Modularized multi-mechanism vibration isolation, energy consumption and vibration absorption integrated device and control method thereof

Abstract

The invention relates to the technical field of vibration isolation and reduction engineering, and discloses a modular multi-mechanism vibration isolation, energy consumption and vibration absorption integrated device and a control method thereof.The modular multi-mechanism vibration isolation, energy consumption and vibration absorption integrated device comprises a support, a plurality of corrugated plates are fixedly connected to the interior of the support, and magnetorheological dampers are fixedly connected to the bottoms of the inner walls of the corrugated plates; a magneto-rheological damper is fixedly connected to the top of the support, a dynamic vibration absorber is fixedly connected to the top of the magneto-rheological damper, a particle damper is fixedly connected to the top of the dynamic vibration absorber, a control unit is arranged on the outer side of the support, the magneto-rheological damper comprises a first shell, and a piston rod is fixedly connected to the inner wall of the first shell. By integrating multiple mechanisms such as corrugated plate vibration isolation, particle damping energy consumption, dynamic vibration absorber vibration absorption and magneto-rheological damper intelligent control, all-directional vibration suppression is achieved, under different frequencies and vibration working conditions, all modules work cooperatively, the vibration reduction effect is greatly improved, and the vibration isolation efficiency gt within the broadband of 20-2000 Hz is ensured; 85%.

Description

Technical Field

[0001] The invention relates to the technical field of vibration isolation and vibration reduction engineering, and in particular to a modular multi-mechanism vibration isolation, energy dissipation and vibration absorption integrated device and a control method thereof. Background Art

[0002] The modular multi-mechanism vibration isolation and energy absorption integrated device belongs to the field of vibration control of industrial machinery or building structures. Its design principle is to add a vibration isolation layer between the structure and the foundation, and install vibration isolation supports to achieve a soft connection with the ground. The device is different from conventional flat plates by setting a corrugated steel plate, which can provide greater horizontal stiffness. The internal center position adopts a modular design and can be customized according to customer needs to select particle dampers, dynamic vibration absorbers and modules with different functions to solve different vibration reduction scenarios and provide vibration reduction and noise reduction solutions for bridge buildings or industrial equipment.

[0003] In the prior art, the vibration isolation platform support mainly isolates vibration and dissipates energy through its own structural damping, and the means are relatively simple. Due to the complexity of the external excitation of each building or equipment, the existing vibration reduction platform cannot solve the complex variable frequency, wide-band vibration and impact composite working conditions in a single form. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a modular multi-mechanism vibration isolation and energy absorption integrated device and a control method thereof, which solves the problem that the existing vibration isolation platform cannot adapt to impact composite working conditions.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a modular multi-mechanism vibration isolation and energy absorption integrated device and a control method thereof, including a support, a plurality of corrugated plates are fixedly connected inside the support, a magnetorheological damper is fixedly connected to the bottom of the inner wall of the corrugated plate, a dynamic vibration absorber is fixedly connected to the top of the magnetorheological damper, a particle damper is fixedly connected to the top of the dynamic vibration absorber, and a control unit is arranged on the outside of the support.

[0006] Preferably, the magnetorheological damper includes a first shell, a piston rod is fixedly connected to the inner wall of the first shell, the top of the piston rod is fixedly connected to the dynamic vibration absorber, a sensor body is fixedly connected to the inner wall of the first shell, magnetorheological fluid is arranged inside the first shell, and a coil is arranged on the outside of the piston rod.

[0007] Preferably, the dynamic vibration absorber comprises a first outer cylinder, a guide rod is fixedly connected to the bottom of the inner wall of the first outer cylinder, and a spring is fixedly connected between the top and the bottom of the inner wall of the first outer cylinder.

[0008] Preferably, the particle damper comprises a second outer cylinder, and a plurality of particle bodies are arranged inside the second outer cylinder.

[0009] Preferably, the waveform parameters of the corrugated plate are amplitude H=30 mm, wavelength λ=300 mm, the material combination is 304 stainless steel substrate and high damping rubber interlayer, vertical stiffness Kv=5×10^6 N / m, and horizontal stiffness Kh=2 N / m.

[0010] Preferably, the particle body is made of tungsten alloy, and a collision energy dissipation chamber with a conical reflector is provided inside the second outer cylinder. The filling rate is controlled to be 50-95% by controlling the filling quantity, thereby adjusting the frequency for energy dissipation.

[0011] Preferably, flange-type connecting plates are provided between the dynamic vibration absorber, the magnetorheological damper and the particle damper.

[0012] Preferably, the control unit includes an embedded sensor, an intelligent control algorithm module and a power drive module. The embedded sensor is used to monitor the vibration state of the structure in real time and obtain acceleration, displacement and frequency data; the intelligent control algorithm module adopts fuzzy logic, model prediction or reinforcement learning algorithms, and calculates the optimal magnetic field strength instruction in combination with the preset control target. The power drive module converts the instruction into a precise current signal, inputs it into the coil of the magnetorheological damper, and regulates the damping force output of the magnetorheological fluid.

