June 2019

n TECHNOLOGY June 2019 www.drivesncontrols.com 20 ROCKWELL AUTOMATION has announced an add-on module for its ControlLogix controllers that uses artificial intelligence (AI) to detect production anomalies and alert workers so they can investigate or intervene. The FactoryTalk Analytics LogixAI module fits into a controller chassis and streams data over the backplane to build predictive models. It can monitor production operations continuously, detecting any anomalies. The module – known earlier in its development as Project Sherlock – reduces the burden of existing analytics technologies that require deep expertise of both data science and industrial processes. “The FactoryTalk Analytics LogixAI module makes predictive analytics more accessible to help more workers make better production decisions,” explains product manager, Jonathan Wise. “The module learns your ControlLogix application and tells operators and technicians when things are changing in unexpected ways. This can help them get ahead of product quality issues and protect process integrity.” For example, the module can help operators to spot performance deviations in equipment such as mixers that could affect product quality or lead to downtime. It can also be used as a virtual sensor. Instead of workers taking a reading – such as the humidity of a food product – the module can analyse variables from line assets such as sprayers, dryers and burners to predict the value, virtually. Workers can be notified of problems by alarms on an HMI or dashboard. In the future, the module will help them to focus their problem-solving activities, or to automate process optimisation. The module is the latest addition to Rockwell’s FactoryTalk Analytics portfolio which includes its FactoryTalk Analytics for Devices, which learns about an automation system’s structure and tells users about problems with individual devices. The new module expands on this by learning about an application and helping to detect anomalies in the way it is operating. The products can be used individually but, in future iterations, will benefit each other. The FactoryTalk Analytics platform aggregates multiple sources of data, allowing users to discover new insights. FactoryTalk Analytics for Devices and the LogixAI module will both be data sources for the platform in the future. www.rockwellautomation.co.uk AI module plugs into controls to spot possible production problems Rockwell’s AI module helps industrial workers to predict production issues and improve processes without needing to learn new skills Sick is using deep learning techniques to create “intelligent” sensors that can perform automated detection, testing and classification of objects and features. At the Hannover Fair, it announced an application that uses deep learning to detect whether a sorting tray in a logistics hub is loaded with an object. To achieve this deep learning, Sick is harnessing neural networks. Compared to the classical process for developing algorithms, which is characterised by manual development of feature representation, a neural network is trained to optimise its task and can be retrained with new data to adapt to new circumstances. To train the networks, Sick is collecting and assessing thousands of images and examples. The new deep learning algorithms it generates will be implemented on sensors such as smart cameras, making them failsafe and directly available. By implementing deep learning in certain types of sensors, Sick says it is taking its AppSpace platform to the next level. It envisages artificial intelligence being added eventually to devices such as inductive proximity sensors, photoelectric retro- reflective sensors, ultrasonic sensors and others. Deep learning will allow smart sensors to adapt to their task One application of Sick’s deep learning technology is to detect whether sorting trays in logistics hubs are actually loaded with only one object.