AI-based dust and object automatic identification tracking system intelligent platform
By introducing AI technology and diversified data analysis into the dust identification and tracking system, the problem of reducing detection accuracy after long-term use of dust sensors is solved, and accurate analysis of dust concentration and support for dust removal operations are achieved.
Patent Information
- Application Number
- CN202411895567.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-22
- Publication Date
- 2025-04-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the existing dust identification and tracking system, dust sensors will accumulate dust after long-term use, resulting in a decrease in detection accuracy and unable to provide accurate dust concentration analysis results.
Design an intelligent platform for automatic dust and object recognition and tracking system based on AI. Through the area division module, data recording module, data evaluation module, data analysis module, image acquisition module, dust identification module and tracking module, combining dust sensors and image data, diversified data analysis and image feature extraction are carried out to judge the operating status of the dust sensor and perform real-time calibration.
Through the combination of multiple data analysis and image data, the operating status of the dust sensor can be accurately judged, avoid decreasing detection accuracy, improve the accuracy of dust detection and analysis results in the factory, and ensure the normal progress of dust removal operations.
Smart Images
Figure CN119827365A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of dust identification and tracking, and in particular to an AI-based intelligent platform for automatic dust and object identification and tracking systems. Background Art
[0002] With the continuous development of science and technology, the energy industry is in short supply, and chemical plants and other energy plants will generate a lot of dust during production and processing. The dust and harmful substances in the air will cause serious health hazards if inhaled into the human body. Therefore, in daily work, the dust will be identified through the dust identification and tracking system, and the dust removal equipment will be controlled to perform unmanned dust removal operations based on the identification results.
[0003] In the prior art, common dust identification and tracking systems usually include an identification module and a tracking module. The identification module usually collects dust concentrations at various locations through dust sensors, and analyzes the dust concentrations at different locations in the factory in combination with image information data captured by the camera. The tracking module then records the dust diffusion trajectory and concentration changes to provide data support for subsequent governance, thereby ensuring the normal progress of subsequent dust removal operations.
[0004] The common dust identification and tracking systems in the prior art are generally based on the data collection of dust sensors and cameras to analyze the dust concentration and whether there is dust at different locations in the factory. However, in order to ensure that the dust sensor is in direct contact with the dust during use, a large amount of dust will accumulate inside and on the surface after long-term use, which will cause deviations in its detection accuracy. As a result, the accuracy of the dust detection and analysis results in the factory is reduced, and it is impossible to provide accurate data support for subsequent governance, resulting in the inability to carry out dust removal operations in a timely manner, affecting human safety. Summary of the invention
[0005] The purpose of this invention is to provide an AI-based intelligent platform for dust and object automatic identification and tracking system to solve the following technical problems:
[0006] How to improve the accuracy of dust concentration detection and analysis results within the factory.
[0007] The purpose of the present invention can be achieved through the following technical solutions:
[0008] An AI-based dust and object automatic identification and tracking system intelligent platform, the system comprising:
[0009] The area division module is used to divide the factory area into multiple areas of equal size, and a dust sensor is set at the center of each area;
[0010] A data recording module is used to record the operating status data of the dust sensor each time it is used;
[0011] The data evaluation module is used to calculate the operating status impact index of the dust sensor at that time in combination with the operating status data of the dust sensor at that time recorded by the data recording module, and analyze the data collection accuracy at that time according to the calculation result;
[0012] The data analysis module is used to analyze the volatility of the dust sensor's operating status impact index in combination with the dust sensor's historical operating status data recorded by the data recording module, and further analyze the status of the dust sensor;
[0013] An image acquisition module is used to collect image data of various areas in the field area and extract dust features in the image data;
[0014] The dust identification module is used to analyze the dust concentration and status of different areas in the current factory area by combining the data information collected by the dust sensor with the information collected by the graphic acquisition module when judging that the dust sensor is in good condition;
[0015] The tracking module is used to record the dust diffusion trajectory and concentration change data based on AI technology and the dust concentration analysis results in different areas of the factory, and perform unmanned dust removal operations.
