Generating high resolution inferences in electrical networks

A computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a device to cause the device to:

• collect multiple items of data from one or more data sources arising from a power distribution network, wherein the multiple items of data comprise (i) one or more levels of temporal resolution and (ii) one or more levels of spatial resolution;
• determine a network topology of the power distribution network based on identification of one or more connections between each of multiple components within the power distribution network;
• and
• perform an optimization by iteratively:
• (a) generating a power flow estimate for one or more nodes within the power distribution network at a pre-determined level of spatio-temporal resolution, wherein said generating comprises applying a model to (i) the multiple items of collected data, (ii) the determined network topology, and (iii) one or more relations constraining the power flow estimate at the pre-determined level of spatio-temporal resolution;
• and
• (b) identifying a class of one or more models which characterize the power flow at the pre-determined level of spatio-temporal resolution.

A system comprising:

• a memory;
• and
• at least one processor coupled to the memory and configured for:
• collecting multiple items of data from one or more data sources arising from a power distribution network, wherein the multiple items of data comprise (i) one or more levels of temporal resolution and (ii) one or more levels of spatial resolution;
• determining a network topology of the power distribution network based on identification of one or more connections between each of multiple components within the power distribution network;
• and
• performing an optimization by iteratively:
• (a) generating a power flow estimate for one or more nodes within the power distribution network at a pre-determined level of spatio-temporal resolution, wherein said generating comprises applying a model to (i) the multiple items of collected data, (ii) the determined network topology, and (iii) one or more relations constraining the power flow estimate at the pre-determined level of spatio-temporal resolution;
• and
• (b) identifying a class of one or more models which characterize the power flow at the pre-determined level of spatio-temporal resolution.

A computer-implemented method, comprising:

• obtaining multiple items of data from a power distribution network, wherein the multiple items of data comprise (i) one or more temporal resolutions and (ii) one or more spatial resolutions;
• identifying a network topology of the power distribution network based on identification of one or more connections between each of multiple components within the power distribution network;
• determining (i) one or more nodes within the power distribution network for which power flow is to be estimated and (ii) a spatio-temporal resolution at which the power flow is to be estimated based on one or more items of enterprise-related information;
• performing an optimization by iteratively:
• (a) estimating the power flow for the one or more nodes at the spatio-temporal resolution, wherein said generating comprises modelling one or more relations constraining the power flow estimate at the pre-determined spatio-temporal resolution to the multiple items of collected data based on the determined network topology;
• and
• (b) identifying a class of one or more models which characterize the power flow at the spatio-temporal resolution;
• and
• communicating (i) the power flow estimate for the one or more nodes at spatio-temporal resolution and (ii) the identified class of one or more models, to one or more decision support systems related to the power distribution network;
• wherein the steps are carried out by at least one computing device.