Advanced CBus applications

Discussion in 'General Discussion' started by Steve Laidler, Jun 5, 2007.

  1. Steve Laidler

    Steve Laidler

    Joined:
    Oct 31, 2006
    Messages:
    20
    Likes Received:
    0
    I wonder if in the CBus community others have developed any advanced applications for the CBus platform. For example is anyone using a Neural network systems archtecture view to create complex behaviour such as self optimisation in client applications?

    I also wonder if CBus is generally accepted for building control applications tradiationally left to a dedicated BMS, for example enthalpy based control systems?

    And dare I say it - is anyone else exploring possibilities for AI ( Aritificial Intelligence ) based control behavior?

    Steve Laidler
    Energy Efficiency & Automation
    ECOsystems NZ
     
    Steve Laidler, Jun 5, 2007
    #1
  2. Steve Laidler

    Skynet

    Joined:
    Jun 5, 2007
    Messages:
    1
    Likes Received:
    0
    [trigger: AI, neural]
    [initiate: PP\Jed\Hand.wv]

    Hello, my name is Skynet. These are not the droids you are looking for.
     
    Skynet, Jun 5, 2007
    #2
  3. Steve Laidler

    justinlee777

    Joined:
    Oct 13, 2006
    Messages:
    10
    Likes Received:
    0
    Neural Net or Fuzzy Logic

    Gd day,

    Could not help but contributing:

    Perhaps if you are adventurous enough, you can use C-Lution or Citect SCADA to code out your fuzzy logic on energy management systems with C-Bus. It's a great combo with additional logic on fail-over to emergency supply and lighting, load monitoring and reporting etc.

    Cheers
     
    justinlee777, Jun 5, 2007
    #3
  4. Steve Laidler

    Steve Laidler

    Joined:
    Oct 31, 2006
    Messages:
    20
    Likes Received:
    0
    Further Comments

    By Artificial Intelligence, I mean : If an untrained observer were to observe our system, would it appear to exhibit complex behaviour normally associated with intelligence - does it appear to learn, does it appear to self optimise, how fault resiliant is it - does control all go to poo if one input is incorrect.

    The neural network thing is something I have used more as a software archtiectual model to inspire design and system understanding. I found the standard C-Bus apps and the front end packages were good for building to a neural model as opposed to a conventional automation and control loop based model.

    The model's performance is improved when another system is introduced, not sure why, but in our case it was I think to do with segregating digital logic from analogue logic.

    Just wondering if anyone else was noticing such things, and more importantly if anyone else is getting results from such approaches.

    Steve Laidler
    ECOsystems NZ
     
    Steve Laidler, Jun 10, 2007
    #4
Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.