Method for estimating nugget diameter and weld parameters
View Patent ↗A method for estimating weld nugget diameter includes receiving current and voltage inputs, a time input, and a force input. In certain embodiments, the method also uses other weld parameters including type of material, type of stack, stack thickness, applied force, preheat current, RMS value of weld current, weld level, and weld time. The method further determines a resistance curve based on the current and voltage inputs, and time input. It also derives parameters from the resistance curve, and determines a weld nugget diameter estimate based on both resistance parameters.
1. A method for estimating weld nugget diameter, the method comprising:
receiving current and voltage inputs;
receiving a time input;
receiving a force input;
determining a resistance curve based on the current and voltage inputs, the time input and the force input;
deriving weld parameters from the resistance curve; and
determining a weld nugget diameter estimate based on weld parameters derived from the resistance curve.
2. The method of claim 1 further comprising first determining a controller type and control strategy.
3. The method of claim 2 wherein the controller type is selected from the group consisting of alternating current and mid-frequency direct current.
4. The method of claim 2 wherein the control strategy is selected from the group consisting of constant voltage, constant current and constant heat control.
5. The method of claim 2 wherein determining a weld nugget diameter estimate comprises modeling the weld nugget diameter estimate in response to determining a controller type and control strategy.
6. The method of claim 1 wherein the weld parameters derived from the resistance curve are selected from the group consisting of minimum resistance, maximum resistance, time when maximum resistance is attained, the difference between maximum and minimum resistance, the ratio of maximum resistance to initial resistance, area underneath the resistance curve, maximum negative slope of the resistance curve, steady state value of resistance, time require to attain the steady state value of resistance, RMS value of the resistance, and the difference between maximum resistance and steady state value of resistance.
7. The method of claim 1 wherein the weld parameters are selected from the group consisting of type of material, type of stack, stack thickness, weld cap diameter, preheat current, RMS value of weld current, applied force, weld time, weld level, and energy delivered to nugget.
8. The method of claim 1 further comprising estimating the probability of a stick weld by computing a weighted sum of the weld parameters.
9. The method of claim 8 wherein the weld nugget diameter estimate is determined in response to the stick weld probability.
10. The method of claim 1 wherein determining a weld nugget diameter estimate based on parameters derived from the resistance curve comprises a correction factor due to increased cap diameter resulting from the number of welds performed.
11. The method of claim 1 wherein the weld parameters derived from the resistance curve include at least one of minimum resistance, maximum resistance, time when maximum resistance is attained, the difference between the maximum and minimum resistance, the ratio of maximum resistance to initial resistance, steady state value of resistance, time required to attain steady state value of resistance, RMS value of resistance, or the difference between maximum resistance and steady state value of resistance.
12. The method of claim 1 wherein the weld parameters derived from the resistance curve include at least one of type of material, type of stack, stack thickness, weld cap diameter, preheat current, RMS value of weld current, applied force, weld time, weld level, or energy delivered to nugget.
13. A computerized method for using a neural network to estimate weld nugget diameter, the method comprising:
receiving current and voltage inputs;
receiving a time input;
receiving a force input;
determining a resistance curve based on the current and voltage inputs, the time input, and the force input;
using the resistance curve, deriving weld parameters;
using the weld parameters derived from the resistance curve to operate the neural network; and
operating the neural network to provide a nugget diameter output.
14. A computerized method for using a neural network to estimate weld parameters, the method comprising;
receiving a degree of expulsion input;
receiving a nugget diameter input;
receiving a shear strength input;
receiving an indentation input;
using the degree of expulsion input, nugget diameter input, sheer strength input and indentation input to operate the neural network; and
operating the neural network to provide a weld parameter output.
15. The method of claim 14 further comprising first determining a controller type and control strategy.
16. The method of claim 15 wherein operating a neural network comprises:
modeling the neural network in response to the step of determining a controller type and control strategy.