Thanks for visiting Laser Systems Europe.

You're trying to access an editorial feature that is only available to logged in, registered users of Laser Systems Europe. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Thinking lasers

Share this on social media:

Issue: 

Dr Ben Mills, an EPSRC early career fellow and senior research fellow for the ORC in Southampton, UK, explains how deep learning could be used to enhance laser processing

Advances in lasers now allow the laser-based processing of almost any material, and consequently innovation in this field is becoming heavily focused on making existing processing techniques more precise and efficient. 

Deep learning, a computing paradigm inspired by biological neurons that learns directly from real-world data, has seen a dramatic rise in interest over recent years, due to its capability in solving extremely complex problems. 

The question is: how can these two fields be combined? Our team at the Optoelectronics Research Centre (ORC) at the University of Southampton is addressing this by exploring whether the power of deep learning can be harnessed in order to improve the repeatability, precision, and control of high-precision femtosecond laser machining of objects on the micro- and even nanoscale.

The objective of the research at the ORC is two-fold: Firstly, to explore the application of deep learning for recognising, in real-time, unexpected events – e.g. laser power fluctuations or unexpected debris – whilst also providing process control – e.g. stopping a manufacturing process at exactly the optimal time, even though the process time may be unknown. Secondly, to investigate the application of deep learning for predictive capabilities for laser machining, in order to be able to accurately predict what a sample would look like after machining with any combination of laser machining parameters.

The intention is that these two approaches will be combined, as unexpected events must first be observed before a predictive capability can determine which parameters will need to be changed, in order to compensate for the earlier manufacturing error. The critical point is that this combined capability needs to operate in real-time.

Deep learning, which generally refers to the application of neural networks (NNs), is based on the premise of computers learning how to solve a problem by themselves. This means that equations describing, for example, the interaction of light and matter, or the probabilistic nature of debris production, are not needed, providing a significant advantage – the interaction of laser light with materials, particularly for femtosecond pulses, is extremely complex. Instead, the deep learning approach simply involves the collection of experimental data, e.g. images of laser machined samples for a wide variety of different experimental parameters. The NN is then trained directly and automatically from this experimental data. Once trained, the NN can process input data and provide useful information on a timescale of tens of milliseconds, and hence is applicable for real-time data processing.

An NN can, therefore, be considered as a transfer function, which converts input data to output data. In the case of laser machining, the input data could be, for example, spectral data, camera images and temperature measurements, while the output data could be anything from beam power corrections to 3D predictions of the sample surface after machining.

While NNs have been studied and understood for many decades, the rate of adoption across academia has only been extremely rapid since 2017, due to the available computing power and data – both which are increasing exponentially – having reached a critical level. As such, most previous work in this area, while being academically interesting, has offered limited potential for industrial application. However, the computing power available now is such that accurate and real-time capability can be offered during processing, which is attracting significant industrial interest.

Figure 1: Schematic describing a setup for monitoring and controlling a process via a neural network

The schematic describing our setup for the monitoring and process control via an NN is shown in Figure 1 above. Femtosecond laser pulses were focused down to a spot size of approximately 30 microns, while the sample was continuously imaged by the camera. The camera images were then processed by the NN, which could output a wide range of experimental information, and also control the laser. 

To date, via an NN we are able to determine; physical changes to the material, fluctuations in laser fluence, the number of pulses used for machining, and values quantifying the changes in the shape and position of the laser beam on the sample. This can be done with higher accuracy and speed than the human eye, requiring approximately only 10ms to process each camera image.

The NN was also able to control the laser in order to stop it exactly at the point it would have machined through a random – and hence unknown – thickness of copper, preventing damage to an underlying layer of glass. Such process control could be applied to the laser cleaning of rust, for example, as the thickness of rust at each position would be unknown, while zero laser damage should occur to the underlying surface. The NN was not only able to cease laser machining at exactly the point of complete removal of the copper, but was able to predict the time remaining until task completion. While direct interrogation of the internal workings of an NN is extremely challenging, we suspect that the NN was able to correlate the amount of debris and the appearance of the machined surface, with some measure of the remaining depth of copper left to machine.

Predictive Capability

Finding the optimal combination of laser machining parameters, such as laser power, laser wavelength, beam size and machining time, for a customer design specification, can be a costly and time-consuming process. Typically, a technician will systemically explore all combinations of laser machining parameters. It would therefore be convenient if an NN could instead be used to determine the optimal parameters automatically.

Such an NN would need to comprehend everything from the physical equations describing the interaction of light and matter, heat transfer, and the laws of diffraction, through to the properties of the sample and the distribution and probability of debris and burr, through to the particular nuances of the laser itself. However, this complexity actually isn’t a problem, as the NN demonstrated here was able to learn all this, by itself, exclusively from images of laser machined samples.

Figure 2: Concept showing a neural network using an inputted beam shape to predict the 3D surface map of machined sample.

Although we have currently only used a NN to predict the effect that laser beam shape has on machined samples, our early results have been staggering. Figure 2 above shows the concept of our work – a NN being shown a beam shape, which it then uses to predict how the 3D surface map of a machined sample would look. Figure 3 shows the results of this, and demonstrates how close the predicted 3D surface map is to the actual result. It is important to realise that the NN had never encountered anything like this particular shape during the training process, and therefore the accuracy of the predicted 3D surface demonstrated here would apply to other beam shapes when first show to the NN. The NN was so effective that it is almost impossible to tell which image is the genuine experimental result. Of particular interest are the slightly raised surfaces in the middle of the laser machined regions. The sample material here, nickel, is known to melt and reform when irradiated with femtosecond laser pulses, and hence the NN had also learnt rules equivalent to fluid dynamics.

Figure 3: An inputted beam shape (left), was used to predict the 3D depth profile of a machined sample (centre), which was then observed to be very close to what was experimentally measured (right).

The future

Computing power increases exponentially as we use the technology of today to build the technology of tomorrow, and it is now the graphics processing unit (GPU) paradigm that is driving innovation in deep learning. NNs have demonstrated basic creativity – for example, see AlphaZero and DeepMind – and hence the potential for NNs to discover new approaches for laser machining is probably not that far away – in fact we are already working on this. By devoting an entire network to a specialised task, NNs have already surpassed many specific human capabilities. Comprehending femtosecond laser machining can now also be added to this list. 

Navigation

Navigation

Navigation

Navigation

Navigation

Navigation