Cutter state monitoring method, cutter state monitoring model construction method, cutter state monitoring model construction system and electronic equipment

Abstract

The invention provides a tool state monitoring method, a tool state monitoring model construction method, a tool state monitoring model construction system and electronic equipment. The model construction method comprises the steps of obtaining machining data of a cutter under a normal working condition, wherein the machining data comprises a power signal of a cutter spindle; the power signals are segmented according to the segmentation lines, and the machining time period of the cutter is determined according to the segmentation result; the parting line is used for defining a machining state and a non-machining state of the cutter; extracting characteristic indexes of the processing data in each processing time period; and determining a cutter state monitoring threshold value of each processing period according to the characteristic value of the characteristic index, and constructing a cutter state monitoring model according to the cutter state monitoring threshold value. According to the method, the influence of non-machining period interference and cross-working-condition cross coupling can be effectively eliminated, so that the constructed cutter state monitoring model has higher working condition adaptive capacity, precise evaluation and real-time monitoring of the cutter state are realized, and the accuracy and timeliness of a monitoring result are remarkably improved.

Description

Technical Field

[0001] This disclosure relates to the field of industrial manufacturing technology, and in particular to a method, system, and electronic device for tool condition monitoring and model construction. Background Technology

[0002] As the manufacturing industry moves towards high-end and intelligent manufacturing, the level of intelligence in the core machining processes of high-end CNC machine tools, as a strategic cornerstone supporting industrial upgrading, has become a key bottleneck restricting equipment performance and reliability. Therefore, it is urgent to conduct research on key technologies to improve the intelligence level of high-end CNC equipment. Among these, online tool condition monitoring technology is one of the core components for achieving intelligent perception and control of the machining process. This technology, through real-time perception and intelligent diagnosis of tool wear and breakage during machining, can effectively ensure machining quality, optimize machining efficiency, and prevent equipment damage, making it crucial for overcoming current shortcomings in intelligence.

[0003] In the research and industrial application of real-time tool condition monitoring technology, accurately and efficiently sensing and acquiring tool condition information during the machining process is the core prerequisite for achieving reliable monitoring. Existing technologies are mainly divided into two major technical paths based on differences in data acquisition methods: direct measurement methods and indirect measurement methods.

[0004] The core idea of ​​direct measurement methods is to directly observe or measure the physical morphology of the cutting tool, such as using optical imaging or contact probe measurements, to directly acquire information on the tool's wear pattern, chipping location, and degree of damage. The advantage of these methods is that the acquired status information is intuitive and the diagnostic results are clear. However, during implementation, the machining process usually needs to be paused (i.e., machine stoppage inspection) to complete static contact measurement or imaging observation of the tool. This characteristic makes it unsuitable for meeting the high real-time online response requirements of emergency events such as sudden tool breakage or tool-workpiece collisions in industrial scenarios, greatly limiting its application scenarios.

[0005] Indirect measurement methods represent the mainstream research direction and industrial application solution for real-time tool condition monitoring. Their core principle is not direct observation of the tool itself, but rather the deployment of various industrial sensors to collect real-time, online-accessible physical signals strongly correlated with the tool's condition. By analyzing the characteristic changes in these indirect signals, the operating status of the tool can be inferred and evaluated. This method enables continuous monitoring of the tool's condition without interrupting the machining process. With its core advantages of no downtime, high real-time performance, and continuous monitoring, it has become the primary data source and technological foundation for building practical intelligent tool condition monitoring systems.

[0006] However, current tool condition monitoring methods based on indirect signals still face technical bottlenecks in practical industrial applications. Specifically, the mapping relationship between tool condition and signals is complex, and there are significant nonlinear and time-varying coupling characteristics between different states such as tool wear and breakage and multiple physical quantity signals. It is difficult to establish an accurate and universal condition diagnosis model, which seriously restricts the reliability, diagnostic accuracy and universality of existing tool condition monitoring, making it difficult to achieve large-scale industrial applications. Summary of the Invention

[0007] The technical problem to be solved by this disclosure is to overcome the shortcomings of existing tool condition monitoring methods based on indirect signals, which have significant nonlinear and time-varying coupling characteristics between tool condition and physical quantity signals, making it difficult to establish an accurate and universal condition diagnosis model. This disclosure provides a method, system, and electronic device for tool condition monitoring and model construction.

[0008] This disclosure solves the above-mentioned technical problems through the following technical solution:

[0009] Firstly, a method for constructing a tool condition monitoring model is provided, including:

[0010] Acquire machining data under normal operating conditions of the tool, wherein the machining data includes the power signal of the tool spindle;

[0011] The power signal is segmented according to the dividing line, and the processing time period of the tool is determined according to the segmentation result; the dividing line is used to define the processing state and non-processing state of the tool.

[0012] Extract characteristic indicators of processing data within each processing period;

[0013] The tool status monitoring threshold for each processing period is determined based on the characteristic values ​​of the characteristic indicators, and a tool status monitoring model is constructed based on the tool status monitoring thresholds. The tool status monitoring model is used to analyze and judge the real-time processing data of the tool in order to monitor the status of the tool.

