CN-121978199-A - Environment on-line monitoring method and system for VOC emission

CN121978199ACN 121978199 ACN121978199 ACN 121978199ACN-121978199-A

Abstract

The invention discloses an environment online monitoring method and system for VOC emission, and relates to the technical field of environment monitoring. The method comprises the steps of interactively arranging a monitor array of M points of an exhaust pipeline of a target mine, collecting historical monitoring logs to obtain M point location historical monitoring log sets, traversing the M point location historical monitoring log sets to conduct high-low monitoring frequency identification to determine M point location high-low monitoring frequency group sets, continuously monitoring the M points to obtain M point location high-frequency monitoring data sequence sets and M point location low-frequency monitoring data sequence sets, conducting high-low frequency interaction fusion analysis on the M point location high-frequency monitoring data sequence sets and the M point location low-frequency monitoring data sequence sets, and determining M point location monitoring result sets. The technical problems of low mine VOC emission monitoring efficiency and insufficient monitoring data accuracy in the prior art are solved, and the technical effects of improving the monitoring efficiency and the monitoring data accuracy are achieved.

Inventors

HUANG YUNLU
SHI JIAHUAN

Assignees

南通仁源节能环保科技有限公司

Dates

Publication Date: 20260505
Application Date: 20251205

Claims (9)

1. An environmental on-line monitoring method for VOC emissions, the method comprising: The method comprises the steps of interactively arranging a monitor array of M point positions of an exhaust pipeline of a target mine, collecting historical monitoring logs, and obtaining M point position historical monitoring log sets, wherein M is a positive integer; Traversing the M point location history monitoring log sets by taking a preset environment monitoring parameter set as an index to identify the high and low monitoring frequencies, and determining M point location high and low monitoring frequency group sets; The monitor array is called to continuously monitor the M points according to the M point high-low monitoring frequency group sets to obtain M point high-frequency monitoring data sequence sets and M point low-frequency monitoring data sequence sets; And carrying out high-low frequency interaction fusion analysis on the M point location high-frequency monitoring data sequence sets and the M point location low-frequency monitoring data sequence sets to determine M point location monitoring result sets.
2. The method for on-line environmental monitoring of VOC emissions of claim 1 wherein said set of preset environmental monitoring parameters comprises temperature, stack pressure, flow rate, and humidity.
3. The method for on-line monitoring of VOC emissions according to claim 1, wherein traversing the M point location history monitoring log sets with a preset environmental monitoring parameter set as an index for performing high-low monitoring frequency identification, determining M point location high-low monitoring frequency group sets, comprises: Taking the preset environment monitoring parameter set as an index, extracting parameter abnormal interval duration from the M point location history monitoring log sets, and obtaining M point location parameter abnormal interval duration clusters; Screening the M point location parameter abnormal interval duration clusters to obtain M point location low monitoring frequency sets and M point location high monitoring frequency sets; and mapping and combining the M point location low monitoring frequency sets and the M point location high monitoring frequency sets to obtain M point location high and low monitoring frequency set sets.
4. The method for online monitoring of VOC emissions according to claim 3, wherein screening the M dot-location parameter anomaly interval duration clusters to obtain M dot-location low-monitoring frequency sets and M dot-location high-monitoring frequency sets comprises: Respectively extracting the maximum value of the parameter abnormal interval time length of different environment monitoring parameters from the M point position parameter abnormal interval time length clusters to be used as M point position low monitoring frequency sets; and respectively extracting the minimum value of the parameter abnormality interval duration of different environment monitoring parameters from the M point location parameter abnormality interval duration clusters to be used as M point location high monitoring frequency sets.
5. The method for on-line monitoring of VOC emissions according to claim 1, wherein performing high-low frequency interactive fusion analysis on the M dot-location high-frequency monitoring data sequence sets and the M dot-location low-frequency monitoring data sequence sets, determining M dot-location monitoring result sets, comprises: Pre-constructing a first feature extraction branch and a second feature extraction branch; Performing characteristic convolution analysis on the M point location high frequency monitoring data sequence sets by using a first characteristic extraction branch to determine M point location high frequency time sequence vector sets; performing characteristic convolution analysis on the M point location low frequency monitoring data sequence sets by using a second characteristic extraction branch to determine M point location low frequency space vector sets; Mapping, interacting and fusing the M point high-frequency time sequence vector sets and the M point low-frequency space vector sets, and updating attention weights of the second feature extraction branches respectively to obtain updated M updated second feature extraction branches; Performing feature convolution analysis on the M point location low frequency monitoring data sequence sets by using the M updated second feature extraction branches to determine M updated point location low frequency space vector sets; And respectively analyzing the M updating point position low-frequency space vector sets and the M point position high-frequency time sequence vector sets by using a fully connected network layer to obtain M point position monitoring result sets.
6. The method for on-line monitoring of the environment for VOC emissions of claim 5, wherein the receptive field of said first feature extraction branch is greater than the receptive field of said second feature extraction branch.
7. The method for on-line monitoring of VOC emissions according to claim 5, wherein mapping, interactively fusing the M point high frequency time sequence vector sets and the M point low frequency space vector sets, respectively performing attention weight update on the second feature extraction branches, and obtaining updated M updated second feature extraction branches, comprising: extracting a first point bit high-frequency time sequence vector set and a first point bit low-frequency space vector set from the M point bit high-frequency time sequence vector sets and the M point bit low-frequency space vector sets; The attention weight of the second feature extraction branch is obtained, and the first point high-frequency time sequence vector set and the first point low-frequency space vector set are subjected to interactive fusion analysis by combining with a mapping interactive fusion function to obtain a first updated attention weight; updating the second feature extraction branch by using the first updated attention weight to obtain a first updated second feature extraction branch; And performing mapping interaction analysis on the M point high-frequency time sequence vector sets and the M point low-frequency space vector sets, updating the attention weight of the second feature extraction branch, and obtaining M updated second feature extraction branches.
8. The method for on-line monitoring of the environment for VOC emissions of claim 7, wherein said mapping interactive fusion function is: ; Wherein, the For the first updated attention weight, In order for the attention to be weighted, For the i-th first point low frequency space vector in the set of first point low frequency space vectors, Is a normalized value of the similarity between the i first bit low frequency space vector and the i first bit high frequency timing vector, Is a transpose of the first bit high frequency timing vector, For the dimension of the i-th first bit high frequency timing vector, m is the number of first bit low frequency space vectors in the first bit low frequency space vector set.
9. An on-line environmental monitoring system for VOC emissions, characterized by being adapted to implement the on-line environmental monitoring method for VOC emissions of any one of claims 1-8, the system comprising: The data acquisition module is used for interactively arranging a monitor array of M point positions of the target mine exhaust pipeline, carrying out history monitoring log acquisition and obtaining M point position history monitoring log sets, wherein M is a positive integer; The frequency identification module is used for traversing the M point location history monitoring log sets to identify the high and low monitoring frequencies by taking a preset environment monitoring parameter set as an index, and determining M point location high and low monitoring frequency group sets; the monitoring module is used for calling the monitor array to continuously monitor the M point positions according to the M point position high-low monitoring frequency group sets to obtain M point position high-frequency monitoring data sequence sets and M point position low-frequency monitoring data sequence sets; and the fusion analysis module is used for carrying out high-low frequency interaction fusion analysis on the M point location high-frequency monitoring data sequence sets and the M point location low-frequency monitoring data sequence sets to determine M point location monitoring result sets.

