Robot adaptive machining method and machining system

Abstract

The robot adaptive machining method and machining system solves the problem of ensuring machining efficiency while ensuring machining quality, and belongs to the field of robot application. The present invention includes: during the robot milling process of each layer, collecting the tool position signal of the current layer, and collecting the vibration signal or the spindle torque signal at the same time; correlating the collected signals in time sequence to obtain an index matrix of the correlation signal; when the robot mills each layer, if chatter occurs, determining the chatter occurrence range according to the index matrix, adjusting the current robot feed speed within the chatter occurrence range until the machining reaches a stable state, and recording the adjusted cutting parameters; if chatter is detected when the robot mills the i-th layer and the chatter occurrence range R is determined, i When the robot mills the i+1th layer, the chattering range R i When the robot is milling the i-th layer, it uses the adjusted cutting parameters, and the rest of the positions in the i+1 layer keep the original cutting parameters.

Description

Technical Field

[0001] The invention relates to a robot self-adaptive processing method and processing system, and belongs to the field of robot application. Background Art

[0002] Industrial robots are widely used in milling processes for various components due to their flexibility, wide workspace, and low cost. Process monitoring during robotic machining can effectively determine the current machining status and improve component quality. However, current robotic machining process monitoring typically monitors specific sensors or signals in the time domain, lacking close correlation with the machining process. Signal correlation across multiple sensors and systems could further enhance process monitoring capabilities. The selection of machining parameters during robotic machining is also crucial. Conservative machining parameters, while ensuring machining quality and safety, sacrifice efficiency. Exploiting excessive cutting parameters can cause chatter and compromise machining quality. By combining robotic machining process monitoring with appropriate cutting parameters, adaptive machining can be achieved, ensuring both efficiency and quality.

[0003] The invention patent with publication number CN115509177A, "A Method, Device, Equipment, and Medium for Monitoring Abnormalities in Part Machining," determines whether a part is a first-piece part. If so, the maximum power of each tool in the part's first power monitoring section is obtained, and the maximum power in the first power monitoring section is compared with the maximum monitoring threshold. If the part is not a first-piece part, the maximum power of the tool in the second power monitoring section is obtained, and the maximum power in the second power monitoring section is compared with the monitoring threshold, thereby determining abnormalities in the machining process. However, for a single machining feature, machining process monitoring does not receive enough attention, and the threshold signal used for monitoring is single, making it impossible to obtain more complete and highly correlated monitoring information. The invention patent with publication number CN105867305A, "Real-time Monitoring Method for CNC Machining Status of Complex Structural Parts Based on Machining Features", establishes a monitoring threshold for whether the machining status is normal through machining features, establishes a typical machining feature monitoring signal and threshold library based on cutting experiments, and adds a machining feature monitoring identifier in the CNC machining program to match the machining feature monitoring threshold in the threshold library. For new machining features, samples of machining features are established through real-time learning to realize machining status monitoring. However, it is mainly aimed at machine tool machining, and there is still room for expansion for the same type of monitoring thresholds for a single feature, which can be expanded to monitoring thresholds for different machining positions of a single feature, and then applied to the same type of features. Summary of the Invention

[0004] In order to solve the problem of how to ensure machining efficiency and quality while performing conventional robot machining, the present invention provides a robot adaptive machining method and a machining system.

[0005] A robot adaptive machining method of the present invention comprises:

[0006] During the robot milling process of each layer, the tool position signal of the current layer is collected, and the vibration signal or the spindle torque signal is collected at the same time. During the collection, the tool position signal and the vibration signal or the tool position signal and the spindle torque signal are extracted and aligned respectively, and then the tool position signal and the vibration signal or the tool position signal and the spindle torque signal are correlated in time sequence to obtain an index matrix of the correlation signal.

[0007] When the robot mills each layer, it detects whether chatter occurs based on the vibration signal or spindle torque signal. If so, it determines the chatter range based on the index matrix of the associated signal. The robot feed speed is adjusted within the chatter range until the machining reaches a stable state, and the adjusted cutting parameters are recorded.

