ANFIS-PSO-Based Optimization for THD Reduction in Cascaded Multilevel Inverter UPS Systems
Abstract
:1. Introduction
2. Materials and Methods
2.1. ANFIS and PSO Fundamentals in Multilevel Inverter Optimization
2.1.1. Adaptive Neuro-Fuzzy Inference System
2.1.2. Training of ANFIS Networks
- Supervised or associative—inputs that correspond to specific outputs must be entered (maybe by an external agent or the system itself).
- Unsupervised or self-organized—must find statistical characteristics between groupings of input patterns.
2.1.3. Particle Swarm Optimization Algorithm
2.1.4. PSO Parameter Optimization
- Inertia coefficient (w)—This parameter regulates the influence of the particle’s previous velocity when calculating the new velocity value. It has a direct impact on the scanning capability and the convergence speed. A suitable inertia value allows the PSO to achieve a good initial scan while gradually adjusting the approach toward the optimal solution.
- Cognitive and social acceleration coefficients (c1 and c2)—These coefficients determine the degree of influence exerted by individual knowledge and collective knowledge on each particle. Coefficient c1 regulates the influence of the best position reached by the particle to guide its new direction, and coefficient c2 regulates the influence of the cluster leader in the calculation of the search direction of the particle.
- Number of particles and generations—As in other population-based metaheuristic algorithms, the number of particles and generations affects the accuracy and effectiveness of the search. A larger number of particles and generations can improve the search for the optimal solution, but it also increases the computational cost. To avoid unnecessary processes, the number of generations was limited after the algorithm achieved satisfactory results in a small number of iterations.
- Influence on convergence speed—The configuration favored fast convergence towards high-quality solutions, avoiding oscillations and stagnation at local optima.
- Exploration capacity—The swarm sought to explore the search space wisely, avoiding premature convergence towards suboptimal solutions.
- Balance between exploration and exploitation—The final configuration offers an appropriate balance between exploring new areas of the search space and exploiting the best solutions.
2.2. Combination of ANFIS and PSO for the Optimization of Multilevel Inverters
- Adaptability and learning of ANFIS—the ability to adapt to complex data and the capacity for continuous learning. By being trained with specific data sets, ANFIS can adjust its internal parameters to adapt to the system’s dynamics. This learning capability allows it to interpret complex patterns in the search for switching angles in the presence of some voltage variations or special cases.
- PSO optimization—the capability for global and dynamic optimization. Through the simulation of swarm behavior, PSO seeks to converge to optimal solutions in the search space. The combination of ANFIS and PSO leverages this optimization capability to further refine the adjustments made by ANFIS.
2.3. Problem Statement
2.4. Proposed Solution
