A vacuum melting furnace based on machine learning algorithm and its use method

Abstract

The present invention discloses a vacuum melting furnace based on a machine learning algorithm and a method for using the same, which relates to the field of vacuum melting technology, including a microprocessor, a PLC controller, an actuator, a power supply module, a water cooling module, a wireless communication module, a data acquisition module, a data analysis module, and a fault diagnosis and maintenance module, wherein the power supply module is responsible for providing power to the entire vacuum melting equipment, and the data acquisition module is responsible for collecting the real-time operating parameters of the actuator during the melting process. In the present invention, during the melting process, the automatic temperature measuring device adopts a non-contact infrared ratio temperature measurement method with high anti-interference performance to achieve online measurement of the temperature of the melt in the crucible, and uses a laser rangefinder to measure the melt level height through a laser beam, thereby obtaining the volume change of the melt, and through real-time measurement of the melting process parameters, provides dynamic data support for the machine learning algorithm to accurately control the heating temperature and the insulation time.

Description

Technical Field

[0001] The present invention relates to the technical field of vacuum melting, and in particular to a vacuum melting furnace based on a machine learning algorithm and a method for using the same. Background Art

[0002] A vacuum melting furnace is a special metallurgical equipment for melting metals and alloys in a vacuum environment. It is mainly used to reduce external contamination of materials while improving the purity and performance of metals. Existing vacuum melting furnaces lack online detection equipment, making it difficult to dynamically adjust the operating parameters of the melting process.

[0003] The defects of existing vacuum melting furnaces are:

[0004] 1. Patent document KR102129871B1 discloses a continuous vacuum melting furnace, which mainly considers how to achieve continuous melting and greatly reduce the use of protective gas used to prevent contact with oxygen during the melting process, but does not consider dynamic detection of operating parameters of the melting process to provide data support for precise control of the melting process;

[0005] 2. Patent document CN111504065B discloses a silver-based alloy bonding wire alloy material melting equipment, which mainly considers how to solve the problem that it is difficult to achieve real-time solid-liquid separation when using a general melting furnace, but does not consider how to accurately control the melting process to improve production efficiency and product quality;

[0006] 3. Patent document CN111570779B discloses a vacuum melting furnace for NdFeB production. The main consideration is how to make the crucible flip on both sides to extend the cooling time of the alloy, but it does not consider how to treat the waste gas generated during the melting process in an environmentally friendly way to reduce environmental pollution;

[0007] 4. Patent document CN109855427B discloses a vacuum melting furnace with a stirring function. The main consideration is how to stir the melt during the melting process to improve the fusion of particle reinforcements and metal and reduce the precipitation of particle reinforcements, but it does not consider ensuring that the crucible is clean and free of impurities to reduce the generation of impurities during the melting process. Summary of the invention

[0008] The object of the present invention is to provide a vacuum melting furnace based on a machine learning algorithm and a method of using the same to solve the problems raised in the above background technology.

[0009] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a vacuum melting furnace based on a machine learning algorithm, comprising a microprocessor, a PLC controller, an actuator, a power supply module, a water cooling module, a wireless communication module, a data acquisition module, a data analysis module and a fault diagnosis and maintenance module, wherein the power supply module is responsible for providing power for the entire vacuum melting equipment, the data acquisition module is responsible for collecting the real-time operating parameters of the actuator during the melting process, the data analysis module uses a learning machine algorithm to process and analyze the collected real-time operating parameters, the microprocessor predicts the melting trend according to the analyzed real-time operating parameters, sends an adjustment control instruction to the PLC controller, and controls the actuator to adjust the melting parameters, the water cooling module protects the actuator by circulating cooling water, and the fault diagnosis and maintenance module discovers and solves potential equipment problems according to the data processed by the data analysis module and adopts fault diagnosis technology;

[0010] The actuator comprises a vacuum melting furnace, a connecting shaft is installed on one inner wall of the vacuum melting furnace, an inductor is installed at one end of the connecting shaft, and a crucible is installed inside the inductor;

[0011] The data acquisition module includes a visual detection sensor, a laser rangefinder and an automatic temperature measuring device installed on the top wall of the vacuum melting furnace. The automatic temperature measuring device is located on one side of the visual detection sensor, and the laser rangefinder is located behind the visual detection sensor. A vacuum detection sensor is installed on the inner wall of one side of the vacuum melting furnace.

