Cutting transmission system with composite damping device and control method

Abstract

The invention relates to the field of hard rock heading machines, and discloses a cutting transmission system with a composite damping device and a control method.The cutting transmission system with the composite damping device comprises a cutting cantilever and an intelligent vibration controller, and an output shaft is arranged in a cavity of the cutting cantilever; a first bearing and a second bearing are sequentially arranged between the outer wall of the output shaft and the inner wall of the cutting cantilever, the output shaft penetrates through the first bearing and the second bearing, a damping device is arranged on the outer wall of the output shaft, and a vibration acceleration sensor is fixedly connected to the outer wall of the cutting cantilever. And one side of the cutting cantilever is fixedly connected with an end cover. The vibration reduction system can dynamically adapt to violent changes of cutting loads through the vibration reduction device, the acting force, supporting rigidity and damping of the vibration reduction device are controlled in real time through a hybrid intelligent control algorithm, the output rotating speed and torque of a motor are driven in real time, and cooperative control over cutting power transmission, real-time vibration monitoring and dynamic vibration reduction is achieved.

Description

Technical Field

[0001] This invention relates to the field of hard rock tunneling machine technology, specifically to a cutting transmission system and control method with a composite vibration damping device. Background Technology

[0002] In underground engineering and mining, hard rock tunnel boring machines (TBMs) are core equipment for efficient rock breaking. Their cutting and transmission systems are responsible for power transmission and cutting operations, directly affecting tunneling efficiency, construction accuracy, and equipment lifespan. In actual tunneling, complex geological conditions, lithological differences, and well-developed rock joints cause the cutting head to continuously bear broadband, time-varying random loads. This can easily couple with the system's natural frequency, causing resonance and resulting in severe vibrations. This not only interferes with normal operations but also accelerates component fatigue wear.

[0003] Currently, vibration reduction in cutting transmission systems mainly employs two types of solutions: passive and active. Passive vibration reduction relies on the deformation of elastic elements such as rubber and springs to absorb energy and suppress vibration. It has a simple structure, low cost, and requires no additional energy, making it suitable for stable load conditions. Active vibration reduction uses hydraulic or electromagnetic actuators to monitor vibration in real time and apply reverse control forces to counteract the excitation. For example, in ship shafting control, piezoelectric actuators suppress vibration with lateral force, offering fast response and high precision.

[0004] However, passive vibration damping has fixed stiffness and damping parameters, which cannot adapt to the wide-frequency time-varying characteristics of cutting loads. The vibration damping effect drops sharply near the resonance frequency, making it difficult to withstand load impacts under complex geological conditions. Although active vibration damping has high precision, it has stringent requirements on actuator performance, system response speed, and energy supply. Under harsh working conditions, the complex structure of the tunneling machine leads to insufficient reliability, and it occupies a large space, making it difficult to integrate compactly with the transmission system. A single solution cannot simultaneously achieve wide-frequency vibration damping, load adaptability, and reliability. Its suppression effect on random load resonance is limited, and it cannot meet the high-precision and high-stability operation requirements of the system. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a cutting transmission system and control method with a composite vibration damping device, which solves the problem that the existing cutting transmission system has limited effect on suppressing random load resonance and cannot meet the high precision and high stability requirements of the system.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a cutting transmission system with a composite vibration damping device, comprising a cutting cantilever and an intelligent vibration controller. An output shaft is disposed within the cavity of the cutting cantilever. A first bearing and a second bearing are sequentially disposed between the outer wall of the output shaft and the inner wall of the cutting cantilever. The output shaft passes through the first and second bearings. A vibration damping device is disposed on the outer wall of the output shaft. A vibration acceleration sensor is fixedly connected to the outer wall of the cutting cantilever. An end cap is fixedly connected to one side of the cutting cantilever. The output shaft passes through the end cap. A cutting head is fixedly connected to one end of the output shaft. A cutting reducer is fixedly connected to the other side of the cutting cantilever. A cutting motor is fixedly installed at the input end of the cutting reducer. A coupling is disposed at the output end of the cutting reducer. One side of the coupling is installed at the other end of the output shaft. The intelligent vibration controller is used to control the working force, support stiffness, and damping of the vibration damping device in real time using a hybrid intelligent control algorithm based on the applied load, and to drive the output speed and torque of the motor in real time.