2.5. Case of Study
3. Results
3.1. Network Training
3.2. MATLAB Simulation of the Case Study
4. Discussion
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
Abbreviations
ABC | Artificial Bee Colony |
ADELINE | Adaptive Linear Neuron |
AI | Artificial Intelligence |
ANN | Artificial Neural Networks |
ANFIS | Adaptive Neuro-Fuzzy Inference System |
BSMC | Backstepping Sliding Mode Control |
CHBMLI | Cascaded H-bridge multilevel inverter |
DC | Direct Current |
DTC | Direct Torque Control |
EPO | Emperor Penguin Optimization |
FFT | Fast Fourier Transform |
FLC | Fuzzy Logic Controller |
GA | Genetic Algorithm |
GD | Gradient Descent |
LS-PWM | Level-shifted PWM |
LSTM | Long Short-Term Memory |
MWH | Modified Widrow–Hoff |
NF | Neuro Fuzzy |
NR | Newton–Raphson |
PEC | Packed E-Cell |
P&O | Perturb and Observe |
PSO | Particle Swarm Optimization |
PWM | Pulse Width Modulation |
RMS | Root Mean Square |
SAPF | Series Active Power Filter |
SHAF | Shunt Hybrid Active Power Filter |
SHE | Selective Harmonic Elimination |
SRF | Synchronous Reference Frame |
THD | Total Harmonic Distortion |
UPS | Uninterruptible Power Supply |
References
- Aamir, M.; Ahmed Kalwar, K.; Mekhilef, S. Review: Uninterruptible Power Supply (UPS) system. Renew. Sustain. Energy Rev. 2016, 58, 1395–1410. [Google Scholar] [CrossRef]
- Keyan, S.; Dehong, X. Chapter 15—Operation and control of uninterruptible power supply system. Control. Power Electron. Convert. Syst. 2024, 4, 457–509. [Google Scholar]
- Nasiri, A.; Hamidi, S.A. Uninterruptible Power Supplies. In Power Electronics Handbook (Fourth Edition); Rashid, M.H., Ed.; Butterworth-Heinemann; University of Wisconsin-Milwaukee: Milwaukee, WI, USA, 2018; pp. 641–657. [Google Scholar]
- Godoy, M.P.; Uberti, V.A.; Abaide, A.d.R.; Guidali, G.D.; Prade, L.R.; Keller, A.L. Identifying and reducing harmonic distortion in an industrial uninterruptible power supply system. In Proceedings of the 2020 6th International Conference on Electric Power and Energy Conversion Systems (EPECS), Istanbul, Turkey, 5–7 October 2020; pp. 34–39. [Google Scholar]
- Katir, H.; Abouloifa, A.; Noussi, K.; Lachkar, I.; Giri, F. Cascaded H-Bridge Inverters for UPS Applications: Adaptive Backstepping Control and Formal Stability Analysis. IEEE Control Syst. Lett. 2022, 6, 145–150. [Google Scholar] [CrossRef]
- Kabalci, E. Multilevel Inverters Introduction and Emergent Topologies. In Multilevel Inverters; Kabalci, E., Ed.; Academic Press: Cambridge, MA, USA, 2021; pp. 1–27. [Google Scholar]
- Choudhury, S.; Bajaj, M.; Dash, T.; Kamel, S.; Jurado, F. Multilevel Inverter: A Survey on Classical and Advanced Topologies, Control Schemes, Applications to Power System and Future Prospects. Energies 2021, 14, 5773. [Google Scholar] [CrossRef]
- Azuela, J.H.S.; Cortés, F.R. Inteligencia Artificial Aplicada a Robótica y Automatización; Grupo, A., Ed.; S.A.Dec.V.: México City, Mexico, 2020; p. 382. [Google Scholar]
- Ponce, P. Inteligencia Artificial—Con Aplicaciones a la Ingeniería; Alfaomega: Mexico City, Mexico, 2010; p. 376. [Google Scholar]
