Visual management method of smart park based on digital twin

Abstract

The present application relates to the field of visual management, and more specifically to a visual management method for a smart park based on digital twins. The method comprises: collecting PM 2.5 concentration; the concentration of a region at several moments constitutes a concentration sequence, and the concentration rise abnormal value is obtained according to the concentration value difference and the degree of disorder in the concentration sequence; the abnormal correction value is obtained according to the difference in the concentration standard limit and the concentration rise abnormal value; for each region at the same moment, the adjacent region is obtained, and the abnormal impact value is obtained according to the distance between the region and the adjacent region combined with the abnormal correction value; the abnormal impact value at the same moment constitutes a set, and the number of divided intervals is obtained according to the discrete situation and the number of the set, based on which the abnormal monitoring is completed and visual management is performed. This application improves the monitoring accuracy of air quality anomalies.

Description

Technical Field

[0001] The present application relates to the field of visualization management, and specifically to a visualization management method for smart parks based on digital twins. Background Art

[0002] Digital twins integrate data such as physical models, sensors, and operation history, and use 3D modeling, virtual reality and other technical means to build a digital mirror of the park's physical space in a virtual space. The smart park visualization platform uses digital twin technology to integrate, analyze and display data such as the park's infrastructure, personnel flow, and energy consumption, in order to achieve a comprehensive monitoring and control platform for park operations, management, and decision-making.

[0003] The platform helps the park maintain good environmental quality by monitoring the park's air quality, noise level, temperature and humidity and other environmental indicators, and provides real-time alarm and warning functions, as well as the location and tracking of environmental pollution sources. Ambient air quality monitoring data is the most important basis for understanding air quality. Abnormal data often appears due to reasons such as fossil fuel combustion and automobile exhaust emissions. The visual management of smart parks requires abnormal monitoring of collected data so that managers can promptly discover abnormal behaviors in the smart park and take management measures. However, due to the fluidity of air, traditional abnormal data monitoring methods are difficult to accurately identify abnormal air quality data and cannot meet the needs of quasi-real-time application of massive monitoring data. Summary of the invention

[0004] In order to solve the technical problem of low accuracy of abnormal monitoring, this application provides a visual management method for a smart park based on digital twins. The technical solutions adopted are as follows:

[0005] This application proposes a visual management method for a smart park based on digital twins, which includes the following steps:

[0006] Collect data from different areas at different times concentration;

[0007] For each area at each moment, the concentration at the preset moment before each moment and the concentration at the current moment are combined into a concentration sequence, and the concentration increase abnormal value of each area at each moment is obtained according to the maximum difference in concentration in the concentration sequence and the degree of disorder of the concentration sequence;

[0008] Presets The standard limit is based on each area at each time The concentration and The difference between the standard limit and the abnormal concentration rise value of each area at each moment is combined to obtain the abnormal correction value of each area at each moment;

[0009] For each region at the same time, the K-nearest neighbor algorithm is used to obtain the neighboring regions of each region; the distance between each region and the neighboring region is used as the weight and the neighboring region is used as the weight. The concentration weights are then combined with the anomaly correction value and The concentration of each area is used to obtain the abnormal impact value of each area;

[0010] The abnormal impact values ​​of all areas at the same time are combined into a set. The discrete situation of the abnormal impact value is obtained according to the variance and mean of the abnormal impact value in the set. The number of divided intervals is obtained based on the discrete situation and the number of abnormal impact values ​​in the set. The HBOS algorithm is used to monitor the abnormalities of each area according to the number of divided intervals, and the abnormalities are visualized for management.

[0011] In the above scheme, this application collects air in different areas of the smart park. concentration series, based on the The concentration increase abnormal value is constructed by the fluctuation of the data collected within a period when the concentration increases, and the The data fluctuations in the three situations of concentration increase, stability and decrease are used to correct the abnormal value of concentration increase to obtain the abnormal correction value. The beneficial effect is that the calculated abnormal correction value can better reflect the current moment in the air Secondly, according to When the pollution source diffuses in the air, the degree of influence on different regions is different. The abnormal impact value is constructed, and the number of interval divisions of the data during abnormal data monitoring is calculated based on the abnormal impact value. The beneficial effect is that the abnormal impact value will The concentration can amplify the degree of air pollution. According to the abnormal impact value, it can more accurately judge the concentration of air pollution in different areas of the smart park at each moment. In the case of abnormal concentration, the appropriate number of interval divisions for abnormal data monitoring using the HBOS algorithm is obtained based on the calculation results, which avoids inaccurate monitoring of abnormal air quality data due to air fluidity and improves the monitoring accuracy of abnormal air quality.

