Intelligent control system for energy-saving continuous up-casting furnace based on machine learning

Abstract

The invention discloses an intelligent control system for an energy-saving continuous up-casting furnace based on machine learning, and particularly relates to the technical field of artificial intelligence, which comprises a data acquisition and processing module, an intelligent control module, a composite heating module, a waste heat recovery module and a safety monitoring and protection module, the data acquisition and processing module is used for acquiring, processing and storing data of temperature, flow velocity, pressure and gas components into a database, the intelligent control module is used for dynamically adjusting heating power and traction speed, generating a control instruction and sending the control instruction to the composite heating module, and the composite heating module is used for controlling the traction speed according to the instruction of the intelligent control module. The resistance heating module is used for distributing the proportion of resistance heating and fuel gas heating, the waste heat recovery module is used for recovering waste gas waste heat of a furnace body, driving an ORC power generation unit through the waste heat and converting the waste heat into electric energy, and the safety monitoring and protecting module is used for configuring multiple sensor networks in key links of the casting furnace and monitoring the safe operation state.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and more specifically, to an intelligent control system for an energy-saving continuous upward drawing melting and casting furnace based on machine learning. Background Art

[0002] In industrial production, the melting and casting process is a key link in metal processing. Especially in the production of aluminum alloys and other metals, the continuous upward drawing melting and casting furnace, as an efficient melting and casting equipment, is widely used due to its high production efficiency, low quality fluctuation, and good energy utilization rate.

[0003] Traditional melting and casting furnace control systems usually rely on traditional PID control or empirical models. These control methods are difficult to achieve precise energy efficiency control and adaptive optimization in complex production environments. With the development of industrial automation and information technology, machine learning, as a technology capable of processing large-scale data and extracting potential laws, has gradually been applied to the optimization and intelligent control of industrial processes. By constructing complex mathematical models, machine learning can monitor the working state of the melting and casting furnace in real time, accurately predict key parameters such as furnace temperature and fuel consumption, and implement a more efficient and energy-saving control strategy. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an intelligent control system for an energy-saving continuous upward drawing melting and casting furnace based on machine learning to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solution. An intelligent control system for an energy-saving continuous upward drawing melting and casting furnace based on machine learning includes a data acquisition and processing module, an intelligent control module, a composite heating module, a waste heat recovery module, and a safety monitoring and protection module;

[0006] The data acquisition and processing module arranges multi-parameter sensors at key positions of the melting and casting furnace to collect temperature, flow rate, pressure, and gas composition data, processes the collected data, and stores the processed data in a database;

[0007] The intelligent control module uses a deep learning model to analyze the collected data, constructs a relationship model between temperature, flow rate, and pressure, dynamically adjusts the heating power and drawing speed according to the input of temperature, flow rate, and pressure data, introduces a feedback adjustment mechanism, generates control instructions, and sends them to the composite heating module;

[0008] The composite heating module distributes the ratio of resistance heating and gas heating according to the instructions of the intelligent control module to achieve the goals of rapid temperature rise and temperature uniformity;

[0009] The waste heat recovery module uses a heat pipe heat exchanger to recover the waste heat of the furnace body exhaust gas, and uses medium and low temperature waste heat to drive the ORC power generation unit to convert the waste heat into electric energy, reducing the equipment's dependence on external power supplies;

[0010] The safety monitoring and protection module configures multiple sensor networks at key links of the melting furnace, including thermocouples, pressure sensors, and highly sensitive gas leak detectors, monitors the safe operating status, and sets safety thresholds. When the monitored data exceeds the safety limit, an alarm is immediately issued and the alarm event is recorded.

