Intelligent spring control strategy and system

Abstract

This invention discloses an intelligent spring control strategy and system, relating to the field of active vibration reduction control technology. It aims to achieve vibration reduction control of intelligent springs under different excitation conditions, solving the control problems of nonlinear structural parameters and varying excitation under different operating conditions. The process includes: building a three-dimensional model of the intelligent spring; simulating the influence of vibration reduction performance parameters; obtaining the control objectives for control input and output variables; and using a transfer function as the model for control simulation. A fuzzy controller is designed based on the working principle of the intelligent spring and the application conditions, enabling real-time adjustment of the intelligent spring's working state by continuously adjusting the preload under varying excitation conditions. The control strategy of this invention achieves real-time response of the intelligent spring to changes in excitation amplitude, realizing rapid and stable vibration suppression control under varying excitation conditions, effectively ensuring that the intelligent spring is in the optimal state of vibration reduction performance under different excitation conditions.

Description

Technical Field

[0001] This invention relates to the field of active vibration reduction control technology, and in particular to a control strategy and system for intelligent spring vibration reduction. Background Technology

[0002] Currently, helicopters face complex unsteady aerodynamic environments during actual flight, and the rotor aerodynamic load exhibits significant time-varying characteristics. For intelligent spring-loaded variable-pitch linkages (such as those shown in Chinese invention patent application number 202410893898.3), achieving vibration reduction hinges on the ability to adjust the operating state in real time according to changes in rotor load. Traditional control methods, such as PID control, require retuning of PID parameters when system parameters change (e.g., material friction coefficient, initial preload). However, in actual applications, the friction coefficient between the baffle and the inner spindle, as well as the initial preload during each installation, vary, resulting in a degree of nonlinearity in the intelligent spring's structural parameters.

[0003] Therefore, considering the nonlinearity of the structural parameters of the intelligent spring and the different excitation changes under various working conditions, how to achieve stable control has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] To address the above problems, this invention proposes an intelligent spring control strategy and system to achieve vibration reduction control of intelligent springs under different excitation conditions, thus solving the control problems of nonlinear structural parameters of intelligent springs and different excitation variations.

[0005] The technical solution of the present invention includes the following steps:

[0006] S1. Import the intelligent spring model into Ansys software for dynamic simulation to obtain its frequency response function under different working states and different excitations. Based on the vibration reduction performance law, obtain the control target and set the intermediate state as the control target. In the intermediate state, the preload of the intelligent spring reaches 0.5 times the preload required for the fully engaged state.

[0007] S2. The frequency response function is transformed by Laplace to obtain the transfer function of the smart spring under different working states, and the equivalent transfer function model of the smart spring is constructed by interpolation.

[0008] S3. Determine the universe of discourse of the intelligent spring control input and control output variables based on the frequency response function and operating conditions, and construct the membership function of the intelligent spring input and output variables;

[0009] S4. Input the control target and input / output variables into the fuzzy logic controller for fuzzification processing, formulate fuzzy control rules suitable for the intelligent spring working condition, and defuzzify the fuzzy control quantity to generate the adjustment command of the piezoelectric actuator output force.

[0010] The design of fuzzy control rules follows these basic principles:

[0011] When the vibration intensity is large and the preload deviation is negative, a larger piezoelectric actuator output force should be output to reduce the preload and return it to the intermediate state.

[0012] When the vibration intensity is large and the preload deviation is positive, a smaller piezoelectric actuator output force or an opposite adjustment amount should be output to increase the preload and return it to the intermediate state.

[0013] When the preload deviation is near zero, the output is moderate or zero adjustment to keep the system near the optimal preload.

[0014] By using the above rules and combining them with the membership function settings in step S3, a fuzzy control output suitable for different excitation amplitude changes can be obtained, and then the fuzzy control quantity is defuzzified.

[0015] S5. Real-time monitoring of the preload force, excitation force, and output force waveform of the intelligent spring bolt; using RMS to obtain the excitation amplitude over a short period of time, determining the current working condition type, and obtaining the preload force required for the intelligent spring to achieve optimal vibration reduction performance under the current working condition;

[0016] S6. Based on the real-time monitored intelligent spring bolt preload and the excitation force, a fuzzy rule is selected, and the output force of the piezoelectric actuator is tracked in real time according to the magnitude of the preload required for the intelligent spring to achieve optimal vibration reduction performance, so as to ensure that the vibration reduction performance of the intelligent spring remains optimal under different working conditions.

[0017] In step S1, the frequency response function (FRF) of the intelligent spring model is constructed as follows: ;

[0018] In the formula, For displacement amplitude, For force amplitude, For frequency, For phase;

[0019] Equivalent state response data exported from ANSYS software, including frequency. Displacement amplitude Phase The FRF dataset is obtained by combining the formula. ;

[0020] Through simulation comparison, it was determined that when the preload of the intelligent spring reaches 0.5 times the preload required for the fully engaged state, the system is in an intermediate working state and its vibration reduction performance is better than that of the soft spring. Therefore, the preload corresponding to this intermediate state is taken as the control target, that is, the intelligent spring is controlled to track the target value in real time.

[0021] The formula for calculating the transfer function in step S2 is: ;

[0022] In the formula, For the Laplace operator, The parameters of the intelligent spring system are determined by the intelligent spring structure, stiffness, damping, etc. The gain parameter of the intelligent spring system affects whether the output amplifies or attenuates the input. The displacement amplitude is in complex field form. For the complex field form of the force amplitude, For transfer functions;

[0023] Establish a continuous mapping between preload and transfer function so that the transfer function of the control simulation object can change continuously with the change of preload. Use piecewise linear interpolation to construct an equivalent transfer function.

