacent sampling points exceeds 50%). When loading vibration threshold range data, the system verifies the construction stage identifier to prevent incorrect calls to threshold parameters from other stages. When calculating vibration boundary differences, the program's built-in logic determines whether the current vibration parameters actually exceed the threshold range, avoiding unnecessary calculations on normal data.

[0060] After the vibration offset ratio is generated, the system compares it with preset response thresholds for each level. A ratio less than 0.3 is marked as a slight offset, 0.3 to 0.6 as a moderate offset, and a ratio exceeding 0.6 as a severe offset. Different processing strategies are triggered for different levels: slight offsets are only logged without triggering adjustments; moderate offsets send suggested adjustment signals to the strategy correction unit; severe offsets immediately trigger system alarms and activate the emergency response protocol. All level determination results, along with the original vibration data and calculation process data, are packaged together to form a complete analysis report and transmitted to cloud storage.

[0061] The system employs special logic when processing continuous vibration anomaly signals. When the same monitoring point experiences three or more vibration anomalies within 10 minutes, the anomaly analysis unit initiates a cumulative effect assessment. This assessment weights and sums the offset proportions of multiple vibrations, with the weighting coefficients decaying over time (the most recent anomaly has a higher weight), ultimately yielding a comprehensive offset index. When this index exceeds a critical value, even if the individual offset proportion is not high, the system will still increase its response level. This design effectively identifies intermittent vibration patterns with cumulative risk.

[0062] The calculation results from the anomaly analysis unit are output through a standardized interface. Vibration offset ratios are transmitted in floating-point format, accompanied by a timestamp, sensor location code, and calculation version number. The system maintains an independent data cache for each monitoring point, storing the most recent 20 offset ratio calculations for trend analysis. When network transmission is interrupted, the calculation results are temporarily stored on local solid-state storage and automatically synchronized to the central database upon connection restoration. All intermediate data generated during the calculation process, such as vibration boundary differences and reference averages, are recorded in the debugging log but are not transmitted during normal operation to reduce network load.

[0063] The unit includes a computing resource management module. Under high system load, this module automatically adjusts the calculation frequency: a full calculation is performed every 5 seconds under normal conditions, while under high load, it switches to calculating the offset ratio of key monitoring points every 30 seconds. Calculation tasks are scheduled using a priority queue, with tasks for monitoring points currently experiencing vibration executed first, and historical data analysis tasks automatically postponed. This flexible calculation mechanism ensures that the system can maintain the real-time performance of core functions even when resources are limited.

[0064] The vibration offset ratio data ultimately serves the system's dynamic adjustment strategy. After receiving this data, the strategy correction unit, in conjunction with other monitoring parameters and system status, decides whether to adjust the data acquisition frequency or the execution intensity. The anomaly analysis unit continuously monitors the effect of the strategy correction, indirectly evaluating the strategy's effectiveness by comparing the change in the offset ratio before and after adjustment. This feedback mechanism forms a closed-loop control, enabling the system to adapt to changing conditions at the construction site. The entire calculation process is fully automated, requiring no manual intervention; all algorithm parameters are managed through configuration files, supporting remote updates and maintenance.

[0065] Example 3: In the construction safety and environmental monitoring system, the anomaly analysis unit is responsible for quantitatively evaluating the work efficiency of the adjustment execution unit. This unit continuously monitors the operating status of the adjustment execution unit, records the start and end times of each adjustment operation, and evaluates the timeliness of the adjustment process through precise time difference calculation. The system adopts a high-precision clock synchronization mechanism to ensure that the acquisition error of all timestamps is controlled within the millisecond range, providing reliable time reference data for subsequent analysis.

