Digital data management system and method based on semiconductor electroplating equipment
Through neural network model, the expected yield rate and equipment adaptability of semiconductor electroplating equipment are predicted, which solves the problem of unreasonable equipment selection in the existing technology, improves the adaptability of equipment utilization and task completion, and realizes scientific equipment and Task matching.
Patent Information
- Application Number
- CN202411251368.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-07
- Publication Date
- 2024-11-26
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The lack of in-depth consideration of task requirements in the selection of existing semiconductor electroplating equipment, resulting in low equipment utilization, and the inability to comprehensively evaluate task volume and equipment errors. The equipment evaluation is only dependent on yield rates, and the appropriate equipment cannot be accurately selected.
Using a digital data management system based on neural networks, the training model predicts the expected yield rate of the current task by collecting historical electroplating task data, and selects appropriate equipment based on task similarity and equipment adaptation, and sets the yield rate deviation range for evaluation. Implement adaptive matching between devices and tasks.
It improves the scientificity and rationality of equipment selection, improves the adaptability of task completion and the accuracy of yield rate evaluation, and realizes the comprehensive analysis and judgment of equipment tasks.
Smart Images

Figure CN119026872A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data analysis, and in particular to a digital data management system and method based on semiconductor electroplating equipment. Background Art
[0002] The current chip market competition is increasingly fierce, and semiconductor electroplating is an extremely critical part of the industry's production and manufacturing process. The selection of different semiconductor electroplating equipment is also extremely important; In the current selection of semiconductor electroplating equipment, most of them are manually selected based on the actual working conditions of the equipment and the accuracy requirements of the workpiece, lacking in-depth consideration of the task requirements. When selecting equipment, it is also impossible to make reasonable electroplating equipment selection based on historical electroplating task data, and the utilization rate of historical task data of each equipment is low; in the actual equipment task evaluation stage, the equipment task is evaluated solely through the yield rate threshold, and it is impossible to make a comprehensive evaluation and judgment on the impact of the task volume and the reasonable error of the equipment itself; Therefore, a digital data management system and method based on semiconductor electroplating equipment is needed to solve the above problems. Summary of the invention
[0003] The object of the present invention is to provide a digital data management system and method based on semiconductor electroplating equipment to solve the problems raised in the above background technology.
[0004] To achieve the above object, the present invention provides the following technical solutions: The digital data management method based on semiconductor electroplating equipment includes the following analysis steps: Step S100: Collecting the working parameter data of the electroplating equipment in the historical electroplating tasks, and synchronously collecting the electroplating yield rate in each task; Step S200: Use the electroplating equipment working parameter data and the corresponding electroplating yield rate data in the historical electroplating tasks to train the neural network model respectively; Step S300: obtaining the working parameters of the equipment in the current electroplating task, and using the neural network model corresponding to each electroplating equipment to analyze the current electroplating task and select the expected yield rate of each electroplating equipment; Step S400: Filter out all electroplating equipment that can complete the current electroplating task according to the functions of each electroplating equipment as the electroplating equipment to be selected, filter out the electroplating task data whose electroplating equipment selection results are within the range of the electroplating equipment to be selected in the historical electroplating tasks, analyze the electroplating task similarity comparison parameters, and calculate the similarity between the current electroplating task and each electroplating task , and calculate the equipment evaluation priority based on the results of the electroplating equipment completing historical electroplating tasks , and then obtain the historical electroplating task data of each candidate electroplating equipment, and calculate the compatibility between the current electroplating task and each candidate electroplating equipment , select the candidate equipment with the largest fitness value as the electroplating equipment selection result for the current electroplating task; Step S500: Obtain the actual yield data of the current electroplating task, compare and analyze it with the expected yield data, and evaluate the yield of the selected electroplating equipment task.
