Search

CN-120949694-B - Dynamic precision compensation control method based on intelligent PCB drilling machine

CN120949694BCN 120949694 BCN120949694 BCN 120949694BCN-120949694-B

Abstract

The application relates to a dynamic precision compensation control method based on an intelligent PCB drilling machine, which comprises the steps of collecting a dynamic parameter set in real time through a multi-source sensor group arranged on a drilling spindle, wherein the dynamic parameter set comprises displacement deviation caused by axial deformation of the spindle, thermal deformation amount based on interlayer temperature gradient analysis and spectral characteristics reflecting cutter-material coupling vibration, dynamically synthesizing a space trajectory correction instruction and a motion parameter optimization instruction according to the multi-dimensional compensation vector and current drilling process parameters, reconstructing a real-time interpolation path through a B spline curve by the trajectory correction instruction, receiving the space trajectory correction instruction and the motion parameter optimization instruction by a motion controller, completing at least 3 compensation iterations in a single drilling period, and driving a linear motor and a spindle motor to execute online precision compensation, wherein the execution delay of the compensation instruction is less than 2ms.

Inventors

  • YE YANFEI

Assignees

  • 梅州市鸿宇电路板有限公司

Dates

Publication Date
20260508
Application Date
20250909

Claims (7)

  1. 1. The dynamic precision compensation control method based on the intelligent PCB drilling machine is characterized by comprising the following steps: Collecting a dynamic parameter set in real time through a multi-source sensor group arranged on a drill spindle, wherein the dynamic parameter set comprises displacement deviation caused by axial deformation of the spindle, thermal deformation amount based on interlayer temperature gradient analysis and spectral characteristics reflecting cutter-material coupling vibration; The multisource sensor group at least comprises a displacement sensor for acquiring the displacement deviation, a temperature sensor array for acquiring a temperature gradient and an acceleration sensor for acquiring a vibration spectrum; Inputting the dynamic parameter set into a machine learning model of online incremental training, wherein the machine learning model is a long-term and short-term memory network embedded with physical constraints, the physical constraints comprise thermal expansion coefficients and material stiffness matrixes, and the machine learning model synchronously decouples interaction errors of a thermal engine coupling effect and a vibration transmission chain to generate multidimensional compensation vectors in real time; according to the multidimensional compensation vector and the current drilling process parameters, the drilling process parameters comprise a plate laminated structure, a target aperture and a hole depth, a real-time interpolation path is reconstructed through a B spline curve based on the rigidity modulus distribution of the plate laminated structure, a space track correction instruction is dynamically synthesized, and a motion parameter optimization instruction is dynamically synthesized based on a power spectrum density analysis of a vibration spectrum and a hole wall roughness prediction model; and the motion controller receives the space track correction instruction and the motion parameter optimization instruction, completes at least 3 compensation iterations in a single drilling period, and drives the linear motor and the spindle motor to execute online precision compensation, wherein the whole-course delay of the compensation instruction from generation to execution is less than 2ms.
  2. 2. The method for controlling dynamic accuracy compensation based on an intelligent PCB drilling machine according to claim 1, wherein the multi-source sensor group comprises an acceleration sensor, a temperature sensor and a torque sensor; The real-time acquisition process comprises the following steps: the sensor data is subjected to fast Fourier transform to convert frequency domain signals, and denoising is performed through moving average filtering and wavelet transform; and extracting root mean square characteristics of vibration signals, average temperature characteristics of temperature data and peak torque characteristics of torque data.
  3. 3. The method for controlling dynamic accuracy compensation based on an intelligent PCB drilling machine according to claim 1, wherein the multi-source sensor group further comprises a displacement sensor, an infrared thermometer and an acoustic sensor; the displacement deviation is processed smoothly by a Kalman filtering algorithm, the thermal deformation is analyzed by calculating the temperature difference between adjacent measuring points, and the frequency spectrum characteristic is extracted by short-time Fourier transformation.
  4. 4. The method for controlling dynamic precision compensation based on an intelligent PCB drilling machine according to claim 1, wherein the machine learning model is a long-term and short-term memory network, the hidden layer is set to 128 units, the physical constraint comprises a thermal expansion coefficient of an embedded loss function and a material stiffness matrix, the weight is updated once per minute in online incremental training, and the incremental data batch size is 64.
  5. 5. The method for controlling dynamic accuracy compensation based on an intelligent PCB drilling machine according to claim 1, wherein the process of dynamically synthesizing the spatial trajectory correction command comprises: based on the stiffness modulus distribution of the plate laminated structure, reconstructing a path through a B spline curve by using a 5-order basis function, and triggering Kalman filtering to optimize the path in real time when the interpolation parameter deviates from a threshold value.
  6. 6. The method for controlling dynamic accuracy compensation based on an intelligent PCB drilling machine according to claim 1, wherein the process of generating the motion parameter optimization instruction comprises: and initializing the spindle rotating speed and the feeding speed according to the target aperture and the aperture depth, dynamically adjusting the spindle rotating speed based on the power spectral density analysis of the vibration spectrum, and iteratively optimizing the axial feeding acceleration through a hole wall roughness prediction model.
  7. 7. The method for controlling dynamic precision compensation based on the intelligent PCB drilling machine according to claim 1, wherein the compensation iteration process comprises dividing a drilling period into a plurality of time periods through time sequence segmentation, calculating a driving signal set of the linear motor and the spindle motor for each time period, and adopting sliding window filtering to optimize signal values in real time when the signal set exceeds a threshold value.

