Information Technology at Rusal
The article is devoted to the development of information systems "Rusal"
Content |
2024
Rusal allocated 20 billion rubles for import substitution of automation systems
On June 21, 2024, it became known that Rusal, one of the largest aluminum producers in the world, allocated 20 billion rubles for the import substitution program of automation systems at its enterprises. The program is designed until 2031 and is aimed at replacing controller equipment and software at aluminum and alumina plants.
According to the press service of Rusal, the implementation of the program began with the replacement of equipment and software in six technological areas. By the end of 2024, it is planned to launch projects at another 20 sites. As a result of the program, domestic manufacturers of equipment and software for automation of technological processes will be able to significantly increase their competencies and competitiveness.
The representative of the company noted that the cost of the import substitution program may PCS increase as it is implemented. This large-scale project will allow Russian manufacturers of automation systems to invest in expanding production and mastering new technologies.
 | No technological process is now complete without a digital control system, which is essentially the "brain" of a technological installation. As with brain surgery, replacing the control system requires a particularly responsible approach, "said Mikhail Grinishin, Director of Production Automation at the Rusal Engineering and Technological Center. |  |
Russian controller equipment and software have already passed successful tests at the experimental site of the Krasnoyarsk Aluminum Plant. In the near future, the next stage of testing will start at the expanded experimental section of the Sayanogorsk aluminum plant, which was confirmed by the company's website.
The import substitution program initiated by Rusal is an important step in the development of Russian industry. It not only improves the efficiency and reliability of production processes, but also opens up new opportunities for Russian manufacturers in the field of automation.[1]
Rusal has strengthened cyber defense and launched its own SOC
Rusal has created a Security Operation Center (SOC), the company said in its sustainability report published in May 2024. SOC operates on the basis of the Information Security Incident Monitoring and Response Department . Now it receives "several tens of thousands of information security events per second" for processing: they are collected from information protection tools, information systems, computers, servers and other equipment.
In the monitoring process, SOC constantly records incidents in the field of information security, but, Rusal claims, the level of security of the company allows you to avoid incidents of high criticality.
Specialists in information security "Rusal" daily record attempts at cyber attacks of various types:
- Scanning the network perimeter
- phishing;
- DDoS attacks;
- attempts to exploit vulnerabilities;
- viral infection.
The number of such attacks is in the thousands, so Rusal specialists regularly analyze the effectiveness of security systems and take measures to increase the level of cyber resistance.
In 2023, the company was not spared by the growing trend of attacks through supply chains. The sustainability report provides an example of one such prevented attack. Thus, the company's specialists discovered an infection of the counterparty's website, to which its employees regularly referred in correspondence with Rusal. The company acted as follows: temporarily blocked access to the infected resource of the counterparty, notified him of the hack and the need to take action. After the "treatment" and elimination of vulnerabilities, Rusal specialists restored access to the site.
To reduce vulnerability to computer attacks, Rusal has developed business continuity planning (BCP) and DRP (Disaster recovery planning) measures for key systems.
To ensure the uninterrupted functioning of its critical information infrastructure (CII) facilities, Rusal conducted cyber training to respond to computer incidents at such facilities. Within the framework of them, actions were worked out to respond to computer incidents, take measures to eliminate the consequences of computer attacks, and provide information to the NCCC.
In addition, in 2023, at 12 enterprises-subjects of KII "Rusal" created information security groups. They provide information security, including detection, prevention and elimination of the consequences of computer attacks, response to computer incidents. Before the creation of such groups, these functions were performed by information security managers of Rusal enterprises.
Rusal invests 1.6 billion rubles in AI. What technologies does the company use?
In mid-May 2024, Rusal announced the investment of 1.6 billion rubles in the development of artificial intelligence technologies at its enterprises. The company has been applying AI solutions since 2018.
According to the press service of Rusal, the AI project worth 1.6 billion rubles involves the introduction of a system for monitoring electrolysis shops using machine vision at five aluminum plants in the group. The Rusal report emphasizes that the technology was developed by the company's specialists, and it should be the first example of the use of artificial intelligence on an industrial scale in aluminum production.
It is planned to introduce a new AI solution in the electrolysis workshops of the Krasnoyarsk, Bratsk, Novokuznetsk, Irkutsk and Volgograd aluminum plants Rusal until 2027.
