Grinding fluid performance optimization method and system based on data analysis

Abstract

The invention relates to the technical field of grinding fluid, in particular to a grinding fluid performance optimization method and system based on data analysis, and the method comprises the following steps: based on collected grinding fluid data, identifying key parameters influencing the performance of the grinding fluid, carrying out time sequence analysis on cutting speed, temperature, viscosity index and pH value data, determining key change nodes of the performance, and calculating the performance of the grinding fluid according to the key change nodes. And performing reanalysis on the node data to obtain a performance key node identifier. According to the method, by applying time sequence analysis and a real-time feedback mechanism, the adjusting process of the grinding fluid is remarkably optimized, key parameters such as the cutting speed, the temperature, the viscosity index and the pH value in the grinding process are allowed to be continuously monitored and analyzed, performance change nodes are accurately recognized, and the grinding efficiency is improved. The proportion and chemical components of the grinding fluid can be adjusted according to real-time data, so that the optimal performance of the grinding fluid is ensured, the cutting efficiency and the machining quality of a workpiece are enhanced, and the performance of the grinding fluid is kept in the optimal state by continuously optimizing the proportion.

Description

Technical Field

[0001] This invention relates to the field of polishing slurry technology, and in particular to a method and system for optimizing polishing slurry performance based on data analysis. Background Technology

[0002] Polishing fluid technology involves liquid media used in physical or chemical machining processes to improve the efficiency and effectiveness of grinding, polishing, or other surface treatments. Polishing fluids not only act as coolants to reduce thermal damage but also as lubricants to reduce friction between the workpiece and the polishing medium, improving machining accuracy and extending tool life. Furthermore, polishing fluids help remove debris generated during machining, keeping the machined surface clean and thus improving machining quality.

[0003] Among them, the data analysis-based grinding fluid performance optimization method refers to the use of data analysis technology to collect, analyze and evaluate the effect of grinding fluid, thereby identifying performance bottlenecks and proposing improvement measures. By collecting data such as temperature, pressure and grinding speed during the grinding process, the relationship between data and grinding effect is analyzed using algorithm models to find the optimal grinding fluid usage parameters, reduce operating costs and make the grinding process more efficient.

[0004] Traditional techniques often rely on preset parameters for slurry preparation, lacking the ability to adapt to changes in cutting conditions. When dealing with complex or uncommon machining environments, they cannot provide necessary adjustments, resulting in suboptimal grinding efficiency and machining quality. For example, unequally adjusted slurry can cause workpiece overheating or wear after prolonged use, affecting product precision and surface quality. Furthermore, fixed ratios cannot be optimized based on actual machining feedback, leading to low efficiency of the slurry in practical use and increasing production costs and maintenance difficulties. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a data analysis-based method and system for optimizing the performance of grinding fluids.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a data analysis-based method for optimizing the performance of grinding slurries, comprising the following steps:

[0007] S1: Based on the collected grinding fluid data, identify the key parameters affecting the performance of the grinding fluid, perform time series analysis on the cutting speed, temperature, viscosity index and pH value data to determine the key change nodes of performance, and re-analyze the node data to obtain the key performance node identifiers.

[0008] S2: Based on the performance key node identifier, mine the nonlinear relationship in the data, including the influence of additives on the performance index of the polishing fluid, identify the chemical components that have the greatest impact on the performance of the polishing fluid, and generate component influence mapping information.

[0009] S3: Based on the component influence mapping information, dynamically adjust the temperature stability and the proportion of chemical additives, optimize the grinding fluid ratio by combining real-time cutting feedback, monitor the adjusted ratio in real time, analyze the impact on cutting efficiency and quality, and obtain the ratio effect evaluation results.

[0010] S4: Based on the evaluation results of the ratio effect, optimize the grinding fluid parameter settings again, refine the chemical component ratio, and verify whether the grinding fluid performance meets the predetermined standard, including experimental verification and data calibration, to obtain the optimal performance configuration of the grinding fluid.

[0011] The present invention is improved in that the identification step of the key parameter is specifically as follows:

[0012] S111: Based on the collected grinding fluid data, including cutting speed, temperature, viscosity index and pH value, data cleaning and outlier processing are performed to obtain the processed performance dataset;

[0013] S112: Based on the processed performance dataset, and considering the impact of each parameter fluctuation on performance, the following formula is used:

[0014]

[0015] Calculate the weighted correlation coefficient r weighted The parameters associated with the performance fluctuations of the grinding slurry were identified, and a preliminary list of key parameters was obtained, including x. i y represents the value of the independent variable at the i-th monitoring point. i This represents the dependent variable value of the i-th monitoring point. and For the average value, w i Let i be the weight of the i-th monitoring point;

[0016] S113: Based on the preliminary list of key parameters, conduct experimental verification, adjust individual parameters and monitor their impact on the performance of the polishing slurry, and determine the key parameters affecting the performance of the polishing slurry.