- Vas, P. Artificial-Intelligence-Based Electrical Machines and Drives Application of Fuzzy, Neural, Fuzzy-Neural, and Genetic-Algorithm-based Techniques; Oxford Science Publicactions: Oxford, UK, 1999. [Google Scholar]
- Blum, C.; Roli, A.; Alba, E. An Introduction to Metaheuristic Techniques. In Parallel Metaheuristics A New Class of Algorithms; Zomaya, A.Y., Alba, E., Eds.; A John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2005. [Google Scholar]
- Shuai, Z.; Frede, B.; Huai, W. Chapter 8—Artificial Intelligence–Assisted Data-Driven Control of Power Electronics Systems; Academic Press: Cambridge, MA, USA, 2024; Volume 4, pp. 219–239. [Google Scholar]
- Shihabudheen, K.V.; Pillai, G.N. Recent advances in neuro-fuzzy system: A survey. Knowl.-Based Syst. 2018, 152, 136–162. [Google Scholar] [CrossRef]
- de Campos Souza, P.V. Fuzzy neural networks and neuro-fuzzy networks: A review the main techniques and applications used in the literature. Appl. Soft Comput. 2020, 92, 106275. [Google Scholar] [CrossRef]
- Muthuramalingam, M.; Manoharan, P.S. Comparative analysis of distributed MPPT controllers for partially shaded stand alone photovoltaic systems. Energy Convers. Manag. 2014, 86, 286–299. [Google Scholar] [CrossRef]
- Karuppusamy, P.; Natarajan, A.M.; Vijeyakumar, K.N. An Adaptive Neuro-Fuzzy Model to Multilevel Inverter for Grid Connected Photovoltaic System. J. Circuits Syst. Comput. 2015, 24, 1550066. [Google Scholar] [CrossRef]
- Nageswara Rao, G.; Chandra Sekhar, K.; Sangameswararaju, P. An Effective Technique for Reducing Total Harmonics Distortion of Multilevel Inverter. J. Intell. Syst. 2018, 27, 433–446. [Google Scholar] [CrossRef]
- Sudha Letha, S.; Thakur, T.; Kumar, J. Harmonic elimination of a photo-voltaic based cascaded H-bridge multilevel inverter using PSO (particle swarm optimization) for induction motor drive. Energy 2016, 107, 335–346. [Google Scholar] [CrossRef]
- Hemachandu, P.; Reddy, V.C.V. A Critical Evaluation of Advanced Multi-Carrier Modulation Scheme for 15-Level Inverter via PSO-PID Controller. Procedia Technol. 2016, 23, 240–247. [Google Scholar] [CrossRef]
- Francis, R.; Meganathan, D. An Improved ANFIS with Aid of ALO Technique for THD Minimization of Multilevel Inverters. J. Circuits Syst. Comput. 2018, 27, 1850193. [Google Scholar] [CrossRef]
- Das, S.R.; Ray, P.K.; Mohanty, A.; Das, H. Performance evaluation of multilevel inverter based hybrid active filter using soft computing techniques. Evol. Intell. 2019, 14, 345–355. [Google Scholar] [CrossRef]
- Lova Lakshmi, T.; Gopichand Naik, M. Soft-Computing Techniques for Voltage Regulation of Grid-Tied Novel PV Inverter at Different Case Scenarios. In Soft Computing and Signal Processing; Advances in Intelligent Systems and Computing; Springer: Singapore, 2019; pp. 181–190. [Google Scholar]
- Das, S.R.; Ray, P.K.; Mishra, A.K.; Mohanty, A. Performance of PV integrated multilevel inverter for PQ enhancement. Int. J. Electron. 2020, 108, 945–982. [Google Scholar] [CrossRef]
- Toumi, T.; Allali, A.; Meftouhi, A.; Abdelkhalek, O.; Benabdelkader, A.; Denai, M. Robust control of series active power filters for power quality enhancement in distribution grids: Simulation and experimental validation. ISA Trans. 2020, 107, 350–359. [Google Scholar] [CrossRef]
