AI automatic monitoring and early warning system for oil refining chemical plant area and use method of AI automatic monitoring and early warning system

Abstract

The invention discloses an AI automatic monitoring and early warning system for an oil refining chemical plant area and a use method thereof, and relates to the technical field of artificial intelligence and video analysis, and the AI automatic monitoring and early warning system comprises a data acquisition module, a preprocessing module, an AI analysis module, a danger awareness and early warning module and a feedback learning module, the data acquisition module comprises an image acquisition module, a sound acquisition module, a pressure acquisition module and a temperature acquisition module. The system is integrated with an existing safety system, an existing monitoring system, an existing equipment management system and the like in an oil refining chemical plant area to form a unified data platform and a unified management center, a visual user interface is designed, and the system has the functions of real-time video display, early warning information prompt, system state monitoring, historical data analysis and the like; therefore, effective monitoring and operation by management personnel are facilitated, the safety condition of the oil refining chemical plant area can be monitored in real time, potential risks can be automatically identified and pre-warned, the production process can be optimized, and the performance can be continuously improved through deep learning.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and video analysis technology, and specifically to an AI automatic monitoring and early warning system for an oil refinery and chemical plant area and a method for using the system. Background Art

[0002] A refinery is a complex industrial facility that converts crude oil into usable petroleum products. It also involves the production of chemical raw materials. The rapid development of artificial intelligence technology, especially breakthroughs in computer vision and deep learning, has greatly improved the intelligent analysis capabilities of machines for images and videos. However, in heavy industries such as refineries and chemicals, despite the continuous improvement of automation, they still face severe challenges in safety monitoring, environmental management, and production efficiency optimization. Traditional monitoring systems rely on manual supervision and simple rule triggering, making it difficult to accurately identify potential risks in complex scenarios in real time, and are unable to provide in-depth data analysis and predictive insights.

[0003] With the rapid development of the petroleum refining and chemical industry, the demand for safe production is also increasing. In the existing technology, most petroleum refining and chemical plant areas use traditional video surveillance systems for safety monitoring. These systems rely on manual monitoring, cannot achieve automatic identification and early warning, are inefficient, and are prone to omissions. Summary of the invention

[0004] The purpose of the present invention is to provide an AI automatic monitoring and early warning system for a petroleum refinery and a method of using the system, which has the advantage of automatic monitoring and solves the problem that most petroleum refineries and petroleum refineries use traditional video surveillance systems for security monitoring in the prior art. These systems rely on manual monitoring, cannot achieve automatic identification and early warning, are inefficient, and are prone to omissions.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an AI automatic monitoring and early warning system for a petroleum refinery and chemical plant area and a method of using the same, comprising a data acquisition module, a preprocessing module, an AI analysis module, a hazard awareness and early warning module and a feedback learning module, the data acquisition module comprising an image acquisition module, a sound acquisition module, a pressure acquisition module and a temperature acquisition module, the preprocessing module comprising a video formatting and compression unit, an image enhancement processing unit, the AI ​​analysis module comprising a behavior recognition unit, a state recognition and state monitoring unit, the hazard awareness and early warning module comprising a real-time early warning unit and an early warning level classification unit, the feedback learning module comprising a feedback unit and a continuous learning and optimization unit.

[0006] Preferably, the image acquisition module deploys high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area to monitor 24 hours a day, covering key and sensitive areas, and collects sensor data of sound, temperature, and pressure through acoustic sensors, pressure sensors, and temperature sensors.

[0007] Preferably, the video formatting and compression unit is used to format and compress the collected original video stream, and the image enhancement processing unit enhances the video image quality by adjusting brightness, contrast and sharpening.

[0008] Preferably, the behavior recognition unit uses a method combining convolutional neural networks and recurrent neural networks to identify human behavior and equipment behavior in the video, the status monitoring unit uses a deep learning algorithm to analyze the equipment operating status, and the danger detection unit uses specific image recognition and pattern analysis technology to quickly and accurately issue an early warning.

[0009] Preferably, the real-time warning unit is used to immediately send warning information to the central control room and automatically initiate response measures when the AI ​​analysis core identifies abnormal behavior or potential danger. The warning level classification unit can classify the warning information into multiple levels according to the identified risk level.