[0008] If the robot detects chatter when milling the i-th layer and determines the chatter range R i When the robot mills the i+1th layer, the chattering range R i When the robot is milling the i-th layer, it uses the adjusted cutting parameters, and the rest of the positions in the i+1 layer keep the original cutting parameters.

[0009] Preferably, the method further comprises:

[0010] During the robot milling process of each layer, the robot joint position signals are also collected;

[0011] When the robot mills the ath layer, a>1, when chatter occurs and the current overall structure stiffness change coefficient When the current feed speed α0 is adjusted to λ is the set threshold;

[0012]

[0013] in, k=1,a, K1 represents the overall structural stiffness when processing the first layer, K a Indicates the overall structural stiffness of the robot when milling the a-th layer; the tool position in the machining program is determined according to the robot joint position signal, K tool,j is the robot end stiffness corresponding to the jth tool position in the corresponding machining program, M1 is the number of tool positions in the machining program corresponding to the first layer, and M a is the number of tool locations in the machining program corresponding to layer a.

[0014] Preferably, while collecting, the tool position signal and the vibration signal or the tool position signal and the spindle torque signal are extracted and aligned respectively, and then the tool position signal and the vibration signal or the tool position signal and the spindle torque signal are correlated in time sequence to obtain an index matrix of the correlation signal:

[0015] A sliding window is used to collect data at each moment. While collecting the signal, the data in the sliding window at the previous moment are correlated. The sampling time is recorded as:

[0016] t p ={t p,0 ,t p,1 ,…,t p,m},m=N1

[0017] t v ={t v,0 ,t v,1 ,…,t v,n},n=N2

[0018] t t ={t t,0 ,t t,1 ,…,t t,s},s=N3

[0019] Among them, t p , t v and t t are the sampling time series of tool position signal, vibration signal and spindle torque signal respectively. N1, N2 and N3 are the number of sampling points. The corresponding sampling point information is recorded as:

[0020] N p ={N p,0 ,N p,1 ,…,N p,m},m=N1

[0021] N v ={N v,0 ,N v,1 ,…,N v,n},n=N2

[0022] N t ={N t,0 ,N t,1 ,…,N t,s},s=N3

[0023] Among them, N p 、N v and N t They are the sampling point information sequences of tool position signal, vibration signal and spindle torque signal, according to the time sequence t p -tv -t t The optimal matching relationship of the correlation signal is established by the index matrix M p-v-t :

[0024] M p-v-t =[0,…,j,…,N4]

[0025] Among them, N4 is the number of sampling points after matching and alignment;

[0026] The aligned signal sampling points are expressed as:

[0027]

[0028] in, and They are the sampling point information sequences of the tool position signal, vibration signal and spindle torque signal after alignment;

[0029] When the robot detects chatter during milling, the index matrix M is calculated based on the aligned vibration signal or spindle torque signal. p-v-t Find the index value M in c , according to the index value M c Determine the associated tool position signal and then determine the range of chatter occurrence R i for:

[0030]

[0031] As a preferred method, the method of detecting whether chatter occurs according to the vibration signal is:

[0032]

[0033] Among them, N v,i is the sampling information corresponding to the jth sampling point of the collected vibration signal, c is the empirical threshold coefficient, N v Represents the sampling point information sequence of the collected vibration signal, y c Indicates the chatter threshold of the vibration signal.

[0034] Preferably, the method further comprises:

[0035] After the robot mills each layer, a color map of the vibration signal distribution at the processing position after alignment is output;

[0036] The methods for obtaining the color map include:

[0037] All vibration signals N collected from the i-th layer v,j The amplitude in the time domain is divided into P intervals according to the linear function, each interval represents a color index. v,j Normalization:

[0038] y rule =(N v,j -minN v ) / (maxN v -minN v )

[0039] y rule is the vibration signal N v,j Normalized value;

[0040] Calculate the index value y of the interval color index :

[0041] y index =floor(P*y rule -y rule )+1

[0042] After the robot mills the i-th layer, according to the index matrix M p-v-t with y index , a color map of the vibration signal distribution on the processing position after alignment on the current i-th layer is drawn for offline analysis of the milling status.

