Digital simulation used to be the preserve of the seriously well heeled — whether you were talking about complex CNC process simulation or factory simulation — but several developments have converged to bring affordable sophisticated factory simulation to SMEs.
Like most recent developments everything seems to hinge around Microsoft, and the software giant recently announced a significant robotics initiative. While this attempts to introduce a common standard to robotics, it has ramifications for factory simulation for layouts, assembly cells or lines, and robot systems.
The announcement included a link to the AGEIA PhysX physics accelerator. To date, this has been mainly used in the gaming market, making 'game play' simulation more sophisticated. It has modules for Rigid Body Physics, which makes collisions look realistic, as well as for fluid and cloth simulation and particle effects for explosions and smoke.
has been working independently of Microsoft for the past few months putting the accelerator to use in the industrial field. AMTRI is recognised by AGEIA, the developer of the PhysX Accelerator and SDK (Software Development Kit), as an official developer.
It has already developed interfaces between industrial simulation software and the PhysX SDK. The company has demonstration material to show what can be done with PhysX in an industrial context, and it is ready to exploit the potential of the latest version of the PhysX SDK, 2.4, and the newly-released PhysX accelerator itself.
AMTRI believes that the accelerator is complementary to industrial simulation. Through a combination of industrial simulation technology, the PhysX SDK and a range of consultancy and integration services, the company's aim is to help SME manufacturers improve competitive edge, productivity and profitability. A key part of this endeavour is product and process simulation.
What this all means is that soon, SMEs will have low-cost access to software tools comparable to those marketed by the major CAD and PLM vendors.
Visualisation of complex systems has been based on Finite Element methods or graphic video game approximations. PhysVis — the first 'demonstration' product from AMTRI combining these techniques — utilises novel games software and hardware tools coupled with FE material modelling to visualise dynamic interactions between system elements.
Aerospace composite lay-up and spraying can be simulated by the inclusion of physics within a low-cost simulation tool and material mapping from micro through meso to macro scale. This provides an affordable, high-performance tool kit with game-like animation and visualisation, but accurate representations of material and manufacturing processes.
PhysVis addresses the challenges posed by the handling and visualisation of large data sets associated with complex systems. It uses a low-cost, highly-graphic virtual environment with innovative software and hardware solutions based around a proven simulation tool, 3Dcreate (from Visual Components), and leading FE techniques coupled with the physics acceleration system, PhysX.
By using 3DCreate — which was developed specifically for designing production systems and assessing risk — novel automated processing applications in the aerospace and spraying sectors can serve as graphic exemplars of the visualisation mapping being applied to dissimilar scientific techniques such as FE and fluid computations.
3DCreate also supports remote visualisation and shared immersive reality through its standard GUI, so the possibility for operator training on machines or processes is available as a free by-product. PhysVis therefore primarily addresses material and machine dynamics, visualisation and simulation.
The PhysX Flag demo is the first step in the research project to produce a 3D simulation of an automated carbon fibre lay-up system. The demo consists of a Kawasaki FS45 robot moving a flag. It shows the flag moving due to the physical forces and gravity. The physics calculations are performed using the PhysX Accelerator, and the simulation is displayed using 3Dcreate.
The recent announcement of Microsoft's robotics initiative will doubtless speed up investment in this market sector. Thanks to progress made by 'low-cost' simulation vendors such as Visual Components and others, coupled with AGEIA's migration from computer gaming into the industrial sector, developments in 'lightweight' portable CAD data and interoperability standards, and you have an interesting combination — especially for SMEs.
Similarly, CNC process simulation is moving apace — as is shown by the introduction of Vericut's Version 6. Vericut has been around since the early days of CAD/CAM in the 1980s. It became established as thede facto
market leader because it would always simulate the 'tape' file (the post-processed CNC output) independently rather than the toolpath file.
This has been developed into what Vericut developer CGTech describes as CNC program verification, not merely toolpath verification/simulation. This is because the software simulates the entire CNC machining process.
Most CAM systems have some sort of toolpath simulation, but they do not provide full process simulation. toolpath simulation is a bit like a professional body investigating itself. You can only have confidence in a simulation if it truly models what the machine tool sees (the tape file). This is essential if 'fresh air' prove-out is to be eliminated.
Additionally, most CAM systems do not have a full kinematic model of the machine tool which does not simulate tool changes, indexing moves and other non-cutting actions that can cause machine tool collisions.
As if to confirm this, Vericut is increasingly used interfaced with other CAD/CAM/PLM systems, such as Catia, UGS, PTC, EdgeCAM and MasterCAM, even though each of these systems has some form of CNC simulation included in the CAM offerings.
Vericut is also only independent simulation system that contains a full kinematic model of the machine tool and control system. This virtual machining environment is essential to fully eliminate prove-outs and eliminate CNC machine tool collisions.
So as simulation gets cheaper, easier and more rigorous, The 'modernise or die' mantra can be substituted by 'simulate or die'.