[0012] In one embodiment, the acquisition of different areas at different times The concentration method is:

[0013] Divide the smart park into areas of equal size and place The concentration monitor collects data once every preset time, and the preset number of times is regarded as a cycle.

[0014] In one embodiment, the method of forming a concentration sequence by combining the concentration at a preset time before each time point and the concentration at the current time point is:

[0015] For each moment in a region, the data collected in the previous cycle including that moment are The concentration of composes a time series and is recorded as concentration series.

[0016] In one embodiment, the method for obtaining the concentration rise abnormal value of each region at each moment according to the maximum difference of the concentration in the concentration sequence and the degree of disorder of the concentration sequence is:

[0017] Get the concentration range in the concentration series;

[0018] Calculate the information entropy of the concentration sequence, and the expression of abnormal concentration increase is:

[0019] , express The information entropy of represents the abnormal value of concentration increase at the tth time in the i-th region, Represents the concentration range at the tth time in the ith region.

[0020] In one embodiment, the expression of the abnormal correction value is:

[0021] , Indicates Moment Air in the area concentration; Indicates that the air in the smart park The standard limit of Indicates Moment The anomaly correction value of the area, It represents the abnormal value of concentration rise at the tth time in the ith region.

[0022] In one embodiment, for each region at the same time, the method of obtaining the neighboring region of each region by using the K-nearest neighbor algorithm is:

[0023] Calculate the distance between each area and all other areas at the same time, sort the absolute values ​​of all distances from small to large, and select a preset number of areas as neighboring areas. The distance is the Euclidean distance between the coordinates of the center points of the areas.

[0024] In one embodiment, the expression of the abnormal impact value is:

[0025] , represents the Euclidean distance between the ith region and its jth neighboring region, represents the jth neighboring region of the ith region at the tth time The concentration of represents the i-th region at the t-th time The concentration of represents the number of neighboring regions, represents the abnormal impact value of the ith region at the tth moment, Indicates Moment The anomaly correction value for each region.

[0026] In one embodiment, the method of forming a set of abnormal impact values ​​of all regions at the same time and obtaining the discrete situation of the abnormal impact values ​​according to the variance and mean of the abnormal impact values ​​in the set is:

[0027] The abnormal impact values ​​of all regions at the same time are combined into a set and recorded as the abnormal impact set;

[0028] Let the ratio of the standard deviation and the mean of all abnormal impact values ​​in the abnormal impact set be the discrete case of the abnormal impact set.

[0029] In one embodiment, the method for obtaining the number of partition intervals based on the discrete cases and the number of abnormal impact values ​​in the set is:

[0030] , represents the standard discrete case of the set affected by the anomaly, Indicates the number of abnormal impact values ​​in the abnormal impact set. represents the rounding function, Indicates the number of partition intervals in the HBOS algorithm.

[0031] In one embodiment, the method of using the HBOS algorithm to monitor the abnormality of each area by dividing the number of intervals is:

[0032] The number of sample points in each divided interval is counted to generate a histogram, and the concentration anomaly score of each area is obtained through the HBOS algorithm. The anomaly threshold is preset, and the concentration with a concentration anomaly score greater than the anomaly threshold is recorded as a concentration anomaly.