- Bihari, S.P.; Sadhu, P.K. Design analysis of high level inverter with EANFIS controller for grid connected PV system. Analog. Integr. Circuits Signal Process. 2020, 103, 411–424. [Google Scholar] [CrossRef]
- Mehrasa, M.; Babaie, M.; Zafari, A.; Al-Haddad, K. Passivity ANFIS-Based Control for an Intelligent Compact Multilevel Converter. IEEE Trans. Ind. Inform. 2021, 17, 5141–5151. [Google Scholar] [CrossRef]
- Rupesh, M.; Vishwanath, D.T.S. Fuzzy and ANFIS Controllers to Improve the Power Quality of Grid Connected PV System with Cascaded Multilevel Inverter. Int. J. Electr. Electron. Res. 2021, 9, 89–95. [Google Scholar] [CrossRef]
- Nikouei, S.Y.; Dehkordi, B.M.; Niroomand, M. A Genetic-Based Hybrid Algorithm Harmonic Minimization Method for Cascaded Multilevel Inverters with ANFIS Implementation. Math. Probl. Eng. 2021, 2021, 15. [Google Scholar] [CrossRef]
- Muralikumar, K.; Ponnambalam, P. Comparison of Fuzzy and ANFIS Controllers for Asymmetrical 31-Level Cascaded Inverter With Super Imposed Carrier PWM Technique. IEEE Access 2021, 9, 82630–82646. [Google Scholar] [CrossRef]
- Banu, J.B.; Jeyashanthi, J.; Ansari, A.T. DTC-IM drive using adaptive neuro fuzzy inference strategy with multilevel inverter. J. Ambient Intell. Humaniz. Comput. 2021, 13, 4799–4821. [Google Scholar] [CrossRef]
- Goggi Narasimha; Gummadavelli Naveen; Shireesha, K.; Kartheek, K.; Kumar, B.P. Anfis based power quality improvement in cascaded h-bridge multilevel inverter. Int. Res. J. Mod. Eng. Technol. Sci. 2022, 4, 10. [Google Scholar]
- Loaena, Y.G.; Biru, I.G.; Reddy, C.S.; Iqbal, K. Design and Analysis of Front Side Modular Multilevel Converter for Smart Transformer in 15 kV Arba Minch Distribution Network Using Diverse Controllers and Multicarrier Modulation. J. Eng. 2022, 2022, 7678241. [Google Scholar] [CrossRef]
- Potturi, S.; Reddy, K.J.; Dash, R.; Jena, R.; Subburaj, V.; Dhanamjayulu, C.; Sun, X. Direct Torque Control of Induction Motor Using ConvLSTM Based on Gaussian Pillbox Surface. Math. Probl. Eng. 2022, 2022, 4408271. [Google Scholar] [CrossRef]
- Chenchireddy, K.; Jegathesan, V. ANFIS Based Reduce Device Count DSTATCOM. J. Appl. Sci. Eng. 2023, 26, 1657–1666. [Google Scholar] [CrossRef]
- Patil, M.D.; M., P.B. Performance Enhancement of Multi-Level Inverters with PSO-ANFIS Computational Techniques. J. Electr. Syst. 2024, 20, 4263–4272. [Google Scholar]
- Vargas, O.S.; Aldaco, S.E.D.L.; Alquicira, J.A.; Vela-Valdés, L.G.; Núñez, A.R.L. Adaptive Network-Based Fuzzy Inference System (ANFIS) Applied to Inverters: A Survey. IEEE Trans. Power Electron. 2024, 39, 869–884. [Google Scholar] [CrossRef]
- Prasad, T.N.; Devakirubakaran, S.; Muthubalaji, S.; Srinivasan, S.; Karthikeyan, B.; Palanisamy, R.; Bajaj, M.; Zawbaa, H.M.; Kamel, S. Power management in hybrid ANFIS PID based AC–DC microgrids with EHO based cost optimized droop control strategy. Energy Rep. 2022, 8, 15081–15094. [Google Scholar] [CrossRef]
- 519-2022; IEEE Recommended Practice and Requirements for Harmonic Control in Electric Power Systems. IEEE Poer and Energy Society: New York, NY, USA, 2022.
