Mold Level Control Research Articles

Neuro-Fuzzy System for Compensating Slow Disturbances in Adaptive Mold Level Control

Good slow disturbances attenuation in a mold level control with stopper rod is very important for avoiding several product defects and keeping down casting interruptions. The aim of this work is to improve the accuracy of the diagnosis and compensation of an adaptive mold level control method for slow disturbances related to changes of stopper rod. The advantages offered by the architecture, called Adaptive-Network-based Fuzzy Inference System, were used for training a previous model. This allowed learning based on the process data from a steel cast case study, representing all intensity levels of valve erosion and clogging. The developed model has high accuracy in its functional relationship between two compact input variables and the compensation coefficient of the valve gain variations. The future implementation of this proposal will consider a combined training of the model, which would be very convenient for maintaining good accuracy in the Fuzzy Inference System using new data from the process.

Eddy current method for steel billet mould level control

The relevant issue of metal level control in a mould of continuous casting machine (CCM) is improving the measurement accuracy when adding slag powders in a mould. The main disadvantage of the widely used radiometric metal level control devices (MLCDs) is the dependence of the output signal on the amount of slag in a mould. Eddy current (EC) sensor for metal level control is not sensitive to the slag. It detects metal level indirectly by temperature of copper mould wall, which is in contact with liquid metal. As a consequence, eddy current MLCD speed of response is limited to the time constant of thermal transient processes in a mould wall, which is insufficient to modern steel industry. This paper deals with a mould CCM metal level control method by means of the EC sensor based on signal detection from the liquid metal through a copper mould wall. The direct method provides required signal speed of response of eddy current MLCD in the case of steel casting with the slag.

A fuzzy PID controller with nonlinear compensation term for mold level of continuous casting process

The mold level is the key to stable continuous casting production process, and its fluctuation determines the quality of the slab. It is of great economic and scientific value to stabilize the mold level in numerous disturbances and complicated operating modes. To achieve high-precision and fast-response control of mold level control in engineering applications, a fuzzy PID controller with nonlinear compensation term is designed in this paper. By analyze the dynamic characteristics of the mold level, a fuzzy PID controller is designed. The parameters of fuzzy controller is optimized by using PSO algorithm, which achieves a better control effect in abnormal operating conditions. More importantly, a nonlinear compensation term is introduced to compensate the output of the fuzzy PID controller, thus further improving the anti-interference ability of the system. To verify the superiority of our method, a series of simulation studies are carried out. Finally, the method is applied to industrial sites. The running results show that, compared with the traditional control method, our method effectively improves the anti-interference ability under normal operating conditions and better suppresses the fluctuation of casting opening.

Mold-level prediction based on long short-term memory model and multi-mode decomposition with mutual information entropy

The mold is referred to as the heart of the continuous casting machine. Mold-level control is one of the keys to ensuring the quality of a high-efficiency continuous casting slab. This article addresses the failure of the mold-level prediction model in the actual production process to overcome the impact of noise. To improve the accuracy of mold-level prediction, a novel method for mold-level prediction based on the multi-mode decomposition method and the long short-term memory model is proposed. First, empirical mode decomposition of the mold-level data is performed. The actual eigenmode component number K is obtained through the calculation of the mutual information entropy of the eigenmode components. Then, we perform a K-based variational mode decomposition on the mold-level data. The noise dominant component is denoised by the calculation of the mutual information entropy of the eigenmode components. Moreover, the long short-term memory model is used to predict the noise dominant component and the information dominant component after denoising. Finally, the predicted result is subjected to variational mode decomposition reconstruction to obtain the predicted mold-level data. The experimental results show that compared with the other methods tested, the model has better prediction efficiency, prediction accuracy, and generalization ability. It provides a new idea for mold-liquid-level prediction and continuous casting blank quality assurance.