[0066] The calculation of the regulation efficiency value is based on strict time data management. When the anomaly analysis unit receives the regulation task start signal from the regulation execution unit, it immediately records the current system time as the regulation start time. When the adjustment execution unit sends an adjustment completion confirmation signal, the system records the timestamp again as the adjustment end time. These two time points are transmitted through a dedicated time data channel to avoid time errors caused by network latency. The system calculates the time difference ΔT using the following formula:

[0067]

[0068] in: This indicates the adjustment efficiency value. This represents the preset adjustment time reference. This is the actual adjustment time ΔT. (Time base value) Different adjustment timeframes are set for different types of adjustment tasks; for example, a vibration adjustment benchmark is set to 20 seconds, and a settlement adjustment benchmark is set to 40 seconds. These benchmark values ​​are stored in the system configuration database and can be dynamically adjusted according to the construction stage. When the actual adjustment time ΔT is less than the benchmark value, [the system will adjust accordingly]. A value greater than 1 indicates high adjustment efficiency; when ΔT is greater than the reference value, A value less than 1 indicates low regulation efficiency.

[0069] The system employs a multi-factor verification mechanism for calculating efficiency values. When acquiring time data, it automatically checks if T_start is earlier than T_end, excluding time records with logical errors. During calculation, the program verifies that the denominator ΔT is zero to prevent division by zero errors that could cause system anomalies. The efficiency value calculation result is limited to a reasonable range of 0-2; values ​​exceeding this range are automatically corrected to boundary values ​​and an exception log is generated. All intermediate calculation data, including original timestamps and time differences, are stored in a debug cache for troubleshooting purposes.

[0070] The tiered evaluation of efficiency values ​​employs dynamic threshold management. The system maintains an efficiency reference value. This value is obtained by statistically averaging the efficiency values ​​of the most recent 100 similar adjustment tasks. Current efficiency value. and After comparison, they were divided into three levels: when >1.2 The time stamp indicates an efficient state; when 0.8 ≤ ≤1.2 This is the normal state; when <0.8 When the condition is deemed inefficient, it is automatically classified as such. This dynamic classification method can automatically adapt to efficiency fluctuations under different construction environments, avoiding misjudgments caused by fixed thresholds.

[0071] After receiving the regulation efficiency value data, the strategy correction unit performs corresponding system parameter optimization. For regulation tasks that remain in an inefficient state, the system gradually increases the regulation intensity coefficient of the regulation execution unit. This coefficient initially has a value of 1.0, increasing by 0.1 each time an inefficient state is detected, up to a maximum of 2.0. The regulation intensity coefficient directly affects the output amplitude of the regulation command; a larger coefficient means a stronger regulation action will be generated for the same parameter deviation. Simultaneously, for regulation tasks that exhibit an efficient state three times consecutively, the system appropriately reduces the regulation intensity coefficient (decreasing it by 0.05 each time) to avoid resource waste caused by over-regulation.

[0072] The efficiency data is also used to optimize system resource allocation. The anomaly analysis unit calculates the average efficiency of each monitoring point and establishes an efficiency distribution heatmap. For monitoring points that are consistently in low-efficiency areas, the system automatically increases their data acquisition frequency to detect parameter change trends in advance; for monitoring points in high-efficiency areas, the sampling frequency is appropriately reduced to conserve system resources. This dynamic resource allocation based on efficiency data allows the system to concentrate limited computing power and communication bandwidth on the most needed monitoring areas.

[0073] The system implements special response strategies when dealing with sudden efficiency drops. If the regulation efficiency of a monitoring point drops by more than 30% within a short period (e.g., 10 minutes), the anomaly analysis unit immediately initiates an emergency diagnostic process. This process first checks the quality of sensor data to rule out false regulation requests caused by measurement errors; secondly, it checks the status of the actuators to confirm whether there are mechanical faults; and finally, it analyzes the communication link to investigate data transmission delays. The diagnostic results generate a detailed report to guide on-site maintenance personnel in carrying out targeted repairs.

