Temperature control method for semiconductor metrology inspection machine and related apparatus

Abstract

Embodiments of the present application disclose a temperature control method and device for a semiconductor metrology and inspection machine, and a computer readable storage medium, for improving the accuracy of temperature control of the semiconductor metrology and inspection machine. The method comprises inputting a current temperature and a target temperature into a temperature PID control algorithm to obtain a first power corresponding to a heating capacitor output by the temperature PID control algorithm, inputting the target temperature into a deep learning model to obtain a second power corresponding to the heating capacitor output by the deep learning model, determining a target power based on the first power and the second power, generating a power control command for the heating capacitor based on the target power, and sending the power control command for the heating capacitor to the semiconductor metrology and inspection machine, so that the semiconductor metrology and inspection machine controls the power of the heating capacitor to the target power based on the power control command.

Description

Technical Field

[0001] This application relates to the field of temperature control for semiconductor metrology and testing equipment, and more specifically, to temperature control methods, temperature control devices, and computer-readable storage media for semiconductor metrology and testing equipment. Background Technology

[0002] Semiconductor metrology and inspection equipment is a device used for the measurement and defect detection of semiconductor devices. In the semiconductor manufacturing process, semiconductor metrology and inspection equipment is widely used in the wafer testing stage. Specifically, it utilizes a probe to detect the wafer surface, a computer processes the detected images to extract defects, and stores the processed image data in the computer. The computer can then display the wafer defect detection images and determine the distribution of defects on the wafer based on the detection images.

[0003] However, semiconductor metrology and testing equipment can experience temperature instability, which can affect the accuracy of related measuring instruments within the equipment. For example, lasers can be affected by the temperature instability, leading to lower accuracy in measurement data. The current solution is to manually control the temperature of the semiconductor metrology and testing equipment based on experience. However, manual control has a large margin of error. Therefore, there is an urgent need for a temperature control method for semiconductor metrology and testing equipment to improve the accuracy of temperature control. Summary of the Invention

[0004] This application provides a temperature control method, a temperature control device, and a computer-readable storage medium for a semiconductor metrology and testing equipment, which improves the accuracy of temperature control in the semiconductor metrology and testing equipment.

[0005] In a first aspect, embodiments of this application provide a temperature control method for a semiconductor metrology and testing machine, comprising:

[0006] Determine the current temperature and target temperature of the semiconductor measurement and testing equipment;

[0007] The current temperature and target temperature are input into the temperature PID control algorithm, which adjusts the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm; wherein, the first power represents the power required for the heating capacitor to reach when the error is less than or equal to a preset error threshold; and the heating capacitor is a component in the circuit of the semiconductor measurement and testing machine.

[0008] The target temperature is input into a pre-trained deep learning model. The pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature and predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a steady state. This yields the second power of the heating capacitor output by the pre-trained deep learning model. The second power represents the power required for the heating capacitor to reach the target temperature when the temperature of the semiconductor measurement and testing instrument is reached.

[0009] The target power is determined based on the first power and the second power, and a power control command for the heating capacitor is generated based on the target power.

[0010] The power control command for the heating capacitor is sent to the semiconductor measurement and testing equipment, so that the semiconductor measurement and testing equipment controls the power of the heating capacitor to the target power based on the power control command, thereby controlling the temperature of the semiconductor measurement and testing equipment.

[0011] Secondly, embodiments of this application provide a temperature control device, including:

[0012] The determination unit is used to determine the current temperature and target temperature of the semiconductor measurement and testing equipment;

[0013] The input unit is used to input the current temperature and the target temperature into the temperature PID control algorithm. The temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required for the heating capacitor to reach when the error is less than or equal to a preset error threshold. The heating capacitor is a component in the circuit of the semiconductor measurement and testing machine.

[0014] The input unit is also used to input the target temperature into a pre-trained deep learning model. The pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature and predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a steady state, so as to obtain the second power of the heating capacitor output by the pre-trained deep learning model. The second power represents the power required for the heating capacitor to reach the target temperature when the temperature of the semiconductor measurement and testing instrument reaches the target temperature.