[0043] As a preference, when the robot mills the i+1th layer, the robot mills into the vibration generating range R i When the robot uses the adjusted cutting parameters when milling the i-th layer, the method includes:

[0044] When the robot is milling the i+1th layer, the safety distance for adjusting the robot feed speed is set to L, and the tool is close to the vibration range R in the feed direction. i When the distance is L, the feed speed is gradually reduced to the recorded value, away from the vibration range R i The robot feed speed is gradually restored. If the distance between the two vibration ranges in the i-th layer is less than L, the two vibration ranges are merged into one vibration range when the robot feed speed is adjusted in the i+1-th layer.

[0045] The present application also provides a robot adaptive machining system, comprising:

[0046] The signal acquisition module is used to collect the tool position signal of the current milling layer during the robot milling process, and simultaneously collect the vibration signal or spindle torque signal and send it to the online monitoring module;

[0047] The online monitoring module is used to extract and align the tool position signal and the vibration signal or the tool position signal and the spindle torque signal respectively during acquisition, and then correlate the tool position signal and the vibration signal or the tool position signal and the spindle torque signal in time sequence to obtain an index matrix of the correlated signals; it is also used to detect whether chatter occurs based on the vibration signal or the spindle torque signal when the robot mills each layer, and if so, determine the chatter occurrence range based on the index matrix of the correlated signals and send it to the adaptive control module; it is also used to output a color map of the distribution of the aligned vibration signals at the processing position to the offline analysis module after the robot mills each layer;

[0048] The adaptive control module is used to control the robot to mill each layer and adjust the current robot feed speed within the range of vibration until the processing reaches a stable state, and record the adjusted cutting parameters; it is also used to detect vibration when the robot mills the i-th layer and determine the vibration range R i When the robot mills the i+1th layer, the chattering range R i When the robot is milling the i-th layer, the robot uses the adjusted cutting parameters, and the rest of the positions in the i+1 layer keep the original cutting parameters;

[0049] The offline analysis module is used to analyze the vibration distribution of each layer generated during the robot milling process based on the color map of the vibration signal distribution at the processing position.

[0050] Preferably, the signal acquisition module is further used to collect the robot joint position signal during the robot milling process of each layer;

[0051] The adaptive control module is also used when the robot mills the ath layer, a>1, when chatter occurs and the current overall structure stiffness change coefficient When the current feed speed α0 is adjusted to λ is the set threshold;

[0052]

[0053] in, k=1,a, K1 represents the overall structural stiffness when processing the first layer, K a Indicates the overall structural stiffness of the robot when milling the a-th layer; the tool position in the machining program is determined according to the robot joint position signal, K tool,j is the robot end stiffness corresponding to the jth tool position in the corresponding machining program, M1 is the number of tool positions in the machining program corresponding to the first layer, and M a is the number of tool locations in the machining program corresponding to layer a.

[0054] The beneficial effect of the present invention is that the feed speed of the robot does not need to be calculated by formula or obtained by intelligent algorithm every time. Due to the similarity of actual processing characteristics, in the actual processing process, the range of vibration occurring in the current i-th layer and the adjusted stable cutting parameters are used. When milling the i+1 layer, the feed parameters can be adjusted in advance within the recorded range. When milling the i+n layer, if there is a large difference in the processing position, the cutting parameters are adaptively adjusted through the robot stiffness, thereby ensuring the processing efficiency while ensuring the processing quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flow chart of the data association method;

[0056] Figure 2 It is a schematic diagram of the robot milling different layers;

[0057] Figure 3 This is the working condition diagram at a certain moment when the robot is milling the first layer;

[0058] Figure 4 It is the working state diagram at a certain moment when the robot is milling the a-th layer;

[0059] Figure 5 This is a schematic diagram of the principle of the method for calculating the color index value of the vibration distribution;

[0060] Figure 6 It is a color diagram of the vibration distribution at the processing position during the square hole processing process. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0062] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0063] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0064] The robot adaptive machining method of this embodiment includes:

[0065] Step 1: During the robot milling process of each layer, the tool position signal, robot joint position signal, vibration signal and spindle torque signal of the current layer are collected;

[0066] The multi-channel signals collected are obtained through different sensors and acquisition systems, so the sampling frequencies are different. The sampling frequencies of the tool position signal and the robot joint position signal are recorded as f1, and the sampling frequencies of the vibration signal and the spindle torque signal are recorded as f2 and f3 respectively. All signals are collected synchronously.

