Big data evaluation system based on spherical precise ultrasonic honing

Abstract

The invention discloses a big data evaluation system based on spherical precision ultrasonic honing, which relates to the technical field of precision manufacturing, and comprises a big data acquisition module for collecting spherical precision ultrasonic honing processing data from various data sources; the honing quality evaluation module is used for judging whether the honing quality under the current material type and process parameter combination is stable or not; the technological parameter preparation module is used for calculating the relevancy between technological parameters and quality characteristics for the material types and technological parameter combinations with unstable honing quality, and establishing a mapping relation; the process parameter adjusting module is used for selecting one group as a preliminary scheme based on the material type with stable honing quality and the process parameter combination, searching for optimization variables through a mapping relation, and constructing a regression model; and the optimization control module is used for solving through an optimization algorithm to obtain an optimized process parameter combination, and processing is performed by adopting the optimized process parameter combination, so that the stability and controllability of the processing quality are improved.

Description

Technical Field

[0001] The invention relates to the field of precision manufacturing technology, and in particular to a big data evaluation system based on spherical precision ultrasonic honing. Background Art

[0002] Spherical precision ultrasonic honing technology is an important surface treatment technology, which is widely used in the processing of high-precision parts in aerospace, automobile manufacturing, medical equipment, etc. This technology realizes micro-machining of the workpiece surface through the combined action of high-frequency vibration and abrasives, so as to improve the surface quality, shape accuracy and dimensional tolerance. However, the spherical precision ultrasonic honing process involves a variety of material types and complex process parameters, such as material hardness, toughness, ultrasonic frequency, amplitude, honing speed, abrasive type and concentration, etc. These factors are intertwined, making the stability and controllability of processing quality a technical difficulty.

[0003] Traditional spherical precision ultrasonic honing evaluation methods mainly rely on experience judgment and offline detection. These methods are not only time-consuming and labor-intensive, but also difficult to fully and accurately reflect the quality changes in the processing process. With the rapid development of big data technology, it has become possible to use big data to evaluate and optimize the quality of the processing process. However, how to effectively collect, process and analyze the large amount of data generated during the spherical precision ultrasonic honing process, extract key quality features, and formulate and optimize the processing parameters based on these data is still a technical problem that needs to be solved urgently.

[0004] In the existing technology, although there are some data-based processing optimization methods, most of these methods are targeted at specific processing processes or material types, and lack the versatility and adaptability for spherical precision ultrasonic honing technology. In addition, these methods still have many shortcomings in data processing, quality feature extraction, and process parameter optimization, such as incomplete data collection, unreasonable quality feature selection, and inefficient optimization algorithm, which leads to the inability to effectively improve the stability and controllability of processing quality. Summary of the invention

[0005] 1. Technical issues to be resolved In view of the shortcomings of the prior art, the present invention provides a big data evaluation system based on spherical precision ultrasonic honing, which can comprehensively collect and process data in the processing process, accurately extract key quality features, and optimize the processing parameters based on these data to achieve improved stability and controllability of processing quality.

[0006] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: A big data evaluation system based on spherical precision ultrasonic honing, comprising: The big data acquisition module collects spherical surface precision ultrasonic honing processing data from multiple data sources, aggregates them into a spherical surface ultrasonic honing big data set, and divides the processing data into several sub-data sets based on different combinations of material types and process parameters; The honing quality assessment module determines the honing quality assessment indicators and extracts the corresponding quality characteristics, calculates the process mean, standard deviation and process capability index of the quality characteristics under each material type and process parameter combination, and determines whether the honing quality is stable under the current material type and process parameter combination; The process parameter preparation module extracts the corresponding sub-datasets for the material types and process parameter combinations with unstable honing quality, obtains the process capability index of all quality characteristics in the sub-datasets, calculates the correlation between process parameters and quality characteristics after grouping by material type, and establishes a mapping relationship; The process parameter adjustment module selects one of the combinations of material types and process parameters with stable honing quality as a preliminary plan, implements processing and collects data to calculate the process capability index. If the process capability index is less than Cpk Threshold, find the optimization variables through mapping relationship and build regression model; The optimization control module uses the optimization algorithm to solve and obtain the optimized process parameter combination, uses the optimized process parameter combination for processing, and calculates the process capability index of all quality characteristics. Cpk The threshold is compared and corresponding measures are taken according to the comparison results.

