Distributed data acquisition and control system for batched production of wooden boxes

Abstract

The invention relates to the technical field of industrial automation control, and discloses a distributed data acquisition and control system for wooden box batch production, which comprises upstream and downstream logic nodes connected through an industrial Ethernet bus, and each node is provided with a distributed edge controller, a stress memory clamp array and a micro-vibration impedance execution terminal. The distributed edge controller utilizes a shift register logic module to control a data packet containing a unique identity label, a stress field reference state vector and a material dynamic impedance distribution map to synchronously transfer along with the material. According to the method, the stress field is frozen at the upstream station, and the boundary constraint is reconstructed at the downstream station according to the data packet, so that the secondary clamping deformation deviation is eliminated; the impedance spectrum of the material is obtained through micro-vibration scanning, feed-forward self-adaptive control is executed, compliance processing according to the heterogeneous characteristic of wood is achieved, and the precision and quality of batched production of wooden boxes are improved.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, specifically a distributed data acquisition and control system for mass production of wooden boxes. Background Technology

[0002] Wooden crate packaging products are widely used in the logistics and warehousing industries, and mass production of wooden crates typically involves multi-process collaborative operations. Existing automated production lines usually employ independent control architectures, lacking a synchronization mechanism between physical material transfer and production data flow. When workpieces move between different workstations, process parameters and status information cannot be inherited. Downstream processing equipment struggles to obtain the physical characteristic history generated by previous processes and must perform initialization settings for each work cycle. This logical isolation reduces the overall collaborative efficiency of mass production.

[0003] Wood raw materials possess biological characteristics, exhibiting anisotropy and heterogeneity. Compared to metal materials, wood has lower stiffness and is prone to elastic deformation under external forces. In multi-station machining processes, workpieces undergo repeated clamping and releasing operations. Conventional fixtures primarily rely on fixed positions for clamping. When a workpiece is transferred to a subsequent station for secondary clamping, the internal stress distribution changes. The disappearance of the original boundary constraints leads to elastic recovery deformation. Inconsistent boundary conditions between processes cause positioning datum deviations, reducing the final assembly accuracy.

[0004] Natural wood contains internal defects and density variations, resulting in a randomly distributed grain. Traditional CNC machining systems typically use constant feed rates and spindle speeds to execute preset toolpaths. The control algorithms lack the ability to detect sudden changes in local material properties ahead of the toolpath. When the tool enters high-hardness areas, the rigid cutting strategy induces vibrations and generates excessive and unstable cutting forces. This interaction mechanism accelerates tool wear and creates defects on the workpiece surface. Existing open-loop control modes cannot dynamically adjust servo parameters based on material impedance properties. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a distributed data acquisition and control system for mass production of wooden boxes. It solves the problems of existing wooden box production lines, such as the lack of a synchronization mechanism between physical flow and data flow, which leads to the discontinuity of status information between workstations, the deviation in processing benchmark consistency caused by stress release accompanying wood elastic deformation and secondary clamping, and the lack of feedforward perception and adaptive adjustment of the heterogeneous characteristics of materials, which leads to processing defects.

[0006] To achieve the above objectives, the present invention provides a distributed data acquisition and control system for mass production of wooden boxes.

[0007] This invention provides a distributed data acquisition and control system for mass production of wooden crates, comprising multiple logical nodes and an industrial Ethernet bus. The multiple logical nodes are physically divided into upstream and downstream workstations and established communication connections through the industrial Ethernet bus. The upstream workstation is configured to perform initial processing and physical state acquisition of the wooden crate materials, while the downstream workstation is configured to receive the wooden crate materials and perform subsequent processing based on the data. Each logical node includes a distributed edge controller, a stress memory fixture array, and a micro-vibration impedance execution terminal. The distributed edge controller establishes electrical connections with the stress memory fixture array and the micro-vibration impedance execution terminal, respectively.

[0008] The distributed edge controller is internally configured with a shift register logic module to map the production line as a data pipeline and control the synchronous transfer of data packets corresponding to the wooden crate materials between distributed edge controllers on different logic nodes. The data packets contain physical characteristic data of the wooden crate materials in the current production cycle, including at least a unique identification code for the wooden crate materials, a stress field reference state vector collected and frozen at the upstream workstation, and a dynamic impedance distribution map of the material generated by scanning through the micro-vibration impedance execution terminal.

[0009] The stress field reference state vector records the boundary constraint states applied to the wooden box material by all stress memory fixture arrays in the upstream station, including the force vector reading and position coordinate vector reading output by the stress memory fixture array when it reaches steady-state clamping. The material dynamic impedance distribution spectrum is two-dimensional scalar field data, configured to map the local dynamic stiffness values ​​of the wooden box material at different planar coordinate points; these local dynamic stiffness values ​​are generated by the distributed edge controller parsing the feedback signal from the micro-vibration impedance execution terminal.

[0010] The stress memory fixture array comprises multiple servo fixture units arranged around the processing area. Each servo fixture unit includes a servo drive motor, a precision ball screw, a force sensor, a clamping contact, and a position encoder. The servo drive motor drives the precision ball screw to bring the clamping contact into contact with the wooden crate material; the force sensor is located at the interface between the clamping contact and the precision ball screw to measure the normal contact pressure; the position encoder provides feedback on the linear position of the clamping contact.

[0011] The micro-vibration impedance actuator includes a main drive mechanism, a vibration isolation flange, a piezoelectric ceramic stack, and a machining tool head, all connected in series axially along the direction of gravity. The main drive mechanism is a linear servo actuator. The vibration isolation flange, located between the main drive mechanism and the piezoelectric ceramic stack, is made of a high-damping viscoelastic composite material and is configured to attenuate high-frequency mechanical vibrations. The piezoelectric ceramic stack is configured to transmit mechanical micro-vibration signals and receive force response signals from the wooden crate material. The piezoelectric ceramic stack can be a micro-vibration detection module with a prestressed encapsulation structure, including an upper connecting flange, a lower output flange, a multi-layer piezoelectric ceramic ring, and a central preload bolt. An axial static preload force greater than the dynamic inertial force under maximum working acceleration is applied through the central preload bolt. Alternatively, the micro-vibration impedance actuator may also include an air-coupled ultrasonic transceiver module, which uses an ultrasonic transmitting unit and an ultrasonic receiving unit to calculate the sound pressure reflection coefficient, then inverts the acoustic impedance and converts it into an equivalent dynamic elastic modulus to generate the dynamic impedance distribution spectrum of the material.

[0012] The distributed edge controller integrates a digital signal generator module and a digital lock-in amplifier module, and is externally connected to a high-voltage power amplifier and a charge amplifier. The digital signal generator module generates a micro-vibration excitation voltage signal, which drives the piezoelectric ceramic stack to generate mechanical vibration via the high-voltage power amplifier. The charge amplifier converts the charge signal output by the piezoelectric ceramic stack into a force response signal. The digital lock-in amplifier module performs quadrature demodulation on the force response signal, extracts the in-phase and quadrature components, and calculates the amplitude and phase lag of the force response signal relative to the excitation signal. The amplitude and phase lag are then used to calculate the local dynamic stiffness of the contact point.

[0013] The distributed edge controller is also configured to execute specific process control logic:

[0014] The stress field freezing process is performed at the upstream workstation: the global stress change rate of the stress memory fixture array is monitored. When it is lower than the preset steady-state judgment threshold, the wooden box material is judged to have reached mechanical steady state, triggering the synchronous latching command, reading and recording the steady-state force value and steady-state coordinate value of all servo fixture units to construct the stress field reference state vector.

[0015] The stress field reconstruction process is performed at the downstream workstation: the stress field reference state vector in the data packet is parsed, the steady-state force value and steady-state coordinate value are set as the control target value, and the deviation between the current physical state and the stress field reference state vector is minimized by adjusting the feed position of the servo fixture unit until the global objective function is less than the preset convergence threshold.

