CN114442512A - Chemical safety monitoring system - Google Patents
Chemical safety monitoring system Download PDFInfo
- Publication number
- CN114442512A CN114442512A CN202011194039.3A CN202011194039A CN114442512A CN 114442512 A CN114442512 A CN 114442512A CN 202011194039 A CN202011194039 A CN 202011194039A CN 114442512 A CN114442512 A CN 114442512A
- Authority
- CN
- China
- Prior art keywords
- data
- chemical
- video
- abnormal
- detection data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0428—Safety, monitoring
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24024—Safety, surveillance
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Alarm Systems (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The application provides a chemical safety monitoring system, includes: the system comprises a video acquisition end, a data acquisition end and a data analysis end; the video acquisition end is used for acquiring video images of chemical fields and sending the video images to the data analysis end; the data acquisition end is used for acquiring and storing detection data of a chemical field in real time and sending the detection data to the data analysis end in real time; the data analysis end is used for analyzing abnormal states; the method comprises the steps of analyzing the video images and the detection data, and generating alarm data files and uploading the alarm data files of the relevant detection data and the relevant video images when abnormal conditions are found. The system effectively improves the early warning pushing efficiency for abnormal conditions and reduces the accident rate of major hazard sources; the problem of mass data transmission caused by directly transmitting the field detection data and the corresponding image video to the server can be avoided, and the transmission pressure of data streams is reduced.
Description
Technical Field
The application relates to the technical field of computers, in particular to a chemical safety monitoring system.
Background
In the industrial development, the production of chemicals tends to be large-scale, modularized and automatic, and chemicals, especially dangerous chemicals, generally have the properties of being flammable and explosive, toxic and harmful, etc., and there is often a high risk in the place where the chemicals exist, and if a chemical accident occurs, a great influence range is caused, so that it is very important to perform remote monitoring on the major dangerous source of the dangerous chemicals.
In the prior art, most chemical production enterprises or laboratories are provided with monitoring systems for monitoring the safety states and potential risk factors of chemicals (especially dangerous chemicals) in the processes of production, use and storage. However, most monitoring systems are independent internal monitoring systems, and enterprises often cannot sufficiently pay attention to safety supervision of dangerous chemicals; when an accident occurs, the internal supervision system cannot upload the accident situation to all levels of relevant supervision departments in time, and accordingly all levels of relevant supervision departments cannot master the safety situation of the major hazard source in real time to monitor and early warn the major hazard source before the accident occurs.
Therefore, the above prior art is not comprehensive in remote monitoring of the major hazard source of the hazardous chemicals, and fails to timely push the warning information for the abnormal condition of the hazardous chemicals.
Disclosure of Invention
The application provides a chemical safety monitoring system to solve the problem that abnormal conditions to dangerous chemicals that exist among the prior art can not in time push early warning information.
The application provides a chemical safety monitoring system, includes: the system comprises a video acquisition end, a data acquisition end and a data analysis end; the video acquisition end is used for acquiring video images of chemical fields and sending the video images to the data analysis end; the data acquisition end is used for acquiring and storing detection data of a chemical field in real time and sending the detection data to the data analysis end in real time; the data analysis end is used for analyzing abnormal states; the method comprises the steps of analyzing the video images and the detection data, and generating alarm data files and uploading the alarm data files of the relevant detection data and the relevant video images when abnormal conditions are found.
Optionally, the data analysis end finds the abnormal condition by analyzing the detection data, and specifically includes the following steps: classifying the detection data to obtain data of various types of detection data in a preset time period; and respectively analyzing various types of detection data according to a preset analysis method and parameters, and performing comprehensive analysis by combining different types of data to determine whether abnormal conditions occur.
Optionally, the data analysis end finds an abnormal situation by analyzing the video image, and specifically includes the following steps: providing the video image to a safety state video identification system trained in advance for safety state identification; the safety state video recognition system is an intelligent image recognition system obtained by training in advance, and the intelligent recognition system can judge whether an abnormal condition occurs or not by reading a video image; and judging whether the video image has abnormal conditions or not according to the safety state identification result of the safety state identification system.
Optionally, if the data analysis end finds an abnormal situation by analyzing the video image, a relevant time period of the detection data is determined according to the type of the abnormal situation, and the detection data in the time period is acquired as the relevant detection data.
Optionally, the data analysis end is further configured to determine an influence level of the abnormal condition on the safety of the chemical, and determine an uploading level of the alarm data file according to the influence level.
Optionally, the data analysis end further includes a basic data input end, an equipment parameter input end, and a chemical data input end; the basic data input end is used for inputting the on-site basic data of the dangerous chemicals and sending the on-site basic data to a dangerous chemical registration management system, a dangerous chemical GIS application system and a chemical safety knowledge base; the equipment parameter input end is used for inputting equipment parameters of equipment related in the storage and use processes of chemicals; the chemical data input end is used for inputting data of raw material chemicals and intermediate chemical data generated in the chemical processing process; and the data analysis end receives the input data and uses the input data in the abnormal state analysis process.
