Forecasting the reliability of wind-energy systems: A new approach using the RL technique

Two of the most significant challenges in the 21st century will be to improve energy security and reduce the greenhouse gas emissions associated with energy consumption. A co-beneficial solution to these challenges is seen as increasing the use of renewable energy for the production of electricity. Some renewable sources, such as wind are often presented as a way to reduce greenhouse gas emissions; however, since wind’s variability increases uncertainty and risk in expected generation, it can be detrimental to energy security. One of the ways in which wind’s contribution to a jurisdiction’s energy security and greenhouse gas reduction strategies can be improved is to employ a forecasting method that can help reduce risks. This paper proposes a method that applies risk and reliability analysis techniques to obtain the most-likely RL (Resistance–Load) scenario using a set of historical data for wind-supply or generation and load. RL estimates the reliability of a wind-energy system by simulating an anticipated resistance (the electrical generation) attempting to meet a load (the electricity demand) for a future year. The method is demonstrated through a case study and its results are compared with real-time data from a 12 MW wind farm to prove its efficacy.

► A method of forecasting reliability for wind-energy systems is developed.
► Uncertainties of random variables, such as, wind speeds, system losses, and load are addressed using probability distributions.
► Simulating future generation and load (or R and L) scenarios through random variate generation and Monte-Carlo simulation.
► Applying RL technique or reliability equations to forecast probability of R > L for wind-energy systems.