Abstract
Energy storage systems (ESSs) can benefit distribution networks by providing multiple services to the distribution system operator and contributing to system reliability. Given the high capital costs of ESS, it is beneficial to optimally design them for their intended applications; however, this can be computationally expensive due to the non-convex formulation of the AC power flow, the complexity of business use cases and the simulation of multi-period operations. This study develops a comprehensive model to size the ESS to minimise the system lifetime costs and maximise reliability. The non-convex AC power flow model is modified using a convex relaxation, which yields efficient and globally optimal results. The battery degradation is simulated by a global wear coefficient considering the effect of depth of discharge. The sizing study is implemented to fulfil multiple applications on the Gussing distribution network in Austria.
Highlights
Distribution Networks face challenges driven by increasing penetration of renewable generation and changes in electricity demand due to the decarbonisation of heat and transport
This paper develops a comprehensive model to size the Energy Storage Systems (ESS) to minimize the system lifetime costs and maximise reliability
These challenges can be addressed in part by employing a Battery Energy Storage System (BESS) to provide services to the Distribution System Operator (DSO) including demand peak shaving and other ancillary services [1]
Summary
Distribution Networks face challenges driven by increasing penetration of renewable generation and changes in electricity demand due to the decarbonisation of heat and transport. In microgrids with the capability to island, reliability may be driven by network failures and by generation adequacy and the capability to provide a power unbalance response, which can be mitigated using a BESS. Neglecting these aspects when designing the BESS might hinder the optimal operation of the grid. This paper employs a convex formulation of the radial network model [6] to transfer the non-convex AC power flow model via Second Order Cone Programming (SOCP) This enables commercially available solvers to efficiently solve the multi-period sizing problem with high accuracy.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
