Abstract

There are several difficulties found to estimate the excitation current and optimum input parameters of synchronous motors. Heuristic methods are frequently used to weight the problem’s parameters or optimum coefficients. As a result, a neural network model is modified in this study to explore the best parameters and estimate the excitation current of a synchronous motor with minimal prediction errors for both the testing dataset and cross validation. Excitation current variations are affected by four input factors, including load current, power factor, error, and changes in excitation current, when training this model. The experimental results reflect that the proposed neural network predicts the new data set effectively as well as enables to predict best weighted value for optimum excitation current of synchronous motors.

Keywords: Synchronous machine, exciting current, neural network, prediction

INTRODUCTION In this study, synchronous machine saturation is modelled using an innovative feedforward artificial neural network technique. The modelling procedure considers the motor locations, excitation levels, and machine loading circumstances [1]. By adjusting the synchronous motor’s available excitation current, the power system can provide the best answer for the need for reactive power. The excitation current estimate issue for synchronous motors is addressed in this study using a successful implementation of Neural Network [2]. In order to achieve quick reaction and high accuracy performances as well as to ensure the system’s tolerance to external disturbance and parameter uncertainty, this article suggests a unique decoupling strategy for a bearing less permanent-magnet synchronous motor. The suggested control strategy uses internal model controllers with two degrees of freedom and the neural network inverse methodology [3]. To maintain the smooth and high-quality functioning of the synchronous machine itself, it is crucial to continually monitor any potential value changes in the excitation current, a crucial parameter of the synchronous machine [4]. The excitation current of synchronous motors may be modeled simply using this paper adaptive artificial neural network-based technique. The network layout is straightforward, and there are fewer processing units (nodes) than in a traditional ANN. This methodology is designed to increase the effectiveness of the traditional ANN-based approach while estimating the excitation current, helping architects easily model excitation current, and helping them create complicated driver software with little programming work [5]. A unique index is presented in this paper for the identification of static as well as dynamic eccentricity faults in permanent magnet synchronous motors. The classification of the findings shows that the proposed index may be used to properly identify the kind, classify, and predict the degree of eccentricity [6].

Keyworde: Synchronous machine, exciting current, neural network, prediction

Full Text PDF Enter a number greater than 18 Submit

Refrences:

1. Tsai H, Keyhani A, Demcko JA, Selin DA. Development of a neural network based saturation model for synchronous generator analysis. IEEE Trans Energy Convers. 1995; 10(4): 617–624.
2. Kaplan O, Elik E. Simplified model and genetic algorithm based simulated annealing approach for excitation current estimation of synchronous motor. Adv Electr Comput Eng. 2018; 18(4): 75–84.
3. Sun X, Chen L, Jiang H, Yang Z, Chen J, Zhang W. High-performance control for a bearingless permanent-magnet synchronous motor using neural network inverse scheme plus internal model controllers. IEEE Trans Ind Electron. 2016; 63(6): 3479–3488.
4. Glučina M, Anđelić N, Lorencin I, Car Z. Estimation of Excitation Current of a Synchronous Machine Using Machine Learning Methods. Computers. 2022; 12(1): 1.
5. Bayindir R, Colak I, Sagiroglu S, Kahraman HT. Application of adaptive artificial neural network method to model the excitation currents of synchronous motors. In 2012 IEEE 11th International Conference on Machine Learning and Applications. 2012 Dec; 2: 498–502.
6. Ebrahimi BM, Roshtkhari MJ, Faiz J, Khatami SV. Advanced eccentricity fault recognition in permanent magnet synchronous motors using stator current signature analysis. IEEE Trans Ind Electron. 2013; 61(4): 2041–2052.
7. Kahraman HT, Bayindir RAMAZAN, Sagiroglu S. A new approach to predict the excitation current and parameter weightings of synchronous machines based on genetic algorithm-based k-NN estimator. Energy Convers Manag. 2012; 64: 129–138.
8. Tsai H, Keyhani A, Demcko JA, Selin DA. Development of a neural network-based saturation model for synchronous generator analysis. IEEE Trans Energy Convers. 1995; 10(4): 617–624.
9. Del Angel A, Geurts P, Ernst D, Glavic M, Wehenkel L. Estimation of rotor angles of synchronous machines using artificial neural networks and local PMU-based quantities. Neurocomputing. 2007; 70(16–18): 2668–2678.
10. Guo H, Ding Q, Song Y, Tang H, Wang L, Zhao J. Predicting temperature of permanent magnet synchronous motor based on deep neural network. Energies. 2020; 13(18): 4782.
11. Chow MY, Thomas RJ. Neural network synchronous machine modeling. In IEEE International Symposium on Circuits and Systems. 1989 May; 495–498.
12. Pillutla S, Keyhani A, Kamwa I. Neural network observers for on-line tracking of synchronous generator parameters. IEEE Trans Energy Convers. 1999; 14(1): 23–30.
13. Kron G. Equivalent Circuits of Electric Machinery. New York, NY, USA: Wiley; 1951.
14. Traxler-Samek G, Zickermann R, Schwery A. Cooling airflow, losses, and temperatures in large air-cooled synchronous machines. IEEE Trans Ind Electron. 2009; 57(1): 172–180. [CrossRef]
15. Dong Z, Weili L, Feng Z, Yunpeng H. Numerical calculation of air gap magnetic field of a salient synchronous generator with the consideration of the effect of turn insulation. In Proceedings of the International Conference on Power System Technology, Kunming, China. 2002 Oct 13–17; 2: 774–778.
16. Fernandez D, Hyun D, Park Y, Reigosa DD, Lee SB, Lee DM, Briz F. Permanent magnet temperature estimation in PM synchronous motors using low-cost hall effect sensors. IEEE Trans Ind Appl. 2017; 53(5): 4515–4525. [CrossRef]
17. Sonmez Y, Guvenc U, Kahraman HT, Yilmaz C. A comparative study on novel machine learning algorithms for estimation of energy performance of residential buildings. In 2015 IEEE 3rd International Istanbul Smart Grid Congress and Fair (ICSG). 2015 Apr; 1–7.