A construction site dust monitoring and prediction method and system based on data analysis

Abstract

The present invention relates to the field of environmental monitoring technology, specifically to a method and system for monitoring and predicting dust at a construction site based on data analysis, comprising the following steps: collecting data on wind speed, humidity, temperature and dust particle concentration at the construction site, eliminating outliers for each data item, correcting measurement deviations using linear calibration, and generating a preliminary screening data set. In the present invention, by dynamically calculating the slope and curvature in the environmental data of the construction site, the generation and diffusion trend of dust can be accurately predicted, and by comparing with historical data of the same period, the prediction accuracy of future dust events can be improved, so that managers can formulate effective dust control strategies in advance based on the prediction results, and by setting automatic alerts, an early warning is immediately issued when the dust concentration reaches a preset threshold, emergency intervention measures are initiated, and the response speed and processing capacity of the construction site to environmental changes are improved, ensuring the environmental friendliness of construction activities and the safety of public health.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and in particular to a method and system for monitoring and predicting dust at a construction site based on data analysis. Background Art

[0002] The field of environmental monitoring technology involves the use of various devices and methods to continuously track and evaluate specific factors in the environment, such as air and water quality, noise, temperature, humidity, etc. By analyzing real-time collected data, it is possible to promptly detect environmental changes and predict potential environmental problems, thereby providing necessary early warning information. It is widely used in many fields such as urban planning, industrial production, agriculture, transportation and public safety, especially in dealing with pollution prevention and control and climate change issues.

[0003] Among them, the construction site dust monitoring and prediction method based on data analysis focuses on using advanced data processing technology to predict and monitor dust emissions in construction sites. Its main purpose is to accurately predict the generation and diffusion trends of dust by analyzing the collected construction site environmental data, thereby helping construction managers to formulate effective dust control measures, which can significantly reduce the impact of construction activities on the surrounding environment, ensure public health and safety, and assist construction projects in complying with relevant environmental protection laws and regulations to optimize operating processes and environmental management strategies.

[0004] Although existing technologies can continuously monitor environmental factors, they often lack the ability to conduct advanced analysis of environmental data, which makes it impossible to effectively foresee and prevent the occurrence of environmental problems. As a result, environmental management strategies rely more on data that has already occurred rather than predicted data, and the result is often a response to problems rather than prevention. For example, in the monitoring of dust at construction sites, existing technologies often cannot identify high-incidence periods and areas of dust in advance due to the lack of dynamic prediction capabilities, resulting in delayed intervention measures and the inability to fundamentally reduce or avoid the risks of environmental and public health problems. The reactive processing mode not only increases the difficulty of environmental governance, but may also exacerbate the problem under adverse weather or other uncontrollable factors, causing long-term impacts on the health and safety of surrounding communities. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a construction site dust monitoring and prediction method based on data analysis.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a construction site dust monitoring and prediction method based on data analysis, comprising the following steps:

[0007] S1: Collect wind speed, humidity, temperature and dust particle concentration data of the construction site, remove outliers for each data item, use linear calibration to correct the measurement deviation, and generate a preliminary screening data set;

[0008] S2: Based on the preliminary screening data set, time series analysis is performed to calculate the slope and curvature of the data points respectively, and key time points are identified according to the slope and curvature calculation results, and the change trend is analyzed to capture the key dynamics of the dust concentration at the construction site and generate key variable extraction results;

[0009] S3: Based on the extraction results of the key variables, compare the current data with the historical data of the same period, evaluate the possibility of future dust events at the construction site, quantitatively analyze the time and location of potential dust events according to the evaluation results, predict the spatiotemporal distribution of dust, and generate future dust event prediction results;

[0010] S4: According to the prediction results of future dust events, a trigger warning threshold is set. When the dust concentration reaches the warning threshold, a warning message is automatically issued to notify the construction site manager to intervene and output an emergency management plan for dust at the construction site.

[0011] As a further embodiment of the present invention, the step of obtaining the preliminary screening data set is:

[0012] S111: Collect original wind speed, humidity, temperature and dust particle concentration data from the construction site, record the timestamp of each data point, and generate an original data set with a time stamp;

[0013] S112: performing outlier elimination on the original data set with time stamp, identifying and removing data points exceeding three standard deviations, and obtaining a data set after outlier processing;

[0014] S113: Performing linear calibration on the data set after outlier processing, using the formula:

[0015] Calculate the correction value , generate a linearly calibrated data set, where represents the data point after outlier processing, is the calibration factor, is the bias calibration value, which is used to adjust the denominator to avoid zero division errors. is the mean value of the data set;

[0016] S114: synthesizing the linearly calibrated data sets, unifying the data formats, integrating and outputting the preliminary screening data sets.

