15 آذر 1401
علي شانقي

علی شانقی

مرتبه علمی: دانشیار
نشانی:
تحصیلات: دکترای تخصصی / مهندسی موادخوردگی
تلفن:
دانشکده: دانشکده فنی مهندسی

مشخصات پژوهش

عنوان
Investigating the corrosion performance of Ti-6Al-4V biomaterial alloy with hydroxyapatite coating by artificial neural network
نوع پژوهش مقاله چاپ شده
کلیدواژه‌ها
Ti-6Al-4V alloy, Hydroxyapatite, Corrosion, Artificial neural network, Gene expression programming, Response surface methodology
سال
2022
مجله Materials Science and Engineering B-Advanced Functional Solid-State Materials
پژوهشگران علی شانقی

چکیده

As a result of its excellent mechanical characteristics and bio-compatibility, the Ti-6Al-4V alloy offers a wide-ranging applications in the medical industry. However, long-term application of these alloys may lead to a decline in corrosion resistance as well as other serious problems in these alloys. To resolve these problems, it is possible to modify the implant surface to enhance the corrosion and bio-compatibility properties. On the Ti-6Al-4V substrate surface, the TiN coating was applied by the Plasma-Assisted Chemical Vapour Deposition (PACVD) method together with the hydroxyapatite coating (HA) using the sol–gel method. Furthermore, the corrosion performance of bio-active hydroxyapatite layer coating on Ti-6Al-4V implant surface was evaluated through the application of the Artificial Neural Network (ANN) and Gene Expression Programming (GEP) models. The generated models were analysed and compared to a response surface methodology (RSM). The hydroxyapatite sol–gel parameters on Ti-6Al-4V implant materials were used as independent variables (input variables). The Ecorr values computed from potentiodynamic polarization measurements were used as a dependent variable (output variable) in the models. The results of the ANN technique (sigmoid activation function, LM training algorithm and 9 neurons in hidden layer) were in good consistent with the experimental Ecorr values, indicating this model is slightly better than the GEP and RSM models. Finally, the findings demonstrated that the ANN technique is an appropriate and robust tool for quantitative modeling of corrosion resistance, and that it can be applied to a variety of surface engineering applications.