[0007] By adopting the above technical solution, the output shaft is mounted on the cutting cantilever via the first and second bearings. The vibration damping device enables the system to dynamically adapt to drastic changes in the cutting load. A vibration acceleration sensor collects vibration signals in real time, and the cutting head is driven by the cutting motor. An intelligent vibration controller, based on the applied load, uses a hybrid intelligent control algorithm to control the vibration damping device's driving force, support stiffness, and damping in real time, and also drives the motor's output speed and torque in real time. This achieves coordinated control of cutting power transmission, real-time vibration monitoring, and dynamic vibration damping, allowing the cutting transmission system to maintain stable operation under complex load conditions. This solves the problem that existing cutting transmission systems have limited vibration damping effectiveness in suppressing random load resonance, failing to meet the system's high-precision and high-stability operational requirements.

[0008] Preferably, the vibration damping device includes a third bearing, the inner wall of which is disposed on the outer wall of the output shaft, and a composite vibration damping structure is disposed between the outer wall of the third bearing and the inner wall of the cutting cantilever.

[0009] Preferably, the composite vibration damping structure includes a vibration damping ring shell, the outer wall of which is fixedly connected to the inner wall of the cut cantilever, and a magnetorheological elastomer is uniformly fixedly connected to the inner wall of the vibration damping ring shell. The end of the magnetorheological elastomer away from the vibration damping ring shell is fixedly connected to the inner ring of the vibration damping ring.

[0010] Preferably, V-shaped support springs are uniformly fixedly connected to both sides of the outer shell and inner ring of the vibration damping ring, and a rhombic displacement amplifier is fixedly connected to one side of the two V-shaped support springs facing each other, and a piezoelectric ceramic is fixedly connected inside the rhombic displacement amplifier.

[0011] Preferably, both the V-shaped support spring and the rhomboid displacement amplifier have arc-shaped slots at their bends, and cylindrical rubber strips are fixedly connected to the arc-shaped slots of both the V-shaped support spring and the rhomboid displacement amplifier.

[0012] Preferably, the outer wall of the output shaft is threaded with a round nut, which is located on one side of the second bearing.

[0013] A cutting transmission control method with a composite vibration damping device, applied to the aforementioned cutting transmission system with a composite vibration damping device, includes the following steps: Solving and Training: Establish a dynamic model of the cutting transmission system, derive the system dynamic differential equation, constrain the control parameters and input parameters and sample them within the constraint interval, solve the system dynamic response data corresponding to each set of sampled parameters, and use deep learning to train the LSTM time-series prediction surrogate dynamic model F and state prediction model G of the cutting transmission system based on the dynamic response data. State prediction: Using the LSTM time-series prediction surrogate dynamics model F and state prediction model G, an MPC system prediction model and state observer are constructed. Based on the measurement values ​​of the vibration acceleration sensor, the system state is predicted through the state observer. The predicted state and control input are combined to predict the change in system state within a preset time. Optimization setting: Set the objective function and constraints, and use the optimization algorithm to minimize the objective function to obtain the change in system control input at the next time step; Solution Iteration: Add the change in system control input to the system control input at the previous moment to obtain the system control input at the current moment and apply it to the system. Substitute the control input at the current moment into the LSTM time-series prediction surrogate dynamics model F to solve for the change in system state at the current moment. Use the actual measured vibration acceleration to predict the change in system state through the state observer as the input to the system prediction model at the next moment. Iterate cyclically to obtain the change in control input at all moments.

[0014] Preferably, in the solution training step, the sampling uses Latin hypercube sampling to sample the control parameters and input parameters, and the specific expressions of the LSTM time-series prediction surrogate dynamics model F and the state prediction model G are as follows: ; in, x 0 is the initial value for the dynamic response. u For control parameters, x ( t The output response of the time series is... x c To be able to measure part of the system's dynamic response, x qThis refers to the overall dynamic response of the system predicted based on the measured values.

[0015] Preferably, in the state prediction step, the prediction time domain for predicting the system state change is: N p The predicted state is ,in, x ( k+i|k ) indicates at time k For future moments k + i The state prediction is performed, and the predicted states are stacked into a vector form: ; And expressed as ,in, U ( k ) is a stacked vector of future control inputs.