[0017] As a further solution of the present invention, the steps for calculating the slope and curvature of the data points are:

[0018] S211: Based on the preliminary screening data set, perform time series analysis, smooth the data, and generate a smoothed data set;

[0019] S212: According to the smoothed data set, using the formula:

[0020] Calculate the slope for each data point , output slope data set, where Indicates the difference in dust concentration between consecutive data points. is the time interval;

[0021] S213: Based on the slope data set, using the formula,

[0022] Calculating curvature , evaluate the rate of change of the slope and generate a curvature data set where, It is the difference value of the slope, which is used to reveal the second-order characteristics of data changes.

[0023] As a further solution of the present invention, the step of obtaining the key variable extraction result is:

[0024] S221: Using the curvature data set, identifying the time point when the curvature value changes significantly, and setting the threshold parameter Filter the curvature data, all the data greater than The time points corresponding to the curvature values ​​are marked as key time points, and a marked key time point list is generated;

[0025] S222: Based on the marked key time point list and in combination with the slope data set, analyzing the slope change of the previous and next data points, determining the change trend of each time point, and obtaining a trend description record of the key time point;

[0026] S223: Integrate the trend description records of the key time points and the marked key time point list, record the date, time and corresponding change trend of each key time point, and generate key variable extraction results.

[0027] As a further solution of the present invention, the steps for evaluating the possibility of future dust events at the construction site are:

[0028] S311: Based on the key variable extraction results, combined with the real-time monitoring data of the current construction site, a comprehensive data set is integrated and output;

[0029] S312: Perform time series analysis on the comprehensive data set, compare the difference between current data and historical data of the same period, analyze the deviation index of key variables, and obtain a data difference description record;

[0030] S313: Based on the data difference description record, predict the possibility of future dust events, using the formula,

[0031] Calculate the probability of dust events occurring , and obtain the dust event possibility assessment results, among which, Represents the time decay factor, which is used to adjust the impact of time span on probability. Representative The weight of each key variable reflects the relative importance of the variable in the occurrence of dust events. Representative The deviation index of a key variable indicates the degree of deviation between the current value and the historical value of the same period. Represents the number of key variables.

[0032] As a further solution of the present invention, the step of obtaining the prediction result of future dust events is:

[0033] S321: Based on the dust event possibility assessment result, using the environmental data monitored at the construction site and historical data from the same period, analyzing the risk level of the location, determining the risk area and the time of occurrence of the dust event, and generating a spatiotemporal risk map;

[0034] S322: Based on the spatiotemporal risk map, using a deep learning model to analyze data, combining historical dust occurrence records and corresponding meteorological conditions, predicting the spatiotemporal distribution of dust in the future, and obtaining a dust distribution prediction result;

[0035] S323: Combining the dust distribution prediction results with the current construction activity data, a multivariate impact analysis is performed using the formula:

[0036] Calculate dust forecast intensity , and obtain the prediction results of future dust events, among which, represents soil moisture, represents wind speed, represents the intensity of construction activities, , , The weight coefficients correspond to soil moisture, wind speed and intensity of construction activities respectively.

[0037] As a further solution of the present invention, the steps for obtaining the emergency management plan for dust at the construction site are:

[0038] S411: Analyze the prediction results of future dust events, extract the extreme values ​​and fluctuation ranges of dust concentration in a future period of time, and generate future dust concentration estimation data;

[0039] S412: According to the future dust concentration estimation data, a trigger warning threshold is set, using the formula:

[0040] Calculate and output the trigger warning threshold ,in, represents the maximum estimated dust concentration, Represents the average volatility concentration of historical data for the same period, is the safety factor;

[0041] S413: Based on the trigger warning threshold, when the monitoring data shows that the dust concentration reaches and exceeds the warning threshold, the early warning mechanism is automatically activated, an alarm is issued, and the construction site manager is notified to intervene and establish an emergency management plan for dust at the construction site.