[0013] Preferably, when the vibration signal detected by the embedded sensor is judged to be an impact load, the intelligent control algorithm module adopts a model predictive control strategy, when the dominant frequency is less than 5 Hz, a fuzzy PID control strategy is adopted, and when it is a broadband vibration, a reinforcement learning control strategy is adopted.

[0014] Preferably, a control method for a modular multi-mechanism vibration isolation and energy dissipation vibration absorption integrated device is used for a modular multi-mechanism vibration isolation and energy dissipation vibration absorption integrated device, comprising the following steps:

[0015] S1. Vibration signal acquisition: Use embedded sensors to monitor the vibration state of the structure in real time, obtain acceleration, displacement and other data, and collect data every 10ms to provide a basis for subsequent control strategies;

[0016] S2. Frequency domain feature extraction: Perform fast Fourier transform on the collected acceleration signal to obtain frequency domain features for judging vibration conditions;

[0017] S3. Working condition identification and control strategy selection:

[0018] Impact load condition: When the acceleration signal is detected to have impact characteristics, the model predictive control strategy is selected;

[0019] Low-frequency vibration condition: If the dominant frequency is less than 5Hz, the fuzzy PID control strategy is adopted;

[0020] Wideband vibration conditions: In other cases, reinforcement learning control strategy is used;

[0021] S4. Damping force calculation: According to the selected control strategy, the target damping force is calculated by combining the frequency domain characteristics and displacement data. Fuzzy PID control calculates the PID parameter adjustment amount based on the frequency deviation and amplitude deviation through fuzzy rule reasoning, and then obtains the target damping force. Model predictive control builds a prediction model. Within the prediction time domain of 3 control cycles, the target damping force is calculated by combining the constraint condition that the current change rate is ≤10A / ms with the goal of minimizing the future displacement prediction value. Reinforcement learning control is based on the state space and action space, and the target damping force is calculated through Q value estimation and ε-greedy action selection;

[0022] S5, magnetic field control: according to the calculated target damping force, the current query table is searched to obtain the corresponding current value, and the command is converted into a precise current signal through the power drive module to change the magnetic field generated by the electromagnetic coil in the magnetorheological damping module to control the damping force output of the magnetorheological fluid;

[0023] S6. Effect evaluation: During the operation of the device, the vibration suppression effect is monitored in real time to evaluate whether the preset vibration reduction target is achieved;

[0024] S7, parameter self-learning: online optimization is performed every 1 second. Reinforcement learning control is updated through experience playback and the strategy is optimized according to the reward function. Other control strategies can also adjust internal parameters according to the effect evaluation results to adapt to changing working conditions and improve control performance.

[0025] The present invention provides a modular multi-mechanism vibration isolation and energy absorption integrated device and a control method thereof. It has the following beneficial effects:

[0026] 1. The present invention realizes all-round suppression of vibration by integrating multiple mechanisms such as corrugated plate vibration isolation, particle damping energy dissipation, dynamic vibration absorber vibration absorption and magnetorheological damper intelligent control. Under different frequencies and vibration conditions, each module works together to greatly improve the vibration reduction effect, ensuring that the vibration isolation efficiency is >85% in the wide frequency band of 20-2000Hz, effectively reducing the vibration response of the structure.

[0027] 2. The present invention uses a control unit based on real-time collected vibration data, and uses intelligent algorithms such as fuzzy logic, model prediction or reinforcement learning to accurately control the damping force of the magnetorheological damper. It can respond quickly according to different vibration conditions. For example, under impact loads, the model prediction control strategy can quickly adjust the damping force to reduce vibration impact; in low-frequency and wide-band vibrations, fuzzy PID control and reinforcement learning control can also achieve precise control, thereby improving the system's adaptability to complex vibration environments.

[0028] 3. The present invention adopts a modular design. The dynamic vibration absorber, magnetorheological damper and particle damper are connected by a flange-type connecting plate, which is convenient for rapid replacement of modules according to different application scenarios. Different module combinations can be selected for different scenarios such as subway superstructures, steel structure bridges, power equipment foundations, etc. to meet diverse vibration reduction needs, thereby improving the versatility and flexibility of the device.

[0029] 4. The present invention uses a particle damper to control the movement of tungsten alloy particles in the collision energy dissipation chamber and a filling rate of 50-95% to achieve efficient energy dissipation. The maximum energy dissipation of a single module can reach 500kJ / cycle. The mass-spring system of the dynamic vibration absorber can be tuned steplessly, and combined with the damping control of the magnetorheological fluid, the frequency adaptation range of the device can reach 0.5-200Hz (TMD adjustable frequency band 0.5-15Hz), which can effectively cope with vibrations of different frequencies.

[0030] 5. The present invention can monitor the vibration suppression effect in real time through the effect evaluation and parameter self-learning links in the control method, and optimize the control strategy online every 1 second based on the evaluation results. The reinforcement learning control is updated through experience playback and the strategy is continuously adjusted according to the reward function. Other control strategies can also adaptively adjust parameters to ensure that the device always maintains the best vibration reduction performance during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is the main figure of the present invention;

[0032] Figure 2 is a cross-sectional schematic diagram of a support of the present invention;

[0033] Figure 3 It is a flow chart of the present invention.