[0017] The present invention is improved in that the step of obtaining the performance key node identifier is specifically as follows:

[0018] S121: Perform time series analysis on the cutting speed, temperature, viscosity index and pH value data, use the ARMA model to smooth the data and highlight long-term trends, and identify potential change nodes;

[0019] S122: Based on the potential change nodes, identify trend changes or fluctuation enhancement points through Bai-Perron breakpoint detection to obtain preliminary key change nodes;

[0020] S123: Reanalyze the initial key change nodes, compare the performance data differences before and after the key nodes, and verify the effectiveness of the nodes and their impact on performance to obtain performance key node identifiers.

[0021] The present invention is improved in that the specific steps for mining the nonlinear relationship are as follows:

[0022] S211: Based on the aforementioned performance key node identifiers, remove data points that do not meet the standards, explore the relationship between additive concentration and performance indicators, and use the following formula:

[0023]

[0024] Determine the additive concentration level, where RS represents a measure of non-linear relationship, and CS i It is the additive concentration at the i-th data point, CS avg is the average concentration of the additive, CS0 is the median concentration, and N is the total number of data points;

[0025] S212: Calculate the performance changes corresponding to various additive concentration levels, compare and verify statistical significance, and obtain the nonlinear relationship analysis results.

[0026] The present invention is improved in that the step of obtaining the component influence mapping information is specifically as follows:

[0027] S221: Based on the aforementioned nonlinear relationship, analyze the data between the additive components and the performance indicators of the grinding fluid, using the following formula:

[0028]

[0029] The chemical components with the greatest impact were identified, and the chemical component impact index IY was obtained, where Py i It is the performance metric value corresponding to the i-th data point, Py avg It is the average of the performance metrics, Py std It is the standard deviation of the performance index, wy i The weights are adjusted based on the concentration of chemical components, and N is the total number of data points;

[0030] S222: Based on the chemical component influence index, quantify the influence of each key chemical component to obtain component influence mapping information.

[0031] The present invention is improved in that the step of optimizing the grinding fluid ratio is specifically as follows:

[0032] S311: Based on the component influence mapping information, monitor the temperature data and additive reaction during the cutting process in real time, and adjust the proportion of chemical additives using the following formula:

[0033] ORnew =OR old ×(1+k O ×(OT target -OT actual ))

[0034] The adjusted additive ratio OR new Among them, OR old It is the original additive ratio, OT target It is the target temperature stability value, OT. actual It is the currently measured temperature, k. O It is an adjustment factor;

[0035] S312: Based on the adjusted additive ratio, the grinding fluid ratio is adjusted in combination with real-time feedback data to match different cutting conditions and obtain the grinding fluid ratio configuration.

[0036] The present invention is improved in that the steps for obtaining the evaluation results of the formulation effect are specifically as follows:

[0037] S321: Conduct cutting experiments, collect key performance data in real time, and obtain experimental monitoring datasets;

[0038] S322: Based on the aforementioned experimental monitoring dataset, the following formula is used:

[0039]

[0040] The deviation of performance in each experiment was quantified to obtain preliminary analysis results of the formulation, where EH represents the evaluation score of the formulation effect, and XH... i μ is the performance index of the i-th experiment. h It is the average value of the performance index, σ h It is the standard deviation, N H It is the number of experiments;

[0041] S323: Based on the preliminary analysis results of the ratio, evaluate the performance of the new ratio under different cutting conditions, determine the effective ratio parameters, and obtain the ratio effect evaluation results.

[0042] The present invention is improved in that the step of obtaining the optimal performance configuration of the polishing slurry is specifically as follows:

[0043] S411: Based on the evaluation results of the formulation effect, further optimize the grinding fluid parameter settings and refine the chemical component ratio;

[0044] S412: Based on the stated chemical composition ratio, test whether the performance of the grinding slurry meets the predetermined standard, using the formula:

[0045]

[0046] The effects of the mixing ratio were evaluated to obtain performance verification results, where VX represents the variance of the grinding fluid performance deviation, and PX... actual,i This is the current performance parameter of the polishing slurry in the i-th test, PX target It is the target performance parameter, n b It is the number of tests;

[0047] S413: Based on the performance verification results, analyze and correct the collected data, determine whether the polishing fluid has reached the optimal state, and obtain the optimal performance configuration of the polishing fluid.

[0048] A data analysis-based system for optimizing the performance of grinding slurries, the system comprising:

[0049] The key parameter analysis module, based on the collected grinding fluid data, collects the performance indicators and cutting condition data of the grinding fluid, performs time series analysis on cutting speed, temperature, viscosity index and pH value, identifies the key change nodes affecting performance, and re-analyzes the node data to obtain the key performance node identifiers.

[0050] Based on the performance key node identifiers, the data relationship mining module mines nonlinear relationships in the data, analyzes the impact of additives on the performance indicators of the grinding fluid, identifies key chemical components that affect the performance of the grinding fluid, and generates component impact mapping information.

[0051] The formulation optimization module utilizes the component influence mapping information to dynamically adjust the temperature stability and chemical additive ratio, and combines it with real-time cutting feedback to optimize the grinding fluid formulation and obtain dynamic formulation adjustment information.