[0042] A construction site dust monitoring and prediction system based on data analysis, comprising:

[0043] The data preprocessing module collects the original wind speed, humidity, temperature and dust particle concentration data of the construction site, records the timestamp of each data point, removes outliers, performs linear calibration, unifies the data format, and integrates and outputs the preliminary screening data set;

[0044] The data feature analysis module performs time series analysis based on the preliminary screening data set, calculates the slope and curvature of each data point, evaluates the rate of change of the slope, reveals the second-order characteristics of the data change, and outputs a slope data set and a curvature data set;

[0045] The key point marking module uses the curvature data set to identify and mark key time points, combines the slope data set to analyze the slope changes of the previous and next data points, determines the change trend of each time point, and integrates to generate key variable extraction results;

[0046] The dust probability prediction module performs time series analysis based on the extraction results of the key variables and the real-time monitoring data of the current construction site, analyzes the deviation index of the key variables, calculates the probability of dust events, predicts the possibility of future dust events, and obtains the possibility assessment results of dust events;

[0047] The dust spatial analysis module analyzes the risk level of the location based on the dust event possibility assessment result, determines the risk area and the time of dust event occurrence, predicts the spatiotemporal distribution of dust in the future, calculates the dust prediction intensity in combination with the current construction activity data, and obtains the prediction result of future dust events;

[0048] The emergency response module analyzes the prediction results of future dust events, extracts the extreme values ​​and fluctuation ranges of dust concentrations in the future, calculates and sets the trigger warning thresholds, and automatically activates the early warning mechanism when the monitoring data shows that the dust concentration reaches and exceeds the warning threshold, notifying the construction site manager to intervene and establish an emergency dust management plan for the construction site.

[0049] Compared with the prior art, the advantages and positive effects of the present invention are:

[0050] In the present invention, by dynamically calculating the slope and curvature in the environmental data of the construction site, the generation and diffusion trend of dust can be accurately predicted. At the same time, by comparing with the historical data of the same period, the prediction accuracy of future dust events is improved, so that managers can formulate effective dust control strategies in advance according to the prediction results. By setting automatic alarms, an early warning is immediately issued when the dust concentration reaches a preset threshold, thereby initiating emergency intervention measures, improving the response speed and processing capacity of the construction site to environmental changes, optimizing the environmental management strategy, and ensuring the environmental friendliness of construction activities and the safety of public health. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a main step flow chart of the present invention;

[0052] Figure 2 A flowchart for obtaining a preliminary screening data set for the present invention;

[0053] Figure 3 is a flow chart for calculating the slope and curvature of a data point of the present invention;

[0054] Figure 4 The flowchart for obtaining the key variable extraction results of the present invention;

[0055] Figure 5 A flow chart for evaluating the possibility of future dust events occurring at a construction site of the present invention;

[0056] Figure 6 A flowchart for obtaining the prediction results of future dust events of the present invention;

[0057] Figure 7 The present invention is a flowchart for obtaining the emergency management plan for dust at the construction site. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0060] See also Figure 1 , a construction site dust monitoring and prediction method based on data analysis, comprising the following steps:

[0061] S1: Collect wind speed, humidity, temperature and dust particle concentration data of the construction site, remove outliers for each data item, use linear calibration to correct the measurement deviation, and generate a preliminary screening data set;

[0062] S2: Based on the preliminary screening data set, time series analysis is performed to calculate the slope and curvature of the data points respectively. According to the slope and curvature calculation results, key time points are identified, the change trend is analyzed, the key dynamics of dust concentration in the construction site are captured, and key variable extraction results are generated;

[0063] S3: Based on the extraction results of key variables, compare the current data with the historical data of the same period, evaluate the possibility of future dust events at the construction site, quantitatively analyze the time and location of potential dust events based on the evaluation results, predict the spatiotemporal distribution of dust, and generate future dust event prediction results;

[0064] S4: According to the prediction results of future dust events, set the trigger warning threshold. When the dust concentration reaches the warning threshold, it will automatically issue a warning message to notify the construction site manager to intervene and output the construction site dust emergency management plan.

[0065] The preliminary screening data set includes corrected wind speed data, corrected humidity data, corrected temperature data and corrected dust particle concentration data. The key variable extraction results include slope and curvature calculation results, change trend analysis results and key time point identification records. The prediction results of future dust events include the probability of event occurrence, expected time of occurrence, expected location of occurrence and spatial distribution map. The construction site dust emergency management plan includes the warning threshold setting results, warning signal issuance records and management intervention measures.

[0066] See also Figure 2 ,The steps to obtain the preliminary screening data set are,

[0067] S111: Collect original wind speed, humidity, temperature and dust particle concentration data from the construction site, record the timestamp of each data point, and generate an original data set with a time stamp;

[0068] At the construction site, data collection involves a variety of environmental parameters, such as wind speed, humidity, temperature, and dust particle concentration. To effectively collect this data, it is first necessary to develop a detailed sensor deployment plan. Taking into account the geographical and environmental characteristics of the site, select the appropriate sensor type and installation location to avoid the shielding effect of terrain and buildings. Sensors should be installed in representative locations with minimal impact on the environment. In addition, set a suitable data collection frequency to adapt to the rapid changes in environmental conditions and capture key data changes. Ensure that each data point collected is accompanied by an accurate timestamp. The data points will be collected and recorded in the time-stamped raw data set for subsequent processing and analysis.

