Regional dynamic risk perception method and system

Abstract

The invention relates to the technical field of security perception, in particular to a regional dynamic risk perception method and system. A regional dynamic risk perception method comprises the following steps: S1, after collecting the sound degree of a target region, dividing the sound degree of the target region to obtain a sub-regional sound degree table; s2, obtaining the sound degree of the target area, and comparing the sound degree with the regional sound degree table to obtain the sound deviation degree of the target area; s3, analyzing the sound degree and the sound deviation degree in the target area to obtain the risk degree in the target area; and S4, collecting N risk detection levels in the area, and performing comprehensive analysis on the N risk detection levels and the sound safety level to obtain a comprehensive risk distribution diagram. According to the invention, the sound risk degree in the target area is obtained through the sound level type, the total deviation number and the sound degree of the target area, and comprehensive analysis is carried out through the danger levels of different danger data.

Description

Technical Field

[0001] The present invention relates to the field of security perception technology, and in particular to a regional dynamic risk perception method and system. Background Art

[0002] An electronic fence is a system that establishes a virtual boundary in a designated area to monitor and alarm objects or people that enter or cross the area. In some complex environments, such as light conditions, obstructions, and signal interference, the perception accuracy of traditional electronic fence systems may be reduced. Therefore, electronic fence systems that rely solely on a single perception method are no longer able to meet market demand. In order to solve these problems, researchers have begun to explore the combination of multiple perception technologies, such as sound-assisted perception, to improve the reliability and perception accuracy of electronic fence systems in complex environments.

[0003] With the rapid development of artificial intelligence and big data technology, regional risk perception through sound analysis has gradually become an effective supplementary means. When precise sound collection devices are not installed in the factory, it will not be possible to accurately analyze the sound data through deep learning, thereby filtering out the impact of background sounds. For example, the level of sound collected on rainy days will generally increase, thereby affecting the perception of risk levels. Therefore, comprehensive analysis of various sources of danger has become increasingly important. Summary of the invention

[0004] In order to overcome the shortcoming of low accuracy of regional risk perception, the present invention provides a regional dynamic risk perception method and system.

[0005] The technical solution is: a regional dynamic risk perception method, comprising the following steps:

[0006] S1: After collecting the sound levels of the target area, the sound levels of the target area are divided to obtain a regional sound level table;

[0007] S2: Obtain the sound level of the target area, compare it with the regional sound level table, and obtain the sound deviation of the target area;

[0008] S3: Analyze the sound level and sound deviation in the target area to obtain the risk level in the target area;

[0009] S4: Collect N risk detection levels in the area, and after comprehensive analysis with the sound safety level, obtain a comprehensive risk distribution map.

[0010] Preferably, after collecting the sound levels of the target area, the sound levels of the target area are divided to obtain a regional sound level table, including: performing data cleaning on the sound levels in the target area, obtaining the standard sound levels in the target area through a sound level calculation formula, and obtaining a regional sound level table based on the standard sound levels in each target area, wherein the sound level calculation formula is:

[0011]

[0012] Where L total is the total sound pressure level in the target area, L i is the sound pressure level of each noise source at distance d0, d i is the distance from the noise source i to the observation point, and d0 is the reference distance.

[0013] Preferably, the sound level of the target area is obtained, and it is compared with the regional sound level table to obtain the sound deviation of the target area, including: obtaining the sound level in each target area, and inputting the sound level into a sound deviation formula to obtain the sound deviation within a preset time period in each target area, taking a sound deviation greater than or equal to a preset sound threshold as an abnormal sound value, and taking a sound deviation less than the preset sound threshold as a normal sound value, wherein the sound deviation formula is:

[0014] M=ω|Z 声 -Z 标 |

[0015] Where M is the sound deviation, ω is the adjustment weight of the sound deviation, and Z 声 is the sound level in the target area, Z 标 It is the standard sound level of the sub-region sound level table in the target area.

[0016] Preferably, the sound level and sound deviation in the target area are analyzed to obtain the sound risk level in the target area, including: using the FIFO method to sort the historical sound deviations in the target area within a first preset time period to obtain a deviation sequence, and analyzing based on the historical sound deviations to obtain a sound level sequence, and obtaining the sound risk level in the target area according to the sound level sequence and the deviation sequence.

