A force-position hybrid control method for low-damage milling and grinding of brittle materials

Abstract

The application relates to the technical field of precision machining, and provides a force-position hybrid control method for brittle material low-damage milling and grinding, which comprises the following steps: planning a machining path and a force-position hybrid control strategy based on target brittle material; performing milling and grinding on the target brittle material; collecting multi-modal sensing data in real time during the milling and grinding; identifying a machining state and calculating a key control deviation based on the multi-modal sensing data; and adjusting the force-position hybrid control strategy in real time based on the machining state and the key control deviation. The application can realize low-damage high-quality milling and grinding of brittle material by sensing the machining state in real time and dynamically adjusting the control strategy while ensuring the precision of the machining path and effectively inhibiting the machining force.

Description

Technical Field

[0001] This invention belongs to the field of precision machining technology, and in particular relates to a force-position hybrid control method for low-damage milling of brittle materials. Background Technology

[0002] Brittle materials are widely used in aerospace, optical components, semiconductors and other fields. Their hard and brittle properties make them prone to defects such as microcracks, chipping, and subsurface damage during conventional processing.

[0003] Traditional CNC milling and grinding processes mostly adopt position control mode, that is, the tool processes according to a preset trajectory and depth of cut. It cannot intelligently identify the current processing state and make optimal control decisions. When the processing state changes instantaneously, simple control mode switching may lead to system instability due to response lag or frequent switching, which may aggravate material damage or reduce processing quality.

[0004] Therefore, there is an urgent need in this field for an intelligent method that can sense the processing status in real time and dynamically adjust the control strategy accordingly, so as to effectively suppress the processing force while ensuring the accuracy of the processing path, thereby achieving low-damage and high-quality milling of brittle materials. Summary of the Invention

[0005] In view of the above-mentioned deficiencies of the prior art, this invention proposes a force-position hybrid control method for low-damage milling of brittle materials, particularly a control method for achieving low-damage, high-precision machining of brittle materials such as ceramics, glass, and silicon crystals through force-position hybrid control during milling. The technical solution of this invention includes the following steps:

[0006] S1: A mixed force-position control strategy for planning processing paths based on the target brittle material;

[0007] S2: Perform milling on the target brittle material;

[0008] S3: Real-time acquisition of multimodal sensor data during milling;

[0009] S4: Identify processing status and calculate key control deviations based on multimodal sensor data;

[0010] S5: Real-time adjustment of force-position hybrid control strategy based on processing status and key control deviations;

[0011] The force-position hybrid control strategy in S1 includes a position control mode, which uses the tool position as feedback to track the machining path; a force control mode, which uses the cutting force as feedback to maintain constant force machining; and a switching logic, which switches between the position control mode and the force control mode.

[0012] Preferably, the planned processing path in S1 includes:

[0013] Computer-aided manufacturing software is used to generate tool paths during the milling process based on the brittle properties of the target brittle material.

[0014] Preferably, the multimodal sensing data in S3 includes:

[0015] Force sensor data, acceleration sensor data, and position sensor data.

[0016] Preferably, the identification processing state in S4 includes:

[0017] S41: The data sample set composed of multimodal sensing data is divided based on a preset clustering algorithm. Each data point is a multidimensional sensor reading vector at a single sampling time. After division, multiple data groups are formed.

[0018] S42: Count the number of data points contained in each data group and compare it with the preset threshold for the number of dominant states;

[0019] S43: Groups with a number of data points greater than or equal to the threshold of the number of dominant states are determined as dominant state groups representing stable processing states.

[0020] S44: Groups with a number of data points less than the threshold of the number of dominant states are identified as transient state groups that characterize the instantaneous processing state.

[0021] Preferably, the cluster center update method of the preset clustering algorithm in S41 includes:

[0022] Define the search trajectory, construct the objective function to guide the optimization, and update the cluster centers by finding the extreme points on the search trajectory that minimize the objective function.

[0023] The search trajectory is defined by the following formula:

[0024]

[0025] The objective function is defined as follows:

[0026]

[0027] The search trajectory seeks the extreme point that minimizes the objective function, as shown in the following formula:

[0028]

[0029] In the formula, Let be the coordinate vector of the cluster center. This is the updated cluster center coordinate vector. For the i-th data point, The total number of data points in the data sample set. This is an auxiliary function used to find the minimum point. For gradient operators, The set of all data points contained in a data group. For data groups The average vector of coordinates of all data points in the dataset. This is the matrix transpose operator.