The essence of the technology is that models of machine vision through specialized video cameras detect violations of the tightness of electrolyzers around the clock and call the operator. As part of the implementation of the pilot project in 2018-2020. Rusal specialists on hundreds of hours of video trained the neural network to recognize cases of depressurization, for example, smoke, and distinguish them from flashing beacons of technological transport, solar bunnies, reflections, buckets with hot metal, the press service of Rusal quoted Mikhail Grinishin, director of production automation at Rusal ITC.
| | It [technology] proved its effectiveness in the course of experimental operation in two electrolysis workshops of the Krasnoyarsk Aluminum Plant (KrAZ), thanks to it, the time of depressurization of electrolyzers was halved compared to monitoring by personnel during scheduled rounds, - said Viktor Mann, technical director of Rusal. | |
According to Mikhail Grinishin, after the introduction of AI technology in the workshops of KrAZ, it is planned to begin designing development for BrAZ and other plants, and in 2027 it is planned to complete the replication of this system. In addition, by May 2024, the group began to develop systems based on artificial intelligence in other production processes.
As Rusal told Vedomosti, by May 2024, the company has more than 30 projects using the capabilities of artificial intelligence. Some technologies are already used in industrial production, and one of them has reached the stage of large-scale replication.
In the work on AI projects, specialists use solutions in the field of machine learning operations. We are talking about the MLOps approach, in which, as explained in Rusal, continuous development practices are used. Machine learning models are tested and developed in isolated experimental systems, and then ready-made and validated algorithms are run in a production environment. The company uses a three-circuit option for developing new AI solutions:
- primary circuit - internal circuit. MLOps platform "Rusal" is implemented on the basis of technologies open source with (open source framework) SinaraML;
- secondary circuit - intermediate circuit. An instance of the Rusal MLOps platform with the ability to dynamically manage resources has been deployed at the Yandex Cloud facilities;
- the third contour is the outer contour. Contractors of companies developing digital products with AI independently deploy the SinaraML open source framework at their facilities, develop, train and package AI models for Rusal in microservices.
The company intends to consider other options for using Yandex Cloud services, such as cloud access to high-performance graphics processors (GPUs).[2]
2019: RUSAL implements project to manage aluminium production using BigData at Siberian plants
On June 26, 2019, RUSAL, one of the world's largest aluminum producers, announced that as part of a strategy to improve production efficiency, it is developing an electrolysis process control system using modern scientific and methodological approaches based on predictive analytics (BigData).
The development and implementation of the new system will improve the efficiency of aluminum production by reducing the number of deviations of technological parameters from the target values and improving the quality of control of the electrolytic cell. In 2019, more than 20 million rubles will be invested in the project.
In aluminum production, diagnosis of the causes of inefficient operation of electrolyzers is carried out after the fact, while for efficient electrolyzers there is no control over operation in the optimal technological window. These factors are restraining to achieve the best technical and economic indicators, both the electrolysis series and the entire enterprise. In this regard, the Technical Directorate of RUSAL and the Department of Information Technologies initiated a project to develop a system for controlling the technological process of electrolysis using modern scientific and methodological approaches based on predictive analytics (BigData), which will allow diagnosing the causes of a decrease in the performance of electrolyzers, selecting optimal conditions depending on the current state of the equipment and predicting violations.
The pilot project is being implemented at several Siberian plants of RUSAL. The system being developed involves replication without additional costs to other production sites of the company.
| | "The use of forecast analytics resources in the management of electrolysis production is only part of the larger task that we set within the framework of this R&D," said Viktor Mann, Technical Director of RUSAL. "If it is successfully implemented, similar approaches can be used to analyze and improve efficiency at other stages of the production chain and, in general, will contribute to reducing energy consumption and increasing the productivity of aluminum production." | |
2018: Five IT solutions that improve the efficiency of aluminum production
The world's largest aluminum producer Rusal, like other companies in the industry, is interested in improving the quality of products, reducing its cost and minimizing the human factor in production. Rusal's technical director, Victor Mann, told TAdviser in February which ITs in production processes help solve these problems.
Sensor network
At the Rusala plant in Sayanogorsk, a system for measuring current distribution along the anodes was designed. The system was tested at the experimental section of the RA-400 and RA-550 electrolyzers.
| | The system is a large measuring network of 60 sensors on a single electrolytic cell, which provides great opportunities for diagnostics and automatic control of the technology. At the moment, using this system, violations at the anode are determined with an accuracy of 95%, and the development of undesirable processes - anode effects - is also diagnosed and eliminated at an early stage, says Victor Mann. | |
The system was developed as part of the third generation APCS project, as a new stage in the development of the electrolysis production automation system already developed by the metallurgist, the company explained to TAdviser. The project to create a new APCS started in 2015 and was completed in 2016. Its main goals were to improve the accuracy and efficiency of the system, the company added.
Mobile devices
Rusal also introduces mobile devices at production sites. This gives faster access to information and management. Crane operators in electrolysis production use plates to promptly obtain a task for pouring metal or rearranging anode pins from an information and technological system. Thus, the driver himself can switch the control mode of the electrolytic cell to perform the necessary technological operations, Mann said.
| | In the foundry, mobile devices are used by quality service personnel to enter information at control points of different stages of production and on the parameters of finished products. All recorded information immediately enters the unified database for analysis by the system. If some parameters do not meet the requirements, the system will automatically block the transition to the next stage, - said Victor Mann. | |
Mobile devices complement the control system and make information more accessible to all users and from different places. The applied software used has been developed by Rusal specialists and is constantly developing its functionality.