[0014] Optionally, it also includes:

[0015] The power signal is smoothed using a gradually increasing first sliding window to obtain a power sequence; each element in the power sequence corresponds to the calculation result of the first sliding window.

[0016] The power sequence is subjected to quantile calculation, and the dividing line is determined based on the calculation result.

[0017] Optionally, before the step of calculating the quantiles of the power sequence, the method further includes:

[0018] Determine whether the number of elements contained in the power sequence reaches the target number;

[0019] In response to the fact that the number of elements does not reach the target number, the number of elements in the power sequence is padded to the target number using the last element contained in the power sequence;

[0020] The step of calculating the quantiles of the power sequence includes:

[0021] Quantiles are calculated for the padded power sequence.

[0022] Optionally, feature indicators of the processing data within each processing period can be extracted, including:

[0023] The machining data within the machining period is segmented using a second sliding window, and characteristic indicators of the machining data within each second sliding window are calculated; wherein, the window length of the second sliding window is negatively correlated with the spindle speed, and / or the step size of the second sliding window is negatively correlated with the spindle speed.

[0024] Optionally, the window length of the second sliding window is determined by the following steps:

[0025] The initial window length is determined based on the spindle speed;

[0026] Determine whether the initial window length is greater than the preset minimum window length;

[0027] In response to the initial window length being greater than the minimum window length, the initial window length is used as the window length of the second sliding window;

[0028] In response to the initial window length being less than the minimum window length, the minimum window length is used as the window length of the second sliding window.

[0029] Optionally, characteristic indicators of the processed data within each second sliding window are calculated, including:

[0030] Perform a Fast Fourier Transform on the processed data within the second sliding window;

[0031] Calculate the characteristic indicators of the processing data within the effective frequency band.

[0032] Optionally, the processing data may also include a vibration signal acquired synchronously with the power signal, and the characteristic index may include the root mean square of the vibration signal.

[0033] Secondly, a tool condition monitoring method is provided, including:

[0034] Acquire machining data under normal operating conditions of the tool, the machining data including synchronously acquired power signal and vibration signal of the tool spindle;

[0035] The power signal is segmented according to the dividing line, and the processing time period of the tool is determined according to the segmentation result; the dividing line is used to define the processing state and non-processing state of the tool.

[0036] Extract feature indicators from the processing data within the processing period;

[0037] The tool condition monitoring threshold is determined based on the characteristic values ​​of the aforementioned characteristic indicators;

[0038] The real-time machining data of the tool is analyzed and judged based on the tool condition monitoring threshold in order to monitor the tool condition.

[0039] Thirdly, a system for constructing a tool condition monitoring model is provided, including:

[0040] The acquisition module is used to acquire machining data under normal operating conditions of the tool, including the power signal of the tool spindle;

[0041] The segmentation module segments the power signal according to the segmentation line and determines the processing time period of the tool based on the segmentation result; the segmentation line is used to define the processing state and non-processing state of the tool.

[0042] The extraction module is used to extract feature indicators of processing data within each processing period.

[0043] The monitoring module is used to determine the tool status monitoring threshold for each processing period based on the feature values ​​of the feature indicators, and to construct a tool status monitoring model based on the tool status monitoring thresholds; the tool status monitoring model is used to analyze and judge the real-time processing data of the tool in order to monitor the tool status.

[0044] Fourthly, a tool condition monitoring system is provided, comprising:

[0045] The acquisition module is used to acquire machining data under normal operating conditions of the tool, including synchronously acquired power signals and vibration signals of the tool spindle;

[0046] The segmentation module is used to segment the power signal according to the segmentation line, and determine the processing period of the tool according to the segmentation result; the segmentation line is used to define the processing state and non-processing state of the tool.

[0047] The extraction module is used to extract feature indicators of the processing data within the processing period.

[0048] The monitoring module is used to determine the tool status monitoring threshold based on the feature value of the feature index, and to analyze and judge the real-time machining data of the tool based on the tool status monitoring threshold, so as to monitor the status of the tool.

[0049] Fifthly, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and for running on the processor, wherein the processor executes the computer program to implement the method described in any of the preceding claims.

[0050] Sixthly, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described in any of the preceding claims.

[0051] In a seventh aspect, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method as described in any of the preceding claims.

[0052] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of this disclosure.

[0053] The positive and progressive effects of this disclosure are as follows: First, this disclosure automatically segments independent processing periods based on power signals, and then uses the processing data within these processing periods for targeted analysis to extract the characteristics of the processing data under normal working conditions and construct a tool condition monitoring model. Thus, through the integrated design of "signal segmentation-feature extraction-state determination," on the one hand, it binds processing data with processing conditions, effectively eliminating interference from non-processing periods and the influence of cross-condition coupling. The tool condition monitoring model built on this data foundation can accurately assess and monitor the tool condition in real time, significantly improving the accuracy and timeliness of monitoring results. On the other hand, it effectively solves the technical pain points of insufficient model generalization ability and significant performance degradation when processing conditions change in the prior art. This disclosure endows the tool condition monitoring model with stronger working condition adaptability, ensuring that it can maintain stable and reliable monitoring performance when processing conditions such as machine tool model, workpiece material, and process parameters are dynamically adjusted, greatly expanding the scope of application. Attached Figure Description

[0054] Figure 1 A flowchart illustrating a method for constructing a tool condition monitoring model, provided as an exemplary embodiment of this disclosure;

[0055] Figure 2 A flowchart illustrating a tool condition monitoring method provided as an exemplary embodiment of this disclosure;

[0056] Figure 3 This is a schematic diagram of the structure of an electronic device provided as an exemplary embodiment of the present disclosure. Detailed Implementation

[0057] The present disclosure is further illustrated below by way of embodiments, but the present disclosure is not limited to the scope of the embodiments described herein.

