Monte Carlo Analysis is a statistical analysis that calculates the response of a circuit when the component parameters vary randomly according to a given statistical distribution (e.g. normal (Gaussian) distribution, uniform distribution, etc.).
For example, Monte Carlo Analysis can be used to evaluate the impact of the variation of key components on the output of a power converter. Component variation can be a result of manufacturing tolerance or degradation of the component over time.
Monte Carlo Analysis can be used to help engineers decide what tolerance key components should have.
Sensitivity Analysis studies how the outputs of a circuit vary when one or several of its components change. It is often used to determine the most sensitive key components in a circuit.
For example, the figure on the right shows the Sensitivity Analysis report of a buck converter. With a 5% variation on both the inductance and capacitance, the inductor current ripple is more sensitive to the inductance.
Fault Analysis studies the performance of a circuit under fault conditions such as a short-circuit or open-circuit of a component.
Fault Analysis helps evaluate if the design can pass various fault tests and if a specific fault can cause catastrophic failure.
A typical workflow of the analysis tools would be to use Sensitivity Analysis to first identify which key components are most sensitive to output performance. Then carry out a Monte Carlo Analysis on these components to determine the proper tolerance levels. Finally, perform Fault Analysis to evaluate the reliability of the design.
To learn more on how these analysis tools are used in a practical example, refer to the Application Note “Monte Carlo, Sensitivity, and Fault Analysis of Resonant LLC Converter (AN009).pdf” in the “Application Notes” folder in PSIM.