[0007] Furthermore, the process mean, standard deviation and process capability index are calculated, including: Get the sub-dataset corresponding to each material type and process parameter combination in the spherical ultrasonic honing big data set, using the formula Calculate the process mean for each quality characteristic in the subdataset, where μ represents the process average, x represents the value of each data point, n represents the number of data points; calculate the process standard deviation for each quality characteristic in the subdataset: ,in, σ represents the process standard deviation, For degrees of freedom.

[0008] Furthermore, using the formula Calculate the process capability index for each quality characteristic in the sub-dataset, where Cpk represents the process capability index, USL represents the upper specification limit, LSL Indicates the lower specification limit.

[0009] Furthermore, it is determined whether the honing quality is stable under the current material type and process parameter combination, including: Pre-set Cpk The process capability index of each quality characteristic is compared with the threshold value. Cpk If the process capability index is greater than or equal to Cpk threshold, it is determined that the honing quality is stable under the current material type and process parameter combination; if the process capability index is less than Cpk If the honing quality is unstable under the current combination of material type and process parameters, the honing quality is judged to be unstable.

[0010] Furthermore, a mapping relationship is established, including: Set the process parameters according to the preliminary plan and carry out processing. Collect various data during the processing, including the actual measured values ​​of all quality characteristics. Calculate the process mean, standard deviation and process capability index of each quality characteristic based on the collected processing data. Compare the calculated process capability index with the preset Cpk If the process capability index of all quality characteristics is greater than or equal to Cpk Threshold, no further optimization is required; if the process capability index of any quality characteristic is less than Cpk Threshold value is optimized.

[0011] Furthermore, for process capability index less than Cpk The quality characteristic of the threshold is calculated by the process capability index of different process parameter combinations of the same material type in the sub-dataset, and the correlation between each process parameter and the quality characteristic is calculated. The calculation formula is as follows: ; in, r Indicates the relevance, x i and y i denote the observed values ​​of process parameters and process capability index, respectively. and Respectively x i and y i The mean of A correlation threshold is preset, and when the correlation between any process parameter and the current quality feature is greater than the correlation threshold, a mapping relationship between the process parameter and the current quality feature is established, otherwise, no operation is performed.

[0012] Furthermore, the optimization variables are found through the mapping relationship, and a regression model is constructed, including: For process capability index less than CpkThe quality characteristics of the threshold value are used to find the process parameters corresponding to the quality characteristics through the established mapping relationship between the process parameters and the quality characteristics. The found process parameters are used as optimization variables, and the process capability index is used as the response variable. Based on the historical processing data of the current equipment, a regression model is constructed: ; in, Cpk represents the process capability index, represents the optimization variable, represents the regression coefficient, is the error term.

[0013] Furthermore, the solution through the optimization algorithm includes: Set constraints and objective function. Objective function: Minimize the coefficient of variation of the process capability index of the quality characteristic. Coefficient of variation: ,in, CV represents the coefficient of variation, σ The process standard deviation represents the quality characteristic, μ Represents the process average of a quality characteristic.

[0014] Furthermore, a genetic algorithm is selected, and initial parameters of the genetic algorithm are set, including population size, iteration number crossover probability, and mutation probability. The regression model is integrated into the genetic algorithm, and the genetic algorithm is run to solve the problem. After several iterations, a set of optimized process parameter combinations is obtained.

[0015] Furthermore, the process capability index is compared with the Cpk threshold, and corresponding measures are taken according to the comparison results, including: If the quality characteristic process capability index is greater than or equal to Cpk threshold, continue to use the optimized process parameter combination for processing; if the quality characteristic process capability index is still less than Cpk threshold, rebuild the regression model and adjust the optimization algorithm.

[0016] (III) Beneficial effects The present invention provides a big data evaluation system based on spherical precision ultrasonic honing, which has the following beneficial effects: (1) By determining clear honing quality evaluation indicators and extracting corresponding quality characteristics, the honing quality under different material types and process parameter combinations can be accurately evaluated. Cpk Threshold value can determine the stability of honing quality under the current combination of material type and process parameters.

