Petrochemical enterprise emergency command cooperation system based on artificial intelligence

Abstract

The invention discloses a petrochemical industry enterprise emergency command cooperative system based on artificial intelligence, which comprises a data integration module, a cooperative command module, an accident training module and an accident analysis module, and is characterized in that the data integration module comprises a data acquisition module, a data processing module, an anomaly detection module and a data adjustment module; and the data integration module is used for collecting, analyzing and processing data. According to the invention, through cooperation of the data integration module, the cooperative command module, the accident training module and the accident analysis module, the effects of real-time monitoring and multi-stage emergency command can be achieved, field data can be collected and analyzed through the data integration module, and multi-stage emergency command can be performed through the cooperative command module. A complete accident database can be established through the accident training module and the accident analysis module, corresponding treatment measures can be taken in time when an accident occurs, and on-site safe operation is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency command and coordination systems for petrochemical enterprises, and in particular to an emergency command and coordination system for petrochemical enterprises based on artificial intelligence. Background Art

[0002] The petrochemical industry faces extremely high safety risks due to its own characteristics. Frequent accidents not only cause huge economic losses, but also seriously threaten human life and environmental quality. In this context, enterprises and society have an increasingly urgent need to improve emergency management, prevent and control risks, which has promoted the development of intelligent emergency command technology.

[0003] Therefore, in recent years, breakthroughs in artificial intelligence technology, especially deep learning, natural language processing, computer vision and other fields, have provided a strong technical foundation for complex data analysis, decision support, and intelligent interaction. At the same time, the mature application of technologies such as the Internet of Things (IoT), big data, and cloud computing has made it possible to collect, process, and analyze massive amounts of data in real time, laying a technical foundation for building an efficient emergency command system.

[0004] However, the emergency command system relies on high-quality data input, including sensor data, historical accident records, environmental monitoring data, etc. Therefore, data acquisition may be limited by the accuracy, coverage and maintenance of the sensors, and the integrity and consistency of the data are often challenged. In addition, the interoperability issues between technical systems of different manufacturers and different standards may lead to data silos and affect the overall effectiveness of the system. In some areas, there is still a lack of clear regulatory guidance and industry standards for the application of artificial intelligence in emergency management systems, resulting in a lack of regulations in the implementation process of enterprises. Summary of the invention

[0005] The purpose of the present invention is to provide an artificial intelligence-based emergency command and coordination system for petrochemical enterprises, which has the advantages of being able to operate collaboratively, realize real-time monitoring of on-site conditions and multi-level emergency command, and solve the problem that the existing emergency command and coordination system for petrochemical enterprises is affected by front-end data, is prone to inaccurate data, and affects its use.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an artificial intelligence-based emergency command and coordination system for petrochemical enterprises, comprising a data integration module, a coordinated command module, an accident training module and an accident analysis module, wherein the data integration module comprises a data acquisition module, a data processing module, an anomaly detection module and a data adjustment module, the data integration module is used to collect and analyze data, the coordinated command module comprises a visualization interface, a coordination module and an emergency APP, the coordinated command module is used to dynamically adjust personnel task allocation according to on-site conditions, the accident training module comprises an accident simulation module and an intelligent evaluation module, the accident training module is used for emergency drills and personnel training, the accident analysis module comprises a database module and a learning feedback module, and the accident analysis module is used to optimize early warning models, decision algorithms and coordination mechanisms.

[0007] Preferably, the data acquisition module includes equipment operation data, environmental monitoring data and video monitoring data, the equipment operation data is used to predict equipment failure, the environmental monitoring data is used to identify potential dangerous situations, and the video monitoring data is used to provide on-site images.

[0008] Preferably, the data collected by the data acquisition module is analyzed and processed by a data processing module, the data processing module adopts an intelligent gateway and is deployed on an edge node, the anomaly detection module adopts a deep learning algorithm to establish an anomaly detection model and analyze the real-time data stream, and the deep learning algorithm adopts LSTM.