[0016] In the downstream station, feedforward adaptive machining control is performed: the predicted machining point coordinates for future moments are calculated based on the current motion state of the main drive mechanism, and the predicted machining point coordinates are mapped to the material dynamic impedance distribution spectrum in the data packet, and the feedforward stiffness parameters of the corresponding grid cells are extracted; based on the feedforward stiffness parameters, the target servo stiffness coefficient and target servo damping coefficient of the main drive mechanism are dynamically adjusted to change the compliance characteristics of the machining tool head with the material changes of the wooden box.

[0017] This invention provides a distributed data acquisition and control system for mass production of wooden crates. It offers the following advantages:

[0018] 1. This invention maps the production line into a data pipeline by configuring a shift register logic module inside the distributed edge controller, and controls the synchronous transfer of data packets containing unique identification codes, stress field reference state vectors, and material dynamic impedance distribution maps between logic nodes. This achieves spatiotemporal synchronization of the physical entity flow of wooden box materials and the transmission of digital feature information, eliminating the physical state information gap in the multi-station discrete manufacturing process.

[0019] 2. This invention constructs a stress field reference state vector that records steady-state force and coordinate values ​​by performing a stress field freezing process at the upstream station, and performs a stress field reconstruction process at the downstream station based on the data packet parsing results. It controls the servo fixture unit to adjust the feed position to minimize physical state deviation and reproduces the boundary constraint conditions of the wooden box material at different processing nodes, reducing the impact of elastic deformation caused by secondary clamping on the consistency of processing reference.

[0020] 3. This invention generates a dynamic impedance distribution map of the material by scanning the terminal through micro-vibration impedance execution, and uses the dynamic solution module to predict the coordinates of the processing point to extract the feedforward stiffness parameters of the corresponding grid unit. Then, it dynamically adjusts the target servo stiffness coefficient and damping coefficient of the main drive mechanism, realizing compliant processing based on the non-homogeneous characteristics of the wooden box material, and reducing the risk of processing defects caused by sudden changes in local hardness of the material. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall architecture of the distributed data acquisition and control system for mass production of wooden boxes according to the present invention;

[0022] Figure 2 This is a diagram illustrating the data structure definition and signal processing logic framework of the present invention.

[0023] Figure 3 This is a flowchart illustrating the overall process control of the system according to the present invention.

[0024] Figure 4 This is a block diagram of the adaptive impedance control loop and algorithm logic of the present invention.

[0025] Among them, 100 is a multiple logic node; 110 is a distributed edge controller; 111 is a shift register logic module; 120 is a stress memory fixture array; 121 is a servo fixture unit; 122 is a force sensor; 123 is a position encoder; 124 is a servo drive motor; 126 is a clamping contact; 125 is a precision ball screw; 130 is a micro-vibration impedance actuator terminal; 131 is a machining tool head; 132 is a piezoelectric ceramic stack; 133 is a main drive mechanism; 134 is a vibration isolation flange; 200 is an industrial Ethernet bus; 201 is an upstream station; 202 is a downstream station; 300 is a micro-vibration detection module; 310 is an upper connecting flange; 320 is... Polymer insulating gasket; 330, Multilayer piezoelectric ceramic ring; 340, Center preload bolt; 350, Lower output flange; 360, Protective housing; 410, Digital signal generator module; 420, Digital lock-in amplifier module; 430, High-voltage power amplifier; 440, Charge amplifier; 510, Dynamics calculation module; 520, Impedance synthesis module; 610, Air-coupled ultrasonic transceiver module; 611, Ultrasonic transmitting unit; 612, Ultrasonic receiving unit; 620, Fluid-driven flexible clamp module; 621, Fluid actuation unit; 622, Double-acting fluid cylinder; 623, Integrated linear displacement sensor; 624, Cavity pressure sensor. Detailed Implementation

[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] See attached document Figure 1 This invention provides a distributed data acquisition and control system for mass production of wooden crates, comprising: multiple logical nodes 100 and an industrial Ethernet bus 200. The logical nodes 100 physically correspond to different processing stations on the wooden crate production line. The multiple logical nodes 100 establish a data connection through the industrial Ethernet bus 200.

[0028] The system divides multiple logical nodes 100 into upstream workstations 201 and downstream workstations 202. Upstream workstation 201 is responsible for the initial processing and physical status acquisition of the wooden crate materials. Downstream workstation 202 is responsible for receiving the wooden crate materials and performing subsequent processing. An industrial Ethernet bus 200 is used to transmit digital signals between the logical nodes 100.

[0029] Each logical node 100 includes: a distributed edge controller 110, a stress memory fixture array 120, and a micro-vibration impedance execution terminal 130. The distributed edge controller 110 establishes electrical connections with the stress memory fixture array 120 and the micro-vibration impedance execution terminal 130, respectively. The distributed edge controller 110 is responsible for acquiring data and outputting control commands.

[0030] The stress memory fixture array 120 comprises multiple independent servo fixture units 121. The servo fixture units 121 are arranged around the processing area. Each servo fixture unit 121 integrates a force sensor 122 and a position encoder 123. The force sensor 122 is used to detect the force vector applied when the servo fixture unit 121 contacts the wooden crate material. The position encoder 123 is used to detect the spatial coordinates of the servo fixture unit 121.

[0031] The micro-vibration impedance actuator 130 includes: a machining tool head 131, a piezoelectric ceramic stack 132, and a drive mechanism 133. The piezoelectric ceramic stack 132 is mechanically connected in series between the machining tool head 131 and the drive mechanism 133. The piezoelectric ceramic stack 132 is used to transmit mechanical micro-vibration signals and receive force response signals.

[0032] The distributed edge controller 110 is internally configured with a shift register logic module 111. The shift register logic module 111 maps the entire production line as a data pipeline. When the wooden crate material is transferred from the upstream station 201 to the downstream station 202, the shift register logic module 111 controls the synchronous transfer of the data packet corresponding to the wooden crate material between different distributed edge controllers 110.

[0033] The data packet is defined as a digital accompanying token. The digital accompanying token contains all physical characteristic data of the wooden crate material in the current production cycle. The data structure of the digital accompanying token... The following relationship must be satisfied:

[0034] ;

[0035] In the formula: For the first Each wooden crate material corresponds to a digital accompanying token; A unique identification code for wooden crates; The stress field reference state vector acquired and frozen at upstream station 201; The dynamic impedance distribution spectrum of the material generated by the terminal 130 scan through the micro-vibration impedance execution; This is a time synchronization stamp used for clock alignment between workstations.

[0036] Stress field reference state vector Record the boundary constraint states applied to the wooden crate material by all servo fixture units 121 in upstream station 201. Stress field reference state vector. The following relationship must be satisfied:

[0037] ;

[0038] In the formula: This represents the total number of servo fixture units 121 in the workstation. For the first The state feature sub-vectors of each servo clamp unit 121.

[0039] State feature vector It contains combined information of force vector and position vector. State feature sub-vectors. The following relationship must be satisfied:

[0040] ;

[0041] In the formula: For the first When the servo clamping unit 121 reaches steady-state clamping at the upstream station 201, the force vector reading output by the force sensor 122 is displayed. For the first When the servo clamping unit 121 reaches steady-state clamping at the upstream station 201, the position coordinate vector reading output by the position encoder 123 is obtained.

[0042] Material dynamic impedance distribution spectrum This is two-dimensional scalar field data. Material dynamic impedance distribution spectrum. Mapping wooden crate materials at different planar coordinate points The local dynamic stiffness value at the location. The distributed edge controller 110 generates a material dynamic impedance distribution map by analyzing the feedback signal from the micro-vibration impedance execution terminal 130. .