[0058] The prefixes such as "first" and "second" used in this disclosure are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this disclosure does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not be construed as an unnecessary limitation. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0059] Figure 1 A flowchart illustrating a method for constructing a tool condition monitoring model, provided as an exemplary embodiment of this disclosure, includes the following steps:

[0060] Step 101: Obtain machining data under normal tool operating conditions.

[0061] The machining data includes the power signal of the tool spindle. A power sensor is placed on the tool spindle, and the power signal is acquired based on the power sensor.

[0062] In addition to the power signal, the machining data may also include signals reflecting the cutting force, wear, and temperature of the tool, such as vibration signals, spindle current signals, servo motor load, temperature signals, tool spindle speed, and acoustic emission signals. These signals are acquired synchronously with the power signal. These signals are acquired by deploying appropriate types of sensors on the machine tool, such as placing vibration sensors on the tool spindle to acquire vibration signals. The number and arrangement of various sensors on the machine tool can be set according to actual conditions, and this disclosure does not impose any particular limitations on this.

[0063] In step 101, the signal can be received directly from the sensor or obtained from the memory; the signal stored in the memory is the signal collected and stored by the sensor.

[0064] During CNC machine tool machining, multiple tools and multiple machining operations may be used to process the workpiece. During the machining data acquisition process, each acquired data point is bound to the tool identifier (Tn) used in machining and the machining operation being executed (represented by the program number Pm). Therefore, when each workpiece is completed, a complete set of machining data can be obtained, and each data segment is uniquely identified by the combined label "Tn-Pm" to identify its process background.

[0065] In step 101, the machining data for each tool to be monitored is obtained, and the machining data is bound to the tool identifier and machining process. It should be noted that the user can specify the tools to be monitored, which can be all tools on the machine tool or only some of them.

[0066] Step 102: Divide the power signal according to the dividing line, and determine the processing time of the tool based on the division result.

[0067] The dividing line is used to define the machining and non-machining states of the tool. In other words, the dividing line represents the typical quantization level of the power signal when the machine tool is running under no-load or in standby mode.

[0068] The power of the tool spindle directly reflects the total energy required to drive the tool's rotation and overcome cutting resistance. When the tool begins to contact the workpiece, the load increases instantaneously, and the power of the tool spindle shows a significant upward trend. This power change and the transition of machining states are physical phenomena with a clear causal relationship, providing the most direct and reliable basis for determining the tool's machining period. This embodiment automatically identifies the machining period through power signals, realizing fully automatic and high-precision segmentation of continuous signals into independent machining segments with clear technological meanings. This effectively replaces the traditional manual setting or fixed time window segmentation mode, improving the level of intelligence in machining process monitoring.

[0069] Step 103: Extract the characteristic indicators of the processing data within the processing period.

[0070] In this embodiment, by eliminating machine tool idle interference signals during the machining period and removing those during the non-machining period, effective machining data feature indicators within the machining period are extracted as the data basis for tool condition monitoring. This method can effectively improve the mapping correlation between signals and tool conditions, and ultimately provide the tool condition monitoring model with input data that is structurally unified, has high signal purity, and is comparable, thus solving the technical problem of insufficient model accuracy and universality caused by the nonlinearity, time-varying nature, and coupling characteristics of indirect signals.

[0071] Step 104: Determine the tool condition monitoring threshold based on the characteristic values ​​of the characteristic indicators, and construct a tool condition monitoring model based on the tool condition monitoring threshold.

[0072] Feature indicators are quantitative representations that are extracted from machining data and are strongly correlated with tool condition. Tool condition monitoring thresholds are determined based on the feature values ​​of the feature indicators, so that the tool condition monitoring thresholds have dynamic adaptability and are updated synchronously with the tool life cycle and machine tool condition, thus ensuring accurate and stable monitoring of tool condition.

[0073] The tool condition monitoring model represents the tool condition monitoring thresholds corresponding to each tool identifier, machining process, and machining time period. These thresholds are precisely matched to the tool's cutting conditions and lifecycle stage. This model is used to perform targeted analysis and judgment on real-time machining data, thereby achieving online dynamic monitoring of the tool condition. The specific implementation process is as follows: First, based on the currently monitored tool identifier, machining process, and corresponding machining time period, a tool condition monitoring threshold is matched from the tool condition monitoring model. Then, the collected real-time machining data of the tool is quantitatively compared with the matched tool condition monitoring threshold. Finally, based on the comparison results (such as whether the real-time data exceeds the threshold range and the extent of the exceedance), the working state of the tool is determined. Examples of tool working states include normal cutting, slight wear, severe wear, and chipping.

