Customers: Novolipetsk Steel Mill, NLMK Novolipetsk; Metallurgical industry Contractors: Data Center Automation Product: Data Center Automation: Expert Base Enterprise Analytical SystemSecond product: Data Center Automation: Data-Track Digital Platform Project date: 2020/03 - 2022/08
|
2022
MES (RAS) Roll Lifecycle Visualization Service
The project includes the following MES technologies (according to MES-11): RAS (Resource Allocation and Status) - status control and resource allocation in terms of procurement management of replaceable parts of process equipment.
Monitoring the condition of process equipment is one of the most common tasks in industry. The task of determining the state of units during their regular operation is becoming more and more relevant. Currently, the global problem with the operation of equipment is to maintain its operability by economically optimal measures. Its solution is important for both operational and repair personnel. The first set the mode of operation of the units taking into account its condition: they detect the wear of the equipment during inspection and control, determine the feasibility of its further operation with significant wear. The latter participate in the maintenance of equipment in order to determine the required amount of repair. Replacement of rolling rolls (transshipment) in rolling production is necessary for normal operation of the technological process and prevention of roll breakage.
It is believed that no more than 3.5-5 thousand tons of metal can be rolled before changing the rolls. Of course, you can do more, but there is always a risk that something will go wrong and the production process will slow down for several hours. Considering that all metal passes through the mill, such an incident will lead to at least a shift in the shipping dates of all orders. Since warehouses have a limited capacity, you will have to reduce the smelting of slabs in converter shops, and this will lead to the need to stop the units. Therefore, it is so important to analyze the behavior of rolling rolls in production, monitor their condition in real time, monitor the life cycle of rolls, and analyze the behavior of rolls from different suppliers. The implemented service provides a detailed resource history and ensures that the equipment is properly prepared for operation. The service also makes it possible to analyze the behavior of rolls from different suppliers and optimally manage the purchases of rolls.
MES Dispatching and Logistics Optimization System in Converter Shop 1
The project includes the following technologies MES (according to MES-11): function (DCA Data Collection/Acquisition) - data collection and storage, PTG (Product Tracking and Genealogy) - product tracking and genealogy, ODS (Operations/Detail Scheduling) - operational detailed planning and optimization, DPU (English Dispatching Production Units) - production dispatching. Provides the ability to obtain information about the status and location of an order at any time. Status information includes who is performing the task, components, materials, and their suppliers, lot number, serial number, current manufacturing conditions, process disturbance warnings, reprocessing data, and other product-related events.
Real-time tracking creates historical data about the progress of production processes. This data ensures the traceability of the components and their use in each final product. The dispatching information is presented in the sequence in which the work is to be performed and changes in real time as events occur at the workshop level. This allows you to change the specified schedule. The solution is based on the DATA-TRACK platform, operational planning and optimization are based on the DATA-PLAN software module as part of the DATA-TRACK platform.
In the end, the customer made a profit by implementing optimal operational planning in that part of the metallurgical redistribution (dispatching and logistics in converter production), where generally accepted means of automated planning are not used or not effective. At the first stages of the project, a confirmed economic effect of 75 million rubles per year was shown due to optimal planning of the movement of cranes and steel carriers.
At metallurgical enterprises, it is not always possible to build an optimal schedule of equipment for a number of areas and units. The reasons lie in the fact that it is often impossible to obtain a slice of the current state of equipment using traditional methods, as well as in the high complexity of building algorithms and mathematical models to solve the problem of optimal planning. The converter shop is an example of such a complex object.
We solved this problem for converter production by combining process control and planning into one system based on vision (neural network) and the use of advanced combinatorial algorithms (planning). The mathematical support of the project was developed by the "DATA-CENTER Automation" under the leadership of the president of the company and the head of the basic department "Big Data Analytics and Video Analysis Methods" of the Ural Federal University, Doctor of Physics and Mathematics. Gainanova D.M. The participation of the first person of the company among the performers is also a sign of the uniqueness of this project.
The technical vision methods for solving the monitoring problem, the construction of mathematical optimization methods indicate not only the novelty of the project, but also make the project significant for the IT industry of building MES systems, and therefore relevant.
The project has two main subsystems:
First subsystem at all times (delay <1 сек) мониторит текущие координаты кранов и сталевозов, а также положение стальковшей с плавкой, загруженность оборудования. В основе мониторинга лежит платформа DATA-TRACK, разработанная фирмой «ДАТА-ЦЕНТР Автоматика», которая опирается на техническое зрение.
Without the widespread use of technical vision, solving the problem of total monitoring of the movement of buckets and smelters in the converter shop, as well as determining in real time the position of buckets and steel carriers for the subsequent calculation of their optimal movement is an impossible or, in any case, very expensive task. In total, 38 cameras located on cranes and in the workshop were used to solve the problem, and neural networks and deep machine learning were used to process data.
The second subsystem implements original planning algorithms based on modern methods of mathematical programming, combinatorial optimization and graph theory. The developed algorithms guarantee accurate solutions to the assigned tasks, providing forecast planning for a day or more and prompt rescheduling of the movement of smelts based on the data generated by the monitoring subsystem. Logistics optimization takes place constantly and in real time. The economic effect is achieved by saving the costs of electricity and aluminum rod, associated with the optimization of the logistics of the movement of smelters in the converter shop.