Optionally, the video acquisition end is further configured to adjust the brightness of a light supplement lamp in the video acquisition end if the brightness of the obtained video image is relatively dark, so as to obtain a video image meeting the brightness required by the data analysis end.
Optionally, the data analysis end is further configured to generate a real-time data file from the detection data and the video image corresponding to the detection data, and store the real-time data file as historical data; and when the abnormal state analysis is carried out, the comparison analysis with the historical data is included.
Optionally, the system further comprises an emergency processing end for generating an emergency processing scheme and performing remote control on the site, including calling video data and performing video inspection through the video acquisition end.
Optionally, the data acquisition end realizes unidirectional transmission of the acquired data to the outside through a unidirectional gateway.
Optionally, the data analysis end is specifically configured to obtain production data information of the chemical field, input the production data information of the chemical field into an abnormal state evaluation model, and obtain abnormal state alarm level information corresponding to the production data information of the chemical field output by the abnormal state evaluation model.
Optionally, the production data information of the chemical field includes at least one of video data information of the chemical field and detection data information of the chemical field.
Optionally, the abnormal state evaluation model is obtained by pre-training according to the historical production data information and the historical production data abnormal state alarm level corresponding to the historical generation data information.
Optionally, the abnormal state evaluation model includes a smoke abnormal state evaluation model and a water mist abnormal state evaluation model.
Optionally, the data analysis end is further configured to determine whether a duration of a field abnormal state of the chemical corresponding to the abnormal state alarm level information is greater than a preset time threshold; if the judgment result is yes, the abnormal state alarm grade information and the production data information of the chemical field corresponding to the abnormal state alarm grade information are used as negative samples; if the judgment result is negative, taking the abnormal state alarm grade information and the production data information of the chemical field corresponding to the abnormal state alarm grade information as a positive sample; and training the abnormal state evaluation model according to the positive sample and the negative sample.
The application provides a chemical safety monitoring system includes: the system comprises a video acquisition end, a data acquisition end and a data analysis end; the video acquisition end is used for acquiring video images of chemical fields and sending the video images to the data analysis end; the data acquisition end is used for acquiring and storing detection data of a chemical field in real time and sending the detection data to the data analysis end in real time; the data analysis end is used for analyzing abnormal states; the method comprises the steps of analyzing the video images and the detection data, and generating alarm data files and uploading the alarm data files of the relevant detection data and the relevant video images when abnormal conditions are found.
According to the technical scheme, the video images and the detection data of the chemical site are analyzed, the related video images and the related detection data of abnormal conditions are generated into the alarm data file in time and uploaded, the early warning pushing efficiency of the abnormal conditions is effectively improved, and the accident rate of major hazard sources is reduced.
The application provides a chemicals safety monitoring system can realize handling the video image and the detection data on chemical scene in time, generates the alarm data file and uploads, can avoid directly to the problem of the large amount of data transmission that the server transmission scene detection data and the image video that corresponds caused, alleviates the transmission pressure of dataflow.
Drawings
Fig. 1 provides a schematic diagram of a first logical structure of a chemical safety monitoring system.
Fig. 2 provides a flow chart of the data analysis end finding abnormal conditions by analyzing the detected data.
Fig. 3 provides a flow chart of the data analysis end finding abnormal conditions by analyzing video images.
Fig. 4 is a schematic diagram of a relationship between uploading alarm data files by a data analysis end according to an embodiment of the present application.
Fig. 5 provides a schematic diagram of a second logical structure of the chemical safety monitoring system.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. This application is capable of implementation in many different ways than those herein set forth and of similar import by those skilled in the art without departing from the spirit of this application and is therefore not limited to the specific implementations disclosed below.
The embodiment of the application provides a chemical safety monitoring system, which is used for finding out the abnormal condition of a chemical field in time, generating an alarm data file and uploading the alarm data file.
Referring to fig. 1, a schematic diagram of a first logic structure of the chemical safety monitoring system is provided. The following describes the chemical safety monitoring system with reference to the figure, and describes the process of processing the detection data and the video image under the abnormal condition by the chemical safety monitoring system.
As shown in fig. 1, the chemical safety monitoring system includes a video acquisition terminal 101, a data acquisition terminal 102, and a data analysis terminal 103.
And the video acquisition terminal 101 is used for acquiring a video image of a chemical field and sending the video image to the data analysis terminal.
The above is a description of the video acquisition terminal 101 from a functional perspective, and the video acquisition terminal is used for acquiring a video image of a chemical field and sending the video image to the data analysis terminal, so as to provide a video basis for the data analysis terminal to analyze an abnormal situation occurring in the chemical field.
The video images of the chemical site refer to all video images in the chemical site, and are used for recording the change conditions of the environmental change in the chemical site on the chemicals or the change conditions of the chemicals on the environment. The video images of the chemical site collected by the video collecting end comprise time information for recording the video images, so that the data analysis end can conveniently sort the video images according to the time information after receiving the video images, and the video images and the detection data can be conveniently matched according to the time information. The chemical field includes any place where chemicals exist, including an enterprise production line and a storage warehouse for producing chemicals, a scientific research laboratory for using chemicals, a transport vehicle for transporting chemicals, and the like, and is not limited herein.
