October 16, 2018
pSeven 6.14 Release
DATADVANCE development team announces the availability of pSeven 6.14, a new version of our flagship platform for data analysis and optimization. In this major update, we’ve come up with design space exploration and predictive modeling enhancements, extended integration capabilities and connected interactivity between some viewers. We also deployed a few usability updates and introduced a new licensing format to make our platform more powerful and versatile than ever. Read on for highlights on these exciting new additions.
Design Space Exploration Enhancements
This release refines Design Space Exploration capabilities that empower engineers to explore various design alternatives and easily find optimal solutions. We upgraded the Design of experiments block to Design space exploration by adding the majority of functions of the existing Optimizer block. The resulting new block has the following key features:
- The entire setup of optimization and design of experiments (including adaptive) techniques are under the hood. You can easily combine strategies and switch between techniques when solving your design tasks.
- SmartSelection is activated by a button and automatically selects the best approach based on the properties of your design task.
- Four new techniques implemented: Surrogate-based optimization, Gradient-based optimization, Adaptive design for GP and Diagonal sampling.
- Added support for discrete and categorical variables.
- Added support for adaptive generation of designs based only on the initial sample (sample-based ADoE).
- Redesigned interface with improved responsiveness and usability.
Predictive Modeling Enhancements
The latest update further enhances your experience when building and evaluating approximation models. Some notable changes are outlined below:
- We significantly improved Model builder performance in the SmartSelection mode, namely:
- Implemented a new method of model training parallelization which utilizes multi-core CPUs in a much more efficient way, training candidate models in parallel. This feature noticeably speeds up model training in case of multiple outputs or in the presence of categorical variables. It is also supported in the manual training mode and in the ApproxBuilder block, where it can be controlled using the specific option.
- Added an ability to build more accurate models in cases when values of some training outputs are exponentially distributed, thanks to the better support for automatic log transformation of training data.
- Improved the special data analysis algorithm to prevent oscillations of model outputs when a smooth model is required. As a result, the best approximation technique is selected according to the expected landscape of the model.
- Introduced a new Advanced option hint to manually specify arbitrary options used in model training.
- Better support of model’s input and output descriptions in blocks and various tools in Analyze. In general, it means that now you can add extensive details on variables and responses before training a model. For example, specify real names, brief descriptions, physical quantities, and measurement units of inputs and outputs using the respective input ports of the ApproxBuilder block. This information will be saved to the model and can be easily retrieved when you view the model in the Analyze tab or load it into the Approximation model block. The details are also kept when you export the model code, for example, as a Functional Mock-up Unit for Co-Simulation. We also added the ability to copy model details to the pSeven clipboard.
Integration Block Updates
In this version, we also focused on extending direct integration capabilities of the platform. The topical issues here include:
- Added the STAR-CCM+ block which enables integration of STAR-CCM+ simulations into a workflow, supports StarCAD, CFD and Turbo Wizard parameters in STAR-CCM+ and can run simulations on a remote STAR-CCM+ server.
- Introduced the FMI model block to integrate Functional Mock-up Units for Co-Simulation into workflows.
- FloEFD, Solidworks and ANSYSWorkbench blocks now use the latest versions of their products.
With this release, we take a major step toward making the platform an even more powerful tool for visualizing your data. Linked selection option is a long-awaited feature that adds connected interactivity between Sample viewer, 2D plot and Parallel coordinates plot. You can now synchronize selections between these viewers if they share at least one common data source and also have linked selection enabled. It means that selections you make in a viewer also select corresponding data in other viewers linked to it. This handy feature offers almost endless opportunities to explore your data.
pSeven 6.14 comes with some usability improvements aiming to delight our customers. Now you can:
- Navigate to a specific row in the Sample viewer (the command is triggered using the Ctrl + G shortcut) and enjoy the accelerated viewer performance when working with large datasets.
- Set name prefixes to data series when importing data from CSV or Excel files.
- Select multiple items in the model tree of Approximation model and CAD blocks. Selected items are automatically shown on the Variables or Properties panes.
- Add screenshots from the simulation on 2D plots to visually assess the design points.
- Quickly view the results of your testing operations — values of variables read from the template and a preview of changes applied by the Text block to the template.
- Use a new automatic layout for blocks and links in your workflows.
- Maximize all block configuration dialogs.
Command-Line Workflow Integration
The update provides a practical and easy way to integrate pSeven as a part of some external process. Now you can start workflows from the command line using pSeven’s batch mode (p7batch). Its notable features are:
- Workflows are started from a command line; there is no need to open a project in pSeven or click any GUI buttons.
- Running a workflow in batch mode does not show pSeven’s graphical interface so that it can be used even on systems without a graphical environment.
- pSeven running in batch mode could use a pSeven Runner license instead of full license.
This version introduces a new license format that provides a simple and flexible way to deploy and use pSeven for solving various engineering problems.
Moving to pSeven 6.14 requires an updated license file. Licenses issued for the previous versions of pSeven are not compatible with pSeven 6.14 and above.
New licenses files will be sent shortly to customers with active technical maintenance period. Please, contact us if you didn't receive it.
This release also brings many updates to the chapters of the user manual and resolves a number of issues and bugs to improve overall customer experience with the product. For a detailed list of updates, please, see the release changelog.