[0069] S112: performing outlier elimination on the original data set with time stamps, identifying and removing data points exceeding three standard deviations, and obtaining a data set after outlier processing;

[0070] For the data sets collected at the construction site, the first priority is to ensure the accuracy and availability of the data. Using the Z-score method to eliminate outliers is a key step in cleaning the data. By calculating the Z-score of each data point (that is, the deviation of its value from the mean of the entire data set divided by the standard deviation), we can identify and exclude those abnormal data points with a standard deviation greater than three times. First, we calculate the mean and standard deviation from the entire data set, and then apply statistical indicators to evaluate the Z-score of each point. The identified abnormal data will be removed from the analysis data set to ensure the accuracy of subsequent data analysis. Through this process, the generated outlier-processed data set is not only of high quality, but also more suitable for further data analysis and model training.

[0071] S113: Perform linear calibration on the data set after outlier processing, using the formula,

[0072] Calculate the correction value , generate a linearly calibrated data set, where represents the data point after outlier processing, is the calibration factor, is the bias calibration value, which is used to adjust the denominator to avoid zero division errors. is the mean value of the data set;

[0073] set up is the calibration coefficient, which represents the sensitivity of data adjustment and is adjusted through feedback during field test calibration. It is a deviation calibration value, ensuring that the denominator is not zero while providing sufficient adjustment flexibility.

[0074] Collect a specific data point , corresponding to the average value of the data set , substitute into the formula and calculate:

[0075] The results show that the corrected reading will be approximately the original data point value plus 1.644 (i.e. ), which means that by adjusting the formula, it is possible to compensate for deviations in the original readings, thus making the data closer to the actual environmental conditions.

[0076] S114: synthesize the linearly calibrated datasets, unify the data format, and integrate and output the preliminary screening datasets;

[0077] The purpose of integrating and unifying the linearly calibrated data sets is to ensure that all data can be correctly used for subsequent analysis and applications, including standardization of data types to ensure that all readings follow the same standard, such as converting temperature from Fahrenheit to Celsius and wind speed from miles per hour to meters per second. In addition, the integration of data also requires time synchronization of the data sets to ensure that data from different sensors have the same time tags, which is especially important for time series analysis. The final preliminary screening data set will provide an accurate and consistent basis for further data analysis and model building.

[0078] See also Figure 3 , the calculation steps of the slope and curvature of the data point are,

[0079] S211: Based on the preliminary screening data set, perform time series analysis, smooth the data, and generate a smoothed data set;

[0080] Starting from the existing preliminary screening data set, the moving average method is first used to smooth the data to reduce data fluctuations caused by random changes and instantaneous noise. In specific operations, for example, a five-point moving average is selected, and the new value of each data point will be calculated as the average of itself and the two data points before and after it, which effectively smooths short-term extreme fluctuations while retaining the integrity of long-term trends, making the data more stable and providing a more reliable basis for the subsequent slope and curvature calculations. The smoothed data set will show a clearer trend line, providing a key pre-processing step for subsequent time series analysis, ensuring that the calculation of slope and curvature can more accurately reflect the actual data changes.

[0081] S212: Based on the smoothed data set, the formula is used:

[0082] Calculate the slope for each data point , output slope data set, where Indicates the difference in dust concentration between consecutive data points. is the time interval;

[0083] At consecutive time points t1 and t2 ( ), the concentration changes from c1 to c2, we have and For example, if at t1=1 hour, the concentration c1=50 units, and at t2=2 hours, the concentration c2=70 units, substitute the formula to calculate,

[0084] The slope value indicates that the concentration increases by 20 units per hour. The calculation result reveals the rate of change of concentration within a specified time period, which is crucial for identifying the trend of dust concentration changes at construction sites and formulating corresponding measures.

[0085] S213: Based on the slope data set, using the formula,

[0086] Calculating curvature , evaluate the rate of change of the slope and generate a curvature data set where, It is the difference value of the slope, which is used to reveal the second-order characteristics of data changes;

[0087] The slope at time t3 = 3 hours is s3 = 30 units / hour, while the slope at an earlier time t2 = 2 hours is s2 = 20 units / hour. The curvature is calculated as

[0088] The curvature value represents an increase of 10 units in slope per hour. The result indicates the acceleration of slope change, which helps to identify the key turning points of dust concentration changes at construction sites, thereby providing a deeper understanding and prediction of the dynamics of dust concentration at construction sites.