Mold-Level Prediction for Continuous Casting Using VMD–SVR

In the continuous-casting process, mold-level control is one of the most important factors that ensures the quality of high-efficiency continuous casting slabs. In traditional mold-level prediction control, the mold-level prediction accuracy is low, and the calculation cost is high. In order to improve the prediction accuracy for mold-level prediction, an adaptive hybrid prediction algorithm is proposed. This new algorithm is the combination of empirical mode decomposition (EMD), variational mode decomposition (VMD), and support vector regression (SVR), and it effectively overcomes the impact of noise on the original signal. Firstly, the intrinsic mode functions (IMFs) of the mold-level signal are obtained by the adaptive EMD, and the key parameter of the VMD is obtained by the correlation analysis between the IMFs. VMD is performed based on the key parameter to obtain several IMFs, and the noise IMFs are denoised by wavelet threshold denoising (WTD). Then, SVR is used to predict each denoised component to obtain the predicted IMF. Finally, the predicted mold-level signal is reconstructed by the predicted IMFs. In addition, compared with WTD–SVR and EMD–SVR, VMD–SVR has a competitive advantage against the above three methods in terms of robustness. This new method provides a new idea for mold-level prediction.

Mold Level Control of Large Size Ingot Based on Fuzzy Self-Tuning PID

Mold Level Control of Large Size Ingot Based on Fuzzy Self-Tuning PID

Robust nonlinear adaptive mould level control for steel continuous casting

Abstract The mould level control for steel continuous casting is an open research problem. In ACINOX-Las Tunas steelwork there are occasional problems with the amplitude of mould level variations and with the stability of the control system. The control quality should be better satisfied and a robust nonlinear PI controller with guaranteed robustness designed with a new NPI-RR method is proposed. This achieves a good trade-off between robustness and disturbance attenuation and also optimizing the trade-off between disturbance attenuation and control effort. The slow disturbances, related to erosion or clogging of the stopper valve and the tundish level variations, can be compensated through an adaptive scheme with gain scheduling, fuzzy logic and online identification. The simulation scale results illustrate the effectiveness of the proposal.

The consistency control of mold level in casting process

A new multiple periodic disturbance rejection control for casting process is proposed in this paper. Firstly, a high-precision multiple periodic disturbance detection method is designed. The accurate frequencies of the disturbance can be obtained. By using the disturbance observer and a group of filters with additional delay items, multiple periodic disturbance can be suppressed under time-delay environment. Then systematic control design is achieved based on the robust stability analysis. Finally, the efficiency of the proposed method is shown by comparisons between different methods and time-varying disturbance rejection in the consistency control of mold level.

Modelado de Control de Nivel en Molde para Vaciado Continuo de Palanquillas de Acero

This paper presents a model of system for the mould level control in a continuous casting machine for steel billets. The objective was to carry out the modeling of each element of the system including disturbances and transformations of the plant that affect the quality of control in the installation study object. It was obtained and validated an approximate model of the control system using physical laws governing the same one, experimental identification by special signals and parametric identification off- line. The model explicitly includes the phenomena of obstruction and erosion of the valve which have a significant impact on the operation of the plant. For hydraulic servo system achieved a gray box model contributes to a better understanding of the process and evaluates incidence of the multi-frequency implemented under conventional outline. Obtained the model can be studied scale simulation system operation and the effectiveness of advanced control solutions. The developed work is part of an I+D process that in the course of more than 10 years, it has contributed to that in the steelmaking where is carried out the investigation, there is a mould level control system that has positive impact on the quality of steel produced.

Mold level stabilization using a standing wave model in continuous casting

In continuous steel casting plants, it is very important to stabilize the molten steel level in the mold for quality of the final products and productivity. Standing waves in the mold cause periodic molten steel level fluctuations, which can deteriorate the product quality and productivity. Nevertheless, there has not been any effective countermeasure for them, because they are different from other disturbances in the sense that they are not changes in the flow rate of the molten steel and hence should not be controlled by manipulating the mass flow. This paper proposes a standing wave model and a mold stabilization technique that can prevent harmful influence of standing waves. This technique can prevent interactions between the mold level control and the standing waves in the mold by removing standing wave components from the original mold level signal. On-line experimental results have shown good performance and this technique has been in practical use in a continuous steel casting process in JFE.

