A heat treatment equipment intelligent control method and system based on semantic parsing

Abstract

The application discloses a kind of based on semantic analysis's heat treatment equipment intelligent control method and system, method includes: S1, receiving the natural language information input by user, obtains user instruction by semantic recognition technology analysis;S2, based on user instruction, according to the corresponding structured process data of pre-set process logic constraint rule set generation;S3, to structured process data scanning, complete the parameter missing in structured process data, from the structured process data after completion filter out invalid process step belonging to non-controlled physical evolution process;S4, according to the configuration information of heat treatment equipment, from pre-set protocol driver library corresponding driver module is loaded, the structured process data after filtering is converted into the instruction sequence executable by heat treatment equipment, and end code is added, and the instruction sequence is sent to heat treatment equipment execution.The application reduces operation threshold and cognitive load, realizes the general adaptation to heterogeneous equipment.

Description

Technical Field

[0001] This invention relates to the fields of industrial automation control and artificial intelligence application technology, and in particular to an intelligent control method and system for heat treatment equipment based on semantic parsing. Background Technology

[0002] Heat treatment is a key process for improving the properties of metallic materials. Currently, the control of heat treatment equipment in industrial settings mainly faces the following challenges: 1. Existing equipment relies on complex code programming of PID temperature controllers or PLCs, which has a high programming threshold. It cannot understand natural language such as "heating at 10℃ / min", and requires manual calculation of parameters, making it inconvenient to use.

[0003] 2. The communication protocols and control logic (such as terminators) of temperature controllers from different manufacturers are completely different. The control system is usually deeply bound to a single piece of hardware and lacks universality. Summary of the Invention

[0004] To address the technical problems existing in the background art, this invention proposes an intelligent control method and system for heat treatment equipment based on semantic parsing.

[0005] In a first aspect, the present invention proposes an intelligent control method for heat treatment equipment based on semantic parsing, comprising: S1. Receive natural language information input by the user and parse it to obtain the user's instructions through semantic recognition technology; S2. Based on user instructions, generate corresponding structured process data according to a set of preset process logic constraint rules; S3. Scan the structured process data, complete the missing parameters in the structured process data, and filter out invalid steps belonging to uncontrolled physical evolution processes from the completed structured process data. S4. Based on the configuration information of the heat treatment equipment, load the corresponding driver module from the preset protocol driver library, convert the filtered structured process data into an instruction sequence that can be executed by the heat treatment equipment, add an end code, and send the instruction sequence to the heat treatment equipment for execution.

[0006] Furthermore, the semantic recognition technology is implemented through a large language model.

[0007] Furthermore, in S1, the process of parsing to obtain user instructions includes: determining whether the user instruction is a direct instruction mode or a process recommendation mode based on the parsing result.

[0008] Furthermore, in S2, if the direct instruction mode is determined, structured process data is generated according to the preset set of process logic constraint rules; if the process recommendation mode is determined, process information is obtained by querying the knowledge base, accessing the pre-trained data of the large language model, or performing online retrieval, and a process solution that meets the user's intention is matched and fed back to the user. Based on the confirmation signal from the user's feedback, the process solution is converted into structured process data.

[0009] Furthermore, in the process recommendation mode, the matching and generation of process solutions includes: extracting the material grade, workpiece size, and processing type from the user description as key entity features; retrieving a matching standard process template from the knowledge base; and using a large language model to fine-tune the process template parameters based on the key entity features to generate a process solution.

[0010] Furthermore, the set of process logic constraint rules includes at least: a first constraint rule that limits the data format, and a second constraint rule that sets the temperature continuity of adjacent steps in the heat preservation instruction.

[0011] Furthermore, in S3, the missing parameters in the structured process data are supplemented as follows: when a step has variable parameters but lacks process continuity parameters, the physical association model is called to calculate and supplement them; when a step lacks initial state parameters, they are supplemented based on default values.

[0012] Furthermore, the protocol driver library includes a register address mapping table and special function code definition for each type of heat treatment equipment. The filtered structured process data is filled into a specific register address and an end code is generated in the special function code.

[0013] Furthermore, the uncontrolled physical evolution process is either furnace cooling or air cooling.