[0089] See also Figure 4 ,The steps to obtain the key variable extraction results are,

[0090] S221: Using the curvature data set, identify the time point when the curvature value changes significantly and set the threshold parameter Filter the curvature data, all the data greater than The time points corresponding to the curvature values ​​are marked as key time points, and a marked key time point list is generated;

[0091] Using the data obtained from the curvature analysis, we first set the curvature threshold The specific setting method is to count the distribution of all points in the curvature data set and select the value exceeding the 95% quantile as the threshold. Then, scan the curvature value of each time point in the time series data one by one and compare it with the set threshold. If the absolute value of the curvature value exceeds this threshold, mark the time point as a key point. At the same time, record the specific timestamp and curvature value of the time point during the marking process, and associate the corresponding points in the slope data to confirm whether the curvature value is associated with the slope change. Ensure that the abnormal value of the curvature is consistent with the change of the actual data trend. All marks and related data will be integrated into a key time point list for subsequent analysis and processing.

[0092] S222: Based on the marked key time point list and in combination with the slope data set, analyzing the slope change of the previous and next data points, determining the change trend of each time point, and obtaining the trend description record of the key time point;

[0093] Based on the list of key time points obtained, combined with the slope data set, the slope values ​​of each key time point and the time points before and after it are extracted, and the slope differences of the adjacent points before and after each key time point are analyzed in detail. The slope difference before and after the key time point is calculated by reading the values ​​in the slope data set. The difference is the slope value after the key time point minus the slope value before the key time point. If the difference is positive, it means that the dust concentration continues to rise after this time point. If the difference is negative, it indicates that the concentration decreases after this point. If the difference is close to zero, it is considered that the change tends to be stable. All analysis results will be recorded in detail, including the slope change amplitude, change type and corresponding timestamp of each key time point, to ensure that the subsequent dynamic trend report is complete and accurate.

[0094] S223: Integrate the trend description records of key time points and the marked key time point list, record the date, time and corresponding change trend of each key time point, and generate key variable extraction results;

[0095] The trend analysis results obtained above are integrated with the list of key time points. First, the timestamp of each key time point is matched, and the corresponding change type and amplitude are checked. If the slope and curvature changes are consistent, the time point is recorded as a valid key point. Then a detailed dynamic report is generated, and all key time points are arranged in chronological order. Each record contains the timestamp of the key time point, the slope change value, the curvature value and the corresponding trend type. The key variable extraction results are comprehensively output, and the specific change description of each time point is added. It is clearly marked which time points show a significant increase, decrease or stable trend in dust concentration, thereby providing detailed data support for the subsequent adjustment and monitoring strategy of the construction environment.

[0096] See also Figure 5 ,The evaluation steps for the possibility of future dust incidents occurring at the construction site are,

[0097] S311: Based on the key variable extraction results, combined with the real-time monitoring data of the current construction site, integrate and output a comprehensive data set;

[0098] Integrating the key variable extraction results with the real-time monitoring data of the current construction site first involves data collection. The collected data include but are not limited to meteorological conditions, soil type, intensity of construction activities and other factors. After that, data verification is performed, and data integrity verification rules are used to ensure that the data is not missing or repeated. Then data cleaning is performed, including format unification, outlier processing, and the use of median replacement to identify outliers. Finally, the maximum and minimum normalization method is applied to normalize the data to a unified quantitative standard to generate a comprehensive data set. This data set has all the characteristic values ​​required to analyze and predict dust events, ready for subsequent model analysis.

[0099] S312: Perform time series analysis on the comprehensive data set, compare the difference between current data and historical data of the same period, analyze the deviation index of key variables, and obtain a description record of data differences;

[0100] The comprehensive data set was subjected to time series analysis. In this process, the time window of each key variable was first determined. For example, the data of the previous three months were considered, and the autoregressive moving average model (ARMA) was used for preliminary trend and periodicity analysis. The analysis method included the use of sliding average and exponential smoothing techniques to deal with short-term fluctuations. Then the data was seasonally adjusted, the impact of seasonal factors on dust was analyzed, and the deviation index of key variables was calculated ( The deviation index shows the change of the current measurement value relative to the historical average. A positive value indicates that the current concentration is higher than the historical average, and a negative value indicates that the current concentration is lower than the historical average). By comparing the current value with the historical value of the same period, these deviations are quantitatively described, and the deviation index is summarized to form a description of the data difference. This description reflects in detail the specific differences between the current construction site environmental status and historical data, providing the necessary input parameters for the prediction model.

