Energy sectors design and the role of design by rules and analysis and finite element calculations
Technical innovations and innovators are the real drivers of this world. Innovations normally do not depend on a single technical discipline. When the knowledge of all technical disciplines comes together, innovations are born. Consider one of the innovations around the world and examine how it was developed by multiple technical disciplines. The creation of energy development sectors, or sectors that run on energy, has been considered the oldest innovation in the human community. Recently, in a technical job interview, I was asked to answer a general question about the energy sector, "Approximately how many pieces of equipment are there in a single oil and gas production facility and how are they designed, manufactured, and assembled in the facilities? The position I was interviewing for was the ‘Technical Authority’ for the design of oil and gas production facilities and process packages. The technical authority is the person responsible for the technical design of the appropriate equipment and process units for the required oil and gas processing conditions in the facilities. To this simple question, I replied to the interviewer: "We have a large number of pieces of equipment," but in reality, they are countless. Similarly, we also have a variety of packaging units for oil and gas processing. What is an oil-and-gas processing unit? A processing unit is nothing but a set of equipment that is grouped together in a single skid and has to be operated together for specific purposes in the plant. This means that the oil and gas processing unit serves different purposes, and one of the most common purposes is the filtration of water, oil, or gas. Once the crude oil has been extracted from the ground, it has been separated or filtered into three parts: oil, gas, and water. These three are further filtered to obtain pure oil, water, or gas. For this purpose, we need a packaging unit that can fulfill this task well. The packages are normally developed with numerous research and development activities, especially for oil, gas, and water processes and other internal processes as per the needs. Normally, each packaging has its own purpose, developed by different researchers around the world. Most successful inventors of process packages now run their own companies to offer their products for the needs of oil and gas production as well as for the needs of other energy development sectors. In the oil and gas sector, these equipment and packages are handled by the relevant disciplines such as mechanical, static, rotary, packages, piping, stresses, structural, civil, instrumentation, electrical, electronics, telecommunication, etc. What are static and rotary equipment? Static equipment is nothing more than a type of equipment that does not contain rotating components. Rotating devices are reversed, they have a rotating component. Example: pumps, compressors, turbines, etc., packages are, as I have already explained, the assembly of all kinds of suitable devices for specific purposes in a single skid. For example: TEG dehydration, methanol injection package, corrosion protection system, sulphate removal unit, chemical injection system, distillation package, etc. If we consider only mechanical statics, there are several types of equipment among them, but pressure vessels, heat exchangers and storage tanks are the most important ones, but there are several subcategories among them. For example, if we take a pressure vessel, there are several types of pressure vessels, e.g. test separators, boilers, reactors, tall columns, scrubbers, silos, etc. are some of the types of pressure vessels available in the market, and each of them is designed differently for its purpose. The truth is that designing and manufacturing these test separators, boilers, reactors or columns requires an independent design specialist with specialised knowledge and a team to get the job done well. This is because each has its own patent design held by different inventors around the world. Some of the manufacturers undertake the repetitive manufacturing of pressure vessels according to the designs of the patent holders. Normally, the design of mechanical components is categorised into two categories, irrespective of the discipline it belongs to. These are the ‘design by rules’ method and the ‘design by analysis’ method. What is design by rules? And analysis? Design by rules is nothing other than the development of engineering on earth. This means that all technical components, which are not just oil and gas components but apply to all areas, were designed using the empirical formulae developed by the old researchers. For example, the calculation of thickness and the generation of stresses due to loads in all engineering components were calculated based on the applicable empirical formulas developed by the old researchers. What is meant by analysis? An analysis is nothing more than the calculation of specific thicknesses for the design of a component, which cannot be based on a single calculation or value. For example, if we want to design a component to withstand an internal or external load of 10 bar, we can use an empirical equation to determine a thickness of 6 mm to 8 mm when calculating