[0012] Preferably, a bottom box is installed at the bottom of the vacuum melting furnace, an argon gas cylinder and a waste gas treatment box are installed inside the bottom box, and the waste gas treatment box is located on one side of the argon gas cylinder, a small shell and tube condenser is installed on the inner wall of one side of the waste gas treatment box, a small buffer tank is installed at the output end of the small shell and tube condenser, a liquid outlet pipe is installed at the output end of the small buffer tank, and one end of the liquid outlet pipe is located at the rear of the bottom box.

[0013] Preferably, a furnace tilting mechanism is installed on the top of the bottom box, and the furnace tilting mechanism is located on one side of the vacuum melting furnace, the output end of the furnace tilting mechanism is connected to the sensor, and a cleaning component is installed on the back wall of the vacuum melting furnace, and the cleaning component is located between the crucible and the laser rangefinder.

[0014] Preferably, the cleaning component includes a small telescopic rod 1 installed on the back wall of the vacuum melting furnace, a small telescopic rod 2 is installed at the output end of the small telescopic rod 1, a small electric adsorption device is installed at the output end of the small telescopic rod 2, and the small electric adsorption device is electrically connected to the power supply module.

[0015] Preferably, a stirring device and a secondary feeder are installed on the top of the vacuum melting furnace, and the secondary feeder is located on one side of the stirring device, the stirring device is located between the automatic temperature measuring device and the visual detection sensor, the stirring device is located directly above the crucible, and an ingot mold mechanism is installed on the bottom wall of the vacuum melting furnace, and the ingot mold mechanism is located on the side and below the sensor.

[0016] Preferably, a molecular pump is installed on one outer wall of the vacuum melting furnace, a direct-connected mechanical pump is installed at the output end of the molecular pump, and the direct-connected mechanical pump is located on one side of the bottom box, a furnace door is movably installed on the front of the vacuum melting furnace through hinges, and an observation window is installed on the front of the furnace door.

[0017] Preferably, an electrical cabinet is installed on one side outer wall of the bottom box, and a control simulation screen and an emergency stop button are installed on the front of the electrical cabinet, and the emergency stop button is located on one side of the control simulation screen.

[0018] Preferably, a high-frequency power supply is installed on the bottom wall of the electrical cabinet, a cooling fan is embedded in the back wall of the electrical cabinet, a vent is opened on the back of the electrical cabinet, and the vent corresponds to the position of the cooling fan, and a power socket is installed on the back of the electrical cabinet, and the power socket is located on the lower side of the vent.

[0019] Preferably, the method for using the vacuum melting furnace based on the machine learning algorithm comprises the following steps:

[0020] Step S1, before use, using a visual detection sensor to detect whether there is dust and impurities remaining in the crucible, if dust and impurities are detected in the crucible, the cleaning component absorbs the dust in the crucible;

[0021] Step S2: After cleaning, place the metal to be smelted in a crucible, first use a molecular pump and a direct-connected mechanical pump to pump out the air in the vacuum melting furnace, and then start heating for smelting;

[0022] Step S3, after smelting is completed, controlling the furnace tilting mechanism to pour the molten metal in the crucible into the ingot mold mechanism for cooling and forming;

[0023] Step S4: Finally, the gas in the vacuum melting furnace is pumped into a waste gas treatment box for condensation and recovery treatment.

[0024] Preferably, in step S2, the following steps are also included:

[0025] Step S21: During the smelting process, an automatic temperature measuring device, a visual detection sensor and a laser rangefinder are used to detect the smelting operating parameters in real time. At the same time, a machine learning algorithm controls the heating temperature and holding time of the sensor according to the smelting data and requirements;

[0026] In step S4, the following steps are also included:

[0027] Step S41, the gas is pumped into a small shell and tube condenser for preliminary condensation, and then the argon cylinder further cools the vaporized metal into liquid and discharges it through the liquid outlet pipe.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. During the smelting process of the present invention, the automatic temperature measuring device adopts a non-contact infrared ratio temperature measurement method with high anti-interference performance to realize online measurement of the temperature of the melt in the crucible, and uses a laser rangefinder to measure the liquid level of the melt through a laser beam, thereby obtaining the volume change of the melt. Through real-time measurement of melting process parameters, dynamic data support is provided for the machine learning algorithm to accurately control the heating temperature and holding time.

