Crankshaft machining optimization method and system

Abstract

The invention relates to the technical field of automobile part machining, and particularly discloses a crankshaft machining optimization method and system.The method comprises the steps that a grinding force signal and a vibration signal in the grinding process are collected; grinding feature information is obtained according to the grinding force signal, and vibration feature information is obtained according to the vibration signal; material hardness information is generated according to the grinding feature information; grinding stability information is generated according to the grinding feature information and the vibration feature information; on the basis of a preset strategy library, parameter adjustment amount is generated according to the material hardness information and the grinding stability information, and the parameter adjustment amount comprises feeding speed adjustment amount and / or grinding depth adjustment amount; according to the method, a grinding force signal and a vibration signal which are acquired in real time based on a sensor are converted into quantitative information reflecting the material hardness and the grinding stability, and the parameter adjustment amount is dynamically generated according to the real-time information based on a preset strategy library, so that the effect of considering the machining efficiency and the stability is achieved.

Description

Technical Field

[0001] This application relates to the technical field of automotive part processing. Specifically, it relates to a method and system for optimizing crankshaft processing. Background Art

[0002] As a core component of an engine, the automotive crankshaft is processed on a highly automated production line, involving multiple precision processes. With the increasing market demand for automotive production capacity, numerically controlled machine tools achieve efficient processing by presetting machining programs and process parameters. Continuously optimizing the machining cycle has become a key development direction for improving the production line efficiency. Modern manufacturing technology has established a stable foundation for the machining process through standardized process parameters and automated control.

[0003] However, fluctuations in the material properties of the crankshaft blank can lead to dynamic changes in the physical state during the grinding process. Traditional fixed process parameters set based on the worst or average working conditions are difficult to adapt to real-time machining conditions. In areas with higher material hardness, the increased grinding force is likely to cause abnormal tool wear and increased vibration; in areas with lower material hardness, the fixed parameters limit the material removal efficiency. Existing methods cannot dynamically adjust parameters according to the actual grinding state, resulting in difficulty in balancing machining efficiency and stability, and frequent manual intervention will further affect production continuity.

[0004] In response to the above problems, there is currently no effective technical solution. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for optimizing crankshaft processing, so as to dynamically adjust parameters according to the actual grinding state, balance machining efficiency and stability, and improve production continuity.

[0006] In a first aspect, this application provides a method for optimizing crankshaft processing, which is applied to the grinding process of an automotive crankshaft. The method is characterized by including the following steps:

[0007] S1. Based on a sensor assembly, collect the grinding force signal and vibration signal during the grinding process in real time;

[0008] S2. Obtain grinding feature information according to the grinding force signal, and obtain vibration feature information according to the vibration signal;

[0009] S3. Generate material hardness information according to the grinding feature information;

[0010] S4. Generate grinding stability information according to the grinding feature information and the vibration feature information;

[0011] S5. Based on a preset strategy library, generate a parameter adjustment amount according to the material hardness information and the grinding stability information. The parameter adjustment amount includes a feed speed adjustment amount and / or a grinding depth adjustment amount.

[0012] The method of the present application can sense the grinding force and vibration state during the grinding process in real time, and based on this information, evaluate the hardness of the workpiece material and the stability of the grinding process and adjust the processing parameters, solving the problem that traditional fixed process parameters are difficult to adapt to fluctuations in material properties, and achieving the effect of balancing processing efficiency and stability.

[0013] The crankshaft machining optimization method, wherein step S2 comprises:

[0014] S21, intercepting the most recent grinding force signal and vibration signal based on a preset time window;

[0015] S22, calculating and obtaining an average grinding force value, a grinding force variance value, and a grinding force kurtosis value according to the intercepted grinding force signal as grinding feature information;

[0016] S23. Calculate and obtain a vibration peak factor and a vibration frequency according to the intercepted vibration signal as vibration characteristic information, wherein the vibration peak factor is a ratio of a peak value to a root mean square value of the vibration signal.

[0017] The above processing method can accurately and quickly extract key characteristic information that can effectively reflect the grinding status from the grinding force signal and vibration signal collected in real time. These characteristic information comprehensively covers the grinding load, process stability, abnormal impact and vibration characteristics, and provides reliable data support for the subsequent generation of material hardness information and grinding stability information, thereby improving the accuracy and effectiveness of the entire crankshaft processing optimization method.

[0018] The crankshaft machining optimization method, wherein step S3 comprises:

[0019] S31, based on the pre-trained hardness prediction model, taking the grinding feature information as input, outputting the material hardness prediction value at the current moment;

[0020] S32, using a first-order lag filtering algorithm to smooth the material hardness prediction value at the current moment based on the material hardness information at the previous moment, and using the smoothed material hardness prediction value as the material hardness information at the current moment.

[0021] The crankshaft machining optimization method, wherein step S4 comprises:

[0022] S41, calculating a first grinding stability index according to the average grinding force value and the grinding force variance value in the grinding feature information, wherein the first grinding stability index is a ratio of the grinding force variance value to the average grinding force value;

[0023] S42. Calculate a second grinding stability index based on the vibration peak factor in the vibration characteristic information, where the second grinding stability index is the difference between the vibration peak factor and a preset vibration threshold value;

[0024] S43. Calculate a third grinding stability index based on the vibration frequency in the vibration characteristic information, where the third grinding stability index is the ratio of the absolute value of the difference between the vibration frequency and a preset frequency to the preset frequency;

[0025] S44. Perform a weighted sum of the first grinding stability index, the second grinding stability index, and the third grinding stability index based on a preset weight coefficient to obtain the grinding stability information.

[0026] For the crankshaft machining optimization method described above, the preset vibration threshold value is set based on the mean and standard deviation of historical peak factors.

