01. June 2016

Modular Industry Computing Architecture for smart factories

The new Modular Industry Computing Architecture (MICA) is designed to provide existing machinery and systems with intelligence, making it possible to transform manufacturing plants into smart factories and bringing to life the potential of integrated industry. As a result, customers can design their production facilities to be more modular, affordable and less complex. Customers can also implement their Industry 4.0 projects – regardless of size or complexity – quickly and cost effectively. MICA makes it possible to temporarily save, evaluate and process data directly within the machinery and equipment. With its modular open platform, the HARTING IIC MICA can be customised as required, including the hardware, software and interfaces, making it compatible with the world of Integrated Industry.

MICA functions as a small, robust computing unit between physical devices and the higher-level IT system. It provides a cost-effective solution that allows users to run computing power at all levels as well as communicating with central IT systems and the cloud, offering fast data analysis and decision-making at the field level along with storage and data-collection capability to provide relief for computer centres.

Small but powerful

MICA is extremely compact, with dimensions of 13 × 8 × 3.5 cm, small enough to fit onto an electrical cabinet DIN rail. It can therefore be installed discretely and directly onto a machine to save on real estate. The front panel can be customised with M12 and USB connectors to optimise network connectivity (Fig.1). The system is housed in an IP 67 enclosure that is designed to withstand vibration and extreme temperature fluctuations in harsh industrial applications. MICA can be used in metallic environments and is resistant to dust and moisture. As there is no fan, it is maintenance free and can be operated remotely through a web browser.

Customisable and compatible

MICA’s open source approach allows users to completely customise the computer: for example, by choosing the programming language and development environment. Unlike conventional single-board computers, the MICA circuit board is divided into three parts – one of which is customisable to the user’s application. The form factor and degree of protection remain unchanged after hardware adjustments.

Software applications run in virtual, Linux-based containers which hold all the necessary libraries and drivers. As a result, incompatibility and dependency on existing packages are no longer an issue.

The MICA computer features a powerful 1 GHz ARM processor, 1 Gbyte of RAM and 4 Gbyte of eMMC flash memory (plus an additional 32 Gbyte on a micro-SD card). HARTING’s virtual industry computing technology enables multiple programs to run in parallel within the virtual containers. These containers run in individual “sandboxes” which isolate and secure different applications from each other. This enables smart, multi-tasking to occur for productive operations.

The touch-optimised interface is implemented entirely in HTML5 and JavaScript. Designed for both end-users and administrators, the interface is easy to use, and can be customised by developers without any special requirements.

Economic benefits

The one-time investment for MICA is significantly lower compared to complete industrial PCs. Moreover, using MICA involves no licensing or leasing fees, and offers inexpensive prototyping and development because of its open development environment, even for multiple projects. Further savings in operating costs are provided by the very low power consumption: less than 5 W for the base version.

Applications

Potential applications for MICA within the industrial automation environment include:

Retrofitting for predictive maintenance: One of the most effective ways to reduce production equipment downtime and maintenance costs is by the on-time recognition of maintenance needs. For example, within an injection moulding machine application tools can be marked with RFID (Fig.2). The current consumed during the injections is then measured by an inductive current sensor, and the results are combined, saved and processed in the MICA. As a result, a warning is sent whenever the power consumption changes, since this indicates a problem. Extending the replacement intervals enables costs to be reduced. After several months, a neural network can then be trained with the acquired data and installed on the MICA. This enables improved predictions to be made about working life spans. In a further step, the MICA can be connected to the ERP system in order to optimise the processes for ordering spare parts.

Optimising processes at packing stations: Traditionally, customer orders are packed and prepared for shipment at a packing station. Within this process, the employee is required to operate a scale, the packing machine’s PLC, the ERP system and a label printer, as well as manually looking up instructions. However, by using MICA it is possible to combine all these processes and execute the steps automatically, thereby relieving the strain on employees, reducing the chance of errors, and significantly improving efficiency in the packaging process.


Comments (0)