the thickness that must withstand this 10 bar. Is this thickness of 6 to 8 mm really necessary? Or can we make all parts of the component safer under the 10 bar load conditions with this 8 mm thickness? Sometimes the actual thickness required to withstand the 10 bar load is greater than 8 mm or less than 4 mm to 6 mm? What would happen if we use the excess thickness? The cost of design, manufacture, operation, transport, maintenance, etc. would increase. So what is required? The optimal design. Optimal means that the design thickness should not be exceeded or undercut in the required operating and other design scenarios. The thickness could be perfect, i.e. it could be optimal, i.e. it could be more suitable for the 10 bar loading conditions. How is the optimisation done in the design? We have to repeat and optimise the calculation for different possible real-world design cases and then find out which thickness is best suited. Finding out the real design scenario of any technical component is not a simple task, but a challenging one. This is an extensive research topic that I cannot go into in detail here in a single article. The designer must determine the optimum loads for a particular component. This includes the optimum environmental conditions in which the component is assembled, the optimum operating conditions, the optimum choice of material, the optimum shape of the material, the optimum thickness of the material, the optimum geometry of the product, etc. After finding out the optimal conditions for all relevant design parameters, the designer needs to optimise the shape and thicknesses based on the different applicable design load cases. The optimiser should know which potential design factors should be considered when creating the right design for any of the design applications. In some cases, over- or under-design can affect the performance of a particular component or project. "Analysis" and "optimisation" are closer to each other. Optimisation of design components is done by analysing various suitable, accurate design data of the component in question. Analysis requires a variety of data and optimisation determines the right data for the design. Can we perform ‘analysis' and ‘optimisation' using only the ‘design by rules’ method? The answer is yes. In the past, technical data analysis and optimisation were carried out using practical manual calculations. The technical design drawings were also produced by draughtsmen. The list of plant design, operation, and maintained details was kept using various notebooks, drawings and reports. And some of the leading engineering companies still work with hand drawings to ensure the accuracy of the design dimensions. This is not just for the oil and gas or other energy sectors; this type of documentation is used in all engineering sectors, including automotive, aerospace, nuclear, infrastructure, defence, etc. Analysis and optimisation are at the heart of all engineering disciplines used to develop and design their technical components. When the first oil or gas production was launched, ‘analysis' and ‘optimisation' were already commonplace in the engineering sectors. When we create specific designs for energy sector equipment, packaging, or aerospace, or nuclear or other specialised fields, we cannot avoid analysing a lot of data in all areas of development and optimisation. When planning plants in the energy sector, it is important to analyse the data in order to be able to plan correctly. This is similarly true for oil and gas process packages. I would say that typically countless hand-crafted engineering calculations have been done to build a single energy plant for society. In fact, many of the oil and gas facilities in operation today were built before computers were introduced to this world. We can still refer to the old calculations that were done by the old engineering experts by hand and using typewriters in our industries or in reports or books. The reason why I am recalling the old documentation procedures in this article is because there is a reason for it and today's procedures are also in line with the old procedures. After reading this article, it becomes understandable. When the computer came on the market, analysis and optimisation became easier as engineers started storing the data on their computers to optimise it. Then, after the introduction of printers and scanners. The designed files were scanned and stored on the computer for reference. Similarly to the computer, electronic components, instruments, telecommunications, etc. were also introduced, and the design of oil and gas facilities became much smarter than the previous methods. However, when the computer programming languages BASIC, C, C++, etc. were introduced, engineering calculations and designs were computerised. The technical manufacturing processes were also systematised. CAD, CAM, CAE and other programmes were introduced. Specialized software, such as AUTOCAD etc. for drawing on the computer was introduced. However, these were developed based on the old handwritten manuals, procedures, schemes, and theories documented in the olden days. The design based on "rules" and "analyses" required for the drafts was modernised by computers.
What is the Design by Rules method? Where do the rules come from?