[0043] The distributed edge controller 110 at upstream workstation 201 will generate a digital accompanying token. Transmitted via industrial Ethernet bus 200. The distributed edge controller 110 at downstream workstation 202 receives the digital accompanying token. Downstream workstation 202 uses a digital accompanying token. Data execution control strategy in the process.

[0044] The micro-vibration impedance actuator 130 includes: a main drive mechanism 133, a vibration isolation flange 134, a piezoelectric ceramic stack 132, and a machining tool head 131.

[0045] The main drive mechanism 133, vibration isolation flange 134, piezoelectric ceramic stack 132, and machining tool head 131 are connected in series axially along the direction of gravity. The main drive mechanism 133 is a linear servo actuator. The output shaft end face of the main drive mechanism 133 is mechanically fixed to the upper surface of the vibration isolation flange 134. The lower surface of the vibration isolation flange 134 is mechanically fixed to the upper surface of the piezoelectric ceramic stack 132. The lower surface of the piezoelectric ceramic stack 132 is mechanically fixed to the mounting handle of the machining tool head 131.

[0046] The vibration isolation flange 134 is made of a high-damping viscoelastic composite material. The vibration isolation flange 134 is located between the main drive mechanism 133 and the piezoelectric ceramic stack 132. The vibration isolation flange 134 is configured to attenuate the propagation of high-frequency mechanical vibrations generated by the piezoelectric ceramic stack 132 to the main drive mechanism 133.

[0047] The piezoelectric ceramic stack 132 is composed of multiple annular piezoelectric ceramic wafers stacked and packaged axially. An axial preload is applied to the piezoelectric ceramic stack 132 via preload bolts. The electrode leads of the piezoelectric ceramic stack 132 are connected to a distributed edge controller 110. The piezoelectric ceramic stack 132 is configured to perform bidirectional conversion between electrical and mechanical energy.

[0048] The axial output displacement characteristics of the piezoelectric ceramic stack 132 are determined by its physical structural parameters and the input electrical signal. Axial displacement response of the piezoelectric ceramic stack 132 The following relationship must be satisfied:

[0049] ;

[0050] In the formula: The axial mechanical expansion and contraction displacement generated by the piezoelectric ceramic stack 132; The number of layers of piezoelectric ceramic wafers inside the piezoelectric ceramic stack 132; The longitudinal piezoelectric strain constant of the piezoelectric material; This is the driving voltage signal applied to the electrode of the piezoelectric ceramic stack 132; The axial reaction force experienced by the machining tool head 131 during machining; This represents the inherent axial equivalent stiffness of the piezoelectric ceramic stack 132.

[0051] The stress memory fixture array 120 includes multiple servo fixture units 121. The multiple servo fixture units 121 are distributed on the base of the machining station according to a predetermined geometry. Each servo fixture unit 121 includes: a servo drive motor 124, a precision ball screw 125, a force sensor 122, a clamping contact 126, and a position encoder 123.

[0052] The output shaft of the servo drive motor 124 is connected to the precision ball screw 125 via a coupling. The precision ball screw 125 is configured to convert the rotary motion of the servo drive motor 124 into linear motion. A clamping contact 126 is mounted on the nut seat of the precision ball screw 125. The clamping contact 126 is configured to directly contact the side surface of the wooden crate material.

[0053] A force sensor 122 is disposed at the interface between the clamping contact 126 and the precision ball screw 125. The force sensor 122 is configured to measure the normal contact pressure on the clamping contact 126. A position encoder 123 is mounted at the tail of the servo drive motor 124. The position encoder 123 is configured to provide feedback on the linear position of the clamping contact 126 in the workstation coordinate system.

[0054] The distributed edge controller 110 is hardwired to the servo drive motor 124, the force sensor 122, and the position encoder 123. The distributed edge controller 110 is configured to acquire the analog voltage signal output by the force sensor 122 and the digital pulse signal output by the position encoder 123.

[0055] See attached document Figure 2 The distributed edge controller 110 is internally configured with volatile storage. Digital accompanying token. It is stored in this volatile storage unit in the form of binary data packets.

[0056] Digital Accompanying Token It is divided into four independent logical storage areas: identity storage area, boundary state storage area, material property storage area, and timing synchronization storage area. The identity storage area stores a unique identity code. The boundary state storage area stores the stress field reference state vector. The material property storage area stores the dynamic impedance distribution spectrum of the material. The time-series synchronization storage area stores time synchronization stamps. .

[0057] Material dynamic impedance distribution spectrum It is instantiated as a two-dimensional discrete numerical matrix in the material property storage area. This two-dimensional discrete numerical matrix represents a meshed mapping of the physical surface of the wooden box material. Material dynamic impedance distribution spectrum. The data structure satisfies the following relationship:

[0058] ;

[0059] In the formula: The number of grid rows along the length of the wooden crate material; This represents the number of grid columns divided along the width of the wooden crate material. For the first Line 1 Local dynamic stiffness measurements at grid points.

[0060] Local dynamic stiffness measurement value Physical coordinates of the wooden crate material surface There exists a one-to-one mapping relationship. This mapping relationship is determined by the machine coordinate system parameters of upstream station 201.

[0061] Shift register logic module 111 is configured to execute digital accompanying tokens. The inter-station transfer operation. The shift register logic module 111 activates the transfer task by listening to the material arrival trigger signal on the industrial Ethernet bus 200.

[0062] When the wooden crate material is from the first The logical node 100 transmits to the first logical node 100 When there are 100 logic nodes, the shift register logic module 111 performs a data migration operation. The data migration operation follows the following discrete-time state transition equation:

[0063] ;

[0064] In the formula: Indicates at time Stored in the first The digital token data is present in each logical node 100; It is a time variable; Indicates at time Stored in the first The digital token data is present in each logical node 100; This is the physical transfer time interval between two adjacent logical nodes 100 for wooden crate materials.

[0065] Time synchronization stamp Record digital accompanying token The global system clock value at creation time. The distributed edge controller 110 utilizes time synchronization stamps. Calculate the age of the data packets. If the current system time is... If the difference exceeds a preset validity period threshold, the distributed edge controller 110 will flag the digital accompanying token. Invalid data.

[0066] The micro-vibration detection module 300 adopts a pre-stressed encapsulation structure. The micro-vibration detection module 300 includes: an upper connecting flange 310, a polymer insulating gasket 320, a multilayer piezoelectric ceramic ring 330, a central pre-tightening bolt 340, a lower output flange 350, and a protective housing 360.

[0067] The upper surface of the upper connecting flange 310 is configured to be mechanically fixed to the output end of the main drive mechanism 133. A first annular positioning groove is provided on the lower surface of the upper connecting flange 310. A second annular positioning groove is provided on the upper surface of the lower output flange 350. The lower end of the lower output flange 350 is configured to mount the machining tool head 131.

[0068] A multilayer piezoelectric ceramic ring 330 is located between the upper connecting flange 310 and the lower output flange 350. Polymer insulating gaskets 320 are respectively disposed between the upper end face of the multilayer piezoelectric ceramic ring 330 and the upper connecting flange 310, and between the lower end face of the multilayer piezoelectric ceramic ring 330 and the lower output flange 350. The polymer insulating gaskets 320 are configured to block the leakage path of the drive voltage signal to the mechanical body structure.

[0069] The center preload bolt 340 passes through the center through hole of the multilayer piezoelectric ceramic ring 330. The threaded end of the center preload bolt 340 is supported in the countersunk hole of the upper connecting flange 310, and the threaded end is screwed into the center threaded hole of the lower output flange 350. The center preload bolt 340 is configured to apply an axial static preload to the multilayer piezoelectric ceramic ring 330.

