IoT-based intelligent management system for construction site monitoring data

Abstract

The present invention relates to the field of data processing technology, and specifically to an intelligent management system for construction site monitoring data based on the Internet of Things; the degree of abnormality is obtained according to the quantity characteristics and interval characteristics of other data points with the same numerical value as any data point in the historical dust concentration sequence, and the data distribution characteristics within the preset neighborhood range of any data point; the association window of any data point is obtained according to the degree of abnormality; a sliding match is performed within the preset adjacent interval of any data point according to the association window to obtain a matching value; an adjustment degree is obtained according to the difference characteristics and degree of abnormality between any data point and the matching value; any data point is adjusted according to the adjustment degree to obtain a corrected value. The present invention constructs a dust prediction model according to the corrected numerical values ​​corresponding to the data points in the historical dust concentration sequence, predicts dust concentration changes and issues an early warning according to the dust prediction model, and improves prediction accuracy and pollution control effects.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an intelligent management system for construction site monitoring data based on the Internet of Things. Background Art

[0002] Since there are air pollution sources such as dust and exhaust gas at the construction site, environmental pollution and other adverse effects occur. In order to solve the pollution problem at the construction site, it is necessary to monitor the pollution data of the construction site, and take corresponding governance measures for intelligent management by processing and analyzing the monitoring data. In the intelligent management of the construction site, a prediction model can be built through a neural network based on historical pollution data, and future pollution data can be predicted through the prediction model, so as to respond in advance according to the prediction results. However, this method ignores the interference of uncertain factors in the construction site, and the historical pollution data obtained may be noisy, which reduces the stability of the data and reduces the accuracy of the constructed prediction model; ultimately, the prediction results are inaccurate, which affects the environmental governance effect of the construction site. Summary of the invention

[0003] In order to solve the above technical problems that the historical pollution data is unstable, affecting the prediction accuracy and environmental governance, the purpose of the present invention is to provide a construction site monitoring data intelligent management system based on the Internet of Things. The technical solutions adopted are as follows:

[0004] A data acquisition module is used to obtain the historical dust concentration sequence of the construction site;

[0005] An anomaly analysis module, used to obtain a first anomaly coefficient according to the quantity characteristics and interval characteristics of other data points with the same value as any data point in the historical dust concentration sequence; obtain a second anomaly coefficient according to the data distribution characteristics within a preset neighborhood of the any data point; and obtain the degree of anomaly of the any data point according to the first anomaly coefficient and the second anomaly coefficient;

[0006] A data adjustment module; used to obtain a correlation window of the arbitrary data point according to the abnormality degree and the preset correlation range; perform sliding matching within a preset adjacent interval of the arbitrary data point according to the correlation window to obtain a matching value; and obtain an adjustment degree according to the difference characteristics between the arbitrary data point and the matching value and the abnormality degree;

[0007] The data monitoring module is used to adjust the arbitrary data point according to the adjustment degree to obtain a corrected value; construct a dust prediction model according to the corrected value corresponding to each data point in the historical dust concentration sequence; predict dust concentration changes according to the dust prediction model and issue an early warning.

[0008] Furthermore, the step of obtaining the first abnormal coefficient according to the quantity characteristics and interval characteristics of other data points having the same value as any data point in the historical dust concentration sequence includes:

[0009] Calculate the proportion of other data points with the same numerical value as the arbitrary data point in the historical dust concentration sequence to obtain the proportion frequency value; calculate the variance of the adjacent interval lengths of a preset number of the other data points closest to the arbitrary data point and normalize them to obtain the interval distribution characteristic value; negatively correlate the proportion frequency value and calculate the product with the interval distribution characteristic value to obtain the first abnormal coefficient of the arbitrary data point.

[0010] Furthermore, the step of obtaining a second anomaly coefficient according to the data distribution characteristics within a preset neighborhood range of the arbitrary data point includes:

[0011] Calculate the variance of the values ​​of all data points within a preset neighborhood range of the arbitrary data point to obtain a first fluctuation characteristic value; calculate the variance of the values ​​of the data points within a preset neighborhood range of the arbitrary data point excluding the arbitrary data point to obtain a second fluctuation characteristic value; calculate the absolute value of the difference between the first fluctuation characteristic value and the second fluctuation characteristic value and normalize them to obtain a second abnormality coefficient of the arbitrary data point.

