Establishing a Load Model for System Dynamic Simulations using Real and Historical Data

In power system planning, the use of an appropriate load model is a critical factor to the accuracy of results. By increasing the accuracy of the model used for loads, uncertainty is reduced, leading to cost savings from lowered safety margins. This allows for more effective and economical decision making. This project utilises the measurement-based approach to load modelling, with data collected from sources such as SCADA, PMUs, and transient recorders. A number of model structures were selected to be tested in order to select the most suitable structure and corresponding parameters. Least squares optimisation was then used to determine suitable values for the parameters by minimising the error between the data values and modelled values. As Least squares optimisation had a tendency to choose local optima, the use of Particle Swarm Optimisation to determine parameters is underway.