- Jang, J.-S.R. ANFIS: Adap tive-Ne twork-Based Fuzzy Inference System. IEEE Trans. Syst. Man Cybern. 1993, 23, 665–685. [Google Scholar] [CrossRef]
- Sobhani, J.; Najimi, M.; Pourkhorshidi, A.R.; Parhizkar, T. Prediction of the compressive strength of no-slump concrete: A comparative study of regression, neural network and ANFIS models. Constr. Build. Mater. 2010, 24, 709–718. [Google Scholar] [CrossRef]
- Karaboga, D.; Kaya, E. Adaptive network based fuzzy inference system (ANFIS) training approaches: A comprehensive survey. Artif. Intell. Rev. 2018, 52, 2263–2293. [Google Scholar] [CrossRef]
- Jigajinni, V.S.; Upendranath, V. Comparison of ANFIS and ANN Techniques in the Simulation of a Typical Aircraft Fuel System Health Management. Int. J. Artif. Intell. Appl. 2018, 9, 43–55. [Google Scholar] [CrossRef]
- Fausett, L.V. Fundamentals of Neural Networks_ Architectures, Algorithms, and Applications; Prentice-Hall: Upper Saddle River, NJ, USA, 1994. [Google Scholar]
- Ahmed, H.; Glasgow, J. Swarm Intelligence: Concepts, Models and Applications; School of Computing Queen’s University: Kingston, ON, Canada, 2012. [Google Scholar]
- Montes, E.M. Paradigmas emergentes en algoritmos bio-inspirados. In Inteligencia Aritificial; AlfaomegaGrupoEditor, S.A., De, C.V., Eds.; Alfaomega: Mexico City, Mexico, 2006; pp. 504–533. [Google Scholar]
- Gómez Díaz, K.Y.; de León Aldaco, S.E.; Aguayo Alquicira, J.; Vela Valdés, L.G. THD Minimization in a Seven-Level Multilevel Inverter Using the TLBO Algorithm. Eng 2023, 4, 1761–1786. [Google Scholar] [CrossRef]
- Adeyemo, I.A.; Ojo, J.A.; Badajide, D.O. Artificial Intelligence Approach to Real-Time Selective Harmonic Elimination in Voltage Source Multilevel Inverter. Int. J. Soft Comput. Eng. (IJSCE) 2018, 8, 7–13. [Google Scholar]
Ref | Application | Approach | Methods and Results |
---|---|---|---|
[15] | Photovoltaic (PV) | MPPT controller THD minimization | PSO-ANFIS = 29.13%, P&O-ANFIS = 47.54%, ANN = 59.16% |
[16] | PV | Current controller THD minimization | ANN = 5.29%, FUZZY = 1.48%, ANFIS = 1.17% |
[17] | Voltage controller | THD minimization | ANN = 14.57%, NR = 11.17%, PSO = 15.37%, NF = 11.77%, ANFIS = 9.16% |
[18] | Motor | Speed control THD minimization | NR= 5.55%, PSO = 4.79% |
[19] | PV | Variable frequency PWM | PI = 7.67%, FUZZY = 5.03%, ANFIS = 3.62%, PID-PSO = 2.68% |
[20] | THD reduction | Voltage controller THD minimization | ALO-ANN = 12.95%, GA-ANN = 14.91%, PSO-ANFIS = 12.95%, ALO-ANFIS = 14.91% |
[21] | SHAF | Hysteresis controller THD minimization | Without filter = 47.94%, ANN = 2.53%, ANFIS = 2.24% |
[22] | PV | Voltage controller THD minimization | Fuzzy = 3.19%, ANFIS = 3.02% |
[23] | PV | Current controller THD minimization | Without filter = 39.51%, ANFIS = 3.32%, WH ADELINE = 2.28%, MWH ADELINE = 2.20% (RC Load) |
