Ball screw pair state evaluation method and device and computer readable storage medium

Abstract

This disclosure relates to the field of CNC machine tool technology, and particularly to a method, device, computer-readable storage medium, and computer program product for assessing the condition of a ball screw pair. The assessment method includes: acquiring condition data of the ball screw pair; classifying the condition data using a classification model to obtain a classification result, the classification result reflecting the classification confidence level of a specified condition category; converting the classification result into a health index, the health index conforming to the nonlinear degradation law of the ball screw pair; and assessing the condition of the ball screw pair based on the health index.

Description

Technical Field

[0001] This disclosure relates to the field of CNC machine tool technology, and in particular to a condition assessment method, condition assessment device, computer-readable storage medium, and computer program product for ball screw pairs. Background Technology

[0002] With the rapid development of the high-end machinery and equipment manufacturing industry, ball screw pairs have become key functional components of mechanical transmission and actuators due to their outstanding advantages such as high transmission efficiency, strong resistance to environmental loads, and excellent adaptability to working conditions. As the core actuator of a machining center, its operating accuracy and health status directly determine the machining accuracy, operational stability, and reliability of the entire machine. Summary of the Invention

[0003] In related technologies, the assessment of the condition of ball screw pairs mainly relies on manual inspection and alarm rules based on fixed thresholds, which has the problem of low reliability.

[0004] In view of this, the present disclosure provides a method for evaluating the condition of ball screw pairs, which can effectively solve the above problems.

[0005] According to one aspect of this disclosure, a method for assessing the condition of a ball screw pair is provided, comprising: acquiring condition data of the ball screw pair; classifying the condition data using a classification model to obtain a classification result, the classification result reflecting the classification confidence of a specified condition category; converting the classification result into a health index, the health index conforming to the nonlinear degradation law of the ball screw pair; and assessing the condition of the ball screw pair based on the health index.

[0006] In some embodiments, converting the classification result into a health index includes: mapping the classification result to the health index using a mapping function, wherein the mapping function is monotonically increasing and the domain of the mapping function includes a first interval and a second interval, wherein the value of the independent variable in the first interval is greater than the value of the independent variable in the second interval, and the slope of the mapping function in the first interval is less than the slope of the mapping function in the second interval.

[0007] In some embodiments, the mapping function is a cumulative distribution function of an asymmetric beta distribution, and the mapping function includes: a first shape parameter, which controls the curve shape of the mapping function in the first interval and is less than 1; and a second shape parameter, which controls the curve shape of the mapping function in the second interval and is greater than 1.

[0008] In some embodiments, the first shape parameter and the second shape parameter are obtained by fitting based on the historical degradation data of the ball screw pair using the maximum likelihood method.

[0009] In some embodiments, obtaining the state data of the ball screw pair includes: obtaining the real-time rotational speed of the screw in the ball screw pair; determining whether the ball screw pair is in a steady-state operating state based on the relationship between the real-time rotational speed and the theoretical rotational speed of the screw; and obtaining the state data of the ball screw pair in the steady-state operating state.

[0010] In some embodiments, determining whether the ball screw assembly is in a steady-state operating state based on the real-time rotational speed and the theoretical rotational speed of the screw includes: determining that the ball screw assembly is in a steady-state operating state in response to the difference between the real-time rotational speed and the theoretical rotational speed being less than or equal to a first threshold.

[0011] In some embodiments, the theoretical rotational speed is determined based on the structure of the lead screw and the linear feed motion target of the ball screw pair.

[0012] In some embodiments, the state data includes at least one of vibration signals, temperature signals, and current signals.

[0013] In some embodiments, assessing the state of the ball screw assembly based on the health index includes: determining that the ball screw assembly is in a healthy state in response to the health index being greater than or equal to a second threshold; determining that the ball screw assembly is in a warning state in response to the health index being less than the second threshold and greater than or equal to a third threshold; and determining that the ball screw assembly is in a dangerous state in response to the health index being less than the third threshold.

[0014] In some embodiments, the status assessment method further includes: performing time-series smoothing on the health index using an exponentially weighted moving average.

[0015] In some embodiments, the classification model is a support vector machine, a random forest, or a long short-term memory network.

