Current Status and Development Trends of Optimization Technologies in the Global Refining and Petrochemical Sector

Thread Content

The current status and development trends of optimization technologies in the global refining and petrochemical industry: As countries around the world pay increasing attention to optimization efforts in this sector and conduct more in-depth research on optimization technologies, domestic and international research institutions as well as technology companies have made significant progress in the field of refining and petrochemical optimization, particularly in areas such as simulation-based optimization techniques, energy optimization research, and software development. Meanwhile, the application of optimization technologies has also generated considerable economic benefits for refining and petrochemical enterprises. The following sections introduce the current research status of simulation optimization and energy system optimization theories, the current research and application status among domestic and international technology companies, as well as the development trends. (1) Current research status of energy system optimization theory. Research on process energy system theory began in the 1970s. As research progressed, scholars at home and abroad proposed models such as the three-subsystem interaction model, the “onion” model, and the three-stage model to describe the structure, role, and changes of energy within process systems. In 1983, while studying the structured methods involved in process optimization, Townsend and Linnhoff considered a process system to be composed of interacting subsystems in order to simplify the design optimization process. They proposed a three-subsystem interaction model, dividing the process system into three subsystems: processing (reaction, separation), heat exchange networks, and power, in order to qualitatively reflect the inherent relationships and constraints regarding energy conversion and utilization within the process system; however, this model could not accurately represent the essence of the energy structure. In 1988, Linnhoff and others proposed the “onion model” to emphasize the hierarchical nature of process design. Its core is the reaction subsystem, with the separation subsystem, heat exchange network, and utility subsystems following in sequence from the inside out. However, this model only provides a rough description of the energy structure of the process system, without offering a strict, quantitative mathematical model. Since 1982, Hua Ben and others, based on the analysis of the second law of thermodynamics and starting from the laws governing energy changes in process systems, proposed and gradually refined a three-component energy structure model that includes energy conversion, energy utilization, and energy recovery stages. This model provides a rigorous and quantitative topological relationship for the energy flow structure of process systems. The key aspect is to divide the system into functional zones and molecular systems, and to establish exergy-economic models, corresponding objective functions, and boundary conditions for each of them; subsequently, an exergy-economic evaluation and optimization of global improvement solutions are carried out on the basis of coordinated optimization of the subsystems. However, due to the strong theoretical nature of the exergy economic analysis method and its relatively complex implementation process, its application in actual enterprise production is subject to certain limitations. (2) Current status of research and applications both at home and abroad
① International status
In recent years, internationally renowned technology and software companies such as KBC, ASPEN, Shell Global Solutions, Invensys, Honeywell, and Process Integrated Limited (PIL) have, relying on their mature process simulation and optimization technologies, combined with advanced pinch analysis methods and leveraging extensive expert experience, gradually developed their own technical solutions for energy system optimization. These solutions have been successfully applied in several large international oil companies and petrochemical enterprises. http://mmbiz.qpic.cn/mmbiz/xjFI9op4iaCnc9dCDiaNApMFYyfeIia5R1iaoxJWTG6e8t2gT3U8JP2Ah5iaVzrqC0fZlPLD1OmEvGMX7QiadsHb4TMg/0?wx_fmt=png KBC has developed the Petro-SIM process simulation software based on HYSYS.Refinery, and possesses a set of standard reactor model kits that can be used in Petro-SIM, including models for catalytic cracking (FCC-SIM), hydrocracking (HCR-SIM), catalytic reforming (REF-SIM), naphtha hydrogenation (NHTR-SIM), diesel hydrogenation (DHTR-SIM), VGO hydrogenation (VGOHTR-SIM), residue hydrodesulfurization (RHDS-SIM), delayed coking (DC-SIM), alkylation (ALK-SIM), visbreaking (VIS-SIM), and isomerization ； Software such as SuperTarget for heat exchanger network optimization, ProSteam for steam system simulation and optimization, and best energy efficiency technologies based on the BT (Best Technology) index have been developed. The Total Site energy integration technology proposed in this research utilizes plant-wide pinch analysis and the strategic energy conservation development roadmap, RoadMaps, to analyze, optimize, and prioritize the energy consumption and supply within all process units and utility systems throughout the plant, thereby identifying the quickest and most effective improvement measures. To date, KBC has implemented the Total Site plant-wide energy optimization project in over 50 refineries and chemical plants, including those belonging to BP, BASF, ESSO, Chevron, Total, COSMO, and Nippon Energy Global. This has resulted in energy savings