[0067] The tool position signal, robot joint position signal, vibration signal and spindle torque signal generated during the robot milling process are obtained by the robot's internal sensors, the vibration signal is obtained by an external vibration sensor, and the spindle torque signal is obtained by the electric spindle driver;

[0068] Step 2: While collecting, respectively extract and align the tool position signal and the vibration signal or the tool position signal and the spindle torque signal, and then correlate the tool position signal and the vibration signal or the tool position signal and the spindle torque signal in time sequence to obtain an index matrix of the correlation signal;

[0069] In a preferred embodiment, a sliding window is used to collect data at each moment. While collecting signals, the data in the sliding window at the previous moment are correlated. The sampling time is recorded as:

[0070] t p ={t p,0 ,t p,1 ,…,t p,m},m=N1

[0071] t v ={t v,0 ,t v,1 ,…,t v,n},n=N2

[0072] t t ={t t,0 ,t t,1 ,…,t t,s},s=N3

[0073] Among them, t p , t v and t t are the sampling time series of tool position signal, vibration signal and spindle torque signal respectively. N1, N2 and N3 are the number of sampling points. The corresponding sampling point information is recorded as:

[0074] N p ={N p,0 ,N p,1 ,…,N p,m},m=N1

[0075] N v ={N v,0 ,Nv,1 ,…,N v,n},n=N2

[0076] N t ={N t,0 ,N t,1 ,…,N t,s},s=N3

[0077] Among them, N p 、N v and N t They are the sampling point information sequences of tool position signal, vibration signal and spindle torque signal, according to the time sequence t p -t v -t t The optimal matching relationship of the correlation signal is established by the index matrix M p-v-t :

[0078] M p-v-t =[0,…,j,…,N4]

[0079] Among them, M p-v-t is the index matrix of the tool position signal, vibration signal and spindle torque signal that meet the optimal matching relationship in time sequence, and N4 is the number of sampling points after matching and alignment;

[0080] After extraction, the signals are aligned. The purpose of extraction and alignment is signal correlation. The specific correlation method is as follows: Figure 1 As shown, the aligned signal sampling points are expressed as:

[0081]

[0082] in, and They are the sampling point information sequences of the tool position signal, vibration signal and spindle torque signal after alignment;

[0083] Step 3: When the robot mills each layer, it detects whether chatter occurs based on the collected vibration signal or spindle torque signal. If it occurs, the chatter range is determined based on the index matrix of the tool position signal. The current robot feed speed is adjusted within the chatter range until a stable machining state is achieved, and the adjusted cutting parameters are recorded.

[0084] When detecting whether chatter occurs, the time domain characteristics of the vibration signal or the spindle torque signal can be extracted. When the robot mills the i-th layer, the amplitude of the vibration signal or the spindle torque signal in the time domain is extracted.

[0085] Taking the vibration signal as an example, the threshold value for judging whether chatter occurs is obtained based on the processing experiment and is denoted as y c , then we have:

[0086]

[0087] Among them, N v,i is the sampling information corresponding to the jth sampling point of the collected vibration signal, c is the empirical threshold coefficient selected by experienced engineers based on the processing site conditions, N v Represents the sampling point information sequence of the collected vibration signal.

[0088] When the robot detects chatter during milling, the index matrix M is calculated based on the aligned vibration signal or spindle torque signal. p-v-t Find the index value M in c , according to the index value M c exist Determine the associated tool position signal in the tool, and then determine the vibration range R i for:

[0089]

[0090] The above three steps are carried out simultaneously, such as Figure 2 As shown, at different layers, if the robot detects chatter when milling the i-th layer in step 3, the chatter range R is determined. i After that, when the robot mills the i+1th layer, the chattering range R i When the robot is milling the i-th layer, it uses the adjusted cutting parameters, and the rest of the positions in the i+1 layer keep the original cutting parameters.