[0034] Among them, 1. support; 2. corrugated plate; 3. dynamic vibration absorber; 4. magnetorheological damper; 5. control unit; 6. magnetorheological fluid; 7. piston rod; 8. coil; 9. sensor body; 10. first outer tube; 11. spring; 12. guide rod; 13. second outer tube; 14. particle body; 15. particle damper; 16. first outer shell. DETAILED DESCRIPTION

[0035] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the specification of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Please see attached Figure 1 -Attached Figure 3The embodiment of the present invention provides a modular multi-mechanism vibration isolation and energy absorption integrated device and a control method thereof, including a support 1, a plurality of corrugated plates 2 are fixedly connected inside the support 1, a magnetorheological damper 4 is fixedly connected to the bottom of the inner wall of the corrugated plate 2, a dynamic vibration absorber 3 is fixedly connected to the top of the magnetorheological damper 4, a particle damper 15 is fixedly connected to the top of the dynamic vibration absorber 3, and a control unit 5 is arranged on the outside of the support 1.

[0037] Specifically, when the user uses the modular multi-mechanism vibration isolation, energy dissipation and vibration absorption integrated device, the overall structure is supported by the support 1. The corrugated plate 2 inside the support 1 uses its specific waveform parameters and material combination to have a vertical stiffness of Kv=5×10^6N / m and a horizontal stiffness of Kh=2N / m, thereby achieving primary vibration isolation and reducing the impact of external vibration on the device; then the magnetorheological damper 4 begins to play a role, and the sensor body 9 inside it monitors the vibration state in real time, and transmits the acquired acceleration, displacement, frequency and other data to the control unit 5 outside the support 1. After the embedded sensor in the control unit 5 receives the data, the intelligent control algorithm module calculates the optimal magnetic field strength instruction based on fuzzy logic, model prediction or reinforcement learning algorithms, combined with the preset control target, and the power drive module converts the instruction into a precise current signal and inputs it into the coil 8 of the magnetorheological damper 4, thereby changing the magnetic field strength to regulate the magnetic flux. The damping force output of the variable fluid 6 is used to suppress vibration; the dynamic vibration absorber 3 absorbs vibration of a specific frequency through the spring 11 and the guide rod 12 in the first outer cylinder 10, in cooperation with the magnetorheological damper 4; the tungsten alloy particle body 14 in the particle damper 15 moves in the collision energy dissipation chamber of the second outer cylinder 13, and the frequency is adjusted by controlling the filling rate of 50-95% to achieve efficient energy dissipation; during the whole process, the control unit 5 continuously adjusts the working state of each component according to the real-time monitoring data, and through the effect evaluation and parameter self-learning links, the control strategy is optimized online every 1 second to adapt to different vibration conditions, ensuring that the device always maintains the best vibration reduction performance, solving the problems of the existing vibration reduction technology means being single, unable to adapt to complex excitations, complicated design and poor stability, as well as insufficient vibration reduction effect and versatility, and providing efficient, flexible and intelligent vibration reduction and noise reduction solutions for bridge construction, industrial equipment, etc.

[0038] See attached Figure 1 and attached Figure 2 The magnetorheological damper 4 includes a first shell 16, a piston rod 7 is fixedly connected to the inner wall of the first shell 16, the top of the piston rod 7 is fixedly connected to the dynamic vibration absorber 3, a sensor body 9 is fixedly connected to the inner wall of the first shell 16, a magnetorheological fluid 6 is arranged inside the first shell 16, and a coil 8 is arranged outside the piston rod 7.

[0039] Specifically, one end of the piston rod 7 is fixedly connected to the dynamic vibration absorber 3, and during the vibration process, it moves with the movement of the dynamic vibration absorber 3, driving the magnetorheological fluid 6 to flow. The sensor body 9 fixed on the inner wall of the first shell 16 monitors the vibration state of the device in real time, collects data such as acceleration, displacement, frequency, etc., and transmits these data to the control unit 5. After the control unit 5 receives the data, the intelligent control algorithm module therein uses fuzzy logic, model prediction or reinforcement learning algorithms according to the preset control target and the collected data to calculate the optimal magnetic field strength instruction. Then the power drive module converts the instruction into a precise current signal and transmits it to the coil 8 on the outside of the piston rod 7. After the coil 8 is energized, it generates A magnetic field is generated, and the magnetic field acts on the magnetorheological fluid 6, causing it to change from a Newtonian fluid to a quasi-solid state within milliseconds. By adjusting the magnetic field strength, that is, adjusting the current in the coil 8, the damping force of the magnetorheological fluid 6 can be dynamically controlled, and then the vibration reduction performance of the entire device can be adjusted to adapt to different vibration conditions and achieve effective suppression of vibration. The significance of the design of the magnetorheological damper 4 is that it provides the entire device with intelligent and precise damping force control capabilities, so that the device can adjust the damping force in real time according to the actual vibration conditions in a complex and changeable vibration environment, thereby enhancing the vibration reduction effect and adaptability of the device and solving the problem that the existing vibration reduction technology cannot flexibly and accurately control the damping force when dealing with complex vibrations.