[0014] In a second aspect, the present invention also provides an intelligent control system for heat treatment equipment based on semantic parsing, used to implement the method described in the first aspect, comprising: The interaction and feedback unit is used to receive natural language information input by the user and provide feedback on the process plan; The data parsing unit is used to parse user instructions using a large language model and generate structured process parameters; The data verification unit is used to perform parameter calculation and completion and invalid process step filtering on the structured process data; The protocol adaptation unit has a built-in extensible protocol driver library, which is used to convert the filtered structured process data into an executable instruction sequence and attach an end code according to the heat treatment equipment. A communication unit is used to send the device instruction sequence to the heat treatment device.

[0015] The technical effects of this invention are as follows: 1. Reduced operational threshold and cognitive load: By parsing natural language instructions, operators do not need to master complex programming code or manually convert parameters, making it easy to use and with a low barrier to entry.

[0016] 2. Improved determinism and safety of the control process: This invention can automatically calculate and complete parameters through physical correlation models, eliminating errors from manual calculations; and automatically eliminate invalid or uncontrolled steps through semantic filtering, avoiding the risk of misoperation.

[0017] 3. Achieved universal adaptation to heterogeneous devices: The structured data and protocol driver library work together to separate the upper-layer semantic control logic from the lower-layer hardware protocol. One system can be compatible with various brands and models of heat treatment equipment controllers, improving the system's versatility and scalability. Attached Figure Description

[0018] Figure 1 This is a flowchart of an intelligent control method for heat treatment equipment based on semantic parsing proposed in this invention.

[0019] Figure 2 This is a schematic diagram illustrating the process where the user instruction is in the process recommendation mode in a specific embodiment of the present invention.

[0020] Figure 3 This is a schematic diagram illustrating the result after user confirmation in the process recommendation mode, as shown in a specific embodiment of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings and embodiments. The embodiments are implemented based on the technical solution of the present invention, providing detailed implementation methods and specific operating procedures. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them, and the scope of protection of the present invention is not limited to the following embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] like Figure 1 As shown, the present invention proposes an intelligent control method for heat treatment equipment based on semantic parsing, comprising: S1. Receive natural language information input by the user and parse it using semantic recognition technology to obtain the user's instructions.

[0023] Specifically, semantic recognition technology is implemented through a large language model. Users input natural language through a text or voice interface, such as "heat to 300 degrees" or "how to anneal TC4 titanium alloy". After receiving the information through the human-computer interaction interface, the system uses the large language model to perform semantic recognition. Based on the keywords, sentence structure, or user intent classification in the semantic recognition results, the large language model obtains the user command, thereby determining whether the user command is a direct command mode that needs to be executed directly, or a process recommendation mode that requires process suggestions first.

[0024] S2. Based on user instructions, generate corresponding structured process data according to a preset set of process logic constraint rules, wherein the structured process data is a hardware-independent, general-purpose format. In a specific embodiment, based on the determination result of S1, the following two processes are executed: If the system is determined to be in direct instruction mode, it generates corresponding structured process data based on a pre-defined set of process logic constraints. For example, if a user requests to "heat from room temperature to 600 degrees Celsius and then hold for one hour," the system will directly generate structured process data according to the parsed user instruction and the process logic constraints.

[0025] If the process recommendation mode is determined, process information is obtained by querying the knowledge base, accessing the pre-trained data of the large language model, or performing online retrieval. A process solution that matches the user's intent is generated and fed back to the user. Based on the user's confirmation signal, the process solution is transformed into structured process data.

[0026] Specifically, the matching and generation of process solutions includes: extracting the material grade, workpiece size and processing type from the user description as key entity features, matching the standard process template with the highest similarity, and using a large language model to fine-tune the process template parameters based on the key entity features to generate a process solution.

[0027] The process plan is presented to the user, and only after receiving confirmation from the user is the process plan considered valid input and proceed to the aforementioned "direct instruction mode" parsing process, ultimately generating the corresponding structured process data. This confirmation process is primarily used to ensure the accuracy of the input data.