[0017] Preferably, the use of the FIFO method to sort the historical sound deviations within the first preset time period of the target area to obtain a deviation sequence includes: comparing the sound level in the target area with all standard sound levels in the regional sound level table, obtaining the deviation between the sound intensity level in the target area and each sound intensity level, taking the minimum deviation as the deviation in the target area, and using the FIFO method to sort the historical sound deviations to obtain a deviation sequence.

[0018] Preferably, the sound level sequence is obtained by analyzing the historical sound deviation, and before the sound risk level in the target area is obtained according to the sound level sequence and the deviation sequence, it includes: obtaining a first preset number of smooth sound level sequences with smooth sound levels in the target area, and performing weighted averaging according to each smooth sound level to obtain a historical benchmark sound level, and re-obtaining the deviation sequence after judging based on the latest sound level sequence and the historical benchmark sound level.

[0019] Preferably, the method of re-obtaining the deviation sequence after judging based on the latest sound level sequence and the historical benchmark sound level includes: when the number of sound levels in the latest sound level sequence that is greater than or equal to the historical benchmark sound level is greater than a second preset number, the sound levels in the latest sound level sequence are adjusted to re-obtain the deviation sequence, wherein the adjustment is to use the difference between the latest sound level and the historical benchmark sound level as the adjusted sound value, and to recalculate the deviation sequence after subtracting the latest sound level sequence according to the adjusted sound value.

[0020] Preferably, the sound level sequence is obtained by analyzing the historical sound deviation, and the sound risk level in the target area is obtained according to the sound level sequence and the deviation sequence, including: obtaining the sound level in the target area based on the minimum deviation, and obtaining the sound level sequence using the FIFO method according to the sound level of each minimum deviation in the target area, taking the total number of abnormal sound values ​​in the deviation sequence as the total number of deviations, taking the number of types of normal sound values ​​in the sound level sequence as the number of sound types, and obtaining the sound risk level in the target area according to the risk value formula, wherein the risk value formula is:

[0021]

[0022] Where S is the risk value, ω, β is the preset characteristic coefficient, x is the total number of deviations, z is the number of sound types, and k is the degree of the sound most recently added to the sound level sequence.

[0023] Preferably, N risk detection levels in the collection area are collected, and after comprehensive analysis with the sound safety level, a comprehensive risk distribution map is obtained, including: obtaining N different risk detection levels in the target area, and weighting and normalizing the different risk detection levels and the sound risk level to obtain a comprehensive risk distribution map, wherein the sound risk degree is divided to obtain the sound risk level.

[0024] A regional dynamic risk perception system, comprising:

[0025] A sub-regional sound level table acquisition module, used to obtain the standard sound level in the target area using the sound level calculation formula, and obtain the sub-regional sound level table based on the standard sound level in each target area;

[0026] A deviation calculation module, used to obtain the sound level in each target area, and input the sound level into a sound deviation formula to obtain the sound deviation within a preset time period in each target area;

[0027] A deviation sequence acquisition module, used to obtain the sound risk level in the target area;

[0028] An adjustment module is used to re-obtain a deviation degree sequence after judging the latest sound degree sequence and the historical benchmark sound degree;

[0029] A sound risk level acquisition module, used to obtain the sound risk level in the target area using a risk value formula according to the total number of deviations and the number of sound types;

[0030] The comprehensive analysis module is used to obtain a comprehensive risk distribution map after weighting and normalizing different risk detection levels and sound risk levels.

[0031] The beneficial effects of the present invention are as follows: the present invention analyzes the sound data collected by the factory with only simple sound level collection equipment, obtains the total number of deviations in the target area, thereby determining the number of abnormal sounds and conducting a danger level analysis, and conducts an overall sound level analysis on the noise generated when the machinery is working in the target area to obtain the number of sound types, thereby conducting risk perception when the sound in the target area changes frequently, and secondly analyzes the historical stable data of the target area to reduce the impact of rainy days on the overall sound collection, and prevent the inaccuracy of the total number of deviations caused by the increase in the sound level due to continuous environmental sounds when collecting the total number of deviations, and obtains the final comprehensive risk distribution map through comprehensive analysis of multiple danger perceptions. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A flow chart of a regional dynamic risk perception method of the present invention;

[0033] Figure 2 This is an example display diagram of the regional sound level table of the present invention. DETAILED DESCRIPTION

[0034] The present invention will now be described more fully below with reference to the accompanying drawings, in which currently preferred embodiments of the present invention are shown. However, the present invention may be implemented in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided for thoroughness and completeness, and these embodiments fully convey the scope of the present invention to the skilled person.