[0040] See attached Figure 1 and attached Figure 2 The dynamic vibration absorber 3 includes a first outer tube 10 , a guide rod 12 is fixedly connected to the bottom of the inner wall of the first outer tube 10 , and a spring 11 is fixedly connected between the top and the bottom of the inner wall of the first outer tube 10 .

[0041] Specifically, when the device is excited by external vibration, the dynamic vibration absorber 3 will vibrate with the overall structure, and the spring 11 will undergo elastic deformation during the vibration process, storing and releasing energy. When the external vibration frequency is close to the natural frequency of the dynamic vibration absorber 3, resonance will be induced. At this time, the expansion and contraction amplitude of the spring 11 increases, and through the interaction with the vibration source, the vibration energy is absorbed and converted into elastic potential energy of the spring 11, thereby reducing the vibration energy transmitted to the protected structure, thereby achieving the purpose of vibration absorption. During the whole process, the guide rod 12 always guides the movement direction of the dynamic vibration absorber 3 to ensure that it can stably perform vibration absorption work, and cooperates with other components such as the magnetorheological damper 4 and the particle damper 15 to jointly realize the effective suppression of different types of vibrations by the device. The significance of the design of the dynamic vibration absorber 3 is to provide a special vibration absorption function for vibrations of a specific frequency, and to combine with other vibration reduction mechanisms to broaden the vibration reduction frequency range of the device, improve the vibration reduction effect of the device in a complex vibration environment, solve the problem that the existing vibration reduction technology is not effective in dealing with vibrations of a specific frequency, and enhance the adaptability and stability of the entire device.

[0042] See attached Figure 1 and attached Figure 2 The particle damper 15 includes a second outer cylinder 13 , and a plurality of particle bodies 14 are arranged inside the second outer cylinder 13 .

[0043] Specifically, when the device is affected by vibration, the particle body 14 will move in the second outer cylinder 13, and these particle bodies 14 will frequently collide with each other and with the inner wall of the second outer cylinder 13, consuming vibration energy during the collision, converting the mechanical energy of the vibration into other forms of energy such as heat energy, thereby playing the role of energy consumption and vibration reduction. In addition, a collision energy consumption chamber with a conical reflector is provided inside the second outer cylinder 13, which makes the movement path of the particle body 14 more complicated, increases the number and intensity of collisions, and further improves the efficiency of energy dissipation. The user can also achieve 50-9 by controlling the filling amount of the particle body 14. The filling rate is controlled at 5% to adjust the frequency and adapt to different vibration conditions, so that the particle damper 15 can effectively play an energy-consuming role in different vibration environments, and work together with the magnetorheological damper 4, the dynamic vibration absorber 3, etc. to reduce the vibration of the device, providing a highly efficient energy-consuming method for the device. Vibration can be effectively suppressed through the collision energy consumption of particles, and the filling rate can be adjusted according to actual needs to optimize the energy consumption effect, solving the problems of the existing vibration reduction technology in terms of single energy consumption means and poor adaptability, enhancing the vibration reduction ability of the entire device in a complex vibration environment, and improving the reliability and stability of the device.

[0044] See attached Figure 1 and attached Figure 2 The waveform parameters of the corrugated plate 2 are amplitude H = 30 mm, wavelength λ = 300 mm, the material combination is 304 stainless steel substrate and high damping rubber interlayer, vertical stiffness Kv = 5×10^6 N / m, and horizontal stiffness Kh = 2 N / m.

[0045] Specifically, the corrugated plate 2 adopts a waveform design with an amplitude of H = 30mm and a wavelength of λ = 300mm, and is combined with a material combination of a 304 stainless steel substrate and a high-damping rubber interlayer, so that it has the characteristics of vertical stiffness Kv = 5×10^6N / m and horizontal stiffness Kh = 2N / m. Its specific waveform parameters can change the vibration transmission path and energy distribution, and can effectively buffer and disperse vibration energy in both vertical and horizontal directions. The 304 stainless steel substrate provides good structural strength to ensure that it is not easily deformed and damaged during long-term use. The high-damping rubber interlayer further consumes vibration energy by virtue of its own damping characteristics. The vertical stiffness ensures that the device can effectively reduce the vertical load when subjected to vertical loads. The vertical displacement caused by vibration is reduced to maintain the stability of the structure, and the horizontal stiffness resists vibration in the horizontal direction to improve the device's ability to resist lateral vibration. Such a stiffness design enables the device to take the lead in effective primary vibration isolation when facing various complex vibration conditions, reducing the workload of subsequent modules, and working in coordination with the magnetorheological damper 4, the dynamic vibration absorber 3, the particle damper 15, etc., to jointly improve the vibration reduction effect of the entire modular multi-mechanism vibration isolation and energy absorption integrated device, solve the problems of poor primary vibration isolation effect and inability to effectively cope with complex vibration directions in existing vibration reduction technologies, enhance the adaptability of the device to different vibration environments, and ensure the safety and stability of the protected structure in a vibration environment.