of 15%–30%, a reduction in hydrogen demand of 8%–20%, energy savings of 25%–35% for compressors, and a decrease in total costs of 14%–20%. For example, a refinery in Europe applied Total Site technology for pinch analysis and steam system modeling, and identified several energy-saving measures related to utility systems such as turbine upgrades, waste heat utilization, and combined heat and power generation; as a result, the refinery’s Energy Intensity Index (EII) decreased by 5 points after these measures were implemented ； After analyzing its utility systems using the Total Site technology, Sasol Fuel Company in South Africa proposed a number of typical energy-saving measures such as improving the heat exchange network and increasing combined power generation; implementing these measures can result in a reduction of utility requirements by around 30%. Aspen offers a wide range of software products for simulation and optimization. These include Aspen Plus for simulating chemical processes, Aspen RefSYS for simulating oil refining processes, HYSYS for chemical process simulation, Aspen Dynamics for dynamic simulation, HX-Net and Aspen Pinch for heat exchanger network optimization, Aspen Utilities for optimizing utility systems, Aspen PlusOptimizer for real-time optimization, the Aspen HTFS+ series for simulating heat exchange equipment, certain software tools for simulating oil refining reactors, and POLYMER for simulating polymer reactions. Among them, Aspen Plus is a steady-state process simulation tool; it is the only software product that incorporates both sequential modular and simultaneous equation algorithms within a single simulation tool ； HYSYS was originally a product of the Canadian company Hyprotech. After Aspen Corporation acquired this company in 2002, HYSYS became part of Aspen’s product line; it is the only software product that offers both steady-state and dynamic simulation capabilities ； Aspen PlusOptimizer enables users to use simultaneous equation solving algorithms to rapidly and accurately solve optimization problems involving nonlinear objective functions with multiple degrees of freedom. It provides engineers in the process industry with an effective tool for dealing with large, complex process flows, and creates the conditions necessary for implementing closed-loop, real-time optimization systems. As a nested product of Aspen Plus, it offers full support for the degrees of freedom, constraints, as well as the setting and configuration of objective functions available in Aspen Plus, and can be used for both online process optimization and offline simulation. Based on these software products, Aspen Corporation further developed comprehensive energy optimization solutions, and leveraging its expertise, carried out energy system optimization analyses in numerous petrochemical companies. For example, in 2000, Aspen carried out a utility system optimization project at BP’s Bulwer Island refinery, with an investment payback period of less than 1 year ； An energy-saving optimization project was carried out on the ethylene plant at YNCC in Yeosu, South Korea. Through adjustments to operating conditions and equipment modifications, a 10% increase in production output and a 2.2% reduction in specific energy consumption were achieved ； In 2005, an energy system optimization analysis was conducted on the design plan for CNOOC Huizhou Refinery, including analysis and optimization of the hydrogen system, the entire plant’s steam system, the plant’s fuel and power (combined heat and power) systems, as well as pinch point analysis across the whole plant. As a result of these optimizations, the EII value of the design plan dropped from 80 to 61.5, resulting in annual economic benefits of $240 million. Shell GlobalSolutions utilizes its in-house team of expert consultants as well as simulation and optimization software it has developed to carry out energy system optimization analyses. The software products used include Aspen Plus, Unisim, as well as the proprietary tools ECU, SHARC, plan24tune, and EMS. Among them, the ECU software is used for the simulation and optimization of ethylene cracking furnaces, while the SHARC software is used for the simulation and optimization of catalytic cracking units. To date, Shell GlobalSolutions has implemented energy system optimization in 29 refineries and petrochemical plants around the world, reducing energy consumption by approximately 2% to 7% in refineries, and by about 3% to 5% in petrochemical plants. Invensys offers a range of software products for process simulation (steady-state and dynamic simulation), offline and online optimization, and operator training systems, including PRO/II for simulating chemical processes, HEXTRAN for optimizing heat exchangers and networks, DYNSIM for dynamic process simulation, and ROMeo for online simulation optimization. In 2002, the company optimized the ethylene utility systems for the ethylene plant in Daesan, South Korea, and the refinery in Mizushima, Japan ； In 2005, the utility systems of the Ohita ethylene plant and those of the Aichi refinery were optimized and upgraded respectively. In 2004, Honeywell acquired HYSYS, a software product used for steady-state and dynamic simulation optimization. In 2005, based on HYSYS and UOP technologies, it launched UniSim, a new generation of process simulation software system. This system integrates design, online optimization, and operator training functions within a single product environment, while also adding and enhancing technologies such as heat exchanger network optimization and process synthesis. Among them, UniSim Desig is used for steady-state process simulation, Unisim