Multistage Active Distribution Network Planning With Restricted Operation Scenario Selection

Lei Zhao,Li Yang,Ke Sun,Fei Chen,Yixin Huang,Jingjing Huang,Qiongdan Dai,Zhenzhi Lin,Lei Wang

doi:10.1109/access.2019.2936936

Abstract

With the development of active distribution networks (ADNs), more uncertainties need to be considered in distribution network planning (DNP), which increases the difficulties in scenario selection. Given this background, a multi-stage bi-level DNP model is proposed considering coordinated operation of distributed generation (DG), energy storage system (ESS) and controllable load (CL). In this model, a novel selection strategy for restricted operation scenarios based on the shadow price is proposed to reduce the complexity of scenario selection and ease the computational burden substantially in virtue of the decoupling of ADN operation and planning. A set consisting of restricted operation scenarios is simulated as an information feedback from operators to planners. The active distribution network planning (ADNP) model is aimed to minimize the total cost of feeder investments, ESS investments, and the additional operation costs caused by severely constrained network resources. The decision variables include the location and type selections of feeders, and the siting, power and capacity of ESSs at each stage of a given planning horizon. An extended IEEE 33-node distribution system and an actual 62-node distribution system in Zhejiang, China are serviced to verify the proposed model. The results show that the proposed ADNP model can effectively select out the restricted operation scenarios and achieve an optimal ADNP scheme by balancing non-network solutions and network solutions.

Highlights

Nowadays, the contradiction between energy demand and environmental protection is increasingly prominent
With the coordinated operation of the various distributed flexible resources in the active distribution network (ADN), such as wind farm (WF) [1], photovoltaic (PV), micro grid (MG) [2], [3], energy storage system (ESS) [4], electric vehicle (EV) [5] and controllable loads (CL) [6], both the difference between peak and valley loads and the negative effect of high renewable distribution generation (DG) penetration are reduced, which delays the upgrading of feeders, substations and other equipment of distribution network and enhance flexibility of distribution network effectively [7]–[9]
An extended IEEE 33-node distribution system is utilized to verify the effectiveness of the proposed multistage active distribution network planning (ADNP) model considering the restricted operation scenarios

Summary

PARAMETERS δ

Impact coefficient of the selected restricted operation scenarios. Duration of each dispatching stage. Number of operation days at each stage. Charging efficiency, discharging efficiency and self-discharging rate of energy storage. Maximum power supply capacity of feeder type m. Maximum power and energy capacity of ESS at node j. Maximum curtailment rates of load at node j, wind power and photovoltaic. Lower and upper limit rates of state of charge of ESS. Lifetimes of feeders and ESSs. Capital recovery rates for investment in cSn , cPn , cEn. ESSs and feeders. Unit installation costs of power rating and energy reservoir cLij,n, cij,m,n of ESS at stage n. Construction costs per unit capacity and per unit length of new feeder ij, type m, CnIn,mv ax Vj,min, Vj,max stage n. Lower and upper limits of voltage at node j

VARIABLES

INTRODUCTION

COMPLEX OPERATION SCENARIOS FOR MULTISTAGE PLANNING OF ADN

SHADOW PRICE SOLVING METHOD

RESTRICTED OPERATION SCENARIOS SELECTION

CONCLUSION