Last news about IDEKO.

IDEKO shows in China how to achieve more precise and zero-defect high-speed machining

IDEKO shows in China how to achieve more precise and zero-defect high-speed machining
  • The technology center showed its latest advances in the optimization of parts manufacturing processes from 25 to 28 October during the HSM conference, which took place in Nanjing, China.
  • During the event, IDEKO's Dynamics and Control group researcher Xavier Beudaert was part of the scientific committee and presented the center's latest developments in six presentations.

The technology centre IDEKO, a member of the Basque Research and Technology Alliance (BRTA), attended the international conference HSM (High Speed Machining), sponsored by the International Academy of Production Engineering CIRP, which was hold in Nanjing, China, from 25 to 28 October.

For three days, the event became the meeting point of professionals from the world of research and engineering specialized in the field of advanced machining technology to share knowledge and explore the latest advances and developments in this area.

IDEKO, a specialized in advanced manufacturing technologies and a reference in the design and optimization of machining processes, had a prominent presence in the congress through the participation of Xavier Beudaert, researcher of the Dynamics and Control group of the technological center and member of the organizing committee of the event, with his participation in six presentations. Through its interventions it addressed different technological solutions developed by IDEKO to optimize the high-speed machining process ensuring zero defects and more accurate production.

Predictive control techniques

In the first of the six talks, “SIMULATION OF FEEDFORWARD CONTROL TECHNIQUES TO IMPROVE MACHINES FEED DRIVES TRACKING PERFORMANCE”, Beudaert discussed how to improve control of high-speed cutting machines to make them accurate. The presentation also involved representatives of the Technical University of the Czech Republic.

To answer this question, the researcher of the Dynamics and Control group of IDEKO presented the results of the LaserStarLine project, in which a solution based on a simulation model has been developed, which represents in detail the movement control of laser machines, which work at high speed, to test different techniques of feedforward control. These techniques are used to anticipate movement before it occurs and predict any errors to optimize machining accuracy.

“Thanks to this simulation model we can investigate and find ways to make high-speed machines more accurate without the need to perform expensive experiments on the real machine,” said the researcher.

Vibrations in turning parts

For its part, during the second presentation, entitled “ANALYSIS OF THE SURFACE TOPOGRAPHY PATTERNS INDUCED BY A SINGLE FREQUENCY TOOL - WORKPIECE VIBRATION IN TURNING”  Beudaert focused on teaching how vibrations generated in the machine influence during the turning process.

This presentation explained the results obtained in the framework of the InterQ project, whose main objective has been to measure, predict and control the quality of manufactured products, as well as manufacturing processes and data analysis to achieve zero defects production.

Within this initiative, the center has been responsible for analyzing vibrations that arise during turning, both those generated in the machine tool and those produced by contact with the piece, as they cause alterations in the surface of the materials being treated. Moreover, it has been studied that these variations change according to the speed at which the tool vibrates.

In order to guarantee the quality of the pieces, according to IDEKO researcher, “it was studied how vibrations affect the surface of the materials in order to identify patterns that can be used to establish effective machining strategies and ensure an accurate turning”.

Machine learning to optimize production

Fort he third talk, PROCESS CONTROL COMBINING MACHINE LEARNING AND FINGERPRINT APPROACHES”, which is also part of the InterQ project, the researcher presented an innovative machine learning solution developed in collaboration with the Luxembourg company DataThings. All with the aim of guaranteeing the quality of parts after machining developed by cutting machines.

This solution combines two key components. On the one hand, a fingerprint approach that collects data and monitors the manufacturing process in real time while detecting problems that may arise to make a quick decision. On the other hand, it uses a long-term memory algorithm (LSTM), which is responsible for predicting the future of the piece after the manufacturing process.

“The tool takes data from previous productions and uses them to make predictions that can help preserve the quality of the piece being made. This reduces the probability of error and minimizes the loss of time and resources,” said Beudaert.

Realistic visual representation of the piece

We continued in the presentation cycle with the presentation IMPLEMENTING OF PHOTOREALISTIC SURFACE RENDERING IN CAM SIMULATION, related to the InterQ initiative and in which he spoke with representatives of the German company ModuleWorks. During the presentation, Beudaert explained different approaches to observe the behavior of a piece during the whole machining process, such as rendering the information obtained from previous experiences by means of a graphical representation of the results of a CAM simulation (Computer Assisted Manufacturing) of a machined piece.

To achieve a realistic visual representation, it is necessary to take into account both the real movements that the tool performs while machining, and the vibrations produced during the process. Likewise, according to Beudaert, “small details should also be expressed as imperfections or textures that are generated on the material surface and that, in addition, cause light to be reflected in the piece in different ways. This is reflected in the simulation and allows us to achieve a realistic virtual representation".


In the first, Beudaert showed a solution for the aeronautical sector. Specifically, it presented a robotic application for the finishing of complex pieces with complicated shapes.

In the second, the researcher taught a development of the InterQ project on image processing algorithms for digital generation of the shear geometry of broaching tools, a process that takes place in the final phase of the process chain and consists of eliminating material in a linear and progressive way with great precision.