[0011] In a preferred embodiment, the data acquisition and processing module arranges multi-parameter sensors at key positions of the melting furnace to collect temperature, flow rate, pressure, and gas composition data, processes the collected data, and stores the processed data in a database. The specific steps are as follows:

[0012] Step A1, data acquisition: Arrange various types of sensors in the melting zone, heat preservation zone, and flow zone of the melting furnace, including high-precision thermocouples, flow rate sensors, pressure sensors, and gas composition analyzers, to collect the molten metal temperature T, the flow rate V of the molten metal, the furnace internal pressure P, and the gas composition data in the exhaust gas, including oxygen C1, nitrogen C2, and carbon monoxide C3. For each time point n, the collected data is represented as D(n) = {T(n), V(n), P(n), C1(n), C2(n), C3(n)}, where D(n) represents the multi-dimensional data obtained at the nth sampling moment, T(n) represents the molten metal temperature collected at time point n, V(n) represents the molten metal flow rate collected at time point n, P(n) represents the furnace internal pressure collected at time point n, and C1(n), C2(n), C3(n) respectively represent the gas composition concentrations collected at time point n;

[0013] Step A2, data processing: Based on signal processing and data preprocessing algorithms, remove noise, perform real-time filtering and outlier detection on the collected multi-dimensional data, integrate it through multi-channel data fusion technology, and store it in the database according to the time series.

[0014] In a preferred embodiment, the intelligent control module uses a deep learning model to analyze the collected data, constructs a relationship model between temperature, flow rate, and pressure, dynamically adjusts the heating power and traction speed according to the input of temperature, flow rate, and pressure data, and introduces a feedback adjustment mechanism to generate control instructions and send them to the composite heating module. The specific steps are as follows:

[0015] Step B1, deep learning model training: Extract the historical data set from the database as the input of the deep learning model, and use a multi-layer perceptron to predict the control output, including heating power and traction speed. It further includes the following steps:

[0016] Step B101: Extract the historical data set from the database and segment it according to the time series. Each time point n corresponds to a data vector {T(n), V(n), P(n), C1(n), C2(n), C3(n)}, which is used as the input of the deep learning model;

[0017] Step B102: Construct a multi-layer perceptron model, including an input layer, a hidden layer, and an output layer. The number of nodes in the input layer is equal to the number of input features. The output layer has two neurons, which are used to predict the heating power and the traction speed respectively. For the L-layer perceptron network, the input layer is represented as the 0th layer, and the output layer is represented as the Lth layer. The input and output of the lth layer are represented as h l and w l , and the input from the input layer to the 1st layer is h1 = f(W1x(n) + b1). The input from the lth layer to the l+1th layer is h l+1 = f(W l+1 h l + b l+1 ). The output of the output layer is y(n) = W L h L + b L = [P heat (n), V traction (n)], where W L is the weight matrix of the last layer, b L is the bias term of the last layer, h L is the output of the last layer, W l+1 is the weight matrix of the l+1th layer, h l+1 is the output of the l+1th layer, b l+1 is the bias term of the l+1th layer, W1 is the weight matrix of the 1st layer, b1 is the bias term of the 1st layer, f is the activation function, x(n) is the input vector, are the predicted heating power and traction speed;

[0018] Step B103: Use the mean square error loss function to calculate the difference between the predicted value and the true value. The formula is: where, [P heat (n), V traction (n)] is the actual control value, are the heating power and traction speed predicted by the model, and N is the total number of training samples;

[0019] Step B2: Feedback regulation: Generate a preliminary control instruction according to the predicted output of the deep learning model and Adjust the heating power according to the difference between the real-time temperature and the target temperature, and the adjusted heating power is Adjust the traction speed according to the difference between the real-time flow rate and the target flow rate. The adjusted traction speed is where K heat is the feedback gain of the heating power, and K traction is the feedback gain of the traction speed. T target and V target are the target temperature and the target flow rate respectively. T actual (n) and V actual (n) are the actual temperature and flow rate respectively. Send the heating power and traction speed after feedback adjustment as the final control instruction to the composite heating module. The final control instruction is: u(n) = {P′ heat (n), V′ traction (n)}, ensuring that the temperature and flow rate are within the set range during the casting process.

[0020] In a preferred embodiment, the composite heating module distributes the proportion of resistance heating and gas heating according to the instruction of the intelligent control module to achieve the goals of rapid heating and temperature uniformity. The specific steps are as follows:

[0021] Step C1. Set the proportionality factor: Receive the output control instruction u(n) = {P′ heat (n), V′ traction (n)} from the intelligent control module. According to the required heating power P′ heat (n), set the target heating power and set a proportionality factor α for controlling the distribution ratio between resistance heating and gas heating. The proportionality factor α is in the range of [0, 1], representing the power ratio of resistance heating to gas heating. When α = 1, it means that all the heating power is provided by resistance heating; when α = 0, it means that all the heating power is provided by gas heating.

