Login to the System engineering wiki (password protected).
The SKA radio telescope is not only physically large, but also complex and comprised of millions of different parts. The designers of these parts need to know how they will be used and how they will fit together. This is where System Engineering comes in – it is a formal way to ensure that the hardware and software is fit for purpose and is value for money.
SKA Development
Whilst working towards building the SKA, we ‘build’ many SKAs. These are ‘paper’ SKAs, whose behaviour as a system can be analysed, and changes made at low cost. These models can be used for many purposes such as performance assessment, maintenance needs analysis and most importantly, cost. Trade-off studies are carried out using multiple models so that we can see what drives the cost and whether we must make compromises in performance. The SKA is so large and complex that we must study detailed choices because small cost/performance differences in millions of parts have a large effect on the behaviour of the system as a whole.
How do we approach SKA System Engineering?
The SKA is a global project and technology development is going on around the world.
First we consider the scientific objectives of the SKA. These are of paramount importance as they drive what the SKA must do and how it will perform. These are called the science requirements.
Next we take into account the fact that the SKA must be built in the real world, be operated by humans, use existing or projected technology, exist within a legal framework and respect the environment.
These considerations give rise to further requirements and constraints which are taken into account when designing the SKA. We also use experience gained from existing large science facilities to provide additional operational requirements.
The topmost system design architecture is then proposed and alternatives are analysed. The best architecture is broken down and the various elements go through the same outline design and analysis process. This is repeated down to the point where simple parts can be made or procured.
Finally, all chosen designs are made or bought, and assembling them begins. At each level of assembly, testing is carried out so that problems are found as early as possible. For SKA, this is an enormous process which takes several years but leaves nothing to chance. Time and time again, it has been demonstrated that this disciplined approach gets the best final performance for money.
Also in this section