[0074] Historical efficiency data is used for long-term performance analysis. The system generates efficiency trend reports periodically (e.g., weekly), displaying efficiency change curves for each monitoring point. In the report, monitoring points with continuous efficiency improvement are marked in green, those with fluctuating efficiency in yellow, and those with continuous efficiency deterioration in red. These visual analyses help managers understand the overall system performance and identify potential problem areas. All historical efficiency data is stored in a compressed format, supporting rapid retrieval and comparative analysis.

[0075] The anomaly analysis unit employs a layered processing architecture. The bottom-level real-time computing module is responsible for calculating basic efficiency values, the middle layer performs efficiency grading assessments, and the upper layer executes system optimization decisions. This architecture design ensures a reasonable distribution of computational load, guaranteeing that the system maintains its responsiveness even under high-concurrency adjustment tasks. Each processing layer has an independent anomaly handling mechanism, enabling rapid isolation of problems when a failure occurs in one layer, without affecting the normal operation of other layers.

[0076] The regulation efficiency monitoring system achieves data interoperability with other intelligent devices on the construction site. For example, when the system detects that the regulation efficiency in a certain area is consistently low, it can automatically dispatch nearby inspection robots to check; or send an alarm to the intelligent safety helmet system to alert relevant personnel to the risks in that area. This cross-system collaboration capability greatly improves the overall monitoring efficiency, forming a three-dimensional safety protection network.

[0077] The system implements strict encryption protection for the transmission of efficiency adjustment data. All efficiency values ​​are digitally signed before transmission, and the receiving end verifies the validity of the signature before processing the data. The historical efficiency database adopts a partitioned and hierarchical storage strategy, with recently accessed high-frequency data stored on high-speed storage devices and long-term archived data transferred to low-cost storage media. Data backup adopts an incremental backup strategy and is automatically synchronized to the cloud disaster recovery center daily.

[0078] The efficiency analysis algorithm supports remote updates and maintenance. System administrators can upload new efficiency calculation algorithms through a secure channel, completing algorithm upgrades without interrupting monitoring services. Each algorithm update retains historical versions, allowing for quick rollback to older versions when necessary. This design ensures the system can continuously optimize efficiency evaluation methods to adapt to ever-changing construction environment requirements.

[0079] Example 4: See Figure 4 In a construction safety and environmental monitoring system, the model prediction unit analyzes historical data sequences to predict potential risks, providing support for optimized decision-making. This unit acquires continuous monitoring data from the data acquisition unit, including historical records of parameters such as vibration amplitude and settlement, forming a time-series dataset. The system stores monitoring data at 5-minute intervals, with each data point containing a timestamp, parameter value, and sensor location information. The length of the historical data sequence is dynamically adjusted according to prediction needs; short-term predictions use data from the most recent 4 hours, while long-term predictions utilize data from the most recent 7 days.

[0080] The deformation feature extraction module performs frequency domain analysis on historical parameter sequences to identify structural deformation characteristic bands. The module first preprocesses the time series data, including outlier removal, missing data imputation, and smoothing filtering. The processed data is then converted into a frequency domain signal using a Fast Fourier Transform (FFT), decomposing it into different frequency components. The system focuses on the low-frequency band of 0-10 Hz, which reflects the slow deformation trend of the structure. The frequency domain analysis results are stored as a spectrum, with the energy intensity of each band marked. For example, spectrum analysis of settlement data from a foundation pit monitoring point shows a significant energy peak at 0.5 Hz, indicating periodic settlement at that location.

[0081] The risk fusion module integrates frequency domain features with real-time monitoring data to generate a quantitative risk value. The module obtains low-frequency band energy values ​​from the deformation feature extraction module and simultaneously receives real-time vibration amplitude data transmitted from the data acquisition unit. The system uses a weighted fusion algorithm to calculate the predicted collapse risk value, with the low-frequency band weight set at 70% and the real-time vibration weight at 30%. The calculation results are output on a 0-100 scale, with higher scores indicating greater risk. The risk value is updated every 30 seconds and pushed to the optimization decision-making unit in real time.

