Data-Driven Static Load Model Parameter Estimation with Confidence Factor

In this paper, the probability of the estimated load parameters accuracy is evaluated using condition number and fuzzy rule base. The voltage and power measured in the field are buffered to form data vector and the exponential dependence of load parameter on voltage magnitude is computed when the voltage at the load bus changes due to external factors. The condition number of Vandermonde matrix and the normalized residue error are provided to a fuzzy rule base to compute the probability of the estimated load parameter being accurate. Several cases of voltage change (discrete and continuous) at the load bus are considered in this work for evaluating the parameters and probability of load model accuracy. The probability of the accuracy of load model parameter assists the system operator to reliably increase the tie line power flow through the voltage constrained interface which is otherwise restricted due to constant power model during the planning stage.