[0028] In existing technologies, traditional equipment is merely an actuator and cannot provide process guidance for novice operators. When users are unaware of specific parameters, such as how to braze copper, the equipment cannot respond, lacking support for process decision-making. The process recommendation mode proposed in this application can specifically address these problems in existing technologies.

[0029] It should be noted that the set of process logic constraint rules includes at least: a first constraint rule that restricts the data format, and a second constraint rule that sets the temperature continuity between adjacent steps in the heat preservation command. For example, the first constraint rule restricts the output data to a standard key-value pair structure containing the step number, target value, and duration; the second constraint rule, when parsing the heat preservation command, forces the starting temperature of the next step to be consistent with the target temperature of the current step, in order to maintain the temperature continuity of the heat treatment process.

[0030] S3. Scan the structured process data, complete the missing parameters in the structured process data, and filter out invalid steps that belong to uncontrolled physical evolution processes from the completed structured process data.

[0031] Among them, the missing parameters in the structured process data that need to be supplemented include: When a process step has a variable rate parameter but lacks a process duration parameter, the physical correlation model is invoked to calculate and complete the parameter. For example, if a process step has a set temperature change rate K but no set time, the target temperature value for the current process step is extracted. The end temperature value of the previous step Call the physical association model, according to the formula Automatically calculate and complete.

[0032] When an initial state parameter is missing for a process step, it is filled in based on the default value. For example, if the user requests "heating from room temperature to 600 degrees Celsius", the initial temperature is missing, so the default room temperature value (such as 20°C) is automatically filled in.

[0033] Among them, the uncontrolled physical evolution process is furnace cooling or air cooling. The uncontrolled physical evolution process is a step that describes a natural process of equipment that does not require active control. Therefore, it is judged as an invalid step and is not considered a valid instruction.

[0034] S4. Based on the configuration information of the heat treatment equipment, load the corresponding driver module from the preset protocol driver library, convert the filtered structured process data into an executable instruction sequence for the heat treatment equipment, add an end code (including a specific numerical code or a specific status bit), and send the instruction sequence to the heat treatment equipment for execution.

[0035] Specifically, the protocol driver library contains register address mapping tables and special function code definitions for various models of heat treatment equipment. The filtered structured process data is filled into specific register addresses according to the mapping table, thereby forming an executable instruction sequence that can be written into the registers of the heat treatment equipment. Furthermore, at the end of the generated instruction sequence, an end code in the special function code is generated according to the driver definition to mark the termination of the process.

[0036] Finally, the instruction sequence is sent to the heat treatment equipment for execution via an industrial communication interface (such as RS-485).

[0037] The following is a description through specific embodiments.

[0038] Example 1: When the user command is in direct command mode, the method steps are as follows: S1. The user, through voice interaction, requests: "Heat from room temperature to 300 degrees Celsius at a rate of 20 degrees Celsius per minute, then maintain the temperature for 1 hour, then heat to 600 degrees Celsius for 30 minutes, then maintain the temperature for 2 hours, and finally cool with the furnace." Semantic recognition technology interprets the user's command as a direct command mode.

[0039] It should be noted that the temperature unit "degree" mentioned in this application is actually Celsius, but it is displayed as "degree" in order to conform to users' daily language habits.

[0040] S2. Based on user instructions, generate corresponding structured process data according to the set of process logic constraint rules: "SP1: room temperature, heating rate K1: 20 degrees Celsius / minute, SP2: 300 degrees Celsius, t2: 60 minutes, SP3: 300 degrees Celsius, t3: 30 minutes, SP4: 600 degrees Celsius, t4: 120 minutes, SP5: 600 degrees Celsius, cooled with furnace."

[0041] S3. Scan the structured process data, complete SP1 to 20 degrees Celsius, calculate the heating time t1 = (300-20) / 20 = 14 minutes for the temperature range from SP1 to SP2 at the user-given temperature change rate, and remove the furnace cooling requirement based on semantic filtering to obtain the final structured process data: "SP1: 20, t1: 14, SP2: 300, t2: 60, SP3: 300, t3: 30, SP4: 600, t4: 120, SP5: 600".