[0082] The optimization decision-making unit generates a control plan by combining predicted signals and real-time parameters. The unit receives structural deformation prediction signals and collapse risk prediction signals from the model prediction unit, while simultaneously acquiring the current operating status data of the control execution unit. The system establishes a decision matrix, combining and analyzing the predicted risk level with the degree of deviation of real-time parameters, and outputs a targeted optimized control plan. The plan includes suggested control intensity, duration, and priority. For example, when the predicted risk value for a certain area exceeds 70 points and the real-time settlement reaches 90% of the threshold, the system generates a high-strength support control plan, suggesting an immediate increase in hydraulic support pressure of 15%, continuing control until the risk value drops below 50 points.

[0083] The system establishes a quality control mechanism during data processing. All input data must pass validity verification, including range checks (whether the vibration amplitude is within the reasonable range of 0-10 mm), continuity checks (whether the rate of change between adjacent sampling points is abnormal), and consistency checks (whether the differences between data from multiple sensors at the same location are too large). Invalid data is automatically marked and excluded from the analysis process, and a data quality alarm is triggered simultaneously. Historical data storage adopts a hierarchical structure: raw data is stored in a high-speed cache layer, processed feature data is stored in the analysis database, and long-term archived data is compressed and stored.

[0084] The model prediction unit supports switching between prediction modes for multiple scenarios. The system presets three prediction modes: the standard mode uses standard parameter weights and is suitable for most construction stages; the sensitive mode increases the weight of low-frequency bands to 85% and is used in areas with complex geological conditions; the robust mode reduces the weight of real-time vibration to 20% and is suitable for environments with significant mechanical vibration interference. Mode switching can be done manually through the management interface or automatically adjusted by the system based on environmental assessment results. The prediction performance of each mode is recorded in the operation log for subsequent analysis and reference (see Table 1).

[0085] Table 1: The structural deformation prediction data of a certain foundation pit monitoring point for 6 consecutive hours are as follows.

[0086] Timestamp Low-frequency band energy Real-time vibration (mm) Predicted risk value Adjustment suggestions 2023-08-1808:00 0.45 1.2 35 Maintain current support pressure 2023-08-1809:00 0.68 1.5 52 Fine-tune support pressure (+5%) 2023-08-1810:00 0.82 2.1 73 Increase support pressure (+12%) 2023-08-1811:00 0.91 2.3 85 Emergency increase in support pressure (+20%) 2023-08-1812:00 0.75 1.8 65 Maintain current pressure 2023-08-1813:00 0.60 1.4 48 Gradually reduce stress (-8%)

[0087] The system implements closed-loop control of prediction and adjustment. The model prediction unit re-evaluates the prediction results every 5 minutes, and the optimization decision unit adjusts the adjustment scheme according to the latest prediction. The adjustment execution unit provides feedback on the actual adjustment effect, forming a continuously optimizing control loop. For example, when the predicted risk value is consistently higher than expected, the system automatically increases the weight of low-frequency bands to enhance sensitivity to slow deformation; when the real-time adjustment effect is better than expected, the adjustment intensity is appropriately reduced to avoid over-intervention.

[0088] The predictive model parameters support dynamic learning and updates. The system records each prediction result and actual structural change data, and trains the model parameters periodically (e.g., weekly). The training process uses a sliding window mechanism, giving higher weights to the latest data, enabling the model to adapt to changes in construction conditions. Model version management adopts a canary release strategy, where new models are first tested at select monitoring points to verify their effectiveness before being rolled out nationwide. Each model update retains a complete version record and supports quick rollback operations.

[0089] The risk visualization interface displays the predictive analysis results in real time. The large-screen display system in the construction site monitoring center integrates the predictive data and displays the risk level of each area in the form of a heat map. High-risk areas (predicted value > 70) are displayed in red, medium-risk areas (40-70) are displayed in yellow, and low-risk areas (< 40) are displayed in green. The interface also provides a historical prediction curve comparison function, allowing users to view the risk change trend of any monitoring point. All visualized data supports touch interaction, facilitating on-site personnel to gain a deeper understanding of the details.

