Automation in Petro-Chemical Industry

Research on electrical cabinet identification method for intelligent inspection based on OCR

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Label detection of electrical cabinets in the distribution room of chemical plants is an important part of intelligent and autonomous inspection. The traditional manual detection method has the disadvantages of low detection efficiency, high rate of missed detection and high false detection rate, and risky for workers. The OCR-based deep learning model can efficiently detect the position of the text box of the label on the electrical cabinet and accurately identify the text content. On the self-built electrical cabinet dataset, a text detection model DBNet and a text recognition model ABINet are trained to detect the position of the electrical cabinet text box. The accuracy of the DBNet model in detecting text boxes on the test set can reach 96%, the accuracy of ABINet in recognizing text can reach 89%, and the average detection speed of the model can reach 1 image per second under the RTX2080Ti graphics card. The methods applied in the experiment meet the requirements of industrial detection accuracy and detection speed.

Research on Edge Server Deployment and Scheduling Strategy in Oil and Gas Field Station and Storage Scenarios

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

In the construction scenarios of oil and gas field station and storage, in response to issues such as the dispersed distribution of construction elements that are not easy to flexibly allocate, the inability of traditional deployment methods of edge servers to meet the on-site data processing requirements, and the presence of obstacles in data signal transmission, a deployment and scheduling strategy for edge servers in the scenarios of oil and gas field station and storage is proposed. This strategy aims to, when the number and resources of edge servers are limited, minimize the deployment cost and latency of edge servers while maximizing the network throughput, and achieve the reasonable deployment of the two-dimensional coordinates of edge servers and the scientific scheduling of construction element tasks. The experimental results show that the strategy proposed in this paper has superior performance in terms of balancing cost, latency, and network throughput, convergence, and scalability.

Exploration and practice of application of digital delivery technology in oil depot

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Taking a newly built oil depot as an example, this paper introduces the application of digital delivery technology in oil depots. Firstly, it involves establishing a 3D model of the oil depot, intelligent P&ID, structured data, and unstructured document materials, which are interconnected with the digital delivery platform to achieve system management, information collection, information processing, and information handover, providing a data foundation for the construction of a smart oil depot. Secondly, it involves building a full-lifecycle management system for oil depot equipment and an intelligent inspection management system to enhance safety management and equipment management, thereby helping to reduce staff and increase efficiency in the depot area. Thirdly, it involves the organic integration of the digital delivery platform and the oil depot MES management system to achieve nine applications, including equipment information viewing and 3D visualization of the oil depot. Digital delivery will comprehensively improve the management level of enterprises, playing a crucial role in reducing staff and increasing efficiency, while maximizing the efficient and scientific management of assets, equipment, archives, emergency response, and operations and maintenance.

Technology Innovation and Practice Oriented to the Fully Autonomous Operation System of Chemical Plants

Wen-Guang YU

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

The operation process of chemical production equipment usually includes five processes: start-up, load up, normal operation, load reduction, and shutdown. By analyzing the control needs and difficulties under all working conditions, the research group integrates and integrates independent operation technologies such as process mechanism, expert experience and intelligent algorithm based on the intelligent operation management and control system platform, and innovatively develops technical solutions such as operation navigation, loop tuning, and alarm governance for various problems arising during the operation of the device. Through the deployment and implementation of the two companies, the whole process of the plant has been highly autonomous.

A Multi-Agent Collaborative Control Method for Pump Injection Based on Generative Adversarial Imitation Learning

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

With the expansion of unconventional oil and gas development, the demand for advanced fracturing technologies has significantly increased. To address challenges such as the high cost, energy consumption, and limited decision-making capabilities of fracturing equipment, this study proposes a multi-agent pump injection coordination control method based on an improved Generative Adversarial Imitation Learning (GAIL) algorithm. By leveraging field operation data, the method computes optimal pump speeds for multiple fracturing trucks in real time, enhancing the stability of collaborative control. The study constructs a standardized control sequence dataset from historical data and validates the proposed approach in a simulated environment. Experimental results demonstrate that the method achieves near-optimal rewards, with discriminator loss deviations of 7%–10% from the ideal, reflecting expert-level control capabilities. Furthermore, the method supports multi-vehicle scalability, effectively addressing dynamic target tasks, significantly improving operational efficiency, and reducing energy consumption.Despite potential limitations in highly dynamic environments or with insufficient training data, this method represents a significant advancement in autonomous fracturing control and offers broad application potential.