[0046] See attached Figure 1 and attached Figure 2 The particle body 14 is made of tungsten alloy, and a collision energy dissipation chamber with a conical reflector is arranged inside the second outer cylinder 13. The filling rate is controlled to 50-95% by controlling the filling quantity, thereby adjusting the frequency for energy dissipation.

[0047] Specifically, the particle body 14 is made of tungsten alloy, which has a large density and high hardness. It can better consume energy during collision and can suppress vibration more effectively than other materials. The conical reflector collision energy consumption chamber arranged inside the second outer cylinder 13 can change the movement trajectory of the particles, make the collisions between particles and between particles and reflectors more frequent, greatly enhance the energy dissipation effect, and achieve 50-95% filling rate control by controlling the filling quantity. The frequency response of the particle damper 15 can be flexibly adjusted. Under different vibration conditions, the filling rate is adjusted according to actual needs, so that the particle damper 15 can appropriately increase the filling rate to improve the energy consumption effect when it vibrates at a low frequency. When it vibrates at a high frequency, the filling rate is adjusted to make it better adapt to the vibration frequency, and work together with the magnetorheological damper 4, the dynamic vibration absorber 3, etc. to broaden the energy consumption frequency range of the device, improve the vibration reduction performance of the entire modular multi-mechanism vibration isolation, energy consumption and vibration absorption integrated device, solve the problem that the existing vibration reduction technology is difficult to adapt to complex and changeable vibration frequencies in terms of energy consumption, enhance the adaptability of the device to various vibration environments, ensure efficient energy consumption under different vibration conditions, and ensure stable operation of the protected structure.

[0048] See attached Figure 1 and attached Figure 2 Flange-type connecting plates are arranged between the dynamic vibration absorber 3 , the magnetorheological damper 4 and the particle damper 15 .

[0049] Specifically, flange connection plates are arranged between the dynamic vibration absorber 3, the magnetorheological damper 4 and the particle damper 15, which makes the connection between the components more stable and reliable. During the operation of the device, the stability of the dampers working together can be ensured to avoid loosening of components due to loose connection, which affects the vibration reduction effect. It is convenient for quick disassembly and installation. When the application scenario of the device changes or a damper fails, the user can quickly replace the corresponding module, such as replacing it with a more suitable combination of particle damper 15 and dynamic vibration absorber 3 in the subway cover building scenario, which improves the maintenance efficiency of the device, saves maintenance costs and time, and enhances the versatility of the device. The module configuration can be flexibly adjusted according to different working conditions, and different module combinations can be selected in different scenarios such as steel structure bridges and power equipment foundations to achieve more accurate and efficient vibration reduction, solve the problems of inconvenient module replacement and poor versatility of existing vibration reduction equipment, promote modular multi-mechanism vibration isolation and energy absorption integrated devices to better meet diverse vibration reduction needs, and enhance their application value in different fields.

[0050] See attached Figure 1 and attached Figure 2 The control unit 5 includes an embedded sensor, an intelligent control algorithm module and a power drive module. The embedded sensor is used to monitor the structural vibration state in real time and obtain acceleration, displacement and frequency data; the intelligent control algorithm module adopts fuzzy logic, model prediction or reinforcement learning algorithms, and calculates the optimal magnetic field strength instruction in combination with the preset control target. The power drive module converts the instruction into a precise current signal and inputs it into the coil 8 of the magnetorheological damper 4 to adjust the damping force output of the magnetorheological fluid 6.

[0051] Specifically, the embedded sensor monitors the vibration state of the structure in real time, obtains acceleration, displacement, and frequency data, and provides an accurate information basis for the entire control process, allowing the device to perceive external vibration changes in real time. The intelligent control algorithm module uses fuzzy logic, model prediction, or reinforcement learning algorithms, combined with preset control targets to calculate the optimal magnetic field strength instructions. It can select the most appropriate algorithm for analysis and decision-making according to different vibration conditions, such as impact loads, low-frequency vibrations, broadband vibrations, etc., to ensure that accurate judgments and responses can be made in various complex vibration environments, thereby improving the device's adaptability to different vibrations. The power drive module converts the instructions into In order to input the precise current signal into the coil 8 of the magnetorheological damper 4 and regulate the damping force output of the magnetorheological fluid 6, the magnetorheological damper 4 can quickly adjust the damping force according to the control command, thereby realizing real-time and precise control of the vibration, and working in coordination with other components in the device such as the corrugated plate 2, the dynamic vibration absorber 3, and the particle damper 15, the vibration reduction effect of the entire modular multi-mechanism vibration isolation and energy absorption integrated device is effectively improved, and the problem that the existing vibration reduction technology cannot intelligently and accurately respond to complex vibration conditions is solved, and the reliability and stability of the device in different application scenarios are enhanced, meeting the needs of bridge construction, industrial equipment and other fields for efficient vibration reduction.

