Artificial intelligence in SIMATIC
The future of automation
Artificial intelligence (AI) is set to play a key role in reducing the programming and engineering effort required to create automation solutions. It’s also making control logic more agile and production processes more flexible and precise – especially in industry. For example, machine-learning algorithms help systems that perform visual quality checks in production plants or image-guided robot systems react much more flexibly to unexpected situations and to quality defects because they can respond automatically during runtime. As a result, they operate much more efficiently, because expert knowledge – for example, regarding the color, consistency, or quality of a product or process – can be transmitted to automation.
Artificial intelligence is creating new opportunitiesAI makes it possible to perform automation tasks that push traditional solutions to their limits. It’s gradually establishing itself on every level of the Totally Integrated Automation portfolio – from the field to the cloud.
AI and Totally Integrated Automation
The new SIMATIC S7-1500 TM NPU module
With the new S7-1500 TM NPU (neural processing unit) module for the SIMATIC S7-1500 controller and the ET 200MP I/O system, AI is finding its way into automation with Totally Integrated Automation. The module can be seamlessly integrated into the SIMATIC automation system, this way permitting a simple and profitable combination of AI algorithms and PLC logic. This is how inexpensive applications using SIMATIC can be implemented at the machine based on artificial intelligence. The S7-1500 TM NPU module is equipped with the AI-capable Myriad™X Vision processing unit chip from Intel Movidius™ which, among other things, enables the efficient processing of neural networks.
The TM NPU can be installed both centrally behind an S7-1500 CPU and peripherally behind an ET 200MP interface module. By installing multiple modules in succession, the solution can also be scaled as needed regarding the quantity structure and the number of implementable applications.
The S7-1500 TM NPU module operates using a trained neural system on an SD card. Users can connect Gigabit Ethernet- and USB 3.1-compatible sensors like cameras and microphones to the module’s integrated interfaces. CPU data transmitted by the backplane bus can also be used as input data. The processing results then are evaluated in the CPU program.
How you can benefit from AI with the SIMATIC S7-1500 TM NPU
Discover impressive solution approachesThanks to AI, the SIMATIC S7-1500 TM NPU module enables countless new approaches and innovative strategies in automation. What once seemed complex is suddenly surprisingly easy, and what used to be impossible is now possible. Let yourself be inspired.
Smart object recognition of any workpieces
The SIMATIC S7-1500 TM NPU module uses the capabilities of neural networks to recognize complex patterns. Input data is processed with the help of knowledge acquired through machine learning. This means that even unknown components and workpieces can be recognized in any location or position.
This reduces engineering time by eliminating the time-consuming process of learning the individual objects and also increases productivity by preventing waste.
The SIMATIC S7-1500 TM NPU module can be integrated directly into the PLC. This makes smart object recognition extremely economical, because no separate setup is required.
Visual quality check of products
The processing of image data in the SIMATIC S7-1500 TM NPU module’s neural network enables reliable visual detection of quality defects. Using trained algorithms, image data is compared with predefined, extremely complex quality characteristics. This eliminates unexpected errors in the ongoing process and ensures constant product quality.
The QA station creates a bottleneck in many production processes, because checking product quality is often time-consuming.
In addition, checking the quality of more complex products has often needed to be performed manually. Using the SIMATIC S7-1500 TM NPU module – which can be integrated directly into the PLC – operators can now check the quality of ever-more complex products automatically, which also means more quickly, more cost-effectively, and more reliably.
Quality checking using sensor and production data
In many production sequences, the overview and evaluation of production data, such as energy, temperature, pressure, and surface, provide a reliable basis for the early detection of quality defects.
When production data recorded by sensors is transferred from the CPU to an S7-1500 TM NPU module integrated directly into the PLC, the module can use its neural network to analyze and evaluate complex patterns in the data and, if appropriate, cause the relevant product to be rejected.
This not only speeds up the quality assurance process, it also improves its reliability.