Research on Compressor Condition Monitoring Method Based on LSTM and Knowledge Graph

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

The operational status monitoring of natural gas centrifugal compressor units is critical for ensuring equipment safety and efficient operation. Traditional monitoring methods are prone to false alarms and missed detections under complex conditions, making it difficult to meet real-time and accuracy requirements. This paper proposes a monitoring algorithm for compressor status based on Long Short-Term Memory (LSTM) networks and knowledge graphs. The algorithm utilizes the LSTM model to extract deep features from multi-dimensional time-series data (e.g., pressure, temperature, vibration) of compressors and combines knowledge graphs to analyze abnormal states through reasoning, achieving precise monitoring of compressor operating conditions and intelligent diagnosis of fault types. Experimental results demonstrate that the proposed algorithm outperforms traditional methods in fault identification accuracy, robustness, and real-time performance, especially in recognizing complex fault patterns, providing strong support for the intelligent operation and maintenance of natural gas compressor units.

Research on a Mixing Control Scheme for Low Concentration Gas

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

The fuel gas used for heating furnace combustion in a certain petrochemical enterprise is a mixture of low concentration gas and air. The low concentration gas mixing (blending) technology developed by Sinopec (Dalian) Petrochemical Research Institute Co., Ltd. is used to mix gas and air. This article introduces a low concentration gas mixing (blending) control technology scheme, and fits the amount of fuel used for heating furnace combustion with the valve opening of the corresponding regulating valve for gas intake. It focuses on the low concentration gas mixing (blending) control algorithm and designs the device from a safety perspective to ensure the safety of low concentration gas mixing (blending).

Steam and circulating water temperature control strategy for reactor

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Introduces several design schemes and operating methods for temperature control using steam and circulating water in reactor; Select one of the design schemes and introduce the implementation steps of four sequential control strategies, including steam heating, multi-stage steam heating, circulating water cooling, and circulating water pressure return, as well as the key temperature control parameter selection. Demonstrate the temperature control effect based on the actual situation.

People-oriented Crossed Operating Platform Smart Seating System

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

This paper focuses on the design problems of the central control room console in the petrochemical industry and innovatively proposes a humanized cross - platform intelligent seat system. Through the intelligent sliding screen device, this system enables one set of mouse and keyboard to operate multiple hosts, effectively reducing the number of consoles and the occupied area, and significantly improving operation efficiency and safety. Combining with specific project examples, the paper deeply analyzes the system design ideas, technology roadmap, application test, using effect, and explores its promotion prospects.

Application and Exploration of Domestic On-site Intelligent I/O Technology in New Material Production

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

The control system plays the role of the brain and central nervous system in chemical projects, so its importance is self-evident. With the development and progress of technology, more and more chemical projects are choosing to use on-site intelligent I/O technology in their control system. Compared to traditional control system, on-site intelligent I/O technology has inherent advantages. In this artical, the application of on-site intelligent I/O technology in a new material project was introduced and explored. Firstly, the definition of on-site intelligent I/O technology was introduced and its characteristics were elaborated from both hardware and software perspectives. Then, the combination of on-site intelligent I/O technology and practical projects was introduced. Finally, the advantages and economic benefits of on-site intelligent I/O technology were explored.

Discussion on Process Design of High-flow Real Flow Calibration of Natural Gas

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Natural gas high-flow real flow calibration process is mainly divided into bypass process and closed loop process. By comparing the differences between bypass process and closed loop process in calibration process and certification process, this paper analyzes the advantages and disadvantages of the two calibration processes, finds out their respective applicable scope, and provides measures to improve calibration accuracy, providing technical support for the construction of calibration station.