[0016] According to another aspect of this disclosure, a state assessment device for a ball screw pair is provided, comprising: an acquisition module configured to acquire state data of the ball screw pair; a classification module configured to classify the state data using a classification model to obtain a classification result, the classification result reflecting the classification confidence of a specified state category; a conversion module configured to convert the classification result into a health index, the health index conforming to the nonlinear degradation law of the ball screw pair; and an assessment module configured to assess the state of the ball screw pair based on the health index.

[0017] According to another aspect of this disclosure, a condition assessment apparatus for a ball screw pair is provided, comprising: at least one memory; and at least one processor coupled to the at least one memory, the at least one processor being configured to execute a condition assessment method as described in any embodiment of this application based on instructions stored in the at least one memory.

[0018] According to another aspect of this disclosure, a computer-readable storage medium is provided that stores computer instructions thereon, which, when executed by a processor, implement the state assessment method as described in any embodiment of this application.

[0019] According to other aspects of this disclosure, a computer program product is provided that, when run on a computer, causes the computer to implement the state assessment method as described in any embodiment of this application. Attached Figure Description

[0020] The accompanying drawings, which form part of this specification, illustrate embodiments of this disclosure and, together with the specification, serve to explain the principles of this disclosure.

[0021] This disclosure will become clearer with reference to the accompanying drawings and the following detailed description, wherein:

[0022] Figure 1 This is a schematic flowchart illustrating a state assessment method according to some embodiments of the present disclosure;

[0023] Figure 2 This is a schematic diagram illustrating the process of obtaining status data according to some embodiments of the present disclosure;

[0024] Figure 3 This is a function graph illustrating the mapping function according to some embodiments of the present disclosure;

[0025] Figure 4 This is a function graph illustrating the mapping function according to other embodiments of the present disclosure;

[0026] Figure 5 This is a schematic flowchart illustrating a state assessment method according to other embodiments of the present disclosure;

[0027] Figure 6 This is a block diagram illustrating a state assessment apparatus according to some embodiments of the present disclosure;

[0028] Figure 7 This is a block diagram illustrating a state assessment apparatus according to other embodiments of the present disclosure;

[0029] Figure 8 This is a block diagram illustrating a computer system for implementing some embodiments of the present disclosure.

[0030] It should be understood that the dimensions of the various parts shown in the accompanying drawings are not drawn to actual scale. Furthermore, the same or similar reference numerals denote the same or similar components. Detailed Implementation

[0031] Various embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. The descriptions of the embodiments are merely illustrative and are in no way intended to limit the scope of the disclosure or its application or use. The present disclosure may be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided so that this disclosure will be thorough and complete, and will fully express the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specifically stated, the relative arrangement of components and steps set forth in these embodiments should be interpreted as merely illustrative and not as limiting.

[0032] The terms “first,” “second,” and similar words used in this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different parts. Words such as “including” mean that the element preceding the word covers the element listed after the word, and do not exclude the possibility of covering other elements as well.

[0033] It should also be understood that any component, data or structure mentioned in the embodiments of this disclosure can generally be understood as one or more unless expressly defined or given to the contrary in the context.

[0034] All terms used in this disclosure (including technical or scientific terms) have the same meaning as understood by one of ordinary skill in the art to which this disclosure pertains, unless otherwise specifically defined. It should also be understood that terms defined in a general dictionary, such as a dictionary, should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and not as having an idealized or highly formalized meaning, unless expressly defined herein.

[0035] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0036] Traditional methods for assessing the condition of ball screw pairs mainly rely on manual inspections and alarm rules based on fixed thresholds, which results in low reliability of the assessment results.

[0037] In view of this, this disclosure proposes a condition assessment method for ball screw pairs, which can improve the reliability of condition assessment.

[0038] First, combined Figure 1 The state assessment method in this disclosure is described.

[0039] Figure 1This is a schematic flowchart illustrating a state assessment method according to some embodiments of the present disclosure.

[0040] like Figure 1 As shown, the state assessment method includes: step S1, acquiring the state data of the ball screw pair; step S3, classifying the state data using a classification model to obtain a classification result, the classification result reflecting the classification confidence of a specified state category; step S5, converting the classification result into a health index, the health index conforming to the nonlinear degradation law of the ball screw pair; and step S7, assessing the state of the ball screw pair based on the health index.