[0074] In this embodiment, independent processing periods are first automatically segmented based on the power signal. Then, the processing data within each processing period is used for targeted analysis to extract the features of the processing data under normal working conditions and construct a tool condition monitoring model. Through the integrated design of "signal segmentation - feature extraction - state determination," the processing data is bound to the processing conditions, effectively eliminating interference from non-processing periods and the influence of cross-condition coupling, achieving accurate assessment and real-time monitoring of the tool condition, and significantly improving the accuracy and timeliness of the monitoring results. On the other hand, it effectively solves the technical pain points of insufficient model generalization ability and significant performance degradation when processing conditions change in the prior art. This embodiment gives the tool condition monitoring model stronger working condition adaptability, ensuring that it can maintain stable and reliable monitoring performance when processing conditions such as machine tool model, workpiece material, and process parameters are dynamically adjusted, greatly expanding the scope of application.

[0075] The following describes one way to determine the dividing line:

[0076] S1. The power signal is smoothed by using a gradually increasing first sliding window to obtain a power sequence; each element in the power sequence corresponds to the calculation result of the first sliding window.

[0077] The power signal is smoothed by using a gradually increasing first sliding window (i.e., a recursive sliding window) to suppress random noise introduced during signal acquisition while preserving the true trend characteristics of the power signal to the greatest extent possible.

[0078] The following describes the specific process of smoothing:

[0079] 1) The power signal is represented by a one-dimensional discrete signal sequence U of length N, where U = [u0, u1, u2, ..., u...]. N-1 ], where u i ∈R, 0≤i≤N-1.

[0080] 2) To address the boundary issue of insufficient historical data at the beginning of the power signal in the first sliding window, this embodiment employs a cumulative averaging method to smooth the data at the beginning. Let the window length of the first sliding window be K (K is an odd number). For the first Kp = min(K, N) points of sequence U, an increasingly larger window is used for averaging to avoid abrupt changes in the data at the beginning. The calculation formula is as follows:

[0081] where i = 0, 1, ..., K p-1 This ensures a smooth transition at the beginning of the signal, where Yi represents the smoothed result of the initial data.

[0082] 3) Calculate the sum of the first K elements of sequence U as the initial window and S. K-1 S K-1 = Calculate the constant Cinv = 1 / K, calculate the arithmetic mean of the data within the first complete window, and output: ;

[0083] 4) For i=K,K+1,…,N-1, update the window sum and calculate the output according to the following recursive formula:

[0084] S i =S i-1 -u i-K +u i ,

[0085] y i =C inv *S i ;

[0086] y i These are the calculation results for each of the first sliding windows.

[0087] In one embodiment, to ensure that the length of the output power sequence matches the complete cycle of the sliding window processing, the signal ends are smoothly closed based on the sliding window processing mechanism. The target length of the smoothed sequence should be N + K - 1 to ensure that there is a corresponding output throughout the entire data processing period.

[0088] In one embodiment, the power sequence ends are smoothly closed by the following steps:

[0089] S1-1. Determine whether the number of elements contained in the power sequence has reached the target number.

[0090] S1-2. In response to the fact that the number of elements has not reached the target number, the number of elements in the power sequence is padded to the target number using the last element contained in the power sequence.

[0091] Step S1-2 is also the calculation result y of the (N-1)th point. N-1 As a constant, it is copied backward K-1 times:

[0092] y i =y N-1 , where i=N,N+1,…,N+K-2;

[0093] The final smoothed power sequence is Y=[y0,y1,y2,…,y N+K-2 ], where the first Kp points of sequence Y (y0 to y Kp-1 It is calculated from the formula in step 2).

[0094] In this embodiment, a complete smooth process including progressive front-end processing, main body recursive update and tail backward retention is constructed, which can effectively suppress random noise in industrial time-series signals, while fully preserving the original trend characteristics of the signal, and meeting the real-time processing requirements of industrial sites, thereby providing a high-quality data foundation for subsequent signal analysis, feature extraction, status monitoring and other links.

[0095] S2. Calculate the quantiles of the power sequence and determine the dividing line based on the calculation results.

[0096] The following describes one way to determine the dividing line:

[0097] Divider B=Q q (Y), where, This represents the quantile function. Arranging the power sequence Y or its smoothed form in ascending order yields the ordered sequence Y. sorted Take the index as The value at point B is used as the dividing line, where the quantile parameter q is a preset constant with a value range of [0.1, 0.3]. M is the process sequence length of the power signal, which is determined by the sampling frequency of the power signal and the duration of the power signal to be analyzed. M = sampling frequency × duration of power signal. The process sequence should ensure that it contains at least one complete processing cycle.