[0012] Further, the step of obtaining the abnormality degree of the arbitrary data point according to the first abnormality coefficient and the second abnormality coefficient comprises:

[0013] The product of the first abnormality coefficient and the second abnormality coefficient is calculated to obtain the abnormality degree of the arbitrary data point.

[0014] Furthermore, the step of obtaining the correlation window of the arbitrary data point according to the abnormality degree and the preset correlation range includes:

[0015] Calculate the difference between the maximum value of the preset association range and the minimum value of the preset association range to obtain an association reference value; calculate the product of the association reference value and the abnormality degree to obtain a range adjustment value; calculate the sum of the range adjustment value and the minimum value of the preset association range to obtain the window length of the arbitrary data point, and construct the association window according to the window length and at the position of the arbitrary data point closest to the window center.

[0016] Furthermore, the step of performing sliding matching within a preset adjacent interval of the arbitrary data point according to the association window to obtain a matching value includes:

[0017] According to the associated window, sliding with a preset step size within the preset adjacent interval, a sliding window is obtained each time it slides; the dynamic time regularization distance between the sliding window and the data fragment in the original associated window is calculated to obtain the difference between the sliding window and the associated window; in the sliding window corresponding to the minimum value of the difference, the data point value with the same position as any data point in the associated window is used as the matching value.

[0018] Furthermore, the step of obtaining the adjustment degree according to the difference characteristics between the arbitrary data point and the matching value and the abnormality degree includes:

[0019] The numerical difference between the matching value and the arbitrary data point is calculated to obtain a difference value; and the product of the difference value and the abnormality degree is calculated to obtain an adjustment degree of the arbitrary data point.

[0020] Furthermore, the step of adjusting the arbitrary data point according to the adjustment degree to obtain a corrected value includes:

[0021] The sum of the value of the arbitrary data point and the adjustment degree is calculated to obtain a corrected value of the arbitrary data point.

[0022] The present invention has the following beneficial effects:

[0023] In the present invention, obtaining the first abnormal coefficient can determine the abnormal situation of any data point according to the low frequency and randomness of the abnormal data, and obtaining the second abnormal coefficient can determine the abnormal situation of any data point according to the degree of influence of the abnormal data on its neighborhood data; obtaining the degree of abnormality according to the first abnormal coefficient and the second abnormal coefficient can accurately determine the abnormal characteristics of any data point, thereby improving the accuracy of subsequent data correction and construction of dust prediction model. Obtaining the association window can more accurately determine the association range according to the degree of abnormality of any data point, thereby improving the accuracy of matching value and correction data acquisition; obtaining the matching value can determine the value at the same position as the arbitrary data point according to the regularity of the historical dust concentration sequence change, thereby improving the accuracy of the correction value. Obtaining the correction value can make the value of the arbitrary data point closer to the actual situation of the construction site, thereby improving the accuracy of the dust prediction model and the governance effect of the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 A block diagram of an intelligent management system for construction site monitoring data based on the Internet of Things provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, characteristics and effects of a construction site monitoring data intelligent management system based on the Internet of Things proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0027] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0028] The specific scheme of a construction site monitoring data intelligent management system based on the Internet of Things provided by the present invention is described in detail below with reference to the accompanying drawings.

[0029] See also Figure 1 , which shows a block diagram of a construction site monitoring data intelligent management system based on the Internet of Things provided by an embodiment of the present invention, the system includes the following modules:

[0030] The data acquisition module S1 is used to obtain the historical dust concentration sequence of the construction site.

[0031] In the embodiment of the present invention, the implementation scenario is to build a prediction model through processed historical pollution data, so as to improve the accuracy of data prediction and the effect of pollution control. First, the historical dust concentration sequence of the construction site is obtained. In the embodiment of the present invention, continuous dust concentration data within one week of the construction site is collected to build a historical dust concentration sequence; it should be noted that the implementer can determine the length of the historical dust concentration sequence and the collection frequency according to the implementation scenario, and the dust concentration at the construction site needs to be stable and continuous, so that the change pattern can be identified for prediction.