[0016] Furthermore, the data recorded by the data recording module includes the current intensity, data recognition speed, temperature and data transmission speed of the dust sensor each time it is used.
[0017] Furthermore, the evaluation process of the data evaluation module includes:
[0018] By formula Calculate the operating status impact index μ of the dust sensor when it is used for the i-th time i ;
[0019] Where i is any use of the dust sensor, sd i is the data recognition speed of the dust sensor when it is used for the i-th time, sd y is the preset data recognition speed, cs i is the data transmission speed of the dust sensor when it is used for the i-th time, cs y is the preset data transmission speed, wd i is the average temperature of the dust sensor when it is used for the i-th time, wd y is the preset average temperature, a is any time point in any use of the dust sensor, b is the total number of time points in any use of the dust sensor, dl ai is the current intensity at the ath time point when the dust sensor is used for the i-th time, For all dl ai The average value of .
[0020] Furthermore, the evaluation process of the data evaluation module also includes:
[0021] By calculating the operating status influence index μ when the dust sensor is used for the i-th time i The preset operating status impact index threshold μ 01 Make a comparison;
[0022] If μ i ≥μ 01 , the system determines that the dust sensor is not in good operating condition during this use, and the data obtained by its detection has errors, and issues a warning;
[0023] If μ i <μ 01 ,The system judges that the dust sensor is in good operating condition during this use, and there is no error in the data obtained by its detection, and further analyzes it in combination with historical data.
[0024] Furthermore, the analysis process of the data analysis module includes:
[0025] The operating state influence index of the dust sensor obtained by continuous detection by the data evaluation module each time it is used is used to establish the operating state influence index change curve μ(t);
[0026] And through the formula Calculate and obtain the discrete coefficient δ of the dust sensor operation status influence index;
[0027] Where n is the total number of times the dust sensor has been used since the last calibration. For all μ i The standard value of t1 is the first use of the dust sensor after the last calibration, and t2 is the last use of the dust sensor after the last calibration.
[0028] Furthermore, the analysis process of the data analysis module also includes:
[0029] By comparing the discrete coefficient δ of the dust sensor operating status influence index with the preset discrete coefficient threshold δ 01 Make a comparison;
[0030] If δ ≥ δ 01 ,The system determines that the operating status of the dust sensor since the last calibration has a large volatility, indicating that the operating status of the dust sensor is at risk, which means that its detection accuracy is affected, and issues a warning;
[0031] If δ<δ 01,The system judges that the operating status of the dust sensor since the last calibration has a small volatility in the ,impact index, which indicates that the operating status of the dust sensor does not pose ,risks, which means that its detection accuracy is good and dust ,identification and tracking operations can be performed.
[0032] Furthermore, the identification process of the dust identification module includes:
[0033] By formula Calculate the dust impact coefficient ρ in the tth area during a dust identification operation t ;
[0034] Among them, n t is the dust concentration in the tth area during a dust identification operation, nd y is the preset dust concentration, wy t is the area of fog shadow in the tth area during a dust identification operation, wy y is the preset fog area, kel t is the visible particle content in the tth area during a dust identification operation, kel b for kel t is the standard value of , x1 and x2 are weight coefficients.
[0035] Furthermore, the identification process of the dust identification module also includes:
[0036] The dust impact coefficient ρ in the tth area during a dust identification operation is calculated t Compare with the preset dust impact coefficient threshold range [ρ1, ρ2];
[0037] If t <ρ1, the system determines that the dust content in the area is low and does not issue a cleaning instruction;
[0038] If t ∈[ρ1, ρ2], the system determines that the dust content in the area is medium and gives normal cleaning instructions;
[0039] If t >ρ2, the system determines that the dust content in the area is high, marks it as the source of dust pollution, and gives instructions for focused cleaning.
