Good data management democratizes powerful CAE technologies bysimplifying them so that they are available to a wider group of companies.
By Michael Engelman and Scott Del Porte
Nearly every enterprise realizes that it needs to bring more new, better-working products to market faster to gain an advantage over the competition. One way to reachthe goal of increased market share and higher profit margins takes its cue from the automotive and aerospace industries. Enterprises can bring CAE and its array of powerful geometry definition and engineering analysis tools to bear on the design process to improve product performance. And while CAE tools are becoming increasingly important to design and manufacturing as well as more critical to the process, limitations on how CAE tools are used have also become apparent.
Even with recent advances in user interfaces, graphics, computing power, and automated features, complex engineering analysis still requires high levels of skill to pull off. Usually, experts in different physical disciplines are needed to perform each of the various analysis steps since the data generated in these steps is difficult to manage and analyze. Additionally, handing off analysis results from one step to the next is complex and time-consuming.
For example, users of CFD (computational fluid dynamics) must know how to apply proper mesh densities, turbulence models, and boundary conditions to models to achieve meaningful results. Users must also know how to go about translating MCAD geometry into the proper format and to correctly interpret output information. Certain skill levels are also required to fully utilize other complex CAE technologies such as nonlinear FEA (finite element analysis), multibody simulation, acoustics simulation, and so on.
The net result is that these technologies are typically used by large companies that have the requisite resources to support an organization of highly educated and experienced specialists. Even among such companies, however, the lack of integration between these specialties and the large amount of effort required to set up each analysis has largely relegated the use of CFD and other high-end CAE tools to relatively late stages in the product development cycle.
SPDM Helps Apply Analysis Early and Often
A new methodology, Simulation Process and Data Management—or SPDM—aims to simplify the use of these powerful CAE technologies so that they are available to a wider group of companies for evaluating more designs more thoroughly at a much earlier stage in the design process.
Figure 1: SPDM makes high-level CAE available to a broader range of engineering and manufacturing companies by integrating design processes, simulation data, and task- and system-level concerns. Click on image to enlarge.
SPDM takes the next logical step by integrating the processes and data involved in engineering design. The SPDM approach can deliver substantial improvements in product performance and reduce product development lead times and costs.SPDM includes both the process and the data generated by simulation, and the solutions range from full design systems to vertical applications. The solutions can be thought of as addressing system-level or task-level issues. Figure 1 (right) illustrates both system- and task-level SPDM applications. Note that any given solution will be concerned with process or with data, or with some combination of the two.
For example, the CAE portal is a system-level solution that manages the process and data for a particular CAE code. Data vaults are system-level solutions that are primarily used to store and manage simulation data. This data can be archived for later use or it can be mined to look for new trends or new information.
Typically, an interface, often Web-based, is developed specific to the manufacturer’s requirements to simplify the analysis process. It does that by prompting the user for critical parameters and then provides results in a format that makes it easy to evaluate the design. For products that require several different types of engineering analysis, engineering data is moved behind the scenes and results are stored in a common repository that makes them easy to reuse. This opens the door to widespread use of CAE as a decision-support tool throughout the entire design process.
Figure 2: This sequence of pictures depicts the heat sink tool, from process (left) to analysis (middle) to geometry (right). Click on images to enlarge.
From the beginning of the design process, the engineering team can consider all possible designs in an effort to create a truly innovative product. Each alternative is thoroughly evaluated with the use of CAE tools in order to compare its performance to others without incurring the costs of detailed design, prototyping, or physical testing. Every design option is evaluated based on a wide range of criteria, and the number of design possibilities is trimmed based on considerations ranging from performance to cost.
SPDM solutions can consist either of full design systems that integrate the entire product development process or of custom interfaces that streamline one or more analysis steps involved in the design process. Data mining techniques can be incorporated to allow both the detailed analysis of multiple simulation results and comparison with experimental results; and SPDM solutions naturally integrate with broader PLM (product lifecycle management) systems already in place.
SPDM in action
AAVID Thermal Technologies of Concord, NH, distinguishes itself by developing custom heat sinks suited to a particular set of tough thermal-management requirements. It had to overcome three obstacles: fierce competition from overseas commodity manufacturers; design expertise that resided in the heads of just a few engineers (each of whom would likely come up with different designs, resulting in inconsistencies and unnecessary rework); and protecting intellectual property as the final products were manufactured offshore. By overcoming these obstacles it would develop a design system that draws on the expertise of the company’s leading engineers and previous design histories and that houses the system within a secure Web-accessible environment.
The Web portal approach has the advantage of offering a consistent methodology that reuses existing designs whenever possible and brings several different CAE tools to bear to optimize the design beyond a level possible in the past. The company’s sales force is now able to create simpler new designs while its top engineers are far more productive overseeing the process and working with CAE tools, all while the company’s core intellectual property is being captured. It leads to a secure system, a design database that permits data mining and prevents reinventing the wheel, an embedded design process that assures consistency and use of best practices, and integrated high-end CAE analyses that can be performed by nonexperts.
SPDM—The User’s Perspective
The user begins by logging into the system and entering customer requirements such as the integrated circuit’s physical dimensions, heat dissipation, and available airflow within the cabinet (see Figure 2, above). The system then queries the design database and identifies previous designs that would fit the requirements of the new application. It also determines an optimal design based purely on the design rules, searching for a parametric model and setting parameters to the values determined by the rules.
After choosing an existing design or the new optimized design, the user analyzes it in more detail. The user can run FEA and CFD programs (geometry, meshing, and boundary conditions are automatically generated). It then solves the model and passes the critical results—e.g., temperatures at each point on the heat sink—to an FEA-based stress analysis program that solves for mechanical stress on the heat sink.
In the past, engineers rarely had time to perform detailed CAE analysis of their designs and had to rely on their experience to generate a design that they felt confident would work. By largely automating the analysis process, the system greatly reduces the cost and lead time of analysis, making it possible to use CAE technology more intensively.
At any point in the process, the user can modify parameters and come up with a new design and, for each new design, the system automatically estimates the finished cost based on material stock, manufacturing, and fixture databases. Once the user settles on a final configuration, the design can be sent to a CAD system where a finished analysis report is generated for the sales engineer to develop a quote for the customer. Even if the design is not used to make a part, the final design configuration is entered in the design database and can be called up for another case with similar input requirements.
This approach essentially applies data management principles to simulation processes, input data, and results to capture and protect design information. It automates repetitive steps, promotes design reuse, and creates a valuable audit trail of the steps used to create new designs. Further, the SPDM approach allows all groups within a company to interact with design data, contributing and using information as needed, without risking the loss of intellectual property.
Michael Engelman is the CEO at Enductive Solutions and Corporate Vice President at Fluent Inc. Scott Del Porte has more than 20 years of engineering and business development experience.