[0032] The anomaly analysis module S2 is used to obtain a first anomaly coefficient based on the quantity characteristics and interval characteristics of other data points with the same numerical value as any data point in the historical dust concentration sequence; obtain a second anomaly coefficient based on the data distribution characteristics within a preset neighborhood range of any data point; and obtain the degree of anomaly of any data point based on the first anomaly coefficient and the second anomaly coefficient.

[0033] Due to various uncertain factors at the construction site, there will be abnormal noise data in the historical dust concentration sequence, which will lead to deviations in the dust prediction model constructed by training, making the construction site governance inaccurate or untimely; therefore, in order to ensure the dust governance effect at the construction site, it is necessary to pre-process the historical dust concentration sequence, correct the more abnormal data, and improve the accuracy of the dust prediction model. Since the number of abnormal data in the historical dust concentration sequence is small and the appearance is relatively random and irregular, the first abnormal coefficient can be obtained based on the number characteristics and interval characteristics of other data points in the historical dust concentration sequence that have the same value as any data point.

[0034] Preferably, in an embodiment of the present invention, the step of obtaining the first abnormal coefficient includes: calculating the proportion of the number of other data points with the same value as the arbitrary data point in the historical dust concentration sequence to obtain the proportion frequency value; when the proportion is lower, the proportion frequency value is lower, which means that the value of the arbitrary data point appears less frequently in the historical dust concentration sequence, and the arbitrary data point is more likely to be abnormal data. Calculate the variance of the adjacent interval lengths of the preset number of other data points closest to the arbitrary data point and normalize them to obtain the interval distribution characteristic value; in an embodiment of the present invention, the preset number is 10, which can be determined by the implementer according to the implementation scenario; the adjacent interval length is the number of data points between two adjacent other data points. When the variance of the adjacent interval lengths of the 10 other data points with the same value closest to the arbitrary data point is larger, the interval distribution characteristic value is larger, which means that the value of the arbitrary data point is less regular and the time of appearance is more random, and the arbitrary data point is more likely to be abnormal data. The proportion frequency value is negatively correlated and the product with the interval distribution characteristic value is calculated to obtain the first abnormal coefficient of any data point; when the proportion frequency value is smaller and the interval distribution value is larger, the first abnormal coefficient is larger, which means that the frequency of occurrence of the value of the arbitrary data point is lower and the occurrence time is more random, and the arbitrary data point is more likely to be abnormal data.

[0035] Furthermore, since the changes in the historical dust concentration sequence are continuous and regular, the probability of instantaneous volatility is low; therefore, the more the arbitrary data point affects the overall data change trend within its local data range, the more likely it is that the arbitrary data point is abnormal data. Then, the second abnormal coefficient is obtained according to the data distribution characteristics within the preset neighborhood range of the arbitrary data point. Preferably, in the embodiment of the present invention, the step of obtaining the second abnormal coefficient includes: calculating the variance of all data point values ​​within the preset neighborhood range of the arbitrary data point to obtain the first fluctuation characteristic value; in the embodiment of the present invention, the preset neighborhood range is composed of 50 data points before and after the arbitrary data point as the center. If this requirement is not met, the preset neighborhood range is constructed with the 50 data points closest to the center position of the arbitrary data point. The implementer can determine it according to the implementation scenario. Calculate the variance of the data point values ​​within the preset neighborhood range of the arbitrary data point excluding the arbitrary data point to obtain the second fluctuation characteristic value; calculate the absolute value of the difference between the first fluctuation characteristic value and the second fluctuation characteristic value and normalize it to obtain the second abnormal coefficient of the arbitrary data point. The greater the difference between the first fluctuation characteristic value and the second fluctuation characteristic value, the more discrete the value of the arbitrary data point is, the more it affects the calculation result of the variance within the preset neighborhood range, and the more likely the arbitrary data point is to be abnormal data.