[0015] The generation unit is used to determine the target power based on the first power and the second power, and to generate a power control command for the heating capacitor based on the target power;

[0016] The transmitting unit is used to send the power control command of the heating capacitor to the semiconductor measurement and testing equipment, so that the semiconductor measurement and testing equipment controls the power of the heating capacitor to the target power based on the power control command, so as to control the temperature of the semiconductor measurement and testing equipment.

[0017] Thirdly, embodiments of this application provide a temperature control device, including:

[0018] Central processing unit, memory, input / output interfaces, wired or wireless network interfaces, and power supply;

[0019] The memory can be either temporary or permanent storage.

[0020] The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the aforementioned temperature control method for the semiconductor measurement and testing equipment.

[0021] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the aforementioned temperature control method for a semiconductor measurement and testing machine.

[0022] Fifthly, embodiments of this application provide a computer program product containing instructions that, when run on a computer, cause the computer to execute the aforementioned temperature control method for a semiconductor measurement and testing machine.

[0023] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: The current temperature and target temperature can be input into a temperature PID control algorithm to obtain a first power corresponding to the heating capacitor output by the temperature PID control algorithm; the target temperature can be input into a pre-trained deep learning model to obtain a second power corresponding to the heating capacitor output by the pre-trained deep learning model; the target power is determined based on the first and second powers; and a power control command for the heating capacitor is generated based on the target power. This power control command is then sent to the semiconductor measurement and testing equipment, enabling the semiconductor measurement and testing equipment to control the power of the heating capacitor to the target power based on the power control command. Temperature control of the semiconductor measurement and testing equipment can be performed based on a deep learning model and a temperature PID control algorithm, thus improving the accuracy of temperature control in the semiconductor measurement and testing equipment. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the architecture of a temperature control system for a semiconductor measurement and testing machine disclosed in an embodiment of this application;

[0025] Figure 2 This is a schematic flowchart of a temperature control method for a semiconductor metrology and testing machine disclosed in an embodiment of this application;

[0026] Figure 3 This is a schematic flowchart of another temperature control method for a semiconductor metrology and testing machine disclosed in an embodiment of this application;

[0027] Figure 4 This is a schematic diagram of the structure of a temperature control device disclosed in an embodiment of this application;

[0028] Figure 5 This is a schematic diagram of another temperature control device disclosed in an embodiment of this application;

[0029] Figure 6 This is a schematic diagram of the structure of another temperature control device disclosed in the embodiments of this application. Detailed Implementation

[0030] This application provides a temperature control method, a temperature control device, and a computer-readable storage medium for a semiconductor metrology and testing equipment, which improves the accuracy of temperature control in the semiconductor metrology and testing equipment.

[0031] Please see Figure 1 The architecture of the temperature control system for the semiconductor metrology and testing equipment in this embodiment includes:

[0032] Temperature control device 101 and semiconductor measurement and testing machine 102. When performing temperature control on the semiconductor measurement and testing machine 102, temperature control device 101 can obtain the current temperature and target temperature of the semiconductor measurement and testing machine 102, determine the target power corresponding to the heating capacitor based on the current temperature and target temperature, generate a power control command for the heating capacitor, and send the power control command to the semiconductor measurement and testing machine 102 so that the semiconductor measurement and testing machine controls the power of the heating capacitor to the target power based on the power control command, thereby controlling the temperature of the semiconductor measurement and testing machine.

[0033] based on Figure 1 Please refer to the temperature control system of the semiconductor measurement and testing machine shown. Figure 2 , Figure 2 This is a flowchart illustrating a temperature control method for a semiconductor metrology and testing machine disclosed in an embodiment of this application. The method includes:

[0034] 201. Determine the current temperature and target temperature of the semiconductor measurement and testing equipment.

[0035] In this embodiment, when performing temperature control of the semiconductor measurement and testing equipment, the current temperature and target temperature of the semiconductor measurement and testing equipment can be determined.

[0036] 202. Input the current temperature and target temperature into the temperature PID control algorithm. The temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required for the heating capacitor to reach when the error is less than or equal to the preset error threshold. The heating capacitor is a component in the circuit of the semiconductor measurement and testing machine.

