CN-121997215-A - AI-driven environment monitoring and early warning method, medium and equipment

CN121997215ACN 121997215 ACN121997215 ACN 121997215ACN-121997215-A

Abstract

The invention discloses an AI-driven environment monitoring and early warning method, medium and equipment, and relates to the technical field of environment monitoring, wherein the method comprises the steps of constructing an abnormal trend prediction model, and carrying out abnormal event classification and abnormal trend prediction according to monitoring data of a kiln flue; the method comprises the steps of taking an abnormal predicted event as a search engine, searching in an abnormal event analysis library to obtain an abnormal influence factor set, configuring a collection strategy of a monitoring collection device and obtaining abnormal monitoring probability data, inputting the abnormal monitoring probability data into an abnormal recognition early warning model, carrying out abnormal event recognition and risk grade judgment, and generating early warning information according to an abnormal event recognition result and risk grade. The method solves the technical problems that in the prior art, complicated and changeable furnace flue environments are difficult to effectively cope with, abnormal events cannot be predicted in advance, and the accuracy and the timeliness of environmental monitoring and early warning are insufficient, achieves the intelligent environmental monitoring and early warning, and improves the technical effects of the timeliness and the accuracy of the environmental monitoring and early warning.

Inventors

ZHU LINGLING
Qin Jiayao

Assignees

启东市清源环境检测技术有限公司

Dates

Publication Date: 20260508
Application Date: 20251209

Claims (10)

An ai-driven environmental monitoring and early warning method, comprising: An abnormal trend prediction model is constructed, abnormal event classification and abnormal trend prediction are carried out through the abnormal trend prediction model according to the monitoring data of the kiln flue, and an abnormal prediction event and abnormal trend probability are obtained; Searching in an abnormal event analysis library by taking the abnormal prediction event as a search engine to obtain an abnormal influence factor set, wherein the abnormal influence factor set comprises abnormal influence factors and influence probability coefficients; Configuring an acquisition strategy of the monitoring acquisition equipment according to the abnormality prediction event and the abnormality trend probability, the abnormality influence factor and the influence probability coefficient, and acquiring abnormality monitoring probability data based on the acquisition strategy of the monitoring acquisition equipment; And inputting the abnormal monitoring probability data into an abnormal recognition early warning model, recognizing abnormal events and judging risk levels, and generating early warning information according to abnormal event recognition results and risk levels.
2. The AI-driven environmental monitoring and early warning method of claim 1, wherein constructing an abnormal trend prediction model comprises: collecting historical sample data of the kiln, including a normal state and an abnormal state; comparing the furnace flue monitoring data corresponding to the normal state and the abnormal state to determine an abnormal event and an abnormal performance characteristic; Establishing a training data set according to the abnormal event and the abnormal performance characteristics, and training a classifier by using the training data set; fitting time sequence trend characteristics according to time sequence sample data in an abnormal state, and training a time sequence network frame to obtain a trend probability prediction model; And integrating and connecting the classifier with a trend probability prediction model to obtain the abnormal trend prediction model.
3. The AI-driven environmental monitoring and early warning method according to claim 2, wherein determining abnormal events and abnormal performance characteristics by comparing furnace flue monitoring data corresponding to the normal state and the abnormal state comprises: According to the furnace flue monitoring data in the abnormal state, a time sequence monitoring data chain is constructed according to the monitoring time sequence, and the prepositive tracing is carried out based on the identification nodes in the abnormal state in the time sequence monitoring data chain, so that the prepositive data change information is obtained; Determining a comparison time period according to the change time period of the preamble data change information; extracting the period data of the normal state according to the comparison time period to obtain the period data of the normal state; aligning the normal state period data with the comparison time period data of the abnormal state to obtain difference data characteristics, and taking the difference data characteristics as abnormal performance characteristics; And carrying out mapping association on the abnormal performance characteristics according to the abnormal event corresponding to the abnormal state, and determining the abnormal event and the abnormal performance characteristics.
4. The AI-driven environmental monitoring and early warning method according to claim 2, wherein the abnormal event classification is performed by an abnormal trend prediction model according to the monitored data of the kiln flue, previously comprising: According to the characterization factors of the normal state and the abnormal state, arranging monitoring sensors, wherein the monitoring sensors at least comprise monitoring sensors corresponding to temperature, pressure, flow rate, gas concentration, humidity, flame spectrum and vibration spectrum; And collecting monitoring data of the kiln flue in real time according to the arranged monitoring sensors.
5. The AI-driven environmental monitoring and warning method of claim 4, wherein searching in an anomaly event resolution library using the anomaly prediction event as a search engine, previously comprises: analyzing relevant monitoring data and influence coefficients of the monitoring data on the abnormal event based on historical sample data according to the abnormal event and the abnormal performance characteristics; the method comprises the steps of taking an abnormal event as a top node, taking each related monitoring data as a support node, establishing a connection edge of the node and the support node, and obtaining an analytic graph structure of each abnormal event by taking a weight value of the connection edge as an influence coefficient; And combining the analysis graph structures of all the abnormal events in parallel to construct the abnormal event analysis library.
6. The AI-driven environmental monitoring and early warning method of claim 1, wherein configuring the acquisition strategy of the monitoring acquisition device according to the anomaly prediction event and the anomaly trend probability, and the anomaly impact factor, impact probability coefficient, comprises: Acquiring abnormal trend probability of an abnormal prediction event, and configuring an acquisition time window, wherein the acquisition time window is shorter when the abnormal trend probability is larger; Setting acquisition frequency according to the acquisition time window and the influence probability coefficient of the abnormal influence factor; and configuring an acquisition strategy of the monitoring acquisition equipment corresponding to the abnormal influence factor according to an acquisition time window and an acquisition frequency according to the mapping relation between the abnormal prediction event and the abnormal influence factor.
7. The AI-driven environmental monitoring and early warning method according to claim 6, wherein configuring the acquisition strategy of the monitoring acquisition device corresponding to the abnormal impact factor according to the acquisition time window and the acquisition frequency comprises: When the abnormal influence factors of the abnormal prediction events overlap, configuring an acquisition strategy corresponding to the monitoring acquisition equipment according to the acquisition time window and the maximum value of the overlapping area of the acquisition frequency, and establishing the mapping association of the abnormal prediction events and the acquisition strategy of the monitoring acquisition equipment.
8. The AI-driven environmental monitoring and early warning method of claim 1, wherein inputting the anomaly monitoring probability data into an anomaly identification early warning model for anomaly event identification and risk level determination comprises: Cleaning pretreatment is carried out on the abnormal monitoring probability data, local time sequence mutation characteristics and long period dependency relationships are extracted, and the spatial relevance of a sensor network is established, so that monitoring data space-time characteristics are obtained; and inputting the space-time characteristics of the monitoring data into the abnormal recognition early warning model to perform abnormal characteristic recognition judgment, obtaining an abnormal event recognition result, determining a risk level according to the abnormal event recognition result, and taking the abnormal event recognition result and the risk level as output results.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the AI-driven environmental monitoring and warning method according to any one of claims 1 to 8.
10. An electronic device, the electronic device comprising: A memory for storing executable instructions; and the processor is used for realizing the AI-driven environment monitoring and early warning method according to any one of claims 1-8 when executing the executable instructions stored in the memory.