[0070] The protective housing 360 is cylindrical and is fitted around the multilayer piezoelectric ceramic ring 330. The upper end of the protective housing 360 is sealed to the upper connecting flange 310, and an axial micro-movement clearance is maintained between the lower end and the lower output flange 350. The protective housing 360 is configured to prevent external dust from entering the working area of ​​the multilayer piezoelectric ceramic ring 330.

[0071] Overall axial equivalent stiffness of the micro-vibration detection module The stiffness is determined by the combined stiffness of the multilayer piezoelectric ceramic ring 330 and the central preload bolt 340. The stiffness parameters of this mechanical structure are configured to meet the linear transmission requirements of the micro-vibration signal. Overall axial equivalent stiffness... The following relationship must be satisfied:

[0072] ;

[0073] In the formula: The overall equivalent structural stiffness of the micro-vibration detection module along the axial direction; The elastic modulus of the multilayer piezoelectric ceramic ring 330 material; The effective cross-sectional area of ​​the multilayer piezoelectric ceramic ring 330; The total axial stacking height of the multilayer piezoelectric ceramic rings 330 The elastic modulus of the material of the center preload bolt 340; The cross-sectional area of ​​the center preload bolt 340; The effective force-bearing length of the preloaded bolt 340 at the center.

[0074] Axial static preload applied by center preload bolt 340 The dynamic inertial force is set to be greater than that of the micro-vibration detection module under maximum operating acceleration. Axial static preload. The following relationship must be satisfied:

[0075] ;

[0076] In the formula: The axial static preload applied to the center preload bolt 340; The mass of the machining tool head 131 suspended below the lower output flange 350; The mass of the lower output flange 350 itself; The maximum axial acceleration of the main drive mechanism 133 during the return stroke; It is the acceleration due to gravity; For the engineering safety factor, and .

[0077] The distributed edge controller 110 integrates a digital signal generator module 410 and a digital lock-in amplifier module 420. The digital signal generator module 410 is configured to generate high-frequency excitation waveforms. The digital lock-in amplifier module 420 is configured to analyze the amplitude and phase of the echo signal.

[0078] The distributed edge controller 110 is connected to a high-voltage power amplifier 430 and a charge amplifier 440. The input of the high-voltage power amplifier 430 is connected to the output of the digital signal generator module 410, and the output is connected to the driving electrode of the piezoelectric ceramic stack 132. The input of the charge amplifier 440 is connected to the feedback electrode of the piezoelectric ceramic stack 132, and the output is connected to the input of the digital lock-in amplifier module 420.

[0079] The system defines a contact window as the time domain for micro-vibration detection. The contact window begins when the machining tool head 131 contacts the surface of the wooden crate material and ends when the main drive mechanism 133 initiates its high-energy machining action. Within the contact window, the digital signal generator module 410 generates a micro-vibration excitation voltage signal. Micro-twitch excitation voltage signal The following relationship must be satisfied:

[0080] ;

[0081] In the formula: This refers to the instantaneous driving voltage applied to the piezoelectric ceramic stack 132; This is the DC bias voltage used to keep the piezoelectric ceramic stack 132 in a pre-loaded state; The voltage amplitude of the micro-twitch excitation; The carrier frequency for micro-flicker excitation; This is a time variable within the contact window.

[0082] piezoelectric ceramic stack 132 under micro-twister excitation voltage signal Driven by the material, it generates mechanical vibration and receives mechanical reaction force from the wooden crate material. This mechanical reaction force is converted into a piezoelectric charge signal. The charge amplifier 440 converts the piezoelectric charge signal into a force response signal in the form of an analog voltage. .

[0083] Distributed edge controller 110 responds to force signals The data is converted from analog to digital and then transmitted to the digital lock-in amplifier module 420. The digital lock-in amplifier module 420 uses a quadrature demodulation algorithm to extract the in-phase component. and orthogonal components The orthogonal demodulation process satisfies the following relationship:

[0084] ;

[0085] ;

[0086] In the formula: The value of the in-phase component of the force response signal relative to the excitation signal; The orthogonal component value of the force response signal relative to the excitation signal; This represents the total number of discrete sampling points within the sampling period. The sampling time interval for analog-to-digital conversion; For the first Discrete values ​​of the force response signal at each sampling point; This is the index for the sampling points.

[0087] Digital lock-in amplifier module 420 based on in-phase component and orthogonal components Calculate the amplitude of the force response signal and phase lag Amplitude and phase lag The following relationship must be satisfied:

[0088] ;

[0089] ;

[0090] In the formula: The amplitude modulus of the demodulated force response signal characterizes the dynamic fluctuation intensity of the interfacial contact force. The phase difference between the response signal and the excitation signal characterizes the viscoelastic damping properties of the wooden box material at the contact point.

[0091] The distributed edge controller 110 will calculate the amplitude and phase lag The data is stored in a temporary data buffer. These two parameters are then used to calculate the local dynamic stiffness and to generate a dynamic impedance distribution map of the material. The basic input data.

[0092] The processing unit inside the distributed edge controller 110 is configured to execute physical quantity conversion and impedance calculation algorithms. This algorithm converts the electrical signal characteristic values ​​output by the digital lock-in amplifier module 420 into mechanical characteristic parameters of the wooden crate material.

[0093] The dynamic impedance calculation process is based on a coupled model of the inverse and direct piezoelectric effects of the piezoelectric ceramic stack 132. The distributed edge controller 110 first calculates the effective mechanical excitation displacement amplitude of the piezoelectric ceramic stack 132 during the contact window period. Effective mechanical excitation displacement amplitude It is the amplitude of the micro-twitch excitation voltage. The function of effective mechanical excitation displacement amplitude. The following relationship must be satisfied:

[0094] ;

[0095] In the formula: This represents the actual axial vibration displacement amplitude output by the piezoelectric ceramic stack 132 to the surface of the wooden box material. The displacement transfer efficiency coefficient is calculated to account for adhesive layer losses. The number of piezoelectric wafer layers stacked inside the piezoelectric ceramic stack 132; The longitudinal piezoelectric strain constant of the piezoelectric material; The amplitude of the micro-scream excitation voltage is set by the digital signal generator module 410; The axial stiffness of the piezoelectric ceramic stack 132 itself; The equivalent series stiffness of the mechanical structure of the micro-vibration detection module.

[0096] The distributed edge controller 110 is based on the effective mechanical excitation displacement amplitude. Force response signal amplitude and phase lag Calculate the local dynamic stiffness at the contact point. Local dynamic stiffness Extracting the real part of the self-complexing impedance characterizes the material's resistance to elastic deformation. Local dynamic stiffness. The following relationship must be satisfied:

[0097] ;

[0098] In the formula: For the current detection point The local dynamic stiffness value at the location; The amplitude of the dynamic force response signal demodulated by the digital lock-in amplifier module 420; The effective mechanical excitation displacement amplitude is calculated; The phase lag angle of the force response signal demodulated by the digital lock-in amplifier module 420 relative to the excitation voltage signal.

[0099] The distributed edge controller 110 will calculate the local dynamic stiffness Mapped to material dynamic impedance distribution spectrum In the corresponding mesh element. For any coordinate point on the surface of the wooden box material. Its corresponding mesh stiffness value The decision logic is as follows:

[0100] like If so, then mark the area as a high-hardness nodule area;

[0101] like If so, the area is marked as a low-density decay or cracked area;

[0102] like If so, then mark the area as the standard wood matrix area.

[0103] In the formula: The preset threshold for knot hardness; These are preset defect and porosity thresholds. These two thresholds are stored in the parameter register of the distributed edge controller 110 and are initialized according to the wood species.

[0104] By traversing all sampling periods within the contact window, the distributed edge controller 110 completes the discretization reconstruction of the impedance characteristics of the current contact region, generating a dynamic impedance distribution map of the material. It was then encapsulated in a digital companion token. Transmission takes place within the specified time.