[0016] Preferably, in the optimization step, the objective function is used to minimize the system's vibration displacement, velocity, and acceleration. The constraints include constraints on the control input and its increment, and constraints on the system state. The optimization algorithm employs a quadratic programming algorithm, and the objective function is expressed as follows: ,in, X ( k ) represents the stacked vector of predicted states, Δ U ( k ) is a stacked vector that controls the input increment. Q and R This is a weight matrix used to balance the importance of state variables and control input increments. The constraint conditions are expressed as follows: ; ; in, ε This indicates the allowable error range for the state variable.

[0017] This invention provides a cutting transmission system and control method with a composite vibration damping device. It has the following beneficial effects: 1. This invention mounts the output shaft on the cutting cantilever via a first bearing and a second bearing. Through the installation of a vibration damping device, the vibration damping system can dynamically adapt to drastic changes in the cutting load. A vibration acceleration sensor collects vibration state signals in real time, and a cutting motor drives the cutting head. An intelligent vibration controller, based on the applied load, uses a hybrid intelligent control algorithm to control the working power, support stiffness, and damping of the vibration damping device in real time, and also drives the output speed and torque of the motor in real time. This achieves coordinated control of cutting power transmission, real-time vibration monitoring, and dynamic vibration damping, allowing the cutting transmission system to maintain stable operation under complex load conditions. It solves the problem that existing cutting transmission systems have limited effectiveness in suppressing random load resonance and cannot meet the high-precision and high-stability operational requirements of the system.

[0018] 2. This invention collects system vibration signals in real time using a vibration acceleration sensor. Combined with an LSTM time-series predictive surrogate dynamics model and an MPC model predictive control algorithm, it can predict the vibration development trend in advance and dynamically optimize the action force, support stiffness, damping parameters of the vibration reduction device, and the speed and torque of the motor. This achieves closed-loop real-time control of vibration signal acquisition, state prediction, parameter optimization, and execution adjustment, ensuring that vibration reduction measures are synchronously matched with vibration excitation and improving the accuracy and timeliness of vibration suppression.

[0019] 3. This invention integrates the composite vibration damping device inside the cutting cantilever, working in conjunction with the output shaft, bearings, and round nut. This not only avoids occupying additional installation space but also enhances the stability of the shaft system through the combination of the round nut and double bearings, improving the operational reliability of the system in harsh tunneling environments with high dust and strong vibration. At the same time, the vibration suppression effect weakens the severe vibrations caused by random loads and system resonance, reducing the alternating stress on the transmission components. The synergy between the round nut and the vibration damping device also isolates direct friction between components, reducing mechanical wear and thus extending the service life of components such as the cutting motor and reducer, reducing the frequency of equipment downtime for maintenance, and lowering long-term operation and maintenance costs. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of a cutting transmission system with a composite vibration damping device proposed in this invention; Figure 2 for Figure 1 Enlarged schematic diagram of the structure at point A in the middle; Figure 3 for Figure 2 A three-dimensional structural diagram of the composite vibration damping ring at point B; Figure 4 This is a partial structural diagram of a rhombic displacement amplifier for a cutting transmission system with a composite vibration damping device proposed in this invention. Figure 5This is a partial structural diagram of a V-shaped support spring in a cutting transmission system with a composite vibration damping device proposed in this invention. Figure 6 This is a schematic diagram of the control flow of an intelligent vibration controller for a cutting transmission system with a composite vibration damping device proposed in this invention. Figure 7 This is a flowchart of a cutting transmission control method with a composite vibration damping device proposed in this invention.

[0021] The components include: 1. Cutting head; 2. First bearing; 3. Vibration acceleration sensor; 4. Output shaft; 5. Round nut; 6. Cutting reducer; 7. Cutting motor; 8. Cutting cantilever; 9. Coupling; 10. Second bearing; 11. End cover; 12. Third bearing; 13. Vibration damping ring housing; 14. Magnetorheological elastomer; 15. V-shaped support spring; 16. Piezoelectric ceramic; 17. Rhomboid displacement amplifier; and 18. Inner ring of the vibration damping ring. Detailed Implementation