Foundry Management System
In the fall of 2017, Rusal began introducing a unified information and technological foundry system at the company's factories, which has been developed and tested since 2016 at the sites of the Krasnoyarsk Aluminum Plant.
The system will cover all redistributions: from the production of liquid metal from electrolysis production to the shipment of finished products. It will also include planning, quality control and laboratory research operations.
| | Initially, the idea was to create a unified platform that completely eliminates paper document flow, and also includes a system of built-in quality and tracking of basic technological parameters at each stage of production. Thus, the company gets more opportunities to take into account the materials involved and the waste generated, - explained TAdviser in Rusal. | |
The unified electronic reference information base eliminates the possibility of manual entry error. Automatic calculation of the chemical composition of the melt, prompt notifications of the system about deviations in technological parameters, automatic assignment for cutting ingots, etc., make it a convenient tool for personnel and help minimize product defects, says Victor Mann.
"Virtual Electrolyzer"
One of the latest directions of the company's development in the field of information technology was the creation of a mathematical prototype - a digital twin - of the technological process: a dynamic digital model "Virtual Electrolyzer" was created. This software, developed by the ITC (engineering and technical center) of the company together with the Siberian Federal University, allows you to simulate electrolysis processes based on basic mathematical models.
| | It is based on formulas describing fundamental physicochemical processes. At the moment, UC Rusal technologists use it as a platform for optimization, forecasting changes in the technological process, as well as as a training platform. Some modules are used to directly control electrolysis technology and are built into an automated control system, "Mann said. | |
He also noted that computing power is becoming cheaper, which allows the enterprise to develop digital models using a device for processing large amounts of data.
| | The Rusal Data Base been accumulating since 2000. This is an overwhelming array of data for processing by standard methods, but for analyzing and building a mathematical model using Big Data technologies - just right, the company noted. | |
In the future, Rusal is also considering the possibility of implementing a project to build predictive models using big data.
In the global trend
The projects implemented by Rusal correspond to the main task facing the global aluminum industry today: reducing electricity consumption per ton of metal, said Maxim Khudalov, director of the ACRA corporate ratings group.
One of the suppliers of equipment for the aluminum industry Siemens back in 2004 formulated ideas for such projects, American and the aluminum manufacturer Alcoa in 2014 began to actively introduce information technologies to control the production process, Khudalov cites.
2017: Construction of a data center for cryptocurrency mining
On December 1, 2017, it became known about Rusal's plans to build a data processing center (DPC), which, among other things, will be engaged in the extraction of cryptocurrencies. The object is planned to be erected on the territory of the Nadvoitsky plant located in Karelia.
As the Vedomosti newspaper writes with reference to the Rusal message, the company began to design a data center that will provide high-tech IT services for big data analysis, machine learning and cryptocurrency mining.
The electrical power consumption of the equipment will be about 6 MW. Investments in the project were not disclosed. According to Evgeny Grankin, senior analyst at the Department of Strategic Development and Economic Forecasting at Gazprombank, we can talk about an amount of $150 million, which corresponds to 26% of the total capital expenditures of the aluminum holding for 2016.
The first stage of the data center for 5 thousand computing units will be launched by the third quarter of 2018. The first partner of the holding will be the Bulgarian company with Russian roots AA! LAB.
Electricity to the enterprise will be supplied from the Onda hydroelectric station, which since 2014 has also been owned by UC Rusal. By the end of 2017, the capacity of the hydroelectric power station was loaded by no more than 60%, estimates ACRA analyst Maxim Khudalov.
Evgeny Grankin says that the use of a data center mainly for mining cryptocurrencies makes the profitability of the project very unpredictable. The payback period for equipment in home mining is about 9-12 months, said Grankin.
| | Industrial mining is more profitable due to cheaper electricity. In addition, large companies that do this are developing their own mining equipment. This reduces costs, the expert adds. | |
Nikolai Sosnovsky, director of metallurgy and mining at Prosperity Capital Management, also considered Rusal's investments risky.[3]
IT passport of Rusal projects
{{# ITProj:: UC Rusal% 28RusAl - Russian Aluminium% 29}}
Notes
| The site content is translated by machine translation software powered by PROMT. The machine-translated articles are not always perfect and may contain errors in vocabulary, syntax or grammar. Read original article If you find inaccuracies or errors in the results of machine translation, please write to editor@tadviser.ru. We will make every effort to correct them as soon as possible. |