[0041] In step S1, the state data of the ball screw assembly can be acquired using sensors. The state data may include at least one of vibration signals, temperature signals, and current signals. For example, an accelerometer can be used to acquire vibration signals of the ball screw assembly and its supporting components, a temperature sensor can be used to acquire temperature signals of the nut and bearing parts of the ball screw assembly, and a current acquisition module can be used to acquire current signals from the servo motor used to drive the ball screw assembly in the CNC system, thereby achieving multi-dimensional state data acquisition of the ball screw assembly.

[0042] In some embodiments, the division of operating states can be used to avoid invalid data in non-steady-state operating states interfering with subsequent state evaluation. The following will combine... Figure 2 Provide a detailed description.

[0043] Figure 2 This is a schematic diagram illustrating the process of obtaining status data according to some embodiments of the present disclosure.

[0044] like Figure 2 As shown, step S1 includes: step S11, obtaining the real-time rotational speed of the screw in the ball screw pair; step S12, determining whether the ball screw pair is in a steady-state operating state based on the relationship between the real-time rotational speed and the theoretical rotational speed of the screw; step S13, obtaining the state data of the ball screw pair in the steady-state operating state.

[0045] In step S11, the real-time rotational speed of the lead screw can be obtained by converting the voltage signal acquired by the speed sensor. For example, the real-time rotational speed of the lead screw can be calculated based on the ratio of the number of pulses per unit time of the speed sensor to the fixed number of pulses output per revolution.

[0046] In step S12, the operating state of the ball screw pair is divided based on the collected real-time rotational speed and the calculated theoretical rotational speed.

[0047] The theoretical speed can be determined based on the screw structure and the linear feed motion target of the ball screw pair. In some embodiments, the theoretical speed of the screw can be determined based on the screw lead and the feed rate command of the CNC system. For example, the ratio of the CNC system's feed rate command to the screw lead can be used as the theoretical speed of the screw.

[0048] There may be an inherent measurement deviation between the real-time rotational speed and the theoretical rotational speed of the ball screw. Therefore, to avoid misjudging the operating state, a tolerance bandwidth can be set based on the theoretical rotational speed. When the real-time rotational speed is close to the theoretical rotational speed, it indicates that the speed fluctuation is small and the operation is smooth and uniform, so it can be determined as a steady-state operating state. For example, in response to the difference between the real-time rotational speed and the theoretical rotational speed being less than or equal to a first threshold, the ball screw assembly is determined to be in a steady-state operating state. In response to the difference being greater than the first threshold, the ball screw assembly is determined to be in a non-steady-state operating state.

[0049] It should be understood that the first threshold can be set according to the actual situation, and this disclosure does not impose any restrictions on it. For example, the first threshold can be 5 r / min.

[0050] By using a speed sensor to obtain the real-time speed of the lead screw and classifying the operating state, the accuracy of the operating state classification can be improved.

[0051] It should be understood that, in addition to obtaining the screw speed data through a speed sensor, the operating state can also be divided in other ways. For example, the operating state can be divided by directly reading the axis enable signal or feed speed signal from the CNC system, without the need to install a speed sensor. This disclosure does not limit the specific method of dividing the operating state, as long as it can distinguish between the steady-state and unsteady-state operating states of the ball screw pair.

[0052] In step S13, only the state data under steady-state operation is collected, and the state data under non-steady-state operation is no longer included in the subsequent state evaluation.

[0053] Since condition data collected under non-steady-state operating conditions is often accompanied by speed fluctuations, load abrupt changes, or impact disturbances, it is prone to introducing abnormal characteristics and noise. Therefore, by distinguishing between steady-state and non-steady-state operating conditions, the interference of invalid data on condition assessment can be eliminated from the source, thereby improving the reliability of condition assessment.

[0054] Back Figure 1 After obtaining the status data of the ball screw pair, subsequent status recognition operations are performed on the data.

[0055] In step S3, the state data is classified into normal or abnormal states using a classification model to obtain a classification result that reflects the classification confidence level of the normal or abnormal category.