[0022] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0023] Please see the appendix Figure 1 - Appendix Figure 6 This invention provides a cutting transmission system and control method with a composite vibration damping device, including a cutting cantilever 8 and an intelligent vibration controller. An output shaft 4 is disposed within the cavity of the cutting cantilever 8. A first bearing 2 and a second bearing 10 are sequentially disposed between the outer wall of the output shaft 4 and the inner wall of the cutting cantilever 8. The output shaft 4 passes through the first bearing 2 and the second bearing 10. A vibration damping device is disposed on the outer wall of the output shaft 4. A vibration acceleration sensor 3 is fixedly connected to the outer wall of the cutting cantilever 8. An end cover 11 is fixedly connected to one side of the cutting cantilever 8, and the output shaft 4 passes through the end cover 11. A cutting head 1 is fixedly connected to one end of the output shaft 4, and a cutting reducer 6 is fixedly connected to the other side of the cutting cantilever 8. A cutting motor 7 is fixedly installed at the input end of the cutting reducer 6, and a coupling 9 is disposed at the output end of the cutting reducer 6. One side of the coupling 9 is installed at the other end of the output shaft 4. The intelligent vibration controller is used to control the working force, support stiffness, and damping of the vibration damping device in real time according to the applied load using a hybrid intelligent control algorithm, and to drive the output speed and torque of the motor in real time.

[0024] Specifically, the output shaft 4 is mounted on the cutting cantilever 8 via the first bearing 2 and the second bearing 10, achieving stable support and flexible rotation of the output shaft 4; the vibration damping device enables the vibration damping system to dynamically adapt to drastic changes in the cutting load; the vibration acceleration sensor 3 collects the vibration status signal of the cutting transmission system in real time, providing accurate monitoring data support for the intelligent vibration controller; the end cover 11 forms a sealed cavity inside the cutting cantilever 8, preventing external dust and impurities from entering and affecting the operation of internal components; the cutting motor 7 operates, and the cutting reducer 6 achieves speed regulation and torque amplification, which is then transmitted through the cutting reducer 6. Coupling 9 smoothly transmits power to output shaft 4 and drives it to rotate, thereby driving the cutting head 1 to operate and realize the cutting and crushing operation of the rock mass. At the same time, the intelligent vibration controller uses a hybrid intelligent control algorithm to control the working power, support stiffness and damping of the vibration reduction device in real time according to the load, and drives the output speed and torque of the motor in real time. This realizes the coordinated control of cutting power transmission, real-time vibration monitoring and dynamic vibration reduction, so that the cutting transmission system can maintain a stable operating state under complex load conditions. This solves the problem that the existing cutting transmission system has limited vibration reduction and suppression effect on random load resonance, and cannot meet the high precision and high stability operation requirements of the system.

[0025] Please see the appendix Figure 1 - Appendix Figure 5 The vibration damping device includes a third bearing 12, the inner wall of which is disposed on the outer wall of the output shaft 4. A composite vibration damping structure is provided between the outer wall of the third bearing 12 and the inner wall of the cutting cantilever 8. The composite vibration damping structure includes a vibration damping ring housing 13, the outer wall of which is fixedly connected to the inner wall of the cutting cantilever 8. A magnetorheological elastomer 14 is uniformly fixedly connected to the inner wall of the vibration damping ring housing 13. An inner ring 18 of the vibration damping ring is fixedly connected to one end of the magnetorheological elastomer 14 away from the vibration damping ring housing 13. V-shaped support springs 15 are uniformly fixedly connected to both sides of the vibration damping ring housing 13 and the inner ring 18. A rhombic displacement amplifier 17 is fixedly connected to one side opposite to the two V-shaped support springs 15. A piezoelectric ceramic 16 is fixedly connected inside the rhombic displacement amplifier 17. Both the V-shaped support spring 15 and the rhomboid displacement amplifier 17 have arc-shaped slots at their bends, and cylindrical rubber strips are fixedly connected inside the arc-shaped slots of both the V-shaped support spring 15 and the rhomboid displacement amplifier 17.

[0026] Specifically, the vibration damping device is located inside the cutting cantilever 8, tightly surrounding the outer ring of the third bearing 12. Multiple magnetorheological elastomers 14 within the vibration damping device are uniformly distributed circumferentially and vulcanized or bonded to the inner wall of the vibration damping ring housing 13. The vibration damping ring housing 13 is then fixed to the inner wall of the cutting cantilever 8 via an interference fit or bolt connection. The other end of each magnetorheological elastomer 14 is also fixedly connected to the inner ring 18 of the vibration damping ring. The inner ring 18 of the vibration damping ring and the outer ring of the third bearing 12 are interference-fitted, thereby coupling the entire vibration damping structure to the output shaft 4. The magnetorheological elastomers 14 constitute... The damping adjustable unit of the vibration reduction structure allows the vibration energy to be transmitted to the inner ring 18 of the damping ring through the third bearing 12 when the output shaft 4 vibrates radially. This causes the magnetorheological elastomer 14 to undergo shear deformation, consuming the vibration energy. The intelligent vibration controller changes the magnetic field strength applied to the magnetorheological elastomer 14 by energizing the coil surrounding it (not shown in the figure but a well-known technology in the field). It can adjust the shear modulus in real time and continuously, thereby achieving online adjustment of the support damping. When encountering severe impact, the damping can be increased to increase energy dissipation and achieve the purpose of vibration reduction.