The video acquisition terminal is realized by adopting acquisition equipment capable of acquiring the change condition of the chemical on site, and the acquisition equipment comprises various video acquisition equipment, such as a video camera, a panoramic camera, a GPS (global positioning system) positioning machine and the like. The video acquisition end comprises at least one video acquisition device, and the number of the video acquisition devices is not limited in the application. The video acquisition end actually shows the form of above-mentioned all kinds of camera equipment to arrange in the spatial position that can gather the on-the-spot information of chemicals.
The video acquisition end provides a basic video image sample for the data analysis end to analyze whether the abnormal condition occurs in the chemical field or not by acquiring the video image of the chemical field. In the application, the video acquisition end is further used for adjusting the brightness of the light supplement lamp in the video acquisition end to obtain the video image meeting the brightness required by the data analysis end if the brightness of the obtained video image is dark. That is to say, if the current video image collected by the video collection end is dark and cannot meet the luminosity requirement required by the data analysis end for analyzing the video, the video analysis end further comprises a light supplement lamp, and the brightness of the video image is enhanced by adjusting the brightness of the light supplement lamp, so that the brightness requirement when the data analysis end analyzes the video image is met.
And the data acquisition end 102 is used for acquiring and storing the detection data of the chemical field in real time and sending the detection data to the data analysis end in real time.
The above is a description of the data acquisition end 102 from a functional perspective, and the data acquisition end is used for acquiring detection data of a chemical field and sending the detection data to the data analysis end, so as to provide a data basis for the data analysis end to analyze abnormal conditions occurring in the chemical field.
The detection data of the chemical field refers to various detection data recorded in the occasions where chemicals exist, and comprises environment detection data, experimental data and the like; the field detection data of the chemicals can reflect the state of the chemicals, and whether the chemicals are abnormal or not is determined according to the numerical value change of the detection data in a preset time period. The explanation of the chemical field is already explained at the video capturing end 101 and will not be described herein.
The real-time collection and storage of the detection data of the chemical field refers to that in order to completely record the factors influencing the physicochemical properties of the chemicals on the chemical field, the data collection end periodically collects the detection data of the chemical field according to the preset time, so that the data analysis end can conveniently determine the influencing factors of the abnormal conditions of the chemical field by analyzing the change of the detection data. The time interval for the data acquisition end to acquire the detection data can be designed according to specific chemical characteristics, and is not limited herein.
The data acquisition end provides basic data samples for the data analysis end to analyze whether abnormal conditions can occur on the chemical field or not by acquiring and storing the detection data of the chemical field in real time. In this application, the data acquisition end realizes the outside one-way transmission of the data of gathering through one-way gateway. That is to say, the detection data provided by the data acquisition end to the data analysis end is transmitted in a single direction, and the data analysis end does not transmit the data to the data acquisition end. The data acquisition end comprises at least one data acquisition device, and the number of the data acquisition devices is not limited in the application.
The data analysis end 103 is used for performing abnormal state analysis; the method comprises the steps of analyzing the video images and the detection data, and generating alarm data files and uploading the alarm data files of the relevant detection data and the relevant video images when abnormal conditions are found.
The data analysis end obtains the related detection data and the related video images of abnormal conditions by analyzing the video images acquired by the video acquisition end and the detection data acquired by the data acquisition end, generates and uploads an alarm data file, and provides early warning information for the safety supervision of a chemical field.
The abnormal condition refers to data of which the detection data does not belong to the normal range value of the corresponding data, for example, the detection data is greater than or less than a preset data threshold value; or an image in which a display phenomenon different from a normal live condition occurs in the video image, for example, a flame phenomenon occurs in the video image.
When the data analysis end analyzes the abnormal state, the abnormal state of the chemical field can be obtained by analyzing and judging the detection data and the video image respectively. When the data analysis end analyzes that the detection data is larger than or smaller than a preset data threshold value, comprehensive judgment is carried out by combining video images of the detection data in a time period, and when the judgment result belongs to an abnormal condition, an alarm data file is generated and uploaded by the related detection data and the related video images; when the data analysis end analyzes the images with the display phenomena different from the normal scene conditions in the video images and the abnormal conditions of the video images, then the detection data corresponding to the video images are called for comprehensive judgment, and when the judgment result belongs to the abnormal conditions, the related detection data and the related video images are generated into alarm data files and uploaded.
As can be seen from the above description, in order to obtain the abnormal situation occurring in the chemical field, the data analysis end of the present application obtains the abnormal situation by analyzing the detection data and the video image, respectively. Therefore, the data analysis end described in the present application adopts different analysis methods for the detection data and the video image to obtain the abnormal situation. This is described in detail below with reference to fig. 2 and 3, respectively.
Please refer to fig. 2, which provides a flowchart of the data analysis end finding abnormal conditions by analyzing the detected data, and the processing steps are as follows:
step S101: and classifying the detection data to obtain the data of various types of detection data in a preset time period.