[0030] 2. The data analysis module of the present invention uses a machine learning algorithm to process and analyze the collected real-time data. Then the microprocessor combines the characteristics of the metal material and the smelting requirements to predict the smelting trend and generate adjustment control instructions, which are sent to the PLC controller through the wireless communication module. After receiving the instructions, the PLC controller accurately controls key process parameters such as heating temperature, holding time and cooling speed according to the content of the instructions, and dynamically adjusts the smelting parameters through real-time monitored data to ensure the stability and repeatability of the smelting process, ensure that the quality and performance of the final product meet specific requirements, and greatly improve production efficiency and product quality.

[0031] 3. The waste gas treatment box of the present invention is used to effectively recover and treat volatile metal impurities and other waste gases generated during the vacuum smelting process to reduce the impact on the environment. During the vacuum smelting process, volatile metal impurities such as copper and zinc will be vaporized at high temperatures. The waste gas is pumped into a small shell and tube condenser through a pipeline for preliminary condensation, and then the vaporized metal is further cooled into a liquid state through a small buffer tank and discharged through a liquid outlet pipe, thereby reducing the pollution and harm of heavy metal gases to the environment and human body.

[0032] 4. The present invention captures the image of the crucible through a visual detection sensor, and uses image recognition technology to detect whether there are impurities and dust remaining inside the crucible. When impurities or dust are detected attached to the inner surface of the crucible, the microprocessor controls the cleaning component to start adsorbing the impurities or dust, ensuring that the crucible is clean and free of impurities, so as to reduce the generation of impurities and energy consumption during the smelting process.

[0033] Figure 1 It is a schematic diagram of the three-dimensional structure of the present invention;

[0034] Figure 2 It is a schematic diagram of the front cross-sectional structure of the vacuum melting furnace of the present invention;

[0035] Figure 3 For the present invention Figure 2 Schematic diagram of the plane structure of A;

[0036] Figure 4 It is a schematic cross-sectional structure diagram of the exhaust gas treatment box of the present invention;

[0037] Figure 5 It is a schematic diagram of a top cross-sectional structure of a vacuum melting furnace of the present invention;

[0038] Figure 6 It is a schematic diagram of the side cross-sectional structure of the electrical cabinet of the present invention;

[0039] Figure 7 It is a rear perspective structural schematic diagram of the vacuum melting furnace of the present invention;

[0040] Figure 8 It is a system diagram of the present invention.

[0041] In the figure: 1. electrical cabinet; 2. bottom box; 3. vacuum melting furnace; 4. furnace door; 5. observation window; 6. control simulation screen; 7. emergency stop button; 8. stirring device; 9. tilting furnace mechanism; 10. sensor; 11. crucible; 12. vacuum detection sensor; 13. ingot mold mechanism; 14. waste gas treatment box; 15. argon bottle; 16. small shell and tube condenser; 17. small buffer tank; 18. liquid outlet pipe; 19. visual detection sensor; 20. laser rangefinder; 21. small telescopic rod one; 22. small telescopic rod two; 23. small electric adsorption device; 24. high frequency power supply; 25. power socket; 26. cooling fan; 27. vent; 28. molecular pump; 29. ​​direct mechanical pump; 30. automatic temperature measuring device. DETAILED DESCRIPTION

[0042] 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.