[0035] Example 1: A regional dynamic risk perception method, such as Figure 1-2 As shown, the following steps are included:

[0036] S1: After collecting the sound levels of the target area, the sound levels of the target area are divided to obtain a regional sound level table;

[0037] The sound level in the target area is cleaned, and the standard sound level in the target area is obtained through the sound level calculation formula. Based on the standard sound level in each target area, a regional sound level table is obtained, where the sound level calculation formula is:

[0038]

[0039] Where L total is the total sound pressure level in the target area, L i is the sound pressure level of each noise source at distance d0, d i is the distance from the noise source i to the observation point, and d0 is the reference distance.

[0040] It should be noted that after data cleaning of the sound levels in the target area by removing duplicate values, filling missing values, and removing outliers, the sound level calculation formula is used to obtain the standard sound level in the target area, and based on the standard sound level in each target area, a regional sound level table is obtained, where each standard sound level is the standard sound level caused by the sound of different numbers of machines in each factory building in the target area.

[0041] S2: Obtain the sound level of the target area, compare it with the regional sound level table, and obtain the sound deviation of the target area;

[0042] The sound level in each target area is obtained, and the sound level is input into the sound deviation formula to obtain the sound deviation within the preset time period in each target area. The sound deviation greater than or equal to the preset sound threshold is taken as the abnormal sound value, and the sound deviation less than the preset sound threshold is taken as the normal sound value, wherein the sound deviation formula is:

[0043] M=ω|Z 声 -Z 标 |

[0044] Where M is the sound deviation, ω is the adjustment weight of the sound deviation, and Z 声 is the sound level in the target area, Z 标 It is the standard sound level of the sub-region sound level table in the target area.

[0045] It should be noted that the sound level will have a preset sound threshold, and the sound levels within the preset sound threshold can be divided into various sound levels.

[0046] S3: Analyze the sound level and sound deviation in the target area to obtain the risk level in the target area;

[0047] The FIFO method is used to sort the historical sound deviations within the first preset time period of the target area to obtain a deviation sequence, and the sound level sequence is obtained based on the historical sound deviations. According to the sound level sequence and the deviation sequence, the sound risk degree in the target area is obtained.

[0048] It should be noted that the historical sound deviation within the first preset time period of the target area is obtained through the FIFO method, the elements in the sequence are analyzed, and when new data is obtained, the earliest data entering the sequence is replaced, and the new sequence is analyzed.

[0049] The sound level in the target area is compared with all standard sound levels in the regional sound level table to obtain the deviation between the sound intensity level in the target area and each sound intensity level. The minimum deviation is taken as the deviation in the target area, and the historical sound deviations are sorted using the FIFO method to obtain a deviation sequence.

[0050] It should be noted that by comparing the sound level in the target area with all standard sound levels in the regional sound level table and taking the minimum deviation as the deviation in the target area, the sound level closest to that generated by the operation of factory machinery can be found, thereby obtaining a deviation sequence.

[0051] A first preset number of smooth sound level sequences with smooth sound levels in the target area are obtained, and a historical benchmark sound level is obtained by weighted average according to each smooth sound level. A deviation degree sequence is obtained after judging based on the latest sound level sequence and the historical benchmark sound level.

[0052] When the number of sound levels in the latest sound level sequence that is greater than or equal to the historical benchmark sound level is greater than a second preset number, the sound levels in the latest sound level sequence are adjusted to re-obtain the deviation sequence, wherein the adjustment is to use the difference between the latest sound level and the historical benchmark sound level as the adjusted sound value, and to recalculate the deviation sequence by subtracting the latest sound level sequence according to the adjusted sound value.