[0056] The classification model can be a Support Vector Machine (SVM), a Random Forest, or a Long Short-Term Memory (LSTM) network. SVMs are lightweight and efficient, meeting the requirements of online real-time computation and are easy to deploy in embedded systems in industrial settings. Random Forests are insensitive to feature dimensions and have strong noise resistance. LSM networks can directly perform end-to-end classification of raw sensor time series, eliminating the need for manual feature extraction. In practical applications, other suitable classification models can also be selected according to requirements; this disclosure does not impose any restrictions on this.

[0057] The following section uses Support Vector Machine (SVM) as an example to provide a detailed explanation of the overall process of state classification.

[0058] Before classifying states, the state data can be preprocessed. Preprocessing operations can include wavelet thresholding for noise reduction, feature extraction, and standardization.

[0059] First, threshold denoising is performed on the state data of the ball screw assembly. For example, a 4th-order Daubechies (db4) wavelet basis can be used for threshold denoising. The wavelet basis is the basic operation function of wavelet decomposition, which can be adapted to the extraction of weak feature signals in mechanical systems. During the denoising process, the state data can be decomposed into three levels of wavelet coefficients of different frequencies using the wavelet basis, separating the effective feature signals from irrelevant interference noise. Then, soft thresholding is performed on the wavelet coefficients obtained from each level of decomposition. By setting a reasonable threshold, small-amplitude noise coefficients below the threshold are directly reduced to zero, while effective signal coefficients above the threshold are smoothly shrunk. This can gently compress and filter out small-amplitude noise coefficients while retaining effective features and avoiding abrupt data distortion.

[0060] After denoising, feature extraction is performed on the state data, extracting time-domain features, frequency-domain features, and time-frequency-domain features respectively. These features are then combined to form a feature vector for state recognition. Time-domain features may include root mean square value, peak factor, kurtosis, and skewness; frequency-domain features may include frequency center, mean square frequency, and frequency variance; and time-frequency-domain features may include wavelet packet energy entropy.

[0061] The resulting feature vectors are then standardized, for example, by using standard score normalization to map each feature dimension to a standard distribution with a mean of 0 and a standard deviation of 1. During the training phase of the support vector machine (SVM) model, the above preprocessing procedure can be used to preprocess historical normal state data and historical abnormal state data containing typical fault modes such as wear, pitting, and lubrication failure. The preprocessed data is then used for supervised training of the SVM model. During training, the model can use a radial basis function kernel, and cross-validation can be used to optimize the model's regularization parameters and kernel bandwidth parameters to determine the optimal parameter combination.

[0062] When using a trained support vector machine (SVM) model for state classification, the state data is first preprocessed as described above, and then input into the trained SVM model to obtain state classification labels and decision function values. For example, if normal states are labeled as positive categories and abnormal states as negative categories during the training phase, the output decision function value is positively correlated with the classification confidence of the normal state category. In this case, the classification result can be used to reflect the classification confidence of the normal state category.

[0063] By using a data-driven approach to build a classification model, it can autonomously learn the nonlinear decision boundary between the normal and abnormal states of the ball screw pair. Compared with the traditional state recognition method based on fixed thresholds, it has stronger generalization ability and higher sensitivity in recognizing early slight degradation states.

[0064] In step S5, the classification result output by the classification model is converted into a health index. The classification result, for example, is the decision function value output by the support vector machine model. The converted health index conforms to the nonlinear degradation pattern of the ball screw assembly: "stable operation, rapid decline."

[0065] In some embodiments, the classification result is mapped to the health index by a mapping function. The mapping function is monotonically increasing, and the domain of the mapping function includes a first interval and a second interval. The value of the independent variable in the first interval is greater than the value of the independent variable in the second interval, and the slope of the mapping function in the first interval is less than the slope of the mapping function in the second interval.

[0066] The mapping function curve can reflect the nonlinear degradation law of the ball screw pair as a whole. The first interval is the normal region reflecting the "stable operation" stage, in which the function curve changes gently; the second interval is the degradation region reflecting the "rapid decline" stage, in which the function curve shows a steep downward trend.

