Historical AMI Data for Distribution System State Estimator Deployment

Stochastic planning studies that use load consumption as a random variable are performed to determine the optimal placement of sensors in distribution circuits so that advanced applications that rely on Distribution System State Estimation (DSSE) can function efficiently and violation-free. This study derives probability distributions of true load using Advanced Metering Infrastructure (AMI) net load and PV data, and incorporates them in a Monte Carlo simulation that estimates voltage on a real-world distribution circuit. The results are compared to results from a similar Monte Carlo simulations for which AMI datasets are not available and simplistic probability load distribution assumptions adopted from other studies were used instead. The analysis performed based on simplistic assumptions ignored extreme load values and, consequently, missed undervoltage violations that were detected by the more realistic AMI data driven analysis. Furthermore, we discuss how AMI data can be used as pseudo-measurements that DSSEs require for real-time operation.