Strength modeling and optimizing ultrasonic welded parts of ABS-PMMA using artificial intelligence methods

Abstract

The present work deals with modeling and optimization of ultrasonic welding (USW) process parameters including welding time, pressure, and vibration amplitude influencing strength of the welded parts of acrylonitrile butadiene styrene (ABS) and poly(methyl methacrylate) (PMMA) using artificial intelligence (AI) methods. Experiments performed on samples by spot welding workpieces of ABS and PMMA. The experimental data are used for training of artificial neural networks (ANN), adaptive neuro-fuzzy inference systems, and hybrid systems. It is found that ANN had better predictions compared with the other AI methods. The best model was a feed-forward back-propagation network, with uniform transfer functions (TANSIG-TANSIG-TANSIG) and 4/2 neurons in the first/second hidden layers. The best predictor is then presented to genetic algorithm (GA) and particle swarm optimization (PSO), as the fitness function and for optimizing the USW machine parameters. After the optimization, results of this part revealed that GA and PSO have comparable results and the calculated strength increased by 10%, as compared with a non-optimized case. In order to confirm the computational results, validating experiments are performed which their outputs demonstrates good agreement with the optimization result.

Adaptive Neuro – Fuzzy Inference Systems – An Alternative Forecasting Tool for Prosumers

“…Neuro-fuzzy systems are a combination of NN and fuzzy sets and represent a powerful tool to model systems behavior. The NN is used to define the clustering in the solution space, which results in the creation of fuzzy sets [27] and [28]. A particular architecture of neuro-fuzzy systems is represented by the Adaptive Neuro-Fuzzy Inference System (ANFIS) [29].…”

Poli̇meti̇lmetakri̇lat Malzemeni̇n Mi̇krodalga İle Otojen Kaynaği

“…Ultrases kullanarak polimerlerin kaynaklanması en yaygın tekniklerden biridir, bu teknik lazer kaynak tekniğine göre daha düşük maliyetlidir. Bu nedenle daha yaygın olarak özellikle polimerlerin kaynaklanmasında tercih edilmektedir [16][17][18].…”

Adaptive Neuro-Fuzzy Inference Systems as a Strategy for Predicting and Controling the Energy Produced from Renewable Sources

“…A few papers included graphs of the cumulative distribution of the errors [34]. Others suggested this distribution by reporting the percentage of errors above some critical values, percentiles [35], the maximum errors [36,37]. In any case, no single error measure could possibly be enough to summarize the efficiency of the forecast.…”