​
Case study: data analysis of crankshaft machining process
Analysis of data from the crankshaft machining process
  1. Home
  2. /
  3. Case Studies
    4. /
  4. Case study: data analysis of crankshaft machining process

Analysis of data from the crankshaft machining process

Goals
Finding the root cause of differences in cutting tool life
Reduce the number of replacements and extend the life of cutting inserts when machining the base surface in the engine block
Results
Early detection of changes in signals indicative of progressive cutting tool wear
Development of preliminary analytical models prepared for implementation of online process monitoring
Benefits of data analysis with RSIMS system
Identification of causes of accelerated cutting tool wear
Reduce downtime associated with emergency tool changes
Reduce quality problems of the workpiece in the period immediately before tool failure
The challenge of cutting tool data analysis
Determination of the reason for differences in cutting tool life. Currently, the observed life of inserts is noticeably lower than expected (it varies in the range of 1000-1500 cycles with an assumed limit of 2000 cycles).

This required the processing of a large data set covering tens of thousands of runs from single cycles recorded with a precision of up to 1 ms, including the extraction of a specific operation from them at a specific time interval, the beginning and end of which are determined by the values of the recorded parameters.
Solution
The developed solution consisted of several key stages. First, we focused on developing a uniform way to process the files. The next step was a detailed statistical analysis of the signals, which included analyzing the data over a long time horizon, developing appropriate transforms to transform the raw signals, and extracting individual features that could help identify the causes of cutting tool degradation. The final element of our work was to build a high-dimensional model to estimate the degree of degradation of the cutting tool.
CASE STUDY - Assumptions and solution development
Solution development included:
- processing of raw process data over a long time horizon,
- identification of the trend of torque increments during the operation of the cutting tool ,
- identification of the trend of time intervals (interruptions in the operation of the device), followed by a relatively rapid increase in the value of moments - measured on the spindle of the cutting tool.

In addition to statistical analysis and development of transforms of raw signals, a multivariate predictive model was developed. This model takes into account the relationships between all monitored process parameters which made it possible to:
- Earlier and reliable detection of the deteriorating condition of the cutting tool, which enables the maintenance team to react faster, plan tool replacement and thus prevent costly quality problems in the workpiece.
- Detection of intervals when the condition of the cutting tool reflected by a sharp increase in torque allows preliminary identification of causes related to the probable bad effect of changes in tool temperature on its service life.
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.
Contact us
We will be happy to answer your questions
Reliability Solutions Sp. z o. o.
ul. Królewska 57, 30-081 Kraków, Polska
icon phone+48 12 394 11 2 icon mailbiuro@reliasol.ai
Piotr Lipnicki
Chief Executive Officer
icon phone+48 605 241 056 icon mailpiotr.lipnicki@reliasol.ai
Dariusz Broda
Head of Engineering
icon phone+48 517 688 108 icon maildariusz.broda@reliasol.ai
Reliability Solutions Sp. z o. o.
ul. Królewska 57, 30-081 Kraków, Polska
+48 12 394 11 2 biuro@reliasol.ai
Published: 05.04.2024
  • Automotive
  • CNC
  • Crankshafts
Other
Volkswagen Motor Poland – robot pneumatic attachment failure detection (case study update)

Volkswagen Motor Poland – robot pneumatic attachment failure detection (case study update)

READ CASE STUDY
Azoty Group – development of an optimization model

Azoty Group – development of an optimization model

See how we developed an optimization model for the reforming plant based on deep neural networks at Grupa Azoty Tarnów.

READ CASE STUDY
Case study: Analysis of recurring problems with vehicle batteries

Case study: Analysis of recurring problems with vehicle batteries

READ CASE STUDY