OS-E: 0830 Multi-Model Optimization (MMO) – Pattern Repetition of an Excavator Arm

Multi-Model Optimization is demonstrated in this Excavator example using Topology optimization design variables that are linked between the two models.

The design space consists of the Excavator Arm made up of two plates of PSHELL elements for each model (Figure 1). The Excavator Buckets are different for each model.

Figure 1. Two MMO Models with Excavator Arm as the Design Space and Different Excavator Buckets

Model Files

Before you begin, copy the file(s) used in this example to your working directory.

Model Description

Model 1 consists of an excavator bucket and Model 2 is fitted with a hammer. The operational loading involves different number and types of subcases for each model. There are three linear static subcases in Model 1 and four linear static subcases in Model 2 that simulate the possible boundary conditions encountered during operation.

The Excavator arm is selected as the design space for Topology Optimization. Each arm is modeled as a pair of PSHELL element plates. The Topology Optimization setup is two DTPL entries (1 main and 1 secondary) with Pattern Repetition in each model to replicate the design from one plate to the other. One plate is selected as the Main and the second plate is the Secondary in each model.

For excavator design, if the bucket section is intended to be versatile and can incorporate multiple tool sets (bucket, hammer, and so on), then the rest of the excavator structure should be designed to operate under a variety of different boundary conditions. Specifically, in this case, focus on designing an excavator arm that is optimized for both the bucket and hammer configurations.

This simulation is not straightforward in individual model based Topology Optimization since each separate run will result in an optimized design for the arm that is suited for the corresponding specific configuration and use-case.

You can leverage Multi-Model Optimization technology to optimize parts across different models that are subject to different configurations and/or boundary conditions.
FE Model
Base and Joints
Rigid Elements
Arm, Bucket, and Boom
The linear material properties are:
Young’s Modulus
Poisson's Ratio
Initial Density
Multi-Model Optimization Setup
The optimization setup for each model is:
Optimization Type
Topology Optimization
Weighted Compliance (WEIGHT = 1.0)
Volume Fraction
Minimize Weighted Compliance
Volume Fraction (VOLFRAC is constrained at 0.3)

Multi-Model Optimization is a MPI-based parallelization method, requiring OptiStruct MPI executables and MPI installation of choice for it to run. The OptiStruct MPI executables are included, additionally MPI installations are also available in the Solvers installation (mpi folder within the Altair home directory). Existing solver decks do not need any additional input, can be easily included, and are fully compatible with the MMO mode. MMO allows greater flexibility to optimize components across structures. The -mmo run option can be used to activate Multi-Model Optimization in OptiStruct.

The first step is setting up the individual topology optimization models and testing them individually to determine if there are any errors or important warnings during runtime.

A main input deck is created that utilizes the ASSIGN,MMO entry to identify the models to be included in the Multi-Model Optimization. The Model Name can be specified on the third field of the ASSIGN entry (this name can be used to identify and qualify the responses used on the DRESPM continuation line, if applicable). In the current run, the main file is named main.fem and the multiple models are named excavation_hammer.fem and excavation_bucket.fem.
Note: In the main file, the individual models should be located in the same directory (working directory) as the main.fem file.
A minimum of one set of Design Variables or Design Domains should be linked between the multiple models identified for Multi-Model Optimization. Linking is accomplished by using the same Identification Number (ID) for the design variables or domains. This is used to internally apply pattern repetition to match the optimized results into a single model.
Note: The presence of other unlinked design variables and domains can lead to different final optimized designs for the common design space in the multiple models.

Linked design variables and domains should be identical in all respects (except for the COORD and SCALE continuation lines). The applied design parameters and manufacturing constraints should match in all models. This restriction does not apply to unlinked domains. Additionally, to facilitate internal pattern repetition, a COORD continuation line is required on the corresponding design variable or domain entry (DTPL, DSIZE, and so on). This allows identification of the location and direction of the repeated pattern across linked design domains.

The design domains are linked as:

For Model 1, the DTPL entries with ID’s 1 and 2 and for Model 2, the DTPL entries with ID’s 1 and 4 are designated MAIN and corresponding SECOND pairs. This is used for pattern repetition within each model (NOT across models) and is not related to MMO. Such pattern repetition within each model is, of course, optional and depends on the use-case. In the current model, the two plates of the design space is required to have matching patterns so it can be manufactured easily.

The linking across models, is identified by DTPL entries with ID 1 in both models. OptiStruct internally selects one of them as the main and applies internal design mapping across models.

Figure 2. Bulk Data setup

The COORD continuation line allows the identification of the anchor point and mapping of the design among multiple models. This continuation line is mandatory. If individual pattern repletion is not setup (via MAIN/SECOND) in all models, you are required to specify the COORD continuation line independent of the MAIN/SECOND continuation lines.

Multi-Model Optimization is MPI-based and the number of MPI processes (-np) is determined by the number of models. -np should be set equal to total number of models plus one. In the current run, the main file is named main.fem and the multiple models are named excavation_hammer.fem and excavation_bucket.fem.

The -mmo run option is required to identify an MMO run. Since there are two models for optimization, the run option -np should be set to 3. Intel MPI is automatically selected by default. This example is illustrated as a Windows GUI run; to run this model via the command line on Windows or Linux, refer to Multi-Model Optimization in the User Guide.


In Figure 2 the design pattern is the same for both configurations.

Figure 4. Mapped Design Domain Result for Multiple Models with Different Configurations. final designs are identical