[0052] Based on the dynamic ratio adjustment information, the ratio monitoring module monitors the adjusted grinding fluid ratio in real time, analyzes its impact on cutting efficiency and quality, and generates a ratio effect evaluation result.

[0053] Based on the evaluation results of the ratio effect, the performance verification module further optimizes the grinding fluid parameter settings, refines the chemical component ratio, and verifies whether the performance of the grinding fluid meets the predetermined standard through experimental verification and data calibration, thereby obtaining the optimal performance configuration of the grinding fluid.

[0054] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0055] In this invention, by applying time series analysis and a real-time feedback mechanism, the adjustment process of the grinding slurry is significantly optimized. This allows for continuous monitoring and analysis of key parameters during the grinding process, such as cutting speed, temperature, viscosity index, and pH value. It accurately identifies nodes of performance change, enabling the grinding slurry ratio and chemical composition to be adjusted based on real-time data. This ensures optimal performance of the grinding slurry, enhances cutting efficiency and workpiece processing quality, and maintains the grinding slurry performance at its best through continuous optimization of the ratio. This effectively reduces resource waste and improves the quality of the cutting process. Attached Figure Description

[0056] Figure 1 The flowchart of the data analysis-based method for optimizing the performance of grinding fluid is provided in this invention.

[0057] Figure 2 This is a flowchart illustrating the identification of key parameters in this invention.

[0058] Figure 3 This is a flowchart illustrating the process of obtaining the performance-critical node identifier in this invention.

[0059] Figure 4 This is a flowchart of the nonlinear relationship mining process in this invention;

[0060] Figure 5 This is a flowchart illustrating the process of obtaining component influence mapping information in this invention.

[0061] Figure 6 This is a flowchart illustrating the optimization of the grinding slurry ratio in this invention;

[0062] Figure 7 This is a flowchart illustrating the process of obtaining the evaluation results of the formulation effect in this invention.

[0063] Figure 8 This is a flowchart illustrating the process of obtaining the optimal performance configuration of the grinding fluid in this invention. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0065] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0066] Example

[0067] Please see Figure 1 This invention provides a technical solution: a method for optimizing the performance of grinding slurry based on data analysis, comprising the following steps:

[0068] S1: Based on the collected grinding fluid data, including performance index data and cutting condition data, identify the key parameters affecting the performance of the grinding fluid, perform time series analysis on cutting speed, temperature, viscosity index and pH value data to determine the key change nodes of performance, and re-analyze the node data to obtain the key performance node identifiers.

[0069] S2: Based on the identification of key performance nodes, nonlinear relationships in the data are mined, including the impact of additives on the performance indicators of the grinding fluid, identifying the chemical components that have the greatest impact on the performance of the grinding fluid, and generating component impact mapping information; the main components of the grinding fluid include base oil (which can be mineral oil, synthetic oil or vegetable oil, used to provide lubrication and cooling effects), additives: anti-wear additives, extreme pressure additives, rust inhibitors and antioxidants, etc., emulsifiers (to enable the oil and water in the grinding fluid to form a stable emulsion, improving cooling and cleaning performance), corrosion inhibitors, etc.

[0070] S3: Based on the component influence mapping information, the temperature stability and the proportion of chemical additives are dynamically adjusted. The grinding fluid ratio is optimized by combining real-time cutting feedback. The adjusted ratio is monitored in real time to analyze its impact on cutting efficiency and quality, and the ratio effect evaluation results are obtained.

[0071] S4: Based on the evaluation results of the mixing effect, optimize the grinding fluid parameter settings again, refine the chemical component ratio, and verify whether the grinding fluid performance meets the predetermined standard, including experimental verification and data calibration, to obtain the optimal performance configuration of the grinding fluid.

[0072] Key performance nodes include key time points, main influencing parameters, and parameter variability. Component influence mapping information specifically includes main influencing components, component sensitivity, and component interactions. The ratio effect evaluation results include ratio adjustment efficiency, ratio accuracy, and ratio stability. The optimal performance configuration of the grinding fluid specifically refers to the optimal ratio of chemical components, performance verification pass rate, and configuration application range.

[0073] Please see Figure 2 The specific steps for identifying key parameters are as follows:

[0074] S111: Based on the collected grinding fluid data, including cutting speed, temperature, viscosity index and pH value, data cleaning and outlier processing are performed to obtain the processed performance dataset;

[0075] Based on the collected grinding fluid data, including cutting speed, temperature, viscosity index, and pH value, which are obtained in real time from various grinding machines through sensors, the data is automatically cleaned and outlier processed. The cleaning process includes filtering out noisy data, calibrating equipment errors, and eliminating readings that deviate significantly from the normal operating range. For example, data points with abnormal temperatures below 0 degrees or above 100 degrees are discarded. The resulting performance dataset reflects the machine's performance status under normal operating conditions, providing accurate basic data for further analysis.