[0022] Step C2. Distribute the heating power: Calculate the power distribution of resistance heating and gas heating according to the target heating power P′ heat (n) and the proportionality factor α. The specific calculation formula is: P res (n) = α·P′ heat (n), P gas (n) = (1 - α)·P′ heat (n), where P res (n) is the power of resistance heating, and P gas (n) is the power of gas heating. Adjust the power of the resistance heater and the gas heater respectively according to the distributed power, and monitor the temperature and flow rate in real time during the heating process to achieve the required heating effect.

[0023] In a preferred embodiment, the waste heat recovery module uses a heat pipe heat exchanger to recover the waste heat of the furnace body exhaust gas, and uses medium and low temperature waste heat to drive the ORC power generation unit to convert the waste heat into electric energy. The specific steps are as follows:

[0024] Step D1: Use a heat pipe heat exchanger to transfer the waste heat in the exhaust gas to the working medium through the heat pipe for heat recovery. Set the effective heat transfer efficiency of the heat exchanger as η heat , and calculate the recovered heat power from the exhaust gas. The specific formula is where P recover (n) is the recovered heat power from the exhaust gas, is the exhaust gas flow rate, C p is the specific heat capacity of the exhaust gas, T flue,in , T flue,out are the temperature of the exhaust gas when entering the heat exchanger and the temperature of the exhaust gas when leaving the heat exchanger respectively;

[0025] Step D2: Input the recovered heat power into the ORC power generation unit and calculate the electric energy output from the ORC power generation unit. The specific formula is: E output (n) = [η ORC ·P recover (n)]·t, where E output (n) is the conversion efficiency of the ORC power generation unit, t is the time duration, and η ORC is the conversion efficiency of the ORC power generation unit.

[0026] In a preferred embodiment, the safety monitoring and protection module configures a multiple sensor network at the key links of the melting and casting furnace, including thermocouples, pressure sensors and high-sensitivity gas leak detectors, monitors the safe operation status, and sets safety thresholds. When the monitored data exceeds the safety limit, an alarm is immediately issued and the alarm event is recorded. The specific steps are as follows:

[0027] Step E1: Sensor data monitoring: Set the safety thresholds of the temperature sensor as T max and T min , the safety thresholds of the pressure sensor as P max and P min , the safety threshold of the gas concentration as C max , and compare the real-time data obtained by each sensor with its preset safety threshold;

[0028] Step E2, Alarm Triggering and Response: When the monitoring data of any sensor exceeds the set safety threshold, an alarm is immediately triggered, and safety protection measures are initiated. When the alarm is triggered, the alarm event is automatically recorded, including the alarm type, occurrence time, and specific threshold that triggered it. Different response measures are taken according to the level and type of the alarm event. When the temperature and pressure exceed the limit, the casting furnace is automatically shut down, and alarm information is sent to the operator via text message and email.

[0029] The beneficial effects of the present invention are as follows: Multi-parameter sensors are arranged at key positions of the casting furnace to collect temperature, flow rate, pressure, and gas composition data. The collected data is processed and stored in a database. The collected data is analyzed using a deep learning model to construct a relationship model between temperature, flow rate, and pressure. According to the input of temperature, flow rate, and pressure data, the heating power and traction speed are dynamically adjusted, and a feedback adjustment mechanism is introduced to generate control instructions. According to the generated control instructions, the ratio of resistance heating and gas heating is allocated to achieve the goals of rapid heating and temperature uniformity. A heat pipe heat exchanger is used to recover the waste heat of the furnace body exhaust gas, and a medium and low temperature waste heat-driven ORC power generation unit is used to convert the waste heat into electrical energy, reducing the equipment's dependence on external power sources. A multi-sensor network, including thermocouples, pressure sensors, and highly sensitive gas leak detectors, is configured at key links of the casting furnace to monitor the safe operating state and set safety thresholds. When the monitoring data exceeds the safety limit, an alarm is immediately issued and the alarm event is recorded, which can reduce the accident risk and ensure the safety of personnel and equipment. By real-time monitoring and dynamically adjusting the heating power, the present invention can improve the energy utilization efficiency, reduce temperature fluctuations and uneven heating through optimized thermal management, enhance the consistency and quality of casting products, and reduce the defect rate. Brief Description of the Drawings