[0010] Preferably, the feedback unit establishes a feedback mechanism so that the processing results after each warning can be fed back to the system, including corrections of false alarms and missed alarms, and records of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results and self-optimize the algorithm model through machine learning and deep learning technologies.

[0011] Preferably, the method of use is as follows:

[0012] A. The image acquisition module deploys high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area for 24-hour monitoring, covering key and sensitive areas, and collects sensor data of sound, temperature, and pressure through acoustic sensors, pressure sensors, and temperature sensors to enhance the accuracy and dimension of AI analysis;

[0013] B. The video formatting and compression unit is used to format and compress the collected original video stream while ensuring that key details are not lost. The image enhancement processing unit enhances the video image quality by adjusting brightness, contrast, and sharpening to ensure the accuracy of subsequent AI analysis;

[0014] C. The behavior recognition unit uses a combination of convolutional neural networks and recurrent neural networks to identify the behaviors of people and equipment in the video, such as whether they are wearing safety equipment and whether there are any illegal operations. The status monitoring unit uses deep learning algorithms to analyze the operating status of equipment and uses specific image recognition and pattern analysis technologies to quickly and accurately issue warnings for dangerous situations such as leakage, fire, and smoke.

[0015] D. The real-time warning unit is used to send warning information to the central control room immediately when the AI ​​analysis core identifies abnormal behavior or potential danger, and automatically initiate response measures, such as emergency shutdown and activation of the fire protection system. The warning level classification unit can classify the warning information into multiple levels according to the identified risk level, so that managers can take different emergency measures according to the warning level;

[0016] E. The feedback unit establishes a feedback mechanism so that the processing results after each warning can be fed back to the system, including the correction of false alarms and missed alarms, and the record of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results through machine learning and deep learning technologies, self-optimize algorithm models, improve recognition accuracy and reduce false alarm rates. This AI automatic monitoring and early warning system is integrated with the existing safety systems, monitoring systems, equipment management systems, etc. in the refinery and chemical plant area to form a unified data platform and management center. An intuitive user interface is designed, including real-time video display, early warning information prompts, system status monitoring, historical data analysis and other functions to facilitate management personnel to effectively monitor and operate.

[0017] Compared with the prior art, the present invention has the following beneficial effects:

[0018] The present invention deploys high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area for 24-hour monitoring, and collects sensor data such as sound, temperature, and pressure through acoustic sensors, pressure sensors, and temperature sensors, thereby enhancing the accuracy and dimension of AI analysis. The original video stream collected by the video formatting and compression unit is formatted and compressed to ensure that key details are not lost. The image enhancement processing unit adjusts the brightness, contrast, and sharpening to enhance the video image quality and ensure the accuracy of subsequent AI analysis. The behavior recognition unit uses a method combining convolutional neural networks and recurrent neural networks to identify the behavior of personnel and equipment in the video, such as whether safety equipment is worn and whether there is any illegal operation. The status monitoring unit uses a deep learning algorithm to analyze the operating status of the equipment. For dangerous situations such as leakage, fire, and smoke, specific image recognition and pattern analysis technologies are used to quickly and accurately issue warnings. The real-time warning unit is used to immediately send a warning signal to the central control room when the AI ​​analysis core identifies abnormal behavior or potential danger. The warning level classification unit can classify the warning information into multiple levels according to the identified risk level, so that managers can take different emergency measures according to the warning level. The feedback unit establishes a feedback mechanism so that the processing results after each warning can be fed back to the system, including the correction of false alarms and missed alarms, as well as the record of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results through machine learning and deep learning technology, self-optimize algorithm models, improve recognition accuracy and reduce false alarm rate. This AI automatic monitoring and early warning system is integrated with the existing safety system, monitoring system, equipment management system, etc. in the refinery and chemical plant area to form a unified data platform and management center. An intuitive user interface is designed, including real-time video display, warning information prompts, system status monitoring, historical data analysis and other functions, so that managers can effectively monitor and operate. It can monitor the safety status of the refinery and chemical plant area in real time, automatically identify and warn potential risks, optimize production processes, and continuously improve its performance through deep learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the system framework of the present invention;

