Slag Detection System Research Articles

Tapping stream tracking model using computer vision and deep learning to minimize slag carry-over in basic oxygen furnace

<abstract> <p>This paper describes a system that can automatically determine the result of the slag dart input to the converter during tapping of basic oxygen furnace (BOF), by directly observing and tracking the behavior of the pouring molten steel at the tapping hole after the dart is injected. First, we propose an algorithm that detects and tracks objects, then automatically calculates the width of the tapping stream from slag-detection system (SDS) images collected in real time. Second, we develop a time-series model that can determine whether the slag dart was properly seated on the tap hole; this model uses the sequential width and brightness data of the tapping stream. To test the model accuracy, an experiment was performed using SDS data collected in a real BOF. When the number of sequential images was 11 and oversampling was 2:1, the classification accuracy in the test data set was 99.61%. Cases of success and failure of dart injection were quantified in connection with operation data such as ladle weight and tilt angle. A pilot system was constructed; it increases the reliability of prevention of slag carry-over during tapping, and can reduce the operator's workload by as much as 30%. This system can reduce the secondary refining cost by reducing the dart-misclassification rate, and thereby increase the productivity of the steel mill. Finally, the system can contribute to real-time process control and management by automatically linking the task of determining the input of darts to the work of minimizing slag carry-over in a BOF.</p> </abstract>

Vision-based detection system of slag flow from ladle to tundish with the help of the detection of undulation of slag layer of the tundish using an image analysis technique

ABSTRACT It is very important to have a slag detection system (SDS) in steelmaking to detect slag and prevent its flow for improving the quality of steel produced. Slag raking is used for removing slag from liquid iron. Detection and control of slag may be carried out with the help of refractory dart or infrared imaging in the steel making zone such as in a Basic Oxygen Furnace (BOF). Each method has its advantages and limitations. Detection and control of slag flow from ladle to tundish in a casting operation is an important step towards cleanliness of the steel production. Various techniques such as ultrasonic level sensor, conductivity measurement using magnetic field, use of microphone to monitor the sound of steel flow through shroud, monitoring the radiation intensity of the stream of steel flow through a lateral side duct of the ladle shroud and monitoring of the shroud vibration have been explored to minimize the slag flow from ladle to tundish. It is obvious from literatures that non-contact based SDS is much more preferred using advanced instrumentation techniques. The problem with the vibration-based system is reduced metal level in the tundish due to setting up of slide gate of the ladle at low flow level as well as non-detection of the slag flow where vortex did not form prior to slag flow. Research showed improved performance with the vision-based caster SDS. Monitoring the sudden brightening of the slag surface around the pouring tube with the help of a CCD camera and image processing technique is also an effective way to detect slag transfer from ladle to tundish. The major limitation of this method is that a significant amount of slag already enters the tundish due to increased flow rate at the end of the tapping before splashing takes place. This work attempts to overcome this problem. There is an increased trend in the amplitude of the undulation of the slag surface of the tundish before the sudden brightening of the slag surface. If such undulations are identified by image analysis technique, signal can be generated to move the ladle slide gate to limit a significant amount of slag flow into the tundish. This would help to improve the quality of the steel that is produced.

Experimental study on accelerometer-based ladle slag detection in continuous casting process

The share of continuous casting method is increased day by day among steel production methods. In continuous casting plants, steel production is carried out using BOF (basic oxygen furnace) or EAF (electric arc furnace). During the journey of the liquid steel through whole steelmaking stages, the carryover ladle slag can turn into one of the most detrimental element in casters. It is known that a carryover ladle slag consists of mainly oxides such as CaO, SiO2, Al2O3, and MgO. Carryover slag has to be kept under control otherwise one can end up with negative results. With the development of sensor technology, the detection of carryover ladle slag becomes possible at the end of heats. The main advantage of the sensor type is early detection over a human-controlled system. A few seconds can make a big difference on safety, quality, yield, castability, and cost when the casting sequence lengthens due to the amount of slag. Nowadays, steel production via BOF which is performed with very low profit margins makes it important to solve this problem. Especially steelmaking produced by BOF, the detection methods of carryover slag are human controlled, electromagnetic based, ultrasonic based, image processing based, acoustic based, and vibration based. In this study, it is seen that the vibration-based detection system has many advantages over others when considering investment cost, product quality, castability, and accuracy. The scope of the study aimed to design a vibration-based carryover slag detection system on BOF steel production line. The designed detection system has an accelerometer, data processor, and controller system. After the experimental study, the cost, the product quality, and the productivity analyses were performed for the condition of use and non-use of the vibration-based detection system. At the end of this study, it is seen that the vibration-based carryover slag detection system is affordable and feasible.