[0040] Beneficial effects of the present invention:
[0041] (1) The present invention records the current operating status data and historical operating status data of the dust sensor through a data recording module, and analyzes the status of the dust sensor through a data evaluation module and a data analysis module respectively. Based on diversified data and multiple analyses, the present invention can make an accurate judgment on the current operating status of the dust sensor, and timely calibrate and clean the dust sensor with poor operating status, thereby ensuring that the dust sensor performs dust identification operations in a good operating state, thereby avoiding deviations in its detection accuracy and improving the accuracy of dust detection and analysis results in the factory.
[0042] (2) The present invention calculates the operating state influence index μ of the dust sensor when it is used for the i-th time i The preset operating status impact index threshold μ 01 By comparing, through this comparison method, we can first analyze the operating status of the dust sensor during this use, and further judge whether there is any error in the data obtained by the detection. By judging the operating status of the dust sensor, we can avoid performing dust identification operations when the dust sensor is in a poor operating state, thereby avoiding the situation where the detection accuracy is reduced due to its poor operating state, affecting the accuracy of the dust detection and analysis results in the factory.
[0043] (3) The present invention compares the discrete coefficient δ of the dust sensor operating state influence index with the preset discrete coefficient threshold δ 01 Through this comparison method, the volatility of the dust sensor's operating status since the last calibration can be analyzed, and further accurate judgments can be made on whether there are risks in the dust sensor's operating status and whether its detection accuracy is affected. Further analysis and judgment can be made after analyzing the dust sensor's operating status based on the data evaluation module, thereby avoiding dust identification operations when the dust sensor is in poor operating condition to ensure its detection accuracy.
[0044] (4) The present invention obtains dust concentration data in different areas through dust sensor detection. Since the data is obtained based on the detection of dust sensors in good operating condition, the data has high accuracy and reliability. In addition, the dust characteristic data in the image data and the operating condition influence index μ of the dust sensor when it is used for the i-th time are combined. i The calculation results can be corrected with data such as the discrete coefficient δ of the dust sensor operating status influence index to ensure the accuracy of the calculation results, thereby truly reflecting the dust pollution conditions in different areas within the factory and providing data support for subsequent governance to ensure the normal progress of subsequent dust removal operations.
[0045] (5) The present invention calculates the dust influence coefficient ρ in the tth area during a dust identification operation.t By comparing with the preset dust impact coefficient threshold range [ρ1, ρ2], the dust content in different areas can be accurately judged. According to the dust content in different areas, the location of dust can be tracked and the source of pollution can be traced, thereby providing accurate data support for subsequent governance and ensuring timely implementation of dust removal operations. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] The present invention will be further described below in conjunction with the accompanying drawings.
[0047] Figure 1 This is a schematic block diagram of an AI-based dust and object automatic identification and tracking system intelligent platform in the present invention. DETAILED DESCRIPTION
[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0049] See also Figure 1 As shown, in one embodiment, the present application provides an AI-based dust and object automatic identification and tracking system intelligent platform, the system comprising:
[0050] The area division module is used to divide the factory area into multiple areas of equal size, and a dust sensor is set at the center of each area;
[0051] A data recording module is used to record the operating status data of the dust sensor each time it is used;
[0052] The data evaluation module is used to calculate the operating status impact index of the dust sensor at that time in combination with the operating status data of the dust sensor at that time recorded by the data recording module, and analyze the data collection accuracy at that time according to the calculation result;
[0053] The data analysis module is used to analyze the volatility of the dust sensor's operating status impact index in combination with the dust sensor's historical operating status data recorded by the data recording module, and further analyze the status of the dust sensor;
[0054] An image acquisition module is used to collect image data of various areas in the field area and extract dust features in the image data;
[0055] The dust identification module is used to analyze the dust concentration and status of different areas in the current factory area by combining the data information collected by the dust sensor with the information collected by the graphic acquisition module when judging that the dust sensor is in good condition;
[0056] The tracking module is used to record the dust diffusion trajectory and concentration change data based on AI technology combined with the dust concentration analysis results in different areas of the factory, and perform unmanned dust removal operations;
[0057] Through the above technical solution, this embodiment provides a region division module for dividing the factory area into a plurality of regions of the same size, and a dust sensor is arranged at the center of each region, so as to facilitate the subsequent observation of the diffusion trajectory of dust. When identifying dust in the factory area, the operating status data of the dust sensor at the time of use is first recorded by the data recording module, and the operating status data of the dust sensor at that time recorded by the data recording module is combined with the data evaluation module to calculate the operating status impact index of the dust sensor at that time, and the data collection accuracy of the data is analyzed according to the calculation result, and the dust sensor history data recorded by the data recording module is combined with the data analysis module to calculate the operating status impact index of the dust sensor at that time. The historical operation status data is used to analyze the volatility of the dust sensor's operation status index, and the status of the dust sensor is further analyzed. When the dust sensor is judged to be in good condition, the image data of each area of the field area can be collected through the image acquisition module, and the dust features in the image data can be extracted. The dust recognition module combines the data information collected by the dust sensor with the information collected by the graphic acquisition module to analyze the dust concentration and status of different areas in the current factory area. Finally, the tracking module combines the dust concentration analysis results of different areas in the factory area based on AI technology, records the dust diffusion trajectory and concentration change data, and performs unmanned dust removal operations.
