讲座题目：能源信息学和分数阶微积分

（Energy Informatics and Fractional Calculus）

讲座时间：2022年11月16日下午2:30-3:30

讲座地点：Tencent Meeting ID：900-369-203

讲座简介：

Fractional calculus (FC) is about differentiation or integration of non-integer orders. More and more evidences show that FC enables us understand complexity better, control complex systems and process complex signal better. Energy systems are complex with both short and long term memory, as well as seasonality.Energy informaticsis a “research fieldcovering the use ofinformation and communication technologyto address energy utilization and management challenges.Methods used for "smart" implementations often combine IoT sensors withartificial intelligenceandmachine learning.” We have shown that energy informatic and fractional calculus are inherently linked. We show a technical case where a fractional order model known as GARMA (Gegenbauer ARMA) model can be used to characterize both long range dependence and seasonality. Gegenbauer polynomial has a generating function of a real power of a quadratic polynomial. When this quadratic polynomial is thez-transfer function of a second order IIR or FIR, raising this whole z-transfer function to the power of a real number, we will get an irrational z-transfer function whose output is called theGegenbauer processdriven by white noise. It turns out that, Gegenbauer process can exhibit both long range dependence (or long memory) and seasonality. This is particularly attractive in modeling time series in energy informatics such as wind speed, power consumption, energy price etc. where long memory and seasonality both dominate. In this seminar, we will introduce this important yet relatively new class of models known ask-factor Gegenbauer ARMA and its applications in time series modeling and prediction in energy informatics.