In the step, the data analysis end is used for counting the data of various types of detection data in a preset time period, and providing a data basis for subsequent data analysis.
The detection data classification refers to classification according to data attributes, for example, environmental data, including: temperature data, humidity data, visibility data, etc.; process parameter data, comprising: voltage data, current data, liquid level data, etc.; chemical reaction data, comprising: combustible gas data, toxic gas data, and the like.
Classifying the detection data to obtain data of various types of detection data in a preset time period, namely classifying various types of detection data according to data attributes, dividing all data of each type of detection data in the preset time period, and preparing for analyzing the detection data by a subsequent data analysis end to obtain abnormal data.
Step S102: and respectively analyzing various types of detection data according to a preset analysis method and parameters, and performing comprehensive analysis by combining different types of data to determine whether abnormal conditions occur.
In the step, the data analysis end is used for analyzing various detection data according to a preset analysis method to determine whether an abnormal condition occurs.
The preset analysis method is used for analyzing and judging the detection data under the abnormal condition in various detection data. Generally, the preset analysis method adopts a variance calculation method, firstly calculates the variance value of all data of each type of detection data in a preset time period, and then performs comprehensive analysis by combining the variance values of different types of detection data to determine whether an abnormal condition occurs.
In addition, the preset analysis method can also preset a data threshold range of various types of detection data belonging to normal conditions, and determine whether abnormal conditions occur or not by comparing whether all the data of each type of detection data in the preset time range have detection data exceeding the specified threshold range or not and if so, performing comprehensive analysis by combining the detection data exceeding the specified threshold range in other types. In addition, other preset methods may also be used in this step to analyze and determine whether the detected data has an abnormal condition, which is not limited herein.
The abnormal condition of the detection data is judged by the method, the data analysis end further calls the video image of the time period of the detection data, the phenomenon whether the abnormal condition exists or not is comprehensively analyzed by combining the video image, and finally the related detection data and the video image corresponding to the detection data are generated into an alarm data file and uploaded.
The above is a step operation method for discovering abnormal conditions by analyzing and detecting data by the data analysis end. Accordingly, the data analysis end can also find abnormal conditions by analyzing the video images.
Referring to fig. 3, a flowchart of a data analysis end finding an abnormal situation by analyzing the video image is provided, and the processing steps are specifically as follows:
step S201: providing the video image to a safety state video identification system trained in advance for safety state identification; the safety state video recognition system is an intelligent image recognition system obtained through training in advance, and the intelligent recognition system can judge whether an abnormal condition occurs or not by reading a video image.
The method comprises the step of judging whether the video image provided by the video acquisition end is abnormal or not in the safe state video identification system.
The safety state video identification system is a video identification system for judging whether the video image has abnormal conditions, and is obtained through deep learning model training. Before a data analysis end provides a video image for the safety state video recognition system to carry out safety recognition, the safety state video recognition system obtains possible abnormal conditions of a chemical field in advance through a deep learning model. The specific learning mode comprises the following steps: and providing the video images marked as the normal state and the video images marked as the abnormal state to a safe state video identification system set by adopting a deep learning principle in advance for training of the safe state video identification system.
In the training process, the marked video image can be divided into two parts, wherein one part is used as training data, and the other part is used as verification data; the training data is provided for the safe state video recognition system for parameter adjustment, and finally the adjustment is carried out until two video images can be accurately recognized, and the safe state video recognition system can be enabled after the training data is large enough; after the parameters are adjusted, the verification data can be provided for the safe state video identification system, and the safe state video identification system identifies the verification data, judges whether the verification data belongs to an abnormal state, and then compares the verification data with a label carried out in advance to judge whether the identification result is accurate. The method can be used for scoring the trained safety state video recognition system, and if the score is qualified, the safety state video recognition system can be used. The safety state video recognition system can also be a multi-classification system, namely whether the image belongs to an anomaly or not is recognized, the type of the anomaly is further recognized, and the specific training process is similar to the mode described above.
The working essence of the safety state video identification system is to identify specific characteristics of images, and belongs to the field of image identification.
The safety state video identification system can be contained in the data analysis end and can be independent of the data analysis end, and when the video image is detected, the data analysis end transmits the video image to the system through a network.
Step S202: and judging whether the video image has abnormal conditions or not according to the safety state identification result of the safety state identification system.
The method is used for judging whether the video image has abnormal conditions.
The safety state recognition result of the safety state video recognition system refers to the result of recognizing the video image by using the safety state video recognition system described in step S201, and the specific judgment details refer to the description in step S201.
The above is a specific operation description that the data analysis end finds an abnormal situation by analyzing the video image, and if the data analysis end finds that the abnormal situation exists in the video image, the data analysis end determines a relevant time period of the detection data according to the type of the abnormal situation, and acquires the detection data in the time period as the relevant detection data. In other words, after the data analysis end finds an abnormal situation by analyzing the video image, the data analysis end comprehensively analyzes the video image and the related detection data by combining the detection data of the video image in the related time period, determines whether the abnormal situation exists, and generates an alarm data file by using the video image and the related detection data if the abnormal situation exists. The related time period is determined according to the type of the abnormal condition and is possibly related to the abnormal condition of the type; for example, when a combustion abnormality occurs, it is considered that temperature detection data from one minute before the image is generated to the present time is retrieved; and if the gas escape abnormity occurs, retrieving the image from 5 minutes before to the current pressure detection data.