[0105] See attached document Figure 3 The stress field freezing process is primarily executed by the distributed edge controller 110 at the upstream station 201. This process aims to eliminate the influence of wood viscoelastic creep on positioning accuracy and convert steady-state physical boundary conditions into digital signals. When the wood crate material is conveyed to the processing area of ​​the upstream station 201, the distributed edge controller 110 sends a cooperative clamping command to the stress memory fixture array 120. The servo clamping units 121 operate simultaneously, contacting and pressing the sidewalls of the wooden box material along the normal direction. Due to the viscoelastic physical characteristics of wood, the stress state at the contact interface will nonlinearly decay over time in the initial stage of clamping contact.

[0106] The distributed edge controller 110 enters the stress relaxation monitoring phase. In this phase, the distributed edge controller 110 continuously reads the real-time force values ​​of the force sensors 122 in each servo gripper unit 121 at a preset high-frequency sampling rate. The distributed edge controller 110 calculates the global stress change rate of the entire gripper system. Global stress change rate The following relationship must be satisfied:

[0107] ;

[0108] In the formula: For a moment The average rate of change of stress in the system at that time; This represents the total number of servo fixture units 121 in the workstation. For the first Force sensor 122 at time Output instantaneous force measurement value; For the first Force sensor 122 at the previous sampling time Output force measurement value; This represents the time interval between two adjacent samples.

[0109] The distributed edge controller 110 will calculate the global stress change rate. Compared with the preset steady-state determination threshold Compare. When the condition is met. At this point, the distributed edge controller 110 determines that the current wooden crate material has reached a mechanical steady state. At this moment, the distributed edge controller 110 triggers a synchronization latch command. In the instant the synchronization latch command is triggered, the system freezes the current state data of all servo clamping units 121 in parallel. The distributed edge controller 110 then reads the steady-state force value of each force sensor 122. and the steady-state coordinate values ​​of each position encoder 123 .

[0110] The distributed edge controller 110 constructs a stress field reference state vector using the acquired steady-state data. Stress field reference state vector The construction process satisfies the following relationship:

[0111] ;

[0112] In the formula: This is the final generated stress field reference state vector; This represents the merging operation of vector sets; For the first The state feature vector of each servo gripper unit 121 is frozen; For the first The normal clamping force scalar recorded by each servo clamping unit 121 at a steady-state moment; For the first The feed axis position coordinate scalar recorded by each servo clamp unit 121 at a steady state moment.

[0113] The distributed edge controller 110 of downstream station 202 is configured to execute a stress field reconstruction algorithm. The purpose of this algorithm is to restore the boundary constraint state of the wooden box material recorded by upstream station 201 within the physical space of downstream station 202.

[0114] The distributed edge controller 110 at downstream workstation 202 first parses the received digital accompanying token. Extract the stress field reference state vector from it. The distributed edge controller 110 will use the stress field reference state vector steady-state value of force and position steady state value Set as the control target value for the downstream servo fixture array 120.

[0115] The stress field reconstruction process is a multivariable closed-loop feedback control process. The distributed edge controller 110 simultaneously adjusts the downstream station 202... The feed position of each servo gripper unit 121 is determined to minimize the deviation between the current physical state and the reference state. The distributed edge controller 110 constructs a global objective function. To quantify this bias. Global objective function. The following relationship must be satisfied:

[0116] ;

[0117] In the formula: Let be the objective function value of the global state deviation to be minimized; This is a control vector that contains the current position coordinates of all servo gripper units 121; This represents the total number of servo fixture units 121 in the workstation. The weighting coefficient for the force deviation term; For the first A servo gripper unit 121 in the control vector The real-time force value fed back by force sensor 122 under action; Accompanying tokens for digital The first recorded Reference force values ​​for each unit; This represents the weighting coefficient for the position deviation term; For the first The real-time position coordinates fed back by the position encoder 123 of the servo clamp unit 121; Accompanying tokens for digital The first recorded The reference position coordinates of each unit.

[0118] The distributed edge controller 110 uses a gradient descent iterative algorithm to solve the problem. control vector that tends to the minimum Within each control cycle, the distributed edge controller 110 calculates the position update of the servo gripper unit 121 based on the gradient direction of the objective function. The position update control law satisfies the following relationship:

[0119] ;

[0120] In the formula: For the first The control cycle is sent to the first 124 position commands for each servo drive motor; For the first The first control cycle reads the first The actual position of each servo clamp unit 121; The iteration step size coefficient is used to control the algorithm and adjust the convergence speed of the system. For the objective function For the The partial derivatives of the position variables of each servo fixture characterize the error gradient.

[0121] The distributed edge controller 110 calculates the aforementioned position update in real time and drives the servo drive motor 124 to move. After each iteration, the distributed edge controller 110 again collects the current force and position values ​​and updates the objective function. .

[0122] The distributed edge controller 110 determines the termination condition for the completion of stress field reconstruction as the global objective function. Less than the preset convergence threshold The termination condition satisfies the following relationship:

[0123] ;

[0124] In the formula: The threshold for the maximum state reconstruction residual allowed by the system.

[0125] When the termination condition is met, the distributed edge controller 110 locks the position of all servo fixture units 121 and sends an enable signal to the micro-screwing impedance execution terminal 130 to allow processing. At this time, the stress state of the wooden box material in the downstream station 202 is considered to have been transferred undamaged from the upstream station 201 to the downstream station 202.

[0126] The implementation of feedforward adaptive machining control relies on physical space coordinates and digital accompanying tokens. Material dynamic impedance distribution spectrum stored in A precise mapping between them.

[0127] The distributed edge controller 110 is configured to read the position feedback signal of the main drive mechanism 133 in real time. The distributed edge controller 110 establishes a global physical coordinate system for the downstream workstation 202. In this global physical coordinate system, the tip position of the machining tool head 131 is defined as the instantaneous coordinate vector. The feed rate vector of the machining tool head 131 is defined as... .

[0128] To compensate for the response lag of the mechanical actuator, the distributed edge controller 110 performs feedforward prediction calculations. Based on the current motion state, the distributed edge controller 110 calculates the predicted coordinates of the processing point at future moments. Predict the coordinates of the processing point. The following relationship must be satisfied:

[0129] ;

[0130] In the formula: The calculated position vector of the aiming point in the physical coordinate system, including the x-coordinate. and ordinate ; This is the actual physical position vector of the machining tool head 131 at the current moment; The instantaneous feed velocity vector of the machining tool head 131 at the current moment; The feedforward aiming time window set for the system is determined by the servo response bandwidth of the main drive mechanism 133.

[0131] The distributed edge controller 110 will calculate the physical coordinates Mapped to material dynamic impedance distribution spectrum Discrete grid index space. Material dynamic impedance distribution spectrum. Stored in digital companion token In this context, its data structure is a two-dimensional matrix.

[0132] The mapping process is implemented through a coordinate transformation algorithm, converting continuous physical length quantities into discrete matrix row and column indices. Row index and column indexes The calculation satisfies the following relationship:

[0133] ;

[0134] ;

[0135] In the formula: To predict the dynamic impedance distribution spectrum of the material at the processing point The corresponding row index integer value; To predict the dynamic impedance distribution spectrum of the material at the processing point The corresponding column index integer value; This is the floor function operator; To predict the physical x-coordinate of the processing point; To predict the physical ordinate of the processing point; The x-coordinate of the origin of the wooden crate material in the workstation reference coordinate system; The ordinate of the origin of the wooden crate material in the workstation reference coordinate system; The physical resolution of the grid set along the length direction for upstream station 201 when generating the map; The physical resolution of the grid is set along the width direction for upstream station 201 when generating the map.

