Definitions of key terms in FEM, CFD, thermal analysis, control engineering and numerical methods — each linking to a related free tool.
While CAE encompasses many methods, they all rely on solving the governing equations of physics numerically. For structural analysis via Finite Element Method (FEM), the core equation is derived from the principle of virtual work, often represented in a discretized matrix form:
$$[K]\{u\} = \{F\}$$Here, $[K]$ is the global stiffness matrix (built from individual elements), $\{u\}$ is the vector of nodal displacements (what we solve for), and $\{F\}$ is the vector of applied forces. This equation is the mathematical heart of predicting how a structure deforms under load.
For fluid flow analysis via Computational Fluid Dynamics (CFD), the foundational equations are the Navier-Stokes equations. They describe the conservation of momentum for a viscous fluid. The incompressible form (constant density) is:
$$\rho \left( \frac{\partial \mathbf{v}}{\partial t} + \mathbf{v} \cdot \nabla \mathbf{v} \right) = -\nabla p + \mu \nabla^2 \mathbf{v} + \mathbf{f}$$Where $\rho$ is density, $\mathbf{v}$ is the velocity vector, $t$ is time, $p$ is pressure, $\mu$ is the dynamic viscosity, and $\mathbf{f}$ represents body forces (like gravity). Solving this equation for complex geometries is what allows the simulation of airflow over a wing or water in a pipe.
Aerospace & Automotive Design: CAE is indispensable for designing safer, more efficient vehicles. Engineers use CFD to optimize the aerodynamics of a car body to reduce drag and improve fuel economy. Simultaneously, FEA is used to perform crash simulations, ensuring the passenger cabin remains intact and safe during an impact, all before a single prototype is crashed.
Consumer Electronics: When designing a smartphone, engineers use thermal simulation (a type of CAE) to model how heat from the processor spreads through the device. This helps place components and design cooling solutions to prevent the phone from overheating during use, ensuring both performance and user comfort.
Civil & Structural Engineering: Before construction begins on a skyscraper or a long-span bridge, the entire structure is modeled and analyzed using FEA. Engineers simulate extreme loads from wind, earthquakes, and traffic to identify stress concentrations and validate that the design can withstand these forces throughout its lifetime.
Biomedical Engineering: CAE enables the development of custom medical implants, like a knee replacement. Surgeons and engineers can use patient-specific CT scan data to create a 3D model of the bone, then use FEA to simulate the stresses on the implant during walking, ensuring it is strong enough and will integrate properly with the bone.
When you start learning CAE, there are a few common pitfalls you might encounter. The first one is the belief that a finer mesh is always better. While it's true that finer meshes theoretically improve accuracy, computation time increases explosively, proportional to the number of elements. For example, simply doubling the number of divisions along one edge of a cube mesh increases the element count eightfold. In practice, it's crucial to identify the "minimum mesh that meets the required accuracy." If you experiment by changing the mesh coarseness in your tool, you can experience "convergence"—the point beyond which further refinement yields little change in the results.
The second pitfall is errors in setting boundary conditions. This is the leading cause of inaccurate results. For instance, if you model the fixed end of a cantilever beam as a "pinned support" instead of "fully fixed," it will allow rotation at the base, leading to completely different deflection calculations. The golden rule of simulation is "garbage in, garbage out" (GIGO). How you simplify (model) reality is where your skill truly matters.
The third point is to avoid trusting raw numerical results too blindly. Even if a stress value reads 345 MPa, failing to visualize where on the part and in which direction it occurs (using contour plots or vector displays) can lead to critical oversights. Judging as "OK" based on numbers alone is risky. Always ask yourself, "Does this result make physical sense?"
The FEM and CFD technologies covered in this terminology dictionary are connected to a wider range of fields than you might think. For example, in biomedical engineering, FEM is used to analyze stress on artificial joints (implants), and CFD simulates blood flow (hemodynamics) within blood vessels. Learning a term like "non-Newtonian fluid" here can lead you into analyzing fluids like blood, whose viscosity changes with shear rate.
Another connection is with materials science. FEM calculations require material property values like "Young's modulus" or "yield point." Conversely, in new material development, this connects to the concept of "materials informatics," where CAE is used to input virtual material properties and predict product performance. For thermal analysis, predicting heat generation and thermal runaway in battery development is an interdisciplinary field combining electrochemistry and thermofluids.
Furthermore, don't overlook acoustical engineering. Vibration of a structure (calculable by FEM) vibrates the air to create sound. Designing automotive interior noise (NAH) or speakers relies on coupled simulations of structural and acoustic analysis (wave equations). As you can see, principles learned with one tool often come to life in completely different application areas.
Once you've gotten a "feel" for things with the terminology dictionary and simulator, the next recommended step is to gradually solidify your theoretical backbone. A fundamental building block is linear algebra. Handling the stiffness matrix $[K]$, central to FEM (solving systems of linear equations, eigenvalue problems), is entirely built upon linear algebra. Start from the level of understanding that "a matrix is an organized collection of coefficients from a system of equations"—just having that mental image is a good start.
Next, begin by not "solving" but "observing" the governing equations for each physical phenomenon. For example, the basis of heat conduction is Fourier's law and the heat conduction equation $$\frac{\partial T}{\partial t} = \alpha \nabla^2 T$$. When you look at this equation, it's important to develop a physical intuition: "The rate of change in time ($\partial T/\partial t$) is proportional to the sharpness of the temperature distribution curve ($\nabla^2 T$)," meaning "areas where temperature is convex will flatten out over time." Think of it as practice in translating equations into plain language.
A learning path with lower risk of frustration is: 1. Pick up keywords from the terminology dictionary and experiment with the simulator → 2. Look up the simple equation describing that phenomenon and connect it to a physical image → 3. Learn how that equation is discretized and solved by computers (basics of FEM or FVM). For your next topics, tackling nonlinear analysis (material plasticity, large deformations) or coupled analysis (structure-fluid-thermal interaction) will let you touch upon the real depth of CAE practice.