[0042] S4. Based on the temperature controller configured for the heat treatment equipment (e.g., a Yudian temperature controller), and according to the Modbus communication protocol supported by the Yudian temperature controller, convert SP1, t1, SP2, t2, SP3, t3, SP4, t4, SP5 and their values ​​into hexadecimal codes. Add a checksum to each instruction to form 9 hexadecimal message groups executable by the temperature controller. Finally, based on the Yudian temperature controller's settings, add an end code message (t5: -121, which must also be converted to hexadecimal and have a checksum added) and a program start message (Srun: 0, which must also be converted to hexadecimal and have a checksum added) after the 9 heat preservation groups. The final message group is: "01 06 00 50 00 C8 88 4D 01 06 00 51 00 8C D9 BE 01 06 00 52 0B B8 2F 59 01 06 00 53 02 58 79 41 01 06 00 54 0B B8 CF 58 01 06 00 55 00 64 98 31 01 06 00 56 17 70 67 CE 01 06 00 57 04 B0 3B 6E 01 06 00 58 17 70 06 0D 01 06 00 59 FB 46 9B 1B 01 06 00 1B 00 00 F9 CD” Finally, the Modbus message group is transmitted to the Yudian temperature controller through the hardware layer. After receiving the message instructions, the Yudian temperature controller writes them into the program group one by one and finally starts the heat treatment program.

[0043] like Figure 2-3 As shown in Example 2: When the user instruction is in process recommendation mode, the method steps are as follows: S1: The user makes a request through voice interaction, "I want to improve the plasticity of TC4 titanium alloy, please recommend a suitable process". The semantic recognition technology analyzes the user's command into a process recommendation mode.

[0044] S2. Based on user instructions, three annealing process schemes for improving the plasticity of TC4 titanium alloy are retrieved from the knowledge base, including ordinary annealing, double annealing, and β-phase region annealing. All three annealing process schemes meet the user's requirements and are therefore fed back to the user. If the user further confirms the implementation of the ordinary annealing process, based on the user's feedback, the user-confirmed standard annealing process scheme (e.g., "heat TC4 titanium alloy to 750℃ (mid-700-800℃), hold for 1.5 hours (mid-1-2 hours), then air cool") is converted into corresponding structured process data, as follows: "SP1: None, heating time: None, SP2: 750 degrees Celsius, t2: 90 minutes, SP3: 750 degrees Celsius, air cooling."

[0045] S3. Scan the structured process data, complete SP1 to 20 degrees Celsius, calculate the heating time t1 = (750-20) / 10 = 73 minutes for the temperature range from SP1 to SP2 at the user-given temperature change rate (e.g., user-supplemented heating rate is 10 degrees per minute), and remove the furnace cooling requirement based on semantic filtering to obtain the final structured process data: "SP1: 20, t1: 73, SP2: 750, t2: 90, SP3: 750".

[0046] S4. Based on the temperature controller configured for the heat treatment equipment (e.g., a Yudian temperature controller), and according to the Modbus communication protocol supported by the Yudian temperature controller, convert SP1, t1, SP2, t2, SP3 and their values ​​into hexadecimal codes. Add a checksum to each instruction to form 5 hexadecimal message groups executable by the temperature controller. Finally, based on the Yudian temperature controller's settings, add an end code message (t3: -121, which must also be converted to hexadecimal and have a checksum added) and a program start message (Srun: 0, which must also be converted to hexadecimal and have a checksum added) after the 5 heat preservation groups. The final message group is: "01 06 00 50 00 C8 88 4D 01 06 00 51 02 DA 58 E0 01 06 00 52 1D 4C 20 BE 01 06 00 53 03 84 79 48 01 06 00 54 1D 4C C0 BF 01 06 00 55 FB 46 5B 18 01 06 00 1B 00 00 F9 CD” Finally, the Modbus message group is transmitted to the Yudian temperature controller through the hardware layer. After receiving the message instructions, the Yudian temperature controller writes them into the program group one by one and finally starts the heat treatment program.

[0047] In a specific embodiment, the present invention also provides an intelligent control system for heat treatment equipment based on semantic parsing, for implementing the above-mentioned method, including an interaction and feedback unit, a data parsing unit, a data verification unit, a protocol adaptation unit, and a communication unit.

