Note: This article is written and published in Ukrainian and is a translated version of the original published here.
The previous part is available here: "Computer calculations in the educational process. AU2014. Report, Part 2"
What do all these calculations have in common? The fact that they are performed according to accepted standards:
- GOST
- SNiP
- OST
- Regulatory documents
- Accepted techniques
Calculation methods
- "GOST"
- "SNiP"
- "OST"
- Regulatory documents
- Accepted techniques
Here it makes sense to explain why we still use calculations according to GOST, but not according to FEM. Because from the point of view of our realities, it often turns out that GOST calculation methods have a large number of advantages compared to numerical calculations, while computer calculations have significant drawbacks.
So, the advantages of standard methods:
- Availability of standards
- Ease of calculations and control
- No need for software and computers for calculation
- Ease of algorithmization, speed of calculations on a computer
- Independence from the performer
- Reports
- Habit
Advantages of standard techniques
- Availability of standards
- Ease of calculations and control
- No need for software and computers for settlements
- Ease of algorithmization, speed of calculations on a computer
- Independence from the performer
- Reports
- Habit
Having a standard as such is already good. It guarantees the performance of the structure (if it was clearly designed according to the standard). In addition, the standards are quite simple. No matter how complex these techniques are, but, in principle, anyone with minimal reading and math skills can understand the process described in the standard. Although, of course, it is better to get a higher education (especially in this specialty). It (calculation according to the standard) can be performed without the use of computer technology, without expensive software. If necessary, it can be implemented independently on a computer. Even without programming, for example, using a regular spreadsheet editor such as MS Excel.
So, no matter how you look at it, there are some pluses everywhere. Up to the fact that based on the results of this calculation, we immediately receive a ready-made report. And most importantly - all this is familiar and depends on the performer. If we give tasks to two different people, and they perform the same items, they will get the same result from the standard (if, of course, they count correctly).
Why is CAE not used? Because they have a huge number of cons. And this is what I say as a 🙂 solver
Resume:
- Lack of standards and approved methodologies
- They do not work without computers and software that has a high cost
- Complexity of algorithmization, resource intensity
- Dependence on performers:
- The problem of choosing a staging
- Implementation dependency
- Reports
- Lack of habit
Disadvantages of CAE
- Lack of standards and approved methodologies
- Do not work without a computer and software that have a high cost
- Complexity of algorithmization, resource intensity
- Dependence on performers:
- The problem of choosing a staging
- Implementation dependency
- Reports
- Lack of habit
We do not have standards (for numerical calculations using FEM, MCO, etc. for most industries). We don't have the ability to use them if we don't have expensive machinery and we can't buy those expensive apps.
If we try to write a finite element calculation program on our own.... It is probably easier to shoot yourself than to do it. No, of course, there are people who are able to do this, they are very serious people. But not everyone can do it and this is not one man-hour. In general, not everyone is given.
In addition, depending on who counts, depending on his experience, knowledge and proximity to the specifics of the tasks that he calculates, we will get fundamentally different results. Both depending on the method and depending on the chosen setting.
Worse! By doing the same thing point by point in different programs, we will get different numbers. They may not differ much, but they will be different numbers.
Reports look completely different, and habits do not.
So, we have a lot of cons! So why use them (computer calculations)?
There is a main feature of standard techniques, which "follows" the reason why it is necessary to use CAE computer programs:
Quite all standard methods are based on empirical calculation methods. So, it is based on a serious theory: sopromat, the theory of elasticity. But, in addition, they are based on a huge number of correction factors, which are taken, in fact, from experience. But this experience was "taken" 30-40 years ago.
EMPIRICAL (pseudoanalytical) CALCULATIONS
And now it turns out that because of this (due to empiricism with correction factors) there are advantages of CAE that can play a role
- Accuracy
- Visibility
- Versatility
- Independence from the type and complexity of geometry
- Speed of change
- Relationship with CAD geometry
- The ability to take into account new materials, working conditions ...
Advantages of CAE
- Accuracy
- Visibility
- Versatility
- Independence from the type and complexity of geometry
- Speed of change
- Relationship with CAD geometry
- The ability to take into account new materials, working conditions, etc.
These are numerical calculations.
But the first point I put fidelity. Because, despite the errors that computer calculations give, thanks to the fact that we can control the degree of detail and the degree of simplification, we can get much more accurate results than with standard methods. Which completely reduces everything to the beam. Usually - a cantilever pinched beam (relatively speaking).
Visibility. Such results are much more understandable to people, especially modern students, than a couple of graphs.
Versatility. They (CAE programs) don't care what to count – even planes, even ships... And our standard methods are focused on the calculation of clearly defined things. If we take a technique for calculating bolts for aircraft, and try to apply to, say, the same, an ordinary gearbox standing on a foundation, then everything will work. But if on the contrary... Then we will get four times more bolts than the plane can withstand
Hence the fact that computer calculations do not depend on the complexity of geometry (theoretically, there are limitations only in the computing capabilities of the computer). It is also possible to quickly make changes to calculations, geometry, and these calculations have a connection with CAD geometry. Unlike conventional standards, which have no connection with what the designer sees on the screen.
In addition, there is the last point, which we will talk about in more detail and with examples in the next part.
Note: This article is written and published in Ukrainian and is a translated version of the original published here.
2 Comments