[0098] In this embodiment, the segmentation line is determined by quantiles, which can adaptively match the characteristic fluctuation patterns of different processing conditions. It can also adapt to the inherent noise of the equipment, the slight differences of different tools, and the slow changes in the working conditions. This not only significantly improves the robustness and accuracy of signal segmentation, but also improves the accuracy of tool anomaly monitoring during the processing. It effectively reduces production losses caused by false alarms and unplanned downtime, and greatly reduces the time and manpower costs of on-site machine tool debugging. Ultimately, it provides key technical support for the construction of a predictive maintenance system for equipment.

[0099] In other implementations, a threshold method is used to determine the dividing line. Specifically, a fixed power threshold is set, and the processing period begins when the power signal exceeds this threshold for a certain duration. Alternatively, the first derivative (rate of change) of the power signal can be used; when its absolute value exceeds a set threshold, it is marked as a state switching point.

[0100] In other implementations, machine learning models are used to identify the state of power timing signals and automatically segment them into multiple state stages such as "no load", "processing", and "tool change".

[0101] In other implementations, the dividing line is determined based on the bimodal threshold method. Specifically, the following steps are taken: acquire the machine tool's full operating condition signals (including complete power signals for no-load, standby, and machining), plot the frequency histogram / kernel density curve of the signals, identify the bimodal peaks and intermediate valleys, and determine the dividing line based on the signal values ​​corresponding to the valleys.

[0102] In other implementations, the dividing line is determined based on empirical values ​​or simulation experiments.

[0103] The following describes one method for determining the processing time period:

[0104] After performing effective interval segmentation of the power signal based on baseline B, the position indices of n processing time periods are obtained:

[0105] {[s1,e1],[s2,e2],...,[s n ,e n ]}, where s q and e q These represent the start and end indices of the q-th processing time segment in the smoothed signal sequence Y, respectively. To convert the above positional information into physically meaningful time coordinates, the power signal acquisition frequency fs is introduced. The conversion relationship is as follows:

[0106] The start time t of the qth processing period sq With end time t eq Calculated by the following formula:

[0107] t sq = s q / fs, t eq = e q / fs;

[0108] Therefore, all processing periods can be represented by time intervals as follows:

[0109] {[t s1 , t e1 ], [t s2 , t e2 ],...,[t sn , ten ]}.

[0110] In one embodiment, to convert a continuous time-varying signal into a stationary sequence of segments suitable for short-time frequency domain analysis and to extract features that reflect the dynamic changes in the tool's machining state, a second sliding window is used to segment the machining data within the machining period. Specifically:

[0111] The processing data within the processing period is segmented using a second sliding window, and the characteristic indicators of the processing data within each second sliding window are calculated.

[0112] Specifically, the window length of the second sliding window is negatively correlated with the spindle speed, and / or the step size of the second sliding window is negatively correlated with the spindle speed. The window length < the step size, and both the window length and the step size are expressed in points.

[0113] In this embodiment, the window length of the second sliding window is configured inversely proportional to the rotational speed of the tool spindle. When the tool spindle is in a high-speed machining state, the window length of the second sliding window shortens synchronously to achieve effective capture of high-frequency machining features.

[0114] In one embodiment, the reference window length T0 and the reference rotational speed R0 of the tool spindle are defined, and the window length of the second sliding window is determined by the following formula:

[0115] T = T0*(R0 / R).

[0116] In this embodiment, the window length and step size of the second sliding window are adaptively and dynamically adjusted according to the specific physical characteristics of the workpiece processing process. This effectively suppresses signal noise caused by machine tool operation, environmental vibration and electromagnetic interference, greatly reduces the difficulty of signal feature extraction, and ensures the accuracy and reliability of feature parameters.

[0117] In one embodiment, the window length of the second sliding window is determined by the following steps:

[0118] The initial window length is determined based on the spindle speed;

[0119] Determine if the initial window length is greater than the preset minimum window length;

[0120] In response to the initial window length being greater than the minimum window length, the initial window length is used as the window length of the second sliding window;

[0121] In response to the initial window length being less than the minimum window length, the minimum window length is used as the window length of the second sliding window.

[0122] In this embodiment, to ensure the feasibility of frequency domain analysis, a minimum window length constraint N is set. min, then the finally adopted window length W is W = max(T, N min ). The step size S is adaptively set according to the window length. Set the overlap coefficient x (0 < x < 1), S = W * x. The value of x is set according to the actual situation.

[0123] In one embodiment, the characteristic index is calculated based on frequency domain analysis. Specifically, the steps of calculating the characteristic index of the processing data in each second sliding window include: performing a fast Fourier transform on the processing data in the second sliding window, and calculating the characteristic index of the processing data within the effective frequency band range.

[0124] The effective frequency band range [F min , F max is determined according to the rotational speed of the tool spindle and its harmonics. The determination method of the effective frequency band range [F min , F max is as follows: taking the spindle rotation frequency f r = R / 60 and its first n harmonics (n*f r , where n is the preset harmonic order) as the center frequencies; taking Δf above and below each center frequency as the half bandwidth, and jointly forming multiple target frequency bands. The half bandwidth Δf is usually taken as half of the spindle rotation frequency f r , that is, Δf = f r / 2; the final effective frequency band [F min , F max is the union of these target frequency bands. The final characteristic index is the sum of the frequency band characteristic indexes calculated for all target frequency bands respectively.