[0009] Preferably, the LSTM calculation method is:

[0010] The forget gate formula is: t =σ(W f [x t ,h t-1 ]+b f );

[0011] The input gate formula is: t =σ(W i [x t， h t-1 ]+b i );

[0012] The output gate formula is: t =σ(W o [x t ,h t-1 ]+b o );

[0013] where f t is the output of the forget gate, σ is the sigmoid function, W f and b fare the weight and bias of the forget gate, respectively, t is the input at the current moment, h t-1 is the hidden state of the previous moment, i t is the output of the input gate, o t is the output of the output gate.

[0014] Preferably, the LSTM calculation method further includes a unit state update announcement and a hidden state update announcement, wherein the unit state is a core component of the LSTM and is used to store long-term information, and the update announcement is:

[0015] c t =f t ⊙c t-1 +i t ⊙tanh(c t )

[0016] where c t is the cell state at the current moment, ⊙ represents element-wise multiplication, tanh is the hyperbolic tangent function, c t The candidate unit state is calculated as:

[0017] c t =tanh(W c [x t ,h t-1 ]+b c ).

[0018] Preferably, the hidden state update is the output of LSTM, which is passed to subsequent layers, and its update formula is:

[0019] h t =o t ⊙tanh(c t )

[0020] where h t is the hidden state at the current moment.

[0021] Preferably, the visualization interface integrates map services, real-time video streaming and sensor data display, the coordination module integrates message push, task assignment and video conferencing, and the emergency APP is used for location tracking, emergency calls and information reporting.

[0022] Preferably, the accident simulation module constructs an accident simulation environment through VR / AR technology to reproduce historical accidents or simulate possible emergencies in the future. The intelligent evaluation module evaluates the emergency response, decision-making efficiency and teamwork ability of trainees in simulation training through AI analysis technology.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention achieves the effect of real-time monitoring of on-site conditions and multi-level emergency command through the cooperation of a data integration module, a collaborative command module, an accident training module and an accident analysis module. On-site data can be collected and analyzed through the data integration module, multi-level emergency command can be carried out through the collaborative command module, a complete accident database can be established through the accident training module and the accident analysis module, and corresponding processing measures can be taken in time when an accident occurs, effectively improving the safe operation of the site. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the system framework of the present invention;

[0026] Figure 2 This is a schematic diagram of the data integration module framework of the present invention;

[0027] Figure 3 Schematic diagram of the workflow of the data integration module of the present invention. DETAILED DESCRIPTION

[0028] 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.

[0029] Example 1

[0030] like Figure 1-3 As shown, it is the first embodiment of the present invention, which provides an artificial intelligence-based petrochemical enterprise emergency command and coordination system, including a data integration module, a collaborative command module, an accident training module and an accident analysis module. The data integration module includes a data acquisition module, a data processing module, an anomaly detection module and a data adjustment module. The data integration module is used to collect and analyze data. The collaborative command module includes a visualization interface, a coordination module and an emergency APP. The collaborative command module is used to dynamically adjust the personnel task allocation according to the on-site situation. The accident training module includes an accident simulation module and an intelligent evaluation module. The accident training module is used for emergency drills and personnel training. The accident analysis module includes a database module and a learning feedback module. The accident analysis module is used to optimize the early warning model, decision algorithm and coordination mechanism.

[0031] like Figure 1-3As shown in the figure, with the enterprise comprehensive database as the data basis and relying on the GIS platform, the information of emergency teams, emergency resources, fire fighting facilities and materials, major hazardous sources, etc. is integrated and displayed to provide auxiliary decision-making for emergency command and daily monitoring and early warning. The data acquisition module is used to monitor and analyze the on-site environment and equipment operation status in real time. The data processing module is used to clean and preprocess the data. The abnormal detection model is established through the abnormal module to analyze the real-time data stream, so that the abnormal situation on the site can be discovered in time. The warning threshold is dynamically adjusted through historical data and learning model data, so as to improve the accuracy and sensitivity of the warning. The coordination and command module integrates map services, real-time video streams, sensor data display and other functions to realize the centralized display of multi-source information. The coordination module has video and voice communication functions. The danger alarm allows managers to get risk information in the first time, understand it through video calls or voice calls, and make reasonable arrangements according to the plan list. The emergency APP is developed for mobile devices to ensure that on-site personnel can also access the emergency command system to realize location tracking, emergency calls, information reporting and other functions. After the danger occurs, managers can view the most basic information of the hazard source in the first time.