[0027] Multiple V-shaped support springs 15 are symmetrically fixed to the two end faces of the damping ring housing 13 and the inner ring 18 of the damping ring to provide basic support. The two V-shaped support springs 15 located on opposite sides are connected by a rhombic displacement amplifier 17. Each rhombic displacement amplifier 17 has a piezoelectric ceramic 16 pre-tightly installed inside. The piezoelectric ceramic 16 acts as the active actuator and will generate a small and precise displacement under the high voltage electric field applied by the intelligent vibration controller. This is the inverse piezoelectric effect. The rhombic displacement amplifier 17 uses the hinge lever principle to amplify the small axial displacement generated by the piezoelectric ceramic 16 into a large displacement output in the direction of the rhombic diagonal, thereby applying a precise preload to the V-shaped support spring 15. This preload will change the equivalent stiffness of the V-shaped support spring 15, thereby finely adjusting the overall stiffness of the entire damping structure.

[0028] By actively controlling the expansion and contraction of the piezoelectric ceramic 16, an active control force opposite to the vibration direction can be generated, directly counteracting the specific frequency vibration from the output shaft 4, especially the low frequency, thus achieving active compensation. Arc-shaped slots are formed at the bends of each V-shaped support spring 15 and the rhomboid displacement amplifier 17, and cylindrical rubber strips are vulcanized and encapsulated within these slots. The V-shaped support spring 15 itself provides the main elastic restoring force, while the cylindrical rubber strips added at its bends, on the one hand, generate friction and viscous damping during spring deformation, consuming energy; on the other hand, the high damping characteristics of the rubber material can effectively suppress resonance peaks and absorb high-frequency vibrations and transient impacts. Its damping characteristics, combined with the adjustable damping of the magnetorheological elastomer 14, give the system excellent damping characteristics over a wide frequency band.

[0029] The V-shaped support spring 15 and the magnetorheological elastomer 14 enable a wide range of continuous and reversible real-time adjustment of stiffness and damping. The piezoelectric ceramic 16 and the displacement amplifier 17 provide precise and rapid active force to counteract specific vibrations. The three work together to achieve a deep integration of active and semi-active forces, enabling the vibration reduction system to dynamically adapt to drastic changes in cutting load.

[0030] Please see the appendix Figure 1 The outer wall of the output shaft 4 is threaded with a round nut 5, which is located on one side of the second bearing 10.

[0031] Specifically, the round nut 5 is coaxially mounted on the outer wall of the output shaft 4 to press the second bearing 10 and provide axial support for the output shaft 4.

[0032] A cutting transmission control method with a composite vibration damping device, applied to the aforementioned cutting transmission system with a composite vibration damping device, includes the following steps: Solving and Training: Establish a dynamic model of the cutting transmission system, derive the system dynamic differential equation, constrain the control parameters and input parameters and sample them within the constraint interval, solve the system dynamic response data corresponding to each set of sampled parameters, and use deep learning to train the LSTM time-series predictive surrogate dynamic model F and state prediction model G of the cutting transmission system based on the dynamic response data. Furthermore, in the training step, Latin hypercube sampling is used for the control parameters and input parameters. The specific expressions for the LSTM time-series prediction surrogate dynamics model F and the state prediction model G are as follows: ; in, x 0 is the initial value for the dynamic response. u For control parameters, xt The output response of the time series. x c To be able to measure part of the system's dynamic response, x q This refers to the overall dynamic response of the system predicted based on the measured values.

[0033] Specifically, based on the shaft structure of the cutting transmission system, the mechanical characteristics of the vibration damping device, and the component connection relationship, parameters such as the output shaft rotational inertia, bearing support stiffness, and vibration damping device damping are integrated to establish a multi-degree-of-freedom dynamic model, and the system dynamic differential equation with vibration displacement, velocity, and acceleration as state variables is derived.

