Two-Time-Scale Multi-Objective Coordinated Volt/Var Optimization For Active Distribution Networks

The penetration of renewable energy resource and electrical vehicle into distribution networks bring a great challenge to the operation of distribution network cause the intermittent nature of renewable energy resource and the uncertainty of electrical vehicle load.In order to maintain network voltage profile and reduce network loss,volt/var optimization(VVO)is of great significance in active distribution networks.This thesis presents the two-time-scale multi-objective coordinated volt/var optimization for active distribution networks.The main research are as follows:(1)This thesis presents a two-time-scale(hourly and 15-min)multi-objective coordinated volt/var optimization for active distribution networks considering the fluctuation of RES and reaction speeds of different control devices.In the hourly timescale,power loss and number of control actions of on-load tap changers and shunt capacitors for 24 hours are minimized by coordinating on load tap changers,shunt capacitors,and reactive power outputs of renewable energy resource.A three-step method consisting of global search,user preference,and local search,is adopt to solve the corresponding multi-objective mixed-integer nonlinear optimization problem to obtain the user preferred solution.The hourly settings of slow control devices(on load tap changer and shunt capacitor)for 24 hours are determined after hourly VVO.(2)In the 15-min time-scale,the RES reactive powers are optimized with RES uncertainties considered to minimize the pilot-bus voltage deviation under all representative scenarios.Copula function and Latin hypercube sampling are used to generate RES scenarios.Affinity propagation clustering is adopted to reduce RES scenarios to get the representative scenarios.Interior point method solves the 15-min time-scale VVO.In this thesis,the impact of the identified pilot-bus on the entire network voltage profile are evaluated.The IEEE 33-bus distribution network is used to evaluate the proposed two-time-scale volt/var optimization formulations and methods with promising results.(3)In this thesis,a look-ahead reactive power control problem,with capacitors,OLTCs,DGs power factor,and the energy storage unit as control variables,is formulated as a multi-objective mixed-integer nonlinear optimization problem with the maximum allowable daily operating times of discrete control devices taken into account.A composite load model,including plug-in electric vehicles,a static induction motor,and ZIP loads,is adopted.A four-stage solution methodology consisting of priority list,group-based multi-objective particle swarm optimization,fuzzy logic and user evaluation is proposed to solving the corresponding problem.The IEEE 123-bus system is used to evaluate the proposed problem formulation and the effectiveness of the proposed method.