[0037] After determining the current temperature and target temperature of the semiconductor measurement and testing machine, the current temperature and target temperature can be input into the temperature PID control algorithm. The temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required for the heating capacitor to reach when the error is less than or equal to a preset error threshold. The heating capacitor is a component in the circuit of the semiconductor measurement and testing machine.

[0038] One method for obtaining the first power output by the heating capacitor from the temperature PID control algorithm is to adjust the error between the current temperature and the target temperature using the temperature PID control algorithm. This can be achieved by adjusting the current temperature based on the magnitude, duration, and / or rate of change of the error, thereby adjusting the error and obtaining the first power output by the temperature PID control algorithm for the heating capacitor. Other reasonable methods are also possible, and specific methods are not limited here.

[0039] 203. Input the target temperature into the pre-trained deep learning model. The pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature and predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a steady state, so as to obtain the second power of the heating capacitor output by the pre-trained deep learning model; wherein, the second power represents the power required for the heating capacitor to reach when the temperature of the semiconductor measurement and testing instrument reaches the target temperature.

[0040] The current temperature and target temperature are input into a temperature PID control algorithm. The algorithm adjusts the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required by the heating capacitor when the error is less than or equal to a preset error threshold. After the heating capacitor is a component in the circuit of the semiconductor measurement and testing machine, the target temperature can be input into a pre-trained deep learning model. The pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature and predicts the power corresponding to the heating capacitor when the voltage and current of the heating capacitor reach a stable state, thus obtaining the second power corresponding to the heating capacitor output by the pre-trained deep learning model. The second power represents the power required by the heating capacitor when the temperature of the semiconductor measurement and testing machine reaches the target temperature.

[0041] 204. Determine the target power based on the first power and the second power, and generate a power control command for the heating capacitor based on the target power.

[0042] The target temperature is input into a pre-trained deep learning model. The pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature, and then predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a steady state. After obtaining the second power of the heating capacitor output by the pre-trained deep learning model, the target power can be determined based on the first power and the second power, and a power control command for the heating capacitor can be generated based on the target power.

[0043] 205. Send the power control command of the heating capacitor to the semiconductor measurement and testing equipment so that the semiconductor measurement and testing equipment controls the power of the heating capacitor to the target power based on the power control command, so as to control the temperature of the semiconductor measurement and testing equipment.

[0044] After determining the target power based on the first power and the second power, and generating a power control command for the heating capacitor based on the target power, the power control command for the heating capacitor can be sent to the semiconductor measurement and testing equipment. This allows the semiconductor measurement and testing equipment to control the power of the heating capacitor to the target power based on the power control command, thereby controlling the temperature of the semiconductor measurement and testing equipment.

[0045] In this embodiment, the current temperature and target temperature can be input into a temperature PID control algorithm to obtain a first power corresponding to the heating capacitor output by the temperature PID control algorithm. The target temperature is then input into a pre-trained deep learning model to obtain a second power corresponding to the heating capacitor output by the pre-trained deep learning model. Based on the first and second powers, a target power is determined, and a power control command for the heating capacitor is generated based on the target power. This power control command is sent to the semiconductor measurement and testing equipment, enabling the equipment to control the heating capacitor's power to the target power. Temperature control of the semiconductor measurement and testing equipment can be performed based on a deep learning model and a temperature PID control algorithm, improving the accuracy of temperature control in the equipment.

[0046] In this embodiment, there are various methods for adjusting the error between the current temperature and the target temperature using a temperature PID control algorithm to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. Figure 2 The temperature control method of the semiconductor measurement and testing machine shown is described below, and one of the methods is described below.

[0047] In this embodiment, when performing temperature control of the semiconductor measurement and testing equipment, the current temperature and target temperature of the semiconductor measurement and testing equipment can be determined.

[0048] After determining the current temperature and target temperature of the semiconductor measurement and testing machine, the current temperature and target temperature can be input into the temperature PID control algorithm. The temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required for the heating capacitor to reach when the error is less than or equal to a preset error threshold. The heating capacitor is a component in the circuit of the semiconductor measurement and testing machine.

