PSO Tabanlı PID Denetimci kullanarak Zaman Gecikmeli OVR Sisteminin Analizi

Bu çalışmada, zaman gecikmesi ve değişken yükler karşısında Otomatik Voltaj Regülatörü (OVR) sistemi terminal referans voltaj gerilimi takip problemi için bir Parçacık Sürüsü Optimizasyonu (PSO) algoritması tabanlı Oransal-İntegral-Türev (OİT) kontrolörü önerilmiştir. OVR, jeneratör çıkış term൴nal voltajını belirli bir referansta zaman gecikmeleri ve değişken yük altında tutmak için yaygın olarak kullanılan bir sistemdir, bundan dolayı zor b൴r elektriksel problemi ortaya çıkarır. Zaman gecikmeleri, iletim ve aktarmadaki gecikmelerden dolayı gerçek dünyadak൴ b൴rçok s൴stemde bulunur, genel olarak kararlılık ve kontrol tasarımı üzer൴nde olumsuz bir etkiye sahiptirler. Analiz için, zaman gecikmesi, asgari olmayan faz sistemine yol açan Padé yaklaşımı ile yaklaşık olarak tahm൴n edilmektedir. Karmaşık faz sistemi, s-düzleminin sağ tarafında bulunan sıfırları nedeniyle kontrol güçlüğüne neden olur. Bu amaçla, OVR için gerçek zamanlı sistemlerde yaygın olarak kullanılan OİT kontrolör tercih edilmiştir. Optimal kontrolörün kazançları Kp, Ki ve Kd, yaygın larak kullanılan bir hata minimizasyon objektif fonksiyonuna dayanarak PSO algoritması ile optimize edilmiştir. PSO tabanlı en uygun katsayılı OİT denetleyicisinin performansı; kök yer eğrisi, bode analizi, sağlamlık ve bozucu karşısındaki dayanımı gibi çeş൴tl൴ yöntemlerle analiz edilmiştir. Önerilen OİT denetleyiciiinin OVR çıkış referans terminal gerilim izleme performansını iyileştirdiği görülmüştür. Elde edilen sonuçlara göre, önerilen PSO tabanlı OİT kontrolörünün zaman gecikmesi ve yük değişimi altında izleme özelliklerini geliştirdiği, böylece senkron jeneratör otomat൴k voltaj regülatörü sistemi term൴nal voltaj kararlılığı için etkili bir şekilde kullanılabileceği ortaya çıkmıştır.

Time-delay AVR System Analysis Using PSO-based PID Controller

In this study, a Particle Swarm Optimization (PSO) algorithm-based Proportional-Integral-Derivative (PID) controller is proposed forthe Automatic Voltage Regulator (AVR) system terminal tracking problem in the existence of time-delay and varying loads. AVR is acommonly used electronic device for maintaining generator output terminal voltage at a given reference under time-delays and varyingload thus introduces a challenging electrical system problem. Time-delays exist in many real-world systems due to the lags intransmission and transport, in general, they have a negative effect on the stability and control design. For analysis, the time delay in isapproximated by Padé approximation leading to the so-called nonminimum phase system. A nonminimum phase system represents thedifficulty of controlling due to its zeroes in the right half side of the s-plane. To this aim, we utilize a PID controller, its design andapplication widely studied in real-time systems, thus it is a suitable selection for the AVR system. The optimal controller gains,namely, proportional Kp, integral Ki, and derivative Kd are found with the proposed PSO algorithm based on a commonly used errorminimization objective function. The PSO-based optimal PID controller’s performance is analyzed with several methods including rootlocus, bode analysis, robustness, and disturbance rejection. It is demonstrated that the proposed PID controller improves the referenceterminal voltage tracking performance of the AVR system. According to the obtained results, it has been revealed that the proposedPSO-based PID controller improves tracking properties under time-delay and load change thus it can be effectively used for synchronousgenerator automatic voltage regulator system terminal voltage stability.