[0076] S112: Based on the processed performance dataset, and considering the impact of each parameter fluctuation on performance, the following formula is used:

[0077]

[0078] Calculate the weighted correlation coefficient r weighted The parameters associated with the performance fluctuations of the grinding slurry were identified, and a preliminary list of key parameters was obtained, including x. i y represents the value of an independent variable at the i-th monitoring point, such as cutting speed, temperature, viscosity index, or pH value. i The value of the dependent variable at the i-th monitoring point represents the corresponding performance index of the grinding fluid. and For the average value, w i The weight of the i-th monitoring point can be assigned different weights based on the importance or stability of the monitoring point.

[0079] Within a specific monitoring period, the following parameter values ​​were collected: cutting speed x i =150rpm, temperature y i =35 degrees, average cutting speed Average temperature Degree, weight w i =0.5, substitute into the formula to calculate:

[0080] The results show that cutting speed and temperature have a very strong positive correlation under a given weight, which means that these two parameters are closely related to the performance of the grinding fluid.

[0081] S113: Based on the preliminary list of key parameters, conduct experimental verification, adjust individual parameters and monitor their impact on the performance of the polishing slurry, and determine the key parameters that affect the performance of the polishing slurry.

[0082] Based on the preliminary list of key parameters, during the experimental verification process, the experimental design of independently varying each parameter meticulously recorded the specific impact of parameter adjustments on the performance of the grinding fluid. For example, adjusting the cutting speed from 100 rpm to 200 rpm and monitoring changes in temperature, viscosity index, and pH value were conducted. After recording the data, statistical methods were used to analyze the trends of change, identifying which parameter changes were directly related to performance improvement or reduction. This ensured that every parameter considered critical was supported by actual data, confirming the key parameters affecting the performance of the grinding fluid and providing a basis for optimizing the grinding fluid formulation and improving mechanical operating parameters.

[0083] Please see Figure 3 The specific steps for obtaining the performance critical node identifier are as follows:

[0084] S121: Perform time series analysis on cutting speed, temperature, viscosity index and pH value data, use ARMA model to smooth the data and highlight long-term trends, and identify potential change nodes;

[0085] Time series analysis was performed on cutting speed, temperature, viscosity index, and pH data. First, each set of data was smoothed by setting a time window to reduce the impact of short-term fluctuations. For example, the moving average method was used to calculate the local average value using 3 to 5 consecutive points to generate smoothed data for each set of parameters. Then, an autoregressive moving average (ARMA) model was used to construct a trend analysis model for the time series. By modeling the trend and seasonal changes of historical data and performing error analysis on the model residuals, it was ensured that the model could effectively represent the long-term trend and periodic characteristics of the data. Finally, potential change nodes were identified based on the change points in the trend analysis.

[0086] S122: Based on potential change nodes, Bai-Perron breakpoint detection is used to identify trend changes or points of increased volatility, thus obtaining preliminary key change nodes;

[0087] Based on potential change nodes, Bai-Perron breakpoint detection is used for analysis. First, the weighted mean and variance of each data segment are calculated, and the trend of mean change in each data segment is detected. By setting an initial threshold, significant change points in each segment are analyzed. A dynamic adjustment mechanism for breakpoint segmentation is introduced based on breakpoint detection. Breakpoints are optimized and adjusted according to data fluctuations and historical change trends. Significant abrupt changes and trend enhancement points in the time series are analyzed to identify clear trend changes or fluctuation enhancement points, thus obtaining preliminary key change nodes.

[0088] S123: Reanalyze the initial key change nodes, compare the performance data differences before and after the key nodes, and verify the effectiveness of the nodes and their impact on performance to obtain performance key node identifiers.

[0089] The initial key change nodes are reanalyzed. The time series data before and after the nodes are segmented and the differences are compared. By comparing the rate of change of parameters before and after the nodes and their specific impact on performance, statistical methods are used to calculate the impact of key change nodes on the overall trend. At the same time, the behavior of data near the key nodes is simulated by experiments to verify whether the direction of change of the key nodes is significantly related to the performance impact. Based on the change curve of the experimental data and statistical difference analysis, the identifiers of performance key nodes are determined.

[0090] Please see Figure 4 The specific steps for mining nonlinear relationships are as follows:

[0091] S211: Based on the identification of key performance nodes, remove data points that do not meet the standards, explore the relationship between additive concentration and performance indicators, and use the following formula:

[0092]

[0093] Determine the additive concentration level, where RS represents a measure of non-linear relationship, and CS i It is the additive concentration at the i-th data point, CS avg is the average concentration of the additive, CS0 is the median concentration, and N is the total number of data points;

[0094] The additive concentration data points are [10, 20, 30, 40, 50], and the average value is CS. avg Given a median CS0 of 30 and a total of 5 data points N, calculate the expression for each concentration point, such as CS. i When = 10, then:

[0095]

[0096] When CS i =20:

[0097] When CS i =30:

[0098]

[0099] When CS i =40:

[0100]

[0101] When CS i =50:

[0102]

[0103] Find the sum:

[0104] RS=RS1+RS2+RS3+RS4+RS5=400+100+0+100+400=1000

[0105] The results indicate that there is a significant nonlinear relationship between additive concentration and performance, with lower or higher concentrations having a greater impact on performance.