[0030] Preferably, the calculation of key control deviations in S4 includes:

[0031] The multimodal sensing data is filtered and normalized to form a standardized data vector, and the cutting force deviation and position deviation are calculated to output the key control deviation.

[0032] The cutting force deviation is calculated using the following formula:

[0033]

[0034] The positional deviation is calculated using the following formula:

[0035]

[0036] The key control deviation is calculated using the following formula:

[0037]

[0038] In the formula, This is the real-time cutting force vector. For the ideal cutting force vector, This is the actual position vector of the tool. Let the desired position vector be the path to be planned. and These are the weighting coefficients.

[0039] Preferably, S5 includes:

[0040] S51: Real-time input of processing status and key control deviations;

[0041] S52: When in a stable processing state, maintain the current force-position hybrid control strategy;

[0042] S53: When in a momentary machining state, and the cutting force deviation is greater than the position deviation, execute the switching logic to switch to force control mode;

[0043] S54: When in a momentary machining state, and the cutting force deviation is less than or equal to the position deviation, execute the switching logic to switch to the position control mode.

[0044] Preferably, S5 further includes:

[0045] A switching hysteresis interval is defined, wherein the switching hysteresis interval is defined by a first hysteresis threshold and a second hysteresis threshold;

[0046] When the switching logic is executed, the switching hysteresis interval is introduced. When the machine is in an instantaneous machining state and the result of the cutting force deviation minus the position deviation is continuously greater than the first hysteresis threshold, the machine switches to the force control mode. Otherwise, the current force-position hybrid control strategy is maintained.

[0047] When the switching logic is executed, the switching hysteresis interval is introduced. When the machine is in an instantaneous machining state and the result of the position deviation minus the cutting force deviation is continuously greater than the second hysteresis threshold, the machine switches to the position control mode. Otherwise, the current force-position hybrid control strategy is maintained.

[0048] Beneficial effects:

[0049] 1. This application, through the hybrid and intelligent switching of force control and position control, not only ensures the accuracy of the tool path, but also effectively suppresses the peak value of cutting force, preventing damage such as brittle material fragmentation and microcrack propagation caused by excessive force, and significantly improving the integrity and subsurface quality of the machined surface.

[0050] 2. By introducing multimodal sensing data analysis based on clustering algorithms, this application can accurately distinguish between stable processing states and transient abnormal states. This data-driven state recognition method is more robust and accurate than the traditional single threshold judgment, providing a reliable decision basis for real-time adjustment of control strategies.

[0051] 3. Based on the real-time identification of the machining status and the calculation of key control deviations, the system can adaptively execute the optimal control mode switching, ensuring the stability and response quality of the control system and extending the tool life. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating a preferred embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of the identification and processing status flow of a preferred embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of a preferred embodiment of the real-time adjustment force-position hybrid control strategy of the present invention. Detailed Implementation

[0055] The embodiments of the present invention will be described in detail below. The embodiments described below are implemented based on the technical solution of the present invention, and detailed implementation methods and specific operation processes are given. However, the protection scope of the present invention is not limited to the embodiments described below.

[0056] This invention designs a force-position hybrid control method for low-damage milling of brittle materials, such as... Figure 1 As shown, the technical solution includes the following steps, specifically:

[0057] S1: A mixed force-position control strategy for planning processing paths based on the target brittle material;

[0058] S2: Perform milling on the target brittle material;

[0059] S3: Real-time acquisition of multimodal sensor data during milling;

[0060] S4: Identify processing status and calculate key control deviations based on multimodal sensor data;

[0061] S5: Real-time adjustment of force-position hybrid control strategy based on processing status and key control deviations;

[0062] The force-position hybrid control strategy in S1 includes a position control mode, which uses the tool position as feedback to track the machining path; a force control mode, which uses the cutting force as feedback to maintain constant force machining; and a switching logic, which switches between the position control mode and the force control mode.

[0063] Specifically, at the start of machining, the force-position hybrid control strategy sets the initial control mode based on the planned machining path and the target brittle material, and switches between the position control mode and the force control mode based on the real-time milling situation through switching logic. The position control mode is used to ensure that the tool moves accurately along the planned path, while the force control mode is used to maintain a stable cutting force to prevent the brittle material from being damaged by excessive force.

[0064] Preferably, the planned processing path in S1 includes:

[0065] Computer-aided manufacturing software is used to generate tool paths during the milling process based on the brittle properties of the target brittle material.

[0066] Preferably, the multimodal sensing data in S3 includes:

[0067] Force sensor data, acceleration sensor data, and position sensor data.