[0048] The interaction and feedback unit receives natural language input from the user and provides a process plan. It offers a text or voice interface, allowing users to issue explicit commands (e.g., "Heat to 300 degrees Celsius, heating rate 20 degrees Celsius per minute") or vague requests (e.g., "How to improve the plasticity of TC4 titanium alloy"). In the process recommendation mode, the interaction and feedback unit can display the generated recommended process plan to the user in the form of a process curve graph or text description, and await the user's confirmation.

[0049] The data parsing unit is used to parse user instructions using a large language model and generate corresponding structured process parameters. The data parsing unit utilizes the large language model to perform semantic understanding of user input, determining whether the user instruction is a direct instruction mode or a process recommendation mode.

[0050] When the direct instruction mode is determined, the data parsing unit calls the preset set of process logic constraint rules to perform structured parsing of the instruction.

[0051] When the process recommendation mode is determined, the data parsing unit extracts key entity features from the user description, such as material grade, workpiece size, and processing type (e.g., annealing, quenching). Then, it obtains process information through at least one of the following methods: querying the knowledge base, accessing pre-trained data of the large language model, or performing an online search, and matches the most similar standard process template. Utilizing the reasoning capabilities of the large language model, the template parameters are fine-tuned based on specific entity features to generate a personalized recommended process plan (e.g., specific temperature, time, cooling method). Only after the user confirms this process plan is it considered valid input and enters the subsequent instruction parsing process, where it is transformed into structured process data.

[0052] The data verification unit is used to perform parameter calculation and completion on the structured process data and filter out invalid steps. The data verification unit scans the structured process data; when a step has variable parameters but lacks process continuity parameters, it calls the physical correlation model to calculate and complete the parameters. When a step lacks initial state parameters, it completes the parameters based on default values.

[0053] In addition, the data verification unit performs semantic classification of the intent of all steps. For steps identified as uncontrolled physical evolution processes such as "furnace cooling" or "air cooling", since these processes are completed naturally by the equipment or environment and do not require active control by the equipment, the system will mark them as invalid and filter them out in the instruction generation queue to prevent sending invalid or conflicting instructions to the equipment.

[0054] The protocol adaptation unit has a built-in extensible protocol driver library, which is used to convert the filtered structured process data into an executable instruction sequence and attach an end code according to the heat treatment equipment.

[0055] Structured process data is a hardware-independent, universal format. The protocol driver library pre-loads drivers for different brands and models of heat treatment equipment (such as Uden and Shimadzu temperature controllers). Each driver contains a Modbus register address mapping table for the device (e.g., the corresponding register addresses for the target temperature SP and runtime t) and special function code definitions. The system loads the corresponding driver based on the configuration of the heat treatment equipment.

[0056] The protocol adaptation unit fills the general structured process data into specific register addresses according to the driver definition, generating a series of Modbus write register messages in hexadecimal format. Finally, according to the device requirements, a device-specific process termination signature is injected at the end of the message sequence to mark the termination of the process.

[0057] The communication unit is used to send equipment command sequences to the heat treatment equipment. Through industrial communication interfaces such as RS-485 and Ethernet, the communication unit sends the complete Modbus command sequence generated by the protocol adapter unit to the controller of the target heat treatment equipment. After receiving and writing the command sequence into the corresponding registers, the controller begins execution according to the set process program.

[0058] The technical effects of this invention are as follows: 1. Reduced operational threshold and cognitive load: By parsing natural language instructions, operators do not need to master complex programming code or manually convert parameters, making it easy to use and with a low barrier to entry.

[0059] 2. Provides process decision support: Based on demand matching and recommendation, this invention can provide standardized process solutions for users who are not familiar with the process through a large language model.

[0060] 3. Improved determinism and safety of the control process: This invention can automatically calculate and complete parameters through physical correlation models, eliminating errors from manual calculations; and automatically eliminate invalid or uncontrolled steps through semantic filtering, avoiding the risk of misoperation.

