Simply stated, online remote artificial intelligence welding consulting and support are your online remote welding engineering department. It uses email, phone, fax, video conferencing, and a welding engineer who have a welding engineering degree, to help artificial intelligence welding robotic and automation system developers, suppliers and users establish and specify welding requirements, answer technical welding questions, and advise on matters of welding.
It’s a convenient, reliable, time and cost saving resource you can use to get the right answers to your welding questions to prevent and solve welding problems. You benefit by staying on schedule, and within budget by avoiding welding mistakes that can cost millions. In addition, you save the time and cost associated with having a welding engineer travel to your location. You also save the cost of maintaining a full-time welding engineering department and staff.
There is a chronic shortage of welding engineers because the science and technology of welding are not generally taught at the four-year university or college level to engineering students, except to those few in a welding engineering degree program. Therefore, most artificial intelligence welding robotic and automation system developers, suppliers and users have no one on their staff with a welding engineering degree. This limits their ability to apply artificial intelligence systems to welding operations.
Artificial intelligence, robotics and automation systems WILL NOT improve welding quality. Only the welding engineer can improve welding quality by selecting the correct welding variables. Because artificial intelligent systems do not have independent thought or decision-making abilities to change welding variables to improve welding quality unless preprogrammed (trained) to do so by the welding engineer. However, these systems have the potential to produce consistent welding quality, depending on the artificial intelligence technique being used, and the welding variables being selected and controlled. To improve welding quality the welding engineer uses the science and technology of welding to select and improve the welding variables. Otherwise, the welding quality will remain the same. For example, the welding engineer may select a better welding process for the application or select a weld joint design to better accommodate the welding process or select a welding filler metal to be more compatible with the base material. These selections will change the completed weldment from a lower quality to a higher quality level. By making these selections the welding engineer has designed-in welding quality, by doing it right the first time to prevent welding defects.
The welding variables determined by the welding engineer are programmed (trained) in the artificial intelligence system for implementation, and control of the welding robotic and automation equipment, which produces consistent quality, productivity, and costs. In other words, the system builds-in the welding quality that was designed-in by the welding engineer.
Under the umbrella of artificial intelligence for welding process control there are various techniques being researched and developed that have advantages and disadvantages depending on the welding process variables being controlled and end results required. For example, the technique of artificial neural networks or deep learning which is a subset of machine learning, and machine learning being a subset of artificial intelligence according to research can be used to discover difficult to recognize patterns, relationships and trends from known welding principles where humans cannot. These abilities of the technique may be useful in helping the welding engineer improve the properties and performance of the weldments.
A heads-up for robotics and automation system developers and suppliers. Let’s not repeat the welding robotic system boom and bust of the mid 1980s when most robotic system developers and suppliers went bankrupt. These bankruptcies were the result of high equipment cost, lack of welding knowledge and experience by welding system suppliers and developers, no welding engineer on staff for support, and the inability to understand the welding manufacturing customers’ requirements.
The bottom line is artificial intelligence techniques can control welding robotic and automation processes, but it has no creativity, imagination or original ideas and thoughts. This insight comes from the welding engineer. Artificial intelligence techniques are another mathematical modeling of welding phenomena tool in the welding engineer’s toolbox that can be used to prevent and solve welding problems.