Cogniplant, transformation towards an intelligent plant
- Client: Tubacex
- Country: Spain
- Year: 2022
- Applied technologies: Data analytics | Machine learning | Digital Twin | Cloud
Tubacex, a multinational corporation specializing in seamless tube and alloy production, is aiming to convert its factory in Amurrio into a Smart production facility that is sustainable and efficient. The company seeks to significantly reduce energy consumption in furnaces and defects and rejects. The main objective is to fully digitize Aceralava, the Amurrio plant, with the challenge of saving 3.5 GWh per year and reducing CO2 emissions by 6,000 tonnes. To achieve these savings, the company plans to optimize its products including smelting and decarburization furnaces and increase the plant’s process efficiency. This includes improvements in melting, ingot casting, forging, tube production, and addressing quality issues.
At Ibermática, our goal is to take plant digitization to the next level and transform it into an intelligent entity. To achieve this, our team has leveraged advanced analytics, cognitive reasoning, and a revolutionary Digital Twin concept to optimize the operational performance of the Amurrio plant. Through specialized sensors and pre-existing management systems, we’ve collected data from the plant’s six primary processes, which are stored in an internal Datalake. This data is then sent to a cloud platform, serving as a data source for a range of Machine Learning tools utilized in analysis and decision-making processes.
Ayesa’s innovative machine-learning approach to data analysis has aided Acerlava in reducing waste by uncovering hidden relationships between processes and quality issues. The use of machine learning and data from the furnace melting process has resulted in a Digital Twin that maximizes the plant’s energy efficiency and overall effectiveness. This approach streamlines the search for the optimal component melting process while taking into account the required steel grade and casting elements.
Furthermore, we have assisted Aceralava in optimizing the efficiency of its decarburizing furnace by developing machine-learning models that improve energy and gas consumption while providing production optimization recommendations.