TEST YOUR CONTROLS IN A FULLY VIRTUALIZED ENVIRONMENT
By Ian McGregor , Business Development Manager , Emulate 3D
The case for testing control systems offsite using an accurate response model is well known , and the advantages are clear . Connect the actual control hardware to a dynamic 3D model of the system or machine under test , and then introduce virtual loads to trigger virtual sensors . These sensors are connected to the control program which can then drive the model , and you have created a valid testbed for the system . This dynamic digital twin approach enables controls testing to be done off the project ’ s critical path , and carried out earlier in the project cycle than would otherwise be possible . Testing sequences can be carried out safely in an office environment , and in many cases they can be automated . This approach has been successfully employed for several decades , leading to more fully tested systems with lower commissioning , launch , and maintenance costs . Reducing uncertainty around the commissioning and ramp up of a new automated system is clearly of considerable value to all stakeholders . As modeling technology emerged which made its use convenient and accurate , the rapid spread of its use for virtual commissioning came as no surprise .
The past few years have seen a further step in development which removes a constraint from the process and enables a more flexible and persistent solution to be available to users . Most current virtual commissioning systems comprise the system or machine model , which replaces the actual system under construction and the control system , or a part of it . The model is often initially developed for discrete event simulation purposes , where several iterations of layout and equipment choices are tested for suitability before deciding on the best solution according to a set of criteria . The model resulting from this experimentation phase forms the basis of the generally more detailed controls testing model , where the operational logic is supplied by the external control system . As the control system comprises standard pieces of equipment , this is also available to the controls engineers at any point in order to carry out testing of the project ’ s critical path .
However , over the years controls testing teams have signalled two identifiable sets of conditions where it becomes problematic to carry out further virtual testing in this way .
The first of these is when the physical system is at the stage where the real control system can be installed . The controls equipment which , up until that point , was not required on site and could be conveniently used for virtual commissioning is now taken to the site to be installed within the real system . Virtual commissioning which uses hardware in the loop becomes inoperable without the hardware .
The second situation can arise at any time when the operational system needs to be modified for whatever reason – it would make sense to test the proposed changes virtually , and offline , but the controllers needed to do this are operational and cannot be taken offline without consequences and additional cost .
Fortunately we are now seeing the emergence of technology which neatly avoids both of these issues , and provides a robust and readily accessible solution . The solution is the accurate virtualisation of the controllers themselves . Users can now create control programs and upload them to a virtual controller connected to a virtual representation of the machine or system . Development and testing can now be carried out without the need for any industrial hardware in the loop , and virtual systems can remain active and useful during system commissioning , through ramp and beyond .
Users can bring up a configuration of models and virtual controllers mirroring any real system in order to implement and then test changes in the digital twin before intervening in the real system . By testing exhaustively in the virtual system , intervention time to implement changes in the real system
is minimised , reducing the impact and cost of changes on production .
Automated system end users , machine builders , and systems integrators alike benefit from the virtualization of material handling equipment and process machines , the control system , and many of the component parts . This approach shortens the development cycle and its associated costs , yet makes it easier to build , test , and operate robust automated systems . The approach also prolongs the useful life of the system as it reduces the risks and costs of system modifications , making their implementation a more straightforward decision .
8 The Satellite Review www . WestfaliaUSA . com