Quasi-Newton BFGS backpropagation

Estimation of speed, armature temperature, and resistance in brushed DC machines using a CFNN based on BFGS BP

Turkish Journal of Electrical Engineering & Computer Sciences, 26(6), 3181-3191, 2018

Hacene MELLAH , Kamel Eddine HEMSAS, Rachid TALEB, Carlo Cecati.​​​​​​​

​​​​​​​

Estimation of speed, armature temperature and resistance in brushed DC machines using a CFNN based on BFGS BP

Hacene MELLAH , Kamel Eddine HEMSAS, Rachid TALEB, Carlo Cecati.

Abstract: In this paper, a sensorless speed and armature resistance and temperature estimator for Brushed (B) DC machines is proposed, based on a Cascade-Forward Neural Network (CFNN) and Quasi-Newton BFGS backpropagation (BP). Since we wish to avoid the use of a thermal sensor, a thermal model is needed to estimate the temperature of the BDC machine. Previous studies propose either non-intelligent estimators which depend on the model, such as the Extended Kalman Filter (EKF) and Luenberger's observer, or estimators which do not estimate the speed, temperature and resistance simultaneously. The proposed method has been verified both by simulation and by comparison with the simulation results available in the literature.

Keywords

 Cascade-forward neural network, parameter estimation, quasi-Newton BFGS, speed estimation, temperature estimation, resistance estimation

References

[1] Welch RHJ, Younkin GW. How temperature affects a servomotor’s electrical and mechanical time constants. In: IEEE 2002 Industry Applications Conference, Conference Record of the 37th IAS Annual Meeting; 13–18 October 2002; Pittsburgh, PA, USA. New York, NY, USA: IEEE. pp. 1041-1046.

[2] Ali SMN, Hanif A, Ahmed Q. Review in thermal effects on the performance of electric motors. In: 2016 International Conference on Intelligent Systems Engineering; 15–17 January 2016; Islamabad, Pakistan. New York, NY, USA: IEEE. pp. 83-88.

[3] Stone GC, Culbert I, Boulter EA, Dhirani H. Electrical insulation for rotating machines. 2nd ed. New York, NY, USA: Wiley, 2004.

[4] De Abreu JPG, Emanuel AE. Induction motor thermal aging caused by voltage distortion and imbalance: loss of useful life and its estimated cost. IEEE T Ind Appl 2002; 38: 12-20.

[5] Grubic S, Aller JM, Lu B, Habetler TG. A survey on testing and monitoring methods for stator insulation systems of low-voltage induction machines focusing on turn insulation problems. IEEE T Ind Electron 2008; 55: 4127-4136.

[6] Zocholl S. Understanding service factor, thermal models, and overloads. In: 59th Annual Conference for Protective Relay Engineers; 4–6 April 2006; College Station, TX, USA. New York, NY, USA: IEEE. pp. 141-143.

[7] Valenzuela MA, Verbakel PV, Rooks JA. Thermal evaluation for applying TEFC induction motors on short-time and intermittent duty cycles. IEEE T Ind Appl 2003; 39: 45-52.

[8] Gnaciński P. Windings temperature and loss of life of an induction machine under voltage unbalance combined with over-or under voltages. IEEE T Energy Conver 2008; 23: 363-371.

[9] Zhang P, Du Y, Dai J, Habetler TG, Lu B. Impaired-cooling-condition detection using DC-signal injection for soft-starter-connected induction motors. IEEE T Ind Electron 2009; 56: 4642-4650.

[10] Bonnett AH, Soukup GC. Cause and analysis of stator and rotor failures in three-phase squirrel-cage induction motors. IEEE T Ind Appl 1992; 28: 921-937.

[11] IEEE. 119-1974, IEEE Recommended Practice for General Principles of Temperature Measurement as Applied to Electrical Apparatus. New York, NY, USA: IEEE 1975.

[12] Yahoui H, Grellet G. Measurement of physical signals in rotating part of electrical machine by means of optical fibre transmission. Measurement 1997; 20: 143-148.

[13] Compton FA. Temperature limits and measurements for rating of D-C machines. Transactions of the American Institute of Electrical Engineers 1943; 62: 780-785.