[0030] Figure 1 It is the system flow chart of the present invention. Detailed Embodiments

[0031] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0032] In the description of the present application, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present application, the meaning of "a plurality" is two or more, unless otherwise specifically defined.

[0033] In the description of the present application, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present application is not necessarily construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without the use of these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope that conforms to the principles and features disclosed in the present application.

[0034] Embodiment 1

[0035] This embodiment provides an intelligent control system for an energy-saving continuous up-drawing melting and casting furnace based on machine learning as shown in Figure 1 and specifically includes a data acquisition and processing module, an intelligent control module, a composite heating module, a waste heat recovery module, and a safety monitoring and protection module;

[0036] The data acquisition and processing module arranges multi-parameter sensors at key positions of the melting and casting furnace to collect temperature, flow rate, pressure, and gas composition data, processes the collected data, and stores the processed data in a database;

[0037] The intelligent control module uses a deep learning model to analyze the collected data, constructs a relationship model between temperature, flow rate, and pressure, dynamically adjusts the heating power and drawing speed according to the input of temperature, flow rate, and pressure data, introduces a feedback adjustment mechanism, generates control instructions, and sends them to the composite heating module;

[0038] The composite heating module distributes the ratio of resistance heating and gas heating according to the instructions of the intelligent control module to achieve the goals of rapid temperature rise and temperature uniformity;

[0039] The waste heat recovery module uses a heat pipe heat exchanger to recover the waste heat of the furnace body exhaust gas and uses the medium and low temperature waste heat to drive an ORC power generation unit to convert the waste heat into electric energy, reducing the equipment's dependence on external power supplies;

[0040] The safety monitoring and protection module configures a multi-sensor network at key links of the melting furnace, including thermocouples, pressure sensors, and highly sensitive gas leak detectors, monitors the safe operating state, and sets safety thresholds. When the monitored data exceeds the safety limit, an alarm is immediately issued and the alarm event is recorded.

[0041] In this embodiment, it is specifically necessary to describe the data acquisition and processing module. The data acquisition and processing module arranges multi-parameter sensors at key positions of the melting furnace to collect temperature, flow rate, pressure, and gas composition data, and processes the collected data, and stores the processed data in the database. The specific steps are as follows:

[0042] Step A1, data acquisition: Arrange various types of sensors in the melting zone, heat preservation zone, and flow zone of the melting furnace, including high-precision thermocouples, flow rate sensors, pressure sensors, and gas composition analyzers, to collect the molten metal temperature T, the flow rate V of the molten metal, the furnace internal pressure P, and the gas composition data in the waste gas, including oxygen C1, nitrogen C2, and carbon monoxide C3. For each time point n, the collected data is expressed as D(n) = {T(n), V(n), P(n), C1(n), C2(n), C3(n)}, where D(n) represents the multi-dimensional data obtained at the nth sampling moment, T(n) represents the molten metal temperature collected at time point n, V(n) represents the flow rate of the molten metal collected at time point n, P(n) represents the furnace internal pressure collected at time point n, and C1(n), C2(n), C3(n) respectively represent the gas composition concentrations collected at time point n;

[0043] Step A2, data processing: Based on signal processing and data preprocessing algorithms, remove noise and perform real-time filtering and outlier detection on the collected multi-dimensional data, integrate it through multi-channel data fusion technology, and store it in the database according to the time series.