Description

Dynamic precision compensation control method based on intelligent PCB drilling machine Technical Field The application relates to the technical field of data processing, in particular to a dynamic precision compensation control method based on an intelligent PCB drilling machine. Background The drilling quality directly affects the performance and assembly reliability of parts, the current drilling processing faces remarkable technical challenges, displacement deviation caused by axial deformation of a main shaft, thermal deformation caused by interlayer temperature gradient and spectral characteristic interaction caused by cutter-material coupling vibration are caused, so that the aperture precision is unstable and the quality of a hole wall is reduced, the interaction error of a heat engine coupling effect and vibration is particularly prominent in high-speed processing, dynamic change is difficult to adapt to traditional static compensation, and the processing precision is difficult to meet high standard requirements; In the prior art, parameters are optimized by relying on cutter wear prediction and dynamic balance correction and combining visual detection and genetic algorithm, but the method is focused on blade processing, lacks a real-time decoupling and dynamic compensation mechanism for main shaft deformation, thermal deformation and vibration in drilling, and cannot cope with complex dynamic environments of composite material drilling; The key problem of the technical concern is how to decouple the principal axis deformation, the heat engine coupling effect and the interaction error of the cutter-material vibration in real time in the high-precision drilling process, so as to realize the stable control of the aperture precision and the hole wall quality, especially in the multi-layer composite material processing, the difference of the thermal expansion coefficient and the dynamic change of the high-speed processing exacerbate the error accumulation, and a method capable of generating a low-delay compensation instruction in real time is needed, and meanwhile, the repeated iterative correction is ensured to be completed in a single drilling period, so that the requirement of track deviation and vibration suppression under the complex processing condition is met, and the processing efficiency and the part quality are further improved. Disclosure of Invention In order to solve the problems in the prior art, the application aims to provide a dynamic precision compensation control method based on an intelligent PCB drilling machine. The application discloses a dynamic precision compensation control method based on an intelligent PCB drilling machine, which comprises the following steps: s101, acquiring a dynamic parameter set in real time through a multi-source sensor set arranged on a drill spindle; S102, the dynamic parameter set comprises displacement deviation caused by axial deformation of a main shaft, thermal deformation amount based on interlayer temperature gradient analysis and spectral characteristics reflecting cutter-material coupling vibration; s103, inputting the dynamic parameter set into an online incremental training machine learning model, synchronously decoupling interaction errors of a coupling effect of a heat engine and a vibration transmission chain through a neural network architecture embedded with physical constraints, and generating a multidimensional compensation vector in real time, wherein the multidimensional compensation vector comprises a space track offset used for counteracting deformation accumulated errors, a spindle rotating speed dynamic correction coefficient used for restraining resonance, and an axial feeding acceleration compensation value used for optimizing the quality of a hole wall; s104, dynamically synthesizing a space track correction instruction and a motion parameter optimization instruction according to the multidimensional compensation vector and current drilling process parameters (including a plate laminated structure, a target aperture and a hole depth), wherein the track correction instruction reconstructs a real-time interpolation path through a B spline curve; s105, the motion controller receives the space track correction instruction and the motion parameter optimization instruction, completes at least 3 compensation iterations in a single drilling period, and drives the linear motor and the spindle motor to execute online precision compensation, wherein the execution delay of the compensation instruction is smaller than 2ms. Further, in step S101, the collecting, in real time, the dynamic parameter set by the multi-source sensor set installed on the drill spindle includes: the multisource sensor collects spindle rotation speed, vibration and temperature data in real time to form a dynamic stream, a first data stream is generated through denoising and standardization, an analysis foundation is laid, and if the parameters exceed a threshold