[0067] For example, the mapping function is the cumulative distribution function of an asymmetric beta distribution, and the mapping function includes: a first shape parameter, which controls the curve shape of the mapping function in the first interval and is less than 1; and a second shape parameter, which controls the curve shape of the mapping function in the second interval and is greater than 1.

[0068] The decision function value output by a support vector machine (SVM) may not conform to the [0,1] domain required by the cumulative distribution function of an asymmetric beta distribution. Direct input may lead to mapping failure or meaningless results. Therefore, if SVM is used for state classification, normalization is required before mapping. For example, the 95th percentile of the decision function value of the normal state category can be used as the upper bound of the health baseline. The 5th percentile of the decision function value for the abnormal state category is used as the lower bound for degradation. This ensures that 95% of the decision function values ​​for the normal state category are less than or equal to... 95% of the decision function values ​​for the abnormal state category are greater than or equal to The decision function value can be calculated using the following formula. Perform normalization calculations to obtain the raw score. :

[0069]

[0070] The clip function is used to limit the calculation results so that the normalized values ​​all fall within the [0,1] interval.

[0071] After normalization, the raw score is obtained. A health index is mapped from the cumulative distribution function of the asymmetric beta distribution. The calculation formula is as follows:

[0072]

[0073] in, is the beta function, which is the normalization constant of the beta distribution. It is used to standardize the integral result so that the output range of the cumulative distribution function is constrained to the interval [0,1]. These are dummy variables for integration, i.e., temporary variables within the integral, used for continuously iterating through 0 to... For each value between them, the probability density is accumulated point by point; The first shape parameter controls the curve shape of the mapping function on the side where the independent variable approaches 1; This is the second shape parameter, used to control the curve shape of the mapping function on the side where the independent variable approaches 0. It is set by... <1 and >1, which makes the function curve conform to the nonlinear degradation law of "stable operation and rapid decline" of the ball screw pair, and at the same time gives it a higher resolution for distinguishing the normal operation stage.

[0074] For example, the first shape parameter The second shape parameter is 0.5. The value is 3. The mapping function curve in this case is as follows: Figure 3 As shown. Figure 3 This is a function graph illustrating the mapping function according to some embodiments of the present disclosure.

[0075] Depend on Figure 3It can be seen that in the first interval, such as the interval [0.2,1], the curve shape remains flat, corresponding to the "stable operation" stage; while in the second interval, such as the interval [0,0.2), the curve shape is relatively steep, corresponding to the "rapid decline" stage.

[0076] It should be understood that the first shape parameter Second shape parameter Settings can be configured according to actual circumstances; this disclosure does not impose any restrictions on them.

[0077] In some embodiments, provided that historical full-life degradation data of the ball screw pair is available, the first shape parameter and the second shape parameter It is obtained by fitting the data using the maximum likelihood method based on the historical degradation data of the ball screw pair.

[0078] The function curve obtained by fitting historical degradation data can more accurately reflect the nonlinear degradation mode of the ball screw pair, thereby further improving the reliability of the condition assessment.

[0079] Even without historical full-life degradation data, the first shape parameter can be determined based on the prior degradation law of ball screw pairs: "slow degradation during normal operation, rapid decline before failure." Second shape parameter Perform manual settings and iterative corrections.

[0080] Besides using the cumulative distribution function of the asymmetric beta distribution as the mapping function, other methods can be used to transform the classification results into a health index.

[0081] For example, a piecewise linear function can be used as the mapping function. By setting a knee point, the function curve can be divided into a "normal segment" and a "degenerate segment," each assigned a different mapping slope. An example of a mapping function curve would be... Figure 4 As shown. Figure 4 This is a function graph illustrating the mapping function according to other embodiments of the present disclosure.

[0082] like Figure 4 As shown, intervals can be divided by setting a knee point where the curve's slope changes drastically. In the first interval, such as the interval [0.2, 1] after the knee point, the curve's slope is smaller, corresponding to the "stable operation" phase; in the second interval, such as the interval [0, 0.2) before the knee point, the curve's slope is larger, corresponding to the "rapid decline" phase. It should be understood that... Figure 4 The shape of the function is shown only schematically, and no restrictions are placed on the specific values ​​of the slope.