[24] | SAPF | Voltage controller THD minimization | Without control = 44.55%, BSMC = 3.83%, ANFIS-PI = 2.96% |
[25] | PV | Voltage controller THD minimization | PI = 18.19%, PID = 16.11%, ANFIS-EPO = 0.78% |
[26] | PEC9 inverter | Voltage variation THD minimization | ANFIS = 2.2% |
[27] | PV | Voltage and current controller THD minimization | PI = 1.57%, FUZZY = 1.57%, ANFIS = 1.58% |
[28] | Voltage controller | THD minimization | GA + ANFIS = 8.1% |
[29] | ANFIS implementation | Voltage controller and RMS THD minimization | Without control= 4.97%, FUZZY= 4.15%, ANFIS = 3.77% |
[30] | Motor | Speed controller THD minimization | ANFIS + Least squares. 2-levels = 5.78%, 5-levels = 4.04% |
[31] | PV | Voltage controller THD minimization | PI = 8.34%, FLC = 5.65%, ANN = 8.18%, ANFIS = 4.04% |
[32] | Network | THD minimization using LS-PWM | PI = 0.01%. FIS = 0.01%. ANFIS = 0.01% |
[33] | Motor | Torque controller THD minimization | FUZZY = 14.32%, ANFIS = 11.73%, ConvLSTMDTC = 6.61% |
[34] | ANFIS performance | THD minimization with Sliding Mode Controller | Synchronous Reference Frame (SRF) = 3.34%, ANFIS-SRF = 1.56% |
[35] | PV Microgrid | Voltage controller THD minimization | Absence of controller = 1.25% FLC = 4.72% ALO-NN = 1.80% PSO-ANFIS = 1.25% |
Name | Symbol | Quantity |
---|---|---|
Population (number of particles) | NP | 20 |
Generations | GEN | 50 |
Cognitive and social acceleration coefficients | C1 and C2 | 2 |
Inertia coefficient | wmax | 0.9 |
wmin | 0.4 |
Percentage Load % | Cell Voltage | Battery Voltage | Inverter Output Voltage | RMS Voltage |
---|---|---|---|---|
100 | 3.65 | 62.05 | 186.15 | 131.62 |
95 | 3.38 | 57.46 | 172.38 | 121.9 |
50 | 3.276 | 55.692 | 167.076 | 118.14 |
15 | 3.15 | 53.55 | 160.65 | 113.6 |
5 | 2.851 | 48.467 | 145.401 | 102.81 |
Experiment | Voltage of Input Sources | Output Voltage | RMS Voltage | ||
---|---|---|---|---|---|
V1 | V2 | V3 | |||
1 | 62.05 | 62.05 | 62.05 | 186.15 | 131.63 |
2 | 57.46 | 62.05 | 62.05 | 181.56 | 128.38 |
3 | 55.692 | 62.05 | 62.05 | 179.8 | 127.13 |
4 | 53.55 | 62.05 | 62.05 | 177.65 | 125.61 |
5 | 48.467 | 62.05 | 62.05 | 172.56 | 122.02 |
6 | 57.46 | 57.46 | 62.05 | 176.97 | 125.13 |
7 | 55.692 | 57.46 | 62.05 | 175.20 | 123.88 |
8 | 53.55 | 57.46 | 62.05 | 173.06 | 122.37 |
9 | 48.467 | 57.46 | 62.05 | 167.98 | 118.77 |
10 | 55.692 | 55.692 | 62.05 | 173.43 | 122.63 |
11 | 53.55 | 55.692 | 62.05 | 171.29 | 121.12 |
12 | 48.467 | 55.692 | 62.05 | 166.20 | 117.52 |
13 | 53.55 | 53.55 | 62.05 | 169.15 | 119.60 |
14 | 48.467 | 53.55 | 62.05 | 164.06 | 116.01 |
15 | 48.467 | 48.467 | 62.05 | 158.98 | 112.42 |
16 | 57.46 | 57.46 | 57.46 | 172.38 | 121.89 |
17 | 55.692 | 57.46 | 57.46 | 170.61 | 120.64 |
18 | 53.55 | 57.46 | 57.46 | 168.47 | 119.12 |
