Authors:
Asmaa Jamal Awad,Ahmed Abdulrasool Ahmed,Osamah Abdallatif,DOI NO:
https://doi.org/10.26782/jmcms.2020.07.00005Keywords:
Matlab software, Generator,Artificial Intelligent (AI),Abstract
Frequent failure in production systems is one of the most important problems facing maintenance planners. In this paper, the methodology for estimating failure in an electrical energy production system has been proposed.Consisting of a number of related sub-systems, respectively, failure of any one causes the rest to stop producing.Operating data were collected and the type of failure identified, which was classified into three types (mechanical failure, electrical failure, and control failure). The software (Matlab) was used in generating and training an artificial neural network (ANN) to estimate the type of failure, through the data collected for each sub-system of the unit under study, use 90% of the data for training, 5% for testing, and 5% for valuation. The target matrix was built and trained, with a mean square error (MSE) its(6.54 E-16), and regression (91%), and adopted to estimate the type of future failure for subsequent years(2019),conformance results were for the subsequent year between (82%-87%) for all the subsystems. Using the artificial neural network, failure types were estimated for another subsequent year (2020), the failure ratios were for subsystems for every ten days during the year of estimation, were (33%) for the generator, (22%) for the boiler, (31%) for the turbine, and (13%) for the condenser. High percentages, which can be reduced by taking advantage of the proposed methodology that gave an understanding of the type of failure, the time it occurred, and the location of the failure, by building an overlapping preventive maintenance plan whose application is approved in reducing the failuretimes of the unit under study.The proposed methodology can also be applied to all other systems of different productionRefference:
I. Devika Chhachhiya, Amita Sharma, Manish Gupta “Case Study on Classification of Glass Using Neural Network Tool in MATLAB” International Journal of Computer Applications, 0975 – 8887),(2014).
II. D. Bose, G. Ghosh, K. Mandal, S.P. Sau4 and S. Kunar “Measurement and Evaluation of Reliability, Availability and Maintainability of a Diesel Locomotive Engine” International Journal of Engineering Research and Technology, Volume 6, Number 4,pp. 515-534, 2003.
III. Emilia Sipos, Laura-Nicoleta Ivanciu”Failure Analysis and Prediction Using Neural Networks in the Chip Manufacturing Process “ResearchGate, DOI: 10.1109/ISSE.2017.8000931, May 2017
IV. Erdi Tosun, Ahmet C¸ alık”Failure load prediction of single lap adhesive joints using artificial neural networks”Alexandria Engineering Journal vol. 55, pp1341–1346,2016
V. Farhad Hooshyaripora, Ahmad Tahershamsib, and Kourosh Behzadian”Estimation of Peak Outflow in Dam Failure Using Neural Network Approach under Uncertainty Analysis” Pleiades Publishing, Vol. 42, No. 5, 2015
VI. Gustavo Scalabrini Sampaio, Arnaldo Rabello de Aguiar Vallim Filho,Leilton Santos da Silva and Leandro Augusto da Silva” Prediction of Motor Failure Time Using An Artificial Neural Network” Sensors, 19, 4342; doi:10.3390/s19194342, 2019
VII. Laurene V. Fausett, “Fundamentals of Neural Networks: Architecture, Algorithm, and Application”, Florida Institute of Technology, First Edition, December, 1993.
VIII. Mahdi Saghafi , Mohammad B. Ghofrani “Real-time estimation of break sizes during LOCA in nuclear power plants using NARX neural network” Nuclear Engineering and Technology doi.org/10.1016/j.net.2018.11.017
IX. M. Goya-Martinez, “The Emulation of Emotions in Artificial Intelligence,” Emotions, Technology, and Design. pp. 171–186, 2016
X. Walter, E.; Pronzato L., “Identification of Parametric Models from Experimental Data”, London, England: Springer-Verlag, 1997.