[0091] In a preferred embodiment, a safety distance L can also be set. When the robot mills the i+1th layer, the tool is close to the vibration range R in the feed direction. i When the distance is L, the feed speed is gradually reduced to the recorded value, away from the vibration range R i The robot feed rate is gradually restored. If the distance between the two chattering ranges in the i-th layer is less than L, the two chattering ranges are merged into a single chattering range when adjusting the robot feed rate for the i+1 layer. The robot feed rate does not need to be calculated using a formula or intelligent algorithm every time. Due to the similarity of actual machining characteristics, the feed rate can be adjusted in advance within the recorded range when milling the i+1 layer based on the chattering range of the current i-th layer and the adjusted stable cutting parameters.

[0092] During the entire milling process of the robot, the processing monitoring results corresponding to the current processing program can be applied to processing programs with the same type of processing features. When the robot processing positions are greatly different, the robot feed speed needs to be adaptively adjusted in combination with the overall structural stiffness of the robot during the processing process to achieve stable cutting. In a preferred embodiment, the structural stiffness of the robot is used as an adaptive adjustment indicator for the robot feed speed. For the same processing program, the different number of milling layers may cause the robot's posture to have a large difference, such as Figure 3 and Figure 4 As shown in Figure 2, the overall structural stiffness of the robot when milling layer a can be expressed as:

[0093]

[0094] Where a>1, K1 represents the overall structural stiffness when processing the first layer, K a Indicates the overall structural stiffness of the robot when milling the a-th layer; the tool position in the machining program is determined according to the robot joint position signal, K tool,j is the robot end stiffness corresponding to the jth tool position in the corresponding machining program, M1 is the number of tool positions in the machining program corresponding to the first layer, and M a is the number of tool locations in the machining program corresponding to layer a, then:

[0095]

[0096] When the robot is milling layer a and chatter occurs, the robot's feed rate α is readjusted according to the ratio of the robot's overall stiffness until the machining stability is reached. The cutting parameters at this time are recorded. The original feed rate is recorded as α0, and the empirical coefficient λ is obtained from the machining experiment. The adjustment process is as follows:

[0097]

[0098] The method of this embodiment further includes: after the robot mills each layer, outputting a color map of the distribution of the aligned vibration signals at the processing position; the method for obtaining the color map includes:

[0099] All vibration signals N collected from the i-th layer v,j The amplitude in the time domain is divided into P intervals according to the linear function, each interval represents a color index. v,j Normalization:

[0100] y rule =(N v,j -minN v ) / (maxN v -minN v )

[0101] y rule is the vibration signal N v,j Normalized value;

[0102] Calculate the index value y of the interval color index :

[0103] y index =floor(P*y rule -y rule )+1

[0104] After the robot mills the i-th layer, according to the index matrix M p-v-t with y index , draw the color map of the vibration signal distribution on the processing position after alignment on the current i-th layer, such as Figure 6 As shown, it is used for offline analysis of milling status.

[0105] This embodiment also provides a robot adaptive processing system, comprising

[0106] The signal acquisition module is used to collect the tool position signal of the current milling layer during the robot milling process, and simultaneously collect the vibration signal or spindle torque signal and send it to the online monitoring module;

[0107] The online monitoring module is used to extract and align the tool position signal and the vibration signal or the tool position signal and the spindle torque signal respectively during acquisition, and then correlate the tool position signal and the vibration signal or the tool position signal and the spindle torque signal in time sequence to obtain an index matrix of the correlated signals; it is also used to detect whether chatter occurs based on the vibration signal or the spindle torque signal when the robot mills each layer, and if so, determine the chatter occurrence range based on the index matrix of the correlated signals and send it to the adaptive control module; it is also used to output a color map of the distribution of the aligned vibration signals at the processing position to the offline analysis module after the robot mills each layer;

[0108] The adaptive control module is used to control the robot to mill each layer and adjust the current robot feed speed within the range of vibration until the processing reaches a stable state, and record the adjusted cutting parameters; it is also used to detect vibration when the robot mills the i-th layer and determine the vibration range R i When the robot mills the i+1th layer, the chattering range R i When the robot is milling the i-th layer, the robot uses the adjusted cutting parameters, and the rest of the positions in the i+1 layer keep the original cutting parameters;

[0109] The offline analysis module is used to analyze the vibration distribution of each layer generated during the robot milling process based on the color map of the vibration signal distribution at the processing position.