[0049] One method for adjusting the error between the current temperature and the target temperature using a temperature PID control algorithm to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm can be that the temperature PID control algorithm adjusts the current temperature according to the magnitude, duration, and / or rate of change of the error to adjust the error and obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm.

[0050] Specifically, the temperature PID control algorithm is a type of PID control algorithm. The working principle of the PID control algorithm is based on error feedback control. The PID algorithm is a classic control algorithm that minimizes the system response error by continuously adjusting the controller output. The PID algorithm includes three main control parameters: proportional gain, integral gain, and derivative gain. By adjusting these three parameters, the PID controller can adjust the error according to the actual response of the system. The proportional gain adjusts the controller output based on the magnitude of the error. When the error increases, the proportional gain increases the controller output, thus increasing the system response; when the error decreases, the proportional gain decreases the controller output, thus decreasing the system response. The integral gain adjusts the controller output based on the duration of the error. When the error persists, the integral gain increases the controller output, thus accelerating the system response; when the error gradually decreases, the integral gain decreases the controller output, thus decreasing the system response. The derivative gain adjusts the controller output based on the rate of change of the error. When the error rate of change is large, the derivative gain increases the controller output, thereby improving the system's response speed; when the error rate of change is small, the derivative gain decreases the controller output, thereby reducing the system's response speed. It can be understood that by continuously adjusting the proportional gain, integral gain, and derivative gain, the PID controller can minimize the error when the system reaches a steady state and can respond quickly to external disturbances. In summary, the working principle of the PID control algorithm is to adjust the controller output according to the magnitude, duration, and rate of change of the error to minimize the system's response error.

[0051] One method for adjusting the error between the current temperature and the target temperature using a temperature PID control algorithm to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm can be an incremental PID algorithm used to adjust the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. Here, the temperature PID control algorithm is an incremental PID algorithm.

[0052] It is worth mentioning that adjusting the error between the current temperature and the target temperature using an incremental PID algorithm can eliminate tolerances, resulting in higher accuracy in calculating the first power corresponding to the heating capacitor. It is understandable that, in addition to using an incremental PID algorithm to adjust the error between the current temperature and the target temperature, a positional PID algorithm can also be used, or other reasonable temperature PID control algorithms can be employed; specific methods are not limited here.

[0053] After determining the target temperature of the semiconductor measurement and testing equipment, the target temperature can be input into a pre-trained deep learning model. The pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature, and then predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a steady state, so as to obtain the second power of the heating capacitor output by the pre-trained deep learning model. The second power represents the power required for the heating capacitor to reach the target temperature when the temperature of the semiconductor measurement and testing equipment reaches the target temperature.

[0054] It is worth mentioning that the method of inputting the current temperature and target temperature into a temperature PID control algorithm, and having the temperature PID control algorithm adjust the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm, can be as follows: A first task is executed through a first thread. The first task is to input the current temperature and target temperature into the temperature PID control algorithm, and have the temperature PID control algorithm adjust the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. Furthermore, the target temperature is input into a pre-trained deep learning model, and the pre-trained deep learning model... The method for predicting the power of the heating capacitor when its voltage and current reach a steady state after adjusting the voltage and current of the heating capacitor at the target temperature, in order to obtain the second power of the heating capacitor output by the pre-trained deep learning model, can be achieved by executing a second task through a second thread. The second task involves inputting the target temperature into the pre-trained deep learning model, which then adjusts the voltage and current of the heating capacitor based on the target temperature to predict the power of the heating capacitor when its voltage and current reach a steady state, thus obtaining the second power of the heating capacitor output by the pre-trained deep learning model. The first thread and the second thread are executed in parallel.

[0055] Before inputting the current temperature and target temperature into the temperature PID control algorithm, a target temperature sample can be obtained first. The target temperature sample is labeled with the second power corresponding to the heating capacitor. Then, the target temperature sample is input into the deep learning model to obtain the predicted second power corresponding to the target temperature sample output by the deep learning model. Finally, the loss between the predicted second power and the labeled second power is calculated according to the regression loss function. When the loss meets the convergence condition, the trained deep learning model is obtained.

