Networked Condition Monitoring Systems the Intelligent Production Logbook
Companies can detect wear and changes to machines and installations at an early stage by using condition monitoring systems. Actions to avoid production downtimes and quality defects can be taken at an early stage based on the condition monitoring data.
Continuous availability of condition data in monitored systems is one important factor to create the correct tasks in the maintenance solution or to order the right spare parts at an early stage via the commodity management system. When the first condition monitoring solutions were introduced a couple of years ago, the hope was this would finally be an approach for the avoidance of unplanned production downtimes. Machines and installations were equipped with sensors and huge amount of data was recorded.
But the evaluation of the available information raised again and again issues for companies: Which data is important? What do the measured values reveal about the condition of the system? Which values are still tolerable and when do actions have to be taken? Can the set limit values and empirical values be applied to other comparable systems? Often, the technological basis of measurement, the required software solutions and empirical values were lacking for a targeted interpretation of the recorded measured values.
From the Sensor to Maintenance and ERP Systems
Today this basis exists, there are empirical values for many applications and the technology is mastered. There are numerous sensors available for the detection of vibration, pressure, temperatures, forces etc. and data can be easily retrieved from existing controllers.
It is possible to link physical measurements by means of software evaluation rules and to create dependencies. For example, wear is not signalled before the pressure exceeds a certain threshold value, the temperature is in the set range and the vibration values of the bearings have reached a certain limit. In addition to the evaluation IT solutions, interfaces are available today for remote service and maintenance planning even for Enterprise Resource Planning systems (ERP). Companies can for example develop improved maintenance strategies for their machinery pool on the basis of longterm evaluations and also improve the service handling by using automated spare part orders in the ERP system.
The problem is that condition monitoring does not come free of charge.
The equipment of machines with sensors, data acquisition and evaluation requires investments. The results of a system introduction with regards to preventing production downtimes cannot be foreseen and related payback calculations are mostly difficult to perform.
Networked Condition Monitoring System.
On the pulse of the machine: example of trend recording of the monitored diagnosis values in a condition monitoring system.
The Road to Intelligent Condition Monitoring
Meanwhile, experience has shown that it does not make sense to equip all machines and installations with condition monitoring systems, the focus is rather on equipping critical installations whose failure would incur significant cost or considerably affect the product quality. The limit of wear which justifies for example the replacement of a part, is defined on the basis of empirical values.
Today, such limits are mostly fixed and do not adjust according to the changes in the processes in a machine or installation. In this regard, software development needs to optimize the detection of wear in installations by using knowledge databases and for example neural networks, and to create adjusting limit values depending on the changing process parameters. Research and solution providers will probably develop condition monitoring solutions as self-learning systems in the near future. As the increasing computing capacity enables complex calculations, the wear limits can be adapted to the variable process and operating conditions. By setting up additional knowledge databases, the significance and reliability of diagnosis of wear to machines and installations can be considerably increased.
CMS Support for the Machine Operator
In addition to avoiding production downtimes, maintaining quality is an increasingly important issue for condition monitoring solutions particularly in high-precision processes where the product quality today decisively depends on the experience of the machine operator. For example, changes in the operating conditions such as increasing or decreasing temperatures in the factory and changes in raw material qualities may lead to need for readjustment of certain parameters in the machining process.
For a multitude of parameters, it is the machine operator who is solely responsible for deciding if the workpiece to be processed meets the quality requirements. In the future, enhanced evaluation software for condition monitoring systems may support the machine operator in his/her decision.
From the Avoidance of Downtimes to the Optimisation of Quality
Condition monitoring of machines and installations will evolve from the avoidance of production downtimes to quality assurance and quality optimization. The significance and accuracy of prediction of condition monitoring is considerably improved in particular by the variable adjustment of limit values to the production processes. Therefore it can be assumed that condition monitoring will become a standard application particularly in complex machines and installations. The machine manufacturer or installer offers integrated solutions directly connected to IT service systems, for example for remote maintenance and spare part orders.
A Heat Pump Plant Turns the Carbon Footprint of Textile Fibre Production Negative