[0052] See attached Figure 1 and attached Figure 2 ,When the vibration signal detected by the embedded sensor is judged as an impact load, the ,intelligent control algorithm module adopts the model predictive control ,strategy. When the dominant frequency is less than 5Hz, the fuzzy PID control ,strategy is adopted. When it is a wide-band vibration, the reinforcement learning ,control strategy is adopted.

[0053] Specifically, when the vibration signal detected by the embedded sensor is judged as an impact load, the model predictive control strategy is adopted. Because the impact load is instantaneous and high-intensity, the model predictive control strategy can plan the control action in advance based on the prediction of the future state of the system, and quickly adjust the damping force of the magnetorheological damper 4, effectively buffer the impact energy, reduce the damage to the structure caused by vibration, and ensure the safety of the device and related structures. When the dominant frequency is less than 5Hz, the fuzzy PID control strategy is adopted. The change of low-frequency vibration is relatively slow. The fuzzy PID control strategy can use fuzzy logic to adjust the traditional PID control parameters online, and adjust the damping force more accurately according to the frequency deviation and amplitude deviation, adapt to the characteristics of low-frequency vibration, and achieve stability. Vibration reduction control, when it is broadband vibration, reinforcement learning control strategy is adopted. Broadband vibration contains multiple frequency components, and the situation is complex and changeable. Reinforcement learning control strategy can self-optimize in complex vibration environment through continuous trial and error and learning. According to the vibration state and the preset reward function, the control strategy is dynamically adjusted to make the device better adapt to broadband vibration and improve the overall vibration reduction effect. These three strategies are selected according to different vibration conditions to give full play to their respective advantages, so that the modular multi-mechanism vibration isolation, energy dissipation and vibration absorption integrated device can achieve efficient vibration control under various complex vibration conditions, which solves the problem that the existing vibration reduction technology is difficult to cope with various vibration conditions, enhances the adaptability and reliability of the device, and improves its application value in construction, industry and other fields.

[0054] See attached Figure 3 A control method for a modular multi-mechanism vibration isolation and energy dissipation vibration absorption integrated device is provided for use in a modular multi-mechanism vibration isolation and energy dissipation vibration absorption integrated device, comprising the following steps:

[0055] S1. Vibration signal acquisition: Use embedded sensors to monitor the vibration state of the structure in real time, obtain acceleration, displacement and other data, and collect data every 10ms to provide a basis for subsequent control strategies;

[0056] S2. Frequency domain feature extraction: Perform fast Fourier transform on the collected acceleration signal to obtain frequency domain features for judging vibration conditions;

[0057] S3. Working condition identification and control strategy selection:

[0058] Impact load condition: When the acceleration signal is detected to have impact characteristics, the model predictive control strategy is selected;

[0059] Low-frequency vibration condition: If the dominant frequency is less than 5Hz, the fuzzy PID control strategy is adopted;

[0060] Wideband vibration conditions: In other cases, reinforcement learning control strategy is used;

[0061] S4. Damping force calculation: According to the selected control strategy, the target damping force is calculated by combining the frequency domain characteristics and displacement data. Fuzzy PID control calculates the PID parameter adjustment amount based on the frequency deviation and amplitude deviation through fuzzy rule reasoning, and then obtains the target damping force. Model predictive control builds a prediction model. Within the prediction time domain of 3 control cycles, the target damping force is calculated by combining the constraint condition that the current change rate is ≤10A / ms with the goal of minimizing the future displacement prediction value. Reinforcement learning control is based on the state space and action space, and the target damping force is calculated through Q value estimation and ε-greedy action selection;

[0062] S5, magnetic field control: according to the calculated target damping force, the current query table is searched to obtain the corresponding current value, and the command is converted into a precise current signal through the power drive module to change the magnetic field generated by the electromagnetic coil in the magnetorheological damping module to control the damping force output of the magnetorheological fluid;

[0063] S6. Effect evaluation: During the operation of the device, the vibration suppression effect is monitored in real time to evaluate whether the preset vibration reduction target is achieved;

[0064] S7, parameter self-learning: online optimization is performed every 1 second. Reinforcement learning control is updated through experience playback and the strategy is optimized according to the reward function. Other control strategies can also adjust internal parameters according to the effect evaluation results to adapt to changing working conditions and improve control performance.

[0065] Specifically, in the S1 vibration signal acquisition link, embedded sensors are precisely deployed at key positions of the modular multi-mechanism vibration isolation and energy absorption integrated device. These positions are usually sensitive points on the vibration transmission path or representative positions of the overall structural vibration state, such as the area near the magnetorheological damper 4 and the dynamic vibration absorber 3 connected to the support. The sensor uses a high-precision MEMS acceleration sensor and a laser displacement sensor to ensure the accuracy of the acquired data. The MEMS acceleration sensor can accurately measure tiny acceleration changes, and the laser displacement sensor can measure displacement with high precision. Data is collected every 10ms. This time interval is determined after a large number of experiments and theoretical analysis. While ensuring timely acquisition of vibration information, it will not cause excessive data processing burden due to excessively high acquisition frequency, nor will it miss key vibration changes due to too low a frequency.

