Electroslag furnace argon flow and micro-positive pressure linkage control method and system

Abstract

This invention proposes a method for the coordinated control of argon flow rate and micro-positive pressure in an electroslag remelting furnace, comprising: acquiring oxygen content and micro-positive pressure signals within the furnace; predicting future furnace conditions using a deep belief network model; offline optimizing the initial parameters of a fuzzy PID controller using a particle swarm optimization algorithm; online adaptively adjusting the controller parameters based on real-time signals and prediction results; and finally, outputting coordinated control commands for argon flow rate and dust collector valve opening. This invention achieves proactive control by predicting future furnace conditions using a deep belief network model, overcoming the problem of large system lag; offline parameter optimization of the fuzzy PID controller using a particle swarm optimization algorithm provides it with excellent initial performance adaptable to different operating conditions; adaptive adjustment of controller parameters through online fuzzy inference enables the system to be robust to dynamic processes; and finally, precise and stable coordinated control of oxygen content and micro-positive pressure within the furnace is achieved by calculating the coordinated argon flow rate and dust collector valve opening commands.

Description

Technical Field

[0001] This invention relates to the field of metallurgical technology, and in particular to a method and system for the linkage control of argon flow rate and micro-positive pressure in an electroslag furnace. Background Technology

[0002] In the electroslag remelting process, introducing argon gas into the furnace and maintaining a stable, slightly positive pressure environment is a key process to prevent steel oxidation, reduce element loss, and improve the purity of the steel ingot. Simultaneously, adjusting the exhaust volume using a proportional dust removal valve is the core method for maintaining this slightly positive pressure and removing harmful dust generated during the smelting process. Currently, the industry commonly uses traditional PID control or fixed flow mode to regulate the argon gas introduction process, and dust removal systems often operate independently.

[0003] However, this type of method has significant drawbacks:

[0004] First, the control mode is rigid and lagging. The control parameters of argon flow and dust removal valve rely heavily on manual experience to set or keep them fixed. They cannot be adjusted in real time according to the dynamic changes in oxygen content and slight positive pressure in the furnace. This results in a mismatch between argon supply and actual demand. The dust removal system also fails to respond synchronously, causing fluctuations in the furnace atmosphere and insufficient control accuracy.

[0005] Secondly, the argon flow control and the dust removal system pressure regulation are independent of each other, and no coordinated linkage mechanism has been established between the two control loops, which leads to mutual interference during the regulation process and a high risk of system oscillation.

[0006] Finally, the control commands of the existing system rely entirely on the feedback signal at the current moment, lacking the ability to predict the dynamic changes and large lag characteristics of the smelting process, and are unable to implement advanced compensation control, especially when the operating conditions are switched, the system response is slow. Summary of the Invention

[0007] This invention proposes a method and system for the linkage control of argon flow rate and micro-positive pressure in an electroslag furnace, which solves the technical problems of insufficient control accuracy of oxygen content and micro-positive pressure in the furnace, response lag and system oscillation caused by the independent control of argon and dust removal systems, lack of predictive ability and fixed parameters in existing electroslag furnace control systems.

[0008] The technical solution of this invention is implemented as follows:

[0009] The first aspect of this invention provides a method for linkage control of argon flow rate and micro-positive pressure in an electroslag furnace, comprising the following steps:

[0010] S1, collects real-time oxygen content signals and micro-positive pressure signals inside the electroslag furnace;

[0011] S2, based on the collected real-time signals, uses a deep belief network (DBN) model to predict the oxygen content and micro-positive pressure in the furnace at future moments;

[0012] S3, using the particle swarm optimization (PSO) algorithm to optimize the initial parameters of the fuzzy PID controller offline;

[0013] S4. Initialize the fuzzy PID controller using the optimized initial parameters, and adaptively adjust the parameters of the fuzzy PID controller based on the real-time signal and prediction results.

[0014] S5 calculates and outputs argon flow control commands and dust removal valve opening control commands based on the adjusted fuzzy PID controller parameters, in order to work together to maintain the set oxygen content and slight positive pressure inside the furnace.

[0015] Specifically, step S1 includes:

[0016] The collected oxygen content signal is processed by Kalman filtering. The filtering formula is as follows:

[0017] ;

[0018] in, Here is the filtered oxygen content estimate at time k. The measured oxygen content at time k is... Kalman gain;

[0019] The collected pressure signal is processed by moving average filtering. The filtering formula is as follows:

[0020] ;

[0021] in, The pressure estimate at time k is the filtered value. The measured pressure value at time k−i.