[0017] (2) For combinations with unstable honing quality, by calculating the correlation between process parameters and quality characteristics and establishing a mapping relationship, it is possible to reveal which process parameters have a significant impact on the quality characteristics, thereby guiding the adjustment and optimization of process parameters and improving processing efficiency and product quality.

[0018] (3) By formulating a preliminary plan based on the combination of material types and process parameters with stable honing quality and verifying its effect in actual processing, the reliability and stability of the selected plan can be ensured. After the implementation of the preliminary plan, the process capability index is calculated based on the collected processing data, and potential quality problems can be discovered in time. For products with a process capability index less than Cpk The quality characteristics of the threshold can quickly locate the corresponding process parameters through mapping relationships and build a regression model, which helps to avoid blindly adjusting process parameters and thus improve processing efficiency.

[0019] (4) By setting the objective function to minimize the coefficient of variation of the process capability index of the quality characteristic, the quality characteristics with a process capability index below the threshold can be optimized, thereby reducing fluctuations in the processing process and improving the stability of the honing quality. With the help of regression models and optimization algorithms, the key factors affecting the quality characteristics can be accurately found and optimized to improve processing efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the structure of the big data evaluation system based on spherical precision ultrasonic honing of the present invention. DETAILED DESCRIPTION

[0021] 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.

[0022] See also Figure 1 The present invention provides a big data evaluation system based on spherical precision ultrasonic honing, comprising: a big data acquisition module, a honing quality evaluation module, a process parameter preparation module, a process parameter adjustment module and an optimization control module; wherein, The big data acquisition module collects spherical surface precision ultrasonic honing processing data from multiple data sources, aggregates them into a spherical surface ultrasonic honing big data set, and divides the processing data into several sub-data sets based on different combinations of material types and process parameters; Collect processing data of spherical precision ultrasonic honing through multiple data sources, including but not limited to sensor data at the production site, equipment control system records, laboratory test results, historical production records, and public databases, etc., to ensure that the data covers different material types (such as metals, ceramics, composite materials, etc.) and a combination of multiple process parameters (such as ultrasonic frequency, amplitude, honing speed, abrasive type and concentration, etc.); Conduct preliminary screening of the collected processing data to eliminate abnormal or invalid data caused by equipment failure, operating errors or data recording errors. At the same time, check the integrity and consistency of the data to ensure that each record contains the necessary processing parameters and result information; Preprocess missing values ​​or inconsistent formats in the data. For missing values, use interpolation methods (such as linear interpolation, polynomial interpolation, etc.) to fill them according to the trend of adjacent data points, or use statistical methods (such as mean and median substitution) to estimate according to data distribution characteristics; For data with inconsistent formats, standardize them to ensure that all data items meet unified format requirements; The preprocessed processing data are aggregated to form a large data set of spherical ultrasonic honing. According to the different combinations of material types and process parameters (such as ultrasonic frequency, amplitude, honing speed, etc.), the large data set of spherical ultrasonic honing is divided into several sub-data sets. The corresponding material type and process parameter combination are identified for each sub-data set. For example, the stainless steel-low frequency, high amplitude and low speed honing data set: Material type: stainless steel (such as 304 stainless steel), process parameter combination: Ultrasonic frequency: 15 kHz (Low frequency), Amplitude: 50 μm (High amplitude), honing speed: 100 rpm (Low speed); Titanium alloy - high frequency low amplitude high speed honing data set: Material type: Titanium alloy (such as Ti-6Al-4V ), process parameter combination: ultrasonic frequency: 25 kHz (high frequency), amplitude: 20 μm (Low amplitude), Honing speed: 300 rpm (high speed); By collecting data from a variety of data sources, including experimental records, data directly output by production equipment, historical project databases, and public databases, the comprehensiveness and diversity of the data are ensured. It not only covers spherical precision ultrasonic honing data of different material types, but also includes a variety of process parameter combinations, providing a rich data foundation for subsequent evaluation and analysis.

