A multi-objective partitioned design method for integrated energy system

The complexity of a large-scale integrated energy system imposes huge computational burden. Besides, centralised state estimation is not suitable for fast and coordinated optimal management of multi-flow coupled systems and efficient energy utilisation. Furthermore, existing distributed state estimations are dealing with the established static system in the form of partitions. Considering the current method of modelling nonlinear fluids, the final impact on the performance of multiple assessments is non-convex for different partitioning approaches. This paper proposes a multi-objective distributed state estimation design approach for an integrated energy system based on non-dominated sorting genetic algorithm-II and unscented Kalman filter. The integrated energy system model contains the electric-gas-thermal system and various coupled units. By comparing and evaluating the estimation accuracy, calculation time and economic indicators of the system with different partitions of the system load, the optimal Pareto solution set are obtained from the multiobjective optimisation, which then guides the construction layout to satisfy different application requirements. In situations where the specific requirements are not clear, this paper gives the operator an objective method recommendation with the help of the entropy weight and Topsis synthesis assessment method. The validity of the method is verified by several case studies, and the method not only assists the estimation of the existing integrated energy system, but it also offers engineering significance in guiding the construction of the future integrated energy system.