Objective The quality of resistance pressure welding joints, frequently used in a lot of applications (automotive industry, electrical appliances...) is influenced by various factors. Up to now it is impossible to detect during the welding process whether the strength of the welding point is sufficient. Having finished the welding process, even with non-destructive testing using ultrasonic waves or x-rays, good quality cannot definitely be verified. In particularly safety relevant areas random destructive testing is performed and additional "safety points" are welded in order to guarantee that the overall strength of the joint is sufficient. The project aims at guaranteed reliable quality assurance which does not only allow to recognize defective joints during the welding process but also allows to avoid defective joints by on-line interferences (feedback control) during the welding process. Through such a technology it could be expected to reduce drastically the destructive testing, to save time in quality assurance and production (up to 25%) and to improve the welding quality. Well-known methods of artificial intelligence from other areas, especially neural networks shall be employed to evaluate on-line the input data of the welding control system (current, voltage and possibly electrode power) for quality assessment.The quality assurance mechanisms shall be cost effective and capable of being integrated in future control systems without expensive sensor technology. In the case of success there will be a great market in Europe for producers and users of the system and a good opportunity for SMEs in Europe to increase their competitivness. Fields of science engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringcontrol systemsengineering and technologymechanical engineeringvehicle engineeringautomotive engineeringengineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensorsnatural sciencescomputer and information sciencesartificial intelligencecomputational intelligence Programme(s) FP4-BRITE/EURAM 3 - Specific research and technological development programme in the field of industrial and materials technologies, 1994-1998 Topic(s) 0203 - Reliability and quality of materials and products Call for proposal Data not available Funding Scheme EAW - Exploratory awards Coordinator HARMS & WENDE GMBH & CO KG Address Grossmoorkehre 9 Hamburg Germany See on map Links Website Opens in new window EU contribution € 0,00
