37

Beware of overfitting. A more accurate model of gathered data from a system may not be a better predictor of future behavior of a system. The above image shows two models of some data. The linear line is somewhat accurate on the training data (the points on the graph), and (one would expect) it will be somewhat accurate on the testing data (where the ...


26

The most obvious downside is cost, all engineering projects have a finite budget and spending more money than you need to is clearly a bad thing not to mention wasting time. There can also be more subtle issues. Things like FE analysis are always approximations and sometimes adding unnecessary detail can introduce artefacts and make it more difficult to ...


13

There are a few reasons. From a purely pragmatic perspective, it's due to time constraints. The requisite time to solve a model increases far, far faster than the level of precision, and whichever level is adopted is subjective, anyway. This is also affected by the fact that excessive accuracy is mostly useless. After all, your model might be 99.999% ...


10

Yes, you are correct that there are different definitions of Systems Engineer that vary by company. In fact, different business units of the same company may even use the term differently. A job posting on Stack Overflow Careers from Booking.com has a Senior Systems Engineer - Systems Architecture role. This role has responsibilities such as taking "...


8

If for some reason "circular dependency" doesn't work—it seems clear enough to me—you could also say that subsystems A and B are "interdependent" in their design. These are essentially synonyms and in either case you may have to go on to explain exactly what you mean and why. But at least it's a bit more concise and less awkward a construction than "...


7

I don't believe there is any conflict but variation in how Human Resource choose to define system engineer positions within specific organizations. It is my opinion that System Engineer has a broad definition mostly related to the specific industry. In my experience system engineer is an interdisciplinary professional of engineering as describe in your ...


6

Systems Engineering pre-dates IT. The classical Systems Engineering has roots in aerospace industry (for better, or for worse). Projects were were getting multidisciplinary, and complex, and required multiple contractors to complete. So appeared a need for a kind of engineer to keep track of various aspects (such as weight, for example) on a relatively ...


6

An extremely accurate model may require a prohibitive amount of input data. It might be possible to generate an excellent model of weather systems, for example, by taking as input the position and velocity of every gas molecule in the atmosphere. In practice, such a model would not be useful, since there's no realistic way to generate the proper input. A ...


5

"Too accurate" is not monotonic. It can actually create an illusion of fidelity which makes you think it's worth pumping more money into the simulation. This becomes very important when you're presenting data from mixed-fidelity models, where some parts are very detailed and other parts are very coarse. A real life example I had involved sampling ...


5

I thought I would expand a little on the answer offered by Karlo. Long story short, I would not try to calculate the analytical time response of a system to a square wave. That would be a serious pure-maths exercise, and not necessary for most engineering applications. Instead, I would suggest using the step response analytical vs. simulated to validate ...


5

One way to think of MEMS is regular mechanical systems, but at a very small scale. Often these tiny systems are fabricated using technology developed for making silicon electronic chips, like nanometer-scale photolithography and etching. However, describing MEMS as just downscaled regular mechanical systems is doing the concept injustice. Various physical ...


5

If you want to evaluate a continues time transfer function at a specific frequency $\omega$ in rad/s you substitute $s$ with $j\,\omega$. For a discrete time transfer function you substitute $z$ with $e^{T_e\,j\,\omega}$. In order to see why you have to substitute $z$ with $e^{T_e\,j\,\omega}$ you can consider the transfer function $z^{-1}$, which is a ...


4

Another term that would be applicable is mutually reliant.


4

It's a trade-off between the simplicity of the sensor and how comprehensive the reported data is. The locked rotor sensor only reports whether or not the motor is spinning, not how fast it's spinning or how much resistance is on the fan. It is a very simple signal though, it's either on or off. On the other hand, the tachometer signal is a bit more ...


3

For a MIMO system $y(s) = G(s)d(s)$, with $m$ inputs and $l$ outputs. Consider a fixed frequency $\omega$ where $G(j\omega)$ is a constant $l \times m$ complex matrix. For the sake of simplicity the matrix $G(j\omega)$ is written as $G$. In short, the singular value decomposition (SVD) states that any matrix $G$ may be decomposed into an input rotation $V$, ...