[0027] For the crankshaft machining optimization method described above, the preset frequency is set according to the resonance peak of the natural frequency of the machine tool - workpiece - grinding wheel combination obtained through modal testing or finite element simulation.

[0028] For the crankshaft machining optimization method described above, the strategy library includes the mapping relationships between multiple grinding state intervals and parameter adjustment amounts;

[0029] Step S5 includes:

[0030] S51. Establish a grinding state based on the material hardness information and the grinding stability information;

[0031] S52. Analyze the grinding state interval where the grinding state is located, and extract the corresponding parameter adjustment amount in combination with the strategy library.

[0032] For the crankshaft machining optimization method described above, the method further includes a step performed between step S1 and step S2:

[0033] SA. Perform filtering processing on the grinding force signal and the vibration signal.

[0034] For the crankshaft machining optimization method described above, the method further includes a step performed after step S5:

[0035] S6. Obtain the usage duration of the grinding wheel, and determine a feed speed correction factor and / or a grinding depth correction factor according to the usage duration and a preset correction mapping table, where the correction mapping table includes feed speed correction factors and / or grinding depth correction factors matching different usage duration intervals;

[0036] S7. Compensate and adjust the feed rate adjustment amount by using the feed rate correction factor, and / or compensate and adjust the grinding depth adjustment amount by using the grinding depth correction factor.

[0037] In a second aspect, the present application also provides a crankshaft machining optimization system, which is applied to the grinding machining of automotive crankshafts. The system includes:

[0038] An acquisition module, configured to collect the grinding force signal and the vibration signal during the grinding process in real time based on the sensor assembly;

[0039] A feature extraction module, configured to obtain grinding feature information according to the grinding force signal and obtain vibration feature information according to the vibration signal;

[0040] A first calculation module, configured to generate material hardness information according to the grinding feature information;

[0041] A second calculation module, configured to generate grinding stability information according to the grinding feature information and the vibration feature information;

[0042] A strategy adjustment module, configured to generate a parameter adjustment amount based on a preset strategy library according to the material hardness information and the grinding stability information. The parameter adjustment amount includes a feed rate adjustment amount and / or a grinding depth adjustment amount.

[0043] The system of the present application can perceive the grinding force and vibration states during the grinding process in real time, and evaluate the hardness of the workpiece material and the stability of the grinding process based on this information and adjust the machining parameters, solving the problem that traditional fixed process parameters are difficult to adapt to the fluctuations of material characteristics, and achieving the effect of taking into account both machining efficiency and stability.

[0044] As can be seen from the above, the present application provides a crankshaft machining optimization method and system. Among them, the method of the present application converts the grinding force signal and the vibration signal collected in real time based on the sensor into quantitative information reflecting the material hardness and the grinding stability, and based on a preset strategy library, dynamically generates a parameter adjustment amount according to this real-time information; therefore, the method of the present application can perceive the grinding force and vibration states during the grinding process in real time, and evaluate the hardness of the workpiece material and the stability of the grinding process based on this information and adjust the machining parameters. Thus, the machining process can better adapt to the fluctuations of the crankshaft blank material characteristics, appropriately reduce the parameters when the material hardness is high to ensure machining stability and reduce tool wear, and appropriately increase the parameters when the material hardness is low to improve the material removal efficiency, thereby solving the problem that traditional fixed process parameters are difficult to adapt to the fluctuations of material characteristics, and achieving the effect of taking into account both machining efficiency and stability. Description of the Drawings

[0045] Figure 1Flowchart of the crankshaft machining optimization method provided by some embodiments of the present application.

[0046] Figure 2 Flowchart of the crankshaft machining optimization method provided by some other embodiments of the present application.

[0047] Figure 3 Schematic structural diagram of the crankshaft machining optimization system provided by some embodiments of the present application.

[0048] Figure 4 Schematic structural diagram of the crankshaft machining optimization system provided by some other embodiments of the present application.

[0049] Reference numerals: 201, acquisition module; 202, feature extraction module; 203, first calculation module; 204, second calculation module; 205, strategy adjustment module; 206, filtering module; 207, correction and compensation module. Detailed implementation manners

[0050] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application claimed, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0051] It should be noted that: similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, terms such as "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0052] In a first aspect, please refer to Figure 1 and Figure 2 , some embodiments of the present application provide a crankshaft machining optimization method, which is applied to the grinding process of automotive crankshafts. The method includes the following steps:

[0053] S1. Based on the sensor assembly, the grinding force signal and the vibration signal during the grinding process are collected in real time;

[0054] S2. The grinding feature information is obtained according to the grinding force signal, and the vibration feature information is obtained according to the vibration signal;

[0055] S3. Generate material hardness information based on the grinding feature information;

[0056] S4. Generate grinding stability information based on the grinding feature information and the vibration feature information;

[0057] S5. Based on a preset policy library, generate a parameter adjustment amount according to the material hardness information and the grinding stability information, where the parameter adjustment amount includes a feed speed adjustment amount and / or a grinding depth adjustment amount.

[0058] Specifically, the sensor assembly refers to a device used to sense and measure physical quantities and convert them into electrical signals, which can be implemented in various forms, such as force sensors, acceleration sensors, acoustic emission sensors, etc. Its main purpose is to obtain the physical state information during the grinding process in real time.

[0059] More specifically, the grinding feature information refers to the quantitative indicators extracted from the grinding force signal that can reflect the characteristics of the grinding process, which can include the average value, fluctuation degree, frequency components, etc. of the grinding force. Its main purpose is to characterize the force state during the grinding process. The vibration feature information refers to the quantitative indicators extracted from the vibration signal that can reflect the vibration state of the grinding system, which can include the amplitude, frequency, energy distribution, etc. of the vibration. Its main purpose is to characterize the dynamic stability during the grinding process.