[0043] In the description of the present invention, it should be noted that the terms "upper", "lower", "inner", "outer", "front end", "rear end", "two ends", "one end", "the other end" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0044] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "provided with", "connected", etc. should be understood in a broad sense. For example, "connected" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0045] See also Figure 1 , Figure 2 , Figure 3 , Figure 5 and Figure 8 , an embodiment provided by the present invention: a vacuum melting furnace based on machine learning algorithm, comprising a microprocessor, a PLC controller, an actuator, a power supply module, a water cooling module, a wireless communication module, a data acquisition module, a data analysis module and a fault diagnosis and maintenance module, wherein the power supply module is responsible for providing power to the entire vacuum melting equipment, the data acquisition module is responsible for collecting the real-time operating parameters of the actuator during the melting process, the data analysis module uses the learning machine algorithm to process and analyze the collected real-time operating parameters, the microprocessor predicts the melting trend according to the analyzed real-time operating parameters, sends an adjustment control instruction to the PLC controller, and controls the actuator to adjust the melting parameters, the water cooling module protects the actuator by circulating cooling water, and the fault diagnosis and maintenance module discovers and solves potential equipment problems according to the data processed by the data analysis module and adopts fault diagnosis technology;

[0046] The actuator includes a vacuum melting furnace 3, a connecting shaft is installed on one inner wall of the vacuum melting furnace 3, an inductor 10 is installed at one end of the connecting shaft, and a crucible 11 is installed inside the inductor 10;

[0047] The data acquisition module includes a visual detection sensor 19, a laser rangefinder 20 and an automatic temperature measuring device 30 installed on the top wall of the vacuum melting furnace 3. The automatic temperature measuring device 30 is located on one side of the visual detection sensor 19, and the laser rangefinder 20 is located behind the visual detection sensor 19. A vacuum detection sensor 12 is installed on the inner wall of one side of the vacuum melting furnace 3.

[0048] Furthermore, before smelting, the crucible 11 is first image-captured using a visual detection sensor 19, and image recognition technology is used to detect whether there are any impurities and dust remaining inside the crucible 11. After ensuring that the crucible 11 is clean and free of impurities, the metal to be melted can be placed in the crucible 11. During the smelting process, the automatic temperature measuring device 30 adopts a non-contact infrared ratio temperature measurement method with high anti-interference performance to achieve online measurement of the temperature of the melt in the crucible 11, and uses a laser rangefinder 20 to measure the melt level height through a laser beam, thereby obtaining the volume change of the melt. Through real-time measurement of the melting process parameters, dynamic data support is provided for the machine learning algorithm to accurately control the heating temperature and the insulation time. In actual use, a radiation thermometer and an infrared thermal imager can also be used for temperature monitoring. The infrared thermal imager measures the temperature of the object by receiving the infrared radiation emitted by the object being measured, while the radiation thermometer measures the temperature value based on the function of the radiation degree and the temperature of the object.

[0049] First, the cleanliness level before the melting process is measured through the data acquisition module, and then the vacuum level when the air is extracted is measured, and then the temperature and melt volume during the melting process are monitored. Then the data analysis module uses machine learning algorithms to process and analyze the collected real-time data. Then the microprocessor combines the characteristics of the metal material and the smelting requirements to predict the smelting trend and generate adjustment control instructions, which are sent to the PLC controller through the wireless communication module. After receiving the instructions, the PLC controller accurately controls key process parameters such as heating temperature, holding time and cooling speed according to the content of the instructions, and dynamically adjusts the smelting parameters through real-time monitored data to ensure the stability and repeatability of the smelting process, and ensure that the quality and performance of the final product meet specific requirements, greatly improving production efficiency and product quality. At the same time, the fault diagnosis and maintenance module implements a preventive maintenance plan to prevent the occurrence of unexpected failures, and uses vibration analysis and thermal imaging technology to diagnose the entire smelting equipment and smelting process, so as to promptly discover and solve potential problems and improve the production stability of the equipment.

[0050] See also Figure 2 , Figure 4 and Figure 7 An embodiment of the present invention is: a vacuum melting furnace based on a machine learning algorithm and a method for using the same, wherein a bottom box 2 is installed at the bottom of the vacuum melting furnace 3, an argon gas bottle 15 and a waste gas treatment box 14 are installed inside the bottom box 2, and the waste gas treatment box 14 is located on one side of the argon gas bottle 15, a small shell and tube condenser 16 is installed on the inner wall of one side of the waste gas treatment box 14, a small buffer tank 17 is installed at the output end of the small shell and tube condenser 16, a liquid outlet pipe 18 is installed at the output end of the small buffer tank 17, and one end of the liquid outlet pipe 18 is located at the rear of the bottom box 2.