[0033] The beneficial effects of this application are as follows:

[0034] This application collects air in different areas of the smart park. concentration series, based on the The concentration increase abnormal value is constructed by the fluctuation of the data collected within a period when the concentration increases, and the The data fluctuations in the three situations of concentration increase, stability and decrease are used to correct the abnormal value of concentration increase to obtain the abnormal correction value. The beneficial effect is that the calculated abnormal correction value can better reflect the current moment in the air Secondly, according to When the pollution source diffuses in the air, the degree of influence on different regions is different. The abnormal impact value is constructed, and the number of interval divisions of the data during abnormal data monitoring is calculated based on the abnormal impact value. The beneficial effect is that the abnormal impact value will The concentration can amplify the degree of air pollution. According to the abnormal impact value, it can more accurately judge the concentration of air pollution in different areas of the smart park at each moment. In the case of abnormal concentration, the appropriate number of interval divisions for abnormal data monitoring using the HBOS algorithm is obtained based on the calculation results, thus avoiding inaccurate monitoring of abnormal air quality data due to air fluidity. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A flow chart of a visual management method for a smart park based on digital twins provided in one embodiment of the present application;

[0037] Figure 2 For the air in the smart park Implementation flow chart for abnormal monitoring of concentration. DETAILED DESCRIPTION

[0038] In order to further explain the technical means and effects adopted by this application to achieve the predetermined invention purpose, the specific implementation method, structure, features and effects of the smart park visualization management method based on digital twins proposed in this application 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.

[0039] 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 application belongs.

[0040] Implementation example of a visual management method for a smart park based on digital twins:

[0041] The specific scheme of the digital twin-based smart park visualization management method provided in this application is described in detail below with reference to the accompanying drawings.

[0042] See also Figure 1 , which shows a flow chart of a smart park visualization management method based on digital twins provided by an embodiment of the present application, the method comprising the following steps:

[0043] Step S001: Collect data from different areas at different times concentration.

[0044] There are many types of air pollutants in the atmosphere, including but not limited to nitrogen dioxide ( )、Carbon monoxide( )、sulfur dioxide( ),ozone( )、fine particles( ) and inhalable particulate matter ( ). This application is for The concentration data is monitored for abnormalities.

[0045] Refers to particles with a diameter of less than or equal to 2.5 microns. This pollutant is highly active, easily carries toxic substances, stays in the atmosphere for a long time, and travels a long distance. However, due to the fluidity of the air, the air in different locations monitored at the same time may have different The concentration will be different. This application evenly arranges sensors in the smart park and collects air concentration at fixed time intervals. The concentration of The concentration is recorded as the monitoring data set.

[0046] Preferably, in this embodiment, the smart park is divided into areas of equal area, and a Concentration monitor, collects data every half an hour, 48 times in one cycle concentration.

[0047] Preferably, in another embodiment, the smart park is divided into areas of equal area, and a Concentration monitor, collects data every ten minutes, 100 times in one cycle concentration.

[0048] So far, different areas at different times have been obtained concentration.

[0049] Step S002: for each region at each moment, the concentration at the preset moment before each moment and the concentration at the current moment are combined into a concentration sequence, and the concentration rise abnormal value at each region at each moment is obtained according to the maximum difference in concentration in the concentration sequence and the degree of disorder of the concentration sequence.

[0050] With the acceleration of industrialization and the continuous improvement of urbanization, air quality issues have become the focus of public attention. When the concentration of pollutants in the air begins to increase, its trend over time often presents a complex and changeable dynamic process. At first, the concentration of pollutants may rise slowly due to a sudden change in a specific activity, including but not limited to an increase in factory exhaust emissions and the accumulation of automobile exhaust; thus forming a local pollution area. Over time, if the pollution source is not controlled in time, the concentration of harmful substances in the air will increase significantly. Therefore, by reducing the concentration of pollutants in the air in the smart park, the concentration of harmful substances in the air will increase significantly. The changes in concentration over time are visualized in the form of graphs, which can quickly determine the air quality conditions in various parts of the park.

[0051] But even in the air When the concentration is under normal conditions, the sensor monitors the air The concentration of air pollutants in the smart park is not constant. Therefore, it is impossible to determine the concentration abnormality through the analysis of concentration fluctuations in a small range. When the concentration of air pollutants in the smart park increases, the sensor monitors The concentration showed a fluctuating upward trend within a period.

[0052] In each area, for each moment, the data collected in the previous cycle including this moment are The concentration of the air is composed into a time series and recorded as a concentration series. The concentration of will increase significantly, and the concentration in the concentration sequence shows a gradual upward trend. Therefore, the concentration increase abnormal value at each moment can be obtained through the difference between the concentrations in the concentration sequence and the concentration change.