Review: The “butterfly effect” in continuous casting

The continuous casting (CC) mould may appear very peaceful when viewed from above, but the powder bed hides relentless fluctuations in the following phenomena: metal flow, thermal gradients, chemical reactions and multiple phase transformations. When observed separately, some of these phenomena seem to have a ‘simple behaviour’, which may appear easy to control through the main casting parameters (e.g. casting speed, pouring temperature and powder type) and associated control systems (e.g. mould level control, automatic powder feeding and mould oscillation). However, when combined, these phenomena exhibit periodic fluctuations in behaviour, which is both difficult to predict and control. For instance, the combination of casting speed, submerged entry nozzle design and slab size can cause the metal flow pattern to shift from double roll to single roll and back, which can cause unstable fluctuations in metal level, standing waves, etc. In this respect, the CC process closely resembles a meteorological system where both variations and local fluctuations in temperature, humidity, pressure, etc., can result in effects that are difficult to predict in the long term. This is equivalent to the famous Lorenz premise: ‘Does the flap of a butterfly’s wings in Brazil set off a tornado in Texas?’ In this paper, we give some examples of the ‘butterfly effect’ in CC discussed below by using a mathematical model able to predict the slab solidification inside the mould in which various factors affecting the process stability are analysed and the probable sources of fluctuation are identified.

Switching Policies for Mold Level Control in Continuous Casting Plants

The problem of controlling the liquid level in the mold of a continuous casting process is considered. Essentially, the system operates in two different modes: (a) one in which the system is affected only by small-amplitude high-frequency disturbances, and (b) one in which the system is affected by a comparatively large disturbance related to an abrupt unclogging phenomenon. To cope with the first situation, the control system must exhibit a narrow bandwidth, whereas the second situation requires a much larger bandwidth. The two conflicting requirements lead naturally to the design of two distinct controllers and a switching rule that activates either controller according to the operating condition. Due to the time-varying nature of the resulting control system, the controllers should be properly realized to guarantee stability; to this purpose, a recent result on switching systems <citerefgrp xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><citeref refid="ref15"/> </citerefgrp> is invoked. Since the sudden switching between controllers can cause controller and actuator output bumps, a soft switching policy that still ensures stability is also proposed. Simulations and laboratory experiments carried out on an electronic analogue of the process show the effectiveness of the suggested control schemes.

Control of mold level in continuous casting based on a disturbance observer

Regulating molten steel level in continuous casting is difficult because it suffers from various disturbances, among which the bulging disturbance is the most prominent. Moreover, as the casting speed increases in continuous casting, the phase lag due to the system delay increases. To eliminate the bulging disturbance in the presence of system delay, we propose an adaptive sine estimator based disturbance observer and a phase lead adaptive fuzzy controller. We tested the proposed method using a 1:1 scale hardware simulator and confirmed that the method successfully decreased the bulging disturbance effect from the molten steel level.

Next Generation Mold Level Control: Development of LevCon 2.0

Siemens VAI develops the next generation of the successful mold level controller LevCon. The new controller uses the most modern control principles from optimal control theory while being robust enough to operate in the difficult conditions of a steel work. One prerequisite for the application of modern control theory is a good mathematical model of the process. In the past, the models used for mold level control design were quite simple, neglecting some of the important limiting factors such as surface resonance waves of the mold level. For the derivation of a resonance surface wave model, extensive water model trials have been carried out. With the help of the plant model, modern control theory is applied and a mathematically optimal controller feedback is designed. It offers an increased disturbance rejection compared to earlier algorithms, which is significantly reducing the mold level deviation due to disturbances as, e.g., unclogging events. However, there are still disturbances that cannot be eliminated completely, like mold level hunting caused by instationary bulging. In the past, mold level hunting was reported to be a very slow disturbance with a wavelength of approximately one roller pitch and a frequency below 0.1 Hz. During the development work, it turned out that also higher harmonics of this base frequency exist with wavelengths of 1/2, 1/3, up to 1/7 of the roller pitch. These high frequency fluctuations cannot be eliminated by feedback control only. However, an additional feed-forward loop which utilizes the withdrawal drives power for direct compensation of the mold level hunting steel flow is developed. With this approach, it is possible to reduce even fast mold level hunting.