[0058] Through the above technical scheme, the current operating status data and historical operating status data of the dust sensor are recorded through the data recording module, and the status of the dust sensor is analyzed by the data evaluation module and the data analysis module respectively. Based on diversified data and multiple analyses, an accurate judgment can be made on the current operating status of the dust sensor, and dust sensors with poor operating status can be calibrated and cleaned in time, so as to ensure that the dust sensor performs dust identification operations in a good operating state, thereby avoiding deviations in its detection accuracy and improving the accuracy of dust detection and analysis results in the factory.
[0059] The data recorded by the data recording module includes the current intensity, data recognition speed, temperature and data transmission speed of the dust sensor each time it is used;
[0060] Through the above technical solution, this embodiment provides data recorded by the data recording module, including the current intensity, data recognition speed, temperature and data transmission speed of the dust sensor each time it is used. The data reflects the operating parameters of the dust sensor when in use. By combining the parameters, the state of the dust sensor when in use can be analyzed, thereby providing strong data support for subsequent analysis and judgment of the detection sensitivity of the dust sensor, thereby improving the accuracy of the analysis results.
[0061] The evaluation process of the data evaluation module includes:
[0062] By formula Calculate the operating status impact index μ of the dust sensor when it is used for the i-th time i ;
[0063] Where i is any use of the dust sensor, sd i is the data recognition speed of the dust sensor when it is used for the i-th time, sd y is the preset data recognition speed, cs i is the data transmission speed of the dust sensor when it is used for the i-th time, cs y is the preset data transmission speed, wd i is the average temperature of the dust sensor when it is used for the i-th time, wd y is the preset average temperature, a is any time point in any use of the dust sensor, b is the total number of time points in any use of the dust sensor, dl ai is the current intensity at the ath time point when the dust sensor is used for the i-th time, For all dl ai The average value of
[0064] Through the above technical solution, this embodiment provides the operating state influence index μ of the dust sensor when it is used for the i-th time i , can be obtained by formula
[0065]
[0066] Calculated, where the formula The current fluctuation value of the dust sensor when it is used for the i-th time can be calculated. Therefore, it is obvious that when the data recognition speed and data transmission speed of the dust sensor are slower and the average temperature and current fluctuation value are higher, the operating state influence index μ of the dust sensor when it is used for the i-th time is higher. iThe larger the value is, the more abnormal the current state of the dust sensor is. Its detection sensitivity for dust may be reduced, thus affecting the accuracy of the dust detection and analysis results in the factory. On the contrary, when the data recognition speed and data transmission speed of the dust sensor are used for the i-th time, and the average temperature and current fluctuation values are lower, the operating state influence index μ of the dust sensor for the i-th use is i The smaller the value, the more normal the current state of the dust sensor is. The dust detection sensitivity will not change too much and will not affect the accuracy of the dust detection and analysis results in the factory.