In engineering, we divide designs into the development of ‘new products' or ‘new technologies' or the development of ‘existing products' or ‘existing technologies'. However, when we design a component, whether it is a ‘new' or an 'old' product, certain rules and regulations must be followed to carry out and complete the designs if the component is based on old inventions. The old engineering rules were established by old researchers based on experiments and inventions they conducted in their time for their products. The development of technology started when living beings began to live in this world. As people began to invent languages and alphabets, they also began to teach the younger generations in a procedural way. Maths was introduced and construction began. And the rules and regulations for engineering were introduced. Professional organisations and engineering councils were set up. However, due to various failures in the production of engineering products and structures, not all the rules or regulations developed were fully accepted by all engineering professional bodies or councils, but the rules that were accepted were strictly followed by the various engineering associations and research and development centres. Compared to today, the research and innovations in engineering in earlier times were enormous. The creation of new equations, especially for the empirical equations, and the creation of symbols, the creation of numbers and mathematics, the creation of languages, the creation of alphabets, new theorems, inventions in chemistry, inventions in energy production, inventions in electrical engineering, electronics, telecommunications, etc. were unlimited. The creation of a simple empirical formation requires a lot of optimisation to ensure that the equation created is correct. In fact, in earlier times, many years of research were done to develop an empirical equation, as it was based on evident experiments at that time. Considering how many unbounded equations were created in ancient times and the limited tools and equipment available to create the equations, one wonders how these equations were created without computers, other sufficient library records, or the like at that time. We have to give credit to these inventors because their contributions are the reason why most technological inventions are taking place in all sectors today. Take the oil and gas sector: every component used there has its own design history and its own empirical formulae to develop the entire system. Countless research-based equations, diagrams and tables have been invented to develop the equipment. Based on the research data from Design by Rules, various codes for oil and gas equipment design were created. The codes, such as ASME, PVRC, BSI, USCG, CSA, API, UL, AWWA, ANSI, TEMA, ASTM, ICBO, SSPC, ASCE, OSHA, AWS, CGA, EJMA, EPA, HEI, ICC, IMECHE, NBBI, NFBA and others, were the collection of various research findings for our guides. We must now ask ourselves if we are able to recreate a simple empirical equation for any of the components with the resources available to us today? And, even if we do create an equation, how many of them will accept this development for their applications? Nowadays, creating a new equation is a hectic and difficult task, and even if we create one, it is extremely difficult to use this equation for real-time applications. The reason for this is that it is always questionable whether the new equation provides a better solution than the old one Moreover, it is difficult to prove these types of exercises so that they are approved by the professional organisations in the energy sector. The old researchers from the USA and Great Britain such as Mr Hooks, Moller, McCloskey, Harsermann, K.P.Wickman, A.G.Hopper, J.L.Mershon, L.P.Zick, K.Abakians, S. Timoshenko, R.J. Roark, L.E.Brownell, R.H.Young, M.B.Bickel, C.Ruiz, H.H.Bednar, S.S.Gill, J.F.Harvey, M.Hekenyi, M.L.Betterley, W.Soedel, W.Flugge, R.Szllad, Dennis R. Moss, Kohan, Anthony Lawrence and others contributions to the energy sectors were unrivalled. I have only mentioned some of the names of the contributors here. In fact, the number of inventors in this field is innumerable in terms of innovations in the energy sector. Their equations, processes, reports are the repeated references for many of the sectors till date, be it for the development of new technical products or for the development of new technologies or for the overhaul of existing products, or for the development of new existing technologies, whatever it may be. However, if the growth of the development of new technical equations or the development of new codes or rules is decreasing compared to the past, we can understand that the research and development activities in these areas have drastically decreased, and we all need to work on improving these mistakes.
What is design by analysis? And finite element analysis?
