Makine Öğrenimi Kullanarak Bir Mekanik Jiroskobun Yalpalama Tahmininde Zaman Serisi Modeli

Öz

Jjiroskobik tork üretebilmeleri nedeniyle, mekanik jiroskoplar uçak, uzay araçları gibi tamamen askıdaki veya tek/iki tekerlekli kara araçlarının dengelenmesinde sıklıkla kullanılmaktadır. Yüksek hızla dönen volan sayesinde jiroskobik tork üretmektedirler. Bu torkun kontrolü için yalpalama hızı uygulamak gerekmektedir. Çalışmada bir mekanik jiroskobun rijit dinamik analizi yapılarak 415 adet yalpalama hızı verisi toplanmıştır. Bu hızın açık çevrim tahmininde lineer olmayan, otomatik gerilemeli yapay sinir ağı (NAR) kullanılmıştır. Elde edilen modelde korelasyon değeri 0.998 ve hata karelerinin ortalamasının karekökü (RMSE) değeri de 0.016 rad/s olmuştur. Model çıktısı ile veri seti arasında yüksek doğrusal ilişki tespit edilmiştir. NAR ağı, veri üzerine herhangi bir ön işlem yapılması gereksinimini ortadan kaldırmıştır. Kullanılan ağ parametreleri ve bu model ile elde edilen tahmin performansları çalışma içerisinde sunulmuştur.

Anahtar Kelimeler

• Jiroskop
• Makine öğrenimi
• Otomatik gerilemeli yapay sinir ağı
• Yalpalama

Time Series Model to Forecast the Precession of a Mechanical Gyroscope Using Machine Learning

Öz

Due to the gyroscopic torque production ability, mechanical gyroscopes are frequently used for balancing fully suspended or single/two-wheeled land vehicles such as airplanes and spacecraft. They produce gyroscopic torque thanks to the flywheel rotating at high speed. Precession is required to control this torque. In the study, 415 precession data were collected by performing a rigid dynamic analysis of a mechanical gyroscope. A non-linear autoregressive artificial neural network (NAR) is used to estimate this velocity. In the model obtained, the correlation value was 0.998 and the root mean square of error (RMSE) value was 0.016 rad/s. A high linear relationship was detected between the model output and the data set. The NAR network has eliminated the need for any pre-processing on the data. The network parameters used and the estimation performances obtained with this model are presented in the study.

Anahtar Kelimeler

• Gyroscope
• Machine Learning
• Nonlinear autoregressive network
• Precession

Kaynakça

1. Ahmed, A., Adnaik, I., Bhavsar, D., & Sargar, T. S. (2016). Design and Analysis of Gyro Wheel for Stabilization of a Bicycle. International Journal for Scientific Research & Development, 4(04), 349-351.
2. Amini, G., Salehi, F., & Rasouli, M. (2021). Drying kinetics of basil seed mucilage in an infrared dryer: Application of GA-ANN and ANFIS for the prediction of drying time and moisture ratio. Journal of Food Processing and Preservation, 45(3), e15258. doi: 10.1111/jfpp.15258
3. Amiroh, K., Rahmawati, D., & Wicaksono, A. Y. (2021). Intelligent System for Fall Prediction Based on Accelerometer and Gyroscope of Fatal Injury in Geriatric. Jurnal Nasional Teknik Elektro, 10(3). doi: 10.25077/jnte.v10n3.936.2021
4. Anonimouse. (2023). Precession Wikipedia®. en.wikipedia.org: Wikimedia Foundation, Inc.,.
5. Ansys®. (2023). Academic Research Mechanical Products, 2021 R2, Help System, ANSYS Mechanical User's Guide: ANSYS, Inc.
6. Beigi, M., & Torki, M. (2021). Experimental and ANN modeling study on microwave dried onion slices. Heat and Mass Transfer, 57, 787–796.
7. Çavuşlu, M. A., Becerikli, Y., & Karakuzu, C. (2012). Hardware implementation of neural network training with Levenberg-Marquardt algorithm. TBV Journal of Computer Science and Engineering, 5(1), 1-7.
8. Dash, S., & Venkatasubramanian, V. (2000). Challenges in the industrial applications of fault diagnostic systems. Computers & Chemical Engineering, 24(2-7), 785-791.