[0022] Specifically, in step S2, the input features of the Deep Belief Network (DBN) model include real-time signals, historical trends, and smelting stage identifiers, and the output is the predicted oxygen content and pressure for a specific future sampling period; wherein, the historical trend is the rate of change of oxygen content deviation and the rate of change of pressure deviation over the previous several sampling periods.

[0023] Furthermore, step S2 also includes a prediction compensation step:

[0024] When the deviation between the predicted future state and the set value exceeds a preset threshold, feedforward compensation is applied to the basic argon flow rate. The compensation formula is as follows:

[0025] ;

[0026] in, The predicted and compensated argon flow rate, The uncompensated base argon flow rate, To predict gain, This represents the predicted oxygen content at time k+n. Set the oxygen content value.

[0027] Specifically, step S3 includes:

[0028] The Particle Swarm Optimization (PSO) algorithm iteratively optimizes the particle swarm based on historical smelting data, wherein the position code of each particle includes the initial parameters of the fuzzy PID controller and the center value of the fuzzy membership function.

[0029] The particle velocity update formula for the PSO algorithm is as follows:

[0030] ;

[0031] in, Let represent the velocity of the i-th particle in the j-th dimension at iteration k+1; w is the inertia weight. , For learning factors; , A random number in the interval [0,1]. Let be the optimal position of the i-th particle in the j-th dimension. Let i be the position of the i-th particle in the j-th dimension during the k-th iteration. Let be the globally optimal position of all particles in the j-th dimension;

[0032] Through iterative optimization using the PSO algorithm, an optimal initial parameter library corresponding to different steel grades or working conditions is obtained.

[0033] Specifically, in step S4, the method for adaptively adjusting the parameters of the fuzzy PID controller based on the real-time signal and the prediction result includes:

[0034] Based on real-time oxygen content deviation and its rate of change The correction amount of PID parameters is generated in real time through fuzzy inference. And update the parameters according to the following formula:

[0035] ;

[0036] in, Let k be the parameters of the fuzzy PID controller at time k. These are adaptive coefficients.

[0037] Specifically, in step S5, the calculation formula for the argon flow control command is as follows:

[0038] ;

[0039] in, This represents the control value of the argon flow rate at time k; This indicates the stage correction factor determined based on the smelting stage.

[0040] Specifically, in step S5, the calculation formula for the dust collector valve opening command is as follows:

[0041] ;

[0042] in, This represents the control value of the dust collector valve opening at time k; This indicates the initial value of the dust collector valve opening. This is the flow rate change compensation coefficient. This represents the pressure deviation compensation coefficient. This represents the slight positive pressure deviation at time k.

[0043] Preferably, the control method further includes step S6, whereby when the system performance indicators fail to meet the requirements within multiple consecutive smelting cycles, a self-evolution mechanism is triggered to re-execute the PSO algorithm for global parameter optimization and to incrementally train the DBN model using newly added real-time data.

[0044] A second aspect of the present invention provides a linkage control system for argon flow rate and micro-positive pressure in an electroslag furnace, comprising:

[0045] The data acquisition module is used to collect real-time data on oxygen content and micro-positive pressure inside the furnace;

[0046] The controller calculates and outputs argon flow control commands and dust removal valve opening control commands based on the real-time collected data and the control method described above.

[0047] The actuator, including the argon flow control component and the dust removal proportional valve, is used to receive and execute the argon flow control command and the dust removal valve opening control command issued by the controller.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0049] (1) This invention uses a deep belief network (DBN) model to predict future furnace conditions, thereby achieving control advance and overcoming the problem of large system lag; it uses the particle swarm optimization (PSO) algorithm to optimize the fuzzy PID controller offline, providing it with excellent initial performance to adapt to different working conditions; it uses online fuzzy inference to adaptively adjust the controller parameters, enabling the system to have robustness in dealing with dynamic processes; finally, it uses linkage calculation of argon flow rate and dust removal valve opening command to achieve precise and stable linkage control of oxygen content and micro-positive pressure in the furnace.

[0050] (2) By introducing a self-evolution mechanism, the present invention can automatically trigger the re-optimization of the PSO algorithm and the incremental training of the DBN model when the system performance deteriorates, so that the control system has the ability to learn and optimize itself during long-term operation, effectively overcoming the model mismatch and performance degradation caused by equipment aging or raw material changes, and ensuring the long-term consistency and reliability of the control effect. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic flowchart of a method for controlling the linkage between argon flow rate and micro-positive pressure in an electroslag furnace according to the present invention. Detailed Implementation

[0053] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0054] Reference Figure 1 The first aspect of this invention provides a method for linkage control of argon flow rate and micro-positive pressure in an electroslag furnace, comprising the following steps:

[0055] S1, real-time oxygen content signal and micro-positive pressure signal in the electroslag furnace are collected by oxygen content sensor and micro-positive pressure sensor;

[0056] To ensure control accuracy, the original acquired signal needs to be filtered:

[0057] The acquired oxygen content signal is processed by Kalman filtering to suppress random noise and improve signal quality. The filtering formula is as follows:

[0058] ;

[0059] in, Here is the filtered oxygen content estimate at time k. The measured oxygen content at time k is... This is the Kalman gain, typically ranging from 0.1 to 0.3.

[0060] The collected pressure signal is processed by moving average filtering to smooth short-term fluctuations. The filtering formula is as follows:

[0061] ;

[0062] in, The pressure estimate at time k is the filtered value. The measured pressure value at time k−i.

[0063] S2, based on the collected real-time signals, uses a deep belief network (DBN) model to predict the oxygen content and micro-positive pressure in the furnace at future moments, and performs feedforward compensation based on the prediction results;

[0064] The DBN model is a three-layer network structure, consisting of one input layer (6 neurons) and two layers of Restricted Boltzmann Machines (RBMs), each with 12 neurons. Its input features include:

[0065] Real-time parameter: Oxygen content at the current moment Slight positive pressure Argon flow rate Dust collector valve opening ;

[0066] Historical trend: Rate of change in oxygen content deviation over the first 3 sampling periods (corresponding to 0.3 seconds) and pressure deviation change rate ;

[0067] Smelting stage identifiers: Different stages are represented by numbers, such as arc initiation period = 1, arc stabilization period = 2, crystallization period = 3.

[0068] The model was trained using at least 5000 sets of historical data covering different steel grades and furnace conditions. It employed the Adam optimizer with mean squared error (MSE) as the loss function. The model output is a predicted oxygen content for the next two sampling periods (0.2 seconds). and pressure forecast values .

[0069] Furthermore, a predictive compensation step is introduced:

[0070] When the deviation between the predicted future state and the set value satisfies

[0071] or hour,

[0072] Feedforward compensation is applied to the base argon flow rate, and the compensation formula is as follows:

[0073] ;

[0074] in, The predicted and compensated argon flow rate, The uncompensated base argon flow rate, To predict gain, This represents the predicted oxygen content at time k+n. Set the oxygen content value.

[0075] S3, using the particle swarm optimization (PSO) algorithm to optimize the initial parameters of the fuzzy PID controller offline;

[0076] The PSO algorithm is used to solve the problem of blind initial setting of controller parameters. The PSO algorithm iteratively optimizes the particle swarm based on historical smelting data, wherein the position code of each particle contains the initial parameters of the fuzzy PID controller and the center value of the fuzzy membership function; its specific implementation process is as follows:

[0077] Particle encoding: Each particle is a 19-dimensional vector containing the initial parameters of 3 fuzzy PIDs ( ) and the center values ​​of 14 fuzzy membership functions (corresponding to the oxygen content deviation of the two input variables). and slight positive pressure deviation Each variable has 7 fuzzy subsets {NB, NM, NS, ZO, PS, PM, PB}.

[0078] Parameter range: , , The membership function center value is allowed to fluctuate by ±20% from the base value.

[0079] Velocity update: A velocity update formula with adaptive inertia weights is used.

[0080] ;

[0081] in, This represents the velocity of the i-th particle in the j-th dimension at iteration k+1; w is the inertia weight. , Maximum number of iterations; learning factor ; , A random number in the interval [0,1]. Let be the optimal position of the i-th particle in the j-th dimension. Let i be the position of the i-th particle in the j-th dimension during the k-th iteration. Let be the globally optimal position of all particles in the j-th dimension;

[0082] Optimization objective: The expression for minimizing the objective function is:

[0083] ;

[0084] Where T is the end time. Let be the oxygen content deviation at time t. The slight positive pressure deviation at time t. To control the penalty coefficient, This represents the maximum absolute value of the controller's output (argon flow rate or valve opening change command) over the entire time period [0, T].

[0085] The optimization process is based on historical data from 100 heats categorized by steel type, ultimately generating an optimal initial parameter library corresponding to different steel types (such as alloy structural steel and stainless steel).

[0086] S4. Initialize the fuzzy PID controller using the optimized initial parameters, and adaptively adjust the parameters of the fuzzy PID controller based on the real-time signal and prediction results.

[0087] During the real-time control phase (sampling period 0.1s), the system retrieves initial parameters from the parameter library that match the current steel grade. Initialize the fuzzy PID controller.

[0088] Adaptive parameter adjustment is achieved through fuzzy inference:

[0089] Calculate the current oxygen content deviation and its rate of change ;

[0090] Will and Inputting data into the fuzzy inference system, the system generates real-time correction values ​​for PID parameters based on a pre-set fuzzy rule base. ;

[0091] Update the PID parameters using the following formula:

[0092] ;

[0093] in, Let k be the parameters of the fuzzy PID controller at time k. This is an adaptive coefficient used to limit the adjustment range in a single step and avoid system oscillation.

[0094] S5 calculates and outputs argon flow control commands and dust removal valve opening control commands based on the adjusted fuzzy PID controller parameters, in order to work together to maintain the set oxygen content and slight positive pressure inside the furnace.

[0095] The argon flow control command is derived from the predicted and compensated flow rate combined with smelting stage factors, and the calculation formula is as follows:

[0096] ;

[0097] in, This represents the control value of the argon flow rate at time k; This indicates a stage correction factor determined based on the smelting stage, with values ​​ranging from 0.8 to 1.2 depending on the different smelting stages such as arc initiation, arc stabilization, and crystallization.

[0098] The dust collector valve opening control command is calculated through a linkage algorithm to ensure coordinated action with the argon gas flow rate, thereby quickly stabilizing the furnace pressure. The calculation formula is as follows:

[0099] ;

[0100] in, This represents the control value of the dust collector valve opening at time k; This indicates the initial value of the dust collector valve opening. As a flow rate change compensation coefficient, when the argon flow rate increases, the dust removal valve should be closed appropriately in advance to suppress the pressure rise; This represents the pressure deviation compensation coefficient, used to quickly eliminate static deviations. This represents the slight positive pressure deviation at time k.

[0101] Constraint handling: The final instruction must satisfy... , To ensure the safe operation of the implementing agency.

[0102] As an optional technical solution, the control method of the present invention further includes step S6, triggering a self-evolution mechanism to achieve long-term self-optimization of the control system;

[0103] To ensure long-term consistency of control performance, this invention also incorporates a self-evolutionary mechanism. This mechanism is automatically triggered when the system detects that the steady-state error of oxygen content in three consecutive furnaces is greater than 0.5 ppm.

[0104] Re-execute the PSO global optimization (30 iterations) and update the initial parameter library with the new data;

[0105] Meanwhile, the DBN prediction model was incrementally trained using 50 new sets of real-time data to adapt the model to the latest changes in operating conditions.

[0106] A second aspect of the present invention provides a linkage control system for argon flow rate and micro-positive pressure in an electroslag furnace, comprising:

[0107] The data acquisition module, including an oxygen content sensor, a micro-positive pressure sensor, and a signal conditioning circuit, is responsible for real-time acquisition of data inside the furnace, specifically for real-time acquisition of oxygen content and micro-positive pressure data inside the furnace.

[0108] The controller is the core of the system. It can be an industrial PLC or an embedded industrial computer. It has an executable program stored inside. It can calculate and output argon flow control commands and dust removal valve opening control commands based on the control method described above, according to the real-time collected data.

[0109] The actuators, including argon flow control components (such as mass flow meters and electric regulating valves) and dust removal proportional valves, are used to receive and execute instructions issued by the controller.

[0110] To ensure effective control, the furnace structure features a dynamic seal made of Inconel 625 metal bellows, which is resistant to 1000℃ and compensates for displacement of ±15mm. The static seal uses silicone rubber sealing strips with a Shore A70 hardness and a compression set of <10% (200℃×72h), thus providing a reliable hardware foundation for maintaining a slight positive pressure.

[0111] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling the argon flow rate and micro-positive pressure in an electroslag furnace, characterized in that, Includes the following steps: S1 collects real-time oxygen content and micro-positive pressure signals within the electroslag furnace, including: The collected oxygen content signal is processed by Kalman filtering. The filtering formula is as follows: ； in, Here is the filtered oxygen content estimate at time k. The measured oxygen content at time k is... Kalman gain; The collected pressure signal is processed by moving average filtering. The filtering formula is as follows: ； in, The pressure estimate at time k is the filtered value. The measured pressure value at time k−i; S2, based on the acquired real-time signals, predicts the oxygen content and slight positive pressure inside the furnace at future moments using a deep belief network (DBN) model; when the deviation between the predicted future state and the set value exceeds a preset threshold, feedforward compensation is applied to the basic argon flow rate, and the compensation formula is: ； in, The predicted and compensated argon flow rate, The uncompensated base argon flow rate, To predict gain, This represents the predicted oxygen content at time k+n. Set the oxygen content value; S3, using the particle swarm optimization (PSO) algorithm to optimize the initial parameters of the fuzzy PID controller offline; S4, Initialize the fuzzy PID controller using the optimized initial parameters, and adaptively adjust the parameters of the fuzzy PID controller based on the real-time signal and prediction results; including: Based on real-time oxygen content deviation and its rate of change The correction amount of PID parameters is generated in real time through fuzzy inference. And update the parameters according to the following formula: ； in, Let k be the parameters of the fuzzy PID controller at time k. These are adaptive coefficients; S5, based on the adjusted fuzzy PID controller parameters, calculates and outputs argon flow control commands and dust collector valve opening control commands to collaboratively maintain the set oxygen content and slight positive pressure inside the furnace; the calculation formula for the dust collector valve opening command is: ； in, This represents the control value of the dust collector valve opening at time k; This indicates the initial value of the dust collector valve opening. This is the flow rate change compensation coefficient. This represents the control value of the argon flow rate at time k; This represents the pressure deviation compensation coefficient. This represents the slight positive pressure deviation at time k.

2. The method for linkage control of argon flow rate and micro-positive pressure in an electroslag furnace as described in claim 1, characterized in that, In step S2, the input features of the Deep Belief Network (DBN) model include real-time signals, historical trends, and smelting stage identifiers, and the output is the predicted oxygen content and pressure for a specific future sampling period; wherein, the historical trend is the rate of change of oxygen content deviation and the rate of change of pressure deviation over the previous several sampling periods.

3. The method for linkage control of argon flow rate and micro-positive pressure in an electroslag furnace as described in claim 1, characterized in that, Step S3 specifically includes: The Particle Swarm Optimization (PSO) algorithm iteratively optimizes the particle swarm based on historical smelting data, wherein the position code of each particle includes the initial parameters of the fuzzy PID controller and the center value of the fuzzy membership function. The particle velocity update formula for the PSO algorithm is as follows: ； in, Let represent the velocity of the i-th particle in the j-th dimension at iteration k+1; w is the inertia weight. , For learning factors; , A random number in the interval [0,1]. Let be the optimal position of the i-th particle in the j-th dimension. Let i be the position of the i-th particle in the j-th dimension during the k-th iteration. Let be the globally optimal position of all particles in the j-th dimension; Through iterative optimization using the PSO algorithm, an optimal initial parameter library corresponding to different steel grades or working conditions is obtained.

4. The method for linkage control of argon flow rate and micro-positive pressure in an electroslag furnace as described in claim 1, characterized in that, In step S5, the calculation formula for the argon flow control command is as follows: ； in, This represents the control value of the argon flow rate at time k; This indicates the stage correction factor determined based on the smelting stage.

5. The method for linkage control of argon flow rate and micro-positive pressure in an electroslag furnace as described in claim 1, characterized in that, It also includes step S6, which triggers the self-evolution mechanism when the system performance indicators fail to meet the requirements in multiple consecutive smelting cycles, re-executes the PSO algorithm to optimize global parameters, and uses newly added real-time data to incrementally train the DBN model.

6. A linkage control system for argon flow rate and micro-positive pressure in an electroslag furnace, characterized in that, include: The data acquisition module is used to collect real-time data on oxygen content and micro-positive pressure inside the furnace; The controller calculates and outputs argon flow control commands and dust removal valve opening control commands based on real-time collected data and the control method described in any one of claims 1-5. The actuator, including the argon flow control component and the dust removal proportional valve, is used to receive and execute the argon flow control command and the dust removal valve opening control command issued by the controller.

Images