[0106] S212: Calculate the performance changes corresponding to various additive concentration levels, compare and verify statistical significance, and obtain the results of nonlinear relationship analysis;

[0107] After confirming the relationship between additive concentration and the performance indicators of the grinding fluid, a quantitative analysis of the relationship between concentration level and performance changes was conducted. First, multiple concentration ranges were set, such as the percentile range from the lowest 5% to the highest 95%. Each range included specific additive concentration values. Performance indicators were calculated for the data points in each range, such as the mean, median, and coefficient of variation of viscosity and pH value. Then, analysis of variance and t-tests were performed to determine whether the performance changes at different additive concentration levels were statistically significant. The nonlinear relationship analysis results were obtained, which included details of the performance changes corresponding to each concentration level and an evaluation of their statistical significance, helping to understand more deeply how additive concentration affects the performance of the grinding fluid.

[0108] Please see Figure 5 The specific steps for obtaining component influence mapping information are as follows:

[0109] S221: Based on nonlinear relationships, the data between additive components and grinding fluid performance indicators are analyzed using the following formula:

[0110]

[0111] The most influential chemical components are identified, and the chemical component influence index IY is obtained. Py is used to quantify the impact of each component on performance.i It is the performance metric value corresponding to the i-th data point, Py avg It is the average of the performance metrics, Py std It is the standard deviation of the performance index, used to standardize the magnitude of the impact. i The weights are adjusted based on the concentration of chemical components, with higher concentrations receiving greater weights. N is the total number of data points.

[0112] The following data was collected, with performance metrics set to Py. i There are 5 data points: [10, 20, 30, 40, 50]; the average value is Py. avg The standard deviation is 30, and the standard deviation is Py. std The value is 12.9, and the weight is wy. i Given the values ​​[0.1, 0.2, 0.3, 0.2, 0.2], and a total of 5 data points N, calculate the expression for each data point, such as in Python. i =10 o'clock:

[0113]

[0114] When Py i =20:

[0115]

[0116] When Py i =30:

[0117]

[0118] When Py i =40:

[0119] When Py i =50:

[0120]

[0121] Calculate the total IY:

[0122] IY=0.24+0.12+0+0.12+0.48≈0.96

[0123] The results show that calculations can clearly identify the specific impact of each chemical component on performance, thus enabling effective component adjustments.

[0124] S222: Based on the chemical composition influence index, the influence of each key chemical component is quantified, showing how the components specifically affect the performance of the grinding fluid, such as viscosity and thermal stability, and obtaining component influence mapping information;

[0125] Based on the chemical composition influence index, the key chemical components of the grinding fluid are first identified. Relevant chemical parameters are extracted through experimental measurements or existing research results. The components are ranked according to key indicators, and the core components that directly affect properties such as viscosity and thermal stability are screened out. Then, the influence of the core components is quantified one by one. The concentration or ratio of each component is measured item by item, and the changing trend of its effect on the performance of the grinding fluid is recorded. Specifically, different concentrations of the target component are added to the reference grinding fluid, stirred evenly, and the dynamic viscosity is measured. The change value is measured and recorded each time using a rotational viscometer, and the change rate is calculated. For the measurement of thermal stability, a heating experiment with gradually increasing temperature (e.g., from room temperature to 100℃, the change of thermal stability coefficient is recorded every 5℃) is conducted. The temperature threshold of chemical component decomposition and its corresponding mass loss rate are statistically analyzed. The experimental data are statistically analyzed, and a mathematical relationship between concentration and performance change is established through a linear regression model. Finally, a component influence mapping map is generated.

[0126] Please see Figure 6 The specific steps for optimizing the grinding slurry ratio are as follows:

[0127] S311: Based on component influence mapping information, real-time monitoring of temperature data and additive reactions during the cutting process is used to adjust the proportion of chemical additives, employing the following formula:

[0128] OR new =OR old ×(1+k O ×(OT target -OT actual ))

[0129] The adjusted additive ratio OR new Among them, OR old It is the original additive ratio, OT target It is the target temperature stability value, OT. actual It is the currently measured temperature, k. O It is an adjustment coefficient, used to adjust the sensitivity of the additive ratio according to temperature differences;

[0130] Original additive ratio OR old The target temperature stability value OT is 0.05 (5%). target The actual temperature is OT, which is 100 degrees. actual The angle is 95 degrees, and the adjustment factor is k. O Set it to 0.01, substitute it into the formula to calculate:

[0131] OR new =0.05×(1+0.01×(100-95))=0.05×1.05=0.0525

[0132] Therefore, the new additive ratio is 0.0525, or 5.25%. This result shows that by fine-tuning the additive ratio, the temperature response of the grinding fluid can be effectively adjusted to adapt to different cutting conditions.

[0133] S312: Based on the adjusted additive ratio and combined with real-time feedback data, the grinding fluid ratio is adjusted to match different cutting conditions and obtain the grinding fluid ratio configuration.

