Early detection of storage stacker crane failure

Goals

Prediction and early detection of failure of storage stacker crane drive systems

Expansion of diagnostic metering and enabling remote diagnostics of stacker crane condition

Results

Early detection of malfunctions such as fan and running wheel bearing failure

Constant monitoring and insight into the operation of the stacker crane

Reduce the amount and duration of unforeseen downtime

Benefits of RSIMS data analysis

Faster fault detection and prediction of impending storage stacker crane failures

Cyclical reports and email notifications

Reduce breakdown and service costs

Maintain a high degree of automation in the logistics process

Challenge

The challenge was to monitor the status of the stacker crane drives remotely on an ongoing basis while detecting failures early. One of the main technical problems was the short times of steady-state operation - i.e., the time of stable stacker crane speed. This type of analysis required additional data processing and a different way of determining the alarm threshold compared to the diagnostics of continuously operating equipment. An indirect goal of the project was also to increase warehouse efficiency and reduce unplanned costs directly and indirectly related to the need to repair the device.

Solution

The company has implemented the RSIMS system to accurately predict potential failures of a warehouse stacker crane in a high-bay warehouse. The system consists of a classical diagnostic system based on vibration acceleration and temperature measurements, along with an anomaly detection system.

The developed diagnostic system consisted of a one-dimensional component - trained for each signal separately - and a multidimensional component - trained for groups of signals simultaneously. Each training runs on the basis of a reference period of operation, on the basis of which a complex alarm threshold is determined automatically - using statistical tools or ML algorithms.

The automatic determination of the threshold is based on a proprietary algorithm that analyzes, among other things, the distribution of the data and realizes the calculation of the threshold based on the interrelation of quantiles and medians of this distribution. It takes into account not only the value of the signal itself, but also the number and length of exceedances in a unit of time so as to clearly distinguish small changes in the functioning of the device from potentially dangerous anomalies.

CASE 1 Slave drive

Slave drive - fan fault

Exceedances of the PeakVue parameter (a parameter that allows early detection of bearing and gear wear) were detected on the slave drive gearbox. After verification of ReliaSol's expert report by the UR team, it turned out that the fan was defective.

Slave drive - failure of the running wheel bearing

Thickening exceedances of the PeakVue parameter were observed within the slave drive gearbox and motor. After verification of ReliaSol's expert report by the UR team, it turned out that the anomaly was related to the failure of the slave drive's contact wheel bearing. The entire running wheel module was replaced.

CASE 2 Slave drive - second fan fault

Rapidly exceeding rms vibration accelerations within the slave drive gearbox and motor were reported. Upon verification of the ReliaSol expert's report by the UR team, it was found that the bracket holding the fan on the slave was defective. The detection resulted in a recommendation to replace the fan with a new one.

Summary

In the first few months of operation, the system has already helped detect two faults and allowed the detection of anomalies in the operation of the system. Now, thanks to regular data reviews, the system allows deeper insight into the operation of the stacker crane and distinguishes emergency anomalies from insignificant events. The implementation of the RSIMS predictive system has had a significant impact on reducing the cost of failures and maintenance. The company is now able to determine the risk of failure in real time and make effective and rapid business decisions within the monitored crane.

About company

ReliaSol is a company established in response to the growing need to increase the efficiency of machines and installations in industry.

We provide software and services that accelerate the digital transformation process. The applications we create combine real data from machines, sensors, event reports and automate drawing conclusions. The result of their work is data visualization, event prediction and monitoring of the optimal operating range of machines and entire production lines.

We will be happy to answer your questions

Reliability Solutions Sp. z o. o.
ul. Królewska 57, 30-081 Kraków, Polska