[0090] The system establishes a multi-level early warning and response mechanism. Different responses are triggered based on the predicted risk value: a foreman-level alert is sent when the risk value is 30-50; a project-level alert is triggered when the risk value is 50-70; a company-level emergency response is initiated when the risk value is 70-90; and the regulatory department is automatically notified when the risk value exceeds 90. Each level of early warning has corresponding handling procedures and personnel scheduling plans to ensure that response measures are upgraded synchronously when the risk escalates. Early warning information is sent simultaneously through three channels: audible and visual alarms, SMS notifications, and mobile app push notifications, ensuring timely information delivery.

[0091] The model prediction unit achieves data interoperability with third-party geological monitoring systems. The system regularly imports external information such as ground-penetrating radar scan data and groundwater level monitoring data to enrich the input dimensions of the prediction model. The data interface adopts a standardized protocol, supporting automatic matching of timestamps and spatial location information. After normalization, the external data is fused and analyzed with the system's internal monitoring data to improve the accuracy of the prediction results. For example, when rising groundwater levels coincide with low-frequency deformation characteristics, the system will pay special attention to piping risk and adjust the prediction model parameters accordingly.

[0092] The prediction system's operational status is monitored in real time. A dedicated daemon continuously checks key indicators such as resource utilization of the model prediction unit, data processing latency, and prediction result volatility. When anomalies are detected, such as CPU utilization consistently exceeding 80% or prediction results remaining unchanged for 10 consecutive times, the system automatically restarts the prediction service and sends an operational alert. All abnormal events are recorded in the system health log, forming a complete operation and maintenance profile. System health reports are generated periodically, summarizing comprehensive indicators such as hardware status, software performance, and prediction accuracy.

[0093] The system adopts a modular architecture, allowing model prediction units to be deployed and run independently. The core prediction algorithm is encapsulated as a standardized service, interacting with other systems through well-defined interfaces. This design supports horizontal scalability, enabling the deployment of multiple prediction nodes to share the computational load in large construction projects. A service discovery mechanism ensures that newly added prediction nodes automatically register with the system network, achieving elastic resource scheduling. Containerized algorithm deployment allows the system to be quickly migrated to different hardware environments, ensuring service continuity.

[0094] Data security measures are implemented throughout the entire prediction process. All monitoring data is encrypted using TLS during transmission and AES-256 encrypted during storage. Access control employs role-based access management, requiring high-level authorization for training and updating the prediction model. Complete audit logs are maintained for all system operations, recording critical actions such as data access, model modifications, and parameter adjustments. Regular security vulnerability scans and penetration tests are conducted to ensure the system's continuous effectiveness. Data backup employs a 3-2-1 strategy, with three copies stored on two different media, one of which is stored off-site.

[0095] The model prediction unit is designed with full consideration of the special environment of the construction site. The hardware meets IP65 protection standards, adapting to high dust and high humidity environments. The wireless communication module supports 4G / 5G and LoRa multi-mode transmission, ensuring signal coverage in complex building environments. The prediction algorithm has been optimized for efficient operation even on edge devices with limited computing resources. The system supports network interruption recovery; data is temporarily stored locally during network interruptions and automatically synchronized upon network recovery.

[0096] Example 5: In a construction safety and environmental monitoring system, the cloud-based strategy engine continuously receives anomaly type signals transmitted by the risk assessment unit. The signal content includes anomaly category identifiers, occurrence time, monitoring point location, and parameter exceedance values. After parsing the signals, the engine generates a tiered verification strategy based on a preset rule base. The rule base defines a three-level verification mechanism: the first level verifies basic equipment identity information, the second level extends to equipment operation history, and the third level covers the equipment's real-time status and its compatibility with the environment. The strategy generation process uses dynamic templates, automatically adjusting the weight of verification items for different anomaly types. For example, when a settlement anomaly signal triggers the verification strategy, the weight of the third-level verification increases to 70%; for a dust anomaly signal, the weight of the third level decreases to 30%. The generated strategy file is transmitted to the permission verification unit via an encrypted channel.