[0101] S313: Based on the data difference description records, the possibility of future dust events is predicted using the formula,

[0102] Calculate the probability of dust events occurring , and obtain the dust event possibility assessment results, among which, Represents the time decay factor, which is used to adjust the impact of time span on probability. Representative The weight of each key variable reflects the relative importance of the variable in the occurrence of dust events. Representative The deviation index of a key variable indicates the degree of deviation between the current value and the historical value of the same period. represents the number of key variables;

[0103] set up , there are three key variables, and their weights are , , , and the corresponding deviation indices are , , , substitute into the formula and get:

[0104] The results show that based on the current data deviation and key variable weights, the probability of dust incidents is about 51.6%. This probability value has direct guiding significance for decision makers in the management and formulation of preventive measures at construction sites.

[0105] See also Figure 6 ,The steps to obtain the prediction results of future dust events are,

[0106] S321: Based on the dust event possibility assessment results, use the environmental data monitored at the construction site and historical data from the same period to analyze the risk level of the location, determine the risk area and the time of dust event occurrence, and generate a spatiotemporal risk map;

[0107] Based on the results of the dust event possibility assessment, combined with the geographic information system (GIS), the environmental data of on-site monitoring and historical data of the same period were used to systematically analyze the risk level of the location, and the environmental sensitivity of each area and the impact of construction activities were evaluated in detail. The specific location of the high-risk area was determined through the superposition analysis of multi-source data; at the same time, combined with historical weather data and construction logs, the high-incidence time period of dust events was analyzed and determined. The time period is usually closely related to the specific climate conditions and construction stages of the region. The time series analysis technology is used to generate detailed spatiotemporal risk maps, which play a key role in refined management and resource allocation, allowing construction site managers to deploy necessary dust prevention measures in advance and optimize construction plans to reduce potential environmental impacts.

[0108] S322: Based on the spatiotemporal risk map, the deep learning model is used to analyze the data, and the historical dust occurrence records and corresponding meteorological conditions are combined to predict the spatiotemporal distribution of dust in the future and obtain the dust distribution prediction results;

[0109] The spatiotemporal risk map is used as input and a deep learning model is used to analyze the data. The model training is based on historical dust occurrence records and related meteorological conditions, including data such as wind speed, temperature, humidity, and the time and location information of dust events. The deep learning model predicts the probability of dust occurrence in each high-risk area at a certain point in the future by identifying the complex connections and patterns between these data, and accurately predicts the spatiotemporal distribution of dust. The model gradually improves the accuracy of the prediction through continuous iteration and optimization. In addition, the visualization of the prediction results is displayed through heat maps and dynamic charts, so that managers can intuitively understand the trend and scope of dust occurrence, providing a scientific basis and decision-making support for the implementation of effective risk management strategies and emergency response measures.

[0110] S323: Combine dust distribution prediction results with current construction activity data to conduct multivariate impact analysis using the formula,

[0111] Calculate dust forecast intensity , and obtain the prediction results of future dust events, among which, represents soil moisture, represents wind speed, represents the intensity of construction activities, , , weight coefficients corresponding to soil moisture, wind speed and intensity of construction activities respectively;

[0112] Represents soil moisture, expressed as a percentage, and the current value collected is 30%; Represents wind speed in meters per second. The current wind speed is 5m / s. Represents the intensity of construction activities. Judging from the construction log, the current intensity of construction activities is the highest (1-10), which is 10.

[0113] , , Represents the weight coefficient of each factor, which is obtained based on historical data analysis. , , . Substitute into the formula to calculate:

[0114] The results show that under the current environment and construction conditions, the dust intensity is 13.5. This value helps to understand the severity of dust at the construction site under specific conditions and provides a basis for formulating corresponding measures.

[0115] See also Figure 7 , the steps to obtain the emergency management plan for dust at the construction site are:

[0116] S411: Analyze the prediction results of future dust events, extract the extreme values ​​and fluctuation ranges of dust concentration in the future period, and generate the estimated data of future dust concentration;

[0117] By analyzing the prediction results of future dust events, the predicted dust concentration extremes, i.e. the highest and lowest concentrations, are extracted. At the same time, various environmental factors, including the actual impact of temperature, humidity and wind speed on the concentration are evaluated, the corresponding environmental data are collected, and the dust concentration prediction values ​​are corrected to obtain more accurate and real-time adjusted dust concentration extremes and fluctuation ranges. The data will serve as the basis for setting warning thresholds.