[0066] In the S2 frequency domain feature extraction, the fast Fourier transform algorithm uses an optimized library function with high computational efficiency and guaranteed accuracy. It performs frequency domain analysis on the acceleration signal and can decompose the complex vibration signal in the time domain into different frequency components, thereby clearly presenting the frequency structure of the vibration. By analyzing the frequency domain characteristics, it can accurately determine whether the vibration is dominated by a single frequency or contains multiple frequency components, as well as the energy distribution of each frequency component, providing strong support for the subsequent accurate identification of vibration conditions;

[0067] In the S3 working condition identification and control strategy selection part, under the impact load condition, the impact characteristics are judged based on multiple parameters such as the amplitude change rate of the acceleration signal, the peak value, and the duration of the signal, and are determined using threshold comparison and signal pattern recognition algorithms. The model prediction control strategy establishes a dynamic model of the device, considers factors such as the structural characteristics, mass distribution, stiffness and damping parameters of the device, predicts the future vibration state in the prediction time domain, optimizes the control input, adjusts the damping force of the magnetorheological damper in advance, and effectively buffers the impact energy. In low-frequency vibration conditions, the fuzzy rules of the fuzzy PID control strategy are stored in the rule base. When the dominant When the frequency is less than 5Hz, the corresponding fuzzy rules are searched in the rule base according to the frequency deviation and amplitude deviation, and the adjustment amount of the PID parameters is inferred and calculated to achieve precise control of low-frequency vibration. Under wide-band vibration conditions, the state space of the reinforcement learning control strategy contains the current vibration state of the device (such as acceleration, displacement, speed, etc.), the current damping force of the magnetorheological damper, control input and other information; the action space is the possible damping force adjustment value of the magnetorheological damper. Through continuous trial and error learning, the effect of each action is evaluated according to the reward function, and the control strategy is optimized, so that the device can achieve good vibration reduction effect even in a complex wide-band vibration environment.

[0068] In the S4 damping force calculation link, when the fuzzy PID control calculates the target damping force based on the frequency deviation and amplitude deviation, in order to improve the calculation efficiency, parallel computing technology is used, and multi-core processors are used to simultaneously process the reasoning calculations of multiple fuzzy rules. When the model predictive control builds the prediction model, the Kalman filter algorithm is used to estimate the system state to improve the prediction accuracy of the model. When the reinforcement learning control calculates the target damping force, by setting appropriate hyperparameters such as learning rate and discount factor, the exploration and utilization capabilities of the algorithm are balanced to accelerate the convergence to the optimal strategy;

[0069] During the S5 magnetic field control process, the current query table is pre-made according to the characteristic curve of the magnetorheological damper and stored in the control system memory of the device. The power drive module adopts a high-performance power amplifier, which can convert the control command into a high-precision and high-stability current signal to ensure that the electromagnetic coil of the magnetorheological damper generates a stable magnetic field and accurately control the damping force output of the magnetorheological fluid;

[0070] When evaluating the S6 effect, additional vibration monitoring sensors, such as strain gauges and acceleration sensors, are installed at key locations of the device to monitor the vibration response of the device in real time. The preset vibration reduction targets are determined according to different application scenarios and requirements. For example, in building structure vibration reduction, the displacement and acceleration limits of the structure may be used as vibration reduction targets; in industrial equipment vibration reduction, the operating accuracy and noise level of the equipment may be used as vibration reduction targets. The vibration suppression effect monitored in real time is compared with the preset targets to determine whether the device has achieved the expected vibration reduction effect.

[0071] During the S7 parameter self-learning stage, the experience replay mechanism of the reinforcement learning control stores information such as the state, action, reward, and next state of each learning process in the experience replay buffer. When the buffer reaches a certain capacity, a batch of data is randomly selected for learning and updating to avoid correlation between data and improve learning stability and efficiency. When other control strategies adjust internal parameters based on the effect evaluation results, an adaptive adjustment algorithm is used to dynamically adjust the adjustment step and direction of the parameters according to the changing trend of the vibration working condition and the adjustment effect, so that the control strategy can adapt to the changing working conditions more quickly and continuously improve the control performance.

[0072] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A modular multi-mechanism vibration isolation and energy absorption integrated device, comprising a support (1), characterized in that: A plurality of corrugated plates (2) are fixedly connected inside the support (1); a magnetorheological damper (4) is fixedly connected to the bottom of the inner wall of the corrugated plate (2); a dynamic vibration absorber (3) is fixedly connected to the top of the magnetorheological damper (4); a particle damper (15) is fixedly connected to the top of the dynamic vibration absorber (3); and a control unit (5) is arranged outside the support (1).

2. The modular multi-mechanism vibration isolation and energy absorption integrated device according to claim 1 is characterized in that: The magnetorheological damper (4) comprises a first shell (16), a piston rod (7) is fixedly connected to the inner wall of the first shell (16), the top end of the piston rod (7) is fixedly connected to the dynamic vibration absorber (3), a sensor body (9) is fixedly connected to the inner wall of the first shell (16), a magnetorheological fluid (6) is arranged inside the first shell (16), and a coil (8) is arranged outside the piston rod (7).