[0125] In this embodiment, by determining the effective frequency band through the spindle rotational speed and its harmonics and calculating the characteristic index of the processing data within the effective frequency band, the accurate extraction of processing characteristics can be achieved, effectively suppressing the noise effects brought by machine tool operation, environmental vibration and electromagnetic interference, while adapting to the processing conditions of different spindle rotational speeds, ensuring the consistency and reliability of the characteristic index, and meeting the requirements of real-time monitoring of the processing process.

[0126] In other implementation manners, statistics such as the effective value, peak value, and kurtosis of the processing data are calculated as the characteristic index.

[0127] In one embodiment, the characteristic index includes the root mean square of the vibration signal. For each processing time period, the window is slid with the step size S, and the root mean square of the vibration signal of each second sliding window is calculated. For the characteristic value sequence F q = [f q1 , f q2 ,..., f qmi corresponding to the qth processing period, where m i represents the total number of second sliding windows within this processing period, and the characteristic value sequence Fq Each element in the equation corresponds to the root mean square of the vibration signal in a second sliding window.

[0128] Vibration signals are the direct physical response to stress waves and impacts during the contact between the tool and the workpiece. Even minute changes in the tool will generate detectable transient impacts or spectral changes in the vibration signal. Therefore, in this embodiment, independent machining periods are first automatically segmented based on the power signal, thus providing a precise framework for targeted analysis of the vibration signal. Then, within this framework, the vibration signal is used to evaluate and monitor indicators such as tool condition and machining process stability in real time.

[0129] In this embodiment, by conducting collaborative analysis of power signals and vibration signals and configuring an efficient data smoothing method, the automatic and accurate segmentation of the processing period is achieved, while real-time status monitoring of the processing process is completed.

[0130] In one embodiment, the tool condition monitoring threshold is determined based on the mean and standard deviation of the feature value sequence. Specifically: for each feature index to be monitored in the Tn-Pm combination, denoted as F, the baseline mean μ of the feature value sequence over all machining periods is calculated. F and standard deviation σ F The baseline mean μ of the eigenvalue sequence F and standard deviation σ F Characterizes health benchmarks.

[0131] Based on statistical principles, a dynamic early warning threshold T for this feature is set. W With dynamic alarm threshold T A And based on the threshold T W With T A Construct a tool condition monitoring model. Dynamic early warning threshold T. W With dynamic alarm threshold T A It is expressed as follows:

[0132] T W = μ F ± K1σ F , T A = μ F ± K2σ F ;

[0133] Wherein, K1 and K2 represent the set warning coefficient and alarm coefficient, respectively.

[0134] In this embodiment, an independent monitoring benchmark is established for each processing period, and the tool state is monitored using dynamic statistical thresholds, which can adaptively match the characteristic fluctuation rules of different processing conditions, effectively distinguish normal fluctuations from real anomalies, not only significantly improve the accuracy of tool anomaly monitoring during the processing, but also effectively reduce production losses caused by false alarms and unplanned downtime, greatly reduce the time and labor costs for on-site debugging of machine tools, and ultimately provide key technical support for the construction of an equipment predictive maintenance system.

[0135] During the subsequent workpiece processing, real-time online monitoring and anomaly judgment are performed on the new processing process. When each processing period is completed, its Tn-Pm label is automatically identified, and the mean value μ of the eigenvalue sequence within the current processing period in the real-time processing data is calculated in the same way as when constructing the tool state monitoring model. curr .

[0136] In one embodiment, taking K1 = 2 and K2 = 3 as an example, if the mean value μ curr falls within the range [μ W - 2σ F , μ F + 2σ F determined according to T F [[ID=二十]], a warning prompt is given. If the mean value μ curr falls within the range [μ A - 3σ F , μ F F + 3σ F determined according to T

[0137] In one embodiment, the standardized deviation D of the current eigenvalue relative to the healthy benchmark (eigenvalue sequence) is calculated:

[0138] D = ∣μ curr - μ F ∣ / σ F , and the processing state of this processing period is judged according to the size of D:

[0139] If D ≤ K1, it indicates that the process characteristics of the tool are within the normal fluctuation range;

[0140] If K1 < D ≤ K2, it indicates that the process characteristics of the tool show a significant deviation, and a warning prompt is given, such as recording a log and prompting an inspection;

[0141] If D > 3, it indicates that the process characteristics of the tool have exceeded the normal fluctuation limit, and an alarm prompt is given, such as indicating a high probability of anomaly and observing whether there are phenomena such as tool breakage, chipping, and wear on the site. )

[0142] ​This disclosure also provides a tool condition monitoring method, see [link to relevant documentation]. Figure 2 The tool condition monitoring method includes the following steps:

[0143] Step 201: Obtain machining data under normal operating conditions of the tool, wherein the machining data includes synchronously acquired power signals and vibration signals of the tool spindle;

[0144] Step 202: Segment the power signal according to the dividing line, and determine the processing time period of the tool according to the segmentation result; the dividing line is used to define the processing state and non-processing state of the tool;

[0145] Step 203: Extract the feature indicators of the processing data within the processing period;

[0146] Step 204: Determine the tool condition monitoring threshold based on the characteristic values ​​of the aforementioned characteristic indicators;

[0147] Step 205: Analyze and determine the real-time machining data of the tool according to the tool status monitoring threshold to monitor the tool status.