[14] Bucci G, Landi C. Metrological characterization of a contactless smart thrust and speed sensor for linear induction motor testing. IEEE T Instrum Meas 1996; 45: 493-498.

[15] AIEE. Temperature rise values for D-C machines: An AIEE committee report. Electrical Engineering 1949; 68: 581-581.

[16] Ganchev M, Kral C, Oberguggenberger H, Wolbank T. Sensorless rotor temperature estimation of permanent magnet synchronous motor. In: 37th Annual Conference on IEEE Industrial Electronics Society; 7–10 November 2011; Melbourne, Australia. New York, NY, USA: IEEE. pp. 2018-2023.

[17] Fiorucci E, Bucci G, Ciancetta F, Gallo D, Landi C, Luiso M. Variable speed drive characterization: review of measurement techniques and future trends. Adv Power Electron 2013; 2013: 968671.

[18] Gao Z, Habetler TG, Harley R, Colby R. A sensorless rotor temperature estimator for induction machines based on current harmonic spectral estimation scheme. IEEE T Ind Electron 2008; 45: 407-416.

[19] Gao Z, Habetler TG, Harley RG, Colby RS. A sensorless adaptive stator winding temperature estimator for mains-fed induction machines with continuous-operation periodic duty cycles. IEEE T Ind Appl 2008; 44: 1533-1542.

[20] Acarnley PP, Watson JF. Review of position-sensorless operation of brushless permanent-magnet machines. IEEE T Ind Electron 2006; 53: 352-362.

[21] Nestler H, Sattler PK. On-line-estimation of temperatures in electrical machines by an observer. Electr Mach Power Syst 1993; 21: 39-50.

[22] Pantonial R, Kilantang A, Buenaobra B. Real time thermal estimation of a brushed DC motor by a steady-state Kalman filter algorithm in multi-rate sampling scheme. In: IEEE TENCON 2012 Region 10 Conference; 19–22 November 2012; Cebu, Philippines. New York, NY, USA: IEEE. pp. 1-6.

[23] Zhang W, Gadsden SA, Habibi SR. Nonlinear estimation of stator winding resistance in a brushless DC Motor. In: 2013 American Control Conference; 17–19 June 2013; Washington, DC, USA. New York, NY, USA: IEEE. pp. 4706-4711.

[24] French C, Acarnley P. Control of permanent magnet motor drives using a new position estimation technique. IEEE T Ind Appl 1996; 32: 1089-1097.

[25] Acarnley PP, Al-Tayie JK. Estimation of speed and armature temperature in a brushed DC drive using the extended Kalman filter. IEE Proc Electr Power Appl 1997; 144: 13-19.

[26] Julier SJ, Uhlmann JK. New extension of the Kalman filter to nonlinear systems. In: International Symposium on Aerospace Defense Sensing Simulation and Controls; 21 April 1997; Orlando, FL, USA. Bellingham, Washington, USA: SPIE. pp. 182-193.

[27] Bolognani S, Tubiana L, Zigliotto M. Extended Kalman filter tuning in sensorless PMSM drives. IEEE T Ind Appl 2003; 39: 1741-1747.

[28] Peroutka Z, Šmídl V, Vošmik D. Challenges and limits of extended Kalman filter based sensorless control of permanent magnet synchronous machine drives. In: 13th European Conference on Power Electronics and Applications; 8–10 September 2009; Barcelona, Spain. New York, NY, USA: IEEE. pp. 1-11.

[29] Mellah H, Hemsas KE, Taleb R. Intelligent sensor based Bayesian neural network for combined parameters and states estimation of a brushed dc motor. Int J Adv Comput Sci Appl 2016; 7: 230-235.

[30] Ahmad S, L. Khan. A self-tuning neurofuzzy feedback linearization-based damping control strategy for multiple HVDC links. Turk J Elec Eng & Comp Sci 2017; 25: 913-938.

[31] Narendra KS, Parthasarathy K. Identification and control of dynamical systems using neural networks. IEEE T Neural Networ 1990; 1: 4-27.