[0020] Figure 2 This is a schematic diagram of the data acquisition module framework of the present invention;

[0021] Figure 3 This is a schematic diagram of the framework of the preprocessing module of the present invention;

[0022] Figure 4 This is a schematic diagram of the AI ​​analysis module framework of the present invention;

[0023] Figure 5 This is a schematic diagram of the module framework of the hazard identification and early warning module of the present invention;

[0024] Figure 6 It is a schematic diagram of the module framework of the feedback and learning mechanism of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Example 1

[0027] like Figure 1-6 As shown, it is the first embodiment of the present invention, which provides an AI automatic monitoring and early warning system for a refinery and a method of using the system, including a data acquisition module, a preprocessing module, an AI analysis module, a hazard awareness and early warning module and a feedback learning module. The data acquisition module includes an image acquisition module, a sound acquisition module, a pressure acquisition module and a temperature acquisition module. The preprocessing module includes a video formatting and compression unit and an image enhancement processing unit. The AI ​​analysis module includes a behavior recognition unit, a state recognition and state monitoring unit. The hazard awareness and early warning module includes a real-time early warning unit and an early warning level classification unit. The feedback learning module includes a feedback unit and a continuous learning and optimization unit.

[0028] like Figure 1-6As shown in the figure, by deploying high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area for 24-hour monitoring, and collecting sound, temperature, pressure and other sensor data through acoustic sensors, pressure sensors and temperature sensors, the accuracy and dimension of AI analysis can be enhanced, and the original video stream collected by the video formatting and compression unit is formatted and compressed to ensure that key details are not lost. The image enhancement processing unit adjusts the brightness, contrast, and sharpening to enhance the video image quality and ensure the accuracy of subsequent AI analysis. The behavior recognition unit uses a combination of convolutional neural networks and recurrent neural networks to identify the behavior of personnel and equipment in the video, such as whether safety equipment is worn and whether there is any illegal operation. The status monitoring unit uses deep learning algorithms to analyze the operating status of the equipment. For dangerous situations such as leakage, fire, and smoke, specific image recognition and pattern analysis technologies are used to quickly and accurately issue warnings. The real-time warning unit is used to immediately send a warning signal to the central control room when the AI ​​analysis core identifies abnormal behavior or potential dangers. The warning level classification unit can classify the warning information into multiple levels according to the identified risk level, so that managers can take different emergency measures according to the warning level. The feedback unit establishes a feedback mechanism so that the processing results after each warning can be fed back to the system, including the correction of false alarms and missed alarms, as well as the record of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results through machine learning and deep learning technology, self-optimize algorithm models, improve recognition accuracy and reduce false alarm rate. This AI automatic monitoring and early warning system is integrated with the existing safety system, monitoring system, equipment management system, etc. in the refinery and chemical plant area to form a unified data platform and management center. An intuitive user interface is designed, including real-time video display, warning information prompts, system status monitoring, historical data analysis and other functions, so that managers can effectively monitor and operate. It can monitor the safety status of the refinery and chemical plant area in real time, automatically identify and warn potential risks, optimize production processes, and continuously improve its performance through deep learning.

[0029] Example 2

[0030] Reference Figure 2-3 , which is the second embodiment of the present invention, and this embodiment is based on the previous embodiment.

[0031] In this embodiment, the image acquisition module deploys high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area to monitor 24 hours a day, covering key and sensitive areas, and collects sensor data such as sound, temperature, and pressure through acoustic sensors, pressure sensors, and temperature sensors.

[0032] The video formatting and compression unit is used to format and compress the collected original video stream, and the image enhancement processing unit enhances the video image quality by adjusting brightness, contrast and sharpening.

[0033] like Figure 2-3 As shown, by deploying high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area for 24-hour monitoring, and collecting sound, temperature, pressure and other sensor data through acoustic sensors, pressure sensors and temperature sensors, the accuracy and dimension of AI analysis can be enhanced. The original video stream collected by the video formatting and compression unit is formatted and compressed to ensure that key details are not lost. The image enhancement processing unit adjusts the brightness, contrast, and sharpening to enhance the video image quality and ensure the accuracy of subsequent AI analysis.

[0034] Example 3

[0035] Reference Figure 4-6 , which is the third embodiment of the present invention, and this embodiment is based on the previous two embodiments.