DynamicOption is used for dynamic process simulation, UniSim SQP Optimizer Option is used for process simulation optimization, UniSim ExchangerNet is used for pinch analysis and heat exchanger network design, and Profit Optimizer is used for online dynamic optimization. PIL Company has developed technologies and software for the design and renovation of heat recovery heat exchange network systems, namely HEAT-int; technologies and software for the design and operational optimization of plant-wide steam power systems, namely SITE-int; and hydrogen pinch analysis technologies and software for the management and optimization of hydrogen resources across the plant, namely H2-int. Among them, the H2-int software product features a unique method for modeling hydrogenation units along with advanced mathematical optimization techniques. Combined with rigorous process simulation of hydrogen networks, it is capable of accurately describing various unit operations and enabling global optimization of the hydrogen system operations in refineries. To date, the company has implemented several energy system optimization projects, achieving good energy-saving results. For example, an energy system optimization analysis was conducted on 13 oil refining units at BP’s Grangemouth refinery (3 atmospheric and vacuum distillation units, hydrocrackers, catalytic crackers, reformers, alkylation units, and multiple hydrotreating units) as well as 8 petrochemical units. With only operational adjustments made, the annual energy savings amounted to $7.8 million ； Energy system optimization was carried out on 25 units at BP’s Coryton refinery, resulting in annual energy savings of $3.7 million through only operational adjustments. http://mmbiz.qpic.cn/mmbiz/xjFI9op4iaCnc9dCDiaNApMFYyfeIia5R1ia8iciaUHhs4yBzUz9POmic1fIDDFzzyibsmu32WKcoofTF6sb2wcpjAhUFQ/0?wx_fmt=png ② Domestic situation: In recent years, companies such as Youhua Process Technology Company, Dalian University of Technology, the Chinese Academy of Sciences, Tsinghua University, and South China University of Technology have, based on the adoption of some advanced international simulation and optimization software, developed their own proprietary technologies such as deep heat discharge, large-scale thermal integration, and virtual temperature methods. They have created software for optimizing heat exchangers, heat exchange networks, and steam power systems, and have carried out comprehensive energy optimization efforts for individual units, between different units, heat exchange networks, and steam power systems, achieving good economic benefits as a result. The energy system optimization technology of Youhua Process Technology Company is based on the theory and methods of comprehensive energy optimization for the “three stages” of process systems. Using this technology, more than 40 energy-saving improvement projects have been carried out for domestic refining and chemical enterprises. Among them, the appraisal result organized by the Guangdong Provincial Department of Science and Technology—“Engineering application of energy system optimization in the process industry within the petroleum and petrochemical sector”—which was approved in September 2002, has generated annual benefits of 120 million yuan for four enterprises. The company’s proprietary technologies and software products developed include: specialized heat exchanger optimization software (HEDO); energy optimization technology for hot feed in hydrogenation units; proprietary technologies for deep thermal discharge and thermal integration in large-scale systems; comprehensive utilization technology for low-temperature heat in large-scale systems; steam power system optimization software (STOpti); the CWopti process package designed specifically for optimizing the plant-wide circulating water system; a complete set of technologies for reducing power consumption in pumps and compressors; heat exchanger network optimization software (ODHEN), etc. In 2001, an energy system optimization and renovation project was carried out on the atmospheric and vacuum distillation units at Daqing Refining & Chemical Plant. As a result of this optimization, the steam consumption per unit of atmospheric distillation was reduced by 2 t/h to 3 t/h, the final temperature of heat exchange increased by over 10°C, and fuel consumption was reduced by 100 kg/h ； The second set at atmospheric pressure raises the final heat exchange temperature by 6°C to 8°C, reducing fuel consumption by 200 kg/h to 250 kg/h ； It reduces the consumption of atmospheric steam by 8 t/h to 10 t/h, increases low-temperature heat recovery by about 2500 kW, with a total investment of around 80 million yuan, resulting in annual energy savings of 32 million yuan. Energy system optimization was carried out at a petrochemical company with a crude oil processing capacity of 13 million tons per year. Projects such as heat supply and heat integration, optimization of low-temperature heat systems, and optimization of steam power systems were proposed and implemented. The total cost of these projects was 160 million yuan, resulting in a reduction in energy consumption of 14.9 kilograms of standard oil per ton; the annual economic benefit amounted to 360 million yuan. Based on independent innovation, the School of Chemical Engineering at Dalian University of Technology has developed the \"Process System Energy Integration Technology.