Method and system for online quality monitoring and control of a vibration welding process
A method for monitoring and controlling a vibration welding system includes collecting sensory data during formation of a welded joint using sensors positioned with respect to welding interfaces of a work piece. A host machine extracts a feature set from a welding signature collectively defined by the sensory data, compares and correlates the feature set with validated information in a library, and executes a control action(s) when the present feature set insufficiently matches the information. A welding system includes a sonotrode, sensors, and the host machine. The host machine is configured to execute the method noted above.
1. A method for monitoring and controlling a vibration welding system having a sonotrode and an anvil, the welding system being operable for forming a welded joint at welding interfaces of a work piece clamped between the sonotrode and the anvil, wherein forming the welded joint is accomplished via a vibration welding process using high-frequency mechanical vibrations, the method comprising:
collecting a set of sensory data during formation of the welded joint using a plurality of sensors, each of which is positioned with respect to the vibration welding system, including measuring a temperature of the welding system, a dynamic resistance between the anvil and the sonotrode, an acoustic signal, and a displacement of a portion of the welding system;
forming a total weld signature using the sensory data, wherein the total weld signature is defined by a plurality of different welding parameters defining the collected set of sensory data;
extracting, via a computer in the form of a host machine, a present feature set from the total weld signature as a combined grouping of a plurality of the welding parameters;
comparing and correlating the present feature set, via the host machine in real time during the formation of the welded joint, with validated information contained in a library to recognize a pattern in the combined grouping of the present feature set; and
executing a control action when the pattern in the extracted present feature set insufficiently matches the information in the library.
2. The method of claim 1 , wherein executing a control action includes changing a control parameter of the vibration welding system.
3. The method of claim 1 , wherein executing a control action includes temporarily stopping the vibration welding process.
4. The method of claim 1 , wherein the plurality of sensors includes a temperature sensor, the anvil includes a wall defining at least one bore, and the temperature sensor is positioned within the bore such that the temperature sensor is embedded within the anvil, and wherein measuring a temperature of the welding system includes using the embedded temperature sensor to measure the temperature.
5. The method of claim 1 , wherein extracting a present feature set from the weld signature includes using at least one of: a neural network, finite element analysis, statistical regression analysis, and principal component analysis.
6. The method of claim 1 , wherein collecting a set of sensory data during the formation of the welded joint includes collecting the set of sensory data while vibration welding a conductive electrode tab of a multi-cell battery to a conductive interconnecting member of the multi-cell battery.
7. A welding system configured for forming a welded joint at welding interfaces of a work piece using a high-frequency mechanical vibration during a vibration welding process, the welding system comprising:
a plurality of sensors positioned with respect to the welding interfaces, wherein the plurality of sensors are operable for collecting a set of sensory data;
an anvil;
a sonotrode positioned adjacent to the anvil and caused to vibrate during the vibration welding process, wherein the set of sensory data includes a temperature of the welding system, a dynamic resistance between the anvil and the sonotrode, an acoustic signal, and a displacement of a portion of the welding system; and
a computer in the form of a host machine having access to a library of validated information;
wherein the host machine is programmed to:
extract a present feature set from a total weld signature formed using the set of sensory data, wherein the present feature set is a combined grouping of a plurality of welding parameters from the set of sensory data;
compare and correlate the extracted present feature set in real time with validated information contained in the library to recognize a pattern in the combined grouping of the present feature set; and
execute a control action when the pattern in the extracted present feature set insufficiently matches the information in the library.
8. The welding system of claim 7 , wherein the host machine is programmed to execute the control action by changing a control parameter of the welding system or by temporarily stopping the vibration welding process.
9. The welding system of claim 7 , wherein the welding system includes an anvil having a wall which defines a bore immediately adjacent to the welding interfaces, and wherein the plurality of sensors includes a temperature sensor positioned at least partially within the bore.
10. The welding system of claim 7 , wherein the host machine is configured to extract the feature set using principal component analysis.
11. The welding system of claim 7 , further comprising a welding anvil having a plurality of segments, wherein at least one of the sensors is positioned between adjacent segments of the plurality of segments.
12. The welding system of claim 7 , wherein the anvil is segmented, and wherein the plurality of sensors includes a temperature sensor positioned between adjacent segments of the segmented anvil.