[0060] More specifically, the material hardness information refers to the quantitative information reflecting the hardness of the crankshaft blank material in the current grinding area, which can be inferred based on information such as the grinding force. Its main purpose is to evaluate the grinding difficulty of the current workpiece material. The grinding stability information refers to the quantitative information reflecting the stability degree of the current grinding process, which can be evaluated by comprehensively considering information such as the grinding force and vibration. Its main purpose is to determine whether there are any abnormalities or unstable trends in the current grinding process.

[0061] More specifically, the policy library refers to a pre-established data structure containing the corresponding relationships between different grinding states and the corresponding parameter adjustment amounts, which can be presented in the form of a lookup table, a rule set, or a model, etc. Its main purpose is to determine how to adjust the processing parameters according to the real-time evaluated grinding state. The parameter adjustment amount refers to the value determined according to the real-time grinding state and used to correct the current processing parameters (such as the feed speed and the grinding depth), which can include an increase amount or a decrease amount. Its main purpose is to achieve the dynamic optimization of the processing parameters.

[0062] The working principle of the method of the present application is as follows: First, based on the sensor component, the grinding force signal and the vibration signal during the grinding process are collected in real time, and these signals are the basic data reflecting the physical state of the grinding process. Then, according to the collected grinding force signal, the grinding characteristic information is extracted, and at the same time, according to the vibration signal, the vibration characteristic information is extracted, and the original signal is converted into a more representative quantization index. On this basis, the material hardness information is generated according to the grinding characteristic information to evaluate the material properties of the current grinding area; at the same time, according to the grinding characteristic information and the vibration characteristic information, the grinding stability information is generated to evaluate the stability of the current grinding process. Finally, based on the preset strategy library, combined with the material hardness information and the grinding stability information obtained in real time, the corresponding parameter adjustment amount is generated, and this adjustment amount is used to correct the current feed speed and / or grinding depth, thereby forming a dynamic adjustment mechanism for machining parameters based on real-time state feedback. It is precisely because it can perceive and evaluate the grinding state in real time and adjust the parameters accordingly that the machining process can adapt to material fluctuations and avoid efficiency losses or stability problems caused by parameter mismatches.

[0063] It should be noted that step S3 and step S4 can be executed sequentially, or simultaneously, or step S4 is executed first and then step S3. In the embodiments of the present application, it is preferably executed simultaneously.

[0064] The method of the present application converts the grinding force signal and the vibration signal collected in real time based on the sensor into quantization information reflecting the material hardness and the grinding stability, and based on the preset strategy library, dynamically generates the parameter adjustment amount according to this real-time information; therefore, the method of the present application can perceive the grinding force and the vibration state in the grinding process in real time, and evaluate the hardness of the workpiece material and the stability of the grinding process based on this information and adjust the machining parameters. Thus, the machining process can better adapt to the fluctuations of the material properties of the crankshaft blank. When the material hardness is relatively high, the parameters are appropriately reduced to ensure machining stability and reduce tool wear. When the material hardness is relatively low, the parameters are appropriately increased to improve the material removal efficiency, thereby solving the problem that traditional fixed process parameters are difficult to adapt to material property fluctuations and achieving the effect of balancing machining efficiency and stability.

[0065] In some preferred embodiments, step S2 includes:

[0066] S21. Intercept the most recent grinding force signal and vibration signal based on a preset time window;

[0067] S22. Calculate and obtain the average grinding force value, the grinding force variance value, and the grinding force kurtosis value according to the intercepted grinding force signal as the grinding characteristic information;

[0068] S23. Calculate and obtain the vibration peak factor and vibration frequency based on the intercepted vibration signal as vibration characteristic information. The vibration peak factor is the ratio of the peak value to the root mean square value of the vibration signal.

[0069] Specifically, the preset time window refers to a continuous time length for intercepting the signal data collected in real time, and its purpose is to obtain signal data that can reflect the current grinding state and has timeliness; the average grinding force value refers to the arithmetic mean of the grinding force signal within the preset time window, which reflects the overall grinding load level during the grinding process; the grinding force variance value refers to the degree of dispersion of the grinding force signal relative to its average value within the preset time window, which reflects the severity of the grinding force fluctuation; the grinding force kurtosis value refers to the sharpness of the probability density distribution of the grinding force signal within the preset time window, which can sensitively capture abnormal impacts or mutations in the signal; the vibration peak factor refers to the ratio of the peak value to the root mean square value of the vibration signal within the preset time window, which reflects the impact characteristics of the vibration signal; the vibration frequency refers to the main frequency component of the vibration signal within the preset time window, which reflects the vibration source and vibration characteristics of the vibration signal.

[0070] More specifically, the solution of the present application intercepts the most recent grinding force signal and vibration signal based on the preset time window, ensuring the timeliness of the analyzed data and being able to reflect the current grinding state in real time. On this basis, calculate and obtain the average grinding force value, grinding force variance value, and grinding force kurtosis value according to the intercepted grinding force signal. These characteristic parameters respectively reflect the load, stability, and impact characteristics of the grinding process from different dimensions. At the same time, calculate and obtain the vibration peak factor and vibration frequency according to the intercepted vibration signal. These characteristic parameters reflect the vibration impact characteristics and main vibration sources of the grinding process. By comprehensively extracting these representative time-domain and frequency-domain characteristics, the physical state of the grinding process can be described more comprehensively and accurately, providing a reliable data basis for generating material hardness information and grinding stability information subsequently. This multi-dimensional and time-effective feature extraction method enables the entire crankshaft machining optimization method to more effectively perceive changes in the grinding state, thereby achieving more accurate parameter adjustment.

[0071] More specifically, the above processing method can accurately and quickly extract key characteristic information from the grinding force signal and vibration signal collected in real time, which can effectively reflect the grinding state. These characteristic information comprehensively cover grinding load, process stability, abnormal impact, and vibration characteristics, providing reliable data support for the subsequent generation of material hardness information and grinding stability information, thereby improving the accuracy and effectiveness of the entire crankshaft machining optimization method.