\" It introduced the \"Principle of Energy Consistency for Process Systems\" for the first time, unifying large-scale, complex process systems as a whole in terms of energy usage; the problem of energy integration was transformed into an optimal design problem for corresponding constrained heat exchange network systems. The V1.0 version of a software system for optimizing large-scale energy networks was developed, enabling effective resolution of energy integration issues in such systems ； For the first time, the \"virtual temperature method\" (i.e., effective temperature) was proposed for conducting operational and design pinch analyses, in order to accurately describe the distribution of energy flow within the system and thus enable an accurate diagnosis of energy usage conditions ； A “strategy for energy integration in process systems” has been summarized to guide the implementation of energy diagnostics in process systems, identification and removal of bottlenecks, as well as the simultaneous optimization of energy conservation and capacity expansion in production facilities. Currently, this technology has been applied to the analysis of energy consumption and production capacity for over 20 production units at petrochemical enterprises in Dalian, Jilin, Liaoyang, Fushun, Yanshan, Qilu, and other locations. Eight energy-saving and capacity-expansion technical renovation plans proposed based on this technology have been implemented. These measures are expected to result in a 10%–20% reduction in energy consumption and a 20%–30% increase in production capacity, generating an annual economic benefit of 80 million yuan. Among these, the virtual temperature method was used to conduct an energy consumption diagnosis of the cold box system in an ethylene plant in China, in order to identify the energy consumption bottlenecks in that process system. Considering the small heat transfer temperature differences in low-temperature processes and the fact that cold utility systems make use of intermediate utilities, a comprehensive approach based on multi-stream heat exchanger networks for such low-temperature processes was proposed; as a result, the energy consumption of the cold utility systems was reduced by 44.5%. The Institute of Process Systems Engineering in the Department of Chemical Engineering at Tsinghua University has developed the highly flexible and low-sensitive heat exchanger network simulation system HEXTH, as well as the total plant energy system optimization tool TSES; these tools have been applied at companies such as Yanshan Petrochemical, Qilu Petrochemical, and Jinxi Petrochemical. In terms of the simulation and optimization of ethylene plants, the GK-V type simulation and analysis system for ethylene crackers, PYRO-SimP, and the simulation and optimization system for ethylene crackers, EPSOS, have been developed. Additionally, a cracking reaction kinetics model and a full-cycle operation optimization model were developed for the small-scale ethylene plant at Lanzhou Petrochemical. These models have effectively guided plant operations, providing technical support for increasing the yield of both ethylene and propylene as well as reducing the overall energy consumption of the plant. In addition, the HEAT-int software from PIL Company and the HEATSim software from Tsinghua University were utilized to develop a software platform for the optimization and modification of heat recovery systems in industrial processes; this platform enables automatic design and modification of heat exchange networks, as well as simulation and optimization of such networks ； Using PIL Corporation’s SITE-int software and Tsinghua University’s HEATSim software, an optimization and retrofitting software for the plant’s steam power utility systems was developed. This software enables engineering analysis of utility systems, target analysis and optimization for cogeneration, as well as simulation and optimization of utility systems. The Institute of Process Engineering, Chinese Academy of Sciences, in collaboration with Mitsubishi Corporation of Japan, developed a simulation system for the delayed coking process of heavy oil. A carbon composition model was proposed, and a model of the delayed coking reactor was created; through the Pro/II software interface, this model was successfully integrated into Pro/II software ； A dedicated software for simulating the entire process of delayed coking was developed independently, and it can operate completely separately from existing commercial process simulation software ； An exergy-based overall process energy efficiency analysis model and database system have been proposed to analyze the energy utilization efficiency and causes of energy consumption losses in raw materials, streams, units, and the entire process. The system can operate independently or be integrated with process simulation software such as Pro/II and Aspen. (3) Research and application status of key technologies ① Energy consumption evaluation methods: At present, the main indicators for evaluating oil refining energy consumption that are widely recognized by foreign refining companies and oil firms are the Energy Intensity Index (EII) proposed by Solomon Company and the Best Technology (BT) index developed by KBC Company. In addition, Shell also has its own internal energy consumption evaluation indicator, CEL (Correct Energy Loss). However, when comparing energy consumption benchmarks, it is still necessary to incorporate Solomon’s EII index. Among them, the EII indicator is determined by key factors such as the actual total energy consumption of the refinery, the utilization capacity of the units, the energy consumption metrics of those units, and the energy consumption of systems outside the facility. It reflects the impact of the refinery’s unit conditions, feed levels, operating conditions, and product quality on energy consumption; it is a comprehensive energy consumption indicator that