Fig. 2 and fig. 3 describe the specific operation steps of the data analysis end for finding the abnormal situation of the chemical field by analyzing the detection data and the video image respectively.
When analyzing the current detection data or the current video image, the data analysis end also compares and analyzes the current detection data or the current video image with historical data.
The data analysis end is also used for generating a real-time data file from the detection data and the video image corresponding to the detection data and storing the real-time data file as historical data; and when the abnormal state analysis is carried out, the abnormal state analysis comprises comparison analysis with the historical data.
The detection data provided by the data acquisition end and the video image provided by the video acquisition end comprise time information for recording the detection data and the video image, and the corresponding relation between the detection data and the video image is established according to the time information. When the data analysis end receives the detection data and the video images, the detection data and the video images corresponding to the detection data are generated into real-time data files according to the time information and are stored as historical data, and a historical data basis is provided for the data analysis end to analyze abnormal states. In addition, when the data analysis end analyzes the abnormal state, the abnormal state analysis end not only analyzes and compares the abnormal state with the historical data, but also performs comprehensive analysis by combining basic data which is pre-recorded by the data analysis end.
In addition, the data analysis end also comprises an abnormal state evaluation model, and the abnormal state evaluation model is adopted to analyze the production data information of the chemical field, which specifically comprises the following steps:
the data analysis end is specifically used for acquiring production data information of the chemical field, inputting the production data information of the chemical field into an abnormal state evaluation model, and acquiring abnormal state alarm grade information which is output by the abnormal state evaluation model and corresponds to the production data information of the chemical field.
It should be noted that the production data information of the chemical site includes at least one of video data information of the chemical site and detection data information of the chemical site.
The abnormal state evaluation model is obtained by pre-training according to the historical production data information and the historical production data abnormal state alarm level corresponding to the historical production data information. Specifically, the abnormal state evaluation model includes a video data abnormal state evaluation model and a detected data abnormal state evaluation model. The video data abnormal state evaluation model is obtained by pre-training according to historical video data information and historical video data abnormal state alarm levels corresponding to the historical video data; the detection data abnormal state evaluation model is obtained by pre-training according to historical detection data information and the historical detection data abnormal state alarm grade corresponding to the historical detection data information.
Specifically, the abnormal state evaluation model comprises a smoke abnormal state evaluation model and a water mist abnormal state evaluation model. For example, if flame and smoke appear in the video image detected by the video state monitoring end, the data analysis end obtains the video data containing the flame and smoke, inputs the video data abnormal state evaluation model, and obtains the video data abnormal state alarm level corresponding to the video data of the flame and smoke output by the video data abnormal state evaluation model. For example, if the image has a first-level flame, i.e., a small flame, and belongs to a medium-sized abnormal state, the video abnormal state alarm level is a first-level alarm, which indicates that an experiment of a chemical field needs to be stopped, and the position of the first-level smoke on the chemical field is detected and repaired. If secondary flame, namely large flame, appears in the image, dense fog appears in the image, and the image belongs to a severe abnormal state, the video abnormal state alarm grade is secondary alarm, which indicates that the experiment of a chemical field needs to be stopped urgently for emergency accident treatment.
Secondly, the data analysis end is also used for judging whether the duration time of the on-site abnormal state of the chemicals corresponding to the abnormal state alarm grade information is greater than a preset time threshold value; if the judgment result is yes, taking the abnormal state alarm grade information and the production data information of the chemical field corresponding to the abnormal state alarm grade information as negative samples; if the judgment result is negative, taking the abnormal state alarm grade information and the production data information of the chemical field corresponding to the abnormal state alarm grade information as a positive sample; and training the abnormal state evaluation model according to the positive sample and the negative sample.
Specifically, for example, the data analysis end receives that the video image sent by the video acquisition end has the first-level flame, the data analysis end inputs the video data into the video data abnormal state evaluation model, and the abnormal state alarm level information output by the video data abnormal state evaluation model is the first-level alarm. After the preset time, the data analysis end inputs the obtained video image data collected by the video collection end into the video data abnormal state evaluation model, and the result output by the video data abnormal state evaluation model shows that the primary alarm disappears, which represents that the sample is a positive sample. If the result output by the video data abnormal state evaluation model in the data analysis end is displayed or is subjected to primary alarm after the preset time, the evaluation and analysis of the video image data of the video acquisition end by the video data abnormal state evaluation model has errors, and the errors cannot actually represent the actual situation of a chemical field, so that the sample is a negative sample. And training and optimizing the abnormal state evaluation model according to a large number of positive samples and negative samples, so that the evaluation accuracy of the abnormal state evaluation model is improved.
The data analysis end 103 further comprises: a basic data input end 103-1, an equipment parameter input end 103-2 and a chemical data input end 103-3; these ports provide a relevant knowledge background so that relevant parameters can be obtained when analyzing the field situation.
