۱- Introduction

۲- Meta-cognitive interval type-2 neuro fuzzy inference system

۳- Meta-cognitive learning algorithm for IT2FIS

۴- Performance evaluation

۵- Conclusion

References

Abstract

Renewable energy is fast becoming a mainstay in today’s energy scenario. One of the important sources of renewable energy is the wave energy, in addition to wind, solar, tidal, etc. Wave prediction/forecasting is consequently essential in coastal and ocean engineering studies. However, it is difficult to predict wave parameters in long term and even in the short term due to its intermittent nature. This study aims to propose a solution to handle the issue using Interval type-2 fuzzy inference system, or IT2FIS. IT2FIS has been shown to be capable of handling uncertainty associated with the data. The proposed IT2FIS is a fuzzy neural network realizing Takagi-Sugeno-Kang inference mechanism employing meta-cognitive learning algorithm. The algorithm monitors knowledge in a sample to decide an appropriate learning strategy. Performance of the system is evaluated by studying significant wave heights obtained from buoys located in Singapore. The results compared with existing state-of-the art fuzzy inference system approaches clearly indicate the advantage of IT2FIS based wave prediction.

Introduction

Renewable energy is becoming more acceptable as an alternative source of energy. They provide more environmental friendly and cheaper power but the electrical power generation is becoming more complex with inclusion of these sources. One of the main reasons for the complexity is the inability to accurately predict the strength of these sources at a given time. There are various natural as well as artificial causes. As a result, the forecast of the energy generated is very uncertain. This uncertainty leads to unpredictable or unrealistic generation, even leads to financial losses. Hence, realistic forecast of these sources is the need for increased and improved renewable energy usage. In this study, we attempt to forecast wave energy by working on an important characteristic of wave, namely significant wave height. Recently, artificial neural network has been used in predicting wave height [15, 9].