takes into account the complexity of the refinery, and is primarily used for comparing the horizontal competitiveness among different companies. KBC’s BT index is primarily used to rank companies by assessing the gap between their current energy consumption levels and their optimal levels through vertical comparisons. This indicator can reasonably reflect the impact of raw materials, processes, and equipment on energy consumption levels. It also helps identify the weak points in terms of energy use within individual units and across the entire plant, thereby providing guidance for companies to carry out targeted optimizations of their energy systems. Due to the fact that these indicators are relatively objective and facilitate the identification of energy-saving potential as well as the guidance of energy-saving efforts, they are widely used in energy system optimization analyses by foreign refining companies. However, since the specific research contents related to energy utilization evaluation methods are one of the core technologies of the aforementioned company, they are not disclosed or transferred to external parties. Domestic oil refineries generally use the comprehensive energy consumption and single-factor energy consumption indicators for refining proposed by the U.S. company Amoco in the 1980s; meanwhile, the chemical industry has always employed the comprehensive energy consumption indicator for chemical products. In 2005, Sinopec revised the existing standards for calculating energy consumption in oil refining, primarily by adjusting the energy coefficients for various refining units based on the changes in their production processes. In 2006, Sinopec revised the calculation standards for the comprehensive energy consumption of ethylene products, adjusted the scope of calculation, and added indicators such as the energy consumption of dienes. The advantage of the aforementioned energy consumption evaluation indicators is that they simplify the concept of energy consumption, making it easy to compare energy consumption among different units as well as between different refineries. However, they can only be used for analyzing the overall energy usage levels of an entire enterprise or its various units; they cannot reflect the impact of factors such as raw materials, processes, and equipment on energy consumption within these units. In particular, it is not possible to compare key indicators related to energy use systems and equipment within a unit, nor can they be used directly to identify potential areas for energy savings in those units and systems. ②Simulation and optimization techniques and software for petrochemical process flows. The methods used for simulating petrochemical processes mainly fall into three categories: sequential module method, simultaneous equation method, and combined module method. Among them, the sequential module method is a technique for sequentially calculating and solving system models by processing each individual module one after another. Due to its intuitive and easy-to-understand nature, it is the basis upon which most of the widely used simulation software for the entire petrochemical process is developed. The optimization methods for the refining process mainly include mathematical programming, artificial intelligence methods, and pinch analysis (primarily used for heat exchanger network optimization). The simulation optimization software products reviewed below are primarily based on the aforementioned simulation and optimization methods. Petro-SIM is a full-featured graphical process simulator that combines industry-leading KBC Profimatics technology with proven strict refinery process models, and was developed based on the HYSYS.Refinery interface. Petro-SIM is capable of creating a typical “desktop refinery,” and it also provides a fairly comprehensive set of reaction models for refinery processes, including catalytic cracking units, catalytic reforming units, hydrocracking units, hydroprocessing units (covering naphtha hydrogenation, diesel hydrogenation, wax oil hydrogenation, and residue hydrogenation for desulfurization), delayed coking, aromate disproportionation, isomerization, and more. This software has been applied in over 120 refining and chemical companies around the world, achieving excellent results in production plan optimization, PIP, and energy system optimization. Aspen RefSYS is an excellent software developed by Aspen Company based on the mature general-purpose process simulation system HYSYS. It incorporates various refining reaction models from Aspen and the world-renowned oil refining technology company UOP – including catalytic cracking, catalytic reforming, hydrocracking, and hydrofining – to enable full-process simulation in refineries. This software helps refineries improve their operating methods and make more accurate predictions regarding the overall efficiency of the refinery. Currently, 23 of the top 25 oil companies in the world use this product. Aspen Plus is a chemical steady-state process simulation tool developed by Aspen Corporation; it is capable of handling the steady-state mass and energy balances in continuous processes. It features a comprehensive database, a wide range of products, and strong integration capabilities. It is the only simulation software that combines both the sequential module SM and the system of equations EO algorithms within one tool. In addition to components, properties, and equation of state data, it also includes a variety of unit operation modules, as well as powerful functions for model and process analysis. The PRO/II product from Invensys is a general-purpose steady-state process simulation software for the