Using Surrogate Data to Mitigate the Risks of Natural Gas Forecasting on Unusual Days

Energy utilities see higher risk when forecasting for their operating areas (service territories) on days that are high-demand or difficult to forecast. These days often have unusual weather patterns (e.g., colder than normal or significant temperature fluctuation from previous days). Due to their unusual nature, data describing these days are scarce. We present a method that successfully transforms natural gas consumption data from operating areas in vastly different geographic regions and climates, with different customer bases, to make better forecasts for areas that have insufficient historical data. Our surrogate data transformation algorithm results in higher forecast accuracy, thereby reducing the risk to energy utilities