Settlement Differences

While William was working at the Australian Energy Market Operator (AEMO), he performed a quantitative analysis of long-term trends in load data across the entire NEM, based on data from one of the largest databases in the southern hemisphere. The year-long project investigated the reasons and possible solutions to the increasing settlement difference across the NEM, studying correlations between data types and accuracy. It involved the quantitative analysis of market energy data (including load and generation data on half-hour trading intervals). Statistical tools were applied to the data, including frequency analysis (i.e., Fourier analysis), the calculation of correlation coefficients, and forecasting of trends using SQL, Matlab, Access, and Excel. Dr. William Peter gave a presentation on the results at the Settlement Managers Meeting in Sydney, Australia in June 2008, and as a result, the Settlement KPI, a metric of AEMO’s performance, was changed. Data analysis tools developed by William are still being applied at AEMO to track long-term trends in load data.