Day-ahead optimal dispatch of an integrated energy system considering time-frequency characteristics of renewable energy source output

• Published in: Energy (Oxford) Vol. 209; p. 118434
• Main Authors: Zheng, Lingwei, Zhou, Xingqiu, Qiu, Qi, Yang, Lan
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 15.10.2020; Elsevier BV
• Subjects: Adaptive hybrid dispatch time-step; Day-ahead optimal dispatch; Energy sources; Hilbert transformation; Hilbert-Huang transform; Integrated energy system; Integrated energy systems; Mapping; Mixed integer; Mixed integer quadratic programming; Operating costs; Optimization; Predictions; Quadratic programming; Renewable energy; Renewable energy sources; Renewable resources; Schedules; Step functions; Integrated energy system; Mixed integer quadratic programming; Day-ahead optimal dispatch; Adaptive hybrid dispatch time-step; Hilbert-Huang transform
• ISSN: 0360-5442; 1873-6785
• DOI: 10.1016/j.energy.2020.118434

The intermittency and uncertainty output of renewable energy source (RES) poses a challenge for the optimal dispatch of an integrated energy system (IES). Many current studies focus on the time-domain characteristics of the RES prediction, while ignoring the time-frequency characteristics of the prediction, thus the operation is usually scheduled on a uniform time-step. Based on this context, an adaptive hybrid dispatch time-step (AHDT) determination method based on the time-frequency characteristics of the prediction is proposed. Different from the traditional time-step, AHDT is non-uniform. Firstly, the Hilbert-Huang transform is used to convert the predicted RES output data to the time-frequency. Next, a time-step function which reflects the mapping of time-step and instantaneous frequency is constructed to suggest an adaptive time-step in each dispatch segment. Then, an optimal day-ahead dispatch model of a typical IES with AHDT, which is essentially a mixed integer quadratic programming (MIQP) problem, is established. Finally, the day-ahead schedules of each unit are obtained by solving the MIQP problem. The simulations are conducted in a total of twenty randomly selected days. The results show that the AHDT method is superior to the uniform time-step method for its effect in reducing the operating cost and the optimization-calculating time cost. •An adaptive hybrid dispatch time-step (AHDT) determination method is proposed.•AHDT based on time-frequency features of predicted data is non-uniform.•Time-frequency features of predicted renewable source output is extracted by HHT.•The method is proposed for day-ahead dispatch of integrated energy systems.•The simulations conducted in 20 random days show the performance in cost-saving.