[0005] In the above technical solution, step S100 includes the following contents: Collect and obtain the electroplating equipment and working parameter data selected in the historical tasks. The working parameter data is divided into equipment working parameter data and working environment parameter data. Among them, the equipment working parameter data refers to the electroplating process parameters, electroplating liquid parameters and electroplating equipment status parameters of the electroplating equipment when performing the electroplating task, and the environmental parameter data refers to the equipment environment data of the electroplating equipment when performing the electroplating task. By analyzing and screening the equipment and task data involved in the historical electroplating tasks, the connection between the data is fully utilized to analyze the influence of different working parameters of each equipment on the workpiece yield when performing the electroplating task. In the above technical solution, step S200 includes the following contents: The working parameter data of electroplating equipment and the corresponding electroplating yield rate data in the historical electroplating tasks are divided into training set and validation set in proportion, and the loss function is selected: ;in, is the model loss function, is the sample size of the training set, The data in the current electroplating equipment training sample The number of workpieces, is the sum of the number of workpieces in all data in the current electroplating equipment training sample, is the sample number of the training set, is the actual yield rate of the training set samples, The model outputs the estimated yield rate; the proportion of the number of workpieces in each sample data is introduced into the loss function, and a weighted analysis is performed on each sample data to highlight the impact of historical task data with a large number of workpieces on the model parameter estimation results, and to weaken the model parameter accuracy deviation caused by the large yield rate error caused by the small number of workpieces in small historical electroplating tasks; In the above technical solution, step S400 includes the following contents: Calculate the similarity between the current electroplating task and each historical electroplating task , for any historical electroplating task, the calculation formula is as follows: ; in, is the task similarity comparison parameter number, is the number of task similarity comparison parameters, is the task similarity comparison parameter weight coefficient, Compare parameter data for historical electroplating task similarity, Compare parameter data for the current electroplating task similarity; Calculate the equipment evaluation priority based on the results of the electroplating equipment completing historical electroplating tasks , the calculation formula is as follows: ; in, It is the preferred task number among the historical electroplating tasks. is the optimal number of tasks, For the preferred task number of workpieces, The total number of workpieces that have completed historical electroplating tasks for the electroplating equipment. For the preferred task Actual yield rate, For the preferred task Estimated yield rate; the setting of the preferred task is determined based on the yield rate evaluation results of each historical electroplating task when it is completed; Calculate the compatibility between the current electroplating task and each candidate electroplating equipment , for any selected electroplating equipment, the calculation formula is as follows: ; in, To determine the basic compatibility between electroplating tasks and electroplating equipment, To evaluate the optimality of electroplating equipment, The number of the historical electroplating task completed by the current electroplating equipment. The total number of historical electroplating tasks completed by the current electroplating equipment. For current electroplating tasks and historical electroplating tasks Similarity, For historical plating tasks Actual yield rate, For historical plating tasks Yield preset standard, For historical plating tasks number of workpieces, The total number of all workpieces that have completed historical electroplating tasks for the electroplating equipment.
[0006] In the above technical solution, step S500 includes the following contents: Obtain the actual yield rate and expected yield rate data of the selected electroplating equipment's historical tasks, and calculate and set the yield rate deviation range when the selected electroplating equipment performs the electroplating task; Calculate the weighted standard deviation of the current electroplating equipment task yield rate deviation , according to the formula: ; in, The number of the historical electroplating tasks completed by the electroplating equipment. The total number of historical plating tasks completed for the plating equipment, For historical plating tasks number of workpieces, The total number of workpieces that have completed historical electroplating tasks for the electroplating equipment. For historical plating tasks Actual yield rate, For historical plating tasks Expected yield rate, It is the average of the difference between the actual yield rate and the expected yield rate in all historical electroplating tasks of the electroplating equipment; Then, the deviation range of the yield rate of the current electroplating equipment when performing the current electroplating task is set to ;in, The expected yield rate for the current electroplating equipment when performing the current electroplating task; By calculating the weighted standard deviation of the yield rate deviation of each electroplating equipment, the deviation range of the task yield rate is set; the number of workpieces of each task completed by the electroplating equipment and the deviation between the actual yield rate and the expected yield rate are introduced into the calculation, and the reasonable deviation between the actual completion status of the equipment and the expected completion status under different task requirements is comprehensively considered and analyzed, and then the actual completion status of the current task is judged and analyzed and abnormal yield rate feedback is given; If the actual yield rate of the current electroplating task exceeds the yield rate deviation range, the selected electroplating equipment is judged to have failed to complete the task yield rate, and the actual yield rate and the expected yield rate data are recorded and the equipment inspection signal is fed back; if the actual yield rate of the current electroplating equipment task is within the yield rate deviation range, the selected electroplating equipment is judged to have passed the task yield rate, and the actual yield rate and the expected yield rate data are recorded, and no signal is fed back; Set the yield rate optimization threshold , the yield rate deviation range is divided into the optimal degree; if the actual yield rate of the current electroplating task is within the yield rate deviation range The yield evaluation result of the current electroplating task is judged to be excellent, and the current task is set as the preferred task when participating in the equipment selection of subsequent electroplating tasks.