[0056] The target temperature is input into a pre-trained deep learning model. The pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature, and then predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a steady state. After obtaining the second power of the heating capacitor output by the pre-trained deep learning model, the target power can be determined based on the first power and the second power, and a power control command for the heating capacitor can be generated based on the target power.

[0057] One method for determining the target power based on the first power and the second power is to first determine the first weight corresponding to the first power and the second weight corresponding to the second power, and then perform a weighted average of the first power, the second power, the first weight, and the second weight to obtain the target power. Specifically, if both the first weight and the second weight are 50%, then the first power and the second power can be averaged to obtain the target power. It is understood that other reasonable methods for determining the target power based on the first power and the second power are also possible, and no specific limitations are imposed here.

[0058] After determining the target power based on the first power and the second power, and generating a power control command for the heating capacitor based on the target power, the power control command for the heating capacitor can be sent to the semiconductor measurement and testing equipment. This allows the semiconductor measurement and testing equipment to control the power of the heating capacitor to the target power based on the power control command, thereby controlling the temperature of the semiconductor measurement and testing equipment.

[0059] Specifically, a power control command for the heating capacitor is sent to a semiconductor measurement and testing machine, which then controls the power of the heating capacitor to the target power based on the power control command. After controlling the temperature of the semiconductor measurement and testing machine, the control effect of controlling the power of the heating capacitor can be obtained first. Then, the deep learning model is fine-tuned based on the control effect to obtain the fine-tuned deep learning model.

[0060] It is worth mentioning that model fine-tuning can adjust the pre-trained model according to task requirements. Depending on the specific situation, global fine-tuning, hierarchical fine-tuning, or parameter freezing can be selected to achieve better performance and generalization ability.

[0061] For details, please refer to Figure 3 , Figure 3 This application discloses another method for temperature control of a semiconductor metrology and testing machine, which is... Figure 3It is known that the input sensor signals and control state binary signals can be obtained. The temperature PID control algorithm can calculate the ratio, integral, and derivative of the error between the current value (current temperature) and the target value (target temperature) of the control term (temperature term). The temperature PID control algorithm can be an incremental PID, which can calculate the current power of the heating capacitor. A deep learning prediction algorithm models and predicts the power of the heating capacitor. Furthermore, the incremental PID algorithm can run in parallel with the AI ​​prediction algorithm (a pre-trained deep learning model). The first power output from the incremental PID algorithm and the second power output from the AI ​​prediction algorithm can be cross-checked and regularized before being added to merge into a more accurate result (target power). For the AI ​​prediction algorithm, historical data can be obtained to build a model. Reinforcement learning can be used to fine-tune the AI ​​prediction algorithm, and the AI ​​prediction algorithm (AI model) can be continuously updated and fine-tuned based on the actual control situation.

[0062] In this embodiment, the current temperature and target temperature can be input into a temperature PID control algorithm to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. The target temperature is then input into a pre-trained deep learning model to obtain the second power corresponding to the heating capacitor output by the pre-trained deep learning model. Based on the first and second powers, the target power is determined, and a power control command for the heating capacitor is generated based on the target power. This power control command is sent to the semiconductor measurement and testing equipment, enabling the equipment to control the heating capacitor's power to the target power. Temperature control of the semiconductor measurement and testing equipment can be performed based on a deep learning model and a temperature PID control algorithm, improving the accuracy of temperature control and the accuracy of measurement data from related instruments within the equipment. Furthermore, temperature control based on a deep learning model and a temperature PID control algorithm allows for real-time temperature stabilization, improving both the measurement accuracy and the stability of the equipment's measurement performance. Furthermore, it enables efficient, precise, and rapid temperature control, unaffected by ambient temperature, providing real-time calculation and instantaneous feedback for highly sensitive temperature control. Moreover, the fusion of PID control with deep learning improves control accuracy, and cross-validation avoids noise interference and reduces measurement errors. Furthermore, multi-threaded parallel algorithms allow for efficient and rapid batch processing of multi-memory data, achieving speeds down to 1ms, enabling near real-time processing and control, and reducing image generation time. Additionally, incremental PID algorithms can adjust the error between the current and target temperatures, eliminating tolerances and resulting in more accurate calculations of the first power corresponding to the heating capacitor. Finally, model fine-tuning allows adjustments to the pre-trained model based on task requirements, enabling global fine-tuning, hierarchical fine-tuning, or parameter freezing to achieve better performance and generalization capabilities.