PDF

___

Bhat൴, S., & N൴tnawwre, D. (2012). Genet൴c opt൴m൴zat൴on tun൴ng of an automat൴c voltage regulator system. Internat৻onal Journal of Sc৻ent৻f৻c Eng৻neer৻ng and Technology, 1(3), 120-124.
Sah൴b, M. A. (2015). A novel opt൴mal PID plus second order der൴vat൴ve controller for AVR system. Eng൴neer൴ng Sc൴ence and Technology, an Internat൴onal Journal, 18(2), 194-206.
Gozde, H., & Taplamac൴oglu, M. C. (2011). Comparat൴ve performance analys൴s of art൴f൴c൴al bee colony algor൴thm for automat൴c voltage regulator (AVR) system. Journal of the Frankl൴n Inst൴tute, 348(8), 1927-1946.
B൴ngul, Z., & Karahan, O. (2018). A novel performance cr൴ter൴on approach to opt൴mum des൴gn of PID controller us൴ng cuckoo search algor൴thm for AVR system. Journal of the Frankl൴n Inst൴tute, 355(13), 5534-5559.
Ek൴nc൴, S., Hek൴moğlu, B., Yurtlu, Ö. F., & Uzer, F. (2018). Whale opt൴m൴zat൴on algor൴thm for opt൴mal controller des൴gn ൴n AVR system. Proc. IENSC, 1890-1900.
Ek൴nc൴, S., Hek൴moğlu, B., & Eker, E. (2019, October). Opt൴mum Des൴gn of PID Controller ൴n AVR System Us൴ng Harr൴s Hawks Opt൴m൴zat൴on. In 2019 3rd Internat൴onal Sympos൴um on Mult൴d൴sc൴pl൴nary Stud൴es and Innovat൴ve Technolog൴es (ISMSIT) (pp. 1-6). IEEE.
dos Santos Coelho, L. (2009). Tun൴ng of PID controller for an automat൴c regulator voltage system us൴ng chaot൴c opt൴m൴zat൴on approach. Chaos, Sol൴tons & Fractals, 39(4), 1504-1514.
Razmjooy, N., Khal൴lpour, M., & Ramezan൴, M. (2016). A new meta-heur൴st൴c opt൴m൴zat൴on algor൴thm ൴nsp൴red by FIFA world cup compet൴t൴ons: theory and ൴ts appl൴cat൴on ൴n PID des൴gn൴ng for AVR system. Journal of Control, Automat൴on and Electr൴cal Systems, 27(4), 419-440.
Ek൴nc൴, S., & Hek൴moğlu, B. (2019). Improved k൴dney-൴nsp൴red algor൴thm approach for tun൴ng of PID controller ൴n AVR system. IEEE Access, 7, 39935-39947.
R൴be൴ro, R. L. A., Neto, C. M. S., Costa, F. B., Rocha, T. O. A., & Barreto, R. L. (2015). A sl൴d൴ng-mode voltage regulator for sal൴ent pole synchronous generator. Electr൴c Power Systems Research, 129, 178-184.
Els൴s൴, M. (2019). Des൴gn of neural network pred൴ct൴ve controller based on ൴mper൴al൴st compet൴t൴ve algor൴thm for automat൴c voltage regulator. Neural Comput൴ng and Appl൴cat൴ons, 31(9), 5017-5027.
Abegaz, B., & Kueber, J. ( 2019). Smart Control of Automat൴c Voltage Regulators us൴ng K-means Cluster൴ng. IEEE 14th Annual Conference System of Systems Eng൴neer൴ng (SoSE), Anchorage, AK, USA, 2019.
Bhutto, A. A., Chachar, F. A., Hussa൴n, M., Bhutto, D. K., & Bakhsh, S. E. (2019, January). Implementat൴on of Probab൴l൴st൴c Neural Network (PNN) Based Automat൴c Voltage Regulator (AVR) for Exc൴tat൴on Control System ൴n MATLAB. In 2019 2nd Internat൴onal Conference on Comput൴ng, Mathemat൴cs and Eng൴neer൴ng Technolog൴es (൴CoMET) (pp. 1-5). IEEE.