MES Dispatching and Logistics Optimization System in the Mixer Room of Converter Shop No. 2
The project includes the following technologies MES (according to MES-11): function (DCA Data Collection/Acquisition) - collection and, data storage PTG (Product Tracking and Genealogy) - tracking and genealogy of products, ODS (Operations/Detail Scheduling) - operational detailed planning and optimization, DPU (English Dispatching Production Units) - dispatching production. Provides the ability to obtain information about the status and location of an order at any time. Status information includes who is performing the task, components, materials, and their suppliers, lot number, serial number, current manufacturing conditions, process disturbance warnings, reprocessing data, and other product-related events. Real-time tracking creates historical data about the progress of production processes. This data ensures the traceability of the components and their use in each final product. The dispatching information is presented in the sequence in which the work is to be performed and changes in real time as events occur at the workshop level. This allows you to change the specified schedule. The solution is based on the DATA-TRACK platform, operational planning and optimization are based on the DATA-PLAN software module as part of the DATA-TRACK platform.
2021
MES Dispatching and Logistics Optimization System in Converter Shop# 2
The project includes the following technologies MES (according to MES-11): function (DCA Data Collection/Acquisition) - data collection and storage, PTG (Product Tracking and Genealogy) - product tracking and genealogy, ODS (Operations/Detail Scheduling) - operational detailed planning and optimization, DPU (English Dispatching Production Units) - production dispatching. Provides the ability to obtain information about the status and location of an order at any time. Status information includes who is performing the task, components, materials, and their suppliers, lot number, serial number, current manufacturing conditions, process disturbance warnings, reprocessing data, and other product-related events.
Real-time tracking creates historical data about the progress of production processes. This data ensures the traceability of the components and their use in each final product. The dispatching information is presented in the sequence in which the work is to be performed and changes in real time as events occur at the workshop level. This allows you to change the specified schedule. The solution is based on the DATA-TRACK platform, operational planning and optimization are based on the DATA-PLAN software module as part of the DATA-TRACK platform.
In the end, the customer made a profit by implementing optimal operational planning in that part of the metallurgical redistribution (dispatching and logistics in converter production), where generally accepted means of automated planning are not used or not effective.
At metallurgical enterprises, it is not always possible to build an optimal schedule of equipment for a number of areas and units. The reasons lie in the fact that it is often impossible to obtain a slice of the current state of equipment using traditional methods, as well as in the high complexity of building algorithms and mathematical models to solve the problem of optimal planning. The converter shop is an example of such a complex object.
We solved this problem for converter production by combining process control and planning into one system based on vision (neural network) and the use of advanced combinatorial algorithms (planning). The mathematical support of the project was developed by the "DATA-CENTER Automation" under the leadership of the president of the company and the head of the basic department "Big Data Analytics and Video Analysis Methods" of the Ural Federal University, Doctor of Physics and Mathematics. Gainanova D.M. The participation of the first person of the company among the performers is also a sign of the uniqueness of this project.
Technical vision methods for solving the monitoring problem, the construction of mathematical optimization methods indicate not only the novelty of the project, but also make the project significant for the IT industry of building MES systems, and therefore relevant.
The project has two main subsystems:
First subsystem at all times (delay <1 сек.) мониторит текущие координаты кранов и сталевозов, а также положение стальковшей с плавкой, загруженность оборудования. В основе мониторинга лежит платформа DATA-TRACK, разработанная фирмой «ДАТА-ЦЕНТР Автоматика», которая опирается на техническое зрение.
Without the widespread use of technical vision, solving the problem of total monitoring of the movement of buckets and smelters in the converter shop, as well as determining in real time the position of buckets and steel carriers for the subsequent calculation of their optimal movement is an impossible or, in any case, very expensive task. In total, 38 cameras located on cranes and in the workshop were used to solve the problem, and neural networks and deep machine learning were used to process data.
The second subsystem implements original planning algorithms based on modern methods of mathematical programming, combinatorial optimization and graph theory. The developed algorithms guarantee accurate solutions to the assigned tasks, providing forecast planning for a day or more and prompt rescheduling of the movement of smelts based on the data generated by the monitoring subsystem. Logistics optimization takes place constantly and in real time. The economic effect is achieved by saving the costs of electricity and aluminum rod, associated with the optimization of the logistics of the movement of smelters in the converter shop.
The project is the winner of the RB Digital Awards 2021 for the best cases in the field of digital transformation in the "Production" nomination. The project was presented for the award under the name "Digital service HEFEST: in-house logistics and dispatching for steelmaking."
MES Check of operability (efficiency) of the concept for optimization of molten iron overflows in the mixer compartment of the converter shop No. 2
MES modeling was performed to determine the effectiveness of optimization of molten iron overflows. The simulation is based on the EXPERT BASE platform.
MES Development of a minimum viable product (MVP) for the analysis of the cyclogram of the movement of a diesel locomotive with cast-iron buckets using video analytics in the circuit of the mixer compartment and converters in the converter shop No. 1
The project includes the following MES technologies (according to MES-11): the DCA (Data Collection/Acquisition) function - data collection and storage, PTG (English Product Tracking and Genealogy) - product tracking and genealogy. The project provides the ability to obtain information about the status and location of the order at each time.
Status information includes who is performing the task, components, materials, and their suppliers, lot number, serial number, current manufacturing conditions, process disturbance warnings, reprocessing data, and other product-related events. The real-time tracking function creates historical data on the passage of production processes in terms of the movement of a diesel locomotive with cast-iron buckets in the circuit of the mixer compartment and converters in the converter shop No. 1. This data ensures the traceability of the components and their use in each final product. The solution was created on the basis of the DATA-TRACK platform, the movement of the locomotive was recorded by means of technical vision. The analytical industrial platform EXPERT BASE was used to implement the analysis of the cyclogram of the movement of a diesel locomotive with cast-iron buckets.
2020: MES Operability Check of Dispatching and Logistics Optimization Concept in Converter Shops# 1 and# 2
Carried out modeling MES in order to determine the effectiveness of optimization of dispatching and. logistics Modeling was carried out on the basis of the platform. EXPERT BASE