[0097] After the permission verification unit loads the hierarchical verification strategy, it initiates the security level verification process for the construction equipment. The unit first establishes a connection with the equipment registration database at the construction site to obtain the unique identifier of the target equipment. The first-level verification checks whether the equipment identifier exists in the system whitelist, and also checks the validity period and issuing authority of the equipment certificate. Upon successful verification, basic operating permissions are granted, allowing the equipment to execute routine operation commands, such as starting a conveyor belt or turning on the lighting system. If the equipment fails the first-level verification, the system immediately freezes its control interface and sends an identity anomaly alarm.

[0098] The second level of verification targets devices that passed the first level of verification, retrieving their operation logs. The system analyzes the completeness and continuity of the logs, checking for unauthorized operation periods or interruptions in the command sequence. The verification algorithm focuses on detecting the compliance of high-frequency operation commands. For example, if an excavator performs more than 20 bucket actions consecutively within 10 minutes, it needs to verify whether each action is matched with slope sensor data. Devices that pass the verification are granted high-risk operation permissions, unlocking critical functions such as heavy machinery operation and deep foundation pit operation. High-risk operation permissions come with operational restrictions, such as automatically locking the rotation function of tower cranes when the wind speed exceeds level 6.

[0099] The third level of verification integrates multi-source real-time data to assess the compatibility between equipment status and the construction environment. The system obtains equipment coordinates via a GPS positioning module and compares them with the safe zone defined by the electronic fence system. Simultaneously, it accesses equipment sensor data, such as inclinometer readings and hydraulic pressure values, to verify whether the equipment is in a stable operating state. Environmental compatibility calculations consider wind speed and rainfall data transmitted from weather stations to determine whether the current environment allows the equipment to continue operating. Equipment that passes all three levels of verification is added to a safe equipment list, which is stored in system memory using a hash table structure and simultaneously backed up to the cloud. Equipment on the safe equipment list enjoys the highest operating privileges, can execute all preset functions, and is not restricted by routine operations.

[0100] The permission allocation results are fed back to the construction equipment via digital certificates. The system generates a dynamic access token for each piece of equipment. The token's validity period is synchronized with the construction period and automatically expires at the end of each day's work. After receiving the token, the equipment control system unlocks the corresponding permission level's set of operation instructions. The permission status is displayed in real time on the equipment operation interface and is distinguished by color coding: green indicates basic permissions, yellow indicates high-risk permissions, and red indicates safe list permissions. Operators can intuitively understand the current equipment permission scope through interface icons.

[0101] The access control unit implements a proactive monitoring mechanism. For authorized equipment, the system re-performs access control verification every 5 minutes, dynamically adjusting the access level. When equipment moves out of the safe zone or sensors detect an abnormal state, the system automatically downgrades access and triggers an alarm. For example, if a concrete pump truck experiences a sudden 10% drop in outrigger pressure during operation, the system immediately downgrades its access from the safe list level to the basic level and locks the pumping function. Simultaneously, it sends an explanation of the downgrade to the equipment operator, prompting them to inspect the equipment.

[0102] The verification failure handling process involves multiple response layers. Upon initial verification failure, the system automatically initiates a second verification request to rule out the impact of temporary communication failures. Two consecutive failures freeze device operation permissions and generate a detailed failure report. The report includes the failure stage, error code, and suggested handling measures, and is simultaneously pushed to responsible personnel via SMS and the monitoring center's large screen. If on-site maintenance is required, a new verification request must be initiated. The system records the time, personnel involved, and results of each maintenance operation.