[0053] Preferably, in this embodiment, the maximum and minimum values ​​in the concentration sequence are counted, and the concentration range is obtained from them. , represents the concentration sequence corresponding to the tth time in the ith region, represents the maximum value function, represents the minimum function, Indicates the concentration range at the tth moment in the i-th region; at the same time, obtain the probability of each concentration in the concentration sequence in the concentration sequence, and calculate the information entropy of the concentration sequence based on the probability, and obtain the concentration increase anomaly based on the concentration range and information entropy of the concentration sequence , express The information entropy of It represents the abnormal value of concentration rise at the tth time in the ith region.

[0054] So when the air When the concentration starts to increase, the calculated Moment The air in the area The range of concentration becomes larger; at the same time, the information entropy of the concentration sequence is expressed as a degree of chaos, and the greater the uncertainty of the concentration sequence, the greater the concentration sequence is, that is, the information entropy The larger the value, the further the calculated The larger the value, the more Concentration change information is relatively rich, The concentration increases as the outlier value increases.

[0055] At this point, the concentration increase anomaly values ​​at each time in each area are obtained.

[0056] Step S003, preset The standard limit is based on each area at each time The concentration and The difference between the standard limit values ​​and the abnormal concentration rise value of each area at each moment are combined to obtain the abnormal correction value of each area at each moment.

[0057] In the above process, only When the concentration of of the region If the concentration of air continues to be in a high pollution state within a period, the calculated concentration increase abnormal value will be at a lower level; or When the concentration gradually decreases, the calculated concentration increase anomaly will be at a higher level.

[0058] Therefore, the abnormal increase in concentration alone is not enough to explain the time The concentration of air pollution is the real pollution status of the air. The concentration of The abnormal value of concentration increase is corrected according to the standard limit to obtain the abnormal correction value.

[0059] Preferably, in this embodiment, the abnormal correction value is , Indicates Moment Air in the area concentration; Indicates that the air in the smart park The standard limit of Indicates Moment The abnormal correction value of the area. The standard limit value is The national standard for concentration is 75 micrograms per cubic meter.

[0060] In the above embodiment, when the air Moment Air in the area The greater the concentration of The larger the value, the smaller the calculated value. The smaller the air in the smart park, the The concentration rise caused by the concentration remaining at a high level for a long time is at a low level. Amplify it and analyze the air in the smart park The concentration increase caused by the gradual decrease of the concentration is at a higher level, which can be Reduce it. Make the final calculated abnormal correction value Better response Moment Air in the area concentration condition.

[0061] So far, the anomaly correction values ​​of different areas at different times have been obtained.

[0062] Step S004: for each region at the same time, the neighboring region of each region is obtained by using the K-nearest neighbor algorithm; the distance between each region and the neighboring region is used as the weight to compare the neighboring region. The concentration weights are then combined with the anomaly correction value and The concentration of is used to obtain the abnormal impact value of each area.

[0063] When the air When the concentration increases, air pollution has a significant impact on areas near its source. Atmospheric conditions (such as wind) have a greater impact on the spread of pollutants. Strong winds cause air pollutants to spread faster, while a windless environment causes pollutants to spread more slowly in the air. This also shows that the concentrations of air pollutants monitored at the same time in different areas of the smart park are very different. The concentration is high, and the air in places far away from the pollution source is The concentration is low. Therefore, when the air in a certain area of ​​the smart park is When the concentration of The concentration of The degree of impact of abnormal concentration on its neighboring areas.

[0064] For different areas in the smart park obtained at the same time The K-nearest neighbor algorithm selects neighboring regions by the Euclidean distance between different regions. For each region, the distance between the region and all other regions is calculated, and the absolute values ​​of all distances are sorted from small to large. The first K regions are selected as neighboring regions. In this embodiment, the value of K is 12. The distance is the Euclidean distance between the coordinates of the center points of the regions. The abnormal impact value is obtained by combining the concentration weights of different regions with the abnormal correction values.