[0072] In some preferred embodiments, step S3 includes:

[0073] S31. Based on a pre-trained hardness prediction model, using the grinding feature information as input, output the predicted value of the material hardness at the current moment;

[0074] S32. Use a first-order lag filtering algorithm to smooth the predicted value of the material hardness at the current moment based on the material hardness information at the previous moment, and take the smoothed predicted value of the material hardness as the material hardness information at the current moment.

[0075] Specifically, the pre-trained hardness prediction model refers to a mathematical model obtained by training with historical data, which can specifically be a model constructed based on machine learning algorithms, such as a regression model, a neural network model, or a support vector machine model. Its purpose is to establish a mapping relationship between the grinding feature information and the material hardness, so as to be able to quickly predict the material hardness according to the real-time grinding feature information.

[0076] More specifically, the first-order lag filtering algorithm is a commonly used digital signal processing algorithm. Specifically, it can calculate the output value at the current moment by weighted averaging the current input value and the output value at the previous moment. Its purpose is to smooth the time-series data, reduce data fluctuations, and improve the stability of the data. The predicted value of the material hardness refers to the estimated value of the material hardness directly output by the hardness prediction model according to the input grinding feature information, and this value may fluctuate due to the influence of instantaneous grinding state changes. The material hardness information refers to the predicted value of the material hardness processed by the first-order lag filtering algorithm, and this value is smoother and more stable than the original predicted value and can more accurately reflect the actual hardness characteristics of the material.

[0077] More specifically, the first-order lag filtering algorithm is used to smooth this predicted value. This filtering algorithm utilizes the continuity of the material hardness change in the time series, corrects the predicted value at the current moment by introducing the material hardness information at the previous moment, and effectively suppresses the fluctuations of the predicted value caused by instantaneous disturbances or sensor noise during the grinding process. Thus, the output material hardness information is more stable and reliable and can more accurately reflect the actual hardness change trend of the material.

[0078] More specifically, step S32 is preferably smoothed by a first-order lag filtering algorithm expressed as H(k)=α*Hpred(k)+(1-α)*H(k - 1), where H(k) is the material hardness information at the k-th moment, Hpred(k) is the predicted value of the material hardness at the k-th moment, H(k - 1) is the material hardness information at the (k - 1)-th moment, and α is the filtering coefficient, which can be set according to the usage requirements, preferably 0.3. In this way, the material hardness information at the current moment comprehensively considers the current predicted value and the historical smoothed value, thus realizing the smoothing of the predicted value.

[0079] More specifically, the hardness prediction model is specifically designed for the three-dimensional features of the average grinding force value, the variance value of the grinding force, and the kurtosis value of the grinding force. In the embodiments of the present application, it is preferably a 1D CNN-LSTM (one-dimensional convolutional neural network and long short-term memory network) hybrid network architecture to process its temporal characteristics; the input of the hybrid network is a feature matrix (dimension 50×3) containing 50 consecutive sampling points, where each column corresponds to the above three mechanical characteristics (average grinding force value, variance value of the grinding force, and kurtosis value of the grinding force), and each column corresponds to the above three mechanical characteristics; the hardness prediction model is provided with two levels of 1D convolutional layers (number of channels 16→32) combined with ReLU activation, and the local association patterns between the three features are extracted through 5×1 and 3×1 convolutional kernels respectively; a 64-unit LSTM layer is connected at the back end to specifically model the long-range dynamic evolution of the three-feature sequence; finally, the full connection layer fuses the spatio-temporal features to output the predicted value of the material hardness. This architecture particularly strengthens the collaborative analysis ability of the kurtosis value (representing the sharpness of the force signal) and the variance value (representing the intensity of the force fluctuation), ensuring accurate capture of the material hardness characteristics.

[0080] More specifically, the training process of the hardness prediction model is to train based on a pre-constructed sample set, which includes three-dimensional grinding features (average grinding force value, variance value of the grinding force, and kurtosis value of the grinding force) and the true hardness value of the corresponding material; the Huber loss function is used in the training process to mainly suppress the regression deviation of high-variance samples, and feature-level Dropout (random inactivation, probability 0.15) is implanted in the gating mechanism of the LSTM layer to strengthen the generalization ability for the kurtosis mutation pattern.

[0081] In some preferred embodiments, step S4 includes:

[0082] S41. Calculate the first grinding stability index according to the average grinding force value and the variance value of the grinding force in the grinding feature information, and the first grinding stability index is the ratio of the variance value of the grinding force to the average grinding force value;

[0083] S42. Calculate the second grinding stability index according to the vibration peak factor in the vibration feature information, and the second grinding stability index is the difference between the vibration peak factor and the preset vibration threshold;

[0084] S43. Calculate the third grinding stability index according to the vibration frequency in the vibration feature information, and the third grinding stability index is the ratio of the absolute value of the difference between the vibration frequency and the preset frequency to the preset frequency;

[0085] S44. Based on the preset weight coefficients, perform weighted summation on the first grinding stability index, the second grinding stability index, and the third grinding stability index to obtain the grinding stability information.

[0086] Specifically, the first grinding stability index reflects the degree of fluctuation of the grinding force relative to its average level and can be used to characterize the smoothness of the grinding process; the preset vibration threshold can be set according to historical data, experience, or theoretical calculations, and the second grinding stability index is used to determine whether the current vibration exceeds the normal range; the preset frequency can be set according to the characteristics of systems such as the machine tool, workpiece, and grinding wheel, and the third grinding stability index is used to measure the degree of proximity between the actual vibration frequency and the natural frequency or key frequency of the system; the preset weight coefficients refer to the coefficients used for weighted summation of the first grinding stability index, the second grinding stability index, and the third grinding stability index, and these coefficients can be determined according to the influence degree of different indexes on grinding stability, experience, or optimization algorithms. The purpose is to comprehensively consider the importance of different indexes and obtain more comprehensive grinding stability information.