[0007] A digital data management system based on semiconductor electroplating equipment using the digital data management method based on semiconductor electroplating equipment in the above technical solution includes a data analysis module, an electroplating equipment selection module, and an electroplating task evaluation module; The data analysis module is used to train the neural network model based on the working parameter data of the electroplating equipment in the historical electroplating tasks and the data on the completed yield of the electroplating tasks, and then predict the expected yield when selecting each electroplating equipment for the current electroplating task; the electroplating equipment selection module is used to analyze the similarity between the current electroplating task and the historical tasks, calculate the fitness of each electroplating equipment to be selected, and then select the electroplating equipment with the highest fitness for the current electroplating task; the electroplating task evaluation module is used to set the yield deviation range based on the historical task yield data of the selected electroplating equipment, and evaluate and feedback the completed yield of the current electroplating task; The output end of the data analysis module is connected to the input end of the electroplating equipment selection module and the electroplating task evaluation module; In the above technical solution, the data analysis module includes a data acquisition unit, a model training unit, and a yield rate prediction unit; The data acquisition unit is used to collect the working parameter data of the electroplating equipment in the historical electroplating tasks and the workpiece yield data in each task; the model training unit uses the collected data to train the yield prediction neural network model of each device; the yield prediction unit uses the current electroplating task working parameters to analyze the expected yield of each device when completing the current electroplating task according to the yield prediction neural network model of each device; The output end of the data acquisition unit is connected to the input end of the model training unit and the electroplating task evaluation module; the output end of the model training unit is connected to the input end of the yield rate prediction unit; the output end of the yield rate prediction unit is connected to the input end of the electroplating equipment selection module and the electroplating task evaluation module; In the above technical solution, the electroplating equipment selection module includes a candidate equipment analysis unit, a similarity calculation unit, a matching degree calculation unit, and an electroplating equipment selection unit; The candidate equipment analysis unit selects electroplating equipment that can complete the task as the candidate equipment according to the current task requirements; the similarity calculation unit is used to calculate the similarity between the current electroplating task and the historical electroplating tasks; the matching calculation unit is used to calculate and analyze the matching degree between the current electroplating task and each candidate equipment according to the historical electroplating task data of each electroplating equipment; the electroplating equipment selection unit is used to select equipment according to the calculation result of the matching degree between the current electroplating task and each candidate equipment; The output ends of the equipment analysis unit to be selected and the similarity calculation unit are connected to the input end of the matching calculation unit; the input end of the matching calculation unit is connected to the output end of the data analysis module; the output end of the matching calculation unit is connected to the input end of the electroplating equipment selection unit; In the above technical solution, the electroplating task evaluation unit includes a yield rate deviation range setting unit and an electroplating task yield rate evaluation unit; The yield rate deviation range setting unit is used to calculate the weighted standard deviation of the yield rate deviation according to the historical electroplating task data of the selected electroplating equipment, and then set the yield rate deviation range; the electroplating task yield rate evaluation unit evaluates and feeds back the completion result of the current electroplating task according to the yield rate deviation range; The input end of the yield rate deviation range setting unit is connected to the data analysis module; the output end of the yield rate deviation range setting unit is connected to the input end of the electroplating task yield rate control unit.
