Views:40 Author:Site Editor Publish Time: 2021-08-05 Origin:Site
Big data analysis: predictive maintenance of equipment
In traditional China custom pcb assembly manufacturers, production equipment still cannot be connected to the Internet, and repairs can only be made after the equipment fails, or regular maintenance is adopted regardless of the actual operation of the equipment. In the event of an unplanned downtime, it is necessary to temporarily purchase parts and spend a high cost for emergency repairs in order to resume normal production as soon as possible. Even if there is no downtime, when people find that the machine is malfunctioning, it may have manufactured substandard products, causing economic losses to the pcba processing plant.
Uptake, an AI industrial prediction platform in the United States, can collect various operational data of front-end equipment by placing sensors in the equipment of the factory, combined with big data analysis and machine learning technology to provide industrial customers with equipment predictive diagnosis and energy efficiency optimization management Suggest. The factory can monitor the operating status in real time, compare historical data, predict potential equipment failures, and effectively avoid the interruption of normal production.
If the equipment predictive maintenance data is integrated into the ERP system in the future, the company can optimize the production process and reduce the economic loss caused by equipment failure to a minimum by dynamically adjusting the production plan. The integration and analysis of different data sources, production equipment and management systems will become the standard configuration for future manufacturing companies to make decisions.
Automatic quality control: machine vision inspection
Before the development of deep neural networks, machine vision has been applied in industrial automation systems, such as pick and place, object tracking, metering, defect detection, etc. Among them, nearly 80% of industrial vision systems focus on defect detection.
The human eye can also detect the abnormality of the product, even if we have never seen such an abnormality. However, because the eyes are prone to fatigue and human judgment is also very subjective, this will cause inconsistent product inspections and even missed inspections. It is also difficult for the human eye to adapt to the needs of high-speed production. For example, for printed circuit boards with complex graphics, manual inspection takes a long time. Usually it can only be based on sampling inspections, and cannot conduct real-time and comprehensive inspections like an automated system. At present, about 60% of inspection tasks on PCB and IC production lines are done by machine vision.
Machine vision is more and more widely used in modern industry by virtue of its advantages such as speed, accuracy and objectivity. The editor of Pater Technology, for example, on the production line, the automatic inspection system can inspect hundreds of components per minute. If equipped with appropriate resolution cameras and optics, the machine can also check the details of features that cannot be seen by the human eye. In addition, because it eliminates direct contact between people and the inspected component, machine vision reduces the cost of component wear and tear, and it also protects workers from dangerous environments.
But machine vision still faces the challenge of adapting to different industrial production environments, because few companies will deploy automated inspection systems specifically for a certain type of product. In different environments, the orientation of the camera lens, the relative position of the camera and the component, and the strong reflected light on the surface of the component will affect the detection accuracy. Therefore, the visual algorithm itself must have strong adaptability.
Intelligent collaborative robot
Because of the fixed movement path of traditional robots, each movement requires engineers to program, debug and manually configure to adapt to the specific production environment. When the robot has to deal with changing scenes, manual adjustment is useless. Deep learning has brought about a revolution, giving robots "flexible" learning capabilities. Over time, the robot can learn from the data, switch between different tasks autonomously, and the import of new tasks can also be completed within a few minutes. In the end, these robots can not only communicate with each other, but also work with humans safely, and even watch workers demonstrate the production process and learn new skills automatically.
At present, high-end industrial robots are mainly dominated by foreign companies. In addition to direct sales of collaborative robots, many companies are also experimenting with new leasing methods, allowing the use of robots to be charged on time like hiring workers. Traditional robots are not safe and need to be isolated from workers, which not only cannot meet the plug-and-play scenario, but also causes additional deployment costs. The emerging business model of Robot as a Service (RaaS, Robot as a Service) lowers the initial payment threshold and emphasizes the software and services other than hardware products. It can be repeatedly reprogrammed to complete new tasks and help companies deal with the production challenges of small batches and multiple orders.