[0034] Subsequently, constraints were imposed on the control parameters and input parameters: the control parameters were selected as the support stiffness and damping coefficient of the vibration damping device, and the constraint intervals were set to 1×10. 6Up to 5×10⁸ N / m, 100 to 50000 Ns / m; the input parameter is selected as the cutting load, the constraint range is set to 5000 to 20000 N, and 200 sets of parameter samples are extracted using the Latin hypercube sampling method to ensure uniform coverage of the parameter space.

[0035] For each set of sampling parameters, the Newmark-β numerical method is used to solve the dynamic differential equations, obtaining the time-series dynamic response data of the system's vibration displacement, velocity, and acceleration. Then, an LSTM network with 3 hidden layers and 64 neurons per layer is constructed, using the sampling parameters as input and the dynamic response data as output. Samples are allocated to an 80% training set and a 20% validation set, and iterative training is performed using mean squared error as the loss function until convergence. This yields the LSTM time-series prediction surrogate dynamic model F and the state prediction model G, whose expressions are: ,in x 0 represents the dynamic response value of the system in its initial static state. u The control parameters are for the current moment; the acceleration signal acquired by the vibration acceleration sensor is the measurable part of the dynamic response. x c By inputting model G, the overall dynamic response of the system, including the torsional vibration and radial vibration of the output shaft, can be obtained. x q Model F then outputs the time-series vibration response x(t) under the corresponding control parameters.

[0036] State prediction: Using the LSTM time-series prediction surrogate dynamics model F and the state prediction model G, an MPC system prediction model and a state observer are constructed. Based on the measurement value of the vibration acceleration sensor 3, the system state is predicted through the state observer. The predicted state and control input are combined to predict the change in system state within a preset time. Furthermore, in the state prediction step, the prediction time domain for predicting system state changes is... N p The predicted state is ,in, xk+i|k Indicates at time k For future moments k + i The state prediction is performed, and the predicted states are stacked into a vector form: ; And expressed as ,in, Uk A stacked vector for future control inputs; Specifically, the state prediction step utilizes the trained LSTM time-series prediction surrogate dynamics model F and state prediction model G to construct the MPC system prediction model and state observer. The state observer employs a Kalman filter structure, using real-time measurements from the vibration acceleration sensor as input, and combines the output of model G to predict the system state, obtaining the system state variables at the current moment, including the radial vibration displacement velocity and torsional vibration angular velocity of the output shaft.

[0037] The prediction time domain for predicting system state changes is set as follows: N p A value of 10 represents the predicted system state over the next 10 control cycles. The predicted state is calculated according to the following relationship: ,in xk+i|k Indicates at time k For future moments k + i State prediction, initial value x ( k | k ) represents the current system state output by the state observer.

[0038] The predicted states are then stacked into a vector form according to the time series: This vector can be represented as ,in Uk A stacked vector of future control inputs, containing future... N p Control parameters such as the support stiffness and damping coefficient of the vibration reduction device within each control cycle are used to fully represent the changes in the system state over a period of time in vector form. This provides multi-step state prediction data for subsequent optimization steps, ensuring that the MPC system can optimize and adjust control parameters based on continuous state change trends.

[0039] Setting optimization: Define the objective function and constraints, and use an optimization algorithm to minimize the objective function to obtain the change in system control input at the next time step; Furthermore, in the optimization steps, the objective function is used to minimize the system's vibration displacement, velocity, and acceleration. Constraints include constraints on the control input and its increment, as well as constraints on the system state. The optimization algorithm employs a quadratic programming approach, and the objective function is expressed as follows: ,in, X ( k ) represents the stacked vector of predicted states, Δ U ( k ) is a stacked vector that controls the input increment. Q and R This is a weight matrix used to balance the importance of state variables and control input increments. The constraint conditions are expressed as follows: ; ; in, ε This indicates the allowable error range for the state variable.

[0040] Specifically, the optimization steps involve defining an objective function. This objective function aims to minimize the system's vibration displacement, velocity, and acceleration. Its expression is: ,in Q 3 N p The diagonal weight matrix of order 1, with diagonal elements corresponding to future... N p The weights of vibration displacement, velocity, and acceleration at each moment are set to 100, 50, and 20, respectively, to highlight the priority of vibration displacement suppression. R for N p The diagonal weight matrix has diagonal elements of 5, which is used to balance the adjustment range of the control input increment and avoid sudden changes in control parameters.