9. Dong, L., Wang, J., Tseng, M.-L., Yang, Z., Ma, B., & Li, L.-L. (2020). Gyro Motor State Evaluation and Prediction Using the Extended Hidden Markov Model. Symmetry, 12(11). doi:10.3390/sym12111750
10. Fan, Y., Ding, H., Li, M., & Li, J. (2018). Modal Analysis of a Thick-Disk Rotor with Interference Fit Using Finite Element Method. Mathematical Problems in Engineering, 2018, 5021245. doi: 10.1155/2018/5021245
11. He, Z., Wen, T., Zhang, X., Li, H., Chen, X., & Liu, X. (2022, 25-27 Nov. 2022). Multi-physics Coupling and Thermal Network Analysis of MSCMG. Paper presented at the 2022 China Automation Congress (CAC).
12. Heaton, J. (2008). Introduction to Neural Networks with Java: Heaton Research.
13. Heris, M. K. (2015). Time-series prediction using ANFIS. The Yarpiz Project, Fuzzy Systems.
14. HosseinTabari, Kisi, O., Ezani, A., & Talaee, P. H. (2012). SVM, ANFIS, regression and climate based models for reference evapotranspiration modeling using limited climatic data in a semi-arid highland environment. Journal of Hydrology, 444-445, 78-89. doi: 10.1016/j.jhydrol.2012.04.007
15. Huang, F., Wang, Z., Xing, L., & Gao, C. (2022). A MEMS IMU Gyroscope Calibration Method Based on Deep Learning. Ieee Transactions on Instrumentation and Measurement, 71, 1-9. doi: 10.1109/TIM.2022.3160538
16. Ibrahim, M., Badran, K., & Esmat, A. (2023). Anomaly Detection for Agile Satellite Attitude Control System Using Hybrid Deep-Learning Technique. Aiaa Journal, https://doi.org/10.2514/2511.I011280. doi: 10.2514/1.I011280
17. Jamil, F., & Kim, D. (2019). Improving Accuracy of the Alpha–Beta Filter Algorithm Using an ANN-Based Learning Mechanism in Indoor Navigation System. Sensors, 19, 3946. doi: 10.3390/s19183946
18. Kacar, İ., Eroğlu, M. A., & Yalçın, M. K. (2021). Design and development of an autonomous bicycle. Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi, 10(1), 364-372. doi: 10.28948/ngumuh.628580
19. Kathpalia, N., & Gulati, T. (2022). 3 Axis Gyro Accelerometer & Artificial Intelligence based Enhancement of GPS Accuracy. Measurement: Sensors, 100618. doi: https://doi.org/10.1016/j.measen.2022.100618
20. Kostyuchenko, T., & Indygasheva, N. (2018). Computer-aided design system for control moment gyroscope. MATEC Web Conf., 158, 01021.
21. Kownacki, C. (2011). Optimization approach to adapt Kalman filters for the real-time application of accelerometer and gyroscope signals' filtering. Digital Signal Processing, 21(1), 131-140. doi: 10.1016/j.dsp.2010.09.001
22. Masters, T. (1993). Practical Neural Network Recipes in C++. Elsevier Inc. : Academic Press.
23. Montoya-Chairez, J., Santibanez, V., & Moreno-Valenzuela, J. (2019). Adaptive control schemes applied to a control moment gyroscope of 2 degrees of freedom. Mechatronics, 57, 73-85. doi: 10.1016/j.mechatronics.2018.11.011
24. Muthusamy, V., & Kumar, K. D. (2022). Failure prognosis and remaining useful life prediction of control moment gyroscopes onboard satellites. Advances in Space Research, 69(1), 718-726. doi: https://doi.org/10.1016/j.asr.2021.09.016
25. Nikkhah, A., Heydari, P., Khaloozadeh, H., & Heydari, A. (2009). Gyroscope Random Drift Modeling, Using Neural Networks, Fuzzy Neural and Traditional Time-Series Methods. 6.
26. Nl, C. (2023). Gyroscope physics. Cleonis, 1(1), 1.
27. Nuswantoro, F. M., Sudarsono, A., & Santoso, T. B. (2020, 29-30 Sept. 2020). Abnormal Driving Detection Based on Accelerometer and Gyroscope Sensor on Smartphone using Artificial Neural Network (ANN) Algorithm. Paper presented at the 2020 International Electronics Symposium (IES).
28. Öğündür, G. (2019). Overfitting, underfitting and bias-variance contradiction. Retrieved 12.12.2020, 2020, from https://medium.com
29. Osman, M. O. M., Sankar, S., & Dukkipati, R. V. (1982). Design synthesis of a gyrogrinder using direct search optimization. Mechanism and Machine Theory, 17(1), 33-45. doi: 10.1016/0094-114X(82)90022-2
30. Pan, S., Xu, Z., & Zhao, C. (2019). A novel single-gimbal control moment gyroscope driven by an ultrasonic motor. Advances in Mechanical Engineering, 11(4), 1687814019844382. doi: 10.1177/1687814019844382