[0068] Preferably, such as Figure 2 As shown, the identification processing states in S4 include:

[0069] S41: The data sample set composed of multimodal sensing data is divided based on a preset clustering algorithm. Each data point is a multidimensional sensor reading vector at a single sampling time. After division, multiple data groups are formed.

[0070] S42: Count the number of data points contained in each data group and compare it with the preset threshold for the number of dominant states;

[0071] S43: Groups with a number of data points greater than or equal to the threshold for the number of dominant states are identified as dominant state groups representing stable processing states.

[0072] S44: Groups with fewer than the number of data points are identified as transient state groups that represent instantaneous processing states.

[0073] Preferably, the cluster center update method of the preset clustering algorithm in S41 includes:

[0074] Define the search trajectory, construct the objective function to guide the optimization, and update the cluster centers by finding the extreme points on the search trajectory that minimize the objective function.

[0075] The search trajectory is calculated using the following formula:

[0076]

[0077] The objective function is shown in the following formula:

[0078]

[0079] The search trajectory is used to find the extreme point that minimizes the objective function, as shown in the following formula:

[0080]

[0081] In the formula, Let be the coordinate vector of the cluster center. This is the updated cluster center coordinate vector. For the i-th data point, The total number of data points in the data sample set. This is an auxiliary function used to find the minimum point. For gradient operators, The set of all data points contained in a data group. For data groups The average vector of coordinates of all data points in the dataset. This is the matrix transpose operator.

[0082] Specifically, for S41, given initial cluster centers, each data point is assigned to the nearest cluster based on the distance criterion, forming a preliminary data group. Then, based on the current partitioning results, the center position of each cluster is recalculated using a cluster center update method. To better represent the average characteristics of the cluster, the distances to all data points are recalculated and partitioned based on the updated cluster centers. This process is repeated until the positions of the cluster centers no longer change significantly or the maximum number of iterations is reached. Once the cluster centers stabilize, the algorithm stops iterating. At this point, each data point has been stably assigned to a specific group based on its distance from the final cluster center. These resulting groups, with high internal similarity and low inter-group similarity, constitute the multiple data clusters output.

[0083] Furthermore, for S43 and S44, the data points in the data group are a multidimensional vector composed of all multimodal sensor readings at a specific sampling moment. It represents a complete snapshot of the machining system's state at that moment. Ideally, the machining process should always be in a stable and controllable state. Therefore, the vast majority of sensor data points will be tightly clustered in one or a few "normal state" regions. The number of data points in this region will be very large because it represents the dominant stable machining period, i.e., the dominant state group characterizing the stable machining state. Conversely, when abnormalities such as tool chipping, local material inhomogeneity, or increased vibration occur, the process will briefly deviate from the stable state and enter a different abnormal state. This state is usually instantaneous, so the corresponding sensor data will form some data groups with a smaller number of data points, i.e., transient state groups characterizing the instantaneous machining state.

[0084] Preferably, the calculation of key control deviations in S4 includes:

[0085] The multimodal sensing data is filtered and normalized to form a standardized data vector, and the cutting force deviation and position deviation are calculated to output the key control deviation.

[0086] The cutting force deviation is calculated using the following formula:

[0087]

[0088] Positional deviation, calculated using the following formula:

[0089]

[0090] The key control deviation is calculated using the following formula:

[0091]

[0092] In the formula, This is the real-time cutting force vector. For the ideal cutting force vector, This is the actual position vector of the tool. Let the desired position vector be the path to be planned. and These are the weighting coefficients.

[0093] Preferably, such as Figure 3 As shown, S5 includes:

[0094] S51: Real-time input of processing status and key control deviations;

[0095] S52: When in a stable processing state, maintain the current force-position hybrid control strategy;

[0096] S53: When in a momentary machining state, and the cutting force deviation is greater than the position deviation, execute the switching logic to switch to force control mode;

[0097] S54: When in a momentary machining state, and the cutting force deviation is less than or equal to the position deviation, execute the switching logic to switch to the position control mode.

[0098] Preferably, S5 further includes:

[0099] Set a switching hysteresis interval, which is defined by a first hysteresis threshold and a second hysteresis threshold;

[0100] When the switching logic is executed, a switching hysteresis interval is introduced. When the machine is in an instantaneous machining state and the result of the cutting force deviation minus the position deviation is continuously greater than the first hysteresis threshold, the machine switches to the force control mode. Otherwise, the current force-position hybrid control strategy is maintained.