[0053] It should be noted that when the sound deviation greater than or equal to the preset sound threshold is taken as the abnormal sound value, because environmental factors such as rain increase all sound levels, the sound that originally met the sound level classification will become non-compliant, and the sound that originally did not meet the sound level classification will become compliant. For example, the sound levels are 1 and 2, and their standard sound levels are 100 and 200 respectively. The preset sound threshold is 20, the normal sound level is 150, and the sound level clamp of rain is 30. Then the sound threshold of sound level 1 is 120-80, and the sound threshold of sound level 2 is 180-220. The collected sound level is 180, which can be divided into sound level 2. Then the total deviation will no longer be accurate. The most consistent sound level in daily life is obtained by weighted average method, and the difference between the latest sound level and the most consistent sound level in daily life is obtained as the adjusted sound value. The sequence is adjusted according to the adjusted sound value to regain the deviation sequence.

[0054] The sound level in the target area is obtained based on the minimum deviation, and the sound level sequence is obtained using the FIFO method according to the sound level of each minimum deviation in the target area. The total number of abnormal sound values ​​in the deviation sequence is taken as the total number of deviations, and the number of types of normal sound values ​​in the sound level sequence is taken as the number of sound types. The sound risk degree in the target area is obtained according to the risk value formula, where the risk value formula is:

[0055]

[0056] Where S is the risk value, ω, β is the preset characteristic coefficient, x is the total number of deviations, z is the number of sound types, and k is the degree of the sound most recently added to the sound level sequence.

[0057] It should be noted that if the number of times that the sound level in the latest sound level sequence is greater than or equal to the historical reference sound level is greater than a second preset number, the risk value is calculated based on the recovered deviation degree sequence.

[0058] S4: Collect N risk detection levels in the area, and after comprehensive analysis with the sound safety level, obtain a comprehensive risk distribution map.

[0059] N different risk detection levels in the target area are obtained, and the different risk detection levels are weighted and normalized with the sound risk levels to obtain a comprehensive risk distribution map, in which the sound risk degree is divided to obtain the sound risk level.

[0060] It should be noted that after detecting and grading other risks such as wind and heat, the corresponding target areas are added to the sound risk level to obtain a comprehensive risk distribution map, which can fully and accurately analyze the risk level of the factory.

[0061] Embodiment 2: Based on Embodiment 1, a regional dynamic risk perception system includes:

[0062] A sub-regional sound level table acquisition module, used to obtain the standard sound level in the target area using the sound level calculation formula, and obtain the sub-regional sound level table based on the standard sound level in each target area;

[0063] A deviation calculation module is used to obtain the sound level in each target area and input the sound level into a sound deviation formula to obtain the sound deviation within a preset time period in each target area;

[0064] A deviation sequence acquisition module, used to obtain the sound risk level in the target area;

[0065] An adjustment module is used to re-obtain a deviation degree sequence after judging the latest sound degree sequence and the historical benchmark sound degree;

[0066] A sound risk level acquisition module, used to obtain the sound risk level in the target area using a risk value formula according to the total number of deviations and the number of sound types;

[0067] The comprehensive analysis module is used to obtain a comprehensive risk distribution map after weighting and normalizing different risk detection levels and sound risk levels.

[0068] While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments.The scope of the following claims is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.

Claims

1. A regional dynamic risk perception method, characterized in that: The following steps are involved: S1: After collecting the sound levels of the target area, the sound levels of the target area are divided to obtain a regional sound level table; S2: Obtain the sound level of the target area, compare it with the regional sound level table, and obtain the sound deviation of the target area; S3: Analyze the sound level and sound deviation in the target area to obtain the risk level in the target area; S4: Collect N risk detection levels in the area, and after comprehensive analysis with the sound safety level, obtain a comprehensive risk distribution map.

2. A regional dynamic risk perception method according to claim 1, characterized in that: After collecting the sound levels of the target area, the sound levels of the target area are divided to obtain a regional sound level table, including: performing data cleaning on the sound levels in the target area, obtaining the standard sound levels in the target area through a sound level calculation formula, and obtaining a regional sound level table based on the standard sound levels in each target area, wherein the sound level calculation formula is: Where L total is the total sound pressure level in the target area, L i is the sound pressure level of each noise source at distance d0, d i is the distance from the noise source i to the observation point, and d0 is the reference distance.

3. A regional dynamic risk perception method according to claim 1, characterized in that: The sound level of the target area is obtained, and it is compared with the regional sound level table to obtain the sound deviation of the target area, including: obtaining the sound level in each target area, and inputting the sound level into the sound deviation formula to obtain the sound deviation within a preset time period in each target area, taking the sound deviation greater than or equal to the preset sound threshold as the abnormal sound value, and taking the sound deviation less than the preset sound threshold as the normal sound value, wherein the sound deviation formula is: M=ω|Z 声 -Z 标 | Where M is the sound deviation, ω is the adjustment weight of the sound deviation, and Z 声 is the sound level in the target area, Z 标 It is the standard sound level of the sub-region sound level table in the target area.