Research on Autonomous Upgrade and Transformation of GEA Lubricating Oil Purifier Control System in Offshore Oil Production Facilities

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

This study aims to address the frequent failures of lubricating oil purifiers (V603 and X-7031) on an offshore oil platform caused by controller aging and program mismatch, with the goal of breaking through foreign technical barriers and achieving autonomous upgrade of the control system to ensure host equipment safety. By reverse-engineering encrypted program logic and reconstructing control functions with domestic PLC hardware, adaptive slag discharge, fault diagnosis, and remote monitoring modules were added. Key technical challenges, including program encryption, hardware compatibility, and data gaps, were overcome through logic signal capture, machine learning-based parameter prediction, and hardware-in-the-loop simulation testing. The upgraded system demonstrated stable operation, with mean time between failures (MTBF) increasing from 420 to 1500 hours. Direct cost savings reached 260,000 RMB, while potential losses exceeding 6 million RMB were avoided. The research proves that autonomous transformation can effectively bypass foreign technical restrictions, providing a replicable solution for domestic substitution of similar equipment and advancing lean management of offshore platform systems.

Quantitative risk calculation and analysis of consequence of storage tank accident

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Abstract: In order to quantitatively evaluate the severity of accident consequences caused by LNG tank leakage failure, based on the analysis of the evolution process of LNG tank leakage, the model diagram of the consequences of the storage tank accident is obtained. According to the consequence evaluation criteria, the simulation results are combined with the field investigation data, and the severity of consequences is comprehensively considered from the aspects of the damage area, the number of casualties, economic losses and environmental pollution. And put forward the corresponding control measures of different regions. The results show that when steam cloud explosion occurs, the most dangerous range is 10m in summer, 9m in spring and autumn, 6m in winter. When jet fire occurs, it is most dangerous in the range of 17.3m downwind in winter. The risk of steam cloud explosion and jet fire after leakage is greater in summer than in spring, autumn and winter. The accident occurred in summer, which may cause 7 casualties, economic losses of 2.5624 million yuan, vegetation destruction of 30.8m2, carbon emissions of 0.11tCO2. The research results solve the problem that the accident consequences cannot be quantified in the previous studies, and form a multi-dimensional evaluation system.

AXENS Process Diesel Hydrocracking Thermal High Pressure Cut-off Ball Valve Technology Optimization

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Abstract: The emergency shut-off ball valves located in the hot high-pressure separator of the AXENS process diesel hydrocracking unit, as core components of the system, play a crucial role. Their primary function is to effectively prevent high-pressure media from entering low-pressure areas, thereby ensuring the safety and stable operation of the entire high-pressure reaction system. Therefore, these ball valves can be considered as key nodes in the safety defense line of the entire system. After nearly a year of operation, two emergency shut-off ball valves within the hot high-pressure system of the AXENS process diesel hydrocracking unit experienced abnormal leakage at the valve body seams. As the operating time of the unit continued to extend and the production load was dynamically adjusted, the leakage at the seams of these valves gradually worsened, posing a significant threat to the stable operation of the unit. It could potentially force the unit to shut down at any time, thereby adversely affecting the unit's continuous production capacity. Faced with this urgent technical challenge, the instrumentation department and workshop responded swiftly and established a special technical task force. The team thoroughly analyzed the specific structural form of the diesel hydrocracking hot high-pressure separator emergency shut-off ball valves and conducted a comprehensive and detailed assessment and study of the sealing component materials. Ultimately, they formulated a practical and feasible solution. The implementation of this solution provided solid technical support and a strong guarantee for the diesel hydrocracking unit to achieve stable operation with "longevity, full capacity, and excellence."

Calculation of flow coefficient and caliber verification of control valves based on different fluids

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Selecting highly adaptable control valves for different working conditions plays a crucial role in ensuring the quality and safety of industrial production. This article analyzes and summarizes the calculation schemes for the flow coefficient of control valves for different phase media such as liquid, gas, steam, and gas-liquid under standard conditions and actual installation conditions, and forms a calculation book. Based on this scheme, a study on the verification of control valve diameter was carried out, and a set of verification processes was compiled. Combined with practical cases in the process design and development, a detailed discussion was made on how to select the flow coefficient and control valve diameter that meet the design conditions. The work in this article has practical operational significance for the calculation and selection of control valves in the process industry.