And the basic data input end 103-1 is used for inputting the on-site basic data of the dangerous chemicals and sending the on-site basic data to a dangerous chemical registration management system, a dangerous chemical GIS application system and a chemical safety knowledge base.
The field basic data of the dangerous chemicals comprises: information of the place where the dangerous chemicals are located, information of major dangerous sources, chemical information and process information.
The basic data input end sends the on-site basic data of the dangerous chemicals to a dangerous chemical registration management system, a dangerous chemical GIS application system and a chemical safety knowledge base, a shared database is established for the basic data of the dangerous chemicals, and a data base is provided for remote monitoring of the dangerous chemicals.
And the equipment parameter recording end 103-2 is used for recording equipment parameters of the equipment involved in the storage and use processes of the chemicals.
The equipment parameters of the equipment related in the storage and use processes of the chemicals are the equipment adjustment parameter range required by the chemicals set according to the basic physical and chemical properties of the chemicals; different chemicals have different physicochemical properties, and therefore, depending on the basic physicochemical properties of the chemicals, the apparatus should adjust the apparatus parameters corresponding to the chemical properties.
The chemical data input end 103-3 is used for inputting data of raw material chemicals and data of intermediate chemicals generated in the chemical treatment process.
The raw material chemicals are raw materials for synthesizing the chemicals, and the method for synthesizing the chemicals from the raw material chemicals is that the raw material chemicals are generated through a series of chemical reactions under certain experimental conditions, and in the process of the series of chemical reactions, the raw material chemicals can generate at least one intermediate chemical. Therefore, the chemical data entry end needs to record data of raw chemicals and intermediate chemicals and record a chemical synthesis principle.
And after receiving the input data, the data analysis end comprehensively analyzes whether abnormal conditions exist or not by combining the input data when analyzing the detection data and the video data.
When the data analysis end analyzes the abnormal state, the detection data of the chemical field and the video image are combined with the basic data of the dangerous chemical field, the data of the raw material chemicals and the intermediate chemical data generated in the chemical processing process to carry out comprehensive comparison analysis, and a data base is provided for generating an alarm data file.
In addition, after the data analysis end generates an alarm data file from the related video image and the related detection data with the abnormal condition and before the alarm data file is uploaded to the server, the data analysis end is also used for judging the influence level of the abnormal condition on the safety of the chemicals and determining the uploading level of the alarm data file according to the influence level.
That is to say, the greater the degree of influence of the abnormal conditions existing in the video images and the related detection data on the safety of the chemicals, the higher the uploading level of the corresponding alarm data file. And the data analysis end selects the corresponding uploading level according to the influence degree of the abnormal condition on the safety of the chemicals. Please refer to fig. 4, which is a schematic diagram illustrating that the data analysis end uploads according to the uploading level of the alarm data file during the production process of the enterprise chemicals.
For example, (1) when the influence level of the abnormal condition on the safety of the chemical is a first level, the uploading early warning level of the alarm data file is a first early warning level, and the data analysis end is specifically used for pushing the alarm data file with the first early warning level and early warning processing feedback to the enterprise end.
(2) And when the influence level of the abnormal condition on the safety of the chemicals is a second level, the uploading early warning level of the alarm data file is the second early warning level, and the data analysis end is specifically used for pushing the alarm data file of the second early warning level to the chemical safety monitoring system of the county/city level or the following units so as to enable the county/city level unit supervision enterprises to implement early warning processing feedback.
(3) And when the influence level of the abnormal condition on the safety of the chemicals is a third level, the uploading early warning level of the alarm data file is the third early warning level, and the data analysis end is specifically used for pushing the alarm data file of the third early warning level to the chemical safety monitoring system of the provincial level or the following units so that the provincial level units deploy dangerous chemical enterprises and major hazard source risk distribution maps in the district, and data support is provided for emergency handling of accidents.
(4) And when the influence level of the abnormal condition on the safety of the chemicals is a fourth level, the uploading early warning level of the alarm data file is the fourth early warning level, the data analysis end is specifically used for pushing the alarm data file of the fourth early warning level to a provincial-level or lower unit chemical safety monitoring system, and the provincial-level unit chemical safety monitoring system pushes the alarm data file of the fourth early warning level to a department-level unit, so that the department-level unit dynamically analyzes the macroscopic trend risk distribution map of the dangerous chemicals in real time.
In addition, on the basis of the first logic structure schematic diagram of the chemical safety monitoring system, an emergency processing end 104 is added to the second logic structure schematic diagram of the chemical safety monitoring system provided by the application, as shown in fig. 5.
The chemical safety monitoring system further comprises an emergency processing end 104, wherein the emergency processing end 104 is used for generating an emergency processing scheme and remotely controlling the site, and comprises the steps of calling video data and carrying out video inspection through a video acquisition end.
The emergency processing scheme is a processing scheme for remotely controlling the emergency accidents on line by calling video images of the chemical field to analyze in time according to the alarm data file generated by the data analysis end and the corresponding uploading level when the emergency processing scheme is used for processing the emergency accidents on the chemical field.