[0118] S412: According to the estimated data of future dust concentration, set the trigger warning threshold, using the formula:

[0119] Calculate and output the trigger warning threshold ,in, represents the maximum estimated dust concentration, Represents the average volatility concentration of historical data for the same period, is the safety factor;

[0120] Through analysis and extraction, the estimated maximum dust concentration is determined to be 150 ,and It is the average concentration value calculated based on the data of the same period, assuming 30 Safety factor It is set based on on-site safety standards and historical concentration fluctuations, and by evaluating the response effects and safety margins of several past dust events.

[0121] Substitute the formula to calculate the trigger warning threshold:

[0122] The result means that once the monitored dust concentration reaches or exceeds 156 , the early warning system will be automatically triggered, allowing managers to respond in time to reduce the potential harm of dust to the environment and personnel. In this way, the sensitivity and practicality of the early warning system are ensured, and sudden high-concentration dust events can be handled in a timely manner.

[0123] S413: Based on the triggering warning threshold, when the monitoring data shows that the dust concentration reaches and exceeds the warning threshold, the early warning mechanism is automatically activated, an alarm is issued, and the construction site manager is notified to intervene and establish an emergency management plan for dust at the construction site;

[0124] Establish an early warning mechanism and use the set trigger warning threshold to monitor the real-time dust concentration. When the monitoring data shows that the dust concentration reaches or exceeds the warning threshold, the system automatically compares the current concentration with the set threshold. Once an excess is found, the system will immediately activate the alarm mechanism and send an instant notification to the site manager through networked communication equipment (mobile phones and other mobile devices) so that corresponding intervention measures can be quickly taken, including but not limited to increasing the frequency of watering, adjusting construction procedures and time periods, etc., so as to effectively manage and control dust problems on the construction site, protect workers' health and comply with environmental protection regulations.

[0125] A construction site dust monitoring and prediction system based on data analysis, comprising:

[0126] The data preprocessing module collects the original wind speed, humidity, temperature and dust particle concentration data of the construction site, records the timestamp of each data point, removes outliers, performs linear calibration, unifies the data format, and integrates and outputs the preliminary screening data set;

[0127] The data feature analysis module performs time series analysis based on the preliminary screening of the data set, calculates the slope and curvature of each data point, evaluates the rate of change of the slope, reveals the second-order characteristics of the data change, and outputs the slope data set and the curvature data set;

[0128] The key point marking module uses the curvature data set to identify and mark key time points, combines the slope data set to analyze the slope changes of the previous and next data points, determines the change trend of each time point, and integrates to generate key variable extraction results;

[0129] The dust probability prediction module is based on the extraction results of key variables, combined with the real-time monitoring data of the current construction site, to perform time series analysis, analyze the deviation index of key variables, calculate the probability of dust events, predict the possibility of future dust events, and obtain the possibility assessment results of dust events;

[0130] The dust spatial analysis module analyzes the risk level of the location based on the dust event possibility assessment results, determines the risk area and the time of dust event occurrence, predicts the spatiotemporal distribution of dust in the future, calculates the dust prediction intensity based on the current construction activity data, and obtains the prediction results of future dust events;

[0131] The emergency response module analyzes the prediction results of future dust events, extracts the extreme values ​​and fluctuation ranges of dust concentrations in the future, calculates and sets the trigger warning thresholds, and automatically activates the early warning mechanism when the monitoring data shows that the dust concentration reaches and exceeds the warning thresholds, notifying the construction site managers to intervene and establish an emergency management plan for dust at the construction site.

