COMPUTATIONAL MODEL USING ANFIS AND GA : APPLICATION FOR TEXTILE SPINNING PROCESS
The hybrid approach neuro-fuzzy with subtractive clustering and genetic algorithm (ANFIS-GA) technique is developed to model, to simulate and to optimize fibre to yarn spinning process in Textile Industry. Starting with cotton, desired yarn is produced on ring frame. The quality and cost of resulting yarn plays a significant role in determining its end application. The challenging task of any spinner lies in producing a yarn as per customer demand with added cost benefit. ANFIS is developed to predict yarn properties from multi-property fibre. GA searches optimized fiber properties according to customer defined yarn property with less cost. ANFIS acts as a fitness function to GA. Cost and properties are further reduced using mixing of fibre properties using GA. The performance of ANFIS -GA innovative model is superior compared to current manual machine intervention. The present model may be a fine framework for development of similar applications for complex model that require prediction and multi-objective optimization.
Artificial intelligence Adaptive Neuo-Fuzzy Genetic Algorithm Subtractive Clustering Optimization
L. S. Admuthe S.D. Apte
Computer Science & Engg.department Textile and Engineering Institute, Ichalkaranji, India Electronics department Walchand College of Engieering, Sangli, India
国际会议
北京
英文
2667-2771
2009-08-08（万方平台首次上网日期，不代表论文的发表时间）