[0008] Compared with the prior art, the beneficial effects achieved by the present invention are: In the present invention, by analyzing the historical electroplating task data, the similarity between the current electroplating task and all the completed tasks of the candidate equipment is analyzed, and then the matching degree between the current electroplating task and each candidate equipment is calculated, and the task equipment is selected according to the matching result. By making full use of the historical electroplating task data, the intrinsic relationship between the parameters involved in each task and the evaluation results after the task is completed is analyzed, and a unified standard is provided to comprehensively judge the matching degree between the task and each electroplating equipment, so as to provide a reasonable and scientific explanation for the adaptation between the task and the equipment, and realize the adaptive selection and allocation of the task and the equipment. In the task evaluation stage, by analyzing the evaluation results of all historical electroplating tasks completed by the selected equipment, the deviation range of the yield rate of the current electroplating task is defined, so as to realize the comprehensive analysis and judgment of the equipment task completion results. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 It is a flow chart of the digital data management method based on semiconductor electroplating equipment of the present invention; Figure 2 A diagram showing the structural connections of a digital data management system based on semiconductor electroplating equipment according to the present invention. DETAILED DESCRIPTION
[0010] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0011] Example: See Figure 1-Figure 2 , the present invention provides a technical solution, like Figure 1 As shown, the present invention provides a digital data management method based on semiconductor electroplating equipment, comprising the following analysis steps: Step S100: Collecting the working parameter data of the electroplating equipment in the historical electroplating tasks, and synchronously collecting the electroplating yield rate in each task; Step S200: Use the electroplating equipment working parameter data and the corresponding electroplating yield rate data in the historical electroplating tasks to train the neural network model respectively; Step S300: obtaining the working parameters of the equipment in the current electroplating task, and using the neural network model corresponding to each electroplating equipment to analyze the current electroplating task and select the expected yield rate of each electroplating equipment; Step S400: Obtain historical electroplating task data, equipment selection results and yield rate, calculate the compatibility between the current electroplating task and each electroplating equipment, and select the electroplating equipment for the current electroplating task; Step S500: Acquire the actual yield rate data of the current electroplating task, compare and analyze it with the expected yield rate data, and evaluate the yield rate of the selected electroplating equipment task; Step S100 includes the following contents: Collect and obtain the electroplating equipment and working parameter data selected in the historical tasks. The working parameter data is divided into equipment working parameter data and working environment parameter data. Among them, the equipment working parameter data is the electroplating process parameters, electroplating solution parameters and electroplating equipment status parameters of the electroplating equipment when performing the electroplating task, and the environmental parameter data is the equipment environment data of the electroplating equipment when performing the electroplating task; In actual electroplating tasks, electroplating process parameters generally include current density, electrode spacing, and electroplating time, while electroplating solution parameters include electroplating solution pH, temperature, and concentration. Electroplating equipment status parameters refer to various parameters set during the operation of the electroplating equipment, including stirring speed, operating load, etc. Environmental parameter data refers to the temperature and humidity of the environment when the electroplating equipment is performing electroplating tasks. Step S200 includes the following contents: The working parameter data of electroplating equipment and the corresponding electroplating yield rate data in the historical electroplating tasks are divided into training set and validation set in proportion, and the loss function is selected: ;in, is the model loss function, is the sample size of the training set, The data in the current electroplating equipment training sample The number of workpieces, is the sum of the number of workpieces in all data in the current electroplating equipment training sample, is the sample number of the training set, is the actual yield rate of the training set samples, Output the expected yield rate for the model; When selecting a model, the problem can be transformed into a multivariate regression problem, and a convolutional neural network can be selected to analyze the data association between the selected parameters and the workpiece yield rate, and then the yield rate can be predicted using the trained model using the parameter data of the current electroplating task; Step S400 includes the following contents: Calculate the similarity between the current electroplating task and each historical electroplating task , for any historical electroplating task, the calculation formula is as follows: ; in, is the task similarity comparison parameter number, is the number of task similarity comparison parameters, is the task similarity comparison parameter weight coefficient, Compare parameter data for historical electroplating task similarity, Compare parameter data for the current electroplating task similarity; When actually analyzing the similarity between the current electroplating task and the historical electroplating task, the selectable similarity comparison parameters include the type of electroplated