A low-cost non-invasive slag detection system for continuous casting

The majority of steel produced today is made by the technology known in the steel making industry as continuous casting. Deciding when to stop the flow of molten steel from the ladle is not trivial, since terminating the process too early affects yield negatively, while closing the outflow valve too late lets slag enter the casting process. There is a variety of automatic slag detection systems available now, but numerous casting operations still rely on the decision of a human operator. In this paper, we propose a cost-effective non-invasive slag detection system that is based on the vibration signal measured during the casting procedure. In this method, the vibration acceleration data is analyzed by a cumulative sum (CUSUM) control chart in real time, providing a violation signal that can be used to close the ladle outflow valve. The proposed algorithm is implemented in an embedded microcontroller unit and is verified through a simulation study and laboratory experiments. These trials suggest that the technique may perform similarly to the human operator, however, just as in the case of the human operator, the disadvantage is that it only identifies the change when a small amount of slag already enters the tundish. Its advantage lies in its simplicity, low-cost, portable and non-invasive nature; possibly aiding the decision of the operator or, it may be used to create a completely automated ladle outflow valve closing system.

Slag Detection System Based on Infrared Thermography in Steelmaking Industry

Slag Detection System Based on Infrared Thermography in Steelmaking Industry

Slag detection system based on infrared temperature measurement

In order to detect the slag content in the pouring or tapping stream of a basic oxygen furnace (BOF), a slag detection system is designed. The temperature of pouring stream is measured using the infrared temperature measurement system. The molten steel and slag could be distinguished according to their infrared emissivities that have a great difference in far-infrared band. A calibration method for the temperature threshold that is used to distinguish the molten steel and slag in a certain environment is proposed and the amount of slag is calculated in real-time. To avoid the situation that the amount of slag is incapable of being calculated due to the change of BOF angle (moving of pouring stream), a method that can quickly find the edge of the pouring stream is applied to implement the tracking of molten steel pouring stream. Experimental results prove that this system is effective to control the amount of slag that comes into the ladle.

Theoretical Analysis and Experiment on Slag Detection System Based on Eddy Current Effect in Continuous Casting

A new method for slag detection in continuous casting is discussed, which is based on eddy current effect. New system is simulated using finite element method; also experiment is done to verification the feasibility of slag detection. It is demonstrated that the new system is convenient installation and maintenance with higher resolution.

Development of caster slag detection system through imaging technique

Detection and removal of slag from the molten metal is an important aspect of steel making process. The application of infrared imaging, spectroscopic method and other contact type slag detection methods are inappropriate as far as caster slag detection is concerned. The detection of vibration due to slag flow has a major shortcoming. In the present article, a simple imaging technique is presented for caster slag detection. The imaging technique is based on the detection of sudden increase in the light intensity by a CCD camera as the slag floats immediately as it enters the tundish through the shroud in a casting operation. The camera is connected to a computer. As the detected light exceeds the predetermined limit, the system generates the indication of the onset of slag.

Application of Improved HMM Algorithm in Slag Detection System

To solve the problems of ladle slag detection system (SDS), such as high cost, short service life, and inconvenient maintenance, a new SDS realization method based on hidden Markov model (HMM) was put forward. The physical process of continuous casting was analyzed, and vibration signal was considered as the main detecting signal according to the difference in shock vibration generated by molten steel and slag because of their difference in density. Automatic control experiment platform oriented to SDS was established, and vibration sensor was installed far away from molten steel, which could solve the problem of easy power consumption by the sensor. The combination of vector quantization technology with learning process parameters of HMM was optimized, and its revaluation formula was revised to enhance its recognition effectiveness. Industrial field experiments proved that this system requires low cost and little rebuilding for current devices, and its slag detection rate can exceed 95%.

Steel Water Continuous Casting Slag Detection System Based on Wavelet

Based on a large number of investigation and analysis for slag detection system (SDS), a kind of automatic SDS using vibration measurements had been put forward in this paper. After studying casting process from ladle to pouring box and protection casting supporting mechanical structure, vibration dynamics model of supporting structure was established, also variation of steel water impact force and influencing factors of inspecting spot vibration are analyzed. Theories of wavelet, wavelet package and their applications in signal processing are presented. Vibration signal of steel water was processed by the way of wavelet package, then its eigenvector and tolerance range were obtained, and that can realize effective identification for steel water status. Experimental results indicate that this system is very effective and achieves anticipated outcomes.

Steel water continuous casting slag detection system based on wavelet

Based on lots of investigation and analysis for internal and overseas slag detection system (SDS) technology, a kind of SDS realization method using wavelet analysis is brought forward. According to the characteristic of design and realization for SDS, experiments platform for vibration measurement is established. After studying casting process from ladle to pouring box and protection casting supporting mechanical structure, vibration dynamics model of supporting structure is founded, also variation of steel water impact force and influencing factors of inspecting spot vibration are discussed. Primary analysis is done for steel water vibration signal by the way of basal wavelet processing method, and the deeply feature extraction is finished to realize the effective recognition for steel water status using wavelet package method. Experimental results indicate that the method has low realization cost and long operating life, requires little rebuilding for current devices, and its slag detection rate can reaches more than 95 percents.