Systems and methods for sequential power system model parameter estimation

Abstract

A system for sequential power system model calibration is provided. The system includes a computing device in communication with at least one sensor monitoring a power system. The computing device includes at least one processor in communication with at least one memory. The at least one processor is programmed to store a model of a device. The model includes a plurality of parameters. The at least one processor is also programmed to receive a plurality of events associated with the device, filter the plurality of events to generate a plurality of unique events, sequentially analyze the plurality of unique events to determine a set of calibrated parameters for the model, and update the model to include the set of calibrated parameters.

1 . A system for sequential power system model calibration comprising a computing device including at least one processor in communication with at least one memory device, wherein the at least one processor is programmed to:
 store a model of a device, wherein the model includes a plurality of parameters;   receive a plurality of events associated with the device;   filter the plurality of events to generate a plurality of unique events;   sequentially analyze the plurality of unique events to determine a set of calibrated parameters for the model; and   update the model to include the set of calibrated parameters.   
     
     
         2 . The system in accordance with  claim 1 , wherein the at least one processor is further programmed to:
 execute the model based on one or more events of the plurality of events to generate one or more results; and   identify one or more sensitive parameters based on the one or more results.   
     
     
         3 . The system in accordance with  claim 2 , wherein the at least one processor is further programmed to:
 perform a Bayesian optimization on the one or more sensitive parameters to determine updated values for the one or more sensitive parameters.   
     
     
         4 . The system in accordance with  claim 3 , wherein to perform the Bayesian optimization the at least one processor is further programmed to:
 determine the updated values for the one or more sensitive parameters based on a nonlinear optimization, wherein an objective function of the nonlinear optimization includes a first term and a second term;   calculate the first term as the residual between a simulated response based on the set of calibrated parameters and a measured response; and   calculate the second term as a quadratic penalty term for deviations of parameters from one or more previous estimates, wherein one or more weights for the quadratic penalty are derived from a Bayesian argument.   
     
     
         5 . The system in accordance with  claim 4 , wherein the at least one processor is further programmed to derive the quadratic penalty based on a covariance matrix of previous estimated parameters. 
     
     
         6 . The system in accordance with  claim 1 , wherein the at least one processor is further programmed to code each of the plurality of events based on one or more dynamic features of the corresponding event. 
     
     
         7 . The system in accordance with  claim 6 , wherein the plurality of events are each coded into a bit-string. 
     
     
         8 . The system in accordance with  claim 6 , wherein the one or more dynamic features include one or more of peak value, bottom value, overshoot percentage, a rising time, a settling time, a phase shift, a damping ratio, an energy function, and a cumulative deviation in energy, Fourier transformation spectrum information, principal component, and steady state gain of the corresponding event. 
     
     
         9 . The system in accordance with  claim 6 , wherein the at least one processor is further programed to extract the one or more dynamic features from a time series of active power, reactive power, voltage and frequency of the corresponding event. 
     
     
         10 . The system in accordance with  claim 6 , wherein the at least one processor is further programmed to:
 code each of the plurality of events as a binary vector;   compare the plurality of binary vectors using the Taminoto coefficient;   discard similar subsequent events based on a similarity threshold; and   generate the plurality of unique events based on at least one remaining event.   
     
     
         11 . The system in accordance with  claim 1 , wherein the plurality of unique events includes a first event and a second event, wherein the model includes a first set of parameters, and where the at least one processor is further programed to:
 execute the model using the first set of parameters and the first event to generate a first set of results;   analyze the first set of results to generate a second set of parameters;   execute the model using the second set of parameters and the second event to generate a second set of results; and   analyze the second set of results to generate a third set of parameters.   
     
     
         12 . The system in accordance with  claim 11 , wherein the plurality of unique events includes a third event, and where the at least one processor is further programmed to:
 execute the model using the third set of parameters and the third event to generate a third set of results; and   analyze the third set of results to generate a fourth set of parameters.   
     
     
         13 . The system in accordance with  claim 12 , wherein the at least one processor is further programmed to compare the first set of results, the second set of results, and the third set of results to determine the set of calibrated parameters. 
     
     
         14 . The system in accordance with  claim 13 , wherein each set of results includes residual error between a simulated response and a measured response, wherein the at least one processor is further programmed to compare the plurality of residual errors to select the set of calibrated parameters with minimal overall residual error. 
     
     
         15 . The system in accordance with  claim 1 , wherein the plurality of events include sensor data associated with the device during the corresponding event. 
     
     
         16 . The system in accordance with  claim 1 , wherein the device includes a power system and the model simulates behavior of the power system. 
     
     
         17 . A computer-implemented method for sequential power system model calibration, the method implemented by a computing device including at least one processor in communication with at least one memory device, wherein the method includes:
 storing a model of a device, wherein the model includes a plurality of parameters;   receiving a plurality of events associated with the device;   filtering the plurality of events to generate a plurality of unique events;   sequentially analyzing the plurality of unique events to determine a set of calibrated parameters for the model; and   updating the model to include the set of calibrated parameters.   
     
     
         18 . The method in accordance with  claim 17  further comprising:
 executing the model based on one or more events of the plurality of events to generate one or more results; 
 identifying one or more sensitive parameters based on the one or more results; and 
 performing a Bayesian optimization on the one or more sensitive parameters to determine updated values for the one or more sensitive parameters. 
 
     
     
         19 . The method in accordance with  claim 17  further comprising:
 coding each of the plurality of events based on one or more dynamic features of the corresponding event, wherein the plurality of events are each coded into a binary vector; 
 comparing the plurality of binary vectors using the Taminoto coefficient; 
 discarding similar subsequent events based on a similarity threshold; and 
 generating the plurality of unique events based on at least one remaining event. 
 
     
     
         20 . A non-transitory computer-readable storage media having computer-executable instructions embodied thereon, wherein when executed by a computing device having at least one processor coupled to at least one memory device, the computer-executable instructions cause the processor to:
 store a model of a device, wherein the model includes a plurality of parameters;   receive a plurality of events associated with the device;   filter the plurality of events to generate a plurality of unique events;   sequentially analyze the plurality of unique events to determine a set of calibrated parameters for the model; and   update the model to include the set of calibrated parameters.