[0136] The distributed edge controller 110 utilizes the calculated index digital accompanying token Material dynamic impedance distribution spectrum The addressing operation is performed. The distributed edge controller 110 extracts the local dynamic stiffness value within the corresponding mesh cell and defines it as the feedforward stiffness parameter. .

[0137] ;

[0138] In the formula: To predict the material stiffness property value ahead on the processing path; For matrix In the line, number Values ​​stored in columns.

[0139] The distributed edge controller 110 will extract The parameters are input to the parameter adjustment module of the motion control algorithm. If... Indicating the presence of a high-hardness knot area ahead, the distributed edge controller 110 will determine the location before the machining tool head 131 actually reaches that area. Within a given timeframe, the output torque limit and feed speed setting of the main drive mechanism 133 are pre-adjusted. If Indicating the presence of a low-density decay area ahead, the distributed edge controller 110 will pre-decrease the cutting depth command to prevent material spalling.

[0140] See attached document Figure 4 The distributed edge controller 110 is configured based on feedforward stiffness parameters. The servo control parameters of the main drive mechanism 133 are dynamically adjusted. This adjustment process aims to achieve constant cutting force or constant trajectory precision control of the machining tool head 131 in non-uniform wood materials.

[0141] The distributed edge controller 110 integrates a variable impedance controller module. This module receives feedforward stiffness parameters calculated from the preceding logic level. The distributed edge controller 110 utilizes feedforward stiffness parameters. Calculate the target servo stiffness coefficient at the current moment. and target servo damping coefficient The parameter calculation process follows the following adaptive mapping relationship:

[0142] ;

[0143] ;

[0144] In the formula: The dynamic proportional gain coefficient applied to the position loop of the main drive mechanism 133 characterizes the equivalent stiffness of the actuator against load disturbances. This is a preset stiffness matching gain factor used to set the system's sensitivity to changes in material hardness. For digital companion tokens The predicted material stiffness value extracted from it; This is the reference stiffness constant of the system under no-load or air-cutting conditions; The dynamic differential gain coefficient applied to the speed loop of the main drive mechanism 133 characterizes the equivalent damping of the actuator. The preset system damping ratio is set to a value between 0.7 and 1.0 to ensure that the system response has no overshoot; The total equivalent moment of inertia of the main drive mechanism 133 is referred to the motor shaft end.

[0145] The distributed edge controller 110 acquires the real-time physical position vector of the processing tool head 131. and real-time velocity vector The distributed edge controller 110 reads the preset reference machining trajectory position vector. and reference velocity vector The distributed edge controller 110 calculates the position deviation vector. and velocity deviation vector .

[0146] Based on the calculated deviation vector and dynamic servo parameters, the distributed edge controller 110 generates the final torque control command. Torque control command The generation follows the following impedance control law:

[0147] ;

[0148] In the formula: This is the output torque command value sent to the servo drive motor in the main drive mechanism 133; It is the transpose of the kinematic Jacobian matrix of the main drive mechanism 133, used to map the forces in Cartesian space to the torques in joint space. The ideal position coordinate vector is given for the machining trajectory planning; The actual position coordinate vector fed back by the position encoder; The ideal feed rate vector is given for machining trajectory planning; The actual feed rate vector fed back by the speed sensor; This is the feedforward compensation torque for gravity and friction calculated based on the dynamic model.

[0149] When the machining tool head 131 enters the high-hardness nodule region, the feedforward stiffness parameter The value increases. The distributed edge controller 110 increases the target servo stiffness coefficient according to the above formula. This action causes the main drive mechanism 133 to output a larger corrective torque to maintain the straightness of the machining trajectory. When the machining tool head 131 enters the low-density decay region, the feedforward stiffness parameter... The value is reduced. The distributed edge controller 110 reduces the target servo stiffness coefficient. This action makes the machining tool head 131 flexible and compliant, preventing the brittle wood matrix from cracking due to excessive cutting force.

[0150] The distributed edge controller 110 is internally configured with a dynamics calculation module 510 and an impedance synthesis module 520. The dynamics calculation module 510 is configured to calculate the feedforward compensation torque based on the rigid body dynamics model. The impedance synthesis module 520 is configured to generate an impedance correction torque based on the position error and velocity error.

[0151] The dynamics calculation module 510 receives the joint position vectors of the main drive mechanism 133 in real time. and joint velocity vector The dynamics solution module 510 simultaneously reads the reference joint acceleration vector generated by trajectory planning. The dynamics solution module 510 calculates the dynamic feedforward torque used to counteract gravity, inertial forces, and friction. Dynamic feedforward torque The following relationship must be satisfied:

[0152] ;

[0153] In the formula: This is the calculated dynamic feedforward compensation torque vector; The main drive mechanism 133 is in the current joint position The system inertia matrix is ​​as follows; The joint acceleration vector of the reference trajectory; The coupling matrix between Coriolis force and centrifugal force; The joint velocity vector is the reference trajectory; This refers to the gravity load vector associated with the configuration of the main drive mechanism 133. This is the joint friction torque vector based on the Coulomb friction and viscous friction models.

[0154] The impedance synthesis module 520 receives the target servo stiffness coefficient obtained from the logic calculation of the previous stage variable impedance. and target servo damping coefficient The impedance synthesis module 520 incorporates the positional deviation in Cartesian space. and speed deviation Calculate the impedance feedback torque Impedance feedback torque The following relationship must be satisfied:

[0155] ;

[0156] In the formula: This is the impedance feedback torque vector mapped to the joint space; The kinematic Jacobian transpose matrix of the main drive mechanism 133 in the current attitude; To achieve the target servo stiffness coefficient This is the stiffness gain matrix with diagonal elements; Let be the reference position coordinate vector in Cartesian space; The actual feedback position coordinate vector in Cartesian space; To achieve the target servo damping coefficient This is the damping gain matrix with diagonal elements; Let be the reference velocity vector in Cartesian space; Let be the actual feedback velocity vector in Cartesian space.

[0157] The distributed edge controller 110 will generate dynamic feedforward torque. With impedance feedback torque Vector superposition and discretization filtering are performed to generate the final motor drive command. This drive command is used to control the electromagnetic torque output of the servo motor in the main drive mechanism 133. Motor drive command. The following relationship must be satisfied:

[0158] ;

[0159] In the formula: For the first The final torque command vector for each control cycle; For the first The dynamic feedforward torque calculated for each control cycle; For the first Impedance feedback torque calculated for each control cycle; For the first Historical torque command vector for each control cycle; These are the torque smoothing filter coefficients, with values ​​ranging from 0 to 1. Configured to suppress the effect of target servo stiffness coefficient High-frequency jitter in control signals caused by rapid mutations.

[0160] The system's workflow is strictly divided into an upstream data acquisition stage, an intermediate data transmission stage, and a downstream adaptive processing stage. All timing actions are globally synchronized by the real-time clock inside the distributed edge controller 110.

[0161] At any moment The wooden crate material is conveyed to upstream station 201 and triggers the positioning sensor. The distributed edge controller 110 of upstream station 201 immediately sends a closing command to the stress memory fixture array 120. The servo fixture unit 121 contacts the surface of the wooden crate and applies preload. From time... At that time This serves as the stress relaxation monitoring window. During this period, the distributed edge controller 110 continuously monitors the global stress change rate. Until it falls below the steady-state threshold .time This signifies the stabilization of the physical boundary conditions. At this moment, the system performs a stress field freezing operation and records the steady-state force values. and position value .

[0162] From time Initially, the micro-fission impedance execution terminal 130 initiates an impedance spectroscopy scan task. This task continues until time [time missing]. End. Total time taken for the scanning process. It depends on the distribution density of the probe points and the single-point excitation response time. Total scanning time The following relationship must be satisfied:

[0163] ;

[0164] In the formula: The total time required to complete the impedance characteristic detection of the entire wooden box surface; This represents the total number of planned detection points; For the first The duration of the micro-vibration excitation signal at each detection point; For the first Processing time for echo signal calculation and stiffness extraction at each detection point; For micro-fiber impedance execution terminal 130 from the first The point moves to the... Mechanical positioning time for each point.