[0087] More specifically, the solution of the present application evaluates grinding stability by comprehensively considering multiple dimensions of information in the grinding process. First, by calculating the ratio of the grinding force variance value to the average grinding force value, the relative degree of grinding force fluctuation is quantified, which can reflect the smoothness of the grinding process more than simply focusing on the magnitude of the grinding force. Second, by calculating the difference between the vibration peak factor and the preset vibration threshold, it is directly determined whether the vibration amplitude is at an abnormal level, and potential impacts or abnormal vibrations can be quickly identified. At the same time, by calculating the relative deviation between the vibration frequency and the preset frequency, it is evaluated whether there are unstable factors such as approaching the system resonance frequency in the grinding process. Finally, these three indexes reflecting grinding stability from different angles are weighted and synthesized to obtain a more comprehensive and reliable grinding stability information. This method makes full use of various characteristic information extracted from the original grinding force signal and vibration signal, avoids the limitations of a single index, and can more accurately identify unstable states in the grinding process. It is precisely due to this multi-index comprehensive evaluation method that it is possible to more finely grasp the dynamic changes in the grinding process and provide a more reliable basis for subsequent parameter adjustment.

[0088] More specifically, the above processing method comprehensively utilizes grinding force characteristics and vibration characteristics to evaluate the stability of the grinding process from multiple dimensions, overcomes the deficiency of relying only on a single parameter for judgment, makes the evaluation of the grinding state more accurate and comprehensive, can more effectively identify potential unstable factors, thereby providing a more reliable basis for subsequent parameter adjustment, and helps to improve the stability and efficiency of crankshaft grinding.

[0089] In some preferred embodiments, the preset vibration threshold is set based on the mean and standard deviation of historical peak factors.

[0090] Specifically, the preset vibration threshold refers to the boundary value used to determine whether the vibration state in the grinding process exceeds the normal range, and its purpose is to serve as a reference standard for evaluating grinding stability. The mean value of the historical peak factors refers to the arithmetic mean of multiple vibration peak factors collected over a past period (or multiple reasonable vibration peak factors collected in advance), and its purpose is to reflect the overall level of the historical vibration peak factors; the standard deviation of the historical peak factors is the degree of dispersion of multiple vibration peak factors corresponding to the mean value of the historical peak factors relative to their mean value, and its purpose is to reflect the fluctuation range of the historical vibration peak factors.

[0091] More specifically, the setting of the preset vibration threshold is preferably such that: H p =a(P pf +k·σ pf ),H p is the preset vibration threshold, P pf is the mean value of the historical peak factors, σ pf is the standard deviation of the historical peak factors, a is the safety margin coefficient, and its value is preferably 1.1 - 1.2, and k is the proportionality coefficient, and its value is preferably 2.

[0092] More specifically, in this embodiment, the safety margin coefficient is used to increase an additional safety margin when calculating the preset vibration threshold, and its purpose is to reduce the risk of misjudging normal fluctuations as abnormal vibrations; the proportionality coefficient is used to adjust the influence weight of the standard deviation of the historical peak factors in the preset vibration threshold calculation formula, and its purpose is to dynamically adjust the threshold according to the volatility of the vibration.

[0093] More specifically, the preset vibration threshold is dynamically adjusted based on the statistical characteristics of the historical vibration peak factors instead of using a fixed value. This setting method fully considers the average level and fluctuation range of vibrations in the actual machining process, enabling the threshold to better adapt to the current machining environment. Therefore, when evaluating grinding stability, it can more accurately distinguish normal vibration fluctuations from true abnormal vibrations, avoiding misjudgment. This significantly improves the accuracy of grinding stability evaluation, provides a more reliable basis for subsequent parameter adjustment based on stability information, and helps to achieve a more stable and reliable crankshaft grinding process.

[0094] In some preferred embodiments, the preset frequency is set according to the resonance peak of the natural frequency of the machine tool - workpiece - grinding wheel combination obtained through modal testing or finite element simulation.

[0095] Specifically, modal testing refers to an experimental method that determines the dynamic characteristics of a structure, including natural frequencies, modal damping, and modal shapes, by applying excitation to the structure and measuring its response. This can be achieved using devices such as exciters and accelerometers. Finite element simulation refers to a numerical calculation method that discretizes a complex structure into a finite number of elements, establishes a mathematical model, and performs calculations to simulate the mechanical behavior of the structure.

[0096] More specifically, the resonance peaks of the natural frequencies of the machine tool-workpiece-grinding wheel combination refer to the specific frequencies that the machine tool, workpiece, and grinding wheel, as an integrated system in the grinding processing system, have when vibrating freely without the action of external periodic forces, as well as the corresponding peaks on the frequency response function curve. These peaks indicate the degree of strong response of the system to external forces at that frequency.

[0097] More specifically, the resonance peaks represent the frequency points where the system responds strongly to external forces. Therefore, grinding processing near these frequencies is likely to trigger resonance. In this solution, the preset frequency used to calculate the grinding stability information is set to a value far from these resonance peaks, for example, set to 85% of the resonance peak. This setting method places the preset frequency in a relatively safe frequency region, far from the resonance points where the system is unstable. During the actual processing, by real-time monitoring the vibration signal and obtaining the current vibration frequency, comparing it with this preset frequency far from the resonance peak, the third grinding stability index is calculated. When the real-time vibration frequency is close to the preset frequency, it indicates that the current vibration frequency is far from the resonance region and the processing is relatively stable; when the real-time vibration frequency is far from the preset frequency, it may mean that the vibration frequency is drifting to other regions, and it is necessary to comprehensively evaluate the grinding stability by combining other information such as grinding force and vibration peak factor.

