Development of an improved reloading material with increased performance

Wear is one of the toughest problems facing heavy industry. In the mining sector alone, 17% of the energy consumed is used to combat wear, which represents 2.7% of global CO2 emissions.

Welding high alloy consumables onto component surfaces is one of the most common methods of combating wear. These allied consumables are needed to meet stringent cost, safety, performance and environmental impact requirements.

A complex interplay of properties determines wear performance, making it difficult to optimize the cost/benefit ratio of hardfacing materials.

This case study describes the Welding Alloys Group (WAG) process as it applied machine learning to this problem, which resulted in the development of an improved hardfacing material; not only from a cost/benefit and performance point of view, but also from an environmental point of view.

Wear is a very complex phenomenon that poses many challenges in heavy industry. The common and misleading conception is that high hardness guarantees high wear resistance properties. In fact, a complex interplay of chemical and mechanical properties of each material involved in the application defines optimum wear resistance. Composition, toughness, hardness, Young’s modulus, grain size and phase composition all impact wear, while external parameters such as pressure, temperature and humidity also play a role. a critical role.

Image Credit: Intellegens Limited

Finished products with significant performance deviations are the result of random and systematic variations in hardfacing welding wires and procedures. Due to the excessive use of highly polluting Chromium (Cr), environmental costs are also a significant concern, as are increasingly stringent environmental regulations that drive the use of lightweight welding consumables.

The main objective of this project was to take a welding consumable based on cast iron with a high Cr content and, based on the performance measurement of standard abrasion resistant methods, to optimize the cost/benefit ratio. depending on the chemical composition using the Alchemite machine learning toolkit.

Intellegens’ unique deep learning tool, Alchemite™, builds comprehensive models across multiple material properties and compositions from rare experimental data using the power of deep neural networks.

Literature composition and physical property data, as well as historical WAG data, were used to build the model for the hardfacing materials. The measure of wear resistance was weight loss.

Methodology

Step 1:

The most influential performance parameters were set by Alchemite, who suggested the first set of theoretical compositions. Manufacturing has been confirmed and approved by WAG engineers.

2nd step:

Against the predicted values, the formulations were manufactured, tested and validated. This analysis showed acceptable agreement and fell within the calculated uncertainty.

Step 3:

The model was refined by performing additional iterations.

Step 4:

A new formulation has been defined. Lab tests showed comparable performance to existing products, but with a 50% reduction in alloying elements and a price reduction of between 10-15%.

Principle results

  • The predictions were validated experimentally. There was a considerable difference between the compositions selected for validation and the existing materials. These new materials have considerably improved the model.
  • alchemite offered a more cost effective and environmentally friendly alternative to modern hardfacing material.
  • This material has been industrially tested for performance by WAG.

Future opportunities

This result represents a significant step forward for Welding Alloys Group. Successful progress towards the use of advanced computational methods for the design of new welding consumables and the improvement of existing consumables has been demonstrated. This approach is therefore continuing for the other ranges of welding consumables.

About Welding Alloys Group

Welding Alloys Group is strongly committed to building and maintaining close relationships with its customers as a go-to supplier of automated wear protection equipment, advanced welding consumables and engineered wear solutions. It has secured many strong industry partnerships as a complete solution provider – from integrated engineering solutions to consumables and machinery.

About Intelgens

Alchemite™, the unique deep learning engine, was developed by Intellegens to train neural networks with noisy and rare data typical of real-world scientific and business challenges.

First developed at the University of Cambridge, the technology has been used to guide the design of new drugs, develop aerospace alloys and engineer next-generation battery technology. Optimizing products and processes, saving time and cost in discovery and development, and enabling breakthrough insights, the tool is now being deployed to solve various industrial customer problems.

References

  1. Global energy consumption due to friction and wear in the mining industry. International Tribology. Volume 115, November 2017, pages 116-139

This information has been obtained, reviewed and adapted from materials provided by Intellegens Limited.

For more information on this source, please visit Intellegens Limited.

Comments are closed.