[0134] Based on the adjusted additive ratio and combined with real-time feedback data, the grinding fluid ratio is adjusted. First, key parameters, including changes in cutting speed, cutting force, and temperature, are obtained from the real-time feedback during the cutting process. The data is correlated with the current additive ratio and cutting efficiency. By comparing the data, it is determined whether the existing ratio matches the cutting conditions. Then, the adjustment range and parameters are calculated. Dynamic optimization is achieved by modifying the ratio of additives to base fluid. The current ratio is updated using the compared data to ensure that the adjusted ratio can cover the requirements of the current cutting conditions, thus obtaining the grinding fluid ratio configuration.

[0135] Please see Figure 7 The specific steps for obtaining the evaluation results of the formulation effect are as follows:

[0136] S321: Conduct cutting experiments and collect key performance data in real time, such as cutting speed, temperature and surface integrity, to obtain experimental monitoring datasets;

[0137] Cutting experiments were conducted to collect key performance data in real time during the cutting process. Before the experiment, appropriate cutting tools and workpiece materials needed to be selected, and experimental variables, including cutting speed, feed rate, and depth of cut, needed to be set. Sensors were installed on the cutting area and the workpiece surface to monitor cutting temperature, surface roughness, and related vibration characteristics in real time. Temperature sensors were used to collect real-time temperature changes in the cutting area. The data was recorded by a data acquisition card and transmitted to the experimental computer. Surface roughness was dynamically monitored by a surface profilometer at a sampling frequency of one second. The cutting speed was captured by a high-speed photogrammetry device to capture the contact state and trajectory between the tool and the workpiece. The collected data was stored as a segmented time series and classified according to performance indicators. Then, outliers were removed through data cleaning to form a complete experimental monitoring dataset.

[0138] S322: Based on the experimental monitoring dataset, the formula is used:

[0139]

[0140] The deviation of performance in each experiment was quantified to obtain preliminary analysis results of the formulation, where EH represents the evaluation score of the formulation effect, and XH... i μ is the performance index of the i-th experiment. h It is the average value of the performance index, σ hIt is the standard deviation, N H It is the number of experiments;

[0141] The following data was collected, number of experiments N H =5, the average value of the performance index μ h =50, standard deviation σ h =5, Experimental performance index XH i =[48,52,49,53,47], substituting into the formula, we get:

[0142]

[0143] The results indicate that adjusting the grinding fluid ratio resulted in smaller performance deviations, suggesting that the ratio was relatively stable.

[0144] S323: Based on the preliminary analysis results of the proportioning, evaluate the performance of the new proportioning under different cutting conditions, determine the effective proportioning parameters, and obtain the proportioning effect evaluation results;

[0145] Based on the preliminary analysis results of the formulation, experimental and control groups were established for different cutting conditions. By adjusting the component concentration changes under the new formulation, the influence of the formulation on cutting performance was recorded. The cutting speed range was set to 100-1000 m / min, with a step size of 50 m / min. The cutting temperature and surface roughness under the corresponding conditions were measured respectively. The measured data were correlated with the formulation concentration one by one. Multivariate regression method was used to analyze the specific influence of each concentration combination on temperature and surface integrity. The concentration that can meet the balance point of multiple performance indicators was selected as the candidate formulation. The corresponding analysis data are presented in tabular form. The effective formulation parameters were determined, and the formulation effect evaluation results were output.

[0146] Please see Figure 8 The specific steps for obtaining the optimal performance configuration of the grinding slurry are as follows:

[0147] S411: Based on the evaluation results of the mixing effect, further optimize the grinding fluid parameter settings and refine the chemical component ratio;

[0148] Based on the evaluation of the formulation effect, the chemical composition ratio of the grinding fluid was further adjusted and optimized. The goal was to further refine the proportion of each component to achieve better performance. First, the performance data of the current grinding fluid, including viscosity, shear resistance, and chemical stability, were analyzed. By comparing experimental data with theoretical values, components with large performance deviations were identified. Subsequently, the components were adjusted one by one, increasing or decreasing the percentage of specific chemical components. Based on the adjustments, a small-batch trial production was carried out again to test the adjusted grinding fluid samples to verify whether the changed performance met the expected goals. Finally, the optimal chemical composition ratio that can improve the performance of the grinding fluid was determined, resulting in an updated grinding fluid formulation scheme.

[0149] S412: Based on the chemical composition ratio, the performance of the grinding slurry is tested to determine whether it meets the predetermined standards, using the following formula:

[0150]

[0151] The effects of the mixing ratio were evaluated to obtain performance verification results, where VX represents the variance of the grinding fluid performance deviation, and PX... actual,i This is the current performance parameter of the polishing slurry in the i-th test, PX target It is the target performance parameter, n b It is the number of tests;

[0152] There are five tests, with actual performance parameters of [102, 98, 105, 100, 99], and a target performance parameter of 100. The calculation formula is:

[0153]

[0154] This value reflects the average deviation of the performance parameters. The result shows that the overall performance deviation of the tested grinding fluid is small and close to the target performance, indicating that the mixing ratio is relatively ideal.