[0051] Furthermore, a waste gas treatment box 14 and an argon gas bottle 15 are placed in the bottom box 2. The waste gas treatment box 14 is used to effectively recover and treat volatile metal impurities and other waste gases generated during the vacuum smelting process to reduce the impact on the environment. During the vacuum smelting process, volatile metal impurities such as copper, zinc, etc. will be vaporized at high temperatures. The waste gas is pumped into a small shell and tube condenser 16 through a pipeline for preliminary condensation, and then the vaporized metal is further cooled into a liquid state through a small buffer tank 17 and discharged through a liquid outlet pipe 18 to reduce the pollution and harm of heavy metal gases to the environment and human body. In addition, during vacuum operation, argon gas can be introduced in time to prevent loss.

[0052] See also Figure 2 and Figure 5 , an embodiment of the present invention: a vacuum melting furnace based on a machine learning algorithm, wherein a furnace tilting mechanism 9 is installed on the top of the bottom box 2, and the furnace tilting mechanism 9 is located on one side of the vacuum melting furnace 3, and the output end of the furnace tilting mechanism 9 is connected to the sensor 10, and a cleaning component is installed on the back wall of the vacuum melting furnace 3, and the cleaning component is located between the crucible 11 and the laser rangefinder 20;

[0053] The cleaning assembly includes a small telescopic rod 21 installed on the back wall of the vacuum melting furnace 3, a small telescopic rod 22 is installed at the output end of the small telescopic rod 21, a small electric adsorption device 23 is installed at the output end of the small telescopic rod 22, and the small electric adsorption device 23 is electrically connected to the power supply module.

[0054] Furthermore, the furnace tilting mechanism 9 is used to control the sensor 10 and the crucible 11 to tilt steadily to an appropriate angle after the vacuum smelting is completed, so as to pour the molten metal from the crucible 11 into the ingot mold mechanism 13 for cooling and molding. Before the vacuum smelting, the crucible 11 is first imaged by the visual detection sensor 19, and the image recognition technology is used to detect whether there are impurities and dust remaining inside the crucible 11. When it is detected that impurities or dust are attached to the inner surface of the crucible 11, the microprocessor controls the cleaning component to start adsorbing the impurities or dust, controls the small telescopic rod 1 21 to extend, drives the small telescopic rod 22 and the small electric adsorption device 23 to approach the crucible 11, and when the small telescopic rod 22 and the small electric adsorption device 23 are moved closer to the crucible 11, the small telescopic rod 22 and the small electric adsorption device 23 are moved closer to the crucible 11. When the device 23 reaches the top of the crucible 11, the small telescopic rod 22 is controlled to extend, driving the small electric adsorption device 23 to move into the crucible 11. Since the small electric adsorption device 23 is electrically connected to the power supply module, the surface of the small electric adsorption device 23 is charged, and the charge is transported to the surface of the small electric adsorption device 23 through the action of the electric field, so that it has electrostatic properties. When the small electric adsorption device 23 is close to the dust, it is affected by the electrostatic force and approaches the surface of the small electric adsorption device 23 with the opposite charge, so that the dust particles are adsorbed on the surface of the small electric adsorption device 23, thereby cleaning the dust in the crucible 11, ensuring that the crucible 11 is clean and free of impurities, so as to reduce the generation of impurities and energy consumption during the smelting process.

[0055] See also Figure 1 , Figure 2 , Figure 5 and Figure 7 , an embodiment of the present invention: a vacuum melting furnace based on a machine learning algorithm, wherein a stirring device 8 and a secondary feeder are installed on the top of the vacuum melting furnace 3, and the secondary feeder is located on one side of the stirring device 8, the stirring device 8 is located between the automatic temperature measuring device 30 and the visual detection sensor 19, the stirring device 8 is located directly above the crucible 11, and an ingot mold mechanism 13 is installed on the bottom wall of the vacuum melting furnace 3, and the ingot mold mechanism 13 is located on the side and below the sensor 10;

[0056] A molecular pump 28 is installed on one outer wall of the vacuum melting furnace 3, and a direct-connected mechanical pump 29 is installed on the output end of the molecular pump 28, and the direct-connected mechanical pump 29 is located on one side of the bottom box 2. A furnace door 4 is movably installed on the front of the vacuum melting furnace 3 through hinges, and an observation window 5 is installed on the front of the furnace door 4.