[0044] In this embodiment, it is specifically necessary to describe the intelligent control module. The intelligent control module uses a deep learning model to analyze the collected data, constructs a relationship model between temperature, flow rate, and pressure, dynamically adjusts the heating power and traction speed according to the input of temperature, flow rate, and pressure data, and introduces a feedback adjustment mechanism to generate control instructions and send them to the composite heating module. The specific steps are as follows:

[0045] Step B1, deep learning model training: Extract the historical data set from the database as the input of the deep learning model, and use a multi-layer perceptron to predict the control output, including heating power and traction speed. Further, it includes the following steps:

[0046] Step B101: Extract the historical data set from the database and segment it according to the time series. Each time point n corresponds to a data vector {T(n), V(n), P(n), C1(n), C2(n), C3(n)}, which is used as the input of the deep learning model;

[0047] Step B102: Construct a multi-layer perceptron model, including an input layer, a hidden layer, and an output layer. The number of nodes in the input layer is equal to the number of input features. The output layer has two neurons, which are used to predict the heating power and the traction speed respectively. For the L-layer perceptron network, the input layer is represented as the 0th layer, and the output layer is represented as the Lth layer. The input and output of the lth layer are represented as h l and w l , and the input from the input layer to the 1st layer is h1 = f(W1x(n) + b1). The input from the lth layer to the l+1th layer is h l+1 = f(W l+1 h l + b l+1 ). The output of the output layer is y(n) = W L h L + b L = [P heat (n), V traction (n)], where W L is the weight matrix of the last layer, b L is the bias term of the last layer, h L is the output of the last layer, W l+1 is the weight matrix of the l+1th layer, h l+1 is the output of the l+1th layer, b l+1 is the bias term of the l+1th layer, W1 is the weight matrix of the 1st layer, b1 is the bias term of the 1st layer, f is the activation function, x(n) is the input vector, are the predicted heating power and traction speed;

[0048] Step B103: Use the mean square error loss function to calculate the difference between the predicted value and the true value. The formula is: where, [P heat (n), V traction (n)] is the actual control value, are the heating power and traction speed predicted by the model, and N is the total number of training samples;

[0049] Step B2: Feedback regulation: Generate a preliminary control instruction according to the predicted output of the deep learning model and Adjust the heating power according to the difference between the real-time temperature and the target temperature. The adjusted heating power is Adjust the traction speed according to the difference between the real-time flow rate and the target flow rate. The adjusted traction speed is where K heat is the feedback gain of the heating power, K traction is the feedback gain of the traction speed, T target and V target are the target temperature and the target flow rate respectively, T actual (n) and V actual (n) are the actual temperature and flow rate respectively. Send the heating power and traction speed after feedback adjustment as the final control instruction to the composite heating module. The final control instruction is: u(n) = {P′ heat (n), V′ traction (n)}, ensuring that the temperature and flow rate are within the set range during the casting process.

[0050] In this embodiment, specifically, the composite heating module needs to be described. The composite heating module distributes the proportion of resistance heating and gas heating according to the instruction of the intelligent control module to achieve the goals of rapid heating and temperature uniformity. The specific steps are as follows:

[0051] Step C1. Set the proportionality factor: Receive the output control instruction u(n) = {P′ heat (n), V′ traction (n)} from the intelligent control module. Set the target heating power according to the required heating power P′ heat (n). Set a proportionality factor α for controlling the distribution ratio between resistance heating and gas heating. The proportionality factor α is within the range of [0, 1], representing the power ratio of resistance heating to gas heating. When α = 1, it means that all the heating power is provided by resistance heating. When α = 0, it means that all the heating power is provided by gas heating. By adjusting α, the heating rate and temperature uniformity can be flexibly adjusted;

[0052] Step C2. Distribute the heating power: Calculate the power distribution of resistance heating and gas heating according to the target heating power P′ heat (n) and the proportionality factor α. The specific calculation formula is: P res (n) = α·P′ heat (n), P gas (n) = (1 - α)·P′ heat (n), where P res (n) is the power of resistance heating, P gas (n) is the power of gas heating. Adjust the power of the resistance heater and the gas heater respectively according to the distributed power, and monitor the temperature and flow rate in real time during the heating process to achieve the required heating effect.

