Xervon Instandhaltung, a German industrial service provider, is developing digital twins for the predictive maintenance of cooling water systems for customers in the process industry and is cooperating with the SAP consulting company itelligence.

To ensure smooth operation, plants in the process industry rely on cooling systems that provide process-oriented, customized and precise cooling. Xervon Instandhaltung, a subsidiary of Remondis Maintenance & Services, operates complex cooling systems for its customers. These systems consist of cooling towers, a branched network of pipelines and several pumps that supply cooled, demineralized water in the correct quantity and at the required temperature.

In order to ensure the continuous operation of the cooling systems, the industrial service provider also takes on the maintenance and servicing of the cooling systems. Monitoring must be carried out around the clock, as failures must not occur. With purely analog means, this ultimately means that service technicians have to monitor the system on site. The personnel required for this increases the costs of maintenance accordingly. In addition, in the traditional approach, maintenance is carried out at regular intervals and therefore sometimes even when it would not have been necessary for the system to run smoothly.

What IoT technologies bring

The use of IoT and Industry 4.0 technologies promises clear efficiency gains here. However, tailored digitalization strategies not only reduce costs, but also improve the quality of maintenance and thus the service life of cooling systems by implementing predictive measures.

To encourage companies across Germany to develop their own digitalization strategies and translate them into concrete projects, the SAP consulting firm itelligence launched the 'itelligence of Things Initiative'. The aim of this competition was to create lighthouse projects for Industry 4.0 and thus highlight the potential of digital technologies for industrial processes. As a winner of the itelligence of Things initiative, Xervon Instandhaltung is supported by itelligence in the development of a tailor-made digitalization solution for a cooling system in the process industry. A central element of the resulting project between Xervon Instandhaltung and itelligence is the practical testing of a digital twin of a complex cooling system.

This digital twin can best be described as a virtual image of the real system. Its most important advantages are

- Sensor data is continuously sent to the cloud, allowing the digital twin to be quickly and automatically monitored for deviations.

- Wear can be simulated 'at the touch of a button'; operating hours and environmental conditions are simply set - predictive maintenance processes can thus be taken to a new level.

The digital twin enables more efficient control and maintenance of the system. It can be used to optimize units in a digital test run so that electricity costs, wear and unplanned downtime are reduced and maintenance is only carried out when it is really necessary.

Digital twin and real system

Xervon Instandhaltung operates complex cooling systems for its customers, including cooling towers, an intricate network of pipelines and several pumps.

© Xervon

To create a digital twin, technical drawings and specifications such as location, design and agreed performance data must be digitized , but digitizing such data alone is not enough. The digital twin must also be connected to the real world in order to be able to continuously monitor the system. This connection is made via data from sensors. Exactly how many sensors need to be installed and where depends on the use of the digital machine twin.

How complex the creation of a digital twin is always depends on the complexity of the physical original and the parameters that are required to design the digital twin in such a way that it actually reproduces the processes of the original. In the project example, the definition and data collection to specify the parameters took several days. In contrast, the implementation in the system only took a few hours.

The joint project between Xervon Instandhaltung and itelligence is primarily concerned with two application scenarios:

1. operational safety - ensuring sufficient cooling water at the specified temperature.

2. condition monitoring, i.e. the continuous monitoring of the system's operating status.

The pumps that supply the cooling water from the cooling towers to the customer are crucial for both scenarios. Each pump is therefore equipped with three sensors and two more on the gearboxes. These sensors provide data on the pump capacity or flow rate, the speed and torque, the temperature of the pump motor and the energy consumption. In addition to this data, weather data such as outside temperature and humidity are collected, as these also influence the temperature of the cooling water. The picture is rounded off by data from vibration sensors, which measure vibrations at relevant frequencies and different spectra. The vibration data is used to create fault and wear images, and is therefore primarily used for condition monitoring.

Theoretically, the sensor data can be collected in real time. However, the more frequently such measurements are taken, the larger the data volumes and the associated storage requirements. Pre-processing software therefore cleans up the raw data collected at the cooling system. Only then is it sent to the cloud and assigned to the corresponding digital twin. This results in meaningful information that can now be analyzed.