metal cations, the number of workpieces, the size of the workpiece, the number of parallel electroplating, the electroplating speed, the thickness of the coating, etc. The selected parameters are quantified and weight coefficients are assigned according to the influence of the selected parameters on the actual workpiece yield rate, and then the similarity between the electroplating tasks is calculated; wherein, the greater the influence of the selected parameters on the actual workpiece yield rate, the greater the weight coefficient assigned; Calculate the equipment evaluation priority based on the results of the electroplating equipment completing historical electroplating tasks , the calculation formula is as follows: ; in, It is the preferred task number among the historical electroplating tasks. is the optimal number of tasks, For the preferred task number of workpieces, The total number of workpieces that have completed historical electroplating tasks for the electroplating equipment. For the preferred task Actual yield rate, For the preferred task Estimated yield rate; the setting of the preferred task is determined based on the yield rate evaluation results of each historical electroplating task when it is completed; Calculate the compatibility between the current electroplating task and each candidate electroplating equipment , for any selected electroplating equipment, the calculation formula is as follows: ; in, To determine the basic compatibility between electroplating tasks and electroplating equipment, To evaluate the optimality of electroplating equipment, The number of the historical electroplating task completed by the current electroplating equipment. The total number of historical electroplating tasks completed by the current electroplating equipment. For current electroplating tasks and historical electroplating tasks Similarity, For historical plating tasks Actual yield rate, For historical plating tasks Yield preset standard, For historical plating tasks number of workpieces, The total number of workpieces that have completed historical electroplating tasks for the electroplating equipment; The present invention provides the following example for calculating the suitability of electroplating equipment: Set the basic suitability of the electroplating task and the electroplating equipment to , the default yield rate standard for all historical electroplating tasks is , assuming that a certain electroplating equipment has completed electroplating tasks, and the similarity between each electroplating task and the current electroplating task is , and the corresponding task workpiece numbers are , the actual yield rate of the corresponding task is , the corresponding task expected yield data are , where the first historical electroplating task is the optimal task, and the optimality of the electroplating equipment evaluation is calculated as , the matching degree between the current electroplating task and a certain electroplating equipment can be calculated according to the formula: ; Calculate the compatibility between the current electroplating task and all the candidate electroplating equipment, and select the candidate equipment with the largest compatibility value as the electroplating equipment selection result for the current electroplating task; Step S500 includes the following contents: Obtain the actual yield rate and expected yield rate data of the selected electroplating equipment's historical tasks, and calculate and set the yield rate deviation range when the selected electroplating equipment performs the electroplating task; Calculate the weighted standard deviation of the current electroplating equipment task yield rate deviation , according to the formula: ; in, The number of the historical electroplating tasks completed by the electroplating equipment. The total number of historical plating tasks completed for the plating equipment, For historical plating tasks number of workpieces, The total number of workpieces that have completed historical electroplating tasks for the electroplating equipment. For historical plating tasks Actual yield rate, For historical plating tasks Expected yield rate, It is the average of the difference between the actual yield rate and the expected yield rate in all historical electroplating tasks of the electroplating equipment; Then, the deviation range of the yield rate of the current electroplating equipment when performing the current electroplating task is set to ;in, The expected yield rate for the current electroplating equipment when performing the current electroplating task; The present invention sets the yield rate deviation range by using the historical task data of the electroplating equipment in the above electroplating equipment fitness calculation case, and the weighted standard deviation of the current task yield rate deviation can be calculated as , assuming that the current electroplating task is expected to have a yield rate of , the current electroplating task yield rate deviation range can be set to ; If the actual yield rate of the current electroplating task exceeds the yield rate deviation range, the selected electroplating equipment is judged to have failed to complete the task yield rate, and the actual yield rate and the expected yield rate data are recorded and the equipment inspection signal is fed back; if the actual yield rate of the current electroplating equipment task is within the yield rate deviation range, the selected electroplating equipment is judged to have passed the task yield rate, and the actual yield rate and the expected yield rate data are recorded, and no signal is fed back; Set the yield rate optimization threshold , the yield rate deviation range is divided into the optimal degree; if the actual yield rate of the current electroplating task is within the yield rate deviation range The yield rate evaluation result of the current electroplating task is judged to be excellent, and the current task is set as the preferred task when participating in the subsequent electroplating task equipment selection; in the above use case, the preferred yield rate threshold is set according to the formula as , and then determine the first historical task as the preferred task.