[0023] The honing quality assessment module determines the honing quality assessment indicators and extracts the corresponding quality characteristics, calculates the process mean, standard deviation and process capability index of the quality characteristics under each material type and process parameter combination, and determines whether the honing quality is stable under the current material type and process parameter combination; Determine honing quality evaluation indicators, including but not limited to surface roughness, shape accuracy, dimensional tolerance, material removal rate, and surface integrity, extract quality features corresponding to the honing quality evaluation indicators from the spherical ultrasonic honing big data set, and determine the upper and lower specifications for each quality feature; It should be noted that the upper limit specification is the maximum value allowed for a certain quality feature (such as surface roughness) after spherical surface processing, and the lower limit specification is the minimum value allowed for a certain quality feature after spherical surface processing. These specifications are set based on actual needs, industry standards or internal specifications. Among them, surface roughness: measures the smoothness of the surface after processing, shape accuracy: evaluates the deviation between the shape after processing and the ideal shape, dimensional tolerance: checks whether the size after processing is within the allowable range, material removal rate: evaluates the material removal efficiency during the processing process, and surface integrity: considers whether there are defects such as cracks and scratches on the surface after processing; Get the sub-dataset corresponding to each material type and process parameter combination in the spherical ultrasonic honing big data set, using the formula Calculate the process mean for each quality characteristic in the subdataset, where μ represents the process mean, x represents the value of each data point, n represents the number of data points; calculate the process standard deviation for each quality characteristic in the subdataset: ,in, σ represents the process standard deviation, is the degree of freedom; Using the formula Calculate the process capability index for each quality characteristic in the sub-dataset, where Cpk represents the process capability index, USL represents the upper specification limit, LSL Indicates the lower specification limit; Pre-set Cpk The process capability index of each quality characteristic is compared with the threshold value. Cpk If the process capability index is greater than or equal to Cpk Threshold, indicating that the quality feature has high stability and capability in the processing under the current material type and process parameter combination, then it is determined that the honing quality is stable under the current material type and process parameter combination; if the process capability index is less than Cpk Threshold, it means that the quality feature may be unstable or incapable in the machining process under the current material type and process parameter combination, and it is determined that the honing quality is unstable under the current material type and process parameter combination; It should be noted that CpkA value greater than 1.33 is generally considered a qualified standard, indicating that the process capability is good, there is a smaller deviation and variability, the product or service has a higher pass rate within the specification requirements, and the risk is lower. Cpk The higher the value, the stronger the process capability and the more stable the product quality; By determining clear honing quality evaluation indicators and extracting corresponding quality features, the honing quality under different material types and process parameter combinations can be accurately evaluated. Cpk Threshold value can determine the stability of honing quality under the current combination of material type and process parameters.

[0024] The process parameter preparation module extracts the corresponding sub-datasets for the material types and process parameter combinations with unstable honing quality, obtains the process capability index of all quality characteristics in the sub-datasets, calculates the correlation between process parameters and quality characteristics after grouping by material type, and establishes a mapping relationship; For the material type and process parameter combination that is determined to have unstable honing quality, the corresponding sub-dataset is extracted to obtain the process capability index of all quality characteristics in the sub-dataset; The extracted sub-datasets are grouped by material type, and for process capability index less than Cpk The quality characteristics of the threshold are calculated by the process capability index of different process parameter combinations of the same material type in the sub-dataset to calculate the correlation between the process parameters and the quality characteristics. The calculation formula is as follows: ; in, r Indicates the relevance, x i and y i denote the observed values ​​of process parameters and process capability index, respectively. and Respectively x i and y i The mean of It should be noted that, usually, the value of the correlation is between -1 and 1, close to 1 indicates positive correlation, close to -1 indicates negative correlation, and close to 0 indicates no correlation. After the above formula is processed by taking the absolute value, the correlation close to 1 indicates correlation, and close to 0 indicates no correlation; A correlation threshold is set in advance. When the correlation between the process parameters and the quality characteristics is greater than the correlation threshold, it indicates that there is a correlation between the process parameters and the quality characteristics, and a mapping relationship between the process parameters and the quality characteristics is established. When the correlation between the process parameters and the quality characteristics is less than or equal to the correlation threshold, it indicates that there is no correlation between the process parameters and the quality characteristics, and no operation is performed. It should be noted that the setting of the correlation threshold is used to determine whether there is a significant correlation between the process parameters and the quality characteristics. The setting of the correlation threshold is based on statistical principles and practical experience to ensure that when the correlation is greater than the correlation threshold, it is considered that there is a significant correlation between the process parameters and the quality characteristics. For combinations with unstable honing quality, by calculating the correlation between process parameters and quality characteristics and establishing a mapping relationship, it is possible to reveal which process parameters have a significant impact on quality characteristics, thereby guiding the adjustment and optimization of process parameters and improving processing efficiency and product quality.

