The company that "we" worked for at the time ... no longer exists (recent event).
Its IP was bought up by another company.. I don't know any more than that.
I have worked on less ambitious versions with other companies since then.
At present I don't know if there is a similar product being marketed to the general public. I don't think the company (when it still existed) was marketing to the general public.. it was marketing to other companies.
- Mems devices and DSPs for vibration monitoring have become very cheap.
- 32 bit processors with enough I/0 and memory for a really smart AC for less than $10
- Current transformers, temperature sensors, humidity sensors have become very cheap.
What isn't quite so cheap.. variable speed drives and control valves appropriate in size and robustness.
The complete solution has to be very very cheap to get any "traction" in the market place.
People (until very recently) expect ROI times of less than 1 year if you want to get any real market penetration.
The time is fast approaching when all the pieces will be in place.
except... (maybe?) a company that has the market position, resources and the "will" to put it all together.
Thinking j.... if you make such a product now, please tell me where I can learn more about it. I have searched in vain for a commercially available smart heat pump system.
Current energy and maintainance costs are high enough for this to make sense now if it is not too costly. Preeptive maintanance alone would pay for such a system in a short time.
My $50 blender is smart... why do I have to wait for my heat pump system to fail before I know about it? and have to replace a blown compressor that apparently had been overheating due to water flow issues for months???
It sounds like "thinking-J" was part of a really great product that would be in much wider use if power prices had jumped sooner. And he is certainly correct about building automation not being about turning on the lights, (or selecting the music). The real value comes from minimizing climate control system power consumption. Once we have the models for some of the basic parts, and the understanding of how they work togather, the design task becomes an excercise in fitting parts into a building that have the correct capability. Just a little thing like the fan speed /run time difference in power useage is certainly showing how to reduce power consumption without giving up performance. So I would remark to Beth that what is needed are models of the elements that make up these systems, rather than general models. So it is a specialized subset, not an additional superset, that would do the job best.
One more comment is that diagnostics would probably not be so very "mechatronic" in nature, since most problems develop as a shift away from "the way it was initially", which can be determined with much simpler models and algorithms. So while diagnostics are certainly very valuable, they are also quite different.
Using a new term - mechatronics - really doesn't change anything
As to using technology to reduce power consumption.
Been there, done that.
In 1995 ... (in anticipation of electricity de-regulation).
We built a building automation system (6"x8"x2"h, basically a PC with numerous I/O - ethernet, several RS-485 ports, RS-232 ports, Dallas sensor buses, analog and digital inputs/outputs).. ran on customized Linux...24vac...
With it we ran the building lighting/ AC (with variable speed compressors and fans) / security and simulated the building (using DOE's building simulation software originally written in fortran - hey! it was free, and it was good)..
Simulation of the entire building was running in the background - ALL the time.
On the small businesses this was focused on addressing.. we could project 24hrs in advance the power requirement within 1-2% accuracy. It reduced peak power requirements by 50%, and average power requirements by 30%. Truly information for a "smart grid" (power company - with entire communities needs known in advance).
Also allowed detailed evaluation of the maintenance of the building/system.
If a filter needed replacement.. we knew it before it created problems.
If a compressor was having problems... we knew it before it failed.
The system knew : typical occupancy for any day of the week, the local weather (via the web), room usage, performance of the AC system...
Then the pressure to address deregulation was removed (98)..killed further advancement.. though the product (in newer incarnations) stills lives.
It is amazing how much electricity is saved with just a smarter AC system (with variable speed compressors/fans and truely smart controls.
Primary reason: compressability of air.
example: a fan running at full speed with a 50% duty cycle (say a cycle of 5 minutes) will consume twice the power of same fan running at 50% speed 100% duty cycle. (power consumption is cubed as speed increases by 2)
To keep the rest of the system optimal under varying conditions (Humidity, heat load, etc...) requires much more complex controls (to keep the coils from freezing up, etc...)
Pay back period for additional technology being installed? less than 2 years (in 95).. likely 6 months today.
Best you can buy for your home today .. two speed / two compressor based system for (a great improvement old systems.. but could be soo much better)...
Real home automation? Isn't about remote controls of the lights.
Real automation? it about the building maintaining it self with minimal power.
Where energy is concerned, a perfect storm is approaching: less coal and nuclear-based power; more renewables; and more electric cars. Widespread modeling like the kind shown here is going to be critical, if we expect to continue our standard of living. Imagine being able combine this kind of building modeling with broad use of LEDs. The energy savings would be gigantic.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.