19 | 48.467 | 57.46 | 57.46 | 163.38 | 115.53 |
20 | 55.692 | 55.692 | 57.46 | 168.84 | 119.39 |
21 | 53.55 | 55.692 | 57.46 | 166.70 | 117.87 |
22 | 48.467 | 55.692 | 57.46 | 161.62 | 114.28 |
23 | 53.55 | 53.55 | 57.46 | 164.56 | 116.36 |
24 | 48.467 | 53.55 | 57.46 | 159.47 | 112.76 |
25 | 48.467 | 48.467 | 57.46 | 154.39 | 109.17 |
26 | 55.692 | 55.692 | 55.692 | 167.07 | 118.14 |
27 | 53.55 | 55.692 | 55.692 | 164.93 | 116.62 |
28 | 48.467 | 55.692 | 55.692 | 159.85 | 113.03 |
29 | 53.55 | 53.55 | 55.692 | 162.79 | 115.11 |
30 | 48.467 | 53.55 | 55.692 | 157.71 | 111.51 |
31 | 48.467 | 48.467 | 55.692 | 152.62 | 107.92 |
32 | 53.55 | 53.55 | 53.55 | 160.65 | 113.59 |
33 | 48.467 | 53.55 | 53.55 | 155.56 | 110.00 |
34 | 48.467 | 48.467 | 53.55 | 150.48 | 106.41 |
35 | 48.467 | 48.467 | 48.467 | 145.40 | 102.81 |
Experiment | Voltage THD% | Angles | ||
---|---|---|---|---|
Ang1 | Ang2 | Ang3 | ||
1 | 10.43 | 8.69° | 27.89° | 49.81° |
2 | 10.49 | 7.870° | 27.30° | 49.56° |
3 | 10.52 | 7.67° | 27.09° | 49.53° |
4 | 10.58 | 7.47° | 26.81° | 49.53° |
5 | 10.76 | 7.06° | 25.69° | 49.69° |
6 | 10.52 | 8.14° | 27.37° | 49.39° |
7 | 10.55 | 7.88° | 27.15° | 49.33° |
8 | 10.60 | 7.63° | 26.87° | 49.28° |
9 | 10.78 | 7.19° | 25.79° | 49.31° |
10 | 10.58 | 7.97° | 27.17° | 49.26 |
11 | 10.63 | 7.71° | 26.89° | 49.19 |
12 | 10.80 | 7.25° | 25.83° | 49.16° |
13 | 10.66 | 7.8° | 26.91° | 49.08° |
14 | 10.84 | 7.31° | 25.88° | 48.97° |
15 | 10.97 | 7.49° | 25.91° | 48.44° |
16 | 10.43 | 8.69° | 27.89° | 49.81° |
17 | 10.45 | 8.31° | 27.62° | 49.67° |
18 | 10.48 | 7.92° | 27.34° | 49.57° |
19 | 10.62 | 7.37° | 26.62° | 49.55° |
20 | 10.46 | 8.45° | 27.67° | 49.62° |
21 | 10.49 | 8.03° | 27.37° | 49.5° |
22 | 10.62 | 7.43° | 26.65° | 49.43° |
23 | 10.51 | 8.18° | 27.41° | 49.42° |
24 | 10.64 | 7.51° | 26.69° | 49.29° |
25 | 10.73 | 7.72° | 26.74° | 48.98° |
26 | 10.43 | 8.69° | 27.89° | 49.81° |
27 | 10.46 | 8.22° | 27.56° | 49.64° |
28 | 10.57 | 7.51° | 26.87° | 49.53° |
29 | 10.47 | 8.39° | 27.61° | 49.58° |
30 | 10.58 | 7.59° | 26.91° | 49.4° |
31 | 10.65 | 7.84° | 26.97° | 49.12° |
32 | 10.43 | 8.69° | 27.89° | 49.81° |
33 | 10.51 | 7.72° | 27.15° | 49.54° |
34 | 10.56 | 8.01° | 27.22° | 49.29° |
35 | 10.43 | 8.69° | 27.89° | 49.81° |
No. Experiment | THD% Training | THD% Simulation | No. Experiment | THD% Training | THD% Simulation |
---|---|---|---|---|---|
1 | 10.43 | 10.46 | 19 | 10.62 | 10.52 |
2 | 10.49 | 10.47 | 20 | 10.46 | 10.51 |
3 | 10.52 | 10.52 | 21 | 10.49 | 10.44 |
4 | 10.58 | 10.5 | 22 | 10.62 | 10.61 |
5 | 10.76 | 10.76 | 23 | 10.51 | 10.49 |
6 | 10.52 | 10.6 | 24 | 10.64 | 10.55 |
7 | 10.55 | 10.46 | 25 | 10.73 | 10.77 |
8 | 10.60 | 10.41 | 26 | 10.43 | 10.46 |
9 | 10.78 | 10.73 | 27 | 10.46 | 10.48 |
10 | 10.58 | 10.53 | 28 | 10.57 | 10.5 |