[0165] At any moment Upstream station 201 completed the dynamic impedance distribution spectrum of the material. The generation of the stress field reference state vector by the distributed edge controller 110. With material dynamic impedance distribution spectrum Packaging and generating digital companion tokens From time At that time The wooden crates are physically moved to downstream station 202 via conveyor belt. Simultaneously, a digital token is attached. Data is transmitted via industrial Ethernet bus 200 to the distributed edge controller 110 at downstream workstation 202. To ensure that data arrives before materials, the system must meet the following timing constraints:

[0166] ;

[0167] In the formula: This refers to the physical transmission distance between upstream workstation 201 and downstream workstation 202. This refers to the average operating speed of the conveyor belt; Accompanying tokens for digital The total size of the binary data; The available bandwidth for the Industrial Ethernet Bus 200; This refers to the processing delay time of the network communication protocol stack.

[0168] At any moment The wooden crate material arrives at downstream station 202. The distributed edge controller 110 at downstream station 202 initiates a stress field reconstruction program. The servo fixture unit 121 performs closed-loop adjustment based on the resolved reference state. This reconstruction process continues until time [time missing]. At this point, the global objective function Converging to within the allowable error range ensures that the stress state of the wooden crate material is consistent with the upstream time. Maintain consistency.

[0169] From time Initially, the main drive mechanism 133 initiates adaptive machining actions. Throughout the entire machining process ( to The system operates in feedforward preview mode. For any machining moment... Distributed edge controller 110 reads the time window in advance The subsequent material stiffness properties. The timing synchronization of the feedforward control satisfies the following relationship:

[0170] ;

[0171] In the formula: This is the current machining control moment; This is the torque command output to the motor at the current moment; This is the variable impedance control algorithm function; Provides feedback on the actual location at the current moment; For future predictions based on current speed The processing position coordinates at that moment; This indicates the stiffness value indexed from the impedance spectrum.

[0172] At any moment The processing path has been completed. The main drive mechanism 133 returns to the safe position, and the servo clamp unit 121 releases the wooden box material.

[0173] To accommodate wooden crate materials where surface contact is not permitted or high-speed scanning is required, the contact-type piezoelectric ceramic stack of the micro-vibration impedance actuator 130 is replaced with an air-coupled ultrasonic transceiver module 610. The distributed edge controller 110 is configured to execute a material property calculation algorithm based on the acoustic impedance mismatch principle.

[0174] The air-coupled ultrasonic transceiver module 610 includes an ultrasonic transmitting unit 611 and an ultrasonic receiving unit 612. The ultrasonic transmitting unit 611 and the ultrasonic receiving unit 612 are mounted at an angle to the same side, and the sound beam focusing point is set on the upper surface of the wooden crate material. A distributed edge controller 110 controls a digital signal generator to generate a high-frequency pulse train excitation signal to drive the ultrasonic transmitting unit 611. The ultrasonic transmitting unit 611 emits a center frequency of [insert frequency here] onto the surface of the wooden crate material. The incident sound wave.

[0175] The distributed edge controller 110 first calculates the sound pressure reflection coefficient. Sound pressure reflection coefficient The calculation is based on the comparison between the received signal amplitude and the amplitude of the fully rigid reflection reference, satisfying the following relationship:

[0176] ;

[0177] In the formula: The sound pressure reflection coefficient of the wooden box material surface, with a value ranging from 0 to 1; The peak value of the reflected echo voltage output by the ultrasonic receiving unit 612 for the current detection point of the wooden box material; The reference reflected echo voltage peak value obtained when the system is tested on the surface of a standard metal calibration block; This is the geometric path attenuation compensation coefficient, used to correct for path differences caused by minor fluctuations in surface topography.

[0178] The distributed edge controller 110 is based on the calculated sound pressure reflection coefficient. The acoustic impedance of the wooden box material is inverted using the principle of continuity of acoustic boundary conditions. Due to the acoustic impedance of air The acoustic impedance is a known constant value and is much smaller than the wood impedance. The following relationship must be satisfied:

[0179] ;

[0180] In the formula: This represents the acoustic impedance value of the wood material at the current detection point; The acoustic impedance constant of air at standard atmospheric pressure; This is the sound pressure reflection coefficient calculated in the previous step.

[0181] To achieve compatibility with the mechanical stiffness control logic in the aforementioned embodiments, the distributed edge controller 110 uses acoustic impedance. Convert to equivalent dynamic elastic modulus This conversion process assumes that wood is an isotropic medium at the microscale, and that its density distribution is uniform within a local area. Equivalent dynamic elastic modulus. The following relationship must be satisfied:

[0182] ;

[0183] In the formula: The equivalent dynamic elastic modulus obtained by calculation is used to characterize the material's resistance to deformation; The calculated acoustic impedance of the wood; The average bulk density of the wooden crate material is determined by a weighing sensor and profile scanner at upstream station 201 and stored in a digital accompanying token. middle.

[0184] The distributed edge controller 110 will calculate the equivalent dynamic elastic modulus. The stiffness coefficients are mapped to dimensionless values ​​and filled into the material dynamic impedance distribution map according to the same meshing logic as in Example 2. In the middle. When When the high-frequency reflection threshold is exceeded, the system determines that the area is a high-hardness nodule; when When the frequency absorption is below the preset low-frequency absorption threshold, the system determines that the area is internally decayed or hollow.

[0185] The fluid-driven flexible clamping module 620 is composed of It consists of an independently controlled fluid actuation unit 621. Each fluid actuation unit 621 includes a double-acting fluid cylinder 622, an integrated linear displacement sensor 623, and a cavity pressure sensor 624. The distributed edge controller 110 independently regulates the input pressure of each fluid actuation unit 621 through a multi-channel digital proportional pressure reducing valve 622.

[0186] During the stress field freezing phase, the distributed edge controller 110 no longer directly reads the motor current, but instead acquires the steady-state fluid pressure value fed back by the cavity pressure sensor 624. The piston rod extension amount fed back by the linear displacement sensor 623 The distributed edge controller 110 calculates the equivalent gas-liquid stiffness under the current clamping state by utilizing the principle of fluid compressibility or the elastic modulus of fluid. Equivalent gas-liquid stiffness The following relationship must be satisfied:

[0187] ;

[0188] In the formula: For the first The equivalent physical stiffness value of each fluid actuator 621 under the current operating conditions; This is the polytropic index of the fluid medium (1.4 for pneumatic systems and the fluid elastic modulus correction factor for hydraulic systems). For the first The absolute pressure value recorded by the fluid actuation unit 621 during steady-state clamping; This refers to the effective force-bearing area of ​​the piston in the fluid actuation unit 621. This is the initial dead zone volume when the actuator is in the zero-position stroke; For the first Real-time displacement measurement value of a fluid actuation unit 621.

[0189] The distributed edge controller 110 calculates the equivalent gas-liquid stiffness set. and the corresponding set of displacements Write digital accompanying token This replaces the force sensor data in Example 3.

[0190] During the stress field reconstruction phase at downstream station 202, the distributed edge controller 110 parses the digital companion token. The stiffness and location information within the fluid. The distributed edge controller 110 uses the fluid supply source pressure from the downstream station 202. The control opening command of the digital proportional pressure reducing valve 622 is obtained through reverse engineering. To ensure that the fluid actuation unit 621 at the downstream station 202 replicates the stiffness characteristics recorded upstream, the target pressure command... The calculation satisfies the following relationship:

[0191] ;

[0192] In the formula: For downstream workstation 202 Pressure setpoint of fluid actuator 621; For digital companion tokens The reference stiffness value extracted from it; For digital companion tokens The reference position value extracted from it.