Design by analysis is nothing more than designing components based on a detailed analysis with the right researched design data. As we discussed in the first section, the chosen component must withstand exactly 10 bar to achieve the required thickness. The analysis would help the designer find the perfect thickness. For example, what thickness is required for a load of 8 bar? What about a load of 9 and 10 bar? What is the required thickness? What is the upper limit for 13 bar? What are the thickness for 11 and 12 bar loads? What are the maximum and minimum loads to be considered? After obtaining this design data, what thickness would be most appropriate to protect the structural component under the exact loading conditions? The designer is the one who has the opportunity to figure this out and decide. In design, most developments are made on the basis of assumptions and then on the basis of experiments in the real application. The limit value for the safety factor in the design of a material or component is not fixed. It must be determined on the basis of a detailed analysis: "What is the safety limit we should consider for this particular element and what would be the best safety limit? For this purpose, different types of experiments are needed to provide industries with a good design. The research data for the energy sector has only been collected based on this type of in-depth research and development studies. When developing equipment for the energy sector, various statistics need to be taken into account. The FOS, load and temperature limits, permissible design, etc. have been documented based on the research data for various devices. To understand the energy sector data, the designer must have special skills and experience, otherwise, the use of the data would be incorrect. The design decisions for the development of new components for energy equipment are only made by experienced engineers. Also, for the design decisions based on codes and standards, one must have at least ten years of exclusive experience in engineering design and have proper training to use and judge the code-based design and analysis for the industry. This is the requirement that the code itself advises to avoid the misuse of codes and incorrect design practises in the sectors. Most equipment in the energy sector is highly lethal as it deals with toxic liquids and gases. Therefore, these devices usually have higher safety limits and guidelines for the design of their components, and designers leave no room for error in the design of their components. And exclusive software, codes and standards are also recommended for analysing the design. Why is the software exclusive and why are most of the software programmes developed for general purposes are not used for the energy sector? The main reason is that the software must fulfil the guidelines and standards of the energy sector in order to design the energy components. Can the software be developed based on regulations and standards and other research data? Yes. Then the general purpose software cannot be used for the design of energy components? No. It can be used. Actually, all types of software developed for engineering design can be used, but the strict rule is that the designer must comply with the rules and regulations of the code of practise for the software he wants to use for his design. Can the software be certified by the professional organisations such as ASME, BSI, EN etc. as they originally have the codes and standards of the energy sector? No. Although the software has been developed based on international codes and standards, it is not certified by professional organisations. But the software may be used for their designs if it fulfils most of their requirements. Regardless of whether the software is developed based on codes and standards or for general design applications, once the calculation is done with it, it must be verified by independent certification bodies such as LR, DNV, BV, DIN, etc. to ensure that the software has been calculated based on ASME, BSI, or other codes in order to obtain approval for the design performed by the designers. The design should be submitted to third parties for confirmation that it is based on the code rules. These third parties vary from country to country, and there are a number of agencies around the world that are specifically responsible for energy sector designs. Most of the approved design software for the energy sector is fully compliant but is not certified by the above mentioned professional organisations, instead the calculations of these software are verified by third party providers to obtain certification for the particular calculation for approval and manufacture. These third-party providers are approved directly by the professional organisations such as ASME or BSI and are responsible for developments in the energy sector in various locations around the world. Finite element analysis is a vast topic in the energy sector, and how is it carried out and approved in these sectors? Rather than calling it finite element analysis, we would refer to this type of calculation as ‘elasticity stability testing’ of designed components using the finite numbered element method. Finite element analysis can be carried out in almost all areas of engineering. What is meant by an element? An element is nothing more than a derived shape, such as a square, rectangle, triangle, etc., which is available to subdivide the engineering models or components as finite numbered shapes to calculate the strength. The subdivision of the model using the elements is a very complex process in finite element calculations. Typically, these elements are used to transfer or constrain the forces and moments throughout the body of the model. The complexity and challenges of the simulation are due to the meshing of the elements and the