Abstract
Electrically-assisted manufacturing refers to the direct application of electrical current to a workpiece during a manufacturing process. This assistance results in several benefits such as flow stress reduction, increased elongation, reduced springback, increased diffusion, and increased precipitation control. These effects are also associated with traditional thermal assistance. However, for over half a decade it has been argued whether or not these observed effects are due to electroplasticity, a term which describes effects that cannot be fully explained through resistive heating. Several theories have been proposed as to the mechanism responsible for these purported athermal effects. Conflicting results within literature have enabled this debate over electroplasticity since its discovery in the mid 20th century.
While the effects of electrically-assisted manufacturing are clearly characterized throughout literature, there is a lack of research related to control systems which may be used to take advantage of its effects. Typically, control systems are developed using an empirical approach, requiring extensive testing in order to fully characterize the stress-strain behavior at all conditions. Additionally, current research has primarily focused on reducing flow stresses during electrically-assisted processes without regard for the strength of the material subsequent to forming. Therefore, there is a strong need for a control system which can quickly be deployed for new materials and does not significantly reduce the subsequent strength of the material.
Herein, a novel control approach is developed in which electrical pulses are triggered by a predetermined stress level. This stress value would be set according to the manufacturer’s stamping die strength. Once the material reaches this stress value, current is deployed until a minimum stress level is reached. At that point, the electricity is turned off and the material allowed to cool; at that stage the stress begins to elevate and the cycle continues. This approach does not require extensive pre-testing and is robust to a range of strain rate. This type of implementation can also be adapted to different levels of capability. For example, since the current is controlled by force and not by time, a low-current power supply will stay on for each pulse longer than a power supply with higher capabilities; however, each will achieve a similar effect. This study investigates the effect of several different minimum stress levels and strain rates. The strain rates chosen are relatively similar to common stamping process. This system was experimentally tested using 1018 CR steel. This control approach was found to be a successful method of maintaining a desired stress level.
Adaptive Locomotion Control of a Hexapod Robot via Bio-Inspired Learning
