Predictive Maintenance of Equipment

The equipment predictive maintenance of Metal Data team makes full use of its accuracy advantages in model prediction, including not only the functions of existing enterprises in digital system, but also the functions of intelligent system based on high-precision model. In order to achieve high accuracy of the model based on machine learning, the first step is to build the model online. In terms of engineering model, we have at least 3 years of working experience in the fields of process, product, equipment and automation; In terms of machine learning, it is based on the advantages of team members in model development, big data and machine learning after obtaining a doctorate in the United States.

The first-line software engineering team has been in Germany for more than 200 years, and has completed the first-line software manufacturing project in the United States.

Predictive maintenance at the digital level

The service time and life limit of each vulnerable equipment part are recorded in the database. When the equipment room is about to reach the practical limit, let the system alarm and prompt the operator to replace this part in time.

At the same time, it also analyzes the use correlation between devices according to big data, and determines that the equipment is affected by other components in the use process; Thus, the life prediction of equipment parts is obtained.

In addition, make use of a series of advantages of our team in the engineering field, such as equipment use, product production and incoming material quality.

Predictive maintenance at the intelligent level

Here, the high-precision model of our team and the technology of machine learning based on online data are used. For decades, our team has always won with the high precision of our model in the competition with European and American counterparts.

Based on the on-line data collected on site and the idea of historical prediction of the future, an engineering model is established for machine learning, so as to judge the life cycle of equipment parts and the whole equipment. For example, at the digital level, the service life of a piece of equipment in normal use may be 2-4 years; However, the equipment parts may be subjected to specific high-intensity continuous use. Based on the judgment of online data, the service life of equipment parts is different in different service environments, which may be 1-3 years; A specific piece has experienced several different service environments. Based on the correctly established model, it may be judged that its service life is 1.5 years! (the service life of the other part may be 4.5 years in a relatively easy service environment!)

There may be remote logins on site, but many enterprises may close remote logins. Therefore, the main data source may be the data collected by the enterprise from MES or industrial Internet, or other data based on SCADA data acquisition system.

A large number of projects completed by our team are equipment related projects. High accuracy of models based on engineering modeling, machine learning and intelligent system architecture development. For example, in the contract with a customer, it is required to achieve a hit rate of 85%; The actual acceptance reached 98%, and even some acceptance reached 100%! The defect early warning system of our team is one of the key projects promoted by Shenzhen quality month. Therefore, the main designer was also invited to give a special lecture on the opening day of Shenzhen quality month. Information about this product can be found in a special introduction article.

Consultation and development of soft sensing system in defect early warning system

Some parameters are difficult to measure directly and can only be obtained by soft measurement. For example, one of the key parameters defining tool quality is tool notch. However, the value of knife notch can only be measured under a high-power (commonly used 1000 times) microscope; However, in the production process, the cutting tool is wrapped in the sliced pole piece, so it is difficult to see the knife notch, and it is more difficult to use a 100x microscope; At the same time, the cutting speed in the production process is about 100 meters per minute, which corresponds to the speed of 100000 meters per minute under a 1000 times microscope, which also makes it almost impossible to directly measure the knife notch at each time in the production process. Based on these two factors, soft sensing technology must be used to determine the value of knife notch at any time.
 

Advance prediction of equipment failure

Based on the idea of historical prediction of the future, before the equipment fails, it can predict when the equipment will fail under the special use environment of the equipment, so it can give an alarm to the equipment maintenance personnel in advance; The prediction accuracy of the model can be taken into account by setting the accuracy coefficient, and so on.

Planning & Consulting

  Biz discuss, Planning, Maturity, Consulting Area, Predict Maint.
  Defect Early Warning, Models, Manufacturing, Li-Battery, Steel

=========
Contact us: Please contact Linkedin in www.Linkedin.com/in/intelli, OR scan the figure below to add Wechat (e.g. myQQfriend); Tel: (+1) 858 898 1288; E-mail BLiMetaverse@Gmail.com. See Profile of the key consultant.