[32] Oyedotun O, Khashman A. Iris nevus diagnosis: convolutional neural network and deep belief network. Turk J Elec Eng & Comp Sci 2017; 25: 1106-1115.

[33] Bishop CM. Neural Networks for Pattern Recognition. New York, NY, USA: Oxford University Press, 1995.

[34] Shamsfakhr F, Sadeghibigham B. A neural network approach to navigation of a mobile robot and obstacle avoidance in dynamic and unknown environments. Turk J Elec Eng & Comp Sci 2017; 25: 1629-1642.

[35] Kim BS, Calise AJ. Nonlinear flight control using neural networks. J Guid Control Dynam 1997; 20: 26-33.

[36] Kaastra I, Boyd M. Designing a neural network for forecasting financial and economic time series. Neurocomputing 1996; 10: 215-223.

[37] Bozkurt MR, Yurtay N, Yılmaz Z, Sertkaya C. Comparison of different methods for determining diabetes. Turk J Elec Eng & Comp Sci 2014; 22: 1044-1055.

[38] Özer HM, Özmen A, Şenol H. Bayesian estimation of discrete-time cellular neural network coefficients. Turk J Elec Eng & Comp Sci 2017; 25: 2363-2374.

[39] Afram A, Janabi-Sharifi F, Fung AS, Raahemifar K. Artificial neural network (ANN) based model predictive control (MPC) and optimization of HVAC systems: a state of the art review and case study of a residential HVAC system. Energ Buildings 2017; 141: 96-113.

[40] Kaye J, Gouse SW. Thermal analysis of a small D-C Motor; Part I. Dimensional analysis of combined thermal and electrical processes. Transactions of the American Institute of Electrical Engineers Part III 1956; 75: 1463-1467.

[41] Kaye J, Gouse SW, Elgar EC. Thermal analysis of a small D-C Motor; Part II. Experimental study of steady-state temperature distribution in a D-C Motor with correlations based on dimensional analysis. Transactions of the American Institute of Electrical Engineers Part III 1956; 75: 1468-1486.

[42] Li W, Wu X, Jiao W, Qi G, Liu Y. Modelling of dust removal in rotating packed bed using artificial neural networks (ANN). Appl Therm Eng 2017; 112: 208-213.

[43] Nabipour M, Keshavarz P. Modeling surface tension of pure refrigerants using feed-forward back-propagation neural networks. Int J Refrig 2017; 75: 217-227.

[44] Venkadesan A, Himavathi S, Sedhuraman K, Muthuramalingam A. design and field programmable gate array implementation of cascade neural network based flux estimator for speed estimation in induction motor drives. IET Electr Power Appl 2017; 11: 121-131.

[45] Sun C, He W, Ge W, Chang C. Adaptive neural network control of biped robots. IEEE T Syst Man Cyber: Syst 2017; 47: 315-326.

[46] Saeedi E, Hossain MS, Kong Y. Side-channel information characterisation based on cascade-forward back-propagation neural network. J Electron Test 2016; 32: 345-356.

[47] Taghavifar H, Mardani A, Taghavifar L. A hybridized artificial neural network and imperialist competitive algorithm optimization approach for prediction of soil compaction in soil bin facility. Measurement 2013; 46: 2288-2299.

[48] Chayjan RA, Esna-Ashari M. Modeling isosteric heat of soya bean for desorption energy estimation using neural network approach. Chil J Agr Res 2010; 70: 616-625.

[49] Lashkarbolooki M, Shafipour ZS, Hezave AZ. Trainable cascade-forward back-propagation network modeling of spearmint oil extraction in a packed bed using SC-CO2. J Supercrit Fluid 2013; 73: 108-115.

[50] Pwasong A, Sathasivam S. A new hybrid quadratic regression and cascade forward backpropagation neural network. Neurocomputing 2016; 182: 197-209.

[51] Khaki M, Yusoff I, Islami N, Hussin NH. Artificial neural network technique for modeling of groundwater level in Langat Basin, Malaysia. Sains Malays 2016; 45: 19-28.

[52] Filik UB, Kurban M. A new approach for the short-term load forecasting with auto-regressive and artificial neural network models. Int J Comput Intell Res 2007; 3: 66-71.