way it is performed, which leads to drastic changes in the simulation results. FEA with developed software tools always requires a thorough analysis by an experienced designer to find the right solution. Let us discuss this in the last chapter. Since most of the components in the energy sector are very critical, design calculation based on finite element analyses are used in all design areas. Equipment design, package design, structural design, piping design, pipe support design, special saddle design, internals design, column shaft analysis, filter element analysis, coalescer element analysis, anchor block analysis, cable and chain analysis, reinforcement analysis, tower elastic stability, lifting analysis, in-situ analysis, transit analysis, tug pull analysis, wind analysis, seismic analysis, geological expertise, electrical system design, electrical network analysis, civil component analysis, foundation analysis, pile design and analysis, fluid flow analysis, vortex impact testing, sea wave analysis, hull design and analysis, topside design and analysis, mooring system design, temperature analysis, vibration analysis, underground pipework analysis, underwater pipework analysis, Cryogenic analysis, nozzle analysis, blade analysis, structural footprint analysis, corrosion prediction and analysis, paint effect analysis, burner temperature analysis, furnace or oven temperature analysis, welding analysis, welding deformation analysis, spring analysis, spring tube analysis, multi-wall temperature distribution analysis, metal conveyor analysis, shaft and impeller analysis, turbine skid analysis, compressor analysis, blade analysis, Tank deformation analysis, bellows analysis, pump seal analysis, failure mode analysis, maintenance reliability analysis, offshore jetty analysis, offshore crane analysis, offshore ship crane analysis, offshore rig analysis, floating offshore platform analysis, locking mechanism analysis, crack analysis, frequency analysis, noise analysis, fire analysis, brick analysis, burnt component analysis, composite material analysis and much more. If we try to enumerate where FEA calculations are applied in the energy sector, the number of those who use it is almost unlimited and is used in all technical fields. Whenever we design new components, we cannot avoid FEA. Verification of the finite element calculations of the supplier's design is one of the most important requirements of energy sector owners like SHELL, ARAMCO, EXXONMOBILE, CHEVRON, PDO, BP, ONGC, or HP, etc to ensure that the supplier's design is safer for their operation. This requirement applies to all disciplines of energy sector design. Moreover, the designer, the technical authority, or the engineer in charge has the right to perform FEM calculations wherever required or has the right to request the vendors contracted to design the components to provide the relevant FEM calculations wherever the design is critical to ensure the stability and quality of the design. This is actually the responsibility of the designer or technical authority. All O&G owners usually impose this requirement on all their projects. But as I said before, the FEM is based on the rules of the energy sector. Some suppliers refuse to publish their FEA studies due to the confidentiality of their products. However, if they work for the energy sector and the EPCM or the client requires it, they must disclose the results to ensure the quality and safety of the product. FEA is considered an effective tool in technical applications. Many of the devices and systems in the energy sector are only so well designed because FEA is successfully behind them. Performing an FEA calculation is not an easy task, as not all designers can do it easily. It used to be very tedious, as FEM was not done by software but manually by handwritten methods, but today it is done with the help of software. This applies not only to the energy sector, but to all sectors. Why are FEM calculations so difficult? Because in engineering, we usually choose FEM when it comes to more complex designs. FEM is mainly meant for complex engineering problems. Successful FEM only helps to develop innovative products for society. Some problems require multiple simulation iterations to get the correct solution. Therefore, the FEA analyst should have extensive experience in all engineering fields, including areas of expertise, design and development activities, codes and standards, operation and maintenance of engineering components, and applicable software systems as programmed to perform the FEA analysis. There are a number of challenges associated with performing FEA analyses and methods. ASME is the only professional organisation that has the specific codes, rules, and standards for this innovative subject to solve the critical technical problems in the energy sector worldwide. Research and development activities in the field of numerical methods for mechanical strength analysis are rather low worldwide, or only very few organisations are involved in solving these tasks. Compared to the old researchers, the new researchers working in this field are also fewer in number. Moreover, all the old developments in the field of numerical techniques have not been fully implemented on the software side. The design by analysis is therefore nothing more than the initial analysis of various data and the subsequent design using the optimised data. Design by analysis can also be referred to the components that need to be designed and analysed as mentioned before, such as ‘In place Analysis', ‘Transit Analysis’, ‘Vessel Shell Analysis’ or by ‘Software Analysis’ and so on. We will discuss the general difficulties of FEM in the next section. But before that, let us understand what finite numerical analyses and methods are, how they have evolved, and what benefits they have for the design industry.
