Distributed Voltage and Frequency Control Using Electric Springs
Zohaib Akhtar*, Balarko Chaudhuri, Ron Hui
•Future power systems would have low inertia leading to larger deviations
and rates of change of grid frequency (RoCoF).
•Large loss of infeed (e.g. due to a fault in the DC grid) higher than the
spinning reserve could be more frequent.
•Primary frequency control would be challenging.
•Wind farms and possibly, loads (of certain types) could be required to
contribute to primary frequency control.
•Conventional demand response is tailored for peak shaving, peak load
deferring etc. but NOT for primary frequency response provision.
The concept of ‘Electric Spring (ES)’
has been proposed as an effective
means of distributed voltage and
frequency control . It is a
controllable voltage source which is
connected in series with a non-
critical load.The combination of the
ES and the non-critical load is called a
Depending on the injected voltage
magnitude and phase angle, the smart
load is capable of providing inductive,
capacitive, and positive real or
negative real power compensation.
 H. Shu Yuen, L. Chi Kwan, and F. F. Wu, "Electric Springs - A New Smart
Grid Technology," Smart Grid, IEEE Transactions on, vol. 3, pp. 1552-1561,
2014 IEEE PES General Meeting
2. RESEARCH OBJECTIVES
•How to convert normal loads (of
certain types) into SMART LOADS
(SLs) whose power consumption can
be controlled to provide primary
•How to achieve the above without
affecting the voltage across the
sensitive (or critical) loads?
•Validate the effectiveness of SLs
through system studies at distribution
level using realistic power system
A. TEST NETWORK
•IEEE37 bus test feeder was used.
•Upstream network was modelled
as a single source with a P-f droop
connected in series with an
•32 loads were equally divided into
critical and non-critical type.
B. ES VERSION 1:
•ES voltage is locked in quadrature to
the series current,so the ES can only
provide reactive power.
5. FUTURE WORK
•Design of an optimal control to minimize the control effort.
•Load characterization and identification of loads that can serve as non-
•Estimation of power ratings of ESs and level at which ESs are to be
installed (MV/LV) to achieve desired voltage and frequency control.
Fig. 6: 15% decrease in torque in FCM. Fig. 7: 15% increase in torque in FCM.
Fig. 8: TVR ESs ver 1.
•No battery storage required.
•Can work in Voltage Control Mode
(VCM)or Frequency Control Mode
4. SIMULATION RESULTS
C. ES VERSION 2:
•No restriction on ES voltage
•Capable of both active and
reactive power compensation.
•Battery storage needed (but the
main source of storage is non-
•Can control voltage and
Good voltage regulation but poor frequency regulation in VCM.
Frequency regulation in FCM at the cost of voltage variation. The
critical load voltage is still within ±5% of the nominal value.
Fig. 1: Smart load concept.
3. ELECTRIC SPRING (ES)
Fig. 2: ES connection in series with
non-critical load to form smart load.
Fig. 4: 15% decrease in torque in VCM. Fig. 5: 15% increase in torque in VCM.
Fig. 9: TFR ESs ver 1.
Fig. 10: RoCoF ESs ver 1. Fig. 11: Total ESs Q required.
Good frequency and voltage regulations achieved simultaneously.
Compensation required is higher in the case of over frequency
compared to that of under frequency.
Fig. 13: TVR with ESs Ver 2.
Fig. 14: TFR with ESs Ver 2.
Fig. 12: 15% reduction in torque with
ESs ver 2. Fig. 15: Total ESs P and Q required.
*Acknowledgement: Funded by the Commonwealth Scholarship.
Fig. 3: IEEE 37-bus test feeder.