[0032] Example 2

[0033] Reference Figure 1-3 , which is the second embodiment of the present invention, and this embodiment is based on the previous embodiment.

[0034] In this embodiment, the data acquisition module includes equipment operation data, environmental monitoring data and video monitoring data. The equipment operation data is used to predict equipment failures, the environmental monitoring data is used to identify potential dangerous situations, and the video monitoring data is used to provide on-site images.

[0035] The data collected by the data acquisition module is analyzed and processed by the data processing module. The data processing module uses an intelligent gateway and is deployed on the edge node. The anomaly detection module uses a deep learning algorithm to establish an anomaly detection model and analyze the real-time data stream. The deep learning algorithm uses LSTM.

[0036] The LSTM calculation method is:

[0037] The forget gate formula is: t =σ(W f [x t ,h t-1 ]+b f );

[0038] The input gate formula is: t =σ(W i [x t ,h t-1 ]+b i );

[0039] The output gate formula is: t =σ(W o [x t ,h t-1 ]+b o );

[0040] where f t is the output of the forget gate, σ is the sigmoid function, W f and b f are the weight and bias of the forget gate, respectively, t is the input at the current moment, h t-1 is the hidden state of the previous moment, i t is the output of the input gate, o t is the output of the output gate.

[0041] The LSTM calculation method also includes the unit state update announcement and the hidden state update announcement. The unit state is the core component of LSTM and is used to store long-term information. The update announcement is:

[0042] c t =f t ⊙c t-1 +i t ⊙tanh(c t )

[0043] where c t is the cell state at the current moment, ⊙ represents element-wise multiplication, tanh is the hyperbolic tangent function, c t The candidate unit state is calculated as:

[0044] c t =tanh(W c [x t ,h t-1 ]+b c ).

[0045] The hidden state update is the output of LSTM and is passed to subsequent layers. The update formula is:

[0046] h t =o t ⊙tanh(c t )

[0047] where h t is the hidden state at the current moment.

[0048] The visualization interface integrates map services, real-time video streaming and sensor data display. The coordination module integrates message push, task assignment and video conferencing. The emergency APP is used for location tracking, emergency calls and information reporting.

[0049] The accident simulation module uses VR / AR technology to construct an accident simulation environment to reproduce historical accidents or simulate possible emergencies in the future. The intelligent assessment module uses AI analysis technology to evaluate the emergency response, decision-making efficiency and teamwork ability of trainees in simulation training.

[0050] like Figure 1-3 As shown in the figure, the anomaly detection module uses the LSTM learning algorithm to establish an anomaly detection model. The anomaly detection model can analyze and process the data collected on site, so as to timely identify potential abnormal behaviors or dangerous trends. By deploying intelligent gateways at the edge nodes, preliminary data cleaning and preprocessing are carried out to reduce data transmission delays and improve response speed. Temperature, pressure, gas concentration and vibration sensors are used to collect equipment operation data, and monitoring equipment is used for video monitoring data. A complete accident database is established through the database module to collect data on each emergency response, including response time, resource usage, and successful and failed disposal measures. The learning feedback module feeds the accident data back to the model training process, continuously optimizes the early warning model, decision-making algorithm and coordination mechanism, and realizes the self-evolution of the emergency command system. Through simulation and deduction, based on typical accident emergency plans, accident scenarios are constructed for different stage scenarios in combination with plans, cases, knowledge, processes, etc., so as to truly simulate the entire process of accident rescue and disposal.

[0051] When in use, the enterprise comprehensive database is used as the data basis, relying on the GIS platform, the emergency team, emergency resources, fire-fighting facilities and materials, major hazardous sources and other information are integrated and displayed, providing auxiliary decision-making for emergency command and daily monitoring and early warning. The data acquisition module is used to monitor and analyze the on-site environment and equipment operation status in real time, and the data processing module is used to clean and pre-process the data. The abnormal detection model is established through the abnormal module to analyze the real-time data stream, so that abnormal situations on the site can be discovered in time. The warning threshold is dynamically adjusted through historical data and learning model data, so as to improve the accuracy and sensitivity of the warning. The coordination and command module integrates map services, real-time video streams, sensor data display and other functions to realize the centralized display of multi-source information. The coordination module has video and voice communication functions. The danger alarm allows managers to get risk information in the first time, understand it through video calls or voice calls, and make reasonable arrangements according to the plan list. The emergency APP has a reporting function. After a danger occurs, managers can view the most basic information of the hazard source in the first time.