[0155] S413: Based on the performance verification results, analyze and correct the collected data to determine whether the polishing fluid has reached the optimal state and obtain the optimal performance configuration of the polishing fluid;

[0156] After obtaining detailed performance verification results, in-depth data analysis and necessary calibration are performed to determine whether the optimal state has been reached. During the analysis, statistical methods are used to evaluate the consistency and deviation of various performance data. For example, the performance stability of the grinding fluid is evaluated by calculating the standard deviation and coefficient of variation of performance parameters. At the same time, based on the performance test results, relevant parameters, such as viscosity threshold and temperature response curve, are adjusted to more accurately reflect the actual situation. Through comprehensive analysis, it is determined whether the grinding fluid has reached the predetermined optimal performance standard. If it has, the specific parameters of the optimal performance configuration are confirmed and recorded. If it has not, adjustments and tests continue until the performance requirements are met, and the optimal performance configuration of the grinding fluid is obtained.

[0157] A data-driven slurry performance optimization system, comprising:

[0158] The key parameter analysis module, based on the collected grinding fluid data, collects the performance indicators and cutting condition data of the grinding fluid, performs time series analysis on cutting speed, temperature, viscosity index and pH value, identifies the key change nodes affecting performance, and re-analyzes the node data to obtain the key performance node identifiers.

[0159] The data relationship mining module, based on the identification of key performance nodes, mines nonlinear relationships in the data, analyzes the impact of additives on the performance indicators of the grinding fluid, identifies key chemical components that affect the performance of the grinding fluid, and generates component impact mapping information.

[0160] The formulation optimization module utilizes the component influence mapping information to dynamically adjust the temperature stability and chemical additive ratio, and combines it with real-time cutting feedback to optimize the grinding fluid formulation and obtain dynamic formulation adjustment information.

[0161] The proportioning monitoring module monitors the adjusted grinding fluid proportion in real time based on dynamic proportioning adjustment information, analyzes its impact on cutting efficiency and quality, and generates proportioning effect evaluation results.

[0162] Based on the evaluation results of the mixing effect, the performance verification module further optimizes the grinding fluid parameter settings, refines the chemical composition ratio, and verifies whether the performance of the grinding fluid meets the predetermined standard through experimental verification and data calibration, thereby obtaining the optimal performance configuration of the grinding fluid.

[0163] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for optimizing the performance of a polishing liquid based on data analysis, characterized in that: The following steps are involved: Based on the collected grinding fluid data, identify the key parameters that affect the performance of the grinding fluid, conduct time series analysis on the cutting speed, temperature, viscosity index and pH value data, determine the key change nodes of the performance, re-analyze the node data, and obtain the key performance node identification; Based on the key performance node identifiers, nonlinear relationships in the data are mined, including the effects of additives on the performance indicators of the grinding fluid, the chemical components that have the greatest impact on the performance of the grinding fluid are identified, and component impact mapping information is generated; Based on the component impact mapping information, the temperature stability and the ratio of chemical additives are dynamically adjusted, the grinding fluid ratio is optimized in combination with the real-time feedback of cutting, and the adjusted ratio is monitored in real time to analyze the impact on cutting efficiency and quality, and obtain the ratio effect evaluation result; According to the evaluation results of the ratio effect, the grinding liquid parameter settings are optimized again, the chemical composition ratio is refined, and it is verified whether the grinding liquid performance meets the predetermined standards, including experimental verification and data calibration, to obtain the optimal performance configuration of the grinding liquid.

2. The method for optimizing the performance of a polishing liquid based on data analysis according to claim 1, characterized in that: The identification steps of the key parameters are specifically as follows: Based on the collected grinding fluid data, including cutting speed, temperature, viscosity index and pH value, data cleaning and outlier processing are performed to obtain a processed performance data set; Based on the processed performance data set, combined with the impact of each parameter fluctuation on performance, the formula is adopted: Calculate the weighted correlation coefficient r weighted , identify the parameters associated with the fluctuation of grinding fluid performance and obtain a preliminary list of key parameters, where x i Represents the independent variable value of the i-th monitoring point, y i represents the dependent variable value of the ith monitoring point, and is the average value, w i is the weight of the i-th monitoring point; Based on the preliminary key parameter list, experimental verification is carried out to adjust single parameters and monitor their effects on the performance of the polishing liquid to determine the key parameters that affect the performance of the polishing liquid.

3. The method for optimizing the performance of a polishing liquid based on data analysis according to claim 1, characterized in that: The steps for obtaining the performance critical node identifier are specifically as follows: Performing time series analysis on the cutting speed, temperature, viscosity index and pH value data, using an ARMA model to smooth the data and highlight long-term trends, and identify potential change nodes; Based on the potential change nodes, the trend change or volatility enhancement points are identified through Bai-Perron breakpoint detection to obtain preliminary key change nodes; The preliminary key change nodes are re-analyzed, the performance data differences before and after the key nodes are compared, and the validity of the nodes and the impact on the performance are verified to obtain the performance key node identification.

