Computational Intelligence for Demand Response Exchange Considering Temporal Characteristics of Load Profile via Adaptive Fuzzy Inference System

This paper presents a computationally intelligent hybrid approach to incorporate the temporal characteristics of customer baseline load (CBL) in demand response exchange (DRX) mechanism using adaptive fuzzy inference system (FIS). The proposed hybrid approach considers the temporal characteristics of load profile using utilization factor, availability factor alongside the conventional/traditional willingness factor. The relation between load criticality and flexibility in terms of utilization and availability factors has been established and incorporated into the DR seller/customer bids in DRX through dynamic costing. Various models viz. linear, nonlinear, and exponential model etc., are developed to assess varying behavior of customer with respect to the CBL profile. In addition, a FIS is developed in this paper to account for uncertain/indistinct nature of input/information provided by the customer. To improve the performance of DRX market clearing, parameters of membership functions used in FIS are adaptively tuned using heuristic approaches. The performance of proposed hybrid model using FIS is compared with the traditional approach, fuzzy, and nonfuzzy approach without considering temporal characteristics, fuzzy, and nonfuzzy approaches with temporal characteristics only. The simulation results are presented and they demonstrate the superiority of FIS based hybrid model with CBL temporal characteristics through dynamic costing when compared to other models.