As part of its dynamic energy demand management strategy, our client, a leading energy company in Spain and with international business, needed to associate the meters spread around the network in customer facilities with the various transformers in the transformer substations and with their various low voltage outputs, either single-phase or three-phase, on the same transformer.
The algorithm developed was migrated in collaboration with one of Madrid’s most prestigious universities. The aim of this algorithm is to associate, in a probabilistic manner, given a customer meter, its output and phase in the transformer substation to which it is connected. In other words, to identify the low voltage phase and output that supplies each of the meters in the low voltage transformer substations with advanced supervision. All this is achieved by means of a measure of dependence used in the field of statistics and probability theory called correlation distance or distance covariance.
This productivity improvement project was implemented on an AWS architecture. The appropriate processing of the large amount of information produced, the understanding of the monitoring and supervision of network transformers and the detection of incremental voltage changes, and the measurement of consumption in the meters transferred by the PLC network were critical to proper implementation.
Supply records x 6 months historical logs = +720M
Records in transformers x 6 months historical logs = +214MM
Do you want to know more about what we offer and to see other success stories?