[0057] Furthermore, the stirring device 8 promotes the homogenization of high and low melting point elements through strong electromagnetic stirring, reduces the volatilization loss caused by local overheating, and the stirring device 8 can be retracted for easy use and storage. The secondary feeder can more efficiently put the material into the crucible 11, while reducing the operation steps of vacuum extraction. The ingot mold mechanism 13 is used to receive and shape the molten metal after the vacuum melting is completed;

[0058] The molecular pump 28 and the direct-connected mechanical pump 29 are used together to form a bipolar pump exhaust system, which can not only provide higher exhaust efficiency, but also effectively prevent oil vapor reflux, thereby maintaining a clean and vacuum environment to achieve a higher vacuum degree and a cleaner vacuum environment. When extracting air from the vacuum melting furnace 3, the vacuum detection sensor 12 is used to monitor the vacuum level in the vacuum melting furnace 3 in real time, so that the microprocessor can control the operating status of the molecular pump 28 and the direct-connected mechanical pump 29. The observation window 5 is convenient for processing personnel to observe the vacuum melting conditions in the vacuum melting furnace 3 from the outside.

[0059] See also Figure 1 , Figure 6 and Figure 7 , an embodiment provided by the present invention: a vacuum melting furnace based on a machine learning algorithm, an electrical cabinet 1 is installed on one side outer wall of the bottom box 2, a control simulation screen 6 and an emergency stop button 7 are installed on the front of the electrical cabinet 1, and the emergency stop button 7 is located on one side of the control simulation screen 6;

[0060] A high-frequency power supply 24 is installed on the bottom wall of the electrical cabinet 1, a cooling fan 26 is embedded in the back wall of the electrical cabinet 1, a vent 27 is opened on the back of the electrical cabinet 1, and the vent 27 corresponds to the position of the cooling fan 26, and a power socket 25 is installed on the back of the electrical cabinet 1, and the power socket 25 is located below the side of the vent 27.

[0061] Furthermore, the operator can manually input the melting parameters through the control simulation screen 6, and can also view the operating parameters of the vacuum melting in real time through the control simulation screen 6. When encountering an emergency, the operator presses the emergency stop button 7 to stop the operation of the entire vacuum melting equipment and cut off the power supply to prevent serious safety accidents. The high-frequency power supply 24 is mainly used to provide alternating current and generate an alternating magnetic field through the inductor 10, thereby generating eddy currents in the metal material in the crucible 11 for heating and melting. During long-term operation, the high-frequency power supply 24 will generate heat, and the heat dissipation fan 26 is used to dissipate the heat of the high-frequency power supply 24 to extend the service life of the high-frequency power supply 24.

[0062] Furthermore, the method for using the vacuum melting furnace based on the machine learning algorithm comprises the following steps:

[0063] Step S1: Before use, use the visual detection sensor 19 to detect whether there is any dust or impurities remaining in the crucible 11. If dust or impurities are detected in the crucible 11, the cleaning component absorbs the dust in the crucible 11;

[0064] Step S2: After cleaning, the metal to be smelted is placed in the crucible 11, and the air in the vacuum melting furnace 3 is first pumped out using the molecular pump 28 and the direct-connected mechanical pump 29, and then heating is started for smelting;

[0065] Step S3, after smelting is completed, the furnace tilting mechanism 9 is controlled to pour the molten metal in the crucible 11 into the ingot mold mechanism 13 for cooling and forming;

[0066] Step S4, finally, the gas in the vacuum melting furnace 3 is pumped into the waste gas treatment box 14 for condensation and recovery treatment;

[0067] In step S2, the following steps are also included:

[0068] Step S21: During the smelting process, the automatic temperature measuring device 30, the visual detection sensor 19 and the laser rangefinder 20 are used to detect the smelting operation parameters in real time, and the machine learning algorithm controls the heating temperature and the holding time of the sensor 10 according to the smelting data and requirements;

[0069] In step S4, the following steps are also included:

[0070] Step S41 , the gas is drawn into the small shell and tube condenser 16 for preliminary condensation, and then the argon cylinder 15 further cools the vaporized metal into liquid state and discharges it through the liquid outlet pipe 18 .