[0036] The first abnormal data and the second abnormal data can reflect the possibility that the arbitrary data point is abnormal data, so the degree of abnormality of the arbitrary data point can be obtained according to the first abnormal coefficient and the second abnormal coefficient; preferably, in the embodiment of the present invention, the step of obtaining the degree of abnormality includes: calculating the product of the first abnormal coefficient and the second abnormal coefficient to obtain the degree of abnormality of the arbitrary data point. When the values ​​of the first abnormal coefficient and the second abnormal coefficient of the arbitrary data point are larger, the degree of abnormality is larger, and the arbitrary data point is more likely to be abnormal data; the formula for obtaining the degree of abnormality includes:

[0037]

[0038] In the formula, R represents the degree of abnormality, P represents the frequency value, norm represents the normalization function, T represents the preset number, and L t represents the length of the tth adjacent interval, represents the average value of the lengths of adjacent intervals, It represents the variance of the adjacent interval lengths of the T data points with the same value closest to the arbitrary data point. represents the interval distribution characteristic value; N represents the number of data points within the preset neighborhood, D n Represents the value of the nth data point within the preset neighborhood. Indicates the numerical average value of the preset neighborhood range; represents the first fluctuation characteristic value, M represents the number of data points within the preset neighborhood excluding the arbitrary data point, and D n It represents the value of the mth data point within the preset neighborhood, excluding the arbitrary data point. It represents the average value of the preset neighborhood except for the arbitrary data point. represents the second fluctuation characteristic value; represents the first abnormal coefficient, Represents the second outlier coefficient.

[0039] Data adjustment module S3; used to obtain the association window of any data point according to the abnormality degree and the preset association range; perform sliding matching within the preset adjacent interval of any data point according to the association window to obtain the matching value; obtain the adjustment degree according to the difference characteristics and abnormality degree between any data point and the matching value.

[0040] After obtaining the degree of abnormality of different data points in the historical dust concentration sequence, the data point value can be adjusted according to the degree of abnormality. The greater the degree of abnormality, the higher the degree of adjustment. Since the dust changes in the historical dust concentration sequence are relatively stable and regular, it is possible to find a segment with a similar change pattern near the arbitrary data point, and determine the data at a similar position to the arbitrary data point from the segment. When the degree of abnormality of the arbitrary data point is greater, a longer data segment is required for matching to avoid the accuracy of data correction affected by a shorter data segment; therefore, the association window of the arbitrary data point is obtained based on the degree of abnormality and the preset association range.

[0041] Preferably, in an embodiment of the present invention, the step of obtaining the association window includes: calculating the difference between the maximum value of the preset association range and the minimum value of the preset association range to obtain the association reference value; in an embodiment of the present invention, the maximum value of the preset association range is a range of 100 data points, and the minimum value is a range of 10 data points. The implementer can determine it by himself according to the implementation scenario, and the association reference value provides a reference value for adjusting the association window. Calculate the product of the association reference value and the degree of abnormality to obtain the range adjustment value; when the degree of abnormality is greater, a longer data segment is required for data matching, and the range adjustment value is larger, so as to avoid the value of any data point being abnormal and affecting the matching accuracy. The sum of the range adjustment value and the minimum value of the preset association range is calculated to obtain the window length of the arbitrary data point. The association window is constructed according to the window length and with the position of the arbitrary data point closest to the window center. For example, if the window length is 81, the location of the association window is an area consisting of 80 data points before and after the arbitrary data point as the center. When the arbitrary data point is not the center, the data segment closest to the window center of the arbitrary data point is used as the association window. The greater the degree of abnormality, the larger the association window of the arbitrary data point, thereby improving the accuracy of data matching.

[0042] Furthermore, a range that is more closely matched with the data in the associated window can be found within the preset adjacent interval of the arbitrary data point, and thus a sliding match is performed within the preset adjacent interval of the arbitrary data point according to the associated window to obtain a matching value; preferably, in an embodiment of the present invention, the step of obtaining a matching value includes: sliding the associated window with a preset step size within the preset adjacent interval, and obtaining a sliding window each time it slides; in an embodiment of the present invention, the preset step size is 3, and the length of three data points is moved each time. The preset adjacent interval is a range constructed by the most recent 1000 data points of the arbitrary data point, and the implementer can determine it according to the implementation scenario; the dynamic time warping distance between the sliding window and the data fragment in the original associated window is calculated to obtain the difference between the sliding window and the associated window; it should be noted that the dynamic time warping distance is obtained by the existing dynamic time warping algorithm, and the specific steps will not be repeated. The more similar the two data fragments are, the smaller the dynamic time warping distance is. In the sliding window corresponding to the minimum difference value, the value of the data point with the same position as the arbitrary data point in the associated window is taken as the matching value; because the matching value has the same position as the arbitrary data point in the associated window, the matching value should be similar to the value of the arbitrary data point under normal circumstances.