[0041] Next, constraints are set: In the control input constraints, the lower limit of u, umin, is the minimum allowable value of the support stiffness and damping of the vibration damping device, and the upper limit, umax, is its maximum allowable value; the lower limit of the control input increment, Δumin, is -5, and the upper limit, Δumax, is 5, limiting the single adjustment range of the control parameters. In the system state constraints, ε The value is 0.01, representing the allowable error range of the vibration state quantity. The constraint expression is: ; Then, a quadratic programming optimization algorithm is used to solve the objective function: the objective function is transformed into a quadratic optimization problem, so as to... X ( k ) and Δ U ( k To optimize the variables, we substitute the constraints to construct the feasible region and iteratively solve the problem using gradient descent until the objective function converges to its minimum value. The resulting Δ is then calculated. U ( k This refers to the change in system control input at the next moment, ensuring that the control adjustment meets both vibration suppression requirements and system physical operating limitations.

[0042] Solution Iteration: Add the change in system control input to the system control input at the previous moment to obtain the system control input at the current moment, and apply it to the system. Substitute the control input at the current moment into the LSTM time-series prediction surrogate dynamics model F to solve for the change in system state at the current moment. Use the actual measured vibration acceleration to predict the change in system state through the state observer as the input to the system prediction model at the next moment. Iterate cyclically to obtain the change in control input at all moments.

[0043] Specifically, the system control input at the initial control moment is set in the solution iteration steps. Its values ​​are the initial default parameters of the vibration damping device support stiffness and damping, as well as the initial output speed and torque of the cutting motor.

[0044] At the current control moment, the change in system control input obtained in the previous step is added to the system control input value at the previous moment to obtain the system control input at the current moment. This control input is then applied to the cutting transmission system: adjusting the support stiffness and damping parameters of the vibration damping device, while simultaneously regulating the output speed and torque of the cutting motor, achieving real-time intervention of the control input on the system's operating state. Next, the control input at the current moment is substituted into the LSTM time-series predictive surrogate dynamics model, combined with the system state change at the previous moment, to solve for the system's state change at the current moment. Simultaneously, a vibration acceleration sensor collects the system vibration acceleration signal at the current moment, inputs this signal to the state observer, estimates the system state change at the current moment, and uses this estimate as the initial state input for the system prediction model at the next moment.

[0045] After each control cycle is completed, the above process is repeated: calculate the new change in system control input, add it to the current control input to obtain the control input at the next moment, apply it to the system to solve for the corresponding state change, update the initial input of the state observer, and iterate until the truncation operation is completed, and finally obtain the change in system control input at all control moments.

[0046] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A cutting transmission system with a composite vibration damping device, comprising a cutting cantilever (8) and an intelligent vibration controller, characterized in that: An output shaft (4) is provided inside the cavity of the cutting cantilever (8). A first bearing (2) and a second bearing (10) are sequentially arranged between the outer wall of the output shaft (4) and the inner wall of the cutting cantilever (8). The output shaft (4) passes through the first bearing (2) and the second bearing (10). A vibration damping device is provided on the outer wall of the output shaft (4). A vibration acceleration sensor (3) is fixedly connected to the outer wall of the cutting cantilever (8). An end cap (11) is fixedly connected to one side of the cutting cantilever (8). The output shaft (4) passes through the end cap (11). One end of the output shaft (4) is fixedly connected to the cutting head (1), and the other side of the cutting cantilever (8) is fixedly connected to the cutting reducer (6). The input end of the cutting reducer (6) is fixedly installed with the cutting motor (7), and the output end of the cutting reducer (6) is provided with a coupling (9). One side of the coupling (9) is installed at the other end of the output shaft (4). The intelligent vibration controller is used to control the working power, support stiffness and damping of the vibration damping device in real time according to the load, and to drive the output speed and torque of the motor in real time.

2. A cutting transmission system with a composite vibration damping device according to claim 1, characterized in that: The vibration damping device includes a third bearing (12), the inner wall of which is disposed on the outer wall of the output shaft (4), and a composite vibration damping structure is disposed between the outer wall of the third bearing (12) and the inner wall of the cut cantilever (8).