[0065] Preferably, in this embodiment, the abnormal impact value is , represents the Euclidean distance between the ith region and its jth neighboring region, represents the jth neighboring region of the ith region at the tth time The concentration of represents the i-th region at the t-th time The concentration of represents the number of neighboring regions, Represents the abnormal impact value of the i-th region at the t-th moment.

[0066] According to the above embodiment, when the first The area in In the air at all times When the concentration of increases, the calculated abnormal correction value of the air in different areas at different times is larger; when the air When the concentration of pollutants increases, due to the diffusion effect of air, the air near the pollution source The concentration will also increase accordingly. The closer the area is to the pollution source, the greater the impact will be. The larger the value, the greater the abnormal impact value of the region calculated.

[0067] So far, the abnormal impact values ​​of different areas at different times have been obtained.

[0068] Step S005, the abnormal impact values ​​of all areas at the same time are formed into a set, the discrete situation of the abnormal impact value is obtained according to the variance and mean of the abnormal impact value in the set, and the number of divided intervals is obtained based on the discrete situation and the number of abnormal impact values ​​in the set; the HBOS algorithm is used to monitor the abnormality of each area according to the number of divided intervals, and the abnormality is visualized and managed.

[0069] According to the abnormal impact values ​​of different areas at different times obtained in the above steps, the The degree of air pollution is magnified, and the abnormal impact value can be used to more accurately judge the air pollution in different areas of the smart park at different times. abnormal concentration, avoiding direct The concentration of air is analyzed but the abnormal air quality monitoring is inaccurate due to air mobility.

[0070] Through the HBOS algorithm, The concentration of the abnormal concentration is monitored. Since the collected concentration is sorted from large to small when using the HBOS algorithm, the concentration needs to be divided. The number of divided intervals is directly related to the monitoring accuracy of the abnormal concentration. Therefore, the number of divided intervals needs to be obtained through the abnormal impact value. The abnormal impact values ​​of all areas at the same time form a set, which is recorded as the abnormal impact set.

[0071] The number of partition intervals is obtained through the abnormal impact value and quantity in the abnormal impact set.

[0072] Preferably, in this embodiment, the discrete situation of the abnormal impact set is first determined , represents the abnormal impact set at the tth moment, represents the mean of all abnormal impact values ​​in the abnormal impact set at the tth moment, Represents the standard deviation of all abnormal impact values ​​in the abnormal impact set at the tth moment, Indicates the discrete situation of the abnormal impact set at the tth moment. When the discrete situation at the tth moment is less than 1, the discrete situation remains unchanged and is recorded as the standard discrete situation; if the discrete situation is greater than or equal to 1, the discrete situation is modified to the number 1 and recorded as the standard discrete situation. Get the number of partition intervals based on the discrete situation and quantity , represents the standard discrete case of the set affected by the anomaly, Indicates the number of abnormal impact values ​​in the abnormal impact set. represents the rounding function, Indicates the number of partition intervals in the HBOS algorithm.

[0073] Based on the above steps, the number of divided intervals is obtained for each moment; according to all the divided intervals, the number of sample points in each divided interval is counted to generate a histogram, and the concentration anomaly score of each area can be calculated according to the histogram. The specific calculation process can refer to the HBOS algorithm to obtain the concentration anomaly score of each area. In this embodiment, the abnormal threshold is set , the abnormal threshold can also be set to 0.75, and the concentration with a concentration abnormality score greater than the abnormal threshold is recorded as a concentration abnormality; finally, the concentration of air in the smart park can be obtained. The abnormal concentration data set of the air in the smart park is as follows: Figure 2 shown.

[0074] Based on the air in the smart park The concentration anomaly data set is combined with Python and Bokeh visualization tools to realize the visualization of abnormal concentration. This helps managers to Focus on areas with abnormal concentrations and Targeted management measures are taken in abnormally polluted areas. The use of Bokeh visualization tools is a well-known technology, and the specific process will not be repeated here.