[0083] By mapping discrete classification results to a continuous function that conforms to the degradation law of ball screw pairs, the degree of health degradation can be quantitatively described, providing a refined basis for predictive maintenance.

[0084] Besides constructing health scores using nonlinear mapping functions, health assessment can also be achieved by combining exponential models with neural networks. For example, regression exponents can be used to extract features of the slow degradation trend of equipment, which can then be fused with deep features learned by autoencoders. The fused features are then input into a neural network to output a continuous health score. Another example is the use of self-organizing map networks for unsupervised learning of normal state data to construct a benchmark model that characterizes the normal operating state of equipment. By calculating the minimum quantization error of the samples, the deviation between the current state and the normal benchmark can be measured, thus achieving a quantitative description of health. The entire process does not require pre-setting the physical degradation laws of the equipment.

[0085] In some embodiments, the status assessment process further includes time-series smoothing of the health index, as described below. Figure 5 Detailed description.

[0086] Figure 5 This is a schematic flowchart illustrating a state assessment method according to other embodiments of the present disclosure. Figure 5 and Figure 1 The difference is that it also includes step S6. The following only introduces the differences, and the similarities will not be repeated.

[0087] like Figure 5 As shown, the status assessment method further includes: step S6, performing time-series smoothing processing on the health index using an exponentially weighted moving average.

[0088] In step S6, to suppress health index fluctuations caused by noise from a single measurement, the health index can be adjusted using the following formula. Perform time-series smoothing to obtain the smoothed health index. :

[0089]

[0090] in, This serves as a time-series sampling time identifier; For smoothing coefficients, The value range is (0,1], and the specific value can be configured by the user according to the real-time requirements. This disclosure does not impose any restrictions on this.

[0091] By performing time-series smoothing on the health index, the fluctuations in the health index caused by measurement noise can be eliminated, further improving the reliability of the status assessment.

[0092] Back Figure 1After obtaining the health index, the condition of the ball screw pair can be assessed.

[0093] In step S7, the health status of the ball screw assembly can be graded and assessed based on the health index.

[0094] In some embodiments, as described above, the classification results reflecting the confidence level of the normal state have been mapped to a monotonically increasing function to obtain the corresponding health index. The more normal the state data, the larger the corresponding health index, that is, the larger the health index, the healthier the ball screw pair. Therefore, the state of the ball screw pair can be determined based on the threshold range of the health index. For example, in response to the health index being greater than or equal to a second threshold, the ball screw pair is determined to be in a healthy state; in response to the health index being less than the second threshold but greater than or equal to a third threshold, the ball screw pair is determined to be in a warning state; in response to the health index being less than the third threshold, the ball screw pair is determined to be in a dangerous state.

[0095] It should be understood that the specific values ​​of the second and third thresholds can be set according to the actual situation. For example, the second threshold can be 0.7 and the third threshold can be 0.4. This disclosure does not impose any restrictions on this.

[0096] Assessing the health status of ball screw pairs using a health index can improve the resolution of different health statuses, thereby enhancing the reliability of the status assessment.

[0097] The following is for reference. Figure 6 and Figure 7 This invention describes a state assessment apparatus according to embodiments of the present disclosure, used to perform any of the embodiments of the state assessment method described above. Figure 6 This is a block diagram illustrating a state assessment apparatus according to some embodiments of the present disclosure.

[0098] like Figure 6 As shown, the ball screw pair state assessment device 6 includes: an acquisition module 61 configured to acquire state data of the ball screw pair; a classification module 62 configured to classify the state data using a classification model to obtain a classification result, the classification result reflecting the classification confidence of a specified state category; a conversion module 63 configured to convert the classification result into a health index, the health index conforming to the nonlinear degradation law of the ball screw pair; and an assessment module 64 configured to assess the state of the ball screw pair based on the health index.

[0099] The acquisition module 61 of the condition assessment device 6 can, for example, be used to perform... Figure 1 Step S1. The classification module 62 of the state assessment device 6 can, for example, be used to perform... Figure 1Step S3. The conversion module 63 of the state assessment device 6 can, for example, be used to perform... Figure 1 Step S5. The evaluation module 64 of the state evaluation device 6 can, for example, be used to perform... Figure 1 Step S7.