[0132] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A construction site dust monitoring and prediction method based on data analysis, characterized in that: The following steps are involved: Collect data on wind speed, humidity, temperature and dust particle concentration at the construction site, remove outliers from each data item, use linear calibration to correct measurement deviations, and generate a preliminary screening data set; Based on the preliminary screening data set, time series analysis is performed to calculate the slope and curvature of the data points respectively, and key time points are identified according to the slope and curvature calculation results, and the change trend is analyzed to capture the key dynamics of the dust concentration at the construction site and generate key variable extraction results; Based on the extraction results of the key variables, compare the current data with the historical data of the same period, evaluate the possibility of future dust events at the construction site, quantitatively analyze the time and location of potential dust events based on the evaluation results, predict the spatiotemporal distribution of dust, and generate prediction results for future dust events; According to the prediction results of future dust events, a trigger warning threshold is set. When the dust concentration reaches the warning threshold, a warning message is automatically issued to notify the construction site manager to intervene and output an emergency management plan for dust at the construction site; The steps for assessing the possibility of future dust events at the construction site are as follows: Based on the extraction results of the key variables, combined with the real-time monitoring data of the current construction site, a comprehensive data set is integrated and output; Performing time series analysis on the comprehensive data set, comparing the difference between current data and historical data of the same period, analyzing the deviation index of key variables, and obtaining a description record of data differences; Based on the data difference description records, the possibility of future dust events is predicted using the formula: Calculate the probability P of dust event occurrence and obtain the possibility assessment result of dust event occurrence, where λ represents the time attenuation factor, which is used to adjust the influence of time span on probability, and a i represents the weight of the i-th key variable, reflecting the relative importance of the variable in the occurrence of dust events, d i represents the deviation index of the i-th key variable, indicating the degree of deviation between the current value and the historical value of the same period, and n represents the number of key variables; The steps for obtaining the prediction results of future dust events are: Based on the dust event possibility assessment results, the environmental data monitored at the construction site and historical data from the same period are used to analyze the risk level of the location, determine the risk area and the time of dust event occurrence, and generate a spatiotemporal risk map; Based on the spatiotemporal risk map, a deep learning model is used to analyze data, and historical dust occurrence records and corresponding meteorological conditions are combined to predict the spatiotemporal distribution of dust in the future, thereby obtaining a dust distribution prediction result; Combining the dust distribution prediction results with the current construction activity data, a multivariate impact analysis was conducted using the formula: I=w1·h+w2·v+w3·A The predicted dust intensity I is calculated to obtain the prediction results of future dust events, where h represents soil moisture, v represents wind speed, A represents the intensity of construction activities, and w1, w2, and w3 correspond to the weight coefficients of soil moisture, wind speed, and construction activity intensity, respectively.

2. The construction site dust monitoring and prediction method based on data analysis according to claim 1 is characterized in that: The steps for obtaining the emergency management plan for dust at the construction site are as follows: Analyze the prediction results of future dust events, extract the extreme values ​​and fluctuation ranges of dust concentrations in the future, and generate estimated data of future dust concentrations; According to the estimated data of future dust concentration, the trigger warning threshold is set, using the formula: T=C max +k·C avg Calculate and output the trigger warning threshold T, where C max represents the maximum estimated dust concentration, C avg represents the average fluctuation concentration of historical data for the same period, and k is the safety factor; Based on the trigger warning threshold, when the monitoring data shows that the dust concentration reaches and exceeds the warning threshold, the early warning mechanism is automatically activated, an alarm is issued, and the construction site manager is notified to intervene and establish an emergency management plan for dust at the construction site.

3. A construction site dust monitoring and prediction system based on data analysis, characterized in that: The system is used to execute the construction site dust monitoring and prediction method based on data analysis according to any one of claims 1-2, comprising: The data preprocessing module collects the original wind speed, humidity, temperature and dust particle concentration data of the construction site, records the timestamp of each data point, removes outliers, performs linear calibration, unifies the data format, and integrates and outputs the preliminary screening data set; The data feature analysis module performs time series analysis based on the preliminary screening data set, calculates the slope and curvature of each data point, evaluates the rate of change of the slope, reveals the second-order characteristics of the data change, and outputs a slope data set and a curvature data set; The key point marking module uses the curvature data set to identify and mark key time points, combines the slope data set to analyze the slope changes of the previous and next data points, determines the change trend of each time point, and integrates to generate key variable extraction results; The dust probability prediction module performs time series analysis based on the extraction results of the key variables and the real-time monitoring data of the current construction site, analyzes the deviation index of the key variables, calculates the probability of dust events, predicts the possibility of future dust events, and obtains the possibility assessment results of dust events; The dust spatial analysis module analyzes the risk level of the location based on the dust event possibility assessment result, determines the risk area and the time of dust event occurrence, predicts the spatiotemporal distribution of dust in the future, calculates the dust prediction intensity in combination with the current construction activity data, and obtains the prediction result of future dust events; The emergency response module analyzes the prediction results of future dust events, extracts the extreme values ​​and fluctuation ranges of dust concentrations in the future, calculates and sets the trigger warning thresholds, and automatically activates the early warning mechanism when the monitoring data shows that the dust concentration reaches and exceeds the warning threshold, notifying the construction site manager to intervene and establish an emergency dust management plan for the construction site.

Images