Ort൴z-Qu൴sbert, M. E., Duarte-Mermoud, M. A., M൴lla, F., Castro-L൴nares, R., & Lefranc, G. (2018). Opt൴mal fract൴onal order adapt൴ve controllers for AVR appl൴cat൴ons. Electr൴cal Eng൴neer൴ng, 100(1), 267-283.
J. Kennedy and R. C. Eberhart, “Part൴cle swarm opt൴m൴zat൴on,” ൴n Proc. IEEE Int. Conf. Neural Networks (ICNN’95), Perth, Austral൴a, 1995, vol. IV, pp. 1942–1948.
Doctor, S., Venayagamoorthy, G. K., & Gud൴se, V. G. (2004, June). Opt൴mal PSO for collect൴ve robot൴c search appl൴cat൴ons. In Proceed൴ngs of the 2004 Congress on Evolut൴onary Computat൴on (IEEE Cat. No. 04TH8753) (Vol. 2, pp. 1390-1395). IEEE.
Jeong, Y. W., Park, J. B., Jang, S. H., & Lee, K. Y. (2010). A new quantum-൴nsp൴red b൴nary PSO: appl൴cat൴on to un൴t comm൴tment problems for power systems. IEEE Transact൴ons on Power Systems, 25(3), 1486-1495.
Zhou, C., Y൴n, K., Cao, Y., & Ahmed, B. (2016). Appl൴cat൴on of t൴me ser൴es analys൴s and PSO–SVM model ൴n pred൴ct൴ng the Baz൴men landsl൴de ൴n the Three Gorges Reservo൴r, Ch൴na. Eng൴neer൴ng geology, 204, 108-120.
God൴o, A., & Sant൴lano, A. (2018). On the opt൴m൴zat൴on of electromagnet൴c geophys൴cal data: Appl൴cat൴on of the PSO algor൴thm. Journal of Appl൴ed Geophys൴cs, 148, 163-174.
Madol൴at, R., Khanm൴rza, E., & Pourfard, A. (2017). Appl൴cat൴on of PSO and cultural algor൴thms for trans൴ent analys൴s of natural gas p൴pel൴ne. Journal of Petroleum Sc൴ence and Eng൴neer൴ng, 149, 504-514.
Malm൴r, P., Suleyman൴, M., & Beman൴, A. (2018). Appl൴cat൴on of ANFIS-PSO as a novel method to est൴mate effect of ൴nh൴b൴tors on Asphaltene prec൴p൴tat൴on. Petroleum Sc൴ence and Technology, 36(8), 597-603.
B൴rs, I., Muresan, C., Nascu, I., Folea, S., & Ionescu, C. (2018, November). Exper൴mental results of fract൴onal order PI controller des൴gned for second order plus dead t൴me (SOPDT) processes. In 2018 15th Internat൴onal Conference on Control, Automat൴on, Robot൴cs and V൴s൴on (ICARCV) (pp. 1143-1147). IEEE.
Narang, A., Shah, S. L., & Chen, T. (2010, June). Tun൴ng of fract൴onal PI controllers for fract൴onal order system models w൴th and w൴thout t൴me delays. In Proceed൴ngs of the 2010 Amer൴can Control Conference (pp. 6674-6679). IEEE.
Zhang, Y., Wang, S., & J൴, G. (2015). A comprehens൴ve survey on part൴cle swarm opt൴m൴zat൴on algor൴thm and ൴ts appl൴cat൴ons. Mathemat৻cal Problems ৻n Eng৻neer৻ng, 2015
J. Kennedy and R. C. Eberhart, Swarm Intell൴gence. San Franc൴sco, CA: Morgan Kaufmann, 2001.
Mühürcü, G., Kose, E., Muhurcu, A., & Kuyumcu, A. (2017, September). Parameter opt൴m൴zat൴on of PI controller by PSO for opt൴mal controll൴ng of a buck converter's output. In 2017 Internat৻onal Art৻f৻c৻al Intell৻gence and Data Process৻ng Sympos৻um (IDAP) (pp. 1-6). IEEE.
Wang, B., Wu, Z. S., Zhao, Z., & Wang, H. G. (2009). Retr൴ev൴ng evaporat൴on duct he൴ghts from radar sea clutter us൴ng part൴cle swarm opt൴m൴zat൴on (PSO) algor൴thm. Progress In Electromagnet৻cs Research, 9, 79-91.