[0100] It should be understood that the state assessment apparatus may also include other modules for performing other steps in the state assessment method of the embodiments of this disclosure, such as a training module, a processing module, etc.

[0101] The condition assessment apparatus of this disclosure can improve the reliability of condition assessment of ball screw pairs.

[0102] Figure 7 This is a block diagram illustrating a state assessment apparatus according to other embodiments of the present disclosure.

[0103] like Figure 7 As shown, the state assessment apparatus 7 includes: at least one memory 71; and at least one processor 72 coupled to the at least one memory 71, the at least one processor 72 being configured to execute the state assessment method as described in any of the foregoing embodiments based on instructions stored in the at least one memory 71.

[0104] Memory 71 is used to store one or more computer-readable instructions. Memory 71 may include any combination of various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory, including but not limited to random access memory, dynamic random access memory, static random access memory, read-only memory, and flash memory. Memory 71 may, for example, store operating systems, application programs, boot loaders, databases, and other programs, as well as various application programs and various data.

[0105] The processor 72 is configured to execute computer-readable instructions to implement the state evaluation method described in any of the foregoing embodiments. Specific implementations of each step of the method can be found in the above embodiments, for example... Figure 1 The steps involved are repeated here, so the details will not be repeated.

[0106] The processor 72 can be various processing devices, such as a central processing unit (CPU), a network processor, etc.; it can also be a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The CPU can be based on x86 or ARM architectures, etc.

[0107] The processor 72 and the memory 71 can communicate with each other directly or indirectly. For example, the processor 72 and the memory 71 can communicate via a network. The network can include a wireless network, a wired network, and / or any combination of wireless and wired networks. The processor 72 and the memory 71 can also communicate with each other via a system bus, which is not limited in this disclosure.

[0108] It should be noted that Figure 7 The components of the state assessment device 7 shown are merely exemplary and not limiting. Depending on the specific application requirements, the state assessment device 7 may also have other components. The processor 72 can control other components in the state assessment device 7 to perform desired functions. The state assessment device 7 can be implemented in software, firmware, and / or hardware and can be integrated into a device with the relevant application installed.

[0109] The condition assessment device disclosed herein can improve the reliability of condition assessment of ball screw pairs.

[0110] Figure 8 This is a block diagram illustrating a computer system for implementing some embodiments of the present disclosure.

[0111] like Figure 8 As shown, the computer system 8 can be represented in the form of a general computing device. The computer system 8 includes a memory 81, a processor 82, and a bus 80 connecting different system components.

[0112] The memory 81 can be various forms of computer-readable storage media, such as system memory, non-volatile storage media, etc. System memory may store, for example, an operating system, application programs, a bootloader, and other programs. System memory may include volatile storage media, such as random access memory (RAM) and / or cache memory. Non-volatile storage media may store, for example, instructions for executing corresponding embodiments of the state evaluation method. Non-volatile storage media include, but are not limited to, disk storage, optical storage, flash memory, etc.

[0113] The processor 82 can be implemented using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete hardware components such as discrete gates or transistors. Correspondingly, each module can be implemented by the central processing unit (CPU) executing instructions in memory to perform the corresponding steps, or by dedicated circuitry to execute the corresponding steps.

[0114] Bus 80 can use any of the various bus architectures. For example, bus architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, and Peripheral Component Interconnect (PCI) bus.

[0115] The computer system 8 may also include an input / output interface 83, a network interface 84, and a storage interface 85. These interfaces 83, 84, and 85, as well as the memory 81 and processor 82, can be connected via a bus 80. The input / output interface 83 provides a connection interface for input / output devices such as a monitor, mouse, and keyboard. The network interface 84 provides a connection interface for various networked devices. The storage interface 85 provides a connection interface for external storage devices such as floppy disks, USB flash drives, and SD cards.

[0116] According to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product that, when run on a computer, causes the computer to implement the state assessment method described in any of the foregoing embodiments. The computer program product includes computer instructions carried on a computer-readable medium, the computer instructions containing program code for performing the methods shown in the flowcharts.