[0101] When the switching logic is executed, a switching hysteresis interval is introduced. When the machine is in an instantaneous machining state and the result of the position deviation minus the cutting force deviation is continuously greater than the second hysteresis threshold, the machine will switch to the position control mode. Otherwise, the current force-position hybrid control strategy will be maintained.

[0102] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A force-position hybrid control method for low-damage milling of brittle materials, characterized in that, include: S1: A mixed force-position control strategy for planning processing paths based on the target brittle material; S2: Perform milling on the target brittle material; S3: Real-time acquisition of multimodal sensor data during milling; S4: Identify processing status and calculate key control deviations based on multimodal sensor data; S5: Real-time adjustment of force-position hybrid control strategy based on processing status and key control deviations; The force-position hybrid control strategy in S1 includes a position control mode that uses the tool position as feedback to track the machining path. Force control mode, which uses cutting force as feedback to maintain constant force machining; Switching logic allows switching between position control mode and force control mode; The identification processing status in S4 includes: S41: The data sample set composed of multimodal sensing data is divided based on a preset clustering algorithm. Each data point is a multidimensional sensor reading vector at a single sampling time. After division, multiple data groups are formed. S42: Count the number of data points contained in each data group and compare it with the preset threshold for the number of dominant states; S43: Groups with a number of data points greater than or equal to the threshold of the number of dominant states are determined as dominant state groups representing stable processing states. S44: Groups with a number of data points less than the threshold for the number of dominant states are identified as transient state groups that characterize the instantaneous processing state. The cluster center update method of the preset clustering algorithm in S41 includes: Define the search trajectory, construct the objective function to guide the optimization, and update the cluster centers by finding the extreme points on the search trajectory that minimize the objective function. The search trajectory is defined by the following formula: ； The objective function is defined as follows: ； The search trajectory seeks the extreme point that minimizes the objective function, as shown in the following formula: ； In the formula, Let be the coordinate vector of the cluster center. This is the updated cluster center coordinate vector. For the i-th data point, The total number of data points in the data sample set. This is an auxiliary function used to find the minimum point. For gradient operators, The set of all data points contained in a data group. For data groups The average vector of coordinates of all data points in the dataset. This is the matrix transpose operator; The calculation of key control deviations in S4 includes: The multimodal sensing data is filtered and normalized to form a standardized data vector, and the cutting force deviation and position deviation are calculated to output the key control deviation. The cutting force deviation is calculated using the following formula: ； The positional deviation is calculated using the following formula: ； The key control deviation is calculated using the following formula: ； In the formula, This is the real-time cutting force vector. For the ideal cutting force vector, This is the actual position vector of the tool. Let the desired position vector be the path to be planned. and These are the weighting coefficients.

2. The force-position hybrid control method for low-damage milling of brittle materials according to claim 1, characterized in that, The planned processing path in S1 includes: Computer-aided manufacturing software is used to generate tool paths during the milling process based on the brittle properties of the target brittle material.

3. The force-position hybrid control method for low-damage milling of brittle materials according to claim 1, characterized in that, The multimodal sensing data in S3 includes: Force sensor data, acceleration sensor data, and position sensor data.

4. The force-position hybrid control method for low-damage milling of brittle materials according to claim 3, characterized in that, S5 includes: S51: Real-time input of processing status and key control deviations; S52: When in a stable processing state, maintain the current force-position hybrid control strategy; S53: When in a momentary machining state, and the cutting force deviation is greater than the position deviation, execute the switching logic to switch to force control mode; S54: When in a momentary machining state, and the cutting force deviation is less than or equal to the position deviation, execute the switching logic to switch to the position control mode.

5. The force-position hybrid control method for low-damage milling of brittle materials according to claim 4, characterized in that, The S5 also includes: A switching hysteresis interval is defined, wherein the switching hysteresis interval is defined by a first hysteresis threshold and a second hysteresis threshold; When the switching logic is executed, the switching hysteresis interval is introduced. When the machine is in an instantaneous machining state and the result of the cutting force deviation minus the position deviation is continuously greater than the first hysteresis threshold, the machine switches to the force control mode. Otherwise, the current force-position hybrid control strategy is maintained. When the switching logic is executed, the switching hysteresis interval is introduced. When the machine is in an instantaneous machining state and the result of the position deviation minus the cutting force deviation is continuously greater than the second hysteresis threshold, the machine switches to the position control mode. Otherwise, the current force-position hybrid control strategy is maintained.

Images