[0148] The implementation process of steps 201-204 is the same as that of steps 101-104, and will not be repeated here.

[0149] Corresponding to the aforementioned methods for constructing tool condition monitoring models and embodiments of tool condition monitoring methods, this disclosure also provides a system for constructing tool condition monitoring models and embodiments of tool condition monitoring systems.

[0150] This disclosure provides a tool condition monitoring model construction system. This system implements the tool condition monitoring model construction method provided in any of the above embodiments. The system includes:

[0151] The acquisition module is used to acquire machining data under normal operating conditions of the tool, including the power signal of the tool spindle;

[0152] The segmentation module segments the power signal according to the segmentation line and determines the processing time period of the tool based on the segmentation result; the segmentation line is used to define the processing state and non-processing state of the tool.

[0153] The extraction module is used to extract feature indicators of processing data within each processing period.

[0154] The monitoring module is used to determine the tool status monitoring threshold for each processing period based on the feature values ​​of the feature indicators, and to construct a tool status monitoring model based on the tool status monitoring thresholds; the tool status monitoring model is used to analyze and judge the real-time processing data of the tool in order to monitor the tool status.

[0155] Optionally, the system also includes:

[0156] A smoothing module is used to smooth the power signal using a gradually increasing first sliding window to obtain a power sequence; each element in the power sequence corresponds to the calculation result of a first sliding window;

[0157] The calculation module is used to perform quantile calculations on the power sequence and determine the dividing line based on the calculation results.

[0158] Optionally, the system also includes:

[0159] The judgment module is used to determine whether the number of elements contained in the power sequence reaches the target number;

[0160] In response to the fact that the number of elements has not reached the target number, the judgment module calls the padding module to pad the number of elements in the power sequence to the target number using the last element contained in the power sequence;

[0161] The calculation module is specifically used to calculate the quantiles of the padded power sequence.

[0162] Optionally, the extraction module is specifically used for:

[0163] The machining data within the machining period is segmented using a second sliding window, and characteristic indicators of the machining data within each second sliding window are calculated; wherein, the window length of the second sliding window is negatively correlated with the spindle speed, and / or the step size of the second sliding window is negatively correlated with the spindle speed.

[0164] Optionally, the system includes a window length determination module for the second sliding window, specifically used for:

[0165] The initial window length is determined based on the spindle speed;

[0166] Determine whether the initial window length is greater than the preset minimum window length;

[0167] In response to the initial window length being greater than the minimum window length, the initial window length is used as the window length of the second sliding window;

[0168] In response to the initial window length being less than the minimum window length, the minimum window length is used as the window length of the second sliding window.

[0169] Optionally, when calculating the feature indicators of the processed data within each second sliding window, the extraction module is specifically used for:

[0170] Perform a Fast Fourier Transform on the processed data within the second sliding window;

[0171] Calculate the characteristic indicators of the processing data within the effective frequency band.

[0172] Optionally, the processing data may also include a vibration signal acquired synchronously with the power signal, and the characteristic index may include the root mean square of the vibration signal.

[0173] This disclosure also provides a tool condition monitoring system, which is used to implement the tool condition monitoring method of any of the above embodiments. The system includes:

[0174] The acquisition module is used to acquire machining data under normal operating conditions of the tool, including synchronously acquired power signals and vibration signals of the tool spindle;

[0175] The segmentation module is used to segment the power signal according to the segmentation line, and determine the processing period of the tool according to the segmentation result; the segmentation line is used to define the processing state and non-processing state of the tool.

[0176] The extraction module is used to extract feature indicators of the processing data within the processing period.

[0177] The monitoring module is used to determine the tool status monitoring threshold based on the feature value of the feature index, and to analyze and judge the real-time machining data of the tool based on the tool status monitoring threshold, so as to monitor the status of the tool.

[0178] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs.

[0179] Figure 3 This is a schematic diagram of the structure of an electronic device according to an example embodiment of the present disclosure. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the method described in any of the above embodiments. Figure 3 The electronic device 30 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0180] like Figure 3As shown, the electronic device 30 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including memory 32 and processor 31).

[0181] Bus 33 includes a data bus, an address bus, and a control bus.

[0182] The memory 32 may include volatile memory, such as random access memory (RAM) 321 and / or cache memory 322, and may further include read-only memory (ROM) 323.

[0183] The memory 32 may also include a program tool 325 (or utility) having a set (at least one) program module 324, such program module 324 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0184] The processor 31 executes various functional applications and data processing, such as the methods provided in any of the above embodiments, by running computer programs stored in the memory 32.

