Individual-based distributed parallel optimization for operation of integrated energy systems considering heterogeneous structure

Abstract

This paper proposes an individual-based distributed parallel (IBDP) method for the operation optimization of integrated energy systems considering heterogeneous structure. The proposed IBDP method utilizes individual-based models (IBMs) to describe integrated energy systems. In this model, three basic modelling units are summarized, from which, a general three-level model of integrated energy systems is inherited. Utilizing IBMs, the high-dimensional integrated energy system model is decoupled into distributed heterogeneous individuals. The individuals are with local information inside to protect their information privacy and independent operation from the environment. Meanwhile, global information, which provides with environmental boundary conditions, can be obtained through individual interactions. To solve the IBMs of integrated energy systems efficiently, an Proximal Jacobian alternating direction multiplier algorithm (Proximal Jacobian ADMM) is adopted to implement the IBDP where the subproblems of individuals run in parallel. To verify the veracity and stability of the proposed method, the IBDP is applied to optimize the operation strategy of integrated energy systems in four different scales, and the results are compared with that obtained by centralized optimization method.