Technical Computing

Technical Computing Description Topics

• The Power of Technical Computing with MATLAB
• Data Analysis and Exploration
• Mathematical Algorithms to Build On
• Modeling and Simulation
• Visualization and Image Processing
• Programming and Application Development
• Distributed and Parallel Programming

Modeling and Simulation

MATLAB^® offers a unique simulation and prototyping environment. The powerful technical language is both concise and descriptive, allowing you to model complex systems with small sections of easy-to-follow code. MATLAB and companion products offer an array of tools for simulation and modeling, including:

• MATLAB's core foundation of high-performance functions for mathematical computation, data analysis, equation solving, and visualization
• State-of-the-art algorithms via add-on toolboxes
• Multiple simulation approaches using MATLAB, as well as Simulink^® and Stateflow^®
• block-diagram graphical design environments

MATLAB's Simulation and Modeling Foundation

MATLAB provides hundreds of functions for simulation and modeling, from basic matrix operations to ordinary differential equation solvers. These functions provide a solid foundation from which to simulate and model complex systems.

To visualize simulation results as they are calculated or for post-processing, built-in animation functions and fast graphics allow you to view model behavior for analysis, testing and debugging, and presentation.

Since MATLAB is an interpreted language, you can modify your models to see the effects immediately, without the additional overhead of recompilation, as in C. The math algorithms are already developed for you, so the code required to build a model in MATLAB is significantly shorter than the corresponding C or C++ code. This compactness makes your MATLAB code easy to write and maintain over time.

MATLAB is used to visualize the chaotic motion in fluid with both surface and contour plots.

Application-Specific Toolbox Algorithms

Application-specific toolboxes, such as Curve Fitting, Optimization, Control System, Neural Network, and Signal Processing, allow you to quickly build simulations and models for applications across a range of disciplines, with limited coding from scratch. Other potential applications include:

• Using the Statistics Toolbox and its random number generator to simulate behavior associated with particular distributions. This random data can then be used to test hypotheses or models under different conditions.
• Applying the System Identification Toolbox to help determine models, which can be used to predict behavior in times series data, such as exchange rates, or to analyze resonant modes in vibration systems.
• Performing image restoration using morphology and other simulation techniques using the Image Processing Toolbox.
• Using the Partial Differential Equation Toolbox for the study and solution of PDEs with the Finite Element Method. Mathematical models can be developed for problems in electrostatics, AC power electromagnetics, heat transfer, and diffusion.

In this example MATLAB is used to visualize a numeric model of Hurricane Floyd. The cloud-like surface depicts 95% relative humidity, the bottom image represents relative precipitation, and the coneplot indicates fluid flow. Using the Audio Video Interleaved (AVI) movie file format, MATLAB can simulate the behavior of the hurricane over time. (Data courtesy of NOAA / GFDL.)

Multiple Simulation Approaches

While MATLAB offers a familiar programming environment, Simulink and Stateflow provide a graphical design environment for modeling and simulating complex control, DSP, and supervisory logic systems. Built on MATLAB, these products can call any MATLAB function including user-written routines, allowing you to combine the best of both approaches. Functions from toolboxes can also be incorporated into Simulink block-diagram models.