[0043] After obtaining the most similar matching value of the arbitrary data point, the adjustment degree can be obtained according to the difference characteristics and abnormality degree between the arbitrary data point and the matching value; preferably, in an embodiment of the present invention, the step of obtaining the adjustment degree includes: calculating the numerical difference between the matching value and the arbitrary data point to obtain the difference value; when the numerical difference between the matching value and the arbitrary data point is larger, the difference value is closer to 0, and the product of the difference value and the abnormality degree is calculated to obtain the adjustment degree of the arbitrary data point; when the abnormality degree of the arbitrary data point is larger, it means that the deviation of the numerical value of the arbitrary data point is larger, and at the same time, the difference between the arbitrary data point and the corresponding matching value is larger, then the absolute value of the adjustment degree is larger, and the more it is necessary to adjust the numerical value of the arbitrary data point. Since the abnormality of the arbitrary data point may be too large or too small, the adjustment degree may be negative.

[0044] The data monitoring module S4 is used to adjust any data point according to the adjustment degree to obtain a corrected value; construct a dust prediction model according to the corrected value corresponding to each data point in the historical dust concentration sequence; and predict dust concentration changes and issue an early warning based on the dust prediction model.

[0045] After obtaining the adjustment degree of the arbitrary data point, the arbitrary data point can be adjusted according to the adjustment degree to obtain a corrected value, specifically including: calculating the sum of the value of the arbitrary data point and the adjustment degree to obtain the corrected value of the arbitrary data point; when. If the difference between the arbitrary data point and the matching value is greater, and the degree of abnormality is greater, the more the value of the arbitrary data point is adjusted, and the greater the adjustment degree; when the degree of abnormality is smaller or the difference value is smaller, the adjustment degree is smaller, which means that the value of the arbitrary data point is more normal, and the corrected data is closer to the original value of the arbitrary data point. When the value of the arbitrary data point is smaller, the adjustment degree is a positive number, and the corrected value is greater than the original value of the arbitrary data point; when the value of the arbitrary data point is larger, the adjustment degree is complex, and the corrected value is less than the original value of the arbitrary data point. After obtaining the corrected value of each data point in the historical dust concentration sequence, a dust prediction model can be constructed according to the corrected value corresponding to each data point in the historical dust concentration sequence; it should be noted that the construction of a dust prediction model through a neural network based on the corrected data of the historical dust concentration sequence belongs to the prior art, and the specific steps are not repeated here. The dust prediction model constructed by the corrected value can improve the prediction accuracy. According to the dust prediction model, the dust concentration change is predicted and warned, and the dust concentration sequence collected in real time is input into the dust prediction model to more accurately predict the dust concentration at a future moment. According to the comparison between the predicted dust concentration and the dust threshold, the corresponding warning and dust reduction measures are implemented according to the comparison results, thereby improving the dust control effect at the construction site.

[0046] In summary, the embodiments of the present invention provide an intelligent management system for construction site monitoring data based on the Internet of Things; the degree of abnormality is obtained according to the quantity characteristics and interval characteristics of other data points with the same numerical value as any data point in the historical dust concentration sequence, and the data distribution characteristics within the preset neighborhood range of any data point; the association window of any data point is obtained according to the degree of abnormality; a sliding match is performed within the preset adjacent interval of any data point according to the association window to obtain a matching value; the degree of adjustment is obtained according to the difference characteristics and degree of abnormality between any data point and the matching value; any data point is adjusted according to the degree of adjustment to obtain a corrected value. The present invention constructs a dust prediction model according to the corrected numerical values ​​corresponding to the data points in the historical dust concentration sequence, predicts dust concentration changes and issues warnings according to the dust prediction model, and improves prediction accuracy and pollution control effects.