[0067] Through this calculation method, after combining the operating parameter data of the dust sensor when it is used for the first time, the operating state influence index μ of the dust sensor when it is used for the i-th time can be calculated. i Make an accurate judgment. The data reflects the operating status of the dust sensor during the current detection, thereby providing accurate data for subsequent judgment on whether there is an abnormality in the detection accuracy of the dust sensor at that time, thereby ensuring the accuracy of the judgment result.
[0068] The evaluation process of the data evaluation module also includes:
[0069] By calculating the operating status influence index μ when the dust sensor is used for the i-th time i The preset operating status impact index threshold μ 01 Make a comparison;
[0070] If μ i ≥μ 01 , the system determines that the dust sensor is not in good operating condition during this use, and the data obtained by its detection has errors, and issues a warning;
[0071] If μ i <μ 01 , the system judges that the dust sensor is in good operating condition during this use, and the data obtained by its detection does not contain errors, and further analyzes it in combination with historical data;
[0072] Through the above technical solution, this embodiment calculates the operating state influence index μ of the dust sensor when it is used for the i-th time i The preset operating status impact index threshold μ 01 By comparing, through this comparison method, we can first analyze the operating status of the dust sensor during this use, and further judge whether there is any error in the data obtained by the detection. By judging the operating status of the dust sensor, we can avoid performing dust identification operations when the dust sensor is in a poor operating state, thereby avoiding the situation where the detection accuracy is reduced due to its poor operating state, affecting the accuracy of the dust detection and analysis results in the factory.
[0073] The analysis process of the data analysis module includes:
[0074] The operating state influence index of the dust sensor obtained by continuous detection by the data evaluation module each time it is used is used to establish the operating state influence index change curve μ(t);
[0075] And through the formula
[0076]
[0077] Calculate and obtain the discrete coefficient δ of the dust sensor operation status influence index;
[0078] Where n is the total number of times the dust sensor has been used since the last calibration. For all μ i The standard value of t1 is the first use of the dust sensor after the last calibration, and t2 is the last use of the dust sensor after the last calibration;
[0079] Through the above technical solution, this embodiment provides a discrete coefficient δ of the dust sensor operating state influence index, which can be calculated by the formula
[0080]
[0081] It is obtained by calculation. Through this calculation method, after combining the historical operating status impact index of the dust sensor and the change in the operating status impact index and other parameters, the discrete coefficient δ of the dust sensor operating status impact index can be accurately calculated. This data reflects the state fluctuation of the dust sensor since the last calibration, thereby providing accurate data for the subsequent risk judgment of the dust sensor's operating status to ensure the accuracy of the judgment result.
[0082] The analysis process of the data analysis module also includes:
[0083] By comparing the discrete coefficient δ of the dust sensor operating status influence index with the preset discrete coefficient threshold δ 01 Make a comparison;
[0084] If δ ≥ δ 01 ,The system determines that the operating status of the dust sensor since the last calibration has a large volatility, indicating that the operating status of the dust sensor is at risk, which means that its detection accuracy is affected, and issues a warning;
[0085] If δ<δ 01 ,The system determines that the dust sensor’s operating status since the last calibration has a small volatility in the impact index, which means that the dust sensor’s operating status does not pose a risk, which means that its detection accuracy is good and dust identification and tracking operations can be performed;
[0086] Through the above technical solution, this embodiment compares the discrete coefficient δ of the dust sensor operation status influence index with the preset discrete coefficient threshold δ 01 Through this comparison method, the volatility of the dust sensor's operating status since the last calibration can be analyzed, and further accurate judgments can be made on whether there are risks in the dust sensor's operating status and whether its detection accuracy is affected. Further analysis and judgment can be made after analyzing the dust sensor's operating status based on the data evaluation module, thereby avoiding dust identification operations when the dust sensor is in poor operating condition to ensure its detection accuracy.