[0075] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0076] 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 visual management method for smart parks based on digital twins, characterized in that: The method comprises the following steps: Collect data from different areas at different times concentration; For each area at each moment, the concentration at the preset moment before each moment and the concentration at the current moment are combined into a concentration sequence, and the concentration increase abnormal value of each area at each moment is obtained according to the maximum difference in concentration in the concentration sequence and the degree of disorder of the concentration sequence; Presets The standard limit is based on each area at each time The concentration and The difference between the standard limit and the abnormal concentration rise value of each area at each moment is combined to obtain the abnormal correction value of each area at each moment; For each region at the same time, the K-nearest neighbor algorithm is used to obtain the neighboring regions of each region; the distance between each region and the neighboring region is used as the weight and the neighboring region is used as the weight. The concentration weights are then combined with the anomaly correction value and The concentration of each area is used to obtain the abnormal impact value of each area; The abnormal impact values ​​of all regions at the same time are combined into a set. The discrete situation of the abnormal impact value is obtained according to the variance and mean of the abnormal impact value in the set. The number of divided intervals is obtained based on the discrete situation and the number of abnormal impact values ​​in the set. The HBOS algorithm is used to monitor the abnormality of each region according to the number of divided intervals, and the abnormality is visualized for management. The method for obtaining the concentration rise abnormal value of each region at each moment according to the maximum difference of the concentration in the concentration sequence and the degree of disorder of the concentration sequence is: Get the concentration range in the concentration series; Calculate the information entropy of the concentration sequence, and the expression of abnormal concentration increase is: , express The information entropy of represents the abnormal value of concentration increase at the tth time in the i-th region, represents the concentration range at the tth time in the ith region, represents the concentration sequence corresponding to the tth time in the i-th region; The expression of the abnormal correction value is: , Indicates Moment Air in the area concentration; Indicates that the air in the smart park The standard limit of Indicates Moment The anomaly correction value of the area, represents the abnormal value of concentration increase at the tth time in the ith region; The expression of the abnormal impact value is: , represents the Euclidean distance between the ith region and its jth neighboring region, represents the jth neighboring region of the ith region at the tth time The concentration of represents the i-th region at the t-th time The concentration of represents the number of neighboring regions, represents the abnormal impact value of the ith region at the tth moment, Indicates Moment Anomaly correction value for each region; The method for obtaining the number of partition intervals based on the discrete situation and the number of abnormal impact values ​​in the set is: , represents the standard discrete case of the set affected by the anomaly, Indicates the number of abnormal impact values ​​in the abnormal impact set. represents the rounding function, Indicates the number of partition intervals in the HBOS algorithm.

2. The method for visual management of a smart park based on digital twins according to claim 1, characterized in that: The collection of different areas at different times The concentration method is: Divide the smart park into areas of equal size and place The concentration monitor collects data once every preset time, and the preset number of times is regarded as a cycle.

3. The method for visual management of a smart park based on digital twins according to claim 2, characterized in that: The method of forming a concentration sequence by combining the concentration at a preset time before each time point and the concentration at the current time point is as follows: For each moment in a region, the data collected in the previous cycle including that moment are The concentration of composes a time series and is recorded as concentration series.

4. The method for visual management of a smart park based on digital twins according to claim 1, characterized in that: The method of obtaining the neighboring regions of each region by using the K-nearest neighbor algorithm for each region at the same time is: Calculate the distance between each area and all other areas at the same time, sort the absolute values ​​of all distances from small to large, and select a preset number of areas as neighboring areas. The distance is the Euclidean distance between the coordinates of the center points of the areas.

5. The method for visual management of a smart park based on digital twins according to claim 1, characterized in that: The method of forming a set of abnormal influence values ​​of all regions at the same time and obtaining the discrete situation of the abnormal influence values ​​according to the variance and mean of the abnormal influence values ​​in the set is: The abnormal impact values ​​of all regions at the same time are combined into a set and recorded as the abnormal impact set; Let the ratio of the standard deviation and the mean of all abnormal impact values ​​in the abnormal impact set be the discrete case of the abnormal impact set.

6. The method for visual management of a smart park based on digital twins according to claim 1, characterized in that: The method of using the HBOS algorithm to monitor the abnormality of each area by dividing the number of intervals is: The number of sample points in each divided interval is counted to generate a histogram, and the concentration anomaly score of each area is obtained through the HBOS algorithm. The anomaly threshold is preset, and the concentration with a concentration anomaly score greater than the anomaly threshold is recorded as a concentration anomaly.

Images