[0061] 4. Achieves universal adaptation to heterogeneous devices: The structured data and protocol driver library work together to separate the upper-layer semantic control logic from the lower-layer hardware protocol. One system can be compatible with various brands and models of heat treatment equipment controllers, improving the system's versatility and scalability.

[0062] 5. Covers all user scenarios: The dual-modal design of direct command mode and process recommendation mode can meet the precise and efficient control needs of skilled engineers, as well as support the learning and standardized operation needs of novice operators.

[0063] The term "an embodiment" or "embodiment" as used in this invention refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the invention. In the description of this invention, it should be understood that the terms "first," "second," and "third," etc., in the specification, claims, and accompanying drawings are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0064] This specification provides method operation steps as shown in the embodiments or flowcharts, but based on conventional or non-inventive labor, more or fewer operation steps may be included. The order of steps listed in the embodiments is merely one of many possible execution orders and does not represent the only possible execution order. In actual system or server product execution, the method can be executed in the order shown in the embodiments or drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment), or the execution order of steps without timing constraints can be adjusted.

[0065] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for intelligent control of heat treatment equipment based on semantic parsing, characterized in that, include: S1. Receive natural language information input by the user and parse it using semantic recognition technology to obtain the user's instructions; S2. Based on user instructions, generate corresponding structured process data according to a set of preset process logic constraint rules; S3. Scan the structured process data, complete the missing parameters in the structured process data, and filter out invalid steps belonging to uncontrolled physical evolution processes from the completed structured process data. S4. Based on the configuration information of the heat treatment equipment, load the corresponding driver module from the preset protocol driver library, convert the filtered structured process data into an instruction sequence that can be executed by the heat treatment equipment, add an end code, and send the instruction sequence to the heat treatment equipment for execution.

2. The method according to claim 1, characterized in that, The semantic recognition technology is implemented through a large language model.

3. The method according to claim 2, characterized in that, In S1, the process of parsing and obtaining user instructions includes: determining whether the user instruction is a direct instruction mode or a process recommendation mode based on the parsing result.

4. The method according to claim 3, characterized in that, In S2, if the direct instruction mode is determined, structured process data is generated according to the preset set of process logic constraint rules; if the process recommendation mode is determined, process information is obtained by querying the knowledge base, accessing the pre-trained data of the large language model, or performing online retrieval, and a process solution that meets the user's intention is generated and fed back to the user. Based on the confirmation signal from the user's feedback, the process solution is converted into structured process data.

5. The method according to claim 4, characterized in that, In the process recommendation mode, the process scheme generation process includes: extracting the material grade, workpiece size and processing type from the user description as key entity features, matching a standard process template, and using a large language model to fine-tune the process template parameters based on the key entity features to generate a process scheme.

6. The method according to claim 1, characterized in that, The set of process logic constraint rules includes at least: a first constraint rule that limits the data format, and a second constraint rule that sets the temperature continuity of adjacent steps in the heat preservation instruction.

7. The method according to claim 1, characterized in that, In S3, the missing parameters in the structured process data are supplemented as follows: when a step has variable parameters but lacks process continuity parameters, the physical association model is called to calculate and supplement them; when a step lacks initial state parameters, they are supplemented based on default values.

8. The method according to claim 1, characterized in that, The protocol driver library contains register address mapping tables and special function code definitions for various models of heat treatment equipment. The filtered structured process data is filled into a specific register address, and the end code in the special function code is generated.

9. The method according to claim 1, characterized in that, The uncontrolled physical evolution process is either furnace cooling or air cooling.

10. A semantic parsing-based intelligent control system for heat treatment equipment, used to implement the method as described in any one of claims 1-9, characterized in that, include: The interaction and feedback unit is used to receive natural language information input by the user and provide feedback on the process plan; The data parsing unit is used to parse user instructions using a large language model and generate structured process parameters; The data verification unit is used to perform parameter calculation and completion and invalid process step filtering on the structured process data; The protocol adaptation unit has a built-in extensible protocol driver library, which is used to convert the filtered structured process data into an executable instruction sequence and attach an end code according to the heat treatment equipment. A communication unit is used to send the device instruction sequence to the heat treatment device.

Images