[0098] More specifically, by setting the preset frequency at a position far from the resonance peak and incorporating it into the grinding stability evaluation system, this solution can identify and avoid potential resonance risks, thereby maintaining the stability of the processing process under dynamically changing processing conditions. Combined with the method of dynamically adjusting parameters based on real-time signals, the parameter adjustment can respond to the actual processing state, avoid improper operations in the unstable state, and improve the robustness and optimization effect of the entire processing system.

[0099] In some preferred embodiments, the policy library includes the mapping relationships between multiple grinding state intervals and parameter adjustment amounts;

[0100] Step S5 includes:

[0101] S51. Establish a grinding state based on the material hardness information and the grinding stability information;

[0102] S52. Analyze the grinding state interval where the grinding state is located, and extract the corresponding parameter adjustment amount in combination with the strategy library.

[0103] Specifically, the strategy library refers to a data structure that stores the mapping relationship between multiple grinding state intervals and parameter adjustment amounts. It can be implemented in the form of a lookup table, a rule set, or a decision tree, etc. Its purpose is to store the preset parameter adjustment strategies; the grinding state interval refers to the division area in the two-dimensional space composed of material hardness information and grinding stability information. It can be implemented by division methods such as a rectangular area, a polygonal area, or an area divided based on a clustering algorithm, etc. Its purpose is to discretize the continuous grinding state and associate it with specific parameter adjustment strategies; the parameter adjustment amount refers to the parameter adjustment value that needs to be executed under a specific grinding state interval, including the feed speed adjustment amount and the grinding depth adjustment amount. Its purpose is to guide the actual adjustment of machining parameters; the grinding state refers to the current machining state jointly determined by material hardness information and grinding stability information. Its purpose is to comprehensively reflect the current grinding conditions. It can be represented in the form of a two-dimensional coordinate point and can be mapped to a two-dimensional plane for easy interval division and searching.

[0104] More specifically, the solution of the present application obtains the grinding force signal and the vibration signal in real time, and through processing, obtains the material hardness information reflecting the material characteristics and the grinding stability information reflecting the smoothness of the machining process. Based on these two types of information, a grinding state in the form of a two-dimensional coordinate is constructed, quantifying the complex machining conditions and mapping them onto a two-dimensional plane. By analyzing the area where the current grinding state is located, the system can quickly and accurately extract the feed speed adjustment amount and the grinding depth adjustment amount applicable to the current working conditions from the strategy library. This method discretizes the continuously changing grinding state into different intervals and presets optimized parameter adjustment strategies for each interval, thus realizing the refined and dynamic adjustment of machining parameters.

[0105] In some preferred embodiments, the method further includes a step executed between step S1 and step S2:

[0106] SA. Filter the grinding force signal and the vibration signal.

[0107] Specifically, the filtering process refers to removing the unwanted frequency components or noise from the original signal and retaining the signal components reflecting useful information. Its purpose is to improve the signal-to-noise ratio of the signal and make the subsequent analysis based on the signal more accurate and reliable.

[0108] More specifically, after collecting the original grinding force signal and vibration signal, the solution of this application does not directly perform feature extraction, but first performs filtering processing, which solves the noise problem in the original signal and improves the signal quality. The high-quality signal makes the subsequent extracted grinding feature information and vibration feature information more accurate. The material hardness information and grinding stability information generated based on the accurate feature information can more truly reflect the current grinding state. Therefore, the parameter adjustment amount generated according to these accurate state information is more reasonable and accurate, thus realizing the effective optimization of the crankshaft grinding process and improving the stability and efficiency of the processing.

[0109] In some preferred embodiments, step SA includes:

[0110] SA1. Perform wavelet transform on the grinding force signal, decompose it into signal components of multiple frequency bands, and according to the preset main frequency range of the grinding force signal, select the signal components corresponding to the frequency band containing the main grinding information, and reconstruct them into the filtered grinding force signal;

[0111] SA2. Perform empirical mode decomposition on the vibration signal to obtain multiple intrinsic mode function components, calculate the energy proportion of each intrinsic mode function component, select the intrinsic mode function components with an energy proportion higher than the preset energy threshold, and reconstruct them into the filtered vibration signal.

[0112] Specifically, wavelet transform refers to a time-frequency analysis method that can decompose a signal into different frequency scales. It can be implemented by discrete wavelet transform, and its purpose is to perform local analysis of the signal in both the time domain and the frequency domain; decomposing into signal components of multiple frequency bands means the signal components in different frequency ranges obtained after wavelet transform, which can be specifically obtained through a wavelet decomposition algorithm, and its purpose is to separate the features of the signal at different frequencies; the preset main frequency range of the grinding force signal refers to the frequency interval containing the main grinding energy determined according to the physical characteristics and experience of the grinding process, which can be specifically determined by performing spectrum analysis on typical grinding force signals, and its purpose is to focus on the frequency information directly related to the grinding process; the signal components corresponding to the frequency band containing the main grinding information refer to those frequency band components that fall within the preset main frequency range of the grinding force signal, and its purpose is to retain the signal components that have an important impact on the grinding process; reconstruction means synthesizing the selected signal components into the original signal through inverse wavelet transform, and its purpose is to restore the time-domain waveform of the filtered signal.

[0113] More specifically, empirical mode decomposition refers to an adaptive signal decomposition method that decomposes a signal into a series of intrinsic mode functions (IMFs) and a residual component, which can be implemented using Hilbert-Huang transform. The purpose is to decompose the signal into components of different time scales according to the characteristics of the signal itself; the intrinsic mode function component refers to the component with local characteristic scales obtained by empirical mode decomposition. Each IMF represents the oscillation mode of the signal at different time scales, and its purpose is to reveal the internal fluctuation law of the signal at different time scales; the energy proportion refers to the proportion of the energy of each intrinsic mode function component in the total signal energy, reflecting the importance of this component. Specifically, it can be obtained by calculating the sum of squares of each IMF component and dividing it by the sum of squares of the total signal. The purpose is to quantify the contribution degree of each IMF component to the total signal; the preset energy threshold refers to the standard for screening important intrinsic mode function components. Components above this threshold are considered to contain the main signal information. Specifically, a percentage value can be set according to experience or experiment. The purpose is to distinguish signal components and noise components.