Description

Environment on-line monitoring method and system for VOC emission Technical Field The invention relates to the technical field of environment monitoring, in particular to an environment on-line monitoring method and system for VOC emission. Background In the industrial field, particularly in mine mining, the problem of Volatile Organic Compound (VOC) emissions is becoming increasingly important. VOCs, as an important component of atmospheric pollutants, can not only adversely affect the ecological environment, but can also pose a potential threat to the health of the operators. In the prior art, the on-line monitoring means of the environment aiming at the emission of the mine VOC mostly adopts the monitoring mode of fixed frequency and fixed point positions, and has the technical defects of unreasonable monitoring frequency setting, low monitoring resource utilization rate, insufficient response sensitivity, limited data accuracy and the like, so that the actual requirements of efficient and accurate monitoring of the emission of the VOC under the complex working condition of the mine are difficult to meet. Disclosure of Invention The application provides an environment on-line monitoring method and system for VOC emission, which solve the technical problems of low efficiency and insufficient accuracy of monitoring data in mine VOC emission monitoring in the prior art. In a first aspect of the application, there is provided an on-line environmental monitoring method for VOC emissions, the method comprising: The method comprises the steps of interactively distributing a monitor array of M points of an exhaust pipeline of a target mine, carrying out historical monitoring log collection to obtain M point historical monitoring log sets, carrying out high-low frequency interactive fusion analysis on the M point high-frequency monitoring data sequence sets and the M point low-frequency monitoring data sequence sets by taking a preset environment monitoring parameter set as an index, traversing the M point historical monitoring log sets to carry out high-low monitoring frequency identification to determine M point high-low monitoring frequency group sets, calling the monitor array to continuously monitor the M points according to the M point high-low monitoring frequency group sets to obtain M point high-frequency monitoring data sequence sets and M point low-frequency monitoring data sequence sets, and determining M point monitoring result sets. In a second aspect of the application, there is provided an environmental on-line monitoring system for VOC emissions, the system comprising: The system comprises a data acquisition module, a frequency identification module, a monitoring module and a fusion analysis module, wherein the data acquisition module is used for interactively arranging a monitor array of M points in an exhaust pipeline of a target mine, carrying out history monitoring log acquisition to obtain M point history monitoring log sets, M is a positive integer, the frequency identification module is used for carrying out high-low monitoring frequency identification on the M point history monitoring log sets by taking a preset environment monitoring parameter set as an index to determine M point high-low monitoring frequency group sets, the monitoring module is used for calling the monitor array to continuously monitor the M points according to the M point high-low monitoring frequency group sets to obtain M point high-frequency monitoring data sequence sets and M point low-frequency monitoring data sequence sets, and the fusion analysis module is used for carrying out high-low frequency interactive fusion analysis on the M point high-frequency monitoring data sequence sets and the M point low-frequency monitoring data sequence sets to determine M point monitoring result sets. One or more technical schemes provided by the application have at least the following technical effects or advantages: Firstly, a monitor array of M points is interactively distributed on an exhaust pipeline of a target mine, history monitoring log collection is carried out, and M point history monitoring log sets are obtained, wherein M is a positive integer. And then, traversing the M point location history monitoring log sets by taking the preset environment monitoring parameter set as an index to identify the high and low monitoring frequencies, and determining M point location high and low monitoring frequency group sets. And then, continuously monitoring the M points by calling the monitor array according to the M point high-low monitoring frequency group sets to obtain M point high-frequency monitoring data sequence sets and M point low-frequency monitoring data sequence sets. And finally, carrying out high-low frequency interaction fusion analysis on the M point location high-frequency monitoring data sequence sets and the M point location low-frequency monitoring data sequence sets, and determining M point location monito