4. The method for optimizing the performance of a polishing liquid based on data analysis according to claim 1, characterized in that: The steps of mining the nonlinear relationship are specifically as follows: Based on the key performance node identification, the data points that do not meet the standards are removed, and the relationship between the additive concentration and the performance index is explored, using the formula: Determine the additive concentration level, where RS represents a measure of the nonlinear relationship and CS i is the additive concentration at the ith data point, CS avg is the mean value of the additive concentration, CS0 is the median concentration, and N is the total number of data points; The performance changes corresponding to the various concentration levels of the additives are calculated, and the statistical significance is compared and verified to obtain the nonlinear relationship analysis results.

5. The method for optimizing the performance of a polishing liquid based on data analysis according to claim 1, characterized in that: The steps for obtaining the component impact mapping information are specifically as follows: Based on the nonlinear relationship, the data between the additive composition and the grinding fluid performance index are analyzed, and the formula is used: Identify the chemical components with the greatest impact and obtain the chemical component impact index IY, where Py i is the performance index value corresponding to the i-th data point, Py avg is the average value of the performance index, Py std is the standard deviation of the performance indicator, wy i is the weight adjusted according to the concentration of the chemical component, and N is the total number of data points; Based on the chemical component impact index, the impact of each key chemical component is quantified to obtain component impact mapping information.

6. The method for optimizing the performance of a polishing liquid based on data analysis according to claim 1, characterized in that: The steps of optimizing the grinding liquid ratio are specifically as follows: Based on the component impact mapping information, the temperature data and additive reaction during the cutting process are monitored in real time, and the chemical additive ratio is adjusted using the formula: OR new =OR old ×(1+k O ×(OT target -OT actual )) Get the adjusted additive ratio OR new , where OR old is the original additive ratio, OT target is the target temperature stability value, OT actual is the current measured temperature, k O is the adjustment factor; Based on the adjusted additive ratio, the grinding fluid ratio is adjusted in combination with real-time feedback data to match the different cutting conditions to obtain the grinding fluid ratio configuration.

7. The method for optimizing polishing liquid performance based on data analysis according to claim 1, characterized in that: The steps for obtaining the ratio effect evaluation result are specifically as follows: Conduct cutting experiments, collect key performance data in real time, and obtain experimental monitoring data sets; Based on the experimental monitoring data set, the formula is adopted: Quantify the degree of deviation of each experimental performance and obtain the preliminary analysis results of the ratio, where EH represents the evaluation score of the ratio effect, XH i is the performance index of the ith experiment, μ h is the average value of the performance index, σ h is the standard deviation, N H is the number of experiments; According to the preliminary analysis results of the ratio, the performance of the new ratio under different cutting conditions is evaluated, the effective ratio parameters are determined, and the ratio effect evaluation results are obtained.

8. The method for optimizing polishing liquid performance based on data analysis according to claim 1, characterized in that: The steps for obtaining the optimal performance configuration of the grinding liquid are specifically as follows: According to the evaluation results of the mixing ratio, the grinding liquid parameter settings are optimized again to refine the chemical composition ratio; Based on the chemical composition ratio, the performance of the grinding fluid is tested to see whether it meets the predetermined standard, using the formula: Evaluate the ratio effect and obtain the performance verification results, where VX represents the variance of the grinding fluid performance deviation, PX actual,i is the current performance parameter of the grinding fluid in the ith test, PX target is the target performance parameter, n b is the number of tests; Based on the performance verification results, the collected data is analyzed and corrected to determine whether the polishing liquid has reached the optimal state, thereby obtaining the optimal performance configuration of the polishing liquid.

9. The polishing liquid performance optimization system based on data analysis is characterized in that: According to any one of claims 1 to 8, the method for optimizing the performance of a grinding liquid based on data analysis is performed, wherein the system comprises: The key parameter analysis module collects the performance indicators and cutting condition data of the grinding fluid based on the collected grinding fluid data, performs time series analysis on the cutting speed, temperature, viscosity index and pH value, determines the key change nodes that affect the performance, and re-analyzes the node data to obtain the key performance node identification; The data relationship mining module mines the nonlinear relationship in the data based on the key performance node identifier, analyzes the influence of additives on the performance indicators of the grinding fluid, identifies the key chemical components that affect the performance of the grinding fluid, and generates component influence mapping information; The ratio optimization module uses the component impact mapping information to dynamically adjust the temperature stability and the ratio of chemical additives, and optimizes the grinding fluid ratio in combination with the real-time feedback of cutting to obtain dynamic ratio adjustment information; The ratio monitoring module monitors the adjusted grinding fluid ratio in real time based on the dynamic ratio adjustment information, analyzes its influence on cutting efficiency and quality, and generates a ratio effect evaluation result; The performance verification module further optimizes the grinding fluid parameter settings based on the ratio effect evaluation results, refines the chemical composition ratio, verifies whether the grinding fluid performance meets the predetermined standards through experimental verification and data calibration, and obtains the optimal performance configuration of the grinding fluid.

Images