[0025] The process parameter adjustment module selects one of the combinations of material types and process parameters with stable honing quality as a preliminary solution, implements processing and collects data to calculate the process capability index. If the process capability index is less than the Cpk threshold, the optimization variables are found through the mapping relationship and a regression model is constructed. Based on the material type and process parameter combination that are judged to have stable honing quality, one or more groups are selected as preliminary solutions, which include specific material type, process parameters (such as honing time, honing pressure, ultrasonic amplitude, ultrasonic frequency, etc.) and expected processing quality characteristics (such as surface roughness, roundness, etc.); In the actual spherical precision honing process, the process parameters in the preliminary plan are set and processed, and various data in the processing process are collected, including the actual measured values ​​of all quality characteristics. Based on the collected processing data, the process mean, standard deviation and process capability index of each quality characteristic are calculated, and the calculated process capability index is compared with the preset Cpk Threshold comparison: If the process capability index of all quality characteristics is greater than or equal to Cpk threshold, the honing quality under the current preliminary scheme is considered to be stable and no further optimization is required; if the process capability index of any quality characteristic is less than Cpk If the threshold is exceeded, it is considered that the honing quality under the current preliminary scheme is unstable and needs to be optimized and adjusted; For process capability index less than Cpk The quality characteristics of the threshold value are used to find the process parameters corresponding to the quality characteristics through the established mapping relationship between the process parameters and the quality characteristics. The found process parameters are used as optimization variables, and the process capability index is used as the response variable. Based on the historical processing data of the current equipment, a regression model is constructed: ; in, Cpk represents the process capability index, represents the optimization variable, represents the regression coefficient, εThe error term, the regression coefficient calculation: use the least squares method to solve the regression coefficient, that is, find a set of β value, so that the sum of squares of errors between the predicted value and the actual value is minimized; By formulating a preliminary plan based on the combination of material types and process parameters with stable honing quality and verifying its effect in actual processing, the reliability and stability of the selected plan can be ensured. After the implementation of the preliminary plan, the process capability index is calculated based on the collected processing data, and potential quality problems can be discovered in time. For products with a process capability index less than Cpk The quality characteristics of the threshold can quickly locate the corresponding process parameters through mapping relationships and build a regression model, which helps to avoid blindly adjusting process parameters and thus improve processing efficiency.

[0026] The optimization control module uses the optimization algorithm to solve and obtain the optimized process parameter combination, uses the optimized process parameter combination for processing, and calculates the process capability index of all quality characteristics. Cpk The threshold is compared and corresponding measures are taken according to the comparison results.

