Create a Rich 2D & 3D Beam Solving System

[sai udayagiri]

Hello,

I am Sai Udayagiri, a physics graduate from India with a strong interest in computational mechanics and symbolic mathematics. I have been exploring SymPy for over a month now, focusing particularly on the Physics module. While going through the GSOC ideas page, the projects "Implement Specific Forces and Torques" and "Create a Rich 2D Beam Solving System" caught my attention.

To evaluate my suitability for the project, I thoroughly explored the continuum mechanics module for the past 10 days, reviewed the past contributions by Sampad, Jashan, Ishan, Prakhar, Advait, and Ishan Pandhare, and studied NPTEL lectures on beams to deepen my theoretical understanding. Although I had a subject on Solid Mechanics during my college coursework, I found that the work in this module extends beyond my prior knowledge. Hence, I invested time in researching both the theory behind beam mechanics and SymPy's implementation of 2D and 3D beam modules.

Observations & Ideas for Improvement

After analyzing previous work and Current Project Requirements, I observed several areas that could be improved or expanded upon:

Enhanced Plotting Capabilities

• Currently, all plots in the beams module are very simple and lack customization.
• I believe we can enhance default plotting features and allow users to fully customize plots .

Support for More Complex Beam Structures

• The module currently does not allow beams with different material properties at joints.
• I am unsure about the complexity and importance of this feature, so I would love to hear thoughts from the community.

Support for Complex Cross-Sections

• The module does not fully support non-standard cross-sections.
• Integrating with the geometry package could allow defining arbitrary cross-sections for 2D and 3D beams.

Expanded Problem Examples & Documentation

• Adding more real-world problem examples would help users better understand the module.
• Examples showcasing more analytical solutions would be beneficial.

My Technical Experience

Python : I have experience in symbolic computation, data structures, and numerical methods.
Git & Open Source :I am familiar with Git, GitHub workflows, and open-source contribution practices.

I would love to hear the views of the community and potential mentors Advait, Prakhar regarding these ideas. If these are not the most desirable changes, I would appreciate guidance on what improvements are needed for the SymPy Beams Module.

Looking forward to your response!

Thanks & Best Regards,
Sai Udayagiri

[Jason Moore]

The most recent work on that module is: https://github.com/sympy/sympy/pull/27130 which we would like to merge and get it working more generally. This expands the capabilities to more structure types. That would be my hope for any near future work.

Fixing bugs seen in this list of issues is also priority: https://github.com/sympy/sympy/issues?q=is%3Aissue%20state%3Aopen%20label%3Aphysics.continuum_mechanics

Jason

moorepants.info
+01 530-601-9791

--
You received this message because you are subscribed to the Google Groups "sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sympy+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sympy/683d2c01-63ff-4b65-b101-499e8eca0375n%40googlegroups.com.

[sai udayagiri]

Hi Jason,

I appreciate your response and the priorities you shared. Over the past four days, I have been thoroughly exploring the structure2d.py and Beam.py modules. I found structure2d.py exciting, readable, and easy to work with. As part of this process, I solved 80+ problems using different combinations of available functions within the module, covering most of them to understand their capabilities and limitations better. This hands-on experience has given me deeper insight into its functionality and highlighted areas where improvements or additions could be made.

While the module is well-structured and logically sound, I noticed several areas that could be refined:

Key Observations

• Function Behavior Inconsistencies: Some functions lack basic validation checks. For instance, when applying a distributed load without specifying end values, the draw function assumes default values, whereas the apply_load function fails to add it as a distributed load. Additionally, many methods allow users to enter inputs in any order without proper validation, leading to incorrect calculations or failures, especially in complex systems with multiple members.
• Missing Functionalities: Some methods are only partially implemented. For example, the apply_load function currently supports only point and distributed loads, but not ramp or parabolic loads. Additionally, several key methods available in Beam.py are missing in structure2d.py. Adding these would significantly enhance the module's ability to handle more advanced structural problems.
• Lack of Documentation & Examples: Neither the .py file nor the module documentation contains structured examples or detailed comments, making it difficult for users to fully understand and utilize its capabilities.
• Limited Test Cases: The current test suite does not comprehensively cover all cases and functionalities. Expanding test coverage would improve reliability and prevent potential regressions.
• Plotting Limitations: The module lacks proper visualization enhancements. Plots miss key elements such as units, labels, and load values, and they do not always provide a perfectly analyzed square-type representation. Additionally, the default graph often lacks the proper axes which , makes it not neat as the structures overlaps with the top borders.

Proposed Contributions

• Refining the Existing Implementation – Addressing inconsistencies and improving method robustness.
• Expanding Functionality – Implementing missing features and adding key methods from Beam.py to enhance versatility.
• Comprehensive Documentation & Examples – Creating structured documentation and usage examples to improve accessibility.
• Strengthening Test Coverage – Developing a more extensive test suite to validate correctness and reliability.
• Enhancing Plotting & Visualization – Improving graphical outputs by ensuring proper labels, units, load values, and clearer structural representations.

I would love to hear your thoughts and suggestions on these observations. If there are additional aspects you'd like me to focus on, please let me know—I’d be happy to explore them further.

If these align with your expectations, I’ll begin drafting my GSoC proposal and start working on a PR to submit alongside it. I have also compiled a list of specific issues I encountered while solving problems,

• When order=0 and end_x, end_y are not provided, the draw function assigns default end values, but apply_load does not, causing inconsistencies.
• If an invalid order is given, the function fails without a proper error message.
• The apply_load function lacks support for ramp loads and parabolic loads.
• Some loads do not appear correctly in the graphical representation.
• Distributed loads applied at (2.5,5) do not get plotted.
• Certain loads are missing entirely from the visualization.
• If supports are defined before loads, reaction calculations fail.
• Graph defaults to (0, -y) when no supports or loads exist, making visualization unclear.
• Graph lacks auto-adjustment for better readability.
• No proper validation checks when applying loads at points not on the beam.
• Lack of error messages when end_x, end_y values are missing but required.
• Order 5 is incorrectly considered a distributed load without strict checks.

• Graph lacks essential features like units, scale, grids, and borders for analysis.

Looking forward to your feedback!

Best regards,
Udayagiri Saibabu

[sai udayagiri]

Hi Jason,

I wanted to follow up on my previous email regarding my observations and proposed Enhancements for the structure2d.py module. I’d appreciate any feedback on whether the proposed scope aligns with expectations.

If it’s too limited, I’m happy to add more stages and compress my timeline in the GSoC proposal draft. If some of the proposed changes aren’t desirable, I’d love to explore other areas where contributions would be more impactful.

Looking forward to your feedback.

Best regards,
 Sai Udayagiri

[Jason Moore]

HI Sai,

My feedback is very positive. If you really solved 80 problems with the module, then you should have a strong idea of what to improve.

Jason

moorepants.info
+01 530-601-9791

To view this discussion visit https://groups.google.com/d/msgid/sympy/7a14b8ae-459a-4dd4-b611-b1e38ee19c07n%40googlegroups.com.

[sai udayagiri]

Hi Jason,

I have drafted my GSoC proposal for enhancing the Structure2d module and would love to get your feedback, suggestions, and review. Before finalizing the project length, I want to ensure it aligns with the community’s expectations.

  I am happy to refine my proposal based on your suggestions and would greatly appreciate any insights. Additionally, it is clearly mentioned in the registration guidelines that the project length should be selected only after receiving confirmation from the community, so it would be helpful if you could suggest the appropriate length for my proposal. 

Proposal Link: GSoC Proposal sai Udayagiri

Looking forward to your feedback!

Best regards,
Udayagiri Saibabu