Load Calibration and Model Validation Methodologies for Power Distribution Systems

Methodologies to calibrate loads and validate power distribution system models are presented in this paper. The calibration methodology makes use of information contained in historical load and measurement data to adjust loads including the system losses and forecast future load conditions. This unified methodology has been applied to calibrate the loads of distribution systems with varying degrees of complexity. The test results demonstrate that the calibrated loads can be used to provide simulation and prediction of system behavior for both normal and faulted conditions. The validation methodology quantifies the accuracy and uncertainty of the calibrated model and enables both the model developer and the model user to have quantitative support to decide whether the model is adequate for its intended purpose. Validation results are also useful in pinpointing which portion of the model needs to be improved. Load calibration and model validation methodologies are important and timely topics as power systems are entering the era of smart grid and more measurement data will be available for model development, calibration and validation.