[0117] The state assessment method of this disclosure can improve the reliability of the state assessment of ball screw pairs.

[0118] Various embodiments of this disclosure have now been described in detail. To avoid obscuring the concept of this disclosure, some details known in the art have not been described. Those skilled in the art will fully understand how to implement the technical solutions disclosed herein based on the above description.

[0119] While specific embodiments of this disclosure have been described in detail by way of examples, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments or equivalent substitutions can be made to some technical features without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. A method for assessing the condition of a ball screw assembly, comprising: Obtain the status data of the ball screw assembly; The state data is classified using a classification model to obtain classification results, which reflect the classification confidence of a specified state category. The classification results are converted into a health index, which conforms to the nonlinear degradation law of the ball screw pair. The condition of the ball screw assembly is assessed based on the health index.

2. The state assessment method according to claim 1, wherein, Converting the classification results into a health index includes: The classification result is mapped to the health index through a mapping function. The mapping function is monotonically increasing, and the domain of the mapping function includes a first interval and a second interval. The value of the independent variable in the first interval is greater than the value of the independent variable in the second interval. The slope of the mapping function in the first interval is less than the slope of the mapping function in the second interval.

3. The state assessment method according to claim 2, wherein, The mapping function is the cumulative distribution function of an asymmetric beta distribution, and the mapping function includes: The first shape parameter is used to control the curve shape of the mapping function within the first interval, and is less than 1; The second shape parameter is used to control the curve shape of the mapping function within the second interval, and is greater than 1.

4. The condition assessment method according to claim 3, wherein, The first shape parameter and the second shape parameter are obtained by fitting the historical degradation data of the ball screw pair using the maximum likelihood method.

5. The state assessment method according to claim 1, wherein, The status data of the ball screw assembly includes: Obtain the real-time rotational speed of the screw in the ball screw pair; Based on the relationship between the real-time rotational speed and the theoretical rotational speed of the ball screw, it is determined whether the ball screw pair is in a steady-state operating state. Obtain the state data of the ball screw pair under the steady-state operating condition.

6. The condition assessment method according to claim 5, wherein, Determining whether the ball screw pair is in a steady-state operating state based on the real-time rotational speed and the theoretical rotational speed of the ball screw includes: In response to the difference between the real-time rotational speed and the theoretical rotational speed being less than or equal to a first threshold, it is determined that the ball screw pair is in a steady-state operating state.

7. The condition assessment method according to claim 5, wherein, The theoretical rotational speed is determined based on the structure of the lead screw and the linear feed motion target of the ball screw pair.

8. The condition assessment method according to claim 1, wherein, The status data includes at least one of vibration signals, temperature signals, and current signals.

9. The condition assessment method according to claim 1, wherein, The condition of the ball screw assembly is assessed based on the health index, including: In response to the health index being greater than or equal to the second threshold, the ball screw assembly is determined to be in a healthy state. In response to the health index being less than the second threshold and greater than or equal to the third threshold, it is determined that the ball screw pair is in a warning state; In response to the health index being less than the third threshold, the ball screw assembly is determined to be in a dangerous state.

10. The condition assessment method according to claim 1, further comprising: The health index is smoothed over time using an exponentially weighted moving average.

11. The state assessment method according to claim 1, wherein, The classification model is a support vector machine, random forest, or long short-term memory network.

12. A condition assessment device for a ball screw assembly, comprising: The acquisition module is configured to acquire the status data of the ball screw assembly. The classification module is configured to classify the state data using a classification model to obtain a classification result, wherein the classification result reflects the classification confidence of a specified state category; The conversion module is configured to convert the classification result into a health index, which conforms to the nonlinear degradation law of the ball screw pair. The evaluation module is configured to evaluate the condition of the ball screw pair based on the health index.

13. A condition assessment device for a ball screw assembly, comprising: At least one memory; as well as At least one processor coupled to the at least one memory, the at least one processor being configured to execute the state assessment method as described in any one of claims 1 to 11 based on instructions stored in the at least one memory.

14. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the state assessment method as described in any one of claims 1 to 11.

15. A computer program product, when run on a computer, causes the computer to implement the state assessment method as described in any one of claims 1 to 11.

Images