[0103] The list of security devices is managed dynamically. Devices on the list undergo a heartbeat check every 30 minutes to confirm their online status and data reporting continuity. Devices that fail to respond to more than three heartbeat checks are automatically removed from the list and must pass three levels of verification before being added back. The system periodically scans the operation records of devices on the list to detect any unusual command sequences. When abnormal operation patterns are detected, a special audit process is initiated and device permissions are temporarily restricted.

[0104] Access control data is shared with other monitoring systems. When a tower crane obtains high-risk operation permission, the system automatically notifies the personnel positioning system in adjacent areas, prompting them to avoid the lifting route. After concrete batching plant equipment is added to the safety list, the material management system simultaneously unlocks the high-performance mixing mode. This cross-system collaborative mechanism optimizes the overall efficiency of the construction process.

[0105] The permission verification unit adopts a distributed deployment architecture. The core verification service runs in the cloud, while lightweight verification modules are deployed on edge nodes. In the event of a network outage, the edge modules perform basic permission verification based on locally cached data to ensure the basic operational capabilities of the devices. Once the network is restored, verification records are automatically synchronized to the cloud to ensure data integrity. All permission change operations are recorded in the blockchain evidence storage system, forming an immutable operation audit trail.

[0106] The system maintains version management for permission policies. A new version number is generated each time the policy rules are updated, and old versions are retained for 180 days for auditing and traceability. The permission allocation logic supports canary releases; new policies are first tested on 10% of devices, and then fully rolled out after confirming there are no conflicts. The policy rollback mechanism can restore the system to the previous stable version within 30 seconds, minimizing update risks.

[0107] Device control commands implement a two-factor authentication mechanism. Even if the device has already obtained operating permissions, a real-time verification code must still be submitted before executing critical commands. The verification code is dynamically generated by the permission verification unit, is valid for 10 seconds, and can only be used once. This mechanism prevents permission tokens from being illegally copied and abused, improving the overall security of the system. The command execution log records the operation time, permission token, and verification code information in detail, supporting full lifecycle traceability.

[0108] Resource allocation for the permission verification unit is prioritized. When concurrent verification requests exceed system load, verification of devices in areas with abnormal settlement is prioritized. High-priority tasks are processed through dedicated computing channels to ensure response latency does not exceed 200 milliseconds. The system resource monitoring module adjusts the load of computing nodes in real time and automatically scales up to handle peak requests. All resource scheduling records are included in the system health report for optimizing subsequent resource allocation strategies.

[0109] The device permission data visualization interface supports multi-dimensional analysis. The monitoring center console can filter and display the distribution of devices with specific permission levels, or view the permission change history of a single device. The timeline mode displays the entire process from device registration to being added to the security list, helping administrators assess device reliability. The interface supports map location of devices with abnormal permissions, allowing for quick navigation to the area where the problematic device is located.

[0110] The system establishes a disaster recovery backup system for permission verification. In the event of a failure of the primary cloud node, the backup node takes over the service within 15 seconds, with verification policies and permission data switching implemented in seconds. All verification requests are automatically queued and retried during the disaster recovery switchover to prevent data loss. Disaster recovery drills are conducted monthly to test the backup system's takeover efficiency and data consistency.