3. The modular multi-mechanism vibration isolation and energy absorption integrated device according to claim 1 is characterized in that: The dynamic vibration absorber (3) comprises a first outer tube (10), a guide rod (12) is fixedly connected to the bottom of the inner wall of the first outer tube (10), and a spring (11) is fixedly connected between the top and bottom of the inner wall of the first outer tube (10).

4. The modular multi-mechanism vibration isolation and energy absorption integrated device according to claim 1 is characterized in that: The particle damper (15) comprises a second outer cylinder (13), wherein a plurality of particle bodies (14) are arranged inside the second outer cylinder (13).

5. The modular multi-mechanism vibration isolation and energy absorption integrated device according to claim 1 is characterized in that: The waveform parameters of the corrugated plate (2) are amplitude H=30 mm, wavelength λ=300 mm, the material combination is a 304 stainless steel base plate and a high damping rubber interlayer, the vertical stiffness Kv=5×10^6 N / m, and the horizontal stiffness Kh=1.2 N / m.

6. The modular multi-mechanism vibration isolation and energy absorption integrated device according to claim 1 is characterized in that: The particle body (14) is made of tungsten alloy. A collision energy dissipation chamber with a conical reflector is arranged inside the second outer cylinder (13). A filling rate of 50-95% is controlled by controlling the filling quantity, thereby adjusting the frequency for energy dissipation.

7. The modular multi-mechanism vibration isolation and energy absorption integrated device according to claim 1 is characterized in that: Flange-type connecting plates are arranged between the dynamic vibration absorber (3), the magnetorheological damper (4) and the particle damper (15).

8. The modular multi-mechanism vibration isolation and energy absorption integrated device according to claim 1 is characterized in that: The control unit (5) comprises an embedded sensor, an intelligent control algorithm module and a power drive module. The embedded sensor is used to monitor the vibration state of the structure in real time and obtain acceleration, displacement and frequency data. The intelligent control algorithm module uses algorithms such as fuzzy logic, model prediction or reinforcement learning to calculate the optimal magnetic field strength instruction in combination with a preset control target. The power drive module converts the instruction into a precise current signal and inputs it into the coil (8) of the magnetorheological damper (4) to adjust the damping force output of the magnetorheological fluid (6).

9. The modular multi-mechanism vibration isolation and energy absorption integrated device according to claim 8 is characterized in that: When the vibration signal detected by the embedded sensor is judged to be an impact load, the intelligent control algorithm module adopts a model predictive control strategy. When the dominant frequency is less than 5 Hz, a fuzzy PID control strategy is adopted. When it is a broadband vibration, a reinforcement learning control strategy is adopted.

10. A control method for a modular multi-mechanism vibration isolation and energy absorption integrated device, characterized in that: The modular multi-mechanism vibration isolation, energy dissipation and vibration absorption integrated device used in any one of claims 1 to 9 comprises the following steps: S1. Vibration signal acquisition: Use embedded sensors to monitor the vibration state of the structure in real time, obtain acceleration, displacement and other data, and collect data every 10ms to provide a basis for subsequent control strategies; S2. Frequency domain feature extraction: Perform fast Fourier transform on the collected acceleration signal to obtain frequency domain features for judging vibration conditions; S3. Working condition identification and control strategy selection: Impact load condition: When the acceleration signal is detected to have impact characteristics, the model predictive control strategy is selected; Low-frequency vibration condition: If the dominant frequency is less than 5Hz, the fuzzy PID control strategy is adopted; Wideband vibration conditions: In other cases, reinforcement learning control strategy is used; S4. Damping force calculation: According to the selected control strategy, the target damping force is calculated by combining the frequency domain characteristics and displacement data. Fuzzy PID control calculates the PID parameter adjustment amount based on the frequency deviation and amplitude deviation through fuzzy rule reasoning, and then obtains the target damping force. Model predictive control builds a prediction model. Within the prediction time domain of 3 control cycles, the target damping force is calculated by combining the constraint condition that the current change rate is ≤10A / ms with the goal of minimizing the future displacement prediction value. Reinforcement learning control is based on the state space and action space, and the target damping force is calculated through Q value estimation and ε-greedy action selection; S5, magnetic field control: according to the calculated target damping force, the current query table is searched to obtain the corresponding current value, and the command is converted into a precise current signal through the power drive module to change the magnetic field generated by the electromagnetic coil in the magnetorheological damping module to control the damping force output of the magnetorheological fluid; S6. Effect evaluation: During the operation of the device, the vibration suppression effect is monitored in real time to evaluate whether the preset vibration reduction target is achieved; S7, parameter self-learning: online optimization is performed every 1 second. Reinforcement learning control is updated through experience playback and the strategy is optimized according to the reward function. Other control strategies can also adjust internal parameters according to the effect evaluation results to adapt to changing working conditions and improve control performance.

Images