[0114] More specifically, for the grinding force signal, wavelet transform can decompose the signal into different frequency scales. By selecting the signal components within the preset main frequency range of the grinding force signal and reconstructing, the key information related to the grinding process can be effectively retained, while high-frequency noise and low-frequency drift are filtered out, which avoids the loss of useful information caused by traditional filtering. For the vibration signal, empirical mode decomposition is an adaptive method that can decompose the intrinsic mode function components according to the characteristics of the signal itself. By calculating the energy proportion of each intrinsic mode function component and setting a preset energy threshold, the intrinsic mode function components containing the main vibration mode can be identified and retained, effectively removing random noise and interference components, and improving the signal-to-noise ratio of the signal.

[0115] More specifically, the quality of the grinding force signal and the vibration signal processed by the above filtering is significantly improved, providing more accurate data input for obtaining grinding characteristic information based on the grinding force signal and obtaining vibration characteristic information based on the vibration signal subsequently.

[0116] In some preferred embodiments, the method further includes steps performed after step S5:

[0117] S6. Obtain the usage duration of the grinding wheel, and determine the feed speed correction factor and / or the grinding depth correction factor according to the usage duration and a preset correction mapping table. The correction mapping table includes a plurality of feed speed correction factors and / or grinding depth correction factors that match different usage duration intervals;

[0118] S7. Use the feed speed correction factor to compensate and adjust the feed speed adjustment amount, and / or use the grinding depth correction factor to compensate and adjust the grinding depth adjustment amount.

[0119] Specifically, the usage duration of the grinding wheel refers to the cumulative working time of the grinding wheel since it was first put into use, which is used to indirectly quantify the degree of wear of the grinding wheel.

[0120] More specifically, the preset correction mapping table refers to a data structure that stores the corresponding relationships between different usage duration intervals of the grinding wheel and the corresponding feed rate correction factors and / or grinding depth correction factors. It can be implemented using a table, a database, or a functional relationship, and its purpose is to provide a basis for determining the parameter correction amount according to the usage duration of the grinding wheel.

[0121] More specifically, the feed rate correction factor and the grinding depth correction factor refer to the values used to compensate for the feed rate adjustment amount and the grinding depth adjustment amount. They can be implemented using a proportional factor or an offset, and their purpose is to offset the influence of the grinding wheel wear on the machining effect.

[0122] More specifically, the usage duration interval refers to dividing the entire service life of the grinding wheel into several continuous or discrete time periods or machining quantity segments. It can be implemented using equal-interval division or division based on the characteristics of the wear curve, and its purpose is to discretize the continuous usage duration for easy lookup of the corresponding correction factors in the correction mapping table.

[0123] More specifically, the compensation adjustment refers to using the determined feed rate correction factor and / or grinding depth correction factor to perform correction calculations on the original feed rate adjustment amount and / or grinding depth adjustment amount. It can be implemented using multiplication or addition operations, and its purpose is to obtain the parameter adjustment amount finally applied to the machine tool control system.

[0124] More specifically, based on the basic dynamic adjustment method, the solution of the present application further introduces the consideration of the usage state of the grinding wheel. It obtains the usage duration of the grinding wheel and determines the corresponding feed rate correction factor and / or grinding depth correction factor according to the preset correction mapping table. These correction factors reflect the compensation requirements of the wear degree of the grinding wheel in different usage stages for the machining parameters. Using these correction factors to compensate and adjust the feed rate adjustment amount and / or grinding depth adjustment amount generated in step S5 makes the adjustment of the machining parameters more comprehensive and accurate. It can not only cope with the instantaneous changes in material hardness or grinding stability, but also compensate for the influence of long-term wear of the grinding wheel, so that the final machining parameters can more accurately reflect the current actual machining requirements. This helps to maintain stable machining accuracy and surface quality throughout the service life cycle of the grinding wheel, reduce machining abnormalities or product quality fluctuations caused by grinding wheel wear, and thus improve the stability and reliability of production.

[0125] In the second aspect, please refer to Figure 3 and Figure 4Some embodiments of the present application also provide a crankshaft processing optimization system, which is applied in the grinding process of automobile crankshafts. The system includes:

[0126] An acquisition module 201 is used to collect a grinding force signal and a vibration signal of a grinding process in real time based on a sensor component;

[0127] A feature extraction module 202 is used to obtain grinding feature information according to the grinding force signal and obtain vibration feature information according to the vibration signal;

[0128] The first calculation module 203 is used to generate material hardness information according to the grinding feature information;

[0129] A second calculation module 204, used to generate grinding stability information according to the grinding characteristic information and the vibration characteristic information;

[0130] The strategy adjustment module 205 is used to generate parameter adjustment amounts based on a preset strategy library and according to material hardness information and grinding stability information. The parameter adjustment amounts include feed speed adjustment amounts and / or grinding depth adjustment amounts.

[0131] The system of the present application converts the grinding force signal and vibration signal collected in real time by the sensor into quantitative information reflecting the hardness of the material and the grinding stability, and dynamically generates parameter adjustment according to the real-time information based on the preset strategy library; therefore, the system of the present application can sense the grinding force and vibration state in the grinding process in real time, and evaluate the hardness of the workpiece material and the stability of the grinding process based on this information and adjust the processing parameters, thereby, the processing process can better adapt to the fluctuations in the material properties of the crankshaft blank, appropriately reduce the parameters when the material hardness is high to ensure the processing stability and reduce the tool wear, and appropriately increase the parameters when the material hardness is low to improve the material removal efficiency, thereby solving the problem that the traditional fixed process parameters are difficult to adapt to the fluctuations in material properties, and achieving the effect of taking into account both processing efficiency and stability.