[0012] like Figure 2 As shown, the present invention also provides a digital data management system based on semiconductor electroplating equipment, including a data analysis module, an electroplating equipment selection module, and an electroplating task evaluation module; the output end of the data analysis module is connected to the input end of the electroplating equipment selection module and the electroplating task evaluation module; The data analysis module is used to train the neural network model based on the working parameter data of the electroplating equipment in the historical electroplating tasks and the data on the completed yield of the electroplating tasks, and then predict the expected yield when selecting each electroplating equipment for the current electroplating task; the electroplating equipment selection module is used to analyze the similarity between the current electroplating task and the historical tasks, calculate the fitness of each electroplating equipment to be selected, and then select the electroplating equipment with the highest fitness for the current electroplating task; the electroplating task evaluation module is used to set the yield deviation range based on the historical task yield data of the selected electroplating equipment, and evaluate and feedback the completed yield of the current electroplating task; The data analysis module includes a data acquisition unit, a model training unit, and a yield rate prediction unit; the output end of the data acquisition unit is connected to the input end of the model training unit and the electroplating task evaluation module; the output end of the model training unit is connected to the input end of the yield rate prediction unit; the output end of the yield rate prediction unit is connected to the input end of the electroplating equipment selection module and the electroplating task evaluation module; The data acquisition unit is used to collect the working parameter data of the electroplating equipment in the historical electroplating tasks and the workpiece yield data in each task; the model training unit uses the collected data to train the yield prediction neural network model of each device; the yield prediction unit uses the current electroplating task working parameters to analyze the expected yield of each device when completing the current electroplating task according to the yield prediction neural network model of each device; The electroplating equipment selection module includes a candidate equipment analysis unit, a similarity calculation unit, a matching calculation unit, and an electroplating equipment selection unit; the output ends of the candidate equipment analysis unit and the similarity calculation unit are connected to the input end of the matching calculation unit; the input end of the matching calculation unit is connected to the output end of the data analysis module; the output end of the matching calculation unit is connected to the input end of the electroplating equipment selection unit; The candidate equipment analysis unit selects electroplating equipment that can complete the task as the candidate equipment according to the current task requirements; the similarity calculation unit is used to calculate the similarity between the current electroplating task and the historical electroplating tasks; the matching calculation unit is used to calculate and analyze the matching degree between the current electroplating task and each candidate equipment according to the historical electroplating task data of each electroplating equipment; the electroplating equipment selection unit is used to select equipment according to the calculation result of the matching degree between the current electroplating task and each candidate equipment; The electroplating task evaluation unit includes a yield rate deviation range setting unit and an electroplating task yield rate evaluation unit; the input end of the yield rate deviation range setting unit is connected to the data analysis module; the output end of the yield rate deviation range setting unit is connected to the input end of the electroplating task yield rate evaluation unit; The yield rate deviation range setting unit is used to calculate the weighted standard deviation of the yield rate deviation based on the historical electroplating task data of the selected electroplating equipment, and then set the yield rate deviation range; the electroplating task yield rate evaluation unit evaluates and feedbacks the completion result of the current electroplating task based on the yield rate deviation range.
[0013] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
Claims
1. Digital data management method based on semiconductor electroplating equipment, It is characterized in that The analysis steps include: Step S100: Collecting the working parameter data of the electroplating equipment in the historical electroplating tasks, and synchronously collecting the electroplating yield rate in each task; Step S200: Use the electroplating equipment working parameter data and the corresponding electroplating yield rate data in the historical electroplating tasks to train the neural network model respectively; Step S300: obtaining the working parameters of the equipment in the current electroplating task, and using the neural network model corresponding to each electroplating equipment to analyze the current electroplating task and select the expected yield rate of each electroplating equipment; Step S400: Filter out all electroplating equipment that can complete the current electroplating task according to the functions of each electroplating equipment as the electroplating equipment to be selected, filter out the electroplating task data whose electroplating equipment selection results are within the range of the electroplating equipment to be selected in the historical electroplating tasks, analyze the electroplating task similarity comparison parameters, and calculate the similarity between the current electroplating task and each electroplating task , and calculate the equipment evaluation priority based on the results of the electroplating equipment completing historical electroplating tasks , and then obtain the historical electroplating task data of each candidate electroplating equipment, and calculate the compatibility between the current electroplating task and each candidate electroplating equipment , select the candidate equipment with the largest fitness value as the electroplating equipment selection result for the current electroplating task; Step S500: Obtain the actual yield data of the current electroplating task, compare and analyze it with the expected yield data, and evaluate the yield of the selected electroplating equipment task.