[0111] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0112] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A construction safety and environmental monitoring system, characterized in that, include: The data acquisition unit acquires environmental and structural parameters of the construction site at fixed time intervals; The risk assessment unit performs threshold judgments on the environmental and structural parameters and generates anomaly type signals. The adjustment execution unit adjusts the execution parameters according to the type and priority of the abnormal signal. The anomaly analysis unit calculates the parameter offset ratio and regulation efficiency value based on the anomaly type signal and regulation process data; The strategy correction unit adjusts the parameter acquisition frequency of the data acquisition unit according to the parameter offset ratio, and optimizes the adjustment intensity of the adjustment execution unit based on the adjustment efficiency value; The failure early warning unit counts the total number of abnormal signals from the anomaly analysis unit and generates a construction failure signal by combining the construction duration. The risk assessment unit compares the environmental parameters and structural parameters with a preset safety parameter range; If the vibration amplitude exceeds the preset vibration threshold range, an abnormal vibration signal will be generated. If the structural settlement exceeds the preset settlement threshold range, a settlement anomaly signal will be generated. If the dust concentration exceeds the preset dust threshold, a dust abnormality signal will be generated. The risk determination unit triggers the adjustment execution unit first when generating the settlement anomaly signal; The anomaly analysis unit identifies the real-time vibration amplitude that triggers the vibration anomaly signal and records it as the current vibration parameter; The two endpoints of the vibration threshold range are respectively denoted as the lower vibration threshold and the upper vibration threshold. Calculate the first vibration difference between the current vibration parameter and the lower vibration threshold, and the second vibration difference between the current vibration parameter and the upper vibration threshold; The minimum value between the first vibration difference and the second vibration difference is denoted as the vibration boundary difference; The vibration offset ratio is generated by the proportional relationship between the vibration boundary difference and the mean vibration parameter. The anomaly analysis unit records the time difference between the start of adjustment and the completion of adjustment by the adjustment execution unit, and generates an adjustment efficiency value by comparing it with a preset adjustment time benchmark. When the vibration offset ratio exceeds a preset offset threshold, the strategy correction unit shortens the fixed time interval of the data acquisition unit proportionally. And when the adjustment efficiency value is lower than the efficiency benchmark, the adjustment intensity of the vibration damping equipment power adjustment command is increased.

2. The construction safety and environmental monitoring system according to claim 1, characterized in that, The adjustment process of the adjustment execution unit for the abnormal vibration signal includes: selecting the median value of the vibration threshold range as the standard vibration value, calculating the vibration difference between the real-time vibration amplitude and the standard vibration value, and generating a power adjustment command for the vibration reduction equipment; The adjustment process for the settlement anomaly signal includes: selecting the median value of the settlement threshold range as the standard settlement value, calculating the settlement difference between the real-time settlement and the standard settlement value, and generating a pressure adjustment command for the support structure.

3. The construction safety and environmental monitoring system according to claim 1, characterized in that, It also includes a model prediction unit and an optimization decision unit; The model prediction unit receives the historical parameter sequence from the data acquisition unit and outputs structural deformation prediction signal and collapse risk prediction signal through the time series prediction model. The optimization decision-making unit integrates the structural deformation prediction signal and real-time structural parameters to generate an optimized adjustment scheme, which is then sent to the adjustment execution unit.

4. The construction safety and environmental monitoring system according to claim 3, characterized in that, The model prediction unit includes a deformation feature extraction module and a risk fusion module; The deformation feature extraction module performs frequency domain feature decomposition on the historical parameter sequence to extract structural deformation feature bands. The risk fusion module weights and processes the structural deformation characteristic bands and real-time vibration amplitude data to output the quantified risk value of the collapse risk prediction signal.

5. The construction safety and environmental monitoring system according to claim 1, characterized in that, It also includes a cloud-based policy engine and a permission verification unit; the cloud-based policy engine generates a hierarchical verification policy based on the anomaly type signal; The permission verification unit loads the hierarchical verification strategy to perform equipment security level verification, and assigns operation permission levels to the construction equipment based on the verification results.

6. The construction safety and environmental monitoring system according to claim 5, characterized in that, When the construction equipment passes the first-level hierarchical verification strategy, the permission verification unit grants basic operation permissions. When the second-level hierarchical verification strategy is passed, high-risk operation permissions are granted; When the third-level hierarchical verification strategy is passed, the construction equipment is added to the list of safety equipment.

7. A method for analyzing construction safety and environmental monitoring data, characterized in that, It includes all modules and method flows of the construction safety and environmental monitoring system as described in any one of claims 1 to 6.

Images