[0052] 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.

[0053] 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 artificial intelligence-based emergency command and coordination system for petrochemical enterprises, comprising a data integration module, a coordination command module, an accident training module and an accident analysis module, characterized in that: The data integration module includes a data acquisition module, a data processing module, an anomaly detection module and a data adjustment module. The data integration module is used to collect and analyze data. The collaborative command module includes a visualization interface, a coordination module and an emergency APP. The collaborative command module is used to dynamically adjust the personnel task allocation according to the on-site situation. The accident training module includes an accident simulation module and an intelligent evaluation module. The accident training module is used for emergency drills and personnel training. The accident analysis module includes a database module and a learning feedback module. The accident analysis module is used to optimize the early warning model, decision algorithm and collaborative mechanism.

2. According to the artificial intelligence-based petrochemical enterprise emergency command and coordination system of claim 1, it is characterized by: The data acquisition module includes equipment operation data, environmental monitoring data and video monitoring data. The equipment operation data is used to predict equipment failures, the environmental monitoring data is used to identify potential dangerous situations, and the video monitoring data is used to provide on-site images.

3. According to the artificial intelligence-based petrochemical enterprise emergency command and coordination system of claim 1, it is characterized by: The data collected by the data acquisition module is analyzed and processed by the data processing module. The data processing module adopts an intelligent gateway and is deployed on the edge node. The anomaly detection module adopts a deep learning algorithm to establish an anomaly detection model and analyze the real-time data stream. The deep learning algorithm adopts LSTM.

4. According to claim 3, an artificial intelligence-based petrochemical enterprise emergency command and coordination system is characterized by: The LSTM calculation method is: The forget gate formula is: t =σ(W f [x t ,h t-1 ]+b f ); The input gate formula is: t =σ(W i [x t ,h t-1 ]+b i ); The output gate formula is: t =σ(W o [x t ,h t-1 ]+b o ); where f t is the output of the forget gate, σ is the sigmoid function, W f and b f are the weight and bias of the forget gate, respectively, t is the input at the current moment, h t-1 is the hidden state of the previous moment, i t is the output of the input gate, o t is the output of the output gate.

5. The artificial intelligence-based petrochemical enterprise emergency command and coordination system according to claim 3 is characterized by: The LSTM calculation method also includes a unit state update announcement and a hidden state update announcement. The unit state is the core component of the LSTM and is used to store long-term information. The update announcement is: c t =f t ⊙c t-1 +i t ⊙tanh(c t ) where c t is the cell state at the current moment, ⊙ represents element-wise multiplication, tanh is the hyperbolic tangent function, c t The candidate unit state is calculated as: c t =tanh(W c [x, t h t-1 ]+b c )。 6. The artificial intelligence-based petrochemical enterprise emergency command and coordination system according to claim 3 is characterized by: The hidden state update is the output of LSTM, which is passed to subsequent layers. The update formula is: h t =o t ⊙tanh(c t ) where h t is the hidden state at the current moment.

7. The artificial intelligence-based petrochemical enterprise emergency command and coordination system according to claim 1 is characterized by: The visualization interface integrates map services, real-time video streaming and sensor data display, the coordination module integrates message push, task assignment and video conferencing, and the emergency APP is used for location tracking, emergency calls and information reporting.

8. The artificial intelligence-based petrochemical enterprise emergency command and coordination system according to claim 1 is characterized by: The accident simulation module constructs an accident simulation environment through VR / AR technology to reproduce historical accidents or simulate possible emergencies in the future. The intelligent evaluation module uses AI analysis technology to evaluate the emergency response, decision-making efficiency and teamwork ability of trainees in simulation training.

Images