[0063] The temperature control method for the semiconductor metrology and inspection equipment in the embodiments of this application has been described above. The temperature control device in the embodiments of this application is described below. Please refer to [link / reference]. Figure 4 One embodiment of the temperature control device in this application includes:

[0064] The determining unit 401 is used to determine the current temperature and target temperature of the semiconductor measurement and testing equipment;

[0065] Input unit 402 is used to input the current temperature and the target temperature into a temperature PID control algorithm. The temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain a first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required by the heating capacitor when the error is less than or equal to a preset error threshold. The heating capacitor is a component in the circuit of the semiconductor measurement and testing machine.

[0066] The input unit 402 is further configured to input the target temperature into a pre-trained deep learning model, which adjusts the voltage and current of the heating capacitor based on the target temperature, and then predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a stable state, so as to obtain the second power of the heating capacitor output by the pre-trained deep learning model; wherein, the second power represents the power required by the heating capacitor when the temperature of the semiconductor measurement and testing instrument reaches the target temperature;

[0067] The generation unit 403 is used to determine a target power based on the first power and the second power, and to generate a power control command for the heating capacitor based on the target power;

[0068] The sending unit 404 is used to send a power control command of the heating capacitor to the semiconductor measurement and testing equipment, so that the semiconductor measurement and testing equipment controls the power of the heating capacitor to the target power based on the power control command, so as to control the temperature of the semiconductor measurement and testing equipment.

[0069] In this embodiment, the current temperature and target temperature can be input into a temperature PID control algorithm to obtain a first power corresponding to the heating capacitor output by the temperature PID control algorithm. The target temperature is then input into a pre-trained deep learning model to obtain a second power corresponding to the heating capacitor output by the pre-trained deep learning model. Based on the first and second powers, a target power is determined, and a power control command for the heating capacitor is generated based on the target power. This power control command is sent to the semiconductor measurement and testing equipment, enabling the equipment to control the heating capacitor's power to the target power. Temperature control of the semiconductor measurement and testing equipment can be performed based on a deep learning model and a temperature PID control algorithm, improving the accuracy of temperature control in the equipment.

[0070] The temperature control device in the embodiments of this application is described in detail below. Please refer to [link / reference]. Figure 5 Another embodiment of the temperature control device in this application includes:

[0071] The determining unit 501 is used to determine the current temperature and target temperature of the semiconductor measurement and testing equipment;

[0072] Input unit 502 is used to input the current temperature and the target temperature into a temperature PID control algorithm. The temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain a first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required by the heating capacitor when the error is less than or equal to a preset error threshold. The heating capacitor is a component in the circuit of the semiconductor measurement and testing machine.

[0073] The input unit 502 is further configured to input the target temperature into a pre-trained deep learning model, which adjusts the voltage and current of the heating capacitor based on the target temperature, and then predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a stable state, so as to obtain the second power of the heating capacitor output by the pre-trained deep learning model; wherein, the second power represents the power required for the heating capacitor to reach when the temperature of the semiconductor measurement and testing instrument reaches the target temperature;

[0074] The generation unit 503 is used to determine a target power based on the first power and the second power, and to generate a power control command for the heating capacitor based on the target power;

[0075] The sending unit 504 is used to send a power control command of the heating capacitor to the semiconductor measurement and testing equipment, so that the semiconductor measurement and testing equipment controls the power of the heating capacitor to the target power based on the power control command, so as to control the temperature of the semiconductor measurement and testing equipment.

[0076] The input unit 502 is specifically used to adjust the current temperature by the temperature PID control algorithm according to the magnitude, duration and / or rate of change of the error, so as to adjust the error and obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm.

[0077] The input unit 502 is specifically used to adjust the error between the current temperature and the target temperature by the incremental PID algorithm, so as to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm; the temperature PID control algorithm is an incremental PID algorithm.