chemical industry. It originated from SP05, the world’s first petroleum distillation simulator developed by SimSci in 1967. In 1973, SimSci introduced a flowchart-based simulator, and in 1979 it released the Process software for process simulation on PCs (the predecessor of PRO/II). PRO/II features a comprehensive database of material properties, a powerful thermodynamic property calculation system, and over 40 unit operation modules, which can be used for steady-state simulation of processes, property calculations, equipment design, cost estimation/economic evaluation, environmental impact assessment, and other types of calculations. ROMeo is a next-generation online optimization system with an open application architecture, jointly developed by Invensys and Shell. It combines Invensys’ expertise in thermodynamics and unit processes with Shell’s leading technologies in mathematical modeling and optimization algorithms. It is an online simulation and optimization system based on rigorous mechanisms and focused on equation solving, offering a comprehensive software solution for the simulation and optimization of refining processes. ROMeo has become the standard software product for ExxonMobil’s online optimization of process operations. Unisim is a next-generation process simulation software introduced by Honeywell in 2005, based on HYSYS and UOP technologies. It integrates design, simulation, online optimization, and operator training systems within a single product environment, and adds and enhances capabilities such as heat exchanger network optimization and process synthesis. http://mmbiz.qpic.cn/mmbiz/xjFI9op4iaCnc9dCDiaNApMFYyfeIia5R1ia6dtaryM36g6VMibVibJIXF7Nng7BF5ficvYD5mSVT9BRmKZOaiax3Q2JfQ/0?wx_fmt=png ③ Simulation and optimization technologies and software for key refining units. As the key unit in an oil refinery, the atmospheric and vacuum distillation unit mainly consists of complex distillation columns such as the crude distillation column, atmospheric distillation column, and vacuum distillation column, which feature intermediate feed inputs, multiple product outputs, and mid-stage reflux. The simulation methods for such complex distillation columns mainly include the plate-by-plate calculation method, the matrix method, and the F-factor method. The plate-by-plate calculation method is the most commonly used, and corresponding simulation modules have been developed, such as the PetroleumDistillation Column module in Aspen RefSys, DIS-SIM in Petro-SIM, and the PetroFrac module in Aspen Plus. These modules have been applied in many refineries, where they all achieve satisfactory accuracy. The catalytic cracking unit is an important secondary processing unit in refineries, and the core of its simulation optimization lies in the reaction-regeneration system. Currently, the main relevant simulation and optimization software products developed at home and abroad include Aspen’s Aspen FCC and Aspen RefSYSCatCracker, KBC’s FCC-SIM, Shell’s SHARC, as well as Luoyang Petrochemical Engineering Company’s FCCM. Among these, the reaction model component of Aspen FCC software utilizes the currently most comprehensive and mature 21-lump reaction kinetic model for describing catalytic cracking reactions. This model can reasonably depict the reaction process of catalytic cracking. The riser reactor model comprises one pre-lifting section and two reaction sections. For product separation, a correlation-based method is employed to perform fractionation. This setup enables simulation of how variations in feed properties and operating conditions affect both the product distribution and product characteristics. Additionally, it allows for optimization calculations of the unit; the optimization objectives can be either maximizing the yield of a specific product or achieving maximum economic benefits for the unit. The optimization variables include feed preheating temperature, riser outlet temperature, oxygen content in the regenerated flue gas, settler pressure, regenerator pressure, recycle oil flow rate, recycle oil slurry, and catalyst activity, among others. Aspen RefSYSCatCracker is a simple and user-friendly rigorous mechanism model; it is an oil refining unit module within Aspen RefSYS that can be quickly integrated into the overall plant process. KBC’s FCC-SIM is a mechanism simulation model for catalytic cracking units. It establishes a heat balance model based on rigorous coke combustion kinetics principles, and includes detailed kinetic models for elements such as the riser reactor and regenerator coking. The catalyst activity model enables the selection of optimal feedstocks, conversion rates, and catalysts for a specific catalytic cracking unit. SHARC, introduced by Shell in 1995, is also a rigorous mechanism-based model that can be used to predict the impact of changes in feedstock, catalysts, and operating conditions on the operation of catalytic cracking units, thereby optimizing their performance; it has been upgraded 7 times to date. This model first calculates the carbon and heat balance of the plant, and then estimates the impact of changes in raw material and catalyst addition rates as well as operating conditions on yield and product quality, enabling users to improve plant performance through raw material selection and optimization, as well as catalyst evaluation and control ； Shell’s proprietary material property representation techniques are used; for example, using non-volatile carbon (NVC) to predict coke yield is more accurate than using Conradson residue ； It has an open structure, making it easy to integrate with other online optimizers and refinery linear programming models. The FCCM of Luoyang