[0036] In this embodiment, the behavior recognition unit uses a method that combines convolutional neural networks and recurrent neural networks to identify human behavior and equipment behavior in the video, the status monitoring unit uses a deep learning algorithm to analyze the equipment operating status, and the danger detection unit uses specific image recognition and pattern analysis technology to quickly and accurately issue early warnings.

[0037] The real-time warning unit is used to immediately send warning information to the central control room and automatically initiate response measures when the AI ​​analysis core identifies abnormal behavior or potential danger. The warning level classification unit can classify the warning information into multiple levels according to the identified risk level.

[0038] The feedback unit establishes a feedback mechanism so that the processing results after each warning can be fed back to the system, including corrections of false alarms and missed alarms, and records of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results and self-optimize the algorithm model through machine learning and deep learning technologies.

[0039] like Figure 4-6As shown in the figure, the behavior recognition unit uses a method combining convolutional neural networks and recurrent neural networks to identify the human behavior and equipment behavior in the video, such as whether safety equipment is worn and whether there is any illegal operation. The status monitoring unit uses deep learning algorithms to analyze the operating status of the equipment. For dangerous situations such as leakage, fire, and smoke, specific image recognition and pattern analysis technologies are used to quickly and accurately issue warnings. The real-time warning unit is used to immediately send warning information to the central control room when the AI ​​analysis core identifies abnormal behavior or potential danger, and automatically initiate response measures, such as emergency shutdown and starting the fire protection system. The warning level classification unit can classify the warning information into multiple levels according to the degree of risk identified, so that managers can take different emergency measures according to the warning level. The feedback unit establishes a feedback mechanism so that the processing results after each warning can be fed back to the system, including the correction of false alarms and missed alarms, as well as the record of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results through machine learning and deep learning technologies, self-optimize the algorithm model, improve recognition accuracy and reduce false alarm rate.

[0040] An AI automatic monitoring and early warning system for a petroleum refinery and a method for using the same, the method comprising the following steps:

[0041] The image acquisition module deploys high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area for 24-hour monitoring, covering key and sensitive areas. It also collects sound, temperature, pressure and other sensor data through acoustic sensors, pressure sensors and temperature sensors to enhance the accuracy and dimension of AI analysis. The video formatting and compression unit is used to format and compress the collected original video stream while ensuring that key details are not lost. The image enhancement processing unit enhances the video image quality by adjusting brightness, contrast, and sharpening to ensure the accuracy of subsequent AI analysis. The behavior recognition unit uses a combination of convolutional neural networks and recurrent neural networks to identify the behavior of people and equipment in the video, such as whether they are wearing safety equipment and whether there are any illegal operations. The status monitoring unit uses deep learning algorithms to analyze the operating status of the equipment. For dangerous situations such as leakage, fire, and smoke, it uses specific image recognition and pattern analysis technologies to quickly and accurately issue warnings. The real-time warning unit uses When the AI ​​analysis core identifies abnormal behavior or potential danger, it immediately sends an early warning message to the central control room and automatically initiates response measures, such as emergency shutdown and activation of the fire protection system. The early warning level classification unit can classify the early warning information into multiple levels according to the degree of risk identified, so that managers can take different emergency measures according to the early warning level. The feedback unit establishes a feedback mechanism so that the processing results after each early warning can be fed back to the system, including the correction of false alarms and missed alarms, as well as the record of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results through machine learning and deep learning technology, self-optimize algorithm models, improve recognition accuracy and reduce false alarm rate. This AI automatic monitoring and early warning system is integrated with the existing safety system, monitoring system, equipment management system, etc. in the refinery and chemical plant area to form a unified data platform and management center, and design an intuitive user interface, including real-time video display, early warning information prompts, system status monitoring, historical data analysis and other functions, so that managers can effectively monitor and operate.

[0042] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0043] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An AI automatic monitoring and early warning system for a petroleum refining and chemical plant area, comprising a data acquisition module, a preprocessing module, an AI analysis module, a hazard awareness and early warning module and a feedback learning module, characterized in that: The data acquisition module includes an image acquisition module, a sound acquisition module, a pressure acquisition module and a temperature acquisition module; the preprocessing module includes a video formatting and compression unit and an image enhancement processing unit; the AI ​​analysis module includes a behavior recognition unit, a state recognition and state monitoring unit; the danger awareness and warning module includes a real-time warning unit and a warning level classification unit; and the feedback learning module includes a feedback unit and a continuous learning and optimization unit.