[0132] In some preferred embodiments, the system further comprises:

[0133] The filtering module 206 is used to filter the grinding force signal and the vibration signal.

[0134] In some preferred embodiments, the system further comprises:

[0135] A correction compensation module 207 is configured to obtain the usage duration of the grinding wheel, determine a feed speed correction factor and / or a grinding depth correction factor according to the usage duration and a preset correction mapping table. The correction mapping table includes a plurality of feed speed correction factors and / or grinding depth correction factors that match different usage duration intervals, and is used to compensate and adjust the feed speed adjustment amount by using the feed speed correction factor, and / or compensate and adjust the grinding depth adjustment amount by using the grinding depth correction factor.

[0136] In addition, a unit described as a separate component may or may not be physically separated, and a component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or may be distributed to 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.

[0137] Furthermore, in each embodiment of the present application, the functional modules may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.

[0138] In this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0139] The above description is only for the embodiments of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An optimized method for crankshaft machining, which is applied to the grinding process of automotive crankshafts, is characterized in that The method includes the following steps: S1. Based on the sensor component, collect the grinding force signal and vibration signal during the grinding process in real time; S2. Obtain grinding feature information according to the grinding force signal, and obtain vibration feature information according to the vibration signal; S3. Generate material hardness information according to the grinding feature information; S4. Generate grinding stability information according to the grinding feature information and the vibration feature information; S5. Based on a preset strategy library, generate a parameter adjustment amount according to the material hardness information and the grinding stability information, where the parameter adjustment amount includes a feed speed adjustment amount and / or a grinding depth adjustment amount.

2. The optimized method for crankshaft machining according to claim 1, wherein, Step S2 includes: S21. Based on a preset time window, intercept the latest grinding force signal and vibration signal; S22. Calculate and obtain the average grinding force value, grinding force variance value, and grinding force kurtosis value according to the intercepted grinding force signal as the grinding feature information; S23. Calculate and obtain the vibration peak factor and vibration frequency according to the intercepted vibration signal as the vibration feature information, where the vibration peak factor is the ratio of the peak value of the vibration signal to the root mean square value.

3. The crankshaft machining optimization method according to claim 1, characterized in that Step S3 includes: S31. Based on a pre-trained hardness prediction model, use the grinding feature information as input and output the material hardness prediction value at the current moment; S32. Use a first-order lag filtering algorithm to smooth the material hardness prediction value at the current moment based on the material hardness information at the previous moment, and use the smoothed material hardness prediction value as the material hardness information at the current moment.

4. The crankshaft machining optimization method according to claim 2, wherein, Step S4 includes: S41. Calculate a first grinding stability index according to the average grinding force value and grinding force variance value in the grinding feature information, where the first grinding stability index is the ratio of the grinding force variance value to the average grinding force value; S42. Calculate a second grinding stability index according to the vibration peak factor in the vibration feature information, where the second grinding stability index is the difference between the vibration peak factor and a preset vibration threshold; S43. Calculate a third grinding stability index according to the vibration frequency in the vibration feature information, where the third grinding stability index is the ratio of the absolute value of the difference between the vibration frequency and a preset frequency to the preset frequency; S44. Based on a preset weight coefficient, perform weighted summation on the first grinding stability index, the second grinding stability index, and the third grinding stability index to obtain the grinding stability information.

5. The crankshaft machining optimization method according to claim 4, wherein The preset vibration threshold is set based on the mean and standard deviation of historical peak factors.

6. The optimized method for crankshaft machining according to claim 4, wherein The preset frequency is set according to the resonance peak of the natural frequency of the machine tool-workpiece-grinding wheel combination obtained through modal testing or finite element simulation.

7. The crankshaft machining optimization method according to claim 1, characterized in that, The strategy library includes the mapping relationships between multiple grinding state intervals and parameter adjustment amounts; Step S5 includes: S51. Establish a grinding state according to the material hardness information and the grinding stability information; S52. Analyze the grinding state interval where the grinding state is located, and extract the corresponding parameter adjustment amount in combination with the strategy library.

8. The optimized method for crankshaft machining according to claim 1, wherein The method further includes a step executed between step S1 and step S2: SA. Perform filtering processing on the grinding force signal and the vibration signal.

9. The crankshaft machining optimization method according to claim 1, characterized in that The method further includes steps performed after step S5: S6. Obtain the usage duration of the grinding wheel, and determine a feed speed correction factor and / or a grinding depth correction factor according to the usage duration and a preset correction mapping table, where the correction mapping table includes a plurality of feed speed correction factors and / or grinding depth correction factors that match different usage duration intervals; S7. Use the feed speed correction factor to compensatorily adjust the feed speed adjustment amount, and / or use the grinding depth correction factor to compensatorily adjust the grinding depth adjustment amount.

10. A crankshaft machining optimization system is applied to the grinding process of automotive crankshafts, and is characterized in that, The system includes: An acquisition module, configured to collect a grinding force signal and a vibration signal during the grinding process in real time based on a sensor assembly; A feature extraction module, configured to obtain grinding feature information according to the grinding force signal and obtain vibration feature information according to the vibration signal; A first calculation module, configured to generate material hardness information according to the grinding feature information; A second calculation module, configured to generate grinding stability information according to the grinding feature information and the vibration feature information; A strategy adjustment module, configured to generate a parameter adjustment amount based on a preset strategy library according to the material hardness information and the grinding stability information, where the parameter adjustment amount includes a feed speed adjustment amount and / or a grinding depth adjustment amount.

Images