[0053] In this embodiment, it is specifically necessary to explain the waste heat recovery module. The waste heat recovery module uses a heat pipe heat exchanger to recover the waste heat of the furnace body exhaust gas, and uses medium and low temperature waste heat to drive the ORC power generation unit to convert waste heat into electric energy. The ORC power generation unit is a key component for converting waste heat of exhaust gas into electric energy. By improving energy efficiency and utilizing waste heat, it reduces energy consumption and exhaust gas emissions, and helps to reduce the environmental impact during the production process. The specific steps are as follows:

[0054] Step D1: Use a heat pipe heat exchanger to transfer the waste heat in the exhaust gas to the working medium through heat pipes for heat recovery. Set the effective heat transfer efficiency of the heat exchanger as η heat , and calculate the recovered thermal power from the exhaust gas. The specific formula is where P recover (n) is the recovered thermal power from the exhaust gas, is the exhaust gas flow rate, C p is the specific heat capacity of the exhaust gas, T flue,in , T flue,out are respectively the temperature of the exhaust gas when entering the heat exchanger and the temperature of the exhaust gas when leaving the heat exchanger;

[0055] Step D2: Input the recovered thermal power into the ORC power generation unit and calculate the electric energy output from the ORC power generation unit. The specific formula is: E output (n) = [η ORC ·P recover (n)]·t, where E output (n) is the conversion efficiency of the ORC power generation unit, t is the time duration, and η ORC is the conversion efficiency of the ORC power generation unit.

[0056] In this embodiment, it is specifically necessary to explain the safety monitoring and protection module. The safety monitoring and protection module configures a multi-sensor network at the key links of the melting furnace, including thermocouples, pressure sensors, and highly sensitive gas leak detectors, monitors the safe operation status, and sets safety thresholds. When the monitored data exceeds the safety limit, an alarm is immediately issued and the alarm event is recorded. The specific steps are as follows:

[0057] Step E1: Sensor data monitoring: Set the safety thresholds of the temperature sensor as T max and T min , the safety thresholds of the pressure sensor as P max and P min , and the safety threshold of the gas concentration as C max . Compare the real-time data obtained by each sensor with its preset safety threshold;

[0058] Step E2, Alarm Triggering and Response: When the monitoring data of any sensor exceeds the set safety threshold, an alarm is immediately triggered, and safety protection measures are initiated, including shutting down the casting furnace, starting the cooling system, and automatically recording the alarm event when the alarm is triggered, including the alarm type, occurrence time, and specific threshold that triggered it. Different response measures are taken according to the level and type of the alarm event. When the temperature and pressure exceed the limit, the casting furnace is automatically shut down, and alarm information is sent to the operator via text message and email.

[0059] It should be noted that in the above embodiments, each embodiment is described with its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0060] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0061] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0062] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0063] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 or steps for implementing the functions specified in one block or a plurality of blocks.

