AIPOD is a general intelligent optimization design software platform independently developed by Tianwei Software, which is committed to solving the engineering design optimization problems with better performance, lower cost, and lighter weight. In view of many problems in the field of industrial design, such as product indicators reaching bottlenecks, relying on expert experience, and difficulty in taking into account multidisciplinary and multi-objectives, the software has developed advanced intelligent optimization strategies based on artificial intelligence technology, which makes the software use threshold lower, the optimization efficiency is higher, and the optimization effect is better. Whether it's structural, fluid, thermodynamic, electromagnetic, or multiphysics coupling problems, AIPODs can help design teams efficiently find better design solutions.
1. Functions and features of AIPOD
Rich CAD CAE software interfaces.
A cutting-edge and efficient new generation of intelligent optimization algorithms.
Digital expert knowledge mining and optimization assistance.
Convenient graphical optimization process building interface.
Automated process execution engine.
Adapts to multiple types of operating systems and computing clusters.
Agile post-processing visualization exploration.
Domestic independent and controllable.
2. Introduction to version updates
AIPOD 2024R1 has been upgraded in terms of software functions and operation experience, including:
Supports hypermorph-based optimization of non-parametric geometry models.
The new CAD CAE software interface covers more comprehensive structural optimization scenarios.
The variable association function has been upgraded to make variable extraction easier.
The one-click verification function of optimization task results is added to make the optimization solution more credible.
The sampling visualization function has been upgraded to make computing power more efficient.
Other usability upgrades (multi-column sorting, re-execution of studies, task monitoring).
Figure 1 AIPOD 2024R1 startup interface of the intelligent optimization design software.
1) Support the optimization of non-parametric geometry models based on hypermorph
AIPOD 2024R1 adds support for the optimization of non-parametric geometric models. In previous versions, AIPOD was good at handling optimization tasks such as variable geometric parameters and variable boundary conditions. However, in some scenarios (such as OEMs, etc.), the optimization work cannot be carried out because the upstream ** vendor only provides non-parametric geometric models. In the new version, the integration of interfaces that support the Hypermorph function in HyperMesh can help designers easily optimize non-parametric geometry models.
2) Added CAD CAE software interface, covering more comprehensive structure optimization scenarios
AIPOD 2024R1 adds software interface integrations such as ABAQUS, MSC Nastran, and HyperMesh. At present, AIPOD's integrated CAD and CAE software interfaces have comprehensively covered multiple scenarios such as geometric modeling, structure, fluid, and electromagnetic. These software interfaces allow users to upload project files directly, helping users get rid of the complex work of macro scripting, so as to lower the threshold for optimizing design and greatly improve work efficiency.
Figure 2 AIPOD 2024R1 covers multiple scenarios such as geometric modeling, structure, fluid, and electromagnetic.
3) Upgraded the variable association function to make variable extraction easier
In the process of optimizing the process, it is a key link to realize the execution of the automation process by writing variables to the input file of CAD CAE software or extracting variables from the output file. In order to facilitate the design of engineers to realize variable association based on graphical interaction (including input variable writing and output variable extraction), AIPOD version 2024R1 has comprehensively upgraded the variable association function, which now supports users to flexibly implement it based on specifying delimiters, specifying widths, interface calls or regular expressions.
In addition, the upgraded variable correlation feature allows users to control the accuracy of design variables, as well as to extract statistical values (such as averages) from multiple output results to be associated as output variables.
Figure 3 Upgrading the variable association function.
4) Added the one-click verification function of optimization task results, making the optimization scheme more credible
In the process of actually using optimization software to carry out work, design enterprises often use means of sacrificing accuracy to improve optimization efficiency due to the limited design cycle, such as using the same type of fast software, model, ignoring complex constraints, etc. In this mode, due to the deviation of accuracy and the simplification of constraints, the optimal results obtained may not meet the actual business requirements, resulting in untrustworthy optimization results and dare not be used. In this regard, AIPOD 2024R1 adds a one-click verification function for optimization task results, which allows users to build "optimization process" and "verification process" respectively, and after the optimization process is completed, the specified solution can be automatically selected, the output result of the verification process can be invoked, and the results can be presented to users for viewing and comparison.
Figure 4 Added the one-click verification function for optimization task results.
5) Upgrade the sampling visualization function to make computing power more efficient
Sampling is often used as a precursor to optimization, and the results contain a large amount of design space information that can be used. Users can analyze the distribution of constraint values in the design space through uniform sampling results, so as to consider whether to relax the constraints in the optimization process. Users can also analyze the efficient and inefficient areas, so as to reduce the design space in the optimization process and improve the effective efficiency. Therefore, AIPOD 2024R1 adds support for users to quickly analyze the binding targets and constraints of sampling results based on parallel graphs and their interactions, and divide efficient, inefficient, feasible, and infeasible areas.
Figure 5 Upgrading the sampling visualization function.
6) Other usability upgrades (multi-column sorting, re-execution of examples, task monitoring).
New multi-column sorting: Users can sort the design list in multiple columns to weigh multiple target values and constraints.
New case re-execution: Support users to re-execute the case entries that have been executed, so as to avoid the unavailability of the study caused by calculation failure or unsaved engineering files. Cutting-edge and efficient new generation of intelligent optimization algorithm;
Improved task monitoring: The original optimization task monitoring desktop has been improved, and new information such as positive and negative change amounts, benchmark variable values, and optimal case numbers have been added.
Figure 6 Other ease-of-use upgrades (multi-column sorting, re-execution of examples, and task monitoring)
For more details about AIPOD and software trial, please search".AIPOD - Intelligent Optimization Design Platform", go to check it out.