[0110] The signal acquisition module is also used to collect the robot joint position signals during the robot milling process of each layer;

[0111] The adaptive control module is also used when the robot mills the ath layer, a>1, when chatter occurs and the current overall structure stiffness change coefficient When the current feed speed α0 is adjusted to λ is the set threshold;

[0112]

[0113] in, k=1,a, K1 represents the overall structural stiffness when processing the first layer, K a Indicates the overall structural stiffness of the robot when milling the a-th layer; the tool position in the machining program is determined according to the robot joint position signal, K tool,j is the robot end stiffness corresponding to the jth tool position in the corresponding machining program, M1 is the number of tool positions in the machining program corresponding to the first layer, and M a is the number of tool locations in the machining program corresponding to layer a.

[0114] The specific implementation of the functions of each module is the same as the above-mentioned robot adaptive processing method.

[0115] Although the present invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely illustrative of the principles and applications of the invention. It should be understood that many modifications may be made to the illustrative embodiments, and that other arrangements may be devised, without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in ways other than those described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be employed in conjunction with other described embodiments.

Claims

1. A robot adaptive machining method, characterized in that: include: During the robot milling process of each layer, the tool position signal of the current layer is collected, and the vibration signal or spindle torque signal is collected at the same time; During the acquisition, the tool position signal and the vibration signal or the tool position signal and the spindle torque signal are extracted and aligned respectively, and then the tool position signal and the vibration signal or the tool position signal and the spindle torque signal are correlated in time sequence to obtain an index matrix of the correlation signal; When the robot mills each layer, it detects whether chatter occurs based on the vibration signal or spindle torque signal. If so, it determines the chatter range based on the index matrix of the associated signal. The robot feed speed is adjusted within the chatter range until the machining reaches a stable state, and the adjusted cutting parameters are recorded. If the robot detects chatter when milling layer i and determines the range of chatter When the robot mills the i+1th layer, the vibration occurs within the milling range. When the robot is milling the i-th layer, the robot uses the adjusted cutting parameters, and the rest of the positions in the i+1 layer keep the original cutting parameters; The method further comprises: During the robot milling process of each layer, the robot joint position signals are also collected; Robot milling Layer, , when flutter occurs and the current overall structural stiffness change coefficient When the current feed speed Adjust to , To set the threshold; ； in, , , Indicates the overall structural stiffness when processing the first layer, Indicates robot milling The overall structural stiffness of the layer; Determine the tool position in the machining program based on the robot joint position signal. is the robot end stiffness corresponding to the j-th tool position in the corresponding machining program, The number of tool positions in the machining program corresponding to the first layer. For the The number of tool locations in the machining program corresponding to the layer.

2. The robot adaptive processing method according to claim 1, characterized in that: During the acquisition, the tool position signal and vibration signal or the tool position signal and the spindle torque signal are extracted and aligned respectively, and then the tool position signal and the vibration signal or the tool position signal and the spindle torque signal are correlated in time sequence to obtain the index matrix of the correlation signal: A sliding window is used to collect data at each moment. While collecting the signal, the data in the sliding window at the previous moment are correlated. The sampling time is recorded as: in, 、 and are the sampling time series of tool position signal, vibration signal and spindle torque signal, respectively. 、 and is the number of sampling points, and the corresponding sampling point information is recorded as: in, 、 and They are the sampling point information sequences of tool position signal, vibration signal and spindle torque signal, according to the timing - - The optimal matching relationship is established to establish the index matrix of the correlation signal : in, is the number of sampling points after matching alignment; The aligned signal sampling points are expressed as: in, 、 and They are the sampling point information sequences of the tool position signal, vibration signal and spindle torque signal after alignment; When the robot detects chatter during milling, the index matrix is ​​generated based on the aligned vibration signal or spindle torque signal. Find the index value in , according to the index value Determine the associated tool position signal and thus the scope of chatter occurrence for: 。 3. The robot adaptive processing method according to claim 1, characterized in that: Method for detecting whether chatter occurs based on vibration signals: in, is the sampling information corresponding to the jth sampling point of the collected vibration signal, is the empirical threshold coefficient, Represents the sampling point information sequence of the collected vibration signal, Indicates the chatter threshold of the vibration signal.