[0185] Electronic device 30 can also communicate with one or more external devices 34 (e.g., keyboard, pointing device, etc.). This communication can be performed through input / output (I / O) interface 35. Furthermore, electronic device 30 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 36. As shown, network adapter 36 communicates with other modules of electronic device 30 via bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with electronic device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0186] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0187] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method provided in any of the above embodiments.

[0188] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0189] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the above embodiments.

[0190] The program code for executing the computer program product of this disclosure can be written in any combination of one or more programming languages, and the program code can be executed entirely on a user device, partially on a user device, as a stand-alone software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0191] While specific embodiments of this disclosure have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of this disclosure is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of this disclosure, but all such changes and modifications fall within the scope of protection of this disclosure.

Claims

1. A method for constructing a tool condition monitoring model, characterized in that, include: Acquire machining data under normal operating conditions of the tool, wherein the machining data includes the power signal of the tool spindle; The power signal is segmented according to the dividing line, and the processing time period of the tool is determined according to the segmentation result; the dividing line is used to define the processing state and non-processing state of the tool. Extract characteristic indicators of processing data within each processing period; The tool status monitoring threshold for each processing period is determined based on the characteristic values ​​of the characteristic indicators, and a tool status monitoring model is constructed based on the tool status monitoring thresholds. The tool status monitoring model is used to analyze and judge the real-time processing data of the tool in order to monitor the status of the tool.

2. The construction method according to claim 1, characterized in that, Also includes: The power signal is smoothed using a gradually increasing first sliding window to obtain a power sequence; each element in the power sequence corresponds to the calculation result of the first sliding window. The power sequence is subjected to quantile calculation, and the dividing line is determined based on the calculation result.

3. The construction method according to claim 2, characterized in that, Before the step of calculating the quantiles of the power sequence, the method further includes: Determine whether the number of elements contained in the power sequence reaches the target number; In response to the fact that the number of elements does not reach the target number, the number of elements in the power sequence is padded to the target number using the last element contained in the power sequence; The step of calculating the quantiles of the power sequence includes: Quantiles are calculated for the padded power sequence.

4. The construction method according to any one of claims 1-3, characterized in that, Extract characteristic indicators of processing data within each processing period, including: The machining data within the machining period is segmented using a second sliding window, and characteristic indicators of the machining data within each second sliding window are calculated; wherein, the window length of the second sliding window is negatively correlated with the spindle speed, and / or the step size of the second sliding window is negatively correlated with the spindle speed.

5. The construction method according to claim 4, characterized in that, The window length of the second sliding window is determined by the following steps: The initial window length is determined based on the spindle speed; Determine whether the initial window length is greater than the preset minimum window length; In response to the initial window length being greater than the minimum window length, the initial window length is used as the window length of the second sliding window; In response to the initial window length being less than the minimum window length, the minimum window length is used as the window length of the second sliding window.

6. The construction method according to claim 4, characterized in that, Calculate the characteristic indicators of the processed data within each second sliding window, including: Perform a Fast Fourier Transform on the processed data within the second sliding window; Calculate the characteristic indicators of the processing data within the effective frequency band.

7. The construction method according to claim 4, characterized in that, The processing data also includes vibration signals acquired synchronously with the power signal, and the characteristic index includes the root mean square of the vibration signal.

8. A tool condition monitoring method, characterized in that, include: Acquire machining data under normal operating conditions of the tool, the machining data including synchronously acquired power signal and vibration signal of the tool spindle; The power signal is segmented according to the dividing line, and the processing time period of the tool is determined according to the segmentation result; the dividing line is used to define the processing state and non-processing state of the tool. Extract feature indicators from the processing data within the processing period; The tool condition monitoring threshold is determined based on the characteristic values ​​of the aforementioned characteristic indicators; The real-time machining data of the tool is analyzed and judged based on the tool condition monitoring threshold in order to monitor the tool condition.

9. A system for constructing a tool condition monitoring model, characterized in that, include: The acquisition module is used to acquire machining data under normal operating conditions of the tool, including the power signal of the tool spindle; The segmentation module segments the power signal according to the segmentation line and determines the processing time period of the tool based on the segmentation result; the segmentation line is used to define the processing state and non-processing state of the tool. The extraction module is used to extract feature indicators of processing data within each processing period. The monitoring module is used to determine the tool status monitoring threshold for each processing period based on the feature values ​​of the feature indicators, and to construct a tool status monitoring model based on the tool status monitoring thresholds; the tool status monitoring model is used to analyze and judge the real-time processing data of the tool in order to monitor the tool status.

10. A tool condition monitoring system, characterized in that, include: The acquisition module is used to acquire machining data under normal operating conditions of the tool, including synchronously acquired power signals and vibration signals of the tool spindle; The segmentation module is used to segment the power signal according to the segmentation line, and determine the processing period of the tool according to the segmentation result; the segmentation line is used to define the processing state and non-processing state of the tool. The extraction module is used to extract feature indicators of the processing data within the processing period. The monitoring module is used to determine the tool status monitoring threshold based on the feature value of the feature index, and to analyze and judge the real-time machining data of the tool based on the tool status monitoring threshold, so as to monitor the status of the tool.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and for running on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 8.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 8.

13. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-8.

Images