[0027] Set constraints and objective function. Objective function: Minimize the coefficient of variation of the process capability index of the quality characteristic. Coefficient of variation: ,in, CV represents the coefficient of variation, σ The process standard deviation represents the quality characteristic, μ A process average that represents a quality characteristic; Set constraints for optimization variables, such as honing time T exist [T_min, T_max] Range, honing pressure P exist [P_min, P_max] Within the range, ultrasonic amplitude A exist [A_min, A_max] In the range, ultrasonic frequency F exist [F_min, F_ max] within the scope; Select an optimization algorithm, such as genetic algorithm, particle swarm algorithm, simulated annealing algorithm, etc. Set the initial parameters of the algorithm, including population size, number of iterations, crossover probability, and mutation probability. For example, the population size is 50, the number of iterations is 100, the crossover probability is 0.8, and the mutation probability is 0.01. Integrate the regression model into the optimization algorithm and run the optimization algorithm to solve. After several iterations, obtain a set of optimized process parameter combinations, such as honing time. T_opt , Honing pressure P_opt , ultrasonic amplitude A_opt and ultrasonic frequency F_opt ; Use the optimized process parameter combination for processing, and calculate the process mean, standard deviation and process capability index of the quality characteristics. If the process capability index of the quality characteristics is greater than or equal to Cpk threshold, continue to use the optimized process parameter combination for processing; if the quality characteristic process capability index is still less than Cpk Threshold, rebuild the regression model and adjust the optimization algorithm, for example, adjust the population size to 100 and the number of iterations to 150; By setting the objective function to minimize the coefficient of variation of the process capability index of the quality characteristic, the quality characteristics with process capability index below the threshold can be optimized, thereby reducing fluctuations in the processing process and improving the stability of the honing quality. With the help of regression models and optimization algorithms, the key factors affecting the quality characteristics can be accurately found and optimized to improve processing efficiency and product quality.

[0028] The present invention also provides a big data evaluation method based on spherical precision ultrasonic honing, comprising the following steps: Step 1: Collect spherical surface precision ultrasonic honing processing data from multiple data sources, aggregate them to form a large data set of spherical surface ultrasonic honing, and divide the processing data into several sub-data sets based on different combinations of material types and process parameters; Step 2: Determine the honing quality evaluation index and extract the corresponding quality characteristics, calculate the process mean, standard deviation and process capability index of the quality characteristics under each material type and process parameter combination, and determine whether the honing quality is stable under the current material type and process parameter combination; Step 3: For the material type and process parameter combination with unstable honing quality, extract the corresponding sub-dataset, obtain the process capability index of all quality characteristics in the sub-dataset, group by material type, calculate the correlation between process parameters and quality characteristics, and establish a mapping relationship; Step 4: Based on the material type and process parameter combination with stable honing quality, select one group as the preliminary plan, implement processing and collect data to calculate the process capability index. If the process capability index is less than Cpk Threshold, find the optimization variables through mapping relationship and build regression model; Step 5: Solve through the optimization algorithm to obtain the optimized process parameter combination, use the optimized process parameter combination for processing, and calculate the process capability index of all quality characteristics. Cpk The threshold is compared and corresponding measures are taken according to the comparison results.

[0029] In the application, the several formulas involved are all calculated by removing dimensions and taking their numerical values. The formula is a formula of the most recent real situation obtained by collecting a large amount of data and performing software simulation. The coefficients in the formula are set by technicians in this field according to actual conditions.

[0030] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in combination with electronic hardware, computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0031] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0032] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A big data evaluation system based on spherical precision ultrasonic honing, characterized by: include: The big data acquisition module collects the spherical surface precision ultrasonic honing processing data and summarizes them into a spherical surface ultrasonic honing big data set. Based on different combinations of material types and process parameters, the processing data is divided into several sub-data sets; The honing quality assessment module determines the honing quality assessment indicators and extracts the corresponding quality characteristics, calculates the process mean, standard deviation and process capability index of the quality characteristics under each material type and process parameter combination, and determines whether the honing quality is stable under the current material type and process parameter combination; The process parameter preparation module extracts the corresponding sub-datasets for the material types and process parameter combinations with unstable honing quality, obtains the process capability index of all quality characteristics in the sub-datasets, calculates the correlation between process parameters and quality characteristics after grouping by material type, and establishes a mapping relationship; The process parameter adjustment module selects any group as the preliminary plan based on the material type and process parameter combination with stable honing quality, implements processing and collects data to calculate the process capability index. If the process capability index is less than Cpk Threshold, find the optimization variables through mapping relationship and build regression model; The optimization control module uses the optimization algorithm to solve and obtain the optimized process parameter combination, uses the optimized process parameter combination for processing, and calculates the process capability index of all quality characteristics. Cpk The threshold is compared and corresponding measures are taken according to the comparison results.