[0193] The distributed edge controller 110 adjusts the opening of the digital proportional pressure reducing valve 622 through a PID control loop, so that the actual monitored pressure value is consistent with the actual pressure value. Maintaining consistency. This variant utilizes the physical flexibility of the fluid medium itself to achieve both passive adaptation to and active maintenance of the stress state of the wooden crate material, achieving an equivalent stress field transfer function without the need for expensive torque sensors.

Claims

1. A distributed data acquisition and control system for wood box mass production, characterized by, The application relates to a wood box material production line, which comprises a plurality of logic nodes and an industrial Ethernet bus, the plurality of logic nodes are divided into upstream workstations and downstream workstations in a physical space and are communicatively connected through the industrial Ethernet bus; the upstream workstations are configured to perform initial processing and physical state acquisition on wood box materials, and the downstream workstations are configured to receive wood box materials and perform subsequent processing according to data; each logic node comprises a distributed edge controller, a stress memory clamp array and a micro-vibration impedance execution terminal; the distributed edge controller is electrically connected with the stress memory clamp array and the micro-vibration impedance execution terminal respectively; the distributed edge controller is internally provided with a shift register logic module, which is configured to map the production line into a data pipeline and control synchronous transfer of data packets corresponding to wood box materials between distributed edge controllers of different logic nodes; the data packets contain physical characteristic data of the wood box materials in a current production cycle, and the physical characteristic data at least comprises a unique identity code of the wood box material, a stress field reference state vector acquired and frozen at the upstream workstation and a material dynamic impedance distribution map generated through scanning by the micro-vibration impedance execution terminal. The stress field reference state vector records boundary constraint states of all stress memory clamp arrays on the wood box material in the upstream workstation, and the boundary constraint states contain force vector readings and position coordinate vector readings output by the stress memory clamp array when reaching a steady-state clamping; The material dynamic impedance distribution map is two-dimensional scalar field data, which is configured to map local dynamic stiffness values of the wood box material at different plane coordinate points, and the local dynamic stiffness values are generated by the distributed edge controller by analyzing feedback signals of the micro-vibration impedance execution terminal. The stress memory clamp array comprises a plurality of servo clamp units arranged around a processing area; Each servo clamp unit comprises a servo drive motor, a precision ball screw, a force sensor, a clamping contact and a position encoder; An output shaft of the servo drive motor is connected with the precision ball screw, the clamping contact is installed on a nut seat of the precision ball screw and is configured to contact the wood box material; 2. The distributed data acquisition and control system for wood box mass production according to claim 1, wherein, The force sensor is arranged at a connecting interface between the clamping contact and the precision ball screw and is configured to measure normal contact pressure; The position encoder is configured to feed back a linear position of the clamping contact.

3. The distributed data acquisition and control system for wood box mass production according to claim 1, wherein, The micro-vibration impedance execution terminal comprises a main driving mechanism, a vibration isolation flange, a piezoelectric ceramic stack and a processing tool head which are axially connected in series along a gravity direction; The main driving mechanism is a linear servo actuator; The vibration isolation flange is located between the main driving mechanism and the piezoelectric ceramic stack, is made of high-damping viscoelastic composite material and is configured to attenuate propagation of high-frequency mechanical vibration to the main driving mechanism; The piezoelectric ceramic stack is configured to emit mechanical micro-vibration signals and receive force response signals from the wood box material, and the processing tool head is installed on a lower surface of the piezoelectric ceramic stack. ​ 4. The distributed data acquisition and control system for wood box mass production according to claim 1, wherein, ​ ​ ​ ​ 5. The distributed data acquisition and control system for wood box mass production according to claim 4, wherein, The piezoelectric ceramic stack adopts a micro-vibration detection module with a pre-stressed packaging structure, and the micro-vibration detection module comprises an upper connecting flange, a lower output flange, a plurality of piezoelectric ceramic rings, and a center pre-tightening bolt; A protective shell is further arranged between the upper connecting flange and the lower output flange, the plurality of piezoelectric ceramic rings are located between the upper connecting flange and the lower output flange, and the protective shell is sleeved on the periphery of the plurality of piezoelectric ceramic rings; The center pre-tightening bolt passes through the center of the plurality of piezoelectric ceramic rings, the center pre-tightening bolt is configured to apply an axial static pre-tightening force, and the axial static pre-tightening force is set to be greater than a dynamic inertial force of the micro-vibration detection module under a maximum working acceleration.

6. The distributed data acquisition and control system for wood box mass production according to claim 4, wherein, The distributed edge controller internally integrates a digital signal generator module and a digital phase-locked amplifier module, and externally connects a high-voltage power amplifier and a charge amplifier; The digital signal generator module is configured to generate a micro-vibration excitation voltage signal, and the high-voltage power amplifier drives the piezoelectric ceramic stack to generate mechanical vibration; The charge amplifier is configured to convert a charge signal output by the piezoelectric ceramic stack into a force response signal; The digital phase-locked amplifier module is configured to perform quadrature demodulation on the force response signal, extract in-phase and quadrature components, and calculate the amplitude and phase lag of the force response signal relative to the excitation signal, and the distributed edge controller calculates the local dynamic stiffness of the contact point using the amplitude and the phase lag.

7. The distributed data acquisition and control system for wood box mass production according to claim 3, wherein, The distributed edge controller of the upstream station is configured to perform a stress field freezing process; The distributed edge controller is configured to monitor the global stress change rate of the stress memory clamp array, determine that the wooden box material reaches a mechanical steady state when the global stress change rate is lower than a preset steady state judgment threshold, and trigger a synchronous latching instruction; The synchronous latching instruction is configured to control the distributed edge controller to read and record the steady state force value and the steady state coordinate value of all servo clamp units in parallel, and is configured to construct the stress field reference state vector.

8. The distributed data acquisition and control system for wood box mass production according to claim 3, wherein, The distributed edge controller of the downstream station is configured to perform a stress field reconstruction process; The distributed edge controller parses the stress field reference state vector in the received data packet, and sets the steady state force value and the steady state coordinate value recorded therein as the control target value of the servo clamp unit of the downstream station; The distributed edge controller minimizes the deviation between the current physical state and the stress field reference state vector by adjusting the feeding position of the servo clamp unit of the downstream station until the global target function is less than a preset convergence threshold.

9. The distributed data acquisition and control system for wood box mass production of claim 4, wherein, The distributed edge controller of the downstream station is configured to perform feedforward adaptive machining control; The distributed edge controller calculates the predicted machining point coordinates at a future time according to the current motion state of the main drive mechanism, and maps the predicted machining point coordinates to the material dynamic impedance distribution atlas in the data packet, and is configured to extract the feedforward stiffness parameters of the corresponding grid cells; The distributed edge controller is configured to dynamically adjust a target servo stiffness coefficient and a target servo damping coefficient of the main driving mechanism according to the feedforward stiffness parameter, and is configured to change the compliance characteristics of the processing tool head to the material quality change of the wooden box material.

10. The distributed data acquisition and control system for wood box mass production of claim 1, wherein, The micro-vibration impedance execution terminal comprises an air-coupled ultrasonic transmitting and receiving module; The air-coupled ultrasonic transmitting and receiving module comprises an ultrasonic transmitting unit and an ultrasonic receiving unit which are installed obliquely, and a sound beam focusing point is set on the surface of the wooden box material; The distributed edge controller is configured to calculate an acoustic pressure reflection coefficient according to the reflected sound wave signal captured by the ultrasonic receiving unit, and inversely calculate the acoustic impedance of the wooden box material based on the acoustic pressure reflection coefficient, and further convert the acoustic impedance into an equivalent dynamic elastic modulus to generate the material dynamic impedance distribution map.

Images