Der Blätterkatalog benötigt Javascript.
Bitte aktivieren Sie Javascript in Ihren Browser-Einstellungen.
The Blätterkatalog requires Javascript.
Please activate Javascript in your browser settings.
4 Messe Aktuell Mittwoch 13 NoveMber SPS 2024 HALLE 7 STAND 450 Mehr über ctrlX OS erfahren ctrlxos de ONE COMMON UNIVERSE COUNTLESS SOLUTIONS ctrlX OS steht für eine neue Ära der Automatisierung Das industrielle Betriebssystem mit Ökosystem fördert Innovation und Zusammenarbeit indem es die Stärken von Partnern aus unterschiedlichsten Industrien zusammenbringt So entstehen für Anwendende maßgeschneiderte und zukunftssichere Automatisierungslösungen Die Linuxbasierte Technologie von ctrlX OS bietet Flexibilität Sicherheit und Skalierbarkeit für alle Anforderungen Advancing Machine Learning and Predictive Maintenance Andrea Gillhuber Matlab and Simulink are instrumental in developing and deploying Machine Learning ML models offering robust tools for feature engineering training and parameter optimization Dr Rainer Mümmler Principle Application Engineer at Mathworks explains the details in an interview What role do Matlab and Simulink play in training and implementing machine learning models? Matlab and Simulink play a crucial role developing and implementing machine learning models Matlab provides a powerful environment for creating features and training machine learning models Its extensive libraries and tools enable it to train models and optimize parameters efficiently These models can then be integrated into the Simulink Matlab environment to perform detailed simulations Simulink supports synthetic data generation particularly when only a small amount of data is available Specific errors can also be induced to test the robustness of the models This synthetic data helps to better train the models and improve their performance In addition Matlab and Simulink can provide trained models in production systems or on edge devices This enables companies to implement their machine learning solutions efficiently and reliably in real applications How can PdM models seamlessly inte grate into an automation environment to ensure realtime monitoring? There are several approaches for seamlessly integrating PdM predictive maintenance models into an automation environment and ensuring realtime monitoring and rapid fault diagnosis Simulink models including PdM models can be implemented on edge devices using code generation or Functional Mockup Units These digital twins support precise and continuous monitoring of system states in real time Classification and anomaly detection algorithms can also run on programmable logic controllers through automatic code generation This enables fast and efficient fault diagnosis directly in the automation environment What support do you offer for deploying machine learning in edge environments? Mathworks provides comprehensive support for deploying machine learning in edge environments especially when running models on devices with limited computing power and storage capacity Key features and tools include optimized code generation and model compression Matlab and Simulink enable the generation of optimized C C++ and CUDA code that can be deployed on CPUs GPUs and FPGAs These optimized codes ensure that machine learning models run efficiently on edge devices even when computing resources are limited Mathworks offers tools for hyperparameter optimization quantization and network pruning to reduce the size of AI models This makes the models suitable for resourcelimited devices without losing significant accuracy As the ML models are executed directly on the edge device they do not rely on network connectivity This creates new use cases for AI especially in remote or otherwise hardtoaccess locations and enables fast and reliable onsite decisionmaking Dr Rainer Mümmler in Hall 6 Booth 215 Im ag e Uw e Nikl as Pele m ed ia