MATLAB and Simulink R2022b offer new Simscape Battery and updates

MATLAB and Simulink R2022b Deliver New Simscape Battery and Updates to Simplify and Automate Model-Based Design

Platform release also includes new Medical Imaging Toolbox, providing an end-to-end medical image analysis workflow

Beijing, China, September 20, 2022 - MathWorks today announced the release of MATLAB® and Simulink® product family version 2022b (R2022b). R2022b introduces two new products and several enhancements to simplify and automate model-based design, helping engineers and researchers achieve product innovation and breakthroughs for their organizations.

The global battery management system market is expected to reach $13.4 billion in 2026. Bloomberg New Energy Finance says this growth is mainly due to the development of the electric vehicle (EV) market. The organization's latest report shows that by 2040, 58% of global passenger car sales will come from electric vehicles. Simscape Battery™ is one of the major innovations introduced in the R2022b version, which provides design tools and parameterized models for companies designing such battery systems.

Engineers and researchers can use Simscape Battery to create digital twins, run virtual tests of battery pack architectures, design battery management systems, and evaluate battery system behavior under normal and fault conditions. The tool can also automatically create a simulation model that matches the desired battery pack topology and includes cold plate connections to evaluate electrical and thermal responses.

Caption: Design and simulate battery and energy storage systems using Simscape Battery.

“We are excited to launch Simscape Battery at a time when innovation in battery management systems is at an all-time high,” said Graham Dudgeon, principal product manager for electrical systems modeling at MathWorks. “This product includes many design tools designed to simplify and automate model-based design, including Battery Pack Model Builder, a tool that enables engineers to interactively evaluate different battery pack architectures.”

R2022b also provides the new Medical Imaging Toolbox. This toolbox provides tools for designing, testing, and deploying diagnostic and radiomics algorithms that use deep learning networks for medical imaging applications. Medical researchers, scientists, engineers, and device designers can use Medical Imaging Toolbox for multi-volume 3D visualization, multimodal registration, segmentation, and automatic ground truth annotation to train deep learning networks based on medical images.

Figure caption: Medical Imaging Toolbox supports interactive data annotation, semi-automatic or fully-automatic medical imaging data annotation, and exports the annotation results for use in medical imaging AI development.

Figure caption: Medical Imaging Toolbox provides classical or deep learning algorithms to segment 2D images or 3D objects into different regions, such as bones, tumors, or other organs, and evaluate the accuracy of regional segmentation.

“Medical engineers and researchers will benefit greatly from the power of the 3D Annotation app and algorithms for the complete medical image analysis workflow, including I/O, preprocessing, training, and analysis,” said Bruce Tannenbaum, technical product marketing manager at MathWorks. “We are excited to support the full deep learning workflow to automatically find objects of interest in images to segment tissue and detect disease.”

R2022b rolls out updates to the following popular MATLAB and Simulink tools, including:

- AUTOSAR Blockset: Develop service-oriented applications using the client-server ARA approach and deploy them on embedded Linux platforms. The tool supports data types and interfaces in user-defined architecture models.

- Fuzzy Logic Toolbox: Interactively design, analyze, and simulate fuzzy inference systems (FIS) using the updated Fuzzy Logic Designer. In addition, this enhanced toolbox supports engineers and researchers to design Type 2 FIS using command-line functions or Fuzzy Logic Designer.

- HDL Coder: Generate optimized SystemC code from MATLAB for high-level synthesis (HLS) and use frame-to-sample conversion for model and code optimization.

- Model Predictive Control Toolbox: Use neural networks as prediction models for nonlinear model predictive controllers. In addition, the toolbox now supports users to implement predictive controllers compliant with ISO 26262 and MISRA C standards.

- System Identification Toolbox: Create deep learning-based nonlinear state-space models using neural ordinary differential equations (ODEs). Machine learning and deep learning methods can also be used to represent the nonlinear dynamics in nonlinear ARX and Hammerstein-Wiener models.