4. The robot adaptive processing method according to claim 2, characterized in that: The method further comprises: After the robot mills each layer, a color map of the vibration signal distribution at the processing position after alignment is output; The methods for obtaining the color map include: All vibration signals collected at layer i The amplitude in the time domain is divided into P intervals according to the linear function, and each interval represents a color index. Normalization: Vibration signal Normalized value; Calculate the index value of the interval color : After the robot mills the i-th layer, according to the index matrix and , a color map of the vibration signal distribution on the processing position after alignment on the current i-th layer is drawn for offline analysis of the milling status.

5. The robot adaptive processing method according to claim 1, characterized in that: When the robot mills the i+1th layer, the vibration occurs within the milling range. When the robot uses the adjusted cutting parameters when milling the i-th layer, the method includes: When the robot is milling the i+1th layer, the safety distance for adjusting the robot feed speed is set to L, and the tool is close to the vibration range in the feed direction. When the distance is L, the feed speed is gradually reduced to the recorded value, away from the vibration range. The robot feed speed is gradually restored. If the distance between the two vibration ranges in the i-th layer is less than L, the two vibration ranges are merged into one vibration range when the robot feed speed is adjusted in the i+1-th layer.

6. Robotic adaptive machining system, characterized in that, include: The signal acquisition module is used to collect the tool position signal of the current milling layer during the robot milling process, and simultaneously collect the vibration signal or spindle torque signal and send it to the online monitoring module; An online monitoring module is used to extract and align the tool position signal and the vibration signal or the tool position signal and the spindle torque signal respectively during acquisition, and then correlate the tool position signal and the vibration signal or the tool position signal and the spindle torque signal in time sequence to obtain an index matrix of the correlation signal; It is also used to detect whether chatter occurs based on the vibration signal or spindle torque signal when the robot mills each layer. If it does occur, the chatter range is determined based on the index matrix of the associated signal and sent to the adaptive control module. It is also used to output a color map of the aligned vibration signal distribution at the processing position to the offline analysis module after the robot mills each layer. The adaptive control module is used to control the robot to mill each layer and adjust the current robot feed speed within the range of vibration until the processing reaches a stable state, and record the adjusted cutting parameters; it is also used to detect vibration and determine the range of vibration when the robot mills the i-th layer When the robot mills the i+1th layer, the vibration occurs within the milling range. When the robot is milling the i-th layer, the robot uses the adjusted cutting parameters, and the rest of the positions in the i+1 layer keep the original cutting parameters; Offline analysis module, used to analyze the vibration distribution of each layer during the robot milling process based on the color map of the vibration signal distribution at the processing position; The signal acquisition module is also used to collect the robot joint position signals during the robot milling process of each layer; Adaptive control module, also used for robot milling Layer, , when flutter occurs and the current overall structural stiffness change coefficient When the current feed speed Adjust to , To set the threshold; in, , , Indicates the overall structural stiffness when processing the first layer, Indicates robot milling The overall structural stiffness of the layer; according to the robot joint position signal to determine the tool position in the processing program, is the robot end stiffness corresponding to the j-th tool position in the corresponding machining program, The number of tool positions in the machining program corresponding to the first layer. For the The number of tool locations in the machining program corresponding to the layer.

7. A robotic adaptive machining device comprising a storage device, a processor, and a computer program stored in the storage device and executable on the processor, wherein: The processor executes the computer program to implement the steps of the robot adaptive machining method according to any one of claims 1 to 5.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the robot adaptive machining method according to any one of claims 1 to 5 are implemented.

Images