[0047] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A construction site monitoring data intelligent management system based on the Internet of Things, characterized in that: The system includes the following modules: A data acquisition module is used to obtain the historical dust concentration sequence of the construction site; An anomaly analysis module, used to obtain a first anomaly coefficient according to the quantity characteristics and interval characteristics of other data points with the same value as any data point in the historical dust concentration sequence; obtain a second anomaly coefficient according to the data distribution characteristics within a preset neighborhood of the any data point; and obtain the degree of anomaly of the any data point according to the first anomaly coefficient and the second anomaly coefficient; A data adjustment module; used to obtain a correlation window of the arbitrary data point according to the abnormality degree and the preset correlation range; perform sliding matching within a preset adjacent interval of the arbitrary data point according to the correlation window to obtain a matching value; and obtain an adjustment degree according to the difference characteristics between the arbitrary data point and the matching value and the abnormality degree; A data monitoring module is used to adjust the arbitrary data point according to the adjustment degree to obtain a correction value; construct a dust prediction model according to the correction value corresponding to each data point in the historical dust concentration sequence; predict dust concentration changes and issue an early warning according to the dust prediction model; The step of obtaining the first abnormal coefficient according to the quantity characteristics and interval characteristics of other data points with the same value as any data point in the historical dust concentration sequence comprises: Calculate the proportion of other data points with the same value as the arbitrary data point in the historical dust concentration sequence to obtain the proportion frequency value; calculate the variance of the adjacent interval lengths of a preset number of the other data points closest to the arbitrary data point and normalize them to obtain the interval distribution characteristic value; negatively correlate the proportion frequency value and calculate the product with the interval distribution characteristic value to obtain the first abnormal coefficient of the arbitrary data point; The step of obtaining a second abnormality coefficient according to the data distribution characteristics within a preset neighborhood of the arbitrary data point comprises: Calculate the variance of all data point values ​​within a preset neighborhood of the arbitrary data point to obtain a first fluctuation characteristic value; calculate the variance of data point values ​​within a preset neighborhood of the arbitrary data point excluding the arbitrary data point to obtain a second fluctuation characteristic value; calculate the absolute value of the difference between the first fluctuation characteristic value and the second fluctuation characteristic value and normalize them to obtain a second abnormal coefficient of the arbitrary data point; The step of obtaining the associated window of the arbitrary data point according to the abnormality degree and the preset associated range comprises: Calculate the difference between the maximum value of the preset association range and the minimum value of the preset association range to obtain an association reference value; calculate the product of the association reference value and the abnormality degree to obtain a range adjustment value; calculate the sum of the range adjustment value and the minimum value of the preset association range to obtain the window length of the arbitrary data point, and construct the association window according to the window length and at the position of the arbitrary data point closest to the window center.

2. According to the Internet of Things-based construction site monitoring data intelligent management system according to claim 1, it is characterized in that: The step of obtaining the abnormality degree of the arbitrary data point according to the first abnormality coefficient and the second abnormality coefficient comprises: The product of the first abnormality coefficient and the second abnormality coefficient is calculated to obtain the abnormality degree of the arbitrary data point.

3. According to the Internet of Things-based construction site monitoring data intelligent management system according to claim 1, it is characterized in that: The step of performing sliding matching within a preset adjacent interval of any data point according to the association window to obtain a matching value comprises: According to the associated window, sliding with a preset step size within the preset adjacent interval, a sliding window is obtained each time it slides; the dynamic time regularization distance between the sliding window and the data fragment in the original associated window is calculated to obtain the difference between the sliding window and the associated window; in the sliding window corresponding to the minimum value of the difference, the data point value with the same position as any data point in the associated window is used as the matching value.

4. According to the Internet of Things-based construction site monitoring data intelligent management system according to claim 1, it is characterized in that: The step of obtaining the adjustment degree according to the difference characteristics between the arbitrary data point and the matching value and the abnormality degree comprises: The numerical difference between the matching value and the arbitrary data point is calculated to obtain a difference value; and the product of the difference value and the abnormality degree is calculated to obtain an adjustment degree of the arbitrary data point.

5. According to the Internet of Things-based construction site monitoring data intelligent management system according to claim 1, it is characterized in that: The step of adjusting the arbitrary data point according to the adjustment degree to obtain a corrected value comprises: The sum of the value of the arbitrary data point and the adjustment degree is calculated to obtain a corrected value of the arbitrary data point.

Images