JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, MARCH 2020 1
Optimal Design of Home Energy Storage System
Abstract—Home battery plays a significant role in improving
the usage of PV based renewable energy resources while main-
taining the local stability of power grid. This essay aims to design
a smart home battery which can intelligently response to the
price.
I. BACKGROUND (5 MARKS)
With the increasing interest in Net-zero target, it aims
to improve the economic potential of home battery system
through AI technique and optimisation algorithm.
Energy storage can be both a generation and a load. When
there is an excess energy exported from household PV genera-
tions, neighbours without a PV panel can buy it to satisfy their
demands which is referred as p2p energy trading. When there
is a network load congestion during the peak hours, demand
management can be applied for shaving the peak. Aggregated
household PV generations The FACS market allows AMEO
to manage the power system around 50Hz which requires the
supply-demand balance across the national electricity market.
Utility for local distribution network is referred to store
excess energy exported from household PV generations which
may cause overvoltage issues in distribution network. More-
over, during the peak hours, the network load congestion can
be largely relieved by discharging community storage to avoid
an expensive network upgrade. Utility also can come from
keeping the electricity service during outage or maintenance.
Fig. 1 illustrates system structure where a community en-
ergy storage system (CESS) is managed by energy manage-
ment system for providing energy services to the commercial
and the residential customers. These customers can own a roof
PV and small scale CESS which may produce a rebate for
providing service to the grid or his neighbors. The aggregated
PV power output could be given in Fig. 5 while the aggregated
load could be shown in Fig. 3.
II. SMART HOME BATTERY SYSTEM
A. Task 1: Load forecasting (7 MARKS)
Given a dataset from VIC, please choose one house where
a roof PV is installed. Thus it has power injected into the grid.
Please use a time series forecasting algorithm to predict
the load which may include the power injected into grid.
B. Task 2: Home Battery (8 MARKS)
Given the parameter of the battery in Tab. I, please design
an optimisation model for the optimal sizing of the home
battery according to the time-of-use pricing tariff. Here,
we assume that the price will be not changed over the time
and there is no transaction fee for selling energy into the grid
or the neighbours.
C. Task 3: Payback (5 MARKS)
After obtained this optimal size, what is the likely payback
period on this home battery?
JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, MARCH 2020 2
Fig. 1. The architecture of the smart home system
TABLE I
BATTERY MODEL PARAMETERS
Efficiency (discharging/charging) Value SoC Operation Parameter Value
nd/nc 0.95 SoCmax/SoCmin 0.8/0.2
Unit price $700 per kWh Installation cost $ 1000
warranty 10 years Maintenance cost $ 20 per kWh×year
Fig. 2. Generation data.
Fig. 3. Demand data.
Fig. 4. Winter Summer TOU Tariff in NSW.
Fig. 5. Summer TOU Tariff in NSW.

欢迎咨询51作业君