| 摘要 |
PROBLEM TO BE SOLVED: To easily and accurately provide power consumption and the like of individual electric appliances such as variable-load household electric appliances, i.e., household electric appliances whose power consumption (current) varies in accordance with modes, settings or the like.SOLUTION: A state estimation part uses total sum data regarding total current consumed in a plurality of electric appliances to perform state estimation that obtains state probabilities of respective factor states in a Factorial Hidden Markov Model (FHMM). A model learning part uses the state probabilities to perform learning of the FHMM where each factor has three or more states. This technology can be applied, e.g., in order to obtain power consumption of variable-load household electric appliances. |
