Page 95 - IJOCTA-15-4

Advanced frequency control strategy for power systems with high renewable energy penetration
in a drop in solar radiation intensity and corre- the “no transmission” state to active power trans-
sponding power output. Specifically, at 100 s, the mission, raising the frequency for about 30 s af-
power output of PV1 and PV2 drops abruptly ter the fault. The BESS then acts as the pri-
from 500 to 85 MW and from 900 to 333 MW, re- mary frequency controller, with a post-fault re-
spectively. At 150 s, the clouds begin to disperse, sponse time of approximately 1 min. Once the
and the clear sky allows for maximum solar ra- clouds cleared and radiation levels returned to
diation, leading to an increase in power output, normal, the output power increased due to the
which stabilizes at 1542 MW, with PV1 reaching excess power transmitted from the BESS to the
523 MW and PV2 reaching 1019 MW. grid, resulting in a slight increase in system fre-
In such a negative scenario, it is evident that quency.
the power system’s frequency will experience sig- Table 2. The parameters of the proposed control
nificant oscillations. Figure 10 shows the fre- method
quency response at several buses considered most
influential to the system. Upon observing this Parameters Value Unit
figure, it is clear that the frequency fluctuations 1. The active power-frequency control
at some buses do not meet the grid connection K f 0.0154 -
requirements, 40 with bus 25 being the most af- K A 4.837 -
fected. The frequency of bus 25 reaches 48.6283 K E 0.219 -
Hz, which exceeds the allowed frequency limit of P max 1.0 pu
49 Hz. P min −1.0 pu

f db 1.0 pu
2. The active power control loop
4.2.2. Scenario 2 d
T 1 11 s
This case involves a sudden lack of generating ca- T 2 d 0.7 s
pacity. The tested IEEE 39-bus system includes T d 0.3 s
Generator 6, which has a capacity of 650 MW and T 3 d 0.2 s
is connected to the nearby PV solar farms. In T 4 d 0.1 s
this scenario, Generator 6 suddenly experienced 5 d
T 6 0.4 s
a fault and tripped out of the power system for T d 10 s
a period of 50 s. The frequency of some buses K w d 100 -
in the system is depicted in Figure 11. Based on p 1
K d 46 -
this result, it is clear that bus 25 experiences more i 1
significant frequency fluctuations than the other I d max 1.0 pu
buses, with a frequency decrease of 48.6046 Hz, I d min −0.4 pu
falling below the permissible frequency of 49 Hz. I q max 1.0 pu
I q min −1.0 pu
Thus, in both scenarios, the system’s status
is at a dangerously unstable level. Therefore, a 3. The reactive power control loop
q
specific solution is required to improve stability. T 1 q 0.2 s
This paper proposes the installation of the BESS T 2 q 1.4 s
T 0.45 s
at bus 25. The mathematical modeling and con- 3 q
T 0.3 s
trol strategy for the BESS are presented in Sec- 4 q
T 0.1 s
tion 4, with its control technique and parameters 5 q
T 0.4 s
detailed in Figure 6 and Table 2. Additionally, to 6
q
evaluate the effectiveness of the proposed method, T w q 9.0 s
the CBEST method introduced by Yan et al. 34 is K p 1 2.0 -
q
also referenced for simulation and comparison. K i 1 85 -
4. The current controller
Figures 12 and 13 show the simulation results q
for Scenario 1. Figure 12 illustrates the frequency K p 2 0.1 -
q
fluctuations on bus 25. As shown, the system fre- K i 2 10 -
d
quency drops from 50 Hz when cloud cover oc- K p 2 2.5 -
d
curs. The BESS’s active response is shown in K i 2 200 -
Figure 13a, where the BESS generates additional ChargeCur 0.05 pu
active power for the grid to compensate for the SOC min 0.1 %
capacity loss caused by cloud cover affecting the SOC max 100 %
PV1 and PV2 solar farms. During this period, AbsCur max 1.0 pu
the BESS receives the signal and switches from U threshold for I q preference 0.9 pu
637