Petrochemical Engineering Company was developed based on a 13-lump kinetic model for heavy oil catalytic cracking, as well as the intrinsic kinetics of catalyst regeneration and the flow and transfer models of regenerators. It is applicable to various types of reactors and regenerators. Catalytic reforming is one of the important secondary processing units in refineries, and the research and development of reforming reaction kinetic models are the core aspects for achieving full-process simulation of reforming units. Since more than 300 compounds are involved in the reaction, encompassing processes such as naphthenes’ dehydrogenation and aromatization, alkanes’ dehydrogenation and cyclization, isomerization, and hydrocracking, these reactions occur either in series, in parallel, or interact with each other, forming a complex reaction system. Therefore, the key to developing reaction kinetic models lies in reasonably simplifying the aggregated components and constructing a reaction network among them. To date, scholars at home and abroad have developed many lumped reaction kinetics models, including Smith’s 4-lumped model, Ramage’s 13-lumped model, the 16-lumped model proposed by Weng Huixin et al., Jorge’s 24-lumped model, Froment’s 28-lumped model, and others. The most representative one is Ramage’s 13-element lumped model, which is widely used in aspects such as the commissioning of industrial plants, process monitoring, fault diagnosis, operation and design optimization, and research and development guidance, achieving significant economic benefits. Currently, some general-purpose process simulation software already includes catalyst reforming reaction models, such as the Reformer module in Aspen RefSYS and the Ref-SIM module in KBC. The catalytic reforming reaction model built into Aspen RefSYS can simulate continuous reforming and semi-regenerative reforming units, featuring a catalyst deactivation rate based on the catalyst metal loading and the history of the catalyst. KBC’s REF-SIM includes models for continuous reforming, semi-regenerative reforming, and recycle units, with detailed kinetic reactions determined by the carbon number and type. As a key unit for producing high-quality light oil products, hydrocracking units require strict modeling requirements. Currently, several representative kinetic models for hydrocracking reactions include: 1) a four-sum model classified based on the number of products and the boiling range. This model was applied to Romashkin vacuum distillates and Alansk vacuum distillates, and under certain temperature and pressure conditions, the predictions of the reaction model matched the experimental values ； 2) The narrow fraction multi-summation reaction kinetics model – this model predicts the boiling range and yield of products over a wide range of feedstock boiling ranges, and its predictions are quite consistent with pilot plant data; therefore, it is considered suitable for using to interpolate and predict the yields of products that have not been tested ； 3) A seven-sum model classified by boiling point and chemical properties. Currently, the common hydrocracking simulation software includes the Hydrocracker model from Aspen RefSys and the HCR-SIM model from KBC. Among them, the Hydrocracker model includes models for all key reactions, such as hydrodesulfurization, hydrodenitration, demetallization, saturation of olefins and aromatics, ring opening, dealkylation, paraffin cracking, paraffin isomerization, and catalyst deactivation models. KBC’s HCR-SIM model also includes all the key reactions; by using detailed heat balance calculations to determine the rise in reactor bed temperature, it is possible to accurately predict the amounts of products and hydrogen consumed. The hydrorefining process includes reactors, vapor-liquid separators, heat exchangers, stripping towers, centrifugal compressors, etc. The reaction kinetics models include olefin saturation models, desulfurization models, denitration models, deoxygenation models, aromatic saturation models, aromatic ring-opening models, and naphthenic ring-opening models, among others. Currently, the common hydrogenation refining simulation software includes the Hydrotreator model from Aspen RefSys and the DHTR-SIM model from KBC. The Hydrotreater model includes specialized feedstock models for distillate oil, wax oil, and residue oil ； The DHTR-SIM model provides a rigorous kinetic characterization of all key reactions, including hydrodesulfurization, hydrodenitration, aromatic saturation, and cracking, and can accurately describe desulfurization at ultra-low sulfur content levels. There are few literature reports on the mechanism models of delayed coking units. The currently developed simulation software products for coking units mainly include KBC’s DC-SIM, Luoyang Petrochemical Engineering Company’s DCLK, and the Heavy Oil Delayed Coking Process Simulation System developed by the Institute of Process Engineering, Chinese Academy of Sciences. Among them, KBC’s DC-SIM software is a mechanism simulation model for delayed coking units; it is capable of simulating the effects of various key process variables such as coke drum temperature, recycle ratio, operating pressure, and feedstock properties, and can be integrated into Petro-SIM for use. The DCLK software developed by Luoyang Petrochemical Engineering Company is a simulation and optimization tool based on the dynamic mechanism model of the eleventh-stage lumped reaction kinetics for coking, designed to dynamically simulate industrial delayed coking processes. This software is capable of reflecting the unsteady-state nature of industrial delayed coking units, and it can predict the