3. A cutting transmission system with a composite vibration damping device according to claim 2, characterized in that: The composite vibration reduction structure includes a vibration reduction ring shell (13), the outer wall of which is fixedly connected to the inner wall of the cut cantilever (8), and a magnetorheological elastomer (14) is uniformly fixedly connected to the inner wall of the vibration reduction ring shell (13). The end of the magnetorheological elastomer (14) away from the vibration reduction ring shell (13) is fixedly connected to the inner ring (18) of the vibration reduction ring.

4. A cutting transmission system with a composite vibration damping device according to claim 3, characterized in that: V-shaped support springs (15) are uniformly fixedly connected to both sides of the outer shell (13) and inner ring (18) of the damping ring. A rhombic displacement amplifier (17) is fixedly connected to one side opposite to the two V-shaped support springs (15). A piezoelectric ceramic (16) is fixedly connected inside the rhombic displacement amplifier (17).

5. A cutting transmission system with a composite vibration damping device according to claim 4, characterized in that: Both the V-shaped support spring (15) and the rhombus displacement amplifier (17) have arc-shaped slots at their bends, and cylindrical rubber strips are fixedly connected inside the arc-shaped slots of both the V-shaped support spring (15) and the rhombus displacement amplifier (17).

6. A cutting transmission system with a composite vibration damping device according to claim 1, characterized in that: The outer wall of the output shaft (4) is threaded with a round nut (5), which is located on one side of the second bearing (10).

7. A cutting transmission control method with a composite vibration damping device, characterized in that: The cutting transmission system with a composite vibration damping device according to any one of claims 1-6 includes the following steps: Solving and Training: Establish a dynamic model of the cutting transmission system, derive the system dynamic differential equation, constrain the control parameters and input parameters and sample them within the constraint interval, solve the system dynamic response data corresponding to each set of sampled parameters, and use deep learning to train the LSTM time-series prediction surrogate dynamic model F and state prediction model G of the cutting transmission system based on the dynamic response data. State prediction: Using the LSTM time-series prediction agent dynamics model F and state prediction model G, an MPC system prediction model and state observer are constructed. The system state is predicted by the state observer based on the measurement value of the vibration acceleration sensor (3). The system state change within a preset time is predicted by combining the estimated state and control input. Optimization setting: Set the objective function and constraints, and use the optimization algorithm to minimize the objective function to obtain the change in system control input at the next time step; Solution Iteration: Add the change in system control input to the system control input at the previous moment to obtain the system control input at the current moment and apply it to the system. Substitute the control input at the current moment into the LSTM time-series prediction surrogate dynamics model F to solve for the change in system state at the current moment. Use the actual measured vibration acceleration to predict the change in system state through the state observer as the input to the system prediction model at the next moment. Iterate cyclically to obtain the change in control input at all moments.

8. A cutting transmission control method with a composite vibration damping device according to claim 7, characterized in that: In the solution training step, the sampling adopts Latin hypercube sampling method for the control parameters and input parameters. The specific expressions of the LSTM time-series prediction surrogate dynamics model F and the state prediction model G are as follows: ； in, x 0 is the initial value for the dynamic response. u For control parameters, x ( t The output response of the time series is... x c To be able to measure part of the system's dynamic response, x q This refers to the overall dynamic response of the system predicted based on the measured values.

9. A cutting transmission control method with a composite vibration damping device according to claim 7, characterized in that: In the state prediction step, the prediction time domain for predicting the system state change is: N p The predicted state is ,in, x ( k+i| k ) indicates at time k For future moments k + i The state prediction is performed, and the predicted states are stacked into a vector form: ； And expressed as ,in, U ( k ) is a stacked vector of future control inputs.

10. A cutting transmission control method with a composite vibration damping device according to claim 7, characterized in that: In the optimization step, the objective function is used to minimize the system's vibration displacement, velocity, and acceleration. The constraints include constraints on the control input and its increment, and constraints on the system state. The optimization algorithm employs a quadratic programming algorithm, and the objective function is expressed as follows: ,in, X ( k ) represents the stacked vector of predicted states, Δ U ( k ) is a stacked vector that controls the input increment. Q and R This is a weight matrix used to balance the importance of state variables and control input increments. The constraint conditions are expressed as follows: ； ； in, ε This indicates the allowable error range for the state variable.

Images