How do you verify the results?

Question

Accepted Answer

I verify the results by comparing them with experimental data and performing sensitivity analysis.

Compare results with experimental data
Perform sensitivity analysis
Use statistical methods to validate results
Collaborate with experimental team to ensure accuracy
Verify against industry standards and best practices

Accepted Answer

Verifying the results of a mechanical Computer-Aided Engineering (CAE) analysis is a critical step to ensure the accuracy and reliability of your simulation. Verification involves comparing the simulation results with known solutions or experimental data. Here's a step-by-step guide on how to verify the results in mechanical CAE:

Establish a Baseline:
- Before conducting any analysis, establish a baseline by defining the expected outcomes or results. This can be done by:
  - Referring to analytical solutions, if available, for simple problems with known solutions.
  - Identifying specific performance criteria or target values for the analysis.
Select Benchmark Problems:
- Choose benchmark problems or test cases for which analytical solutions or experimental data are available. These problems should be representative of the type of analysis you are conducting.
Conduct the Analysis:
- Perform the CAE analysis using your chosen software and input data.
Compare Results:
- Compare the simulation results with the expected outcomes or benchmark data. The comparison should involve key variables such as stress, displacement, strain, or any other relevant parameters.
Quantitative Comparison:
- Quantitatively assess the agreement between the simulation results and the expected outcomes or benchmark data. Use mathematical metrics, such as error calculations or correlation coefficients, to quantify the level of agreement.
Visual Inspection:
- Visually inspect the results through plots, graphs, and animations to identify any discrepancies or anomalies. Visual inspection can reveal patterns or trends that quantitative analysis might miss.
Sensitivity Analysis:
- Perform sensitivity analyses by varying input parameters within a reasonable range and observing how the results change. This helps identify the impact of uncertainties in inputs on the simulation outcomes.
Mesh Convergence:
- Check the convergence of your finite element mesh. Ensure that the results do not significantly change with mesh refinement, indicating that the solution is not highly sensitive to mesh size.
Code Verification:
- For in-house or customized CAE software, perform code verification by comparing results from different versions of the software or by comparing with known analytical solutions.
Error Sources Identification:
- Identify potential sources of errors in the analysis, such as incorrect material properties, boundary conditions, or modeling assumptions. Address these errors as necessary.
Documentation:
- Document your verification process, including the expected outcomes, benchmark problems, quantitative results, and any deviations or discrepancies. This documentation is essential for transparency and future reference.
Peer Review:
- Involve colleagues or experts in CAE analysis to review your verification process and results. External perspectives can help identify errors or uncertainties.
Iterative Improvement:
- If discrepancies are found, iterate on the analysis by refining the model, adjusting inputs, or making necessary corrections until the results align with expectations or benchmarks.
Validation:
- Validation is a separate process that involves comparing CAE results with physical experiments for real-world systems. While verification compares with known solutions or benchmarks, validation confirms that the CAE model accurately represents the physical behavior of the system.
Quality Assurance:
- Implement quality assurance processes and adhere to established standards and best practices in CAE analysis within your organization.

By following these steps, you can systematically verify the results of your mechanical CAE analysis and ensure the accuracy and reliability of your simulations. Verification is a critical part of the analysis process, as it builds confidence in the modeling and simulation capabilities of your CAE software and methodology.

How do you verify the results?

Popular interview questions of Senior CAE Engineer