Particle swarm optimization based tuning of a modified smith predictor for mould level control in continuous casting

Mould level variations are a serious productivity and quality problem in continuous casting process. This work proposes a level control structure based on the Aström's modified Smith predictor able to improve the bulging effect rejection. Unlike conventional methods, this control strategy decouples the disturbance rejection from the setpoint response and therefore can be independently optimized. Using this scheme, the bulging rejection specifications are reformulated as a H ∞ problem and the tuning parameters are designed through the particle swarm optimization approach. Simulation results confirm that the proposed architecture is more effective than the other ones currently implemented in real plants.

Mold level control of a continuous casting plant by switching control strategies

In continuous casting plants especially at higher casting speeds dynamic bulging, a variation of the magnitude and/or the shape of the bulges, may occur and lead to a fluctuation of the height of molten steel with serious consequences in terms of quality and damage. A steady growth of these fluctuations is called mold level hunting. Presently both phenomena are met with a reduction of the casting speed.This paper proposes a method based on switching control strategies, in particular the method of dynamic output feedback control, to keep the process stable and to increase the troughput of such plants. Simulations with models of different complexity show a productivity increase of around 20%.

Particle Swarm Optimization Based Tuning of a Modified Smith Predictor for Mold Level Control in Continuous Casting

Mold level variations are a serious productivity and quality problem in continuous casting process. This work proposes a level control structure based on the Aström's modified Smith predictor able to improve the reduction of the bulging effect. Unlike conventional methods, this control strategy decouples the disturbance rejection from the setpoint response and therefore can be independently optimized. Within this scheme, the bulging rejection specifications are formulated as a H ∞ problem and the tuning parameters are designed through a particle swarm optimization approach. Simulation results confirm that the proposed architecture is more effective than the ones currently implemented in real plants.

Control issues in continuous casting of steel

Continuous casting plants are highly complex plants whose performance requires reliable mold level control system. Several effects greatly influence the surface quality of the final product and if these are not suppressed sufficiently by the mold level control system, than the production speed, hence the productivity has to be lowered. This paper analyses the most important disturbance and their effects on the control system design, and presents a new solution for the most critical case, mold level hunting.

Re-start Technology for Reducing Sticking-type Breakout in Thin Slab Caster

In order to increase the production rates and to secure the base production technology for the hard casting materials, a re-start casting technology as an active curing methodology has been developed to reduce the number of sticking type breakouts in the thin slab caster with high speed. It consists of the healing casting conditions of breakouts, the mold level control method for preventing the overflow or big change of molten steel near meniscus and the control of cooling rate at second cooling zone for rolling under the limits of rolling force and torque at the directly linked roughing mill. By applying the proposed technology, the automatic restarting rate of breakout is remarkably increased from 0 to 92%.

FUZZY NEURON HYBRID CONTROL FOR CONTINUOUS STEEL CASTING

The fuzzy neuron hybrid control method is proposed for mold level control of the continuous steel casting in this paper. The dynamic characteristics of the continuous steel casting process is described. To the plant with big uncertainties and grave nonlinearities, the fuzzy neuron hybrid control system is set up and the new model-free controller is designed. In this control system, the fuzzy neuron hybrid controller is constructed by the fuzzy PI controller and the neuron controller, the gain of the neuron controller is tuned by using a fuzzy algorithm. The simulation tests of mold level control are made under various conditions with different casting speeds, new gate valve and old gate valve, changing in the valve position loop gain. The results demonstrate that the new controller has good performance, very strong robustness and adaptability.