Application of linear function in radioisotope level meter operation

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

After the replacement of the radioactive source and detector of the radioisotope level meter, Continuous production units often fail to empty the container of the level to be measured in time， to carry out the real empty tank zero calibration for level measurement, or fill the material to carry out the full tank full calibration, and can only do the real empty calibration or full calibration at the appropriate opportunity (such as device maintenance). n this case, the mathematical operation linearization model of the isotope level meter can be used [1] ，to calculate the zero and full point count rate of the isotope level meter, so that the meter can be put into operation in time, and the off-line time and radiation occupational hazards of the meter can be reduced.

The invention relates to an improved design method of regulating valve inner parts to overcome high pressure gas vortex

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

This paper describes an improved design method of regulating valve inner parts to overcome the high pressure gas vortex, which is mainly used in the single component pure high pressure gas medium, the valve front straight section can not meet the requirements, or there is no straight section of the fluid control. Due to the lack of straight pipe section of the valve, the high-pressure gas in front of the valve generates eddy currents, which produces unbalanced force on the valve core, resulting in high-frequency vibration of the valve interior when the valve is greater than a certain opening degree. The unsafe factors caused by high frequency vibration lead to the valve can only operate at a small opening and cannot operate normally. Through the renovation of the valve internals, the valve seat increases the noise reduction function, and the valve sleeve structure is redesigned, effectively overcoming the problem that the valve is not enough because of the straight pipe section in front of the valve, the high pressure gas in front of the valve produces eddy current, and the valve internals are unreasonable in design structure, and the high-frequency vibration is generated when the valve is greater than a certain opening degree.

Research on Intelligent Inspection Robot Technology and Application of Oilfield Enterprises

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

At present, China is vigorously promoting intelligent construction, gradually realizing production automation, unmanned factory, flattening and centralizing production management, and realizing the production of machines instead of manual operations. With the development of intelligent technologies such as machine vision, sensing technology, modern Internet and big data, under the new situation of vigorous development of the new round of scientific and technological revolution and increasing resource and environmental constraints, enterprise managers generally recognize the importance of intelligence and began to use artificial intelligence, intelligent robots and other technologies to obtain the effect of improving production efficiency and production safety, streamlining staff, reducing production costs, and improving the economic benefits of enterprises. This paper analyzes the current situation of intelligent inspection robots in the current petroleum and petrochemical industry, systematically analyzes the current situation and future development of intelligent inspection robots, and then puts forward reference opinions on the innovative development and optimization design of intelligent inspection robots in oilfield enterprises.

Research on remote operation and maintenance technology of electronic control system of drilling rig based on information platform

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Through IoT technology and cloud edge collaboration, the operation data of the drilling rig's electrical control system is remotely collected. Big data analysis is used to make online judgments and management of equipment abnormalities, achieving real-time monitoring of equipment overview, active detection of abnormal faults, rapid feedback and other integrated control capabilities. Research on efficient remote equipment operation and maintenance management processes and a series of business functions based on new generation information technology, to improve the information, precision, and intelligent management capabilities of equipment.

Engineering design of on-line monitoring system for water pollution sources

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

Abstract: With the continuous improvement and strengthening of local governments' monitoring and supervision requirements on the water quality of production wastewater and rainwater discharged by chemical enterprises, the traditional method of manual sampling for water quality monitoring has been unable to meet the requirements, and the setting of online monitoring system for water pollution sources has become inevitable. This paper introduces the practical application of online monitoring system for water pollution sources in engineering projects. The system design, instrument selection, sampling method and data reporting, video monitoring upload, environmental acceptance and so on are described in detail .

Research on Petrochemical Smart Factory in the Context of “Artificial Intelligence +”

2025, 61(4): 0-0.

Asbtract ( )

Related Articles | Metrics

In the context of “Artificial Intelligence +”, the construction of smart factories has become an inevitable choice for the petrochemical industry, and has passed through three stages: data integration, comprehensive integration, and initial intelligence empowerment. Moving forward, the petrochemical industry will accelerate the pace of architecture design, upgrading the existing systems and constructing the new ones, as well as application operation of the third-generation smart factory based on its existing informatization foundation and the advanced experiences of relevant industries. Aiming at the business scenarios of smart factory, this paper proposes the application architecture and technical framework, and sorts out the overall architecture of AI. Finally, the challenges and trends of petrochemical smart factories are summarized.

Table of Content