Systems and methods for optimal synchrophasor data recovery

A system for recovering missing phase measurement unit (PMU) measurements, comprising:

• a plurality of PMUs configured to obtain a plurality of PMU measurements;
• a hierarchical computing system comprising:
• one or more first computing systems configured to:
• receive the plurality of obtained PMU measurements from the plurality of PMUs, wherein each of the plurality of obtained PMU measurements is associated with a time stamp and a sensor reading corresponding to an electrical characteristic;
• and
• provide the plurality of obtained PMU measurements to one or more second computing systems;
• and
• the one or more second computing systems configured to:
• populate a PMU dataset based on the plurality of obtained PMU measurements;
• determine a plurality of missing entries within the PMU dataset, wherein each of the plurality of missing entries indicates a missing PMU measurement within the PMU dataset at a particular time;
• determine a plurality of substitute entries for the plurality of missing entries based on an optimization algorithm that determines differences associated with a missing entry, of the plurality of missing entries, and a first set of PMU measurements, of the plurality of obtained PMU measurements, that are taken immediately prior to the missing entry;
• insert the plurality of substitute entries into the PMU dataset to generate a new PMU dataset;
• calculate a current state for an electrical power system using the new PMU dataset;
• and
• provide the current state to a power generation system of the electric power system, wherein the power generation system increases or decreases power provided by the electrical power system based on the current state,
• wherein the PMU dataset is a matrix indicating the plurality of obtained PMU measurements and the plurality of missing entries, and wherein the one or more second computing systems are configured to determine the plurality of substitute entries by using a column stacking vector and a total variation (TV) algorithm to determine a TV norm definition associated with the plurality of obtained PMU measurements and the plurality of missing entries.

A non-transitory computer-readable medium having processor-executable instructions stored thereon, wherein the processor-executable instructions, when executed by one or more controllers, facilitate:

• receiving a plurality of obtained PMU measurements from a plurality of PMUs, wherein each of the plurality of obtained PMU measurements is associated with a time stamp and a sensor reading corresponding to an electrical characteristic;
• populating a PMU dataset based on the plurality of obtained PMU measurements;
• determining a plurality of missing entries within the PMU dataset, wherein each of the plurality of missing entries indicates a missing PMU measurement within the PMU dataset at a particular time;
• determining a plurality of substitute entries for the plurality of missing entries based on an optimization algorithm that determines differences associated with a missing entry, of the plurality of missing entries, and a first set of PMU measurements, of the plurality of obtained PMU measurements, that are taken immediately prior to the missing entry;
• inserting the plurality of substitute entries into the PMU dataset to generate a new PMU dataset;
• calculating a current state for an electrical power system using the new PMU dataset;
• and
• providing the current state to a power generation system of the electrical power system, wherein the power generation system increases or decreases power provided by the electrical power system based on the current state,
• wherein the PMU dataset is a matrix indicating the plurality of obtained PMU measurements and the plurality of missing entries, and wherein determining the plurality of substitute entries comprises:
• using a column stacking vector and a total variation (TV) algorithm to determine a TV norm definition associated with the plurality of obtained PMU measurements and the plurality of missing entries.