Abstract: The invention relates to the estimation of a time-variable physical quantity by a distributed system comprising a centralized platform and a set of distributed measurement devices each communicating with the platform via a communication network, on the basis of the measurement (100), by each device of the set, of a local datum representative of the physical quantity, the collection (120) by the centralized platform, over a predetermined sampling period, of measurement samples originating from a subset of devices of said set via said communication network, and the estimation (130) of the physical quantity over the sampling period on the basis of the collected samples. According to the invention, each device of said set executes (110), in each sampling period, a local algorithm deciding randomly and based on a sending probability associated with each device, whether the corresponding measurement device is part of said subset.

Abstract: The invention relates to the estimation of a time-variable physical quantity by a distributed system comprising a centralized platform and a set of distributed measurement devices each communicating with the platform via a communication network, on the basis of the measurement (100), by each device of the set, of a local datum representative of the physical quantity, the collection (120) by the centralized platform, over a predetermined sampling period, of measurement samples originating from a subset of devices of said set via said communication network, and the estimation (130) of the physical quantity over the sampling period on the basis of the collected samples. According to the invention, each device of said set executes (110), in each sampling period, a local algorithm deciding randomly and based on a sending probability associated with each device, whether the corresponding measurement device is part of said subset.

Abstract: A method is provided for smoothing of an overall load curve. The method includes aggregation of a plurality of elementary load curves. In one example, the method makes provision to determine (S1) a regulation period di associated with the load i; determine and store each elementary load curve by obtaining (S2) a plurality of measured consumption samples with a sampling period Tech that is a submultiple of each regulation period di. Starting from a reference time common to all of the loads, the method time-shifts (S3) the operation of each load i by a random shift ?ti depending on the sampling period Tech, on the regulation period di determined for the load i and on a random integer value Ni associated with the load i.

Abstract: “A method is provided for smoothing of an overall load curve. The method includes aggregation of a plurality of elementary load curves. In one example, the method makes provision to determine (S1) a regulation period di associated with the load i; determine and store each elementary load curve by obtaining (S2) a plurality of measured consumption samples with a sampling period Tech that is a submultiple of each regulation period di. Starting from a reference time common to all of the loads, the method time-shifts (S3) the operation of each load i by a random shift ?ti depending on the sampling period Tech, on the regulation period di determined for the load i and on a random integer value Ni associated with the load i.