The remote control of the site refers to that the emergency processing end timely processes abnormal conditions of a chemical site according to the alarm data file and the corresponding early warning grade, online video inspection is carried out by calling site video data, and the emergency processing end generates a corresponding emergency processing scheme according to the danger grade of the abnormal conditions appearing in the video image so as to remotely process the abnormal conditions of the chemical site online.
The application provides a chemicals safety monitoring system, through the video image and the detection data of analysis chemicals scene, in time will appear the relevant video image and the relevant detection data generation alarm data file of abnormal conditions and upload, effectively improved the early warning propelling movement efficiency to the abnormal conditions, reduce the accident rate of major hazard source. In addition, the chemical safety monitoring system can process video images and detection data of a chemical field in time, generate and upload alarm data files, can avoid the problem of mass data transmission caused by directly transmitting the field detection data and the corresponding image videos to the server, and reduces the transmission pressure of data streams.
Although the present application has been described with reference to the preferred embodiments, it is not intended to limit the present application, and those skilled in the art can make variations and modifications without departing from the spirit and scope of the present application, therefore, the scope of the present application should be determined by the claims that follow.
Claims (10)
1. A chemical safety monitoring system, comprising:
the system comprises a video acquisition end, a data acquisition end and a data analysis end;
the video acquisition end is used for acquiring video images of chemical fields and sending the video images to the data analysis end;
the data acquisition end is used for acquiring and storing detection data of a chemical field in real time and sending the detection data to the data analysis end in real time;
the data analysis end is used for analyzing abnormal states; the method comprises the steps of analyzing the video images and the detection data, and generating alarm data files and uploading the alarm data files of the relevant detection data and the relevant video images when abnormal conditions are found.
2. The chemical safety monitoring system according to claim 1, wherein the data analysis end finds an abnormal situation by analyzing the detection data, and specifically comprises the following steps:
classifying the detection data to obtain data of various types of detection data in a preset time period;
and respectively analyzing various types of detection data according to a preset analysis method and parameters, and performing comprehensive analysis by combining different types of data to determine whether abnormal conditions occur.
3. The chemical safety monitoring system according to claim 1, wherein the data analysis end finds an abnormal situation by analyzing the video image, and specifically comprises the following steps:
providing the video image to a safety state video identification system trained in advance for safety state identification; the safety state video recognition system is an intelligent image recognition system obtained by training in advance, and the intelligent recognition system can judge whether an abnormal condition occurs or not by reading a video image;
and judging whether the video image has abnormal conditions or not according to the safety state identification result of the safety state identification system.
4. The chemical safety monitoring system according to any one of claims 1 to 3, wherein the data analysis end is further configured to determine an impact level of the abnormal condition on chemical safety, and determine an uploading level of the alarm data file according to the impact level.
5. The chemical safety monitoring system of claim 1, wherein the data analysis end further comprises a basic data entry end, an equipment parameter entry end, a chemical data entry end;
the basic data input end is used for inputting the on-site basic data of the dangerous chemicals and sending the on-site basic data to a dangerous chemical registration management system, a dangerous chemical GIS application system and a chemical safety knowledge base;
the equipment parameter input end is used for inputting equipment parameters of equipment related in the storage and use processes of chemicals;
the chemical data input end is used for inputting data of raw material chemicals and intermediate chemical data generated in the chemical processing process;
and the data analysis end receives the input data and uses the input data in the abnormal state analysis process.
6. The chemical production safety monitoring system of claim 1, wherein the video acquisition end is further configured to adjust the brightness of a fill light in the video acquisition end to obtain a video image satisfying the brightness required by the data analysis end if the brightness of the obtained video image is too dark.
7. The chemical safety production monitoring system according to claim 1, wherein the data analysis end is further configured to generate a real-time data file from the detection data and the video image corresponding to the detection data and store the real-time data file as historical data; and when the abnormal state analysis is carried out, the comparison analysis with the historical data is included.
8. The chemical safety production monitoring system according to claim 1, further comprising an emergency processing terminal for generating an emergency processing scheme and performing remote control on the site, including calling video data and performing video inspection through the video acquisition terminal.
9. The chemical safety production monitoring system according to claim 1, wherein the data acquisition end realizes unidirectional transmission of the acquired data to the outside through a unidirectional gateway.
