There's more than one way to ascertain what loads are drawing what power in your house. Non-intrusive load monitoring could do the job without necessitating the installation of a smart meter. In so doing, NILM might be a way of addressing concerns of those who dislike the idea of installing equipment that give the utility an ability to switch household loads on and off.
The idea behind NILM is to first characterize the way big loads in your house typically behave, then just monitor the main power meter going to the house. An embedded computer equipped with learning algorithms and other logic watches the power consumption at the main meter and makes educated guesses about what loads are acting at any given time.
Of course, NILM can't actively get involved in turning loads on and off the way a smart meter would. But the technique could be a way of performing on-going energy audits in a way that would be more or less painless for the homeowner.
The technique was recently described by researchers at Carnegie Mellon University. Writing in Yale's Journal of Industrial Ecology, Dr. Mario Berges explained, “This form of non-intrusive load monitoring may be able to provide a new type of continuous electrical audit for residential buildings, down to the appliance level. While costs can only be estimated at this point, it is possible that the price of such a system could be similar to that of the whole-house meters currently available on the market, approximately $200 per residence.”
NILM systems generally require a training period during which the different appliances in the home are switched through the different modes of operation to populate a library of power-use signatures. The researchers explored the idea of incorporating these steps into the typical one-time visit of residential energy auditors.
Recognizing the energy signatures of different appliances isn't easy, Says Berges. Several loads in a home are likely to add significant noise to the signal (e.g., by continuously varying the power levels) which obscures smaller transitions, and some variable loads may also be present. In addition, with a larger number of appliances—including those that cycle independently, like the refrigerator—it is more likely that there could be overlapping events. However, initial trials demonstrated that the system was capable of correctly classifying most of the loads in the building, with higher accuracy for larger appliances.
Researchers measured the electric current on both legs of the main electric supply with split-core current transformers. To compute complex power, the voltage was also measured with a DAQ. The DAQ was a National Instruments PCI-6143 16-Bit unit for which researchers wrote a custom LabVIEW program to drive the sampling operations and to compute complex power and other metrics for every three full periods of the signal.
You can read the whole report here: http://onlinelibrary.wiley.com/doi/10.1111/j.1530-9290.2010.00280.x/full