Description

AI-driven environment monitoring and early warning method, medium and equipment Technical Field The application relates to the technical field of environmental monitoring, in particular to an AI-driven environmental monitoring and early warning method, medium and equipment. Background The kiln is widely applied to the fields of steel smelting, glass manufacturing, ceramic firing and the like, the running condition of the kiln directly influences the production efficiency and the product quality, and the flue of the kiln is taken as a key channel for discharging waste gas, so that accurate and real-time monitoring of the flue of the kiln is very important. On the one hand, the conventional environment monitoring relies on fixed threshold judgment, when the running state of the kiln changes in a complex manner due to production process adjustment, equipment aging and the like, the real condition cannot be accurately reflected, abnormal event missed judgment or misjudgment is extremely easy to occur, on the other hand, the periodic inspection is low in efficiency, the full-time and full-direction monitoring of the kiln flue cannot be realized, the instantaneous abnormality in the flue is difficult to capture, the potential problem is difficult to discover in time, and the timeliness, accuracy and comprehensiveness of the environment monitoring of the kiln flue are further influenced. In the related technology at the present stage, the technical problems that the complex and changeable furnace flue environment is difficult to effectively cope with, the abnormal event cannot be predicted in advance, and the accuracy and timeliness of the environment monitoring and early warning are insufficient exist. Disclosure of Invention By providing the AI-driven environment monitoring and early warning method, medium and equipment, the technical problems that the environment monitoring and early warning accuracy and timeliness are insufficient due to the fact that complex and changeable furnace kiln flue environments are difficult to effectively pair and abnormal events cannot be predicted in advance in the prior art are solved, the intelligent environment monitoring and early warning is realized, and the technical effects of the timeliness and the accuracy of the environment monitoring and early warning are improved. The application provides an AI-driven environment monitoring and early warning method, which comprises the steps of constructing an abnormal trend prediction model, carrying out abnormal event classification and abnormal trend prediction through the abnormal trend prediction model according to monitoring data of a kiln flue, obtaining an abnormal prediction event and abnormal trend probability, searching in an abnormal event analysis library by taking the abnormal prediction event as a search engine, obtaining an abnormal influence factor set, wherein the abnormal influence factor set comprises an abnormal influence factor and an influence probability coefficient, configuring an acquisition strategy of monitoring acquisition equipment according to the abnormal prediction event and the abnormal trend probability, the abnormal influence factor and the influence probability coefficient, obtaining abnormal monitoring probability data based on the acquisition strategy of the monitoring acquisition equipment, inputting the abnormal monitoring probability data into an abnormal recognition early warning model, carrying out abnormal event recognition and risk grade judgment, and generating early warning information according to an abnormal event recognition result and a risk grade. In a possible implementation manner, the AI-driven environment monitoring and early warning method is further used for collecting historical sample data of a kiln, including a normal state and an abnormal state, comparing the historical sample data of the kiln corresponding to the normal state and the abnormal state according to kiln flue monitoring data corresponding to the normal state and the abnormal state to determine abnormal events and abnormal performance characteristics, establishing a training data set according to the abnormal events and the abnormal performance characteristics, training a classifier by using the training data set, fitting time sequence trend characteristics according to time sequence sample data of the abnormal state, training a time sequence network frame to obtain a trend probability prediction model, and integrally connecting the classifier with the trend probability prediction model to obtain the abnormal trend prediction model. In a possible implementation manner, the AI-driven environment monitoring and early warning method is further used for performing the following processing of constructing a time sequence monitoring data chain according to furnace flue monitoring data of the abnormal state according to a monitoring time sequence, performing prepositive tracing on the basis of identification nodes of t