[0087] The identification process of the dust identification module includes:
[0088] By formula
[0089]
[0090] Calculate the dust impact coefficient ρ in the tth area during a dust identification operation t ;
[0091] Among them, n t is the dust concentration in the tth area during a dust identification operation, nd y is the preset dust concentration, wy t is the area of fog shadow in the tth area during a dust identification operation, wy y is the preset fog area, kel t is the visible particle content in the tth area during a dust identification operation, kel b for kel t The standard value of , which can be selected and set according to the allowable error in the empirical data, x1 and x2 are weight coefficients, which are set according to empirical fitting;
[0092] Through the above technical solution, this embodiment provides a dust identification operation in the t-th area of the dust influence coefficient ρ t Obviously, when the dust concentration in the t-th area is higher during a dust identification operation, the fog shadow area and the visible particle content in the t-th area are larger, so the dust influence coefficient ρ in the t-th area during a dust identification operation is t The larger the dust concentration is, the more serious the dust pollution in the area is. On the contrary, when the dust concentration in the tth area is lower during a dust identification operation, the fog shadow area and visible particulate matter content in the tth area are less. Then the dust influence coefficient ρ in the tth area during a dust identification operation is t The smaller it is, the lighter the dust pollution in the area is;
[0093] Through this calculation method, the dust concentration data in different areas is obtained through dust sensor detection. Since the data is obtained based on the detection of dust sensors in good operating condition, the accuracy and reliability of the data are high. By combining the characteristic data of dust in the image data and the operating condition influence index μ of the dust sensor when it is used for the i-th time, i The calculation results can be corrected with data such as the discrete coefficient δ of the dust sensor operating status influence index to ensure the accuracy of the calculation results, thereby truly reflecting the dust pollution conditions in different areas within the factory and providing data support for subsequent governance to ensure the normal progress of subsequent dust removal operations.
[0094] The identification process of the dust identification module also includes:
[0095] The dust impact coefficient ρ in the tth area during a dust identification operation is calculated t Compare with the preset dust impact coefficient threshold range [ρ1, ρ2];
[0096] If t <ρ1, the system determines that the dust content in the area is low and does not issue a cleaning instruction;
[0097] If t ∈[ρ1, ρ2], the system determines that the dust content in the area is medium and gives normal cleaning instructions;
[0098] If t >ρ2, the system determines that the dust content in the area is high, marks it as the source of dust pollution, and gives instructions for focused cleaning;
[0099] Through the above technical solution, this embodiment calculates the dust influence coefficient ρ in the tth area during a dust identification operation. t By comparing with the preset dust impact coefficient threshold range [ρ1, ρ2], the dust content in different areas can be accurately judged. According to the dust content in different areas, the location of dust can be tracked and the source of pollution can be traced, thereby providing accurate data support for subsequent governance and ensuring timely implementation of dust removal operations.
[0100] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.
Claims
1. An AI-based dust and object automatic identification and tracking system intelligent platform, characterized in that: The system comprises: The area division module is used to divide the factory area into multiple areas of equal size, and a dust sensor is set at the center of each area; A data recording module is used to record the operating status data of the dust sensor each time it is used; The data evaluation module is used to calculate the operating status impact index of the dust sensor at that time in combination with the operating status data of the dust sensor at that time recorded by the data recording module, and analyze the data collection accuracy at that time according to the calculation result; The data analysis module is used to analyze the volatility of the dust sensor's operating status impact index in combination with the dust sensor's historical operating status data recorded by the data recording module, and further analyze the status of the dust sensor; An image acquisition module is used to collect image data of various areas in the field area and extract dust features in the image data; The dust identification module is used to analyze the dust concentration and status of different areas in the current factory area by combining the data information collected by the dust sensor with the information collected by the graphic acquisition module when judging that the dust sensor is in good condition; The tracking module is used to record the dust diffusion trajectory and concentration change data based on AI technology and the dust concentration analysis results in different areas of the factory, and perform unmanned dust removal operations.
2. According to claim 1, an AI-based dust and object automatic identification and tracking system intelligent platform is characterized in that: The data recorded by the data recording module includes the current intensity, data recognition speed, temperature and data transmission speed of the dust sensor each time it is used.
