Method and system for automatic detection of inefficient household thermal insulation

A system for automatic detection of inefficient household thermal insulation within a group of monitored households, comprising a server and a plurality of household clients, wherein:

• each of the server and the plurality of household clients comprises one or more processors, operatively coupled to non-transitory computer readable storage devices, on which are stored modules of instruction code, wherein execution of the instruction code by the one or more processors implements a function of the server and the plurality of household clients;
• the plurality of household clients are configured to acquire at least part of:
• indoor temperature, outdoor temperature, HVAC compressor activation time, HVAC thermostat convenience temperature settings, HVAC mode of operation, household profile parameters and household profile resident parameters;
• the plurality of household clients are configured to convey the acquired data to the server;
• the server further comprises a preprocessor, configured to identify periods of HVAC down-time, and determine the indoor temperature gained during the identified periods of HVAC down-time;
• the preprocessor is further configured to extract parameters of temperature gained, relating to a measured rate of the indoor temperature gained during the HVAC down-time;
• the server further comprises at least one trainer, configured to train a machine learning algorithm, to create at least one classification model, wherein the monitored households are classified according to the acquired data and the parameters of temperature gained;
• the at least one trainer is further configured to use the at least one classification model to produce expected values for the parameters of temperature gained per each household, according to a household class membership;
• the server further comprises a data analysis module, configured to:
• produce a ratio between the expected values for the parameters of the temperature gained and measured values of the parameters of the temperature gained per each monitored household;
• to compare the ratio per the each monitored household with ratios per peers, wherein the peers are houses of similar profile parameters and/or residents profile parameters;
• and
• to determine, according to the comparison, whether a specific household is suspected to be inefficiently insulated.