I am interested in how many of you guys are directly involved in developing The neccessary modelling and control including software and hardware for various systems.
The questions I am interested in answering are related to:
The staffing structure?
Is there say a modelling team, a hardware team, a software team and an implementation team?
How does the effect of so many (often competing) departments effect the quality of the product and the time it takes for completion?
Is there a Panacea for all of the ills??
Are the modelling, coding and downloading to target hardware (micro or/and DSP) encapsulated in one system?
Once the target hardware is written to - how are changes to model structure and model variables implemented (Is this on-line in Real - Time?)
What are the Industry Standard Softwares used in development?
Varies widely, even within industries. Some of your later questions imply embedded control, which I am much more qualified to talk about than process control.
That depends on the size of the project. Usually if modeling is done at all it is by an individual, unless you're talking about a _big_ project like a jetliner or a power grid. Many control loops are controlled by folks who are flying by the seat of their pants; in that case modeling never really happens in any formal way.
There's often a separate hardware and software team. I'm not sure what you mean by "implementation team" -- generally if hardware and software are not done by one and the same individual (often the same guy who doesn't know enough to do the modeling) the teams will come together to integrate their separate efforts into the whole.
If a design effort is going to go at all smoothly it's because its organized by project team. Even if there are separate 'hardware', 'software', etc., departments the individuals will still be responsible to a project manager for their work delivered.
Of course you can have stupid competition that undermines this -- just remember that Dilbert is not fantasy, it is a reflection of real life that's been toned down to be believable.
Yes, but that sort of genetic engineering isn't currently possible, and will probably be highly illegal if it ever is technically possible -- and who's going to sign off on the functional spec?
Varies widely. That depends. Often from one project to another within a small company.
In many different ways. Usually by the time the target hardware and software is done any control loops are pretty well established; changes are more in the way of tweaking and tuning than wholesale model changes.
Probably MatLab, although I recommend SciLab 'cause it's free and has more potential.
I am including process control as well in the sense that there may be an element of statistical control of say a batch product that can be changed on-line where neccesary e.g an actuator is moved. SCADA systems involving Client/Server and Radio System Comms.
Condition Monitoring Systems and system ID also - if anyone has any contact with this area
I'm thinking perhaps in the development process The proved hardware/software design might be say, commissioned by another department or company.
generally if hardware and software
What would you use and for what projects.?
What do you use for FPGA development environment??
particular look at the Process Automation methodology PDF This covers all aspects of the life cycle from requirements definition through implementation to validation and I think answers many of your questions
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.