Modeling and Level 2 Automation
Sarir Company, backed by a strong technical team, extensive knowledge, and vast experience in the steel and oil industries, can perform modeling, smartening, and level 2 automation in these industries in line with the Fourth Industrial Revolution. This brings about the following benefits and outcomes:
- Improved product quality
- Optimization of energy and raw material consumption
- Reduced operational costs
- Reduced production time and consequently increased production
Implementing this project in the industry involves the following steps:
Step 1: Monitoring and Data Collection
For modeling, model testing, and using machine learning, accurate data from the process is needed. Sometimes this data can be obtained using sensors installed on the production line or stored records, and sometimes due to various reasons such as lack of data, low accuracy, or low sampling frequency of parameters, temporary or permanent data acquisition equipment installation is required.
Using monitoring and data collection modules in Level 2 automation, we receive and record real-time information from industrial processes such as temperature, speed, acceleration, and similar physical data, ultimately collecting and monitoring them. This information forms the basis for decision-making at the automation level and enables us to precisely and optimally control processes.
Monitoring and Data Collection Modules
- Sensors: Using sensors, we measure various physical parameters at different points of the hot rolling line and two-stand tandem mill steel. These sensors record information such as sheet temperature, fluid pressure and flow, and sheet speed.
- Data Acquisition Modules (DAQ): We use data acquisition modules such as data loggers and data acquisition cards to record input data from sensors such as line temperature and acceleration of different parts of the two-stand tandem mill. Then, we convert this data into digital signals and send them to the central system for analysis and control.
- Monitoring and Data Display Module: A module designed as graphical dashboards in the form of SCADA (Supervisory Control and Data Acquisition) for displaying and analyzing temperature data and shower flow of the cooling table, as well as acceleration of different parts of the two-stand tandem mill. In these modules, operators can see the real-time status of processes and make necessary adjustments.
Step 2: Data Analysis
Data analysis and model development systems in Level 2 automation are used to examine and optimize key parameters in industrial processes. These systems process and analyze real-time and historical data to identify important patterns and extract useful information, which helps in making more accurate decisions and applying optimal settings to processes. This section includes the following components:
- Data Collection and Refinement: Cleaning, normalizing, and refining the necessary physical data obtained from sensors to ensure their accuracy and quality
- Data Analysis Using Statistical and Artificial Intelligence Algorithms: Analyzing data using statistical methods and artificial intelligence algorithms such as neural networks and regression to identify patterns and hidden relationships among their amounts (these algorithms also help identify unusual changes and process problems)
Step 3: Process Modeling and Model Development
The third step in implementing the Level 2 module is process modeling. This subsection includes the following parts:
- Process modeling by solving governing equations of the process and simulating them in relevant software
- Using machine learning algorithms on data received from the previous stage to determine uncertainties in the model extracted from the simulation
- Using relationships between physical parameters such as temperature, flow, speed, and other parameters extracted from step two, and the output of the machine learning algorithm from the previous stage, the model is developed to accurately represent the process behavior.
Due to the complexities of performing the above steps and the need for a large amount of CPU, GPU, and RAM, all these steps are carried out on a supercomputer.
Step 4: Offline Software
- Offline Model: For use by process engineers, software is provided to them. In these software programs, it is possible to fully examine the process and study the impact of each parameter and actuator on the process. Additionally, using these software programs, the setpoints of existing control loops in the process can be obtained and provided offline to the operator. In these software programs, the output of the previous step is simplified with machine learning algorithms so that they can run at high speed on personal computers.
- Forecasting and Decision-Making System: Using past data and identified patterns, we can now predict future process behavior and instruct automation controllers to make necessary adjustments. These predictions help prevent potential problems and reduce waste.
- Dashboard and Reporting System: In this section, the information obtained from analysis and optimization is presented to operators and managers in the form of dashboards and reports. These dashboards provide comprehensive and accurate information on the current status of the process and optimization results, which are used for quick and effective decision-making.
Step 5: Implementation
The software designed in step four is implemented in Level 2 automation to receive real-time parameters from the line and sensor data and send the setpoints of existing control loops in the process to Level 1. If the hardware for this task does not exist, it will be implemented first. Additionally, if needed, robust and adaptive control algorithms are added to Level 2 automation to prevent disturbances, uncertainties, and process changes over time.