2. The big data evaluation system based on spherical precision ultrasonic honing according to claim 1 is characterized in that: Calculate process mean, standard deviation, and process capability indices, including: Get the sub-dataset corresponding to each material type and process parameter combination in the spherical ultrasonic honing big data set, using the formula Calculate the process mean for each quality characteristic in the subdataset, where represents the process average, x represents the value of each data point, n represents the number of data points; calculate the process standard deviation for each quality characteristic in the subdataset: ,in, σ represents the process standard deviation, For degrees of freedom.

3. The big data evaluation system based on spherical precision ultrasonic honing according to claim 2 is characterized in that: Using the formula Calculate the process capability index for each quality characteristic in the sub-dataset, where Cpk represents the process capability index, USL represents the upper specification limit, LSL Indicates the lower specification limit.

4. The big data evaluation system based on spherical precision ultrasonic honing according to claim 3 is characterized in that: Determine whether the honing quality is stable under the current material type and process parameter combination, including: Pre-set Cpk The process capability index of each quality characteristic is compared with the threshold value. Cpk If the process capability index is greater than or equal to Cpk threshold, it is determined that the honing quality is stable under the current material type and process parameter combination; if the process capability index is less than Cpk If the honing quality is unstable under the current combination of material type and process parameters, the honing quality is judged to be unstable.

5. The big data evaluation system based on spherical precision ultrasonic honing according to claim 1 is characterized in that: Establish a mapping relationship, including: Set the process parameters according to the preliminary plan and carry out processing. Collect various data during the processing, including the actual measured values ​​of all quality characteristics. Calculate the process mean, standard deviation and process capability index of each quality characteristic based on the collected processing data. Compare the calculated process capability index with the preset Cpk If the process capability index of all quality characteristics is greater than or equal to Cpk Threshold, no further optimization is required; if the process capability index of any quality characteristic is less than Cpk Threshold value is optimized.

6. The big data evaluation system based on spherical precision ultrasonic honing according to claim 5 is characterized in that: For process capability index less than Cpk The quality characteristic of the threshold is calculated by the process capability index of different process parameter combinations of the same material type in the sub-dataset, and the correlation between each process parameter and the quality characteristic is calculated. The calculation formula is as follows: ； in, r Indicates the relevance, x i and y i denote the observed values ​​of process parameters and process capability index, respectively. and Respectively x i and y i The mean of A correlation threshold is preset, and when the correlation between any process parameter and the current quality feature is greater than the correlation threshold, a mapping relationship between the process parameter and the current quality feature is established, otherwise, no operation is performed.

7. The big data evaluation system based on spherical precision ultrasonic honing according to claim 1 is characterized in that: Find optimization variables through mapping relationships and build regression models, including: For process capability index less than Cpk The quality characteristics of the threshold value are used to find the process parameters corresponding to the quality characteristics through the established mapping relationship between the process parameters and the quality characteristics. The found process parameters are used as optimization variables, and the process capability index is used as the response variable. Based on the historical processing data of the current equipment, a regression model is constructed: ； in, Cpk represents the process capability index, represents the optimization variable, represents the regression coefficient, is the error term.

8. The big data evaluation system based on spherical precision ultrasonic honing according to claim 1 is characterized in that: Solving through optimization algorithms includes: Set constraints and objective function. Objective function: Minimize the coefficient of variation of the process capability index of the quality characteristic. Coefficient of variation: ,in, CV represents the coefficient of variation, σ The process standard deviation represents the quality characteristic, μ Represents the process average of a quality characteristic.

9. The big data evaluation system based on spherical precision ultrasonic honing according to claim 8, characterized in that: Select a genetic algorithm, set the initial parameters of the genetic algorithm, including population size, iteration number, crossover probability, and mutation probability, integrate the regression model into the genetic algorithm, run the genetic algorithm to solve, and after several iterations, obtain a set of optimized process parameter combinations.

10. The big data evaluation system based on spherical precision ultrasonic honing according to claim 9, characterized in that: The process capability index and Cpk The threshold is compared and corresponding measures are taken according to the comparison results, including: If the quality characteristic process capability index is greater than or equal to Cpk threshold, continue to use the optimized process parameter combination for processing; if the quality characteristic process capability index is still less than Cpk threshold, rebuild the regression model and adjust the optimization algorithm.

Images