2. The digital data management method based on semiconductor electroplating equipment according to claim 1, It is characterized in that Step S100 includes the following contents: Collect and obtain the electroplating equipment and working parameter data selected in the historical tasks. The working parameter data is divided into equipment working parameter data and working environment parameter data; among them, the equipment working parameter data refers to the electroplating process parameters, electroplating liquid parameters and electroplating equipment status parameters of the electroplating equipment when performing electroplating tasks, and the environmental parameter data refers to the equipment environment data of the electroplating equipment when performing electroplating tasks.
3. The digital data management method based on semiconductor electroplating equipment according to claim 1, It is characterized in that Step S200 includes the following contents: The working parameter data of electroplating equipment and the corresponding electroplating yield rate data in the historical electroplating tasks are divided into training set and validation set in proportion, and the loss function is selected: ; in, is the model loss function, is the sample size of the training set, The data in the current electroplating equipment training sample The number of workpieces, is the sum of the number of workpieces in all data in the current electroplating equipment training sample, is the sample number of the training set, is the actual yield rate of the training set samples, Output the expected yield for the model.
4. The digital data management method based on semiconductor electroplating equipment according to claim 1, It is characterized in that Step S400 includes the following contents: Calculate the similarity between the current electroplating task and each historical electroplating task , for any historical electroplating task, the calculation formula is as follows: ; in, is the task similarity comparison parameter number, is the number of task similarity comparison parameters, is the task similarity comparison parameter weight coefficient, Compare parameter data for historical electroplating task similarity, Compare parameter data for the current electroplating task similarity; Calculate the equipment evaluation priority based on the results of the electroplating equipment completing historical electroplating tasks , the calculation formula is as follows: ; in, It is the preferred task number among the historical electroplating tasks. is the optimal number of tasks, For the preferred task number of workpieces, The total number of workpieces that have completed historical electroplating tasks for the electroplating equipment. For the preferred task Actual yield rate, For the preferred task Estimated yield rate; the setting of the preferred task is determined based on the yield rate evaluation results of each historical electroplating task when it is completed; Calculate the compatibility between the current electroplating task and each candidate electroplating equipment , for any selected electroplating equipment, the calculation formula is as follows: ; in, To determine the basic compatibility between electroplating tasks and electroplating equipment, To evaluate the optimality of electroplating equipment, The number of the historical electroplating task completed by the current electroplating equipment. The total number of historical electroplating tasks completed by the current electroplating equipment. For current electroplating tasks and historical electroplating tasks Similarity, For historical plating tasks Actual yield rate, For historical plating tasks Yield preset standard, For historical plating tasks number of workpieces, The total number of all workpieces that have completed historical electroplating tasks for the electroplating equipment.
5. The digital data management method based on semiconductor electroplating equipment according to claim 1, It is characterized in that Step S500 includes the following contents: Obtain the actual yield rate and expected yield rate data of the selected electroplating equipment's historical tasks, and calculate and set the yield rate deviation range when the selected electroplating equipment performs the electroplating task; Calculate the weighted standard deviation of the current electroplating equipment task yield rate deviation , according to the formula: ; in, The number of the historical electroplating tasks completed by the electroplating equipment. The total number of historical plating tasks completed for the plating equipment, For historical plating tasks number of workpieces, The total number of workpieces that have completed historical electroplating tasks for the electroplating equipment. For historical plating tasks Actual yield rate, For historical plating tasks Expected yield rate, It is the average of the difference between the actual yield rate and the expected yield rate in all historical electroplating tasks of the electroplating equipment; Then, the deviation range of the yield rate of the current electroplating equipment when performing the current electroplating task is set to ;in, The expected yield rate for the current electroplating equipment when performing the current electroplating task; If the actual yield rate of the current electroplating task exceeds the yield rate deviation range, the selected electroplating equipment is judged to have failed to complete the task yield rate, and the actual yield rate and expected yield rate data are recorded and the equipment inspection signal is fed back; if the actual yield rate of the current electroplating equipment task is within the yield rate deviation range, the selected electroplating equipment is judged to have passed the task yield rate, and the actual yield rate and expected yield rate data are recorded, and no signal is fed back; Set the yield rate optimization threshold , divide the yield rate deviation range into optimal levels; if the actual yield rate of the current electroplating task is within the yield rate deviation range The yield evaluation result of the current electroplating task is judged to be excellent, and the current task is set as the preferred task when participating in the equipment selection of subsequent electroplating tasks.