31. Papakonstantinou, C., Daramouskas, I., Lappas, V., Moulianitis, V. C., & Kostopoulos, V. (2022). A Machine Learning Approach for Global Steering Control Moment Gyroscope Clusters. Aerospace, 9(3). doi:10.3390/aerospace9030164
32. Rachmatullah, M. I. C., Santoso, J., & Surendro, K. (2020). A Novel Approach in Determining Neural Networks Architecture to Classify Data With Large Number of Attributes. Ieee Access, 8, 204728-204743. doi: 10.1109/ACCESS.2020.3036853
33. Sartori, M. A., & Antsaklis, P. J. (1991). A simple method to derive bounds on the size and to train multilayer neural networks. IEEE Transactions on Neural Networks, 2(4), 467-471. doi: 10.1109/72.88168
34. Shen, L., Zhu, Y., Liu, C., Wang, W., Liu, H., Kamruzzaman, . . . Zheng, X. (2020). Modelling of moving drying process and analysis of drying characteristics for germinated brown rice under continuous microwave drying. Biosystems Engineering, 195, 64-88.
35. Shi, H., Hu, S., & Zhang, J. (2019). LSTM based prediction algorithm and abnormal change detection for temperature in aerospace gyroscope shell. International Journal of Intelligent Computing and Cybernetics, 12(2), 274-291. doi: 10.1108/IJICC-11-2018-0152
36. Sucuoglu, H. S., Bogrekci, I., Gultekin, A., & Demircioglu, P. (2018). Design, Analysis and Development of Mobile Robot with Flip-Flop Motion Ability. IFAC-PapersOnLine, 51(30), 436-440. doi: https://doi.org/10.1016/j.ifacol.2018.11.323
37. Sun, J., Cai, Z., Sun, J., & Jin, D. (2023). Dynamic analysis of a rigid-flexible inflatable space structure coupled with control moment gyroscopes. Nonlinear Dynamics, 111(9), 8061-8081. doi: 10.1007/s11071-023-08254-8
38. Taheri, S., Brodie, G., & Gupta, D. (2021). Optimised ANN and SVR models for online prediction of moisture content and temperature of lentil seeds in a microwave fluidised bed dryer. Computers and Electronics in Agriculture, 182, 106003. doi: 10.1016/j.compag.2021.106003
39. Tamura, S., & Tateishi, M. (1997). Capabilities of a four-layered feedforward neural network: four layers versus three. IEEE Transactions on Neural Networks, 8(2), 251-255. doi: 10.1109/72.557662
40. Tobon-Mejia, D. A., Medjaher, K., Zerhouni, N., & Tripot, G. (2012). A Data-Driven Failure Prognostics Method Based on Mixture of Gaussians Hidden Markov Models. IEEE Transactions on Reliability, 61(2), 491-503. doi: 10.1109/TR.2012.2194177
41. Venkatasubramanian, V., Rengaswamy, R., Kavuri, S. N., & Yin, K. (2003). A review of process fault detection and diagnosis: Part III: Process history based methods. Computers & Chemical Engineering, 27(3), 327-346. doi: 10.1016/S0098-1354(02)00162-X
42. Wang, J. W., Deng, Z. H., & Shen, K. (2022). Virtual Gyros Construction and Evaluation Method Based on BILSTM. Ieee Transactions on Instrumentation and Measurement, 71. doi: 10.1109/TIM.2022.3212544
43. Wisesa, I., & Mahardika, G. (2019). Fall detection algorithm based on accelerometer and gyroscope sensor data using Recurrent Neural Networks. IOP Conference Series: Earth and Environmental Science, 258, 012035. doi: 10.1088/1755-1315/258/1/012035
44. Xiu, T., Yue-dong, L., Xin-xiao, L., & Er-yong, H. (2021). Structural Engineering Analysis for a Control Moment Gyroscope Framework. Journal of Physics: Conference Series, 1939, 012119. doi: 10.1088/1742-6596/1939/1/012119
45. Yang, P., Yang, C., Lanfranchi, V., & Ciravegna, F. (2022). Activity Graph Based Convolutional Neural Network for Human Activity Recognition Using Acceleration and Gyroscope Data. IEEE Transactions on Industrial Informatics, 18(10), 6619-6630. doi: 10.1109/TII.2022.3142315
46. Yang, X., Wu, X., Yu, X., & Basin, M. V. (2023). Closed-Loop Subspace Predictive Control of Gyroscope. Ieee Transactions on Industrial Electronics, 1-10. doi: 10.1109/TIE.2023.3286008
47. Zhou, Z.-J., & Hu, C.-H. (2008). An effective hybrid approach based on grey and ARMA for forecasting gyro drift. Chaos, Solitons & Fractals, 35(3), 525-529. doi: 10.1016/j.chaos.2006.05.039