Metallurgy

[loosely speaking] experimental works: however, sometime you have to learn some data visualization and post-processing techniques (such as filtering, ...) in experimental PhD research as well. But therein the crux of your work is based on your measurement techniques and results not numerical modeling

Kaveh Zamani you are right. But in some fields numerical modeling is compulsory in addition to experimental works, such as Structural mechanics, CFD and Aerodynamics. So i am asking about the areas where experimental work have enough worth.

All experimental parts of mechanical sciences such as experimental fluid dynamics (wind tunnels, supply pipes and channels, thermal systems), material sciences, testing and experimental validation. Numerical modelling is highly preferred to be provided with experiments, but not compulsory.

However as Kaveh Zamani mentioned, this gap quickly shrinks and you are required to acquire more mathematical modelling skills for maintaining the competitiveness in the job market.

Mohammed Ba Zuhair In last 10/15 years areas such as Composite materials and Wind Energy required only experimental work. But recently experimental work in these area is almost saturated or commercialized so in order to do competitive research we must Need Numerical modeling. But I think still some areas like Battery and Solar cells do not required mathematical modeling,,but it may not be in Mechanical Engineering domain

There are still a lot of things that can be done in a tow tank or wind tunnel that have never been done due to lack of insight and / or funding. What are the lift and drag properties of a single shipping container versus 3D angles of attack? A model railroad scale model can be had for about 20$. What are the lift and drag properties of a human drowning victim as it free falls through the water column?

Probably some good work could be done in one of those indoor skydiving venues and a range of body types could be tested vs angle of attack quite easily.

There is a little bit on both in Hoerner, but not enough to build out a parametric model in CFD. What is the wind loading vs azimuth on a high speed catamaran or trimaran? There are tons of data on mono-hulls but nothing usable in the open literature for Cats and yet there are many of them in service.

There are many different experimental hydro/ aerodymanic tests that can be performed for ordinary slow speed objects without haveing o have very sophisticated testing facilities or massive computational efforts. Just updating Hoerner with a global literature search would be a very welcome addition to the technical literature and the writer would be widely cited for the next century.

Sean Kery Sir of course there are unlimited area in Wind turbine or other Structural and CFD analysis or Fluid/Structural interface. So that is my point of discussion that it involve most of Mathematical modelling or Optimization work. Same in other cases like aerodynamics. I wanted to ask the area that involve mostly experimental work.

Hi

In case of structural mechanics, a new look, the so called state based philosophy, is developed, which considers a phenomenon as the change of state between the intact (origin) and the collapse (destination). This philosophy is able to determine the ultimate strength of beams, columns, stiffened panels, and etc. , exactly, without construction and solution of any governing equation!, but via some simple and explicit relations. Moreover in this method, there is no difference between the deterministic and stochastic methods. That is the results are in comply with both of the deterministic and stochastic approach. Problems of fracture mechanics, damage mechanics, structural steel design, probability, epidemiology were analysed in a simple but exact manner by this method.

For more information look for the state based philosophy in the Researchgate and in the literature.

Material science

Bhaskar Bhatt yes that is why most of researchers are shifting from Engineering towards material sciences specially in Asian countries, but some Engineering institutes in our countries do not consider material science in Mechanical domain,,hahahah

The work I described is strictly experimental and empirical physical model testing. Some minor computer analysis to reduce the data is necessary but only requires excel or a similar spreadsheet. This must be done carefully with some guidance from the ITTC guidelines perhaps but any undergrad engineering student and many high school students can handle the math involved. It doesn't require CFD although it can be used to ground truth / reality check results from CFD which are always a bit Iffy without such validation. While you can do optimization studies by varying selected parameters of the test conditions and or the model geometry and surface roughness for instance, again it does not require advanced computations, only physical model test experiments. Certainly it can be done in conjunction with any level of computational modeling available, but that part is an "also ran", and not a necessary precursor or adjunct. Good luck

Hi Imran Ali

I believe that in all areas of mechanical engineering, you need mathematical equations to compute and write the governing equations. In the linear approach, mathematical equations are less needed, but in nonlinear methods you must master advanced mathematics.

As someone who deals with highly non-linear hydrodynamics all the time, I must disagree. Without physical model tests, it is impossible to calibrate the numerical methods such that they give realistic results. With the empirical results one can calibrate the viscous and damping inputs necessary for the numerical models to come close to the physical reality. Without the empirical calibrations and validations the purely numerical results are always suspect and often grossly wrong. The physical model tests can yield relevant and useful results without ever running a simulation if they are done right. William Froude and David Taylor and the Wright Brothers did some outstanding work that is still relevant today and they did it decades before the first computer software existed other than Lovelace and Babbage.

Welding area requires less calculation than that of others

If you want to make welding in a piece as a designer, you'll definitely need to make analytical calculations, which are derived from analytical calculations of material resistance, mechanical engineering design, and perfect elasticity.

Mechanical engineering consists of machine design, thermal engineering, manufacturing/ production engineering and industrial engineering as broad specializations and in my view are four pillars of mechanical engineering. Out of which manufacturing/ production engineering area involve simpler calculations and the work mostly done on experimental setups which give direct readings. This can be supported by theory from the literature.

The area of mechanical operations that include crushing, grinding and particle size separation has the least involvement of mathematical modeling and numerical analysis. Much of the equations available are obtained empirically based on data collection of input and output variables.