11 | 10.63 | 10.54 | 29 | 10.47 | 10.41 |
12 | 10.81 | 10.78 | 30 | 10.58 | 10.5 |
13 | 10.66 | 10.64 | 31 | 10.65 | 10.59 |
14 | 10.84 | 10.71 | 32 | 10.43 | 10.46 |
15 | 10.97 | 10.87 | 33 | 10.51 | 10.54 |
16 | 10.43 | 10.46 | 34 | 10.56 | 10.49 |
17 | 10.45 | 10.59 | 35 | 10.43 | 10.46 |
18 | 10.48 | 10.39 |
Method | Ref | Levels | Characteristics | THD Obtained |
---|---|---|---|---|
Metaheuristic | [15] | 5 | There is no mention of angles, 15 harmonics, without a filter | PSO-ANFIS = 29.13% |
[35] | 5 | There is no mention of angles, 14 harmonics, with filter | PSO-ANFIS = 1.25% | |
Proposed technique | 7 | Three angles, 50 harmonics, without filter | ANFIS-PSO ranged = 10.39% to 10.87% | |
[28] | 7 | Seven angles, 50 harmonics, without filter | GA-ANFIS = 8.1% | |
[20] | 9 | Four angles, 15 harmonics, without filter | PSO-ANFIS = 12.95% ALO-ANFIS = 14.91% | |
[25] | 27 | Thirteen angles, 15 harmonics, without filter | ANFIS-EPO = 0.78% | |
Classic | [30] | 2 5 | There is no mention of angles and harmonics, with a filter | ANFIS-Least squares. 2-levels = 5.78% 5-levels = 4.04% |
[17] | 7 | Three angles, 20 harmonics, without filter | ANFIS = 9.16% | |
[27] | 9 | There is no mention of angles, 20 harmonics, with a filter | ANFIS = 1.58% | |
[19] | 15 | There is no mention of angles, 150 harmonics, with a filter | ANFIS = 3.62% | |
[31] | 15 | There is no mention of angles, 20 harmonics, No mention a filter | ANFIS = 4.04% | |
[29] | 31 | There is no mention of angles, 20 harmonics, No mention a filter | ANFIS = 3.77% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Sánchez Vargas, O.; Vela Valdés, L.G.; Borunda, M.; Lozoya-Ponce, R.E.; Aguayo Alquicira, J.; De León Aldaco, S.E. ANFIS-PSO-Based Optimization for THD Reduction in Cascaded Multilevel Inverter UPS Systems. Electronics 2024, 13, 4456. https://doi.org/10.3390/electronics13224456
Sánchez Vargas O, Vela Valdés LG, Borunda M, Lozoya-Ponce RE, Aguayo Alquicira J, De León Aldaco SE. ANFIS-PSO-Based Optimization for THD Reduction in Cascaded Multilevel Inverter UPS Systems. Electronics. 2024; 13(22):4456. https://doi.org/10.3390/electronics13224456
Chicago/Turabian StyleSánchez Vargas, Oscar, Luis Gerardo Vela Valdés, Monica Borunda, Ricardo Eliú Lozoya-Ponce, Jesus Aguayo Alquicira, and Susana Estefany De León Aldaco. 2024. "ANFIS-PSO-Based Optimization for THD Reduction in Cascaded Multilevel Inverter UPS Systems" Electronics 13, no. 22: 4456. https://doi.org/10.3390/electronics13224456
APA StyleSánchez Vargas, O., Vela Valdés, L. G., Borunda, M., Lozoya-Ponce, R. E., Aguayo Alquicira, J., & De León Aldaco, S. E. (2024). ANFIS-PSO-Based Optimization for THD Reduction in Cascaded Multilevel Inverter UPS Systems. Electronics, 13(22), 4456. https://doi.org/10.3390/electronics13224456