Recommended by LinkedIn
Numerical analysis: What exactly is it? And what are its advantages?
Another name for finite element analysis is nothing other than finite numerical element analysis. And how does this FEA or numerical analysis find its way into technical designs? Indeed, in engineering designs, components come in different shapes and sizes and under different operating conditions. However, when designing, it is not easy to check the elasticity of each component according to the design conditions. For example, if the component is a simple square, circle or rectangular shape, it is quite easy for the designers to determine the elasticity using formulas. However, if the shapes or sizes of the components change, for example, if the object is a combined square with circles and rectangles, the elasticity of the component is not so easy to determine because detailed elasticity tests are required in all areas where the shapes change and are weaker to understand whether the elasticity is safer in all portions of the component. Can we easily design the complex shaped structures? No. Assume a cantilever beam and assume that one end is fixed and the other end is free and loaded. The deflection and shear due to the loads acting on the beam can be determined using empirical formulae. Consider a simple shaft, regardless of its shape or whether it is hollow inside but experiences a torque or a sudden load or impact at the end, and the shear and ductility effects can be determined using the equations. Consider a trolley rail in a coal-fired power plant that carries heavy loads to transport coal or other materials. Similarly, consider the shafts of pumps assembled with impellers carrying tangential or radial or torque or shock loads and whose elastic stability and bending are to be determined using the formulae. If you want to design a pump motor gearbox with a specific pressure angle or a power transmission capability with the corresponding speed ratio, you can use the formulae. Similarly, different types of gears such as spur gears, bevel gears or pinions can be calculated using the formulas and their elasticity behaviour can be easily checked. The circular shaft that is hollow in the centre is called a pipe and its use in the energy sector is uncountable. There are pipelines that are laid from country to country to transport oil or gas from one country to another. Transnational pipelines are the biggest challenges in the oil and gas industry due to their length. Some of the pipes run through the mainland or through an offshore area as well as underground. Take a vertical filtration column with a length of around 25 to 30 metres, which is laid vertically on the ground. This is also a type of cantilever, but the underside rests on the ground and the shear and transverse moments due to the external and internal loads are determined using the formulae. Similarly, we have formulas and procedures for designing components of any type or shape. However, when we design and manufacture the above engineering problems, we normally do the strength calculations to understand where the thickness needs to be more or less or where reinforcement is required or where a bellows needs to be added or where a water pipe is attached or where a double wall is required, etc. Anyway, if the components are handy or smaller, the equations solving are common practise in all engineering sectors. We can also perform design optimisation using the same formulas for these small and non-complex design problems. For the discussion, let us take the design of a filteration colum with a length of about 30 metres. And let us assume that this 30m is manufactured in the workshop? Can we use the 30m straight pipe as a column for a filtration application on site? No. Actually, the column has a conical shape and the thickness is normally different from bottom to top. However, the non-tapered straight column is also used in the sectors. If we install the column vertically, it must be able to withstand the external, internal and other mechanical loads during operation on site. How are the stressed areas taken into account when designing the columns? Actually, this has to be done section by section. The sections were either provided with uniform or uneven lengths. There are two reasons for this. The first reason is that the tall columns cannot be produced as a single tube. The loads acting on the individual sections provide information about the exact design flexibility requirements for the column. The column I have put up for discussion in this article is a process column in the O&G industry. Let us forget about the production methods of the column. Let us put the 30000 km of pipelines on the ground and assume that the pipes are stressed by the process fluid flowing through the pipes. Suppose the 30000 km of pipelines can be laid straight on the ground? No. Because, as we can see, the ground conditions of a country or state are not flat. They zigzag and run at different heights and depths and under different sanding conditions. How difficult it is to design a pipeline that can run over a long distance. The