3. According to claim 2, the AI-based dust and object automatic identification and tracking system intelligent platform is characterized in that: The evaluation process of the data evaluation module includes: By formula Calculate the operating status influence index μ of the dust sensor when it is used for the i-th time i ; Where i is any use of the dust sensor, sd i is the data recognition speed of the dust sensor when it is used for the i-th time, sd y is the preset data recognition speed, cs i is the data transmission speed of the dust sensor when it is used for the i-th time, cs y is the preset data transmission speed, wd i is the average temperature of the dust sensor when it is used for the i-th time, wd y is the preset average temperature, a is any time point in any use of the dust sensor, b is the total number of time points in any use of the dust sensor, dl ai is the current intensity at the ath time point when the dust sensor is used for the i-th time, For all dl ai The average value of .
4. According to claim 3, the AI-based dust and object automatic identification and tracking system intelligent platform is characterized in that: The evaluation process of the data evaluation module also includes: By calculating the operating status influence index μ when the dust sensor is used for the i-th time i The preset operating status impact index threshold μ 01 Make a comparison; If μ i ≥μ 01 , the system determines that the dust sensor is not in good operating condition during this use, and the data obtained by its detection has errors, and issues a warning; If μ i <μ 01 ,The system judges that the dust sensor is in good operating condition during this use, and there is no error in the data obtained by its detection, and further analyzes it in combination with historical data.
5. According to claim 4, the AI-based dust and object automatic identification and tracking system intelligent platform is characterized in that: The analysis process of the data analysis module includes: The operating state influence index of the dust sensor obtained by continuous detection by the data evaluation module each time it is used is used to establish the operating state influence index change curve μ(t); And through the formula Calculate and obtain the discrete coefficient δ of the dust sensor operation status influence index; Where n is the total number of times the dust sensor has been used since the last calibration. For all μ i The standard value of t1 is the first use of the dust sensor after the last calibration, and t2 is the last use of the dust sensor after the last calibration.
6. The AI-based dust and object automatic identification and tracking system intelligent platform according to claim 5 is characterized in that: The analysis process of the data analysis module also includes: By comparing the discrete coefficient δ of the dust sensor operating status influence index with the preset discrete coefficient threshold δ 01 Make a comparison; If δ ≥ δ 01 ,The system determines that the operating status of the dust sensor since the last calibration has a large volatility, indicating that the operating status of the dust sensor is at risk, which means that its detection accuracy is affected, and issues a warning; If δ<δ 01 ,The system judges that the operating status of the dust sensor since the last calibration has a small volatility in the ,impact index, which indicates that the operating status of the dust sensor does not pose ,risks, which means that its detection accuracy is good and dust ,identification and tracking operations can be performed.
7. The AI-based dust and object automatic identification and tracking system intelligent platform according to claim 6 is characterized in that: The identification process of the dust identification module includes: By formula Calculate the dust impact coefficient ρ in the tth area during a dust identification operation t ; Among them, n t is the dust concentration in the tth area during a dust identification operation, nd y is the preset dust concentration, wy t is the fog shadow area in the tth area during a dust identification operation, wy y is the preset fog area, kel t is the visible particle content in the tth area during a dust identification operation, kel b for kel t is the standard value of , and x1 and x2 are weight coefficients.
8. The AI-based dust and object automatic identification and tracking system intelligent platform according to claim 7 is characterized in that: The identification process of the dust identification module also includes: The dust impact coefficient ρ in the tth area during a dust identification operation is calculated t Compare with the preset dust impact coefficient threshold interval [ρ1, ρ2]; If t <ρ1, the system determines that the dust content in the area is low and does not issue a cleaning instruction; If t ∈[ρ1, ρ2], the system determines that the dust content in the area is medium and gives normal cleaning instructions; If t >ρ2, the system determines that the dust content in the area is high, marks it as the source of dust pollution, and gives instructions for focused cleaning.