2. The AI ​​automatic monitoring and early warning system for a petroleum refinery and chemical plant according to claim 1 is characterized by: The image acquisition module deploys high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area for 24-hour monitoring, covering key and sensitive areas, and collects sensor data of sound, temperature, and pressure through acoustic sensors, pressure sensors, and temperature sensors.

3. The AI ​​automatic monitoring and early warning system for a petroleum refinery and chemical plant according to claim 1 is characterized by: The video formatting and compression unit is used to format and compress the collected original video stream, and the image enhancement processing unit enhances the video image quality by adjusting brightness, contrast and sharpening.

4. The AI ​​automatic monitoring and early warning system for a petroleum refinery and chemical plant according to claim 1 is characterized by: The behavior recognition unit uses a method combining convolutional neural networks and recurrent neural networks to identify human behavior and equipment behavior in the video. The status monitoring unit uses a deep learning algorithm to analyze the equipment operating status. The danger detection unit uses specific image recognition and pattern analysis technology to issue early warnings quickly and accurately.

5. The AI ​​automatic monitoring and early warning system for a petroleum refinery and chemical plant according to claim 1 is characterized by: The real-time warning unit is used to immediately send warning information to the central control room and automatically initiate response measures when the AI ​​analysis core identifies abnormal behavior or potential danger. The warning level classification unit can classify the warning information into multiple levels according to the identified risk level.

6. The AI ​​automatic monitoring and early warning system for a petroleum refinery and chemical plant area according to claim 1 is characterized by: The feedback unit establishes a feedback mechanism so that the processing results after each warning can be fed back to the system, including corrections of false alarms and missed alarms, and records of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results and self-optimize the algorithm model through machine learning and deep learning technologies.

7. A method for using an AI automatic monitoring and early warning system for a petroleum refinery according to any one of claims 1 to 6, characterized in that: The steps for using it are as follows A. The image acquisition module deploys high-resolution cameras in key operating areas, storage areas, pipeline systems, and locations around important equipment in the refinery and chemical plant area for 24-hour monitoring, covering key and sensitive areas, and collects sensor data of sound, temperature, and pressure through acoustic sensors, pressure sensors, and temperature sensors to enhance the accuracy and dimension of AI analysis; B. The video formatting and compression unit is used to format and compress the collected original video stream while ensuring that key details are not lost. The image enhancement processing unit enhances the video image quality by adjusting brightness, contrast, and sharpening to ensure the accuracy of subsequent AI analysis; C. The behavior recognition unit uses a combination of convolutional neural networks and recurrent neural networks to identify the behaviors of people and equipment in the video, such as whether they are wearing safety equipment and whether there are any illegal operations. The status monitoring unit uses deep learning algorithms to analyze the operating status of equipment and uses specific image recognition and pattern analysis technologies to quickly and accurately issue warnings for dangerous situations such as leakage, fire, and smoke. D. The real-time warning unit is used to send warning information to the central control room immediately when the AI ​​analysis core identifies abnormal behavior or potential danger, and automatically initiate response measures, such as emergency shutdown and activation of the fire protection system. The warning level classification unit can classify the warning information into multiple levels according to the identified risk level, so that managers can take different emergency measures according to the warning level; E. The feedback unit establishes a feedback mechanism so that the processing results after each warning can be fed back to the system, including the correction of false alarms and missed alarms, and the record of actual operations. The continuous learning and optimization unit can continuously learn historical data and feedback results through machine learning and deep learning technologies, self-optimize algorithm models, improve recognition accuracy and reduce false alarm rates. This AI automatic monitoring and early warning system is integrated with the existing safety systems, monitoring systems, equipment management systems, etc. in the refinery and chemical plant area to form a unified data platform and management center. An intuitive user interface is designed, including real-time video display, early warning information prompts, system status monitoring, historical data analysis and other functions to facilitate management personnel to effectively monitor and operate.

Images