The effect of humidity in a weather-sensitive peak load forecasting model

When considering mid-term peak load evolution, it is crucial to estimate the impact of economic growth as well as the effect of climatic conditions. In order to take into account the impact of both the above-mentioned parameters, many peak load forecasting approaches are based on the decomposition of loads into non-weather sensitive and weather sensitive components. The most commonly considered weather variable is temperature, while other variables such as relative humidity, wind velocity, cloudiness etc can have a significant effect. In this paper a well-established weather sensitive methodology for peak load forecasting is extended in order to take into account the effect of relative humidity on peak loads. The resulting methodology is applied to the system of Cyprus, a Mediterranean country where high levels of relative humidity are typical. Numerical results are presented and compared to forecast results where the effect of relative humidity is neglected.