[0078] The temperature control device further includes: an acquisition unit 505; and a calculation unit 506.

[0079] The obtaining unit 505 is used to obtain a target temperature sample; wherein, the target temperature sample is labeled with the second power corresponding to the heating capacitor;

[0080] The input unit 502 is further configured to input the target temperature sample into the deep learning model to obtain the predicted second power corresponding to the target temperature sample output by the deep learning model.

[0081] The computing unit 506 is used to calculate the loss between the predicted second power and the labeled second power according to the regression loss function. When the loss satisfies the convergence condition, the trained deep learning model is obtained.

[0082] The temperature control device further includes: a fine-tuning unit 507;

[0083] The obtaining unit 505 is also used to obtain the control effect of controlling the power of the heating capacitor;

[0084] The fine-tuning unit 507 is used to fine-tune the deep learning model based on the control effect to obtain the fine-tuned deep learning model.

[0085] The generation unit 503 is specifically used to determine the first weight corresponding to the first power and the second weight corresponding to the second power, and to perform a weighted average processing on the first power, the second power, the first weight and the second weight to obtain the target power.

[0086] The input unit 502 is specifically used to execute a first task through a first thread; wherein, the first task is to input the current temperature and the target temperature into a temperature PID control algorithm, and the temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain the task corresponding to the first power of the heating capacitor output by the temperature PID control algorithm; and to execute a second task through a second thread; wherein, the second task is to input the target temperature into a pre-trained deep learning model, and the pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature, and predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a stable state, to obtain the task corresponding to the second power of the heating capacitor output by the pre-trained deep learning model; and the first thread and the second thread are executed in parallel.

[0087] In this embodiment, each unit in the temperature control device performs the functions described above. Figure 2 The operation of the temperature control device in the illustrated embodiment will not be described in detail here.

[0088] Please refer to the following: Figure 6Another embodiment of the temperature control device 600 in this application includes:

[0089] Central processing unit 601, memory 605, input / output interface 604, wired or wireless network interface 603, and power supply 602;

[0090] Memory 605 is either a short-term storage memory or a persistent storage memory;

[0091] The central processing unit 601 is configured to communicate with the memory 605 and execute instructions stored in the memory 605 to perform the aforementioned operations. Figure 2 The method in the illustrated embodiment.

[0092] This application also provides a computer-readable storage medium, which includes instructions that, when executed on a computer, cause the computer to perform the aforementioned actions. Figure 2 The method in the illustrated embodiment.

[0093] This application also provides a computer program product containing instructions, which, when run on a computer, causes the computer to perform the aforementioned... Figure 2 The method in the illustrated embodiment.

[0094] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0095] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0096] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0097] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0098] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0099] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A temperature control method for a semiconductor measurement and testing machine, characterized in that, include: Determine the current temperature and target temperature of the semiconductor measurement and testing equipment; The current temperature and the target temperature are input into a temperature PID control algorithm. The temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required by the heating capacitor when the error is less than or equal to a preset error threshold. The heating capacitor is a component in the circuit of the semiconductor measurement and testing machine. The target temperature is input into a pre-trained deep learning model. Based on the target temperature, the pre-trained deep learning model adjusts the voltage and current of the heating capacitor and predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a steady state, so as to obtain the second power of the heating capacitor output by the pre-trained deep learning model; wherein, the second power represents the power required by the heating capacitor when the temperature of the semiconductor measurement and testing instrument reaches the target temperature; The target power is determined based on the first power and the second power, and a power control command for the heating capacitor is generated based on the target power. The power control command of the heating capacitor is sent to the semiconductor measurement and testing equipment, so that the semiconductor measurement and testing equipment controls the power of the heating capacitor to the target power based on the power control command, so as to control the temperature of the semiconductor measurement and testing equipment.

2. The method according to claim 1, characterized in that, The step of adjusting the error between the current temperature and the target temperature using the temperature PID control algorithm to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm includes: The temperature PID control algorithm adjusts the current temperature based on the magnitude, duration, and / or rate of change of the error to regulate the error and obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm.