[0071] Working principle: before vacuum melting, the crucible 11 is first image-captured by the visual detection sensor 19, and the image recognition technology is used to detect whether there are impurities and dust remaining inside the crucible 11. When it is detected that impurities or dust are attached to the inner surface of the crucible 11, the microprocessor controls the cleaning component to start adsorbing the impurities or dust, so as to remove the dust in the crucible 11 and ensure that the crucible 11 is clean and free of impurities. After that, the metal to be melted can be placed in the crucible 11, and the furnace door 4 is closed. The air in the vacuum melting furnace 3 is quickly extracted by the joint use of the molecular pump 28 and the direct-connected mechanical pump 29, and the vacuum detection sensor 12 is used to monitor the vacuum level in the vacuum melting furnace 3 in real time, so that the microprocessor can control the operating status of the molecular pump 28 and the direct-connected mechanical pump 29;

[0072] During the smelting process, the temperature of the melt in the crucible 11 is measured online using an automatic temperature measuring device 30, and the height of the melt level is measured by a laser rangefinder 20 through a laser beam, thereby obtaining the volume change of the melt, so that the microprocessor can combine the characteristics of the metal material and the smelting requirements, predict the smelting trend, and accurately control key process parameters such as the heating temperature, the holding time, and the cooling rate. By dynamically adjusting the smelting parameters, the stability and repeatability of the smelting process are ensured.

[0073] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A vacuum melting furnace based on a machine learning algorithm, characterized in that: It includes a microprocessor, a PLC controller, an actuator, a power supply module, a water cooling module, a wireless communication module, a data acquisition module, a data analysis module and a fault diagnosis and maintenance module. The power supply module is responsible for providing power for the entire vacuum melting equipment. The data acquisition module is responsible for collecting the real-time operating parameters of the actuator during the melting process. The data analysis module uses a learning machine algorithm to process and analyze the collected real-time operating parameters. The microprocessor predicts the melting trend according to the analyzed real-time operating parameters, sends an adjustment control instruction to the PLC controller, and controls the actuator to adjust the melting parameters. The water cooling module protects the actuator by circulating cooling water. The fault diagnosis and maintenance module uses the data processed by the data analysis module and adopts fault diagnosis technology to discover and solve potential equipment problems; The actuator comprises a vacuum melting furnace (3), a connecting shaft being installed on an inner wall of one side of the vacuum melting furnace (3), an inductor (10) being installed at one end of the connecting shaft, and a crucible (11) being installed inside the inductor (10); The data acquisition module comprises a visual detection sensor (19), a laser rangefinder (20) and an automatic temperature measurement device (30) installed on the top wall of the vacuum melting furnace (3), the automatic temperature measurement device (30) being located on one side of the visual detection sensor (19), and the laser rangefinder (20) being located behind the visual detection sensor (19); a vacuum detection sensor (12) is installed on an inner wall of one side of the vacuum melting furnace (3); A cleaning assembly is installed on the back wall of the vacuum melting furnace (3), the cleaning assembly comprising a small telescopic rod 1 (21) installed on the back wall of the vacuum melting furnace (3), a small telescopic rod 2 (22) installed at the output end of the small telescopic rod 1 (21), a small electric adsorption device (23) installed at the output end of the small telescopic rod 2 (22), and the small electric adsorption device (23) is electrically connected to the power supply module; A stirring device (8) and a secondary feeder are installed on the top of the vacuum melting furnace (3), and the secondary feeder is located on one side of the stirring device (8). The stirring device (8) is located between the automatic temperature measuring device (30) and the visual detection sensor (19). The stirring device (8) is located directly above the crucible (11). An ingot mold mechanism (13) is installed on the bottom wall of the vacuum melting furnace (3), and the ingot mold mechanism (13) is located below the side of the sensor (10).