6. A digital data management system based on semiconductor electroplating equipment using the digital data management method based on semiconductor electroplating equipment according to any one of claims 1 to 5, Features: The system includes a data analysis module, an electroplating equipment selection module, and an electroplating task evaluation module; The data analysis module is used to train a neural network model based on the working parameter data of the electroplating equipment in the historical electroplating tasks and the finished product yield data of the electroplating tasks, and then predict the expected yield when selecting each electroplating equipment for the current electroplating task; The electroplating equipment selection module is used to analyze the similarity between the current electroplating task and the historical tasks, calculate the fitness of each selected electroplating equipment, and then select the electroplating equipment with the highest fitness for the current electroplating task; the electroplating task evaluation module is used to set the yield rate deviation range according to the historical task yield rate data of the selected electroplating equipment, and evaluate and feedback the completion yield rate of the current electroplating task; The output end of the data analysis module is connected to the input end of the electroplating equipment selection module and the electroplating task evaluation module.
7. According to the digital data management system based on semiconductor electroplating equipment described in claim 6, Features: The data analysis module includes a data acquisition unit, a model training unit, and a yield rate prediction unit; The data acquisition unit is used to collect the working parameter data of the electroplating equipment in the historical electroplating tasks and the workpiece yield data in each task; the model training unit uses the collected data to train the yield prediction neural network model of each device; the yield prediction unit uses the working parameters of the current electroplating task to analyze the expected yield of each device when completing the current electroplating task according to the yield prediction neural network model of each device; The output end of the data acquisition unit is connected to the input end of the model training unit and the electroplating task evaluation module; the output end of the model training unit is connected to the input end of the yield prediction unit; the output end of the yield prediction unit is connected to the electroplating equipment selection module and the input end of the electroplating task evaluation module.
8. According to the digital data management system based on semiconductor electroplating equipment described in claim 6, Features: The electroplating equipment selection module includes a candidate equipment analysis unit, a similarity calculation unit, a matching degree calculation unit, and an electroplating equipment selection unit; The candidate equipment analysis unit selects electroplating equipment that can complete the task as the candidate equipment according to the current task requirements; The similarity calculation unit is used to calculate the similarity between the current electroplating task and the historical electroplating tasks; The matching degree calculation unit is used to calculate and analyze the matching degree between the current electroplating task and each selected equipment according to the historical electroplating task data of each electroplating equipment; the electroplating equipment selection unit is used to select equipment according to the calculation result of the matching degree between the current electroplating task and each selected equipment; The output ends of the candidate device analysis unit and the similarity calculation unit are connected to the input end of the matching degree calculation unit; The matching degree calculation unit input end is connected to the data analysis module output end; The output end of the matching degree calculation unit is connected to the input end of the electroplating equipment selection unit.
9. According to the digital data management system based on semiconductor electroplating equipment described in claim 6, Features: The electroplating task evaluation unit includes a yield rate deviation range setting unit and an electroplating task yield rate evaluation unit; The yield rate deviation range setting unit is used to calculate the weighted standard deviation of the yield rate deviation according to the historical electroplating task data of the selected electroplating equipment, and then set the yield rate deviation range; the electroplating task yield rate evaluation unit evaluates and feeds back the completion result of the current electroplating task according to the yield rate deviation range; The input end of the yield rate deviation range setting unit is connected to the data analysis module; the output end of the yield rate deviation range setting unit is connected to the input end of the electroplating task yield rate control unit.