3. The method according to claim 1, characterized in that, The temperature PID control algorithm is an incremental PID algorithm; The step of adjusting the error between the current temperature and the target temperature using the temperature PID control algorithm to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm includes: The incremental PID algorithm is used to adjust the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm.

4. The method according to claim 1, characterized in that, Before inputting the target temperature into the pre-trained deep learning model, the method further includes: Obtain a target temperature sample; wherein the target temperature sample is labeled with the second power corresponding to the heating capacitor; The target temperature sample is input into a deep learning model to obtain the predicted second power corresponding to the target temperature sample output by the deep learning model. The loss between the predicted second power and the labeled second power is calculated based on the regression loss function. When the loss satisfies the convergence condition, the trained deep learning model is obtained.

5. The method according to any one of claims 1 to 4, characterized in that, After sending a power control command for the heating capacitor to the semiconductor measurement and testing equipment, so that the semiconductor measurement and testing equipment controls the power of the heating capacitor to the target power based on the power control command to control the temperature of the semiconductor measurement and testing equipment, the method further includes: Achieving a control effect that controls the power of the heating capacitor; The deep learning model is fine-tuned based on the control effect to obtain the fine-tuned deep learning model.

6. The method according to any one of claims 1 to 4, characterized in that, Determining the target power based on the first power and the second power includes: Determine the first weight corresponding to the first power and the second weight corresponding to the second power; The target power is obtained by performing a weighted average of the first power, the second power, the first weight, and the second weight.

7. The method according to any one of claims 1 to 4, characterized in that, The step of inputting the current temperature and the target temperature into a temperature PID control algorithm, and adjusting the error between the current temperature and the target temperature by the temperature PID control algorithm to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm, includes: The first task is to execute the first task through the first thread; wherein the first task is to input the current temperature and the target temperature into the temperature PID control algorithm, and the temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain the task corresponding to the first power of the heating capacitor output by the temperature PID control algorithm. The step of inputting the target temperature into a pre-trained deep learning model, and having the pre-trained deep learning model adjust the voltage and current of the heating capacitor based on the target temperature, predicting the power of the heating capacitor when the voltage and current of the heating capacitor reach a steady state, to obtain the second power of the heating capacitor output by the pre-trained deep learning model, includes: The second task is executed by a second thread; wherein the second task is to input the target temperature into a pre-trained deep learning model, and after the pre-trained deep learning model adjusts the voltage and current of the heating capacitor based on the target temperature, predict the power of the heating capacitor when the voltage and current of the heating capacitor reach a stable state, so as to obtain the task corresponding to the second power of the heating capacitor output by the pre-trained deep learning model; and the first thread and the second thread are executed in parallel.

8. A temperature control device, characterized in that, include: The determination unit is used to determine the current temperature and target temperature of the semiconductor measurement and testing equipment; The input unit is used to input the current temperature and the target temperature into a temperature PID control algorithm. The temperature PID control algorithm adjusts the error between the current temperature and the target temperature to obtain the first power corresponding to the heating capacitor output by the temperature PID control algorithm. The first power represents the power required by the heating capacitor when the error is less than or equal to a preset error threshold. The heating capacitor is a component in the circuit of the semiconductor measurement and testing machine. The input unit is further configured to input the target temperature into a pre-trained deep learning model, which adjusts the voltage and current of the heating capacitor based on the target temperature, and then predicts the power of the heating capacitor when the voltage and current of the heating capacitor reach a stable state, so as to obtain the second power of the heating capacitor output by the pre-trained deep learning model; wherein, the second power represents the power required by the heating capacitor when the temperature of the semiconductor measurement and testing instrument reaches the target temperature; A generation unit is configured to determine a target power based on the first power and the second power, and generate a power control command for the heating capacitor based on the target power; The transmitting unit is used to send a power control command of the heating capacitor to the semiconductor measurement and testing equipment, so that the semiconductor measurement and testing equipment controls the power of the heating capacitor to the target power based on the power control command, so as to control the temperature of the semiconductor measurement and testing equipment.

9. A temperature control device, characterized in that, include: Central processing unit and memory; The memory is either a short-term storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images