product yield at any point during an operating cycle of the coke tower, as well as the immediate composition of various materials within the tower. The heavy oil delayed coking process simulation system developed in collaboration between the Chinese Academy of Sciences and Mitsubishi Corporation of Japan includes a model of the delayed coking reactor, simulation of the entire delayed coking process, an energy efficiency analysis model based on effective energy, as well as a database system. It can operate independently or be integrated with process simulation software such as Pro/II and Aspen. The directions for energy-saving optimization of ethylene plants mainly include simulation optimization of the cracking reaction and simulation optimization of the cold zone. Currently, pyrolysis reaction models can be divided into the following 3 categories: empirical kinetic models represented by the KSF pyrolysis depth function and PONA value to describe the product distribution during the pyrolysis process ； Molecular reaction dynamics models represented by the Kumar model ； Mechanistic models represented by the SPYRO model. The main software products include: SPYRO (TECHNIP), ECU (SHELL), OlefinSim (KBC), PYRO-SimP (Tsinghua University), etc. Among them, the SPYRO model is the most successful; its latest kinetic scheme includes over 7,000 reactions involving approximately 240 types of radicals and molecules (or equivalent components), enabling the prediction of the yields of pyrolysis products ranging from C2 to HVGO. The SPYRO calculation is divided into two stages; in the first stage, a simplified dynamics model is used to calculate the precise pressure distribution and the temperature distribution of the material ； In the second stage, following the free radical reaction mechanism, enthalpy balance and kinetic calculations for the elementary reactions are performed based on the temperature distribution obtained in the first stage. By integrating with process simulation software and enabling iterative calculations via SPYRO, it optimizes the calculation of multiple scenarios during the process as well as the dynamic simulation of production data, thereby providing a more accurate reflection of the production status. This approach helps to optimize the structure of raw materials and products used in ethylene cracking, achieving energy savings and improved efficiency. By using this model for the operational management of cracking furnaces, raw material selection, and optimization of the production process, it is expected that the ethylene yield can be increased by 1% to 4%. For example, after October 2004, the quality of the ethylene feedstock used by Yanshan Petrochemical was gradually deteriorating due to the processing of external oils in the refining units; as a result, the ethylene yield kept dropping and the operating cycle of the cracking furnaces was significantly shortened. By monitoring the cracking feedstock, process conditions, and composition of the cracking products, and by using SPYRO software for simulation calculations, as well as techniques such as available work analysis and pinch analysis, optimization strategies were proposed to improve the separation sequence, heat exchange processes, and operational procedures. The unit began operating in March 2005, and industrial data show that the ethylene yield increased by nearly 1% without significant impact on the operation cycle. ④Heat exchanger network simulation and optimization techniques and software: Heat exchanger network optimization has attracted attention since the 1960s as a sub-problem of integrated process design. Many researchers have conducted in-depth studies on the comprehensive optimization of heat exchanger networks and applied these methods in practical engineering projects, achieving significant economic benefits. Heat exchanger network optimization methods mainly include pinch analysis, mathematical programming, and artificial intelligence. Among them, the pinching technique was proposed by Linnhoff in 1982. Based on the second law of thermodynamics, this technique is primarily used for the comprehensive optimization of heat exchange networks. Starting from the fact that there are limits to energy recovery, this technique identifies the bottlenecks in energy recovery by combining temperature-enthalpy curves or problem tables, establishes an initial network for maximum energy recovery, weighs investment costs against operating costs, and further optimizes the network to obtain an optimal heat exchange network. In 1986, two groups of researchers, Linnhoff and Ahmad on the one hand, and Floudas et al. on the other, independently proposed different methods for solving the optimal pinch temperature difference in heat exchanger network optimization; the former method is known as \"Supertargeting\" ； In 1992, Yin Qinghua, Hua Ben, and others proposed that ignoring the losses in flow exergy and optimizing the matching units might result in solutions that do not represent the true optimal pinch temperature difference. They provided a detailed analysis of the impact of taking into account flow exergy losses, matching unit optimization, and heat transfer enhancement on the optimal pinch temperature difference, and suggested improved optimization strategies ； In the dual minimum heat transfer temperature difference method proposed by Challand et al. (1981) and Clbert (1982), the minimum heat transfer temperature difference of the combined curve is taken as HRAT to determine the value of the external energy supply; whereas in the synthesis of the network, the heat transfer temperature difference for individual heat transfer units is taken as E-MAT, and EMAT