Method and system for optimizing a composite load disaggregation

A system for optimizing a composite load disaggregation in an electrical environment, the system comprising:

• at least one processor;
• and
• a memory coupled to the at least one processor for executing a plurality of modules present in the memory, the plurality of modules comprising
• an input module configured to receive one or more input variables;
• a factor-graph module configured to perform a factor-graph analysis on the one or more input variables to generate confidence measures for one or more appliances in the electrical environment, wherein the confidence measures indicate at least one appliance from the one or more appliances being switched on at an instance, wherein the factor-graph analysis is implemented based on matrix formulation of blind source separation or based on load block diagram;
• a contextual information database configured to store contextual information impacting the usage of the one or more appliances;
• a rule engine, configured to optimize the composite load disaggregation by applying one or more rules to the confidence measures and to the contextual information for identifying at least one appliance from the one or more appliances and wherein conflicts in decision of identifying least one appliance from the one or more appliances is resolved based on the output of the optimized composite load disaggregation through the factor-graph analysis;
• a priori database, configured to store a set of priori data;
• a rule database, configured to store one or more rules for optimizing the composite load disaggregation result.

A non-transitory computer readable medium having embodied thereon a computer program for optimizing a composite load disaggregation in an electrical environment, the computer program product comprising:

• a program code for performing a factor-graph analysis on one or more input variables to generate confidence measures for one or more appliances in the electrical environment, wherein the confidence measures indicates at least one appliance from the one or more appliances being switched on at an instance, wherein the factor-graph analysis is implemented based on matrix formulation of blind source separation or based on load block diagram;
• a program code for retrieving contextual information impacting the usage of the one or more appliances from a contextual information database;
• and
• a program code for optimizing the composite load disaggregation by applying one or more rules to the confidence measures and to the contextual information for identifying at least one appliance from the one or more appliances and wherein conflicts in decision of identifying least one appliance from the one or more appliances is resolved based on the output of the optimized composite load disaggregation through the factor-graph analysis.