[0064] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0065] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent control system for an energy-saving continuous up-drawing melting furnace based on machine learning, characterized in that: It includes a data acquisition and processing module, an intelligent control module, a composite heating module, a waste heat recovery module, and a safety monitoring and protection module; The data acquisition and processing module arranges multi-parameter sensors at key positions of the melting furnace to collect temperature, flow rate, pressure, and gas composition data, processes the collected data, and stores the processed data in a database; The intelligent control module uses a deep learning model to analyze the collected data, constructs a relationship model between temperature, flow rate, and pressure, dynamically adjusts the heating power and traction speed according to the input of temperature, flow rate, and pressure data, introduces a feedback adjustment mechanism, generates control instructions, and sends them to the composite heating module; The composite heating module distributes the proportion of resistance heating and gas heating according to the instructions of the intelligent control module to achieve the goals of rapid temperature rise and temperature uniformity; The waste heat recovery module uses a heat pipe heat exchanger to recover the waste heat of the furnace body exhaust gas and uses medium and low temperature waste heat to drive an ORC power generation unit to convert the waste heat into electric energy; The safety monitoring and protection module configures a multi-sensor network at key links of the melting furnace, including thermocouples, pressure sensors, and high-sensitivity gas leak detectors, monitors the safe operating status, sets safety thresholds, and immediately issues an alarm and records the alarm event when the monitored data exceeds the safety limit; 2. The intelligent control system for an energy-saving continuous upward drawing melting and casting furnace based on machine learning according to claim 1, characterized in that: The data acquisition and processing module arranges multi-parameter sensors at key positions of the melting furnace to collect temperature, flow rate, pressure, and gas composition data, processes the collected data, and stores the processed data in a database. The specific steps are as follows: Step A1, data acquisition: Arrange various types of sensors in the melting area, heat preservation area, and flow area of the melting furnace, including high-precision thermocouples, flow rate sensors, pressure sensors, and gas composition analyzers, to collect the molten metal temperature T, the flow rate V of the molten metal, the furnace internal pressure P, and the gas composition data in the waste gas, including oxygen C1, nitrogen C2, and carbon monoxide C3. For each time point n, the collected data is expressed as D(n) = {T(n), V(n), P(n), C1(n), C2(n), C3(n)}, where D(n) represents the multi-dimensional data obtained at the nth sampling moment, T(n) represents the molten metal temperature collected at time point n, V(n) represents the molten metal flow rate collected at time point n, P(n) represents the furnace internal pressure collected at time point n, and C1(n), C2(n), C3(n) respectively represent the gas composition concentrations collected at time point n; Step A2, data processing: Based on signal processing and data preprocessing algorithms, remove noise, perform real-time filtering and outlier detection on the collected multi-dimensional data, integrate it through multi-channel data fusion technology, and store it in the database according to the time series; 3. The intelligent control system for an energy-saving continuous up-drawing melting furnace based on machine learning according to claim 1, wherein: The intelligent control module uses a deep learning model to analyze the collected data, constructs a relationship model between temperature, flow rate, and pressure, dynamically adjusts the heating power and traction speed according to the input of temperature, flow rate, and pressure data, introduces a feedback adjustment mechanism, generates control instructions, and sends them to the composite heating module. The specific steps are as follows: Step B1. Deep learning model training: Extract the historical data set from the database as the input of the deep learning model, and use a multi-layer perceptron to predict the control outputs, including the heating power and the traction speed. Step B2, Feedback Regulation: Generate a preliminary control command based on the predicted output of the deep learning model and Adjust the heating power according to the difference between the real-time temperature and the target temperature, and the adjusted heating power is Adjust the traction speed according to the difference between the real-time flow rate and the target flow rate, and the adjusted traction speed is where K heat is the feedback gain of the heating power, K traction is the feedback gain of the traction speed, T target and V target are the target temperature and the target flow rate respectively, T actual (n) and V actual (n) are the actual temperature and flow rate respectively. Send the heating power and traction speed after feedback regulation to the composite heating module as the final control command. The final control command is: u(n) = {P’ heat (n), V’ traction (n)}.

4. The intelligent control system for an energy-saving continuous up-drawing melting furnace based on machine learning according to claim 3, characterized in that: In the deep learning model training of step B1, when extracting the historical data set from the database as the input of the deep learning model and using a multi-layer perceptron to predict the control outputs, including the heating power and the traction speed, it further includes the following steps: Step B101. Extract the historical data set from the database and segment it according to the time series. Each time point n corresponds to a data vector {T(n), V(n), P(n), C1(n), C2(n), C3(n)}, which is used as the input of the deep learning model. Step B102: Construct a multi-layer perceptron model, including an input layer, a hidden layer, and an output layer. The number of nodes in the input layer is equal to the number of input features. The output layer has two neurons, which are used to predict the heating power and the traction speed respectively. For an L-layer perceptron network, the input layer is represented as the 0th layer, and the output layer is represented as the Lth layer. The input and output of the lth layer are represented as h l and w l , and the input from the input layer to the 1st layer is h1 = f(W1x(n) + b1), and the input from the lth layer to the (l + 1)th layer is h l+1 = f(W l+1 h l + b l+1 ). The output of the output layer is y(n) = W L h L + b L = [P heat (n), V traction (n)], where W L is the weight matrix of the last layer, b L is the bias term of the last layer, h L is the output of the last layer, W l+1 is the weight matrix of the (l + 1)th layer, h l+1 is the output of the (l + 1)th layer, b l+1 is the bias term of the (l + 1)th layer, W1 is the weight matrix of the 1st layer, b1 is the bias term of the 1st layer, f is the activation function, x(n) is the input vector, are the predicted heating power and traction speed; Step B103: Use the mean squared error loss function to calculate the difference between the predicted value and the true value. The formula is: where, [P heat (n), V traction (n)] is the actual control value, is the heating power and traction speed predicted by the model, and N is the total number of training samples.