[53] AL-Allaf ONA. Cascade-forward vs. function fitting neural network for improving image quality and learning time in image compression system. In: 2012 Proceedings of the World Congress on Engineering; 4–6 July 2012; London, UK. pp. 1172-1178.

[54] Zhou YM, Meng ZJ, Chen XZ, Wu Z. Helicopter engine performance prediction based on cascade-forward process neural network. In: IEEE 2012 Conference on Prognostics and Health Management; 18–21 June 2012; Denver, CO, USA. New York, NY, USA: IEEE. pp. 1-5.

[55] Lo Sciuto G, Cammarata G, Capizzi G, Coco S, Petrone G. Design optimization of solar chimney power plant by finite elements based numerical model and cascade neural networks. In: 2016 International Symposium on Power Electronics, Electrical Drives, Automation and Motion; 22–24 June 2016; Anacapri, Italy. New York, NY, USA: IEEE. pp. 1016-1022.

[56] Hussain W, Hussain F, Hussain O. QOS prediction methods to avoid SLA violation in post-interaction time phase. In: IEEE 2016 11th Conference on Industrial Electronics and Applications; 5–7 June 2016; Hefei, China. New York, NY, USA: IEEE. pp. 32-37.

[57] Serbest K, Bozkurt MR, Eldoğan O. Classification of cardiac arrhythmias with artificial neural networks according to gender differences. J Eng Sci Technol 2015; 10: 1144-1149.

[58] Dao VNP, Vemuri R. A performance comparison of different back propagation neural networks methods in computer network intrusion detection. Differ Equations Dyn Syst 2002; 10: 1-7.

[59] Mukherjee I, Routroy S. Comparing the performance of neural networks developed by using Levenberg-Marquardt and quasi-Newton with the gradient descent algorithm for modelling a multiple response grinding process. Expert Syst Appl 2012; 39: 2397-2407.

[60] Becker S, Le Cun Y. Improving the convergence of back-propagation learning with second order methods. In: Proceedings of the 1988 Connectionist Models Summer School; 17–26 June 1989; Los Angeles, CA, USA. San Mateo, CA, USA: Morgan Kaufmann. pp. 29-37.

[61] Saito K, Nakano R. Partial BFGS update and efficient step-length calculation for three-layer neural networks. Neural Comput 1997; 9: 123-141.

[62] Broyden CG. The convergence of a class of double-rank minimization algorithms. IMA J Appl Math 1970; 6: 222-231.

[63] Fletcher R. A new approach to variable metric algorithms. Comput J 1970; 13: 317-322.

[64] Goldfarb D. A family of variable-metric methods derived by variational means. Math Comput 1970; 24: 23-23.

[65] Shanno DF. Conditioning of quasi-Newton methods for function minimization. Math Comput 1970; 24: 647-647.

[66] Apostolopoulou MS, Sotiropoulos DG, Livieris IE, Pintelas P. A memoryless BFGS neural network training algorithm. In: IEEE 2009 7th International Conference on Industrial Informatics; Cardiff, UK. New York, NY, USA: IEEE. pp. 216-221.

[67] Zhao L, Wang D, Yang Y. The quadratic property of the L-MBFGS methods for training neural networks. In: 2011 International Conference on Mechatronic Science, Electric Engineering and Computer, Electric Engineering and Computer; 19–22 August 2011; Jilin, China. New York, NY, USA: IEEE. pp. 849-852.

[68] Bordes A, Pierre U. SGD-QN: careful Quasi-Newton stochastic gradient descent. J Machine Learn Res 2009; 10: 1737-1754.

[69] Liu DC, Nocedal J. On the limited memory BFGS method for large scale optimization. Math Program 1989; 45: 503-528.

[70] McLoone S, Irwin G. A variable memory quasi-Newton training algorithm. Neural Process Lett 1999; 9: 77-89.

[71] Mokhtari A, Ribeiro A. Global convergence of online limited memory BFGS. J Mach Learn Res 2015; 16: 3151-3181.

[72] IEC. IEC Publication 60034-1, Rotating Electrical Machines. Part 1: Rating and Performance. Geneva, Switzerland: IEC, 2004.