Abstract
Power distribution systems rely on the use of active shunt compensators for power quality (PQ) improvement. This paper proposes the application of a new technique based on Long Short-Term Memory (LSTM) to mitigate PQ problems and achieve shunt compensation. The proposed controller comprises a single LSTM cell and the weights of the LSTM are updated online. The output of the LSTM is trained to determine correctly component of load current corresponding to 50 Hz, which is then used for active filtering and compensation. A Simulink-based MATLAB model is developed for the grid connected power distribution network feeding nonlinear loads. A single-phase voltage source converter configured as H-bridge is controlled as a compensator. Additionally, a photovoltaic source of 5 kW rating is integrated at the DC link of this converter. In both the cases discussed, the proposed LSTM network is found to be successfully track the fundamental load current component and utilize the same for compensation. Simulation results justify the ability of the controller developed using LSTM in achieving power factor correction of supply current and reactive and active power injection.
Similar content being viewed by others
Abbreviations
- v s :
-
Supply voltage (V)
- i s :
-
Grid current (A)
- i l :
-
Load current (A)
- i c :
-
Compensator current (A)
- P pv :
-
Active power of PV array (kW)
- Vpv :
-
PV array voltage (V)
- I pv :
-
Current injected by PV array (A)
- Vdc :
-
DC bus voltage (V)
- Vdcref :
-
Reference voltage at the DC bus (V)
- C dc :
-
Capacitance of the inverter (µF)
- R L 1 :
-
Source resistance (Ω)
- L L 1 :
-
Source inductance (mH)
- L f :
-
Filter inductance (mH)
- L b :
-
Inductor of the boost converter (mH)
- C b :
-
Capacitor of the boost converter (mH)
- x t :
-
Input signal
- y t :
-
Output signal
- d t :
-
Desired output (A)
- d eff :
-
Effective output (A)
- d pv :
-
PV contribution (A)
- u p :
-
Unit template
- W :
-
Weight (A)
- μ :
-
Learning rate
- P, Q :
-
Real and reactive powers (W, var
References
B. Singh, A. Chandra, K. Al-Haddad, Power Quality: Problems and Mitigation Techniques (Wiley, Chichester, 2015)
B. Singh, J. Solanki, An implementation of an adaptive control algorithm for a three-phase shunt active filter. IEEE Trans. Industr. Electron. 56, 2811–2820 (2009)
M.I.M. Montero, E.R. Cadaval, F.B. Gonzalez, Comparison of control strategies for shunt active power filters in three-phase four-wire systems. IEEE Trans. Power Electron. 22, 229–236 (2007)
G. Pathak, D. Mohanty, S.K. Dwivedi, Implementation of MVF-based control technique for 3-Φ distribution static compensator. J. Inst. Eng. India Ser. B 100, 589–597 (2019)
O. Kukrer, H. Komurcugil, R. Guzman, L.G. de Vicuna, A new control strategy for three-phase shunt active power filters based on FIR prediction. IEEE Trans. Industr. Electron. 68, 7702–7713 (2021)
V. Puranik, S.R. Arya, SOGI–FLL based adaptive filter for DSTATCOM under variable supply frequency. J. Inst. Eng. India Ser. B 98, 423–431 (2017)
V.N. Jayasankar, U. Vinatha, Backstepping controller with dual self-tuning filter for single-phase shunt active power filters under distorted grid voltage condition. IEEE Trans. Ind. Appl. 56, 7176–7184 (2020)
S. Mishra, C.N. Bhende, Bacterial foraging technique-based optimized active power filter for load compensation. IEEE Trans. Power Deliv 22, 457–465 (2007)
A. Bhattacharya, C. Chakraborty, A shunt active power filter with enhanced performance using ANN-based predictive and adaptive controllers. IEEE Trans. Industr. Electron. 58, 421–428 (2011)
J. Fei, H. Wang, Y. Fang, Novel neural network fractional-order sliding-mode control with application to active power filter. IEEE Trans. Syst. Man Cybern. Syst. 2021, 1–11
S.R. Arya, B. Singh, Neural network based conductance estimation control algorithm for shunt compensation. IEEE Trans. Industr. Inf. 10, 569–577 (2014)
P. Chittora, A. Singh, M. Singh, Chebyshev functional expansion based artificial neural network controller for shunt compensation. IEEE Trans. Industr. Inf. 14, 3792–3800 (2018)
J. Fei, H. Wang, Experimental investigation of recurrent neural network fractional-order sliding mode control of active power filter. IEEE Trans. Circuits Syst. II Express Briefs 67, 2522–2526 (2020)
J. Fei, Y. Chen, Dynamic terminal sliding-mode control for single-phase active power filter using new feedback recurrent neural network. IEEE Trans. Power Electron. 35, 9904–9922 (2020)
F.A. Gers, J. Schmidhuber, F. Cummins, Learning to forget: continual prediction with LSTM. Neural Comput. 12, 2451–2471 (2000)
G. Dudek, P. Pełka, S. Smyl, A hybrid residual dilated LSTM and exponential smoothing model for midterm electric load forecasting. IEEE Trans. Neural Netw. Learn. Syst. 2021, 1–13
C.H. Liu, J.-C. Gu, M.-T. Yang, A simplified LSTM neural networks for one day-ahead solar power forecasting. IEEE Access 9, 17174–17195 (2021)
Acknowledgements
The author is thankful to the Department of Science and Technology, Govt. of India.
Funding
No funding was received for the completion of this study.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The author declares there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
AC mains: 220 V, 1-phase, 50 Hz. Rs = 0.05 Ω, Ls = 1 mH. Vdc = 400 V. Lf = 3.1 mH. Nonlinear diode rectifier with RL1 = 15 Ω, LL1 = 100 mH, 5 kW PV, Voc = 37.5 V, Isc = 8.7 A, Lb = 5 mH, Cb = 500 µF.
Rights and permissions
About this article
Cite this article
Singh, A. Applications of Adaptive Long Short-Term Memory to Active Filtering. J. Inst. Eng. India Ser. B 103, 737–746 (2022). https://doi.org/10.1007/s40031-021-00685-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40031-021-00685-4