bends, flanges, supports, louvres, valves and associated construction work are the biggest challenges for engineers. Codes such as ASME and BSIs provide specific guidelines for this type of piping design. The peculiarity is that it assumes how the hoop and longitudinal stresses develop on the pipes, in the bends, and in other joints during the journey, and how the lateral stresses propagate through the pipes, so each joint of the pipes requires a special design to avoid loads and moments. And if the pipes corrode or lose their material, how will their design be affected, and will the system start to vibrate? Let us assume that with 30000 km of interconnected pipes, it is impossible to consider the entire length as a single unit and analyse in the system as a whole to avoid vibrations, loads, or other design uncertainties. The solution to this type of problem in the industry is piecewise approximations. Divide the system into different equivalent or non-equivalent parts depending on the requirement, and analyse each of the parts, and transfer the results of one part to the others. This is the method used when developing larger components, if they are larger and in a different development environment. When the component is larger, the behaviour of the component is not so easy to find out in practise due to the design loads. Only the approximate result can be used for the design of such components. The same applies to complex designs. If the design of a component is very complex, the stress transformation of this type of component is also not easily feasible. Therefore, a piecewise approximation is always required to reinforce this type of construction. Piecewise approximations using the empirical formulas are called finite element approximations. We can also refer to this as a numerical approximation. This is the relationship between the finite element calculation based on rules and analysis in industries. As the parts, elements, or numerical values increase in size, the approximation results also vary. Usually, very detailed analyses are required to create the best design using finite element methods. FEM is used in all industries, depending on the design needs. In the energy sector, almost all designs are performed using the finite element method. In this sector, there are simple, complex, and very complex designs, so a piecewise approach is inevitable. Usually, these types of energy problems are categorised into one-dimensional and two-dimensional as well as beam and frame elements, isoperimetric finite element models. Then the problems are categorised according to the type of structure to be solved in the industry. The problem-oriented FEA steps are similar, which we will discuss in the next article with suitable practical examples.
Conclusion:
Finite elements are a method for the numerical solution of a differential equation. It follows that, without differential equations, there would be no finite element equation. Many industries and practicing engineers and scientists use the finite element method as an application for the design of components. The state of the art of finite element analysis a few decades ago was responsible for this. Design by rules and analysis are part of the engineering design of all types of components. This article discusses general information about the design calculations required for designs in the oil and gas energy sector. It also explains what a design by rule and a design by analysis are and what the differences are. It also discusses how finite element calculations are used and what areas are involved in the development of the energy sector, especially in the oil and gas sector. A few areas where numerical calculations are applied are mentioned to help designers learn and understand more about analysis based calculations in the design of equipment and packages in the energy sector.
Author:
A.M. Senthil Anbazhagan is a research scholar with a focus on the growth of development activities in the energy sector. He is currently working with EPCOGEN Private Limited, Hallmark Towers, 9th Floor, No.35, South Phase, Thiru-Vi-Ka Industrial Estate, Guindy, Chennai-600032, Tamil Nadu, India. While executing oil and gas projects, he has held positions such as Engineering Manager, Technical Authority, Principal Engineer, Lead Engineer, etc. He has more than 17 years of professional experience in global oil and gas sector development and has published more than 16 research articles on oil and gas facility design, including its challenges and solutions. His mobile number is +91 7339659661 and his email address is anbazhagan.senthil@gmail.com. The purpose of this paper is to promote the education and development of the energy industry. Company policies, design rules and regulations, EPC contractors, international conventions, and legal requirements should be followed when putting the design criteria into practice.
Lead Mechanical Engineer (Static and Package Equipment) CEng(India), CREP, PMP®
8moInsightful!
Caltec Co-Owner/Director/Innovator
8moExcellent article A. M. Senthil Anbazhagan, PhD., PE., I throughly enjoyed reading it. 👍🏼