It's the driver, it's the sensor

At that time, topics such as conjectural maintenance and hypothetical models based on artificial intelligence (AI) attracted a lot of attention in the field of mechanical engineering design. However, many customers do not have a proper idea of what can be guessed, and are often only concerned with monitoring equipment and processes.

For this reason, at SPS 2019 in Nuremberg, Lenze will be demonstrating how it is possible to intelligently monitor the situation and then provide multifaceted information about the "health" of machines and systems without relying on any other expensive sensor technology.

Situation monitoring and conjecturical maintenance are often seen as synonyms, but in fact they are two very different concepts.

Speculative maintenance is the conjecture of work or work probability, such as whether the probability of a defect in the gearbox will rise to more than 90% in the next 50 working hours. This conjecture can be used to plan the replacement of the gearbox in time before the shortcomings of the machine or plant practice occur.

On the other hand, the prospective phase of situation monitoring promises to paint a more detailed picture of the situation at time through the interpretation of available data. This requires a deeper understanding of equipment and processes in order to generate meaningful information from the "essential" data. Profiling based on machine learning (ML) and AI can help identify anomalies faster.

There's no need to rely on sensor technology

Since the added value of condition monitoring supply does not mean higher hardware costs and does not have to rely on other sensors, it is popular with OEMs. The key to this resolution plan is to extract additional information values from data sources that are already available. Lenze can supply a variety of algorithms for use by guessing and assisting mechanical engineers in translating their process expertise and equipment knowledge into condition monitoring models and then advancing power.

Truss robots

As a supplier of automation systems, Lenze was demonstrating the concept in two different ways at SPS. The first method is based on a model, i.e., by comparing the measured practical values with the values in the mathematical depiction of the expected equipment, assuming that the error exceeds the necessary plan, the device can be considered to have a defect. Anomalies can be detected by detecting changes in the positive values of current and torque, as well as frequency profiling of the practice values.

The other side is based on data. That is, an algorithm learns how the system is operating and how various parameters (e.g., velocity, acceleration, torque, bearing, and current consumption) interact with each other, and can then compare these practical values with the learned depiction to account for errors.

In addition, Lenze simulated collisions on the spindle and increased wear on the belt drive. Whether it is an affirmative increase in the pass value or an anomaly in the frequency profile, the above two cases can be detected by the current and torque values. In both cases, the situation monitoring triggers an alarm and shows the cause on the Kanban board.

In addition to the conceptual differences, the data evaluation methods involved in the above two monitoring methods are also different. Since there is no need for high accounting skills, the evaluation based on the model is generally carried out in the control system. ML and AI profiling for data-based reviews are generally implemented as a cloud.

Lenze's product portfolio provides OEMs with a wide range of options. This includes many different 3D PLCs for model-based condition monitoring. Assuming the powerful C750 cabinet controller, you can also make data-driven reviews locally. In addition, you can use the x500 gateway to get a route to the cloud. When combined with the x4 approach, mechanical engineers can benefit from our turnkey cloud solution program, which covers not only condition monitoring, but also long-distance maintenance of equipment and user-friendly property management services.

conclusion

Lenze's efficient situation monitoring service is based on the interpretation of existing information and does not need to rely on other sensor technology, but rather on the device's own device. With its broad automation portfolio (hardware, software, network and cloud applications) and its expertise in this area, Lenze can provide comprehensive support in data interpretation and help OEMs become data experts for their equipment.

It's the driver, it's the sensor

Sections viewed

Contact

working hours