10. The system for monitoring the safe production of chemicals according to claim 1, wherein the data analysis end is specifically configured to obtain production data information of the chemical field, input the production data information of the chemical field into an abnormal state evaluation model, and obtain abnormal state alarm level information corresponding to the production data information of the chemical field output by the abnormal state evaluation model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011194039.3A CN114442512A (en) | 2020-10-30 | 2020-10-30 | Chemical safety monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011194039.3A CN114442512A (en) | 2020-10-30 | 2020-10-30 | Chemical safety monitoring system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114442512A true CN114442512A (en) | 2022-05-06 |
Family
ID=81357246
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011194039.3A Pending CN114442512A (en) | 2020-10-30 | 2020-10-30 | Chemical safety monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114442512A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115164993A (en) * | 2022-07-26 | 2022-10-11 | 正人智慧科技(苏州)有限公司 | Pump room environment temperature and humidity monitoring and early warning method and system |
| CN117197754A (en) * | 2023-10-19 | 2023-12-08 | 航天正通汇智(北京)科技股份有限公司 | Security monitoring method based on ultra-clear wide-amplitude video |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109637076A (en) * | 2018-11-09 | 2019-04-16 | 无锡市安泰安全技术服务有限公司 | A kind of hazardous chemical major hazard source on-line monitoring pre-alarm system |
| CN110120143A (en) * | 2018-02-07 | 2019-08-13 | 中国石油化工股份有限公司 | A kind of inflammable and explosive region smog intelligent monitor system of Petrochemical Enterprises |
| CN111144700A (en) * | 2019-12-03 | 2020-05-12 | 中国五洲工程设计集团有限公司 | Safety monitoring and early warning system applied to dangerous chemical production |
| CN111464793A (en) * | 2020-05-15 | 2020-07-28 | 哈尔滨工业大学 | Method and system for monitoring and early warning experimental device based on artificial intelligence |
-
2020
- 2020-10-30 CN CN202011194039.3A patent/CN114442512A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110120143A (en) * | 2018-02-07 | 2019-08-13 | 中国石油化工股份有限公司 | A kind of inflammable and explosive region smog intelligent monitor system of Petrochemical Enterprises |
| CN109637076A (en) * | 2018-11-09 | 2019-04-16 | 无锡市安泰安全技术服务有限公司 | A kind of hazardous chemical major hazard source on-line monitoring pre-alarm system |
| CN111144700A (en) * | 2019-12-03 | 2020-05-12 | 中国五洲工程设计集团有限公司 | Safety monitoring and early warning system applied to dangerous chemical production |
| CN111464793A (en) * | 2020-05-15 | 2020-07-28 | 哈尔滨工业大学 | Method and system for monitoring and early warning experimental device based on artificial intelligence |
Non-Patent Citations (2)
| Title |
|---|
| 康荣学: "基于GIS的重大危险源安全监测预警系统研究与开发", 中国安全生产科学技术, no. 03, pages 110 - 115 * |
| 方来华: "危化品生产储存使用全过程安全监控与监管系统", 中国安全生产科学技术, no. 07, pages 114 - 117 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115164993A (en) * | 2022-07-26 | 2022-10-11 | 正人智慧科技(苏州)有限公司 | Pump room environment temperature and humidity monitoring and early warning method and system |
| CN115164993B (en) * | 2022-07-26 | 2024-10-15 | 正人智慧科技(苏州)有限公司 | Pump room environment temperature and humidity monitoring and early warning method and system |
| CN117197754A (en) * | 2023-10-19 | 2023-12-08 | 航天正通汇智(北京)科技股份有限公司 | Security monitoring method based on ultra-clear wide-amplitude video |
| CN117197754B (en) * | 2023-10-19 | 2024-04-30 | 航天正通汇智(北京)科技股份有限公司 | Security monitoring method based on ultra-clear wide-amplitude video |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113138865A (en) | Environment monitoring method, system, platform and storage medium | |
| CN108189043B (en) | Inspection method and inspection robot system applied to high-speed rail machine room | |
| CN111476921B (en) | Intelligent inspection system for machine room | |
| CN110751451B (en) | Laboratory big data management system | |
| CN112115927B (en) | Intelligent machine room equipment identification method and system based on deep learning | |
| CN111080775A (en) | Server routing inspection method and system based on artificial intelligence | |
| CN109564166A (en) | The system and method checked for automatic and artificial combination | |
| CN114442512A (en) | Chemical safety monitoring system | |
| CN117649752B (en) | Monitoring, early warning and disposing method, device, equipment and medium for water supply network | |
| CN111966034B (en) | Intelligent water affair system | |
| CN112258362B (en) | Dangerous source identification method, system and readable storage medium | |
| CN116010826A (en) | Construction safety early warning method and system for building engineering | |
| CN117035419B (en) | Intelligent management system and method for enterprise project implementation | |
| CN114925980A (en) | Production safety management system | |
| CN110865596A (en) | Raise dust on-line monitoring and management and control system | |
| CN104574729A (en) | Alarming method, device and system | |
| CN110148290A (en) | Information-based big data system is supervised in the early warning of Intellisense Mine Safety in Production and prevention and control | |
| CN113835387A (en) | Operation and maintenance management method, system and medium | |
| CN113392706A (en) | Device and method for detecting smoking and using mobile phone behaviors | |
| CN117168536A (en) | Python-based protection area ecological environment monitoring system | |
| CN114768158B (en) | Intelligent fire fighting system and automatic inspection method thereof | |
| CN116055521A (en) | Inspection system and image recognition method for electric inspection robot | |
| CN114266483A (en) | Dangerous waste monitoring system based on Internet of things | |
| CN118400395B (en) | Industrial Internet security monitoring method and system based on equipment user portrait | |
| CN115990327B (en) | Intelligent fire control management system based on thing networking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |