Intelligent metallurgical electric furnace temperature monitoring system
Temperature prediction and control is carried out through multi-layer neural network and adaptive neural network algorithm, combined with improved PID and fuzzy logic control, the problem of insufficient response speed and low control accuracy of the metallurgical electric furnace temperature monitoring system is solved, and the high accuracy of the metallurgical process is achieved Temperature control and long-term adaptability of the system.
Patent Information
- Application Number
- CN202411002308.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-12
AI Technical Summary
The intelligent metallurgical electric furnace temperature monitoring system is insufficient in the face of rapidly changing metallurgy process, resulting in a hysteresis of temperature control, affecting the quality of metallurgical products, and it is difficult to accurately analyze and predict under complex conditions to reduce control accuracy.
Using multi-layer neural network model and adaptive neural network algorithm for real-time temperature analysis and prediction, combined with improved PID control algorithm and fuzzy logic control, optimize model parameters through online learning mechanism and particle swarm optimization algorithm to achieve rapid response and high-precision control .
Real-time accurate prediction and timely adjustment of temperature changes in metallurgical processes, strengthen the response speed and stability of temperature control, improve the adaptability and control accuracy of nonlinear metallurgical processes, ensure the stability of data transmission and the reliability of system sex.
Smart Images

Figure CN118936114A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of electric furnace temperature monitoring, and in particular to an intelligent metallurgical electric furnace temperature monitoring system. Background Art
[0002] The intelligent metallurgical furnace temperature monitoring system usually consists of the following parts: Temperature sensor: responsible for real-time monitoring of the temperature inside the furnace. Commonly used temperature sensors include thermocouples, infrared thermometers, etc. Signal processing module: amplifies, filters and converts the temperature signal collected by the sensor to ensure the accuracy and stability of the data. Data transmission module: transmits the processed temperature data to the central control system. Data transmission methods include wired transmission and wireless transmission. Central control system: receives and processes the transmitted temperature data, and controls the furnace according to the set temperature range to ensure the temperature accuracy during the smelting process. Human-machine interface: provides a friendly user interface, displays temperature data, alarm information, etc., to facilitate operators to monitor and adjust. The system collects the temperature data inside the furnace in real time through the temperature sensor, and transmits the data to the central control system after processing it through the signal processing module. The central control system analyzes and judges the temperature data according to the pre-set parameters, and adjusts the heating power of the furnace through the feedback control loop to ensure the temperature stability of the metallurgical process. The human-machine interface is used to display and record temperature changes for reference by operators.
[0003] Although the intelligent metallurgical furnace temperature monitoring system plays an important role in the metallurgical industry, it still has some defects: the temperature changes very quickly in the metallurgical process, and if the system's response speed is not fast enough, it may cause temperature control lag and fail to adjust the heating power of the furnace in time, thus affecting the quality of metallurgical products. Although modern intelligent systems already have strong data processing capabilities, in the face of complex and changeable metallurgical processes, the system is sometimes difficult to accurately analyze and predict, resulting in reduced control accuracy. Summary of the invention
[0004] In view of the deficiencies in the prior art, the present invention provides an intelligent metallurgical electric furnace temperature monitoring system, which solves the problem that the temperature changes very quickly in the metallurgical process lead to temperature control lag, and the heating power of the electric furnace cannot be adjusted in time, thus affecting the quality of metallurgical products; in the face of complex and changeable metallurgical process systems, it is sometimes difficult to perform accurate analysis and prediction, resulting in a decrease in control accuracy.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent metallurgical furnace temperature monitoring system, comprising:
[0006] Temperature sensor module, used to monitor the temperature inside the electric furnace in real time;
[0007] A signal processing module, connected to the temperature sensor module, for amplifying, filtering and converting the collected temperature signal;
[0008] A data transmission module, connected to the signal processing module, for transmitting the processed temperature data to a central control system;
[0009] A central control system, including real-time analysis, PID control algorithm, online learning mechanism and prediction model, receives the temperature data transmitted by the data transmission module, analyzes and predicts temperature changes, and adjusts the heating power of the electric furnace according to a preset control strategy;
[0010] The human-machine interface module is connected to the central control system and is used to display temperature data, alarm information and system status.
[0011] Preferably, the signal processing module includes an analog-to-digital converter and a digital filter, which are used to convert the analog temperature signal into a digital signal and filter out noise. The temperature sensor module includes multiple thermocouples and infrared thermometers, which are distributed at different positions of the electric furnace to improve the accuracy and coverage of temperature monitoring.
[0012] Preferably, the data transmission module includes a wireless transmission unit for realizing remote data transmission and improving the flexibility of the system.
[0013] Preferably, the real-time analysis and prediction model of the central control system is based on a multi-layer neural network, which can accurately predict the temperature under the complex and changeable conditions of the metallurgical process. The model formula is:
[0014] y(t+1)=f(W,X(t));
[0015] Among them, y(t+1) is the predicted temperature value, f is the neural network function, W is the weight matrix of the neural network, and X(t) is the input temperature data. The multi-layer neural network model has been trained with a large amount of metallurgical process data and can adapt to different metallurgical process conditions to improve the prediction accuracy of the system.
[0016] Preferably, the PID control algorithm can adjust the PID parameters in real time according to the temperature change by introducing an adaptive adjustment mechanism, thereby improving the response speed and stability of temperature control. The control formula is:
[0017]
[0018] Among them, K p , K i and K dis the adaptively adjusted PID parameter, e(t) is the current temperature error, and the intelligent metallurgical furnace temperature monitoring system, wherein the central control system uses a particle swarm optimization algorithm to optimize the parameters of the neural network model. The formula of the particle swarm optimization algorithm is:
[0019]
[0020] in, and are the velocity and position of the ith particle in generation t, is the historical optimal position of the ith particle, g t is the global optimal position, ω, c1, c2 are weight coefficients, r1, r2 are random numbers, and the central control system performs temperature control based on model predictive control. The formula of model predictive control is:
[0021]
[0022] Wherein, J is the objective function, r(t+k) is the reference temperature, y(t+k|t) is the predicted temperature, Δu(t+k) is the control increment, and λ is the weight factor. The prediction model in the model predictive control algorithm is based on the state space equation, and the formula is:
[0023]
[0024] Where x(t) is the state vector, u(t) is the control input, y(t) is the output vector, and A, B, C, and D are state space matrices.
[0025] Preferably, the digital filter in the signal processing module adopts a Kalman filter algorithm to further improve the accuracy of the temperature data. The filtering formula is:
[0026]
[0027] in, is the updated state estimate, K k is the Kalman gain, z k is the measurement value, and H is the measurement matrix.
[0028] Preferably, the wireless transmission unit of the data transmission module adopts LoRa technology to achieve long-distance, low-power data transmission. The central control system is equipped with a redundant control module for taking over control when the main control system fails to ensure the reliability of the system. The temperature sensor module, signal processing module, data transmission module and central control system are connected through a standardized interface to facilitate system maintenance and upgrading.
[0029] Preferably, the PID control algorithm improves the adaptability and control accuracy to nonlinear metallurgical processes by introducing fuzzy logic control. The fuzzy control formula is:
[0030]
[0031] Among them, fuzzy(e) is the error value after fuzzy logic processing.
[0032] Preferably, the adaptive neural network algorithm includes an input layer, multiple hidden layers and an output layer, wherein the input layer receives real-time temperature data, the hidden layer performs feature extraction and pattern recognition, and the output layer generates a predicted temperature value and an adjustment signal. The model formula is:
[0033] y(t+1)=σ(W h σ(W i X(t)+b i )+b h );
[0034] Among them, σ is the activation function, W i and W h are the weight matrices of the input layer and the hidden layer, respectively, i and b h is the bias term, and the formula of the PID control algorithm is:
[0035]
[0036] Among them, K p , K i and K d is the adaptively adjusted PID parameter, e(t) is the current temperature error, and the online learning mechanism continuously updates the weights and biases of the neural network model according to the latest data through an incremental learning algorithm. The incremental learning algorithm formula is:
[0037] W(t+1)=W(t)+ηΔW;
[0038] Among them, W(t) is the current weight, η is the learning rate, and ΔW is the weight update value.
[0039] The present invention provides an intelligent metallurgical furnace temperature monitoring system, which has the following beneficial effects:
[0040] This intelligent metallurgical furnace temperature monitoring system, by using a multi-layer neural network model and an adaptive neural network algorithm, realizes real-time analysis and accurate prediction of temperature changes in the metallurgical process, ensuring the timeliness and accuracy of temperature control. By introducing an improved PID control algorithm and fuzzy logic control, the system can make adaptive adjustments based on real-time temperature data, improving the adaptability and control accuracy of nonlinear metallurgical processes.
[0041] Analog-to-digital converters and digital filters are used to process temperature data, and long-distance, low-power data transmission is achieved through LoRa technology, ensuring the stability and reliability of data transmission. The central control system adopts online learning mechanism and incremental learning algorithm to continuously optimize the weights and biases of the neural network model according to the latest data, thereby improving the long-term adaptability and control accuracy of the system. At the same time, the particle swarm optimization algorithm further optimizes the parameters of the neural network model. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 It is a schematic diagram of the internal structure of the present invention. DETAILED DESCRIPTION
[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0044] like Figure 1 As shown, an embodiment of the present invention provides an intelligent metallurgical furnace temperature monitoring system, including a temperature sensor module for real-time monitoring of the temperature inside the furnace.
[0045] The signal processing module is connected to the temperature sensor module and is used to amplify, filter and convert the collected temperature signal. The signal processing module includes an analog-to-digital converter and a digital filter, which are used to convert the analog temperature signal into a digital signal and filter out noise. The temperature sensor module includes multiple thermocouples and infrared thermometers, which are distributed at different positions of the electric furnace to improve the accuracy and coverage of temperature monitoring. The digital filter in the signal processing module adopts the Kalman filtering algorithm to further improve the accuracy of the temperature data. The filtering formula is:
[0046]
[0047] in, is the updated state estimate, K k is the Kalman gain, z k is the measurement value, and H is the measurement matrix.
[0048] The data transmission module is connected to the signal processing module and is used to transmit the processed temperature data to the central control system. The data transmission module includes a wireless transmission unit for realizing remote data transmission and improving the flexibility of the system. The wireless transmission unit of the data transmission module adopts LoRa technology to realize long-distance, low-power data transmission. The central control system is equipped with a redundant control module for taking over control when the main control system fails to ensure the reliability of the system. The temperature sensor module, signal processing module, data transmission module and central control system are connected through standardized interfaces to facilitate system maintenance and upgrades.
[0049] The central control system, including real-time analysis, PID control algorithm, online learning mechanism and prediction model, receives the temperature data transmitted by the data transmission module, analyzes and predicts the temperature change, and adjusts the heating power of the electric furnace according to the preset control strategy. The real-time analysis and prediction model of the central control system is based on a multi-layer neural network, which can accurately predict the temperature under the complex and changeable conditions of the metallurgical process. The model formula is:
[0050] y(t+1)=f(W,X(t));
[0051] Among them, y(t+1) is the predicted temperature value, f is the neural network function, W is the weight matrix of the neural network, and X(t) is the input temperature data. The multi-layer neural network model has been trained with a large amount of metallurgical process data and can adapt to different metallurgical process conditions and improve the prediction accuracy of the system. The adaptive neural network algorithm includes an input layer, multiple hidden layers and an output layer. The input layer accepts real-time temperature data, the hidden layer performs feature extraction and pattern recognition, and the output layer generates predicted temperature values and adjustment signals. The model formula is:
[0052] y(t+1)=σ(W h σ(W i X(t)+b i )+b h );
[0053] Among them, σ is the activation function, W i and W h are the weight matrices of the input layer and the hidden layer, respectively, i and b h The PID control algorithm introduces an adaptive adjustment mechanism to adjust the PID parameters in real time according to temperature changes, thereby improving the response speed and stability of temperature control. The control formula is:
[0054]
[0055] Among them, K p , K i and K dis the adaptively adjusted PID parameter, e(t) is the current temperature error, and the intelligent metallurgical furnace temperature monitoring system, in which the central control system uses the particle swarm optimization algorithm to optimize the parameters of the neural network model. The formula of the particle swarm optimization algorithm is:
[0056]
[0057] in, and are the velocity and position of the ith particle in generation t, is the historical optimal position of the ith particle, g t is the global optimal position, ω, c1, c2 are weight coefficients, r1, r2 are random numbers, and the central control system performs temperature control based on model predictive control. The formula of model predictive control is:
[0058]
[0059] Among them, J is the objective function, r(t+k) is the reference temperature, y(t+k|t) is the predicted temperature, Δu(t+k) is the control increment, λ is the weight factor, and the prediction model in the model predictive control algorithm is based on the state space equation, and the formula is:
[0060]
[0061] Among them, x(t) is the state vector, u(t) is the control input, y(t) is the output vector, A, B, C and D are the state space matrices. The PID control algorithm improves the adaptability and control accuracy of nonlinear metallurgical processes by introducing fuzzy logic control. The fuzzy control formula is:
[0062]
[0063] Among them, fuzzy(e) is the error value after fuzzy logic processing.
[0064] The human-machine interface module is connected to the central control system to display temperature data, alarm information and system status. The online learning mechanism continuously updates the weights and biases of the neural network model based on the latest data through an incremental learning algorithm. The incremental learning algorithm formula is:
[0065] W(t+1)=W(t)+ηΔW;
[0066] Among them, W(t) is the current weight, η is the learning rate, and ΔW is the weight update value.
[0067] Table 1: Data comparison table
[0068]
[0069] Table 2: Comparative experimental data of the response and control accuracy of the present invention and the prior art under different temperature change conditions
[0070] Time (seconds) Target temperature ℃ The system temperature of the present invention is ℃ Current system temperature ℃ 0 1500 1500 1500 1 1520 1519.8 1517 2 1540 1539.5 1530 3 1560 1559.9 1552 4 1580 1579.6 1571 5 1600 1599.8 1590 6 1580 1579.9 1575 7 1560 1559.7 1553 8 1540 1539.8 1532 9 1520 1519.9 1515
[0071] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. Intelligent metallurgical furnace temperature monitoring system, characterized in that: include: Temperature sensor module, used to monitor the temperature inside the electric furnace in real time; A signal processing module, connected to the temperature sensor module, for amplifying, filtering and converting the collected temperature signal; A data transmission module, connected to the signal processing module, for transmitting the processed temperature data to a central control system; A central control system, including real-time analysis, PID control algorithm, online learning mechanism and prediction model, receives the temperature data transmitted by the data transmission module, analyzes and predicts temperature changes, and adjusts the heating power of the electric furnace according to a preset control strategy; The human-machine interface module is connected to the central control system and is used to display temperature data, alarm information and system status.
2. The intelligent metallurgical furnace temperature monitoring system according to claim 1 is characterized in that: The signal processing module includes an analog-to-digital converter and a digital filter, which are used to convert the analog temperature signal into a digital signal and filter out noise. The temperature sensor module includes a plurality of thermocouples and infrared thermometers, which are distributed at different positions of the electric furnace.
3. The intelligent metallurgical furnace temperature monitoring system according to claim 1 is characterized in that: The data transmission module includes a wireless transmission unit for realizing remote data transmission.
4. The intelligent metallurgical furnace temperature monitoring system according to claim 1 is characterized in that: The real-time analysis and prediction model of the central control system is based on a multi-layer neural network, which can accurately predict the temperature under the complex and changeable conditions of the metallurgical process. The model formula is: y(t+1)=f(W,X(t)); Among them, y(t+1) is the predicted temperature value, f is the neural network function, W is the weight matrix of the neural network, and X(t) is the input temperature data. The multi-layer neural network model has been trained with a large amount of metallurgical process data and can adapt to different metallurgical process conditions.
5. The intelligent metallurgical furnace temperature monitoring system according to claim 1 is characterized in that: The PID control algorithm introduces an adaptive adjustment mechanism to adjust the PID parameters in real time according to temperature changes, thereby improving the response speed and stability of temperature control. The control formula is: Among them, K p , K i and K d is the adaptively adjusted PID parameter, e(t) is the current temperature error, and the intelligent metallurgical furnace temperature monitoring system, wherein the central control system uses a particle swarm optimization algorithm to optimize the parameters of the neural network model. The formula of the particle swarm optimization algorithm is: in, and are the velocity and position of the ith particle in generation t, is the historical optimal position of the ith particle, g t is the global optimal position, ω, c1, c2 are weight coefficients, r1, r2 are random numbers, and the central control system performs temperature control based on model predictive control. The formula of model predictive control is: Wherein, J is the objective function, r(t+k) is the reference temperature, y(t+k|t) is the predicted temperature, Δu(t+k) is the control increment, and λ is the weight factor. The prediction model in the model predictive control algorithm is based on the state space equation, and the formula is: Where x(t) is the state vector, u(t) is the control input, y(t) is the output vector, and A, B, C, and D are state space matrices.
6. The intelligent metallurgical furnace temperature monitoring system according to claim 1 is characterized by: The digital filter in the signal processing module adopts the Kalman filter algorithm to further improve the accuracy of the temperature data. The filtering formula is: in, is the updated state estimate, K k is the Kalman gain, z k is the measurement value, and H is the measurement matrix.
7. The intelligent metallurgical furnace temperature monitoring system according to claim 1 is characterized in that: The wireless transmission unit of the data transmission module adopts LoRa technology to achieve long-distance, low-power data transmission. The central control system is equipped with a redundant control module for taking over control when the main control system fails to ensure the reliability of the system. The temperature sensor module, signal processing module, data transmission module and central control system are connected through a standardized interface to facilitate system maintenance and upgrades.
8. The intelligent metallurgical furnace temperature monitoring system according to claim 1 is characterized by: The PID control algorithm improves the adaptability and control accuracy of nonlinear metallurgical processes by introducing fuzzy logic control. The fuzzy control formula is: Among them, fuzzy(e) is the error value after fuzzy logic processing.
9. The intelligent metallurgical furnace temperature monitoring system according to claim 1 is characterized by: The adaptive neural network algorithm includes an input layer, multiple hidden layers and an output layer, wherein the input layer receives real-time temperature data, the hidden layer performs feature extraction and pattern recognition, and the output layer generates predicted temperature values and adjustment signals. The model formula is: y(t+1)=σ(W h σ(W i X(t)+b i )+b h ); Among them, σ is the activation function, W i and W h are the weight matrices of the input layer and the hidden layer, respectively, i and b h The online learning mechanism continuously updates the weights and biases of the neural network model according to the latest data through an incremental learning algorithm. The incremental learning algorithm formula is: W(t+1)=W(t)+ηΔW; Among them, W(t) is the current weight, η is the learning rate, and ΔW is the weight update value.
Citation Information
Cited By
Temperature control system and method in plastic particle production process
CN119717541A
Environment control system and control method for large-scale server cluster
CN119861575A
Pyrometallurgy copper temperature abnormity identification system and method based on machine learning and platinum-rhodium thermocouple
CN120907679A
Industrial furnace automatic temperature control and safety monitoring system
CN121297494A
Industrial furnace intelligent temperature control system based on data acquisition
CN121557752A