3

MEMS is a semiconductor based process that can create miniature structures in silicon. When these structures (sensors) are subjected to mechanical stresses (pressure, acceleration, yaw etc) they are able generate electrical signals. These signals are conditioned using an Application Specific Integrated Circuit (ASIC) to create a sensor system. Below are ...


3

rlocus() takes the open loop transfer function as an argument, not the closed loop. i.e. G not T. Look at the documentation for rlocus here https://www.mathworks.com/help/control/ref/rlocus.html


3

This is because for all three of of your choices for the gain the closed loop system is unstable. Namely, for just an integrator as controller the gain should be below 1.5 in order for the closed loop system to be stable and all of your gains are above that.


2

sys·tem /ˈsistəm/ noun: a set of connected things or parts forming a complex whole, in particular. a set of things working together as parts of a mechanism or an interconnecting network. plural noun: systems "the state railroad system" synonyms: structure, organization, arrangement, complex, network; informal: setup "a system of canals" en·...


2

First point is that signum is not continuous with the jump at $t=0$. Second because of the absolute value in the integral it has the same $E$ and $P$ as a constant signal $1$. This means that $E_{sgn}=+\infty$ and $P_{sgn}=1$


2

First off, the equations you have are not correct. For example, if equations 1 and 5 are added it results in $V_1+V_5=0$. The l.h.s of the first four equations should be $V_1-V_5$, $V_2-V_6$, $V_3-V_7$, and $V_4-V_8$. Then you are have four equations in the four unknowns $X_1$, $X_2$, $X_3$, and $X_4$ after taking Laplace transforms with zero initial ...


2

I cannot give you the solution by using transfer functions. However I can give you a general form by using the state space representation. I will do it for a square system, i.e. the number of inputs and outputs is equal. For a system with $n$ inputs and $m$ outputs it is getting more messy and a lot harder to solve the problem. The system $$\dot{x}=f(x)+ ...


2

I think what you have is correct, provided the friction and spring constants are inverses. I am more comfortable with the force-voltage analogy and would create the equivalent one as follows.


2

You want to look for the Laplace transform of a square wave. Note: The transfer function $H(s)$ is the ratio of the Laplace transforms of output $Y(s)$ and input $U(s)$: $Y(s)=H(s)\cdot U(s)$ If you take the Laplace transform of a function (input our output), it is not called a transfer function.


2

An impulse has a flat/constant power density for all frequencies. The impulse response of a system can also be used to characterize a system, namely convolution of the input with impulse response, will yield the system response (however usually you would do this in the frequency domain, since then convolution becomes simple multiplication). The downsides of ...


2

In matlab, using c2d (link) and bode (link) functions: s = tf('s'); G_c = 2/(1+s); Ts = 1; G_d = c2d(G_c,Ts,'Tustin'); bode(G_c), hold on, bode(G_d)


2

There are many definitions of bandwidth. Typically it is determined from the open loop transfer function and not from the sensitivity or complementary sensitivity functions. Assuming a SISO feedback loop, with plant $P$ and controller $C$, a common definition of bandwidth is the 0 dB crossover frequency of the open loop transfer function $PC$. Assuming ...


2

To solve this problem I would: Sketch the block diagram of the control network. Derive the required closed loop transfer function in terms of $G(s)$ and $C(s)$. Substitute the expressions for $G(s)$ and $C(s)$ to obtain the closed loop transfer function in terms of $s$. Calculate the poles of the closed loop transfer function using your preferred method. ...


1

You can design an asymptotic output tracker based on feedback linearization if the residual dynamics are stable. The theory for this can be found in the book 'Nonlinear Control Systems' by Isidori [Springer]. You can find examples worked out using Mathematica here and here. Another way is to develop a LQR tracking controller. See, for example, Chapter 4 of ...


1

So from an engineering point of view, besides time (or computing power) why should you avoid that Coming from a mechanical engineering perspective the biggest reason is you only commit to the additional effort if it produces significantly different results. If the level of accuracy in your model is orders of magnitude higher than the level of accuracy ...


Only top voted, non community-wiki answers of a minimum length are eligible