Predictive maintenance – Early diagnostics for targeted maintenance

Predictive maintenance – Early diagnostics for targeted maintenance

3. December 2020

Predictive maintenance as a key innovation

Digital potential stretches across the entire ecosystem like an immense wave. Intelligent networked industry is leading to a huge level of innovation and change in the everyday work environment of industrial companies. Networking and communication between machine and product as well as between machine and humans is therefore an entirely normal evolution. The maintenance of the future – smart maintenance – aims to ensure that all assets operate smoothly for as long as possible with the minimum use of resources. Predictive maintenance is one of the key innovations in this context.

Predictive maintenance

Predictive maintenance is clearly distinct from conventional maintenance approaches, such as preventive maintenance or reactive maintenance. For example: planning regular maintenance intervals may reduce the risk of failure but at the expense of the productive operating time of the plant! And planning is difficult for reactive maintenance given the spontaneous nature of defects, leading to longer maintenance times.

Predictive maintenance is completely different: anomalies can be forecast based on empirical values gained from the condition monitoring. This enables the logical introduction of needs-based planning of maintenance measures. The availability of assets is maximised plus you receive details for targeted maintenance measures at an early stage.

Condition monitoring – Nuanced cognition

Condition monitoring (CM) is the basis for predictive maintenance. CM continuously or regularly monitors the condition of an asset. In a sense, the procedure is similar to a medical process, such as an electrocardiogram (ECG). This is used to categorise a person’s physical fitness, among other things. While this creates a clinical picture for humans, differences in the target and actual values in assets can, in some cases, point to technical or mechanical defects. An important feature of CM systems is that they work with sensors. These typically record variables, such as temperature, speed, oscillations, vibrations, pressure and other values, which provide an insight into the condition of an asset. The sensor system is therefore part of the condition recording, whose measurement of the relevant data enables the subsequent digital documentation of the measurement results.

Take a look at our free eBook “KI-Lösungen – Optimierung von Assets und Anlagen durch intelligente Analytik“

Artificial intelligence (AI) and IIoT sensors

Predictive maintenance methodically collects asset condition data: this creates the basis for all the subsequent steps. After all, generating a data pool without evaluating the data would be pointless. The recording of the asset condition data in real time is called condition monitoring. It converts the sea of data into a valid condition and provides recommendations for action. The underlying technology: AI and IIoT sensors. A large pool of data on the specific asset is required to be able to make valid forecasts for predictive maintenance. This means that data needs to be recorded (continuously or with certain timestamps), digitalised and transmitted. What happens with this data? How can we make optimal use of the data potential? This data then naturally has to be stored, analysed and measured. Afterwards, a forecast for certain events can be calculated.

IIoT sensors – Sushi sensor

Industrial Internet of Things (IIoT) sensors combine smart technology with user-friendly digital services in a cost-effective device. Thanks to the plug & play installation, the first major step into digital transformation takes just a few minutes. The requirements of IIoT sensors are abundantly clear. The Sushi sensor meets the requirements in every respect:

  • Tough

The Sushi sensor fulfils the fundamental requirements in a harsh plant environment: dust- and water-tight (IP 66/67) as well as explosion-protected.

  • Easy installation (plug & play)

The Sushi sensor uses the LoRaWAN communication standard. LoRaWAN is one of the Low Power Wide Area (LPWA) network protocols that is becoming increasingly popular as a wireless communication system for the IoT (for sensor devices).

  • Battery operated

The Sushi sensor is battery operated and designed for low power consumption. The battery is easy to replace and no cabling work is necessary.

  • Extensive range of Sushi sensors for a variety of measurement parameters:

Wireless Vibration Sensor XS770A
NEW: Temperature sensor XS550 and pressure sensor XS530

Separate modules are combined for the new sensor application. The best thing about the modular system: Wireless communication module XS110A is simply combined with the measurement module (temperature or pressure) for the various measurement purposes – easy installation and removal, without annoying cables.

In addition, the Sushi sensor is aligned to the requirements of M+O sensors (“Monitoring + Optimization”) in line with the NOA (Namur Open Architecture) concept, which are currently being formulated by the NAMUR. In this respect, M+O sensors are used to monitor and optimise the plant components, in contrast to sensors that measure the properties of the process medium. This means that the M+O sensors cover both classic measuring principles as well as novel approaches.

Additional functionalities:

  • Interoperability

Standardised interfaces and uniform data structures for easy access to current data and individual histories.

  • Established and available communication technology

The communication technology impresses with its low energy consumption, large ranges of several kilometres as well as the infrastructure costs. It provides a comprehensive insight into the condition of a process plant.

Predictive maintenance – Quintessence

  • Precise error identification

Clear insights into the status of the assets and well-planned maintenance as well as repairs extend the service life of assets.

  • Perfecting maintenance intervals

Poorly planned maintenance work and the resultant downtimes are costly. Condition monitoring reduces the costs for components and personnel.

  • Minimising downtimes

Productivity is enhanced as the assets are always ready for use.

  • Ensuring compliance

The reporting functions and methods that are consequently available help to ensure compliance with all maintenance-relevant standards, such as the standard ISO55000, without any problems. The result: environmentally compatible operational procedures are ensured.

  • Added safety

Thanks to real-time monitoring and troubleshooting, if problems arise, employees can work in safe conditions.

Without question: the relevance and necessity of predictive maintenance for industry will only increase in future. Are you curious but unsure whether predictive maintenance is right for you?

Take a look at our free eBook “KI-Lösungen – Optimierung von Assets und Anlagen durch intelligente Analytik“.

Find out more about the benefits of AI applications in process automation based on specific practical examples.

We would be pleased to provide more details in a personal consultation. We look forward to hearing from you!


Snapshot Innovation and what is the point of cybersecurity?

Smart Maintenance – Artificial Intelligence (AI) technologies neatly streamline all areas of the value chain in companies

[Total: 0   Average: 0/5]

No Comments so far

Jump into a conversation

No Comments Yet!

You can be the one to start a conversation.

Your data will be safe!Your e-mail address will not be published. Also other data will not be shared with third person.