5. The intelligent control system for the energy-saving continuous up-drawing melting furnace based on machine learning according to claim 1, characterized in that: The composite heating module distributes the proportion of resistance heating and gas heating according to the instructions of the intelligent control module to achieve the goals of rapid temperature rise and temperature uniformity. The specific steps are as follows: Step C1. Set the scaling factor: Receive the output control instruction u(n) = {P’ heat (n), V’ traction (n)} from the intelligent control module. Set the target heating power according to the required heating power P’ heat (n). Set a scaling factor α for controlling the distribution ratio between resistance heating and gas heating. The scaling factor α ranges from [0, 1], representing the power ratio of resistance heating to gas heating. When α = 1, it means that all the heating power is provided by resistance heating; when α = 0, it means that all the heating power is provided by gas heating; Step C2, Allocate Heating Power: According to the target heating power P’ heat (n) and the scaling factor α, calculate the power allocation between resistance heating and gas heating. The specific calculation formula is: P res (n) = α · P’ heat (n), P gas (n) = (1 - α) · P’ heat (n), where P res (n) is the power of resistance heating, and P gas (n) is the power of gas heating. According to the allocated power, adjust the powers of the resistance heater and the gas heater respectively, and during the heating process, monitor the temperature and flow rate in real time to achieve the required heating effect.

6. The intelligent control system for an energy-saving continuous up-drawing melting furnace based on machine learning according to claim 1, characterized in that: The waste heat recovery module uses a heat pipe heat exchanger to recover the waste heat of the furnace body exhaust gas and uses the medium and low temperature waste heat to drive the ORC power generation unit to convert the waste heat into electric energy. The specific steps are as follows: Step D1: Use a heat pipe heat exchanger to transfer the waste heat in the waste gas to the working medium through the heat pipe for heat recovery, and set the effective heat transfer efficiency of the heat exchanger as η heat , calculate the recovered heat power from the waste gas. The specific formula is where P recover (n) is the recovered heat power from the waste gas, is the waste gas flow rate, C p is the specific heat capacity of the waste gas, T flue,in , T flue,out are the temperature of the waste gas when it enters the heat exchanger and the temperature of the waste gas when it leaves the heat exchanger, respectively; Step D2: Input the recovered thermal power into the ORC power generation unit and calculate the electrical energy output from the ORC power generation unit. The specific formula is: E output (n) = [η ORC ·P recover (n)]·t, where E output (n) is the conversion efficiency of the ORC power generation unit, t is the time duration, and η ORC is the conversion efficiency of the ORC power generation unit.

7. The intelligent control system for an energy-saving continuous up-drawing melting and casting furnace based on machine learning according to claim 1, characterized in that: The safety monitoring and protection module configures a multi-sensor network at the key links of the melting and casting furnace, including thermocouples, pressure sensors and highly sensitive gas leak detectors, monitors the safe operating status, and sets safety thresholds. When the monitored data exceeds the safety limit, an alarm is immediately issued and the alarm event is recorded. The specific steps are as follows: Step E1, Sensor Data Monitoring: Set the safety threshold of the temperature sensor as T max and T min , the safety threshold of the pressure sensor as P max and P min , the safety threshold of the gas concentration as C max , compare the real-time data obtained by each sensor with its preset safety threshold; Step E2. Alarm triggering and response: When the monitoring data of any sensor exceeds the set safety threshold, the alarm is immediately triggered, and the safety protection measures are started. When the alarm is triggered, the alarm event is automatically recorded, including the alarm type, occurrence time, and specific threshold triggered. Different response measures are taken according to the level and type of the alarm event. When the temperature and pressure exceed the limit, the melting and casting furnace is automatically shut down, and the alarm information is sent to the operator by text message and email.

Images