4. A regional dynamic risk perception method according to claim 1, characterized in that: The sound level and sound deviation in the target area are analyzed to obtain the sound risk level in the target area, including: using the FIFO method to sort the historical sound deviations in the target area within a first preset time period to obtain a deviation sequence, and analyzing based on the historical sound deviations to obtain a sound level sequence, and obtaining the sound risk level in the target area according to the sound level sequence and the deviation sequence.

5. A regional dynamic risk perception method according to claim 4, characterized in that: The method of using the FIFO method to sort the historical sound deviations within the first preset time period of the target area to obtain a deviation sequence includes: comparing the sound level in the target area with all standard sound levels in the regional sound level table, obtaining the deviation between the sound intensity level in the target area and each sound intensity level, taking the minimum deviation as the deviation in the target area, and using the FIFO method to sort the historical sound deviations to obtain a deviation sequence.

6. A regional dynamic risk perception method according to claim 4, characterized in that: The sound level sequence is obtained by analyzing the historical sound deviation, and before the sound risk level in the target area is obtained according to the sound level sequence and the deviation sequence, it includes: obtaining a first preset number of smooth sound level sequences with smooth sound levels in the target area, and obtaining a historical benchmark sound level by weighted average according to each smooth sound level, and re-obtaining the deviation sequence after judging based on the latest sound level sequence and the historical benchmark sound level.

7. A regional dynamic risk perception method according to claim 6, characterized in that: The method of re-obtaining the deviation degree sequence after judging the latest sound level sequence and the historical reference sound level comprises: When the number of sound levels in the latest sound level sequence that is greater than or equal to the historical benchmark sound level is greater than a second preset number, the sound levels in the latest sound level sequence are adjusted to re-obtain the deviation sequence, wherein the adjustment is to use the difference between the latest sound level and the historical benchmark sound level as the adjusted sound value, and to re-calculate the deviation sequence by subtracting the latest sound level sequence according to the adjusted sound value.

8. A regional dynamic risk perception method according to claim 4, characterized in that: The sound level sequence is obtained by analyzing the historical sound deviation, and the sound risk level in the target area is obtained according to the sound level sequence and the deviation sequence, including: obtaining the sound level in the target area based on the minimum deviation, and obtaining the sound level sequence using the FIFO method according to the sound level of each minimum deviation in the target area, taking the total number of abnormal sound values ​​in the deviation sequence as the total number of deviations, taking the number of types of normal sound values ​​in the sound level sequence as the number of sound types, and obtaining the sound risk level in the target area according to the risk value formula, wherein the risk value formula is: Where S is the risk value, ω, β is the preset characteristic coefficient, x is the total number of deviations, z is the number of sound types, k is the sound level most recently added to the sound level sequence.

9. A regional dynamic risk perception method according to claim 1, characterized in that: The N risk detection levels in the collection area are collected and comprehensively analyzed with the sound safety level to obtain a comprehensive risk distribution map, including: obtaining N different risk detection levels in the target area, and weighting and normalizing the different risk detection levels and the sound risk level to obtain a comprehensive risk distribution map, wherein the sound risk degree is divided to obtain the sound risk level.

10. A regional dynamic risk perception system, characterized in that: include: A sub-regional sound level table acquisition module, used to obtain the standard sound level in the target area using the sound level calculation formula, and obtain the sub-regional sound level table based on the standard sound level in each target area; A deviation calculation module, used to obtain the sound level in each target area, and input the sound level into a sound deviation formula to obtain the sound deviation within a preset time period in each target area; A deviation sequence acquisition module, used to obtain the sound risk level in the target area; An adjustment module is used to re-obtain a deviation degree sequence after judging the latest sound degree sequence and the historical benchmark sound degree; A sound risk level acquisition module, used to obtain the sound risk level in the target area using a risk value formula according to the total number of deviations and the number of sound types; The comprehensive analysis module is used to obtain a comprehensive risk distribution map after weighting and normalizing different risk detection levels and sound risk levels.

Images