Engineers use MATLAB®, Simulink®, and Predictive Maintenance Toolbox™ to develop and deploy condition monitoring and predictive maintenance software to enterprise IT and OT systems.
- Access streaming and archived data using built-in interfaces to cloud storage, relational and nonrelational databases, and protocols such as REST, MQTT, and OPC UA.
- Preprocess data and extract features to monitor equipment health using apps for signal processing and statistical techniques.
- Develop machine learning models to isolate root cause of failures and predict time-to-failure and remaining useful life (RUL).
- Deploy algorithms and models to your choice of in-operation systems such as embedded systems, edge devices, and the cloud by automatically generating C/C++, Python, HDL, PLC, GPU , .NET, or Java® based software components.
Using MATLAB and Simulink for Predictive Maintenance
Access Data Wherever It Lives
Data from equipment can be structured or unstructured, and reside in multiple sources such as local files, the cloud (e.g., AWS® S3, Azure® Blob), databases, and data historians. Wherever your data is, you can get to it with MATLAB. When you don’t have enough failure data, you can generate it from a Simulink model of your machine equipment by injecting signal faults, and modeling system failure dynamics.
Clean and Explore Your Data to Simplify It
Data is messy. With MATLAB, you can preprocess it, reduce its dimensionality, and engineer features.
- Align data that is sampled at different rates, and account for missing values and outliers.
- Remove noise, filter data, and analyze transient or changing signals using advanced signal processing techniques.
- Simplify datasets and reduce overfitting of predictive models using statistical and dynamic methods for feature extraction and selection.
Detect and Predict Faults Using Machine Learning
Identify root cause of failures and predict time-to-failure using classification, regression, and time-series modeling techniques.
- Interactively explore and select the most important variables for estimating RUL or classifying failure modes.
- Train, compare and validate multiple predictive models with built-in functions.
- Calculate and visualize confidence intervals to quantify uncertainty in predictions.
Deploy Algorithms in Production Systems
Shorten response times, transmit less data, and make results immediately available to operators on the shop floor by implementing your MATLAB algorithms on embedded devices and in enterprise IT/OT systems.
- Reference Architectures for AWS and Azure Cloud Integration with MATLAB
- Code Generation for Embedded Systems
- Application Deployment to the Web
- Internet of Things with MATLAB and Simulink
- From Acorn to Oak: Seeding Federated Learning with Physical Models (32:29)
- Deploy Trained Predictive Models to Target Systems