2. A vacuum melting furnace based on a machine learning algorithm according to claim 1, characterized in that: A bottom box (2) is installed at the bottom of the vacuum melting furnace (3), an argon gas bottle (15) and a waste gas treatment box (14) are installed inside the bottom box (2), and the waste gas treatment box (14) is located on one side of the argon gas bottle (15), a small shell and tube condenser (16) is installed on the inner wall of one side of the waste gas treatment box (14), a small buffer tank (17) is installed at the output end of the small shell and tube condenser (16), and a liquid outlet pipe (18) is installed at the output end of the small buffer tank (17), and one end of the liquid outlet pipe (18) is located at the rear of the bottom box (2).

3. A vacuum melting furnace based on a machine learning algorithm according to claim 2, characterized in that: A furnace tilting mechanism (9) is installed on the top of the bottom box (2), and the furnace tilting mechanism (9) is located on one side of the vacuum melting furnace (3). The output end of the furnace tilting mechanism (9) is connected to the sensor (10), and the cleaning component is located between the crucible (11) and the laser rangefinder (20).

4. A vacuum melting furnace based on a machine learning algorithm according to claim 2, characterized in that: A molecular pump (28) is installed on one side outer wall of the vacuum melting furnace (3), a direct-connected mechanical pump (29) is installed at the output end of the molecular pump (28), and the direct-connected mechanical pump (29) is located on one side of the bottom box (2), and a furnace door (4) is movably installed on the front side of the vacuum melting furnace (3) through hinges, and an observation window (5) is installed on the front side of the furnace door (4).

5. The vacuum melting furnace based on machine learning algorithm according to claim 2, characterized in that: An electrical cabinet (1) is installed on one side outer wall of the bottom box (2), and a control simulation screen (6) and an emergency stop button (7) are installed on the front of the electrical cabinet (1), wherein the emergency stop button (7) is located on one side of the control simulation screen (6).

6. A vacuum melting furnace based on a machine learning algorithm according to claim 5, characterized in that: A high-frequency power supply (24) is installed on the bottom wall of the electrical cabinet (1), a heat dissipation fan (26) is embedded in the back wall of the electrical cabinet (1), a vent (27) is provided on the back of the electrical cabinet (1), and the vent (27) corresponds to the position of the heat dissipation fan (26), and a power socket (25) is installed on the back of the electrical cabinet (1), and the power socket (25) is located below the side of the vent (27).

7. A method for using a vacuum melting furnace based on a machine learning algorithm, applicable to a vacuum melting furnace based on a machine learning algorithm according to any one of claims 1 to 5, characterized in that: The method for using the vacuum melting furnace based on the machine learning algorithm comprises the following steps: Step S1: Before use, a visual detection sensor (19) is used to detect whether dust and impurities remain in the crucible (11); if dust and impurities are detected in the crucible (11), a cleaning component absorbs the dust in the crucible (11); Step S2: After cleaning, the metal to be smelted is placed in the crucible (11), the air in the vacuum melting furnace (3) is first extracted using a molecular pump (28) and a direct-connected mechanical pump (29), and then heating is started for smelting; Step S3, after smelting is completed, controlling the furnace tilting mechanism (9) to pour the molten metal in the crucible (11) into the ingot mold mechanism (13) for cooling and forming; Step S4: Finally, the gas in the vacuum melting furnace (3) is pumped into the waste gas treatment box (14) for condensation and recovery treatment.

8. The method for using a vacuum melting furnace based on a machine learning algorithm according to claim 7, characterized in that: In step S2, the following steps are also included: Step S21, during the smelting process, an automatic temperature measuring device (30), a visual detection sensor (19) and a laser rangefinder (20) are used to detect the smelting operation parameters in real time, and a machine learning algorithm controls the heating temperature and holding time of the sensor (10) according to the smelting data and requirements; In step S4, the following steps are also included: Step S41: The gas is drawn into a small shell and tube condenser (16) for preliminary condensation, and then the argon cylinder (15) further cools the vaporized metal into a liquid state and discharges it through a liquid outlet pipe (18).

Images