[0024] When the real-time preload is less than the preload required for the intermediate state:

[0025] ;

[0026] in, ;

[0027] When the real-time preload is greater than the preload required for the intermediate state but less than the preload required to achieve the fully engaged state:

[0028] ;

[0029] in, ;

[0030] The continuous control model used in control simulation is constructed;

[0031] In the above, and Based on real-time monitoring of sensor data The value is obtained by calculation. This refers to the amount of preload required to achieve a fully bonded state. , , These are the transfer functions for the smart spring's soft spring state, intermediate state, and fully engaged state, respectively.

[0032] In step S3, determining the universe of discourse for the intelligent spring control input and control output variables includes:

[0033] First, based on the intelligent spring control strategy, the preload of the intelligent spring is adjusted according to the changes in the amplitude of the external load excitation and the amplitude of the support displacement; then, the control input variables are determined to include the preload deviation. and vibration intensity The control output variable is the output force of the piezoelectric actuator. ;

[0034] The membership function is: ;

[0035] in, Indicates membership degree; This represents the normalized target control parameters; a, b, c, d, e, and f are all predefined constants.

[0036] The universe of discourse for the input and output variables of the intelligent spring control is based on the excitation condition of the helicopter rotor pitch linkage. The universe of discourse for the preload deviation is calculated from the preload and preload deviation required for the intelligent spring to be in the fully engaged state and the intermediate state under different excitation amplitudes.

[0037] Based on the ratio of the intelligent spring in the soft spring state, intermediate state to fully combined state under different working conditions in the dynamic simulation, the vibration intensity of the intermediate state under typical working conditions can be close to 1, which is convenient for regular design; when the system tends to be fully combined, the peak value increases significantly, and the vibration intensity can reach the range of 2 to 3, thereby effectively distinguishing the working conditions of small, medium and large vibration.

[0038] Step S4 is as follows:

[0039] S4.1. Use the control input and output variables determined in step S3 as the input and output of the fuzzy controller;

[0040] The control input variable preload deviation determined in step S3 and vibration intensity As input to the fuzzy controller, the preload deviation is denoted as... The preload is obtained from the difference between the preload required for the intermediate state and the real-time preload; the vibration intensity is denoted as... The amplitude can be determined by the ratio of the support displacement amplitude to the reference displacement amplitude; the output variable of the fuzzy controller is the piezoelectric actuator output force adjustment command. or output force increment It is used to adjust the preload of the intelligent spring bolt in real time;

[0041] S4.2 Establish the constraint relationship between the control target parameters and the output force command;

[0042] The following constraints are established: the output force command should meet the system's allowable range, and the output should be prevented from exceeding the limit through a saturation circuit; at the same time, based on the physical structure of the intelligent spring for adjusting the preload, a mapping relationship can be established between the piezoelectric actuator output force and the bolt preload change, so that the output force command can directly act on the preload control quantity.

[0043] S4.3, Construct the logic for determining working conditions and switching rules;

[0044] To simultaneously meet the dual objectives of "vibration reduction under large vibrations" and "rigidity maintenance under small vibrations," a vibration intensity-based design is established. Working condition discrimination threshold ;when The condition is determined to be a low-vibration condition, and the controller output force command is set to zero. This ensures the intelligent spring maintains its fully engaged state corresponding to the initial preload, guaranteeing the rigidity and maneuverability of the variable-pitch tie rod; when When the condition is determined to be vibration reduction control, the fuzzy control rule base is activated to adjust the preload, so that the intelligent spring works near the intermediate state where the vibration reduction performance is better.

[0045] S4.4 Establish a fuzzy rule base and output control variables;

[0046] When entering the vibration reduction control mode, fuzzy inference is performed to obtain the fuzzy control quantity based on the fuzzy set membership degree of the two input variables, preload deviation and vibration intensity.

[0047] S4.5 Defuzzify to generate output force adjustment instructions;

[0048] The fuzzy control quantity is defuzzified using the center averaging method or the centroid method to obtain the numerical piezoelectric actuator output force adjustment command, which is then transmitted to the piezoelectric actuator control module for execution, thereby completing the real-time adjustment of the bolt preload.

[0049] Step S5 is as follows:

[0050] S5.1 Real-time data acquisition and sliding window processing;

[0051] The waveforms of the excitation force on the smart spring, the support displacement response waveform, the bolt preload, and the output force of the piezoelectric actuator are collected in real time. The effective value (RMS) of the excitation signal in a short time is calculated using the sliding window root mean square method to obtain stable short-time amplitude estimation results.

[0052] S5.2 Estimate the excitation amplitude using RMS and classify the operating conditions;

[0053] When the excitation signal is approximately sinusoidal, the short-time amplitude is estimated based on the correspondence between RMS and amplitude, and then the excitation amplitude is divided into different working condition types such as small, medium and large, which are used for subsequent fuzzy rule base selection and target preload update.

[0054] S5.3 Determine the optimal preload target;

[0055] Based on the "variation law of vibration reduction performance with preload" and the "optimal preload in the intermediate state is about 0.5 times the preload required to achieve the fully engaged state" obtained in step S1, the preload target required for the intelligent spring to achieve the best vibration reduction performance under the current working condition is determined by the excitation amplitude of the current working condition, and the preload deviation is obtained by combining the real-time preload, which is used for real-time tracking control in step S6.

[0056] Step S6 is as follows:

[0057] S6.1 Rule base selection and parameter update;

[0058] Based on the current working condition type determined in step S5, select the corresponding fuzzy rule subset or adjust the rule weights; at the same time, update the target preload and preload deviation so that the controller can reflect the characteristic of "optimal preload changing with working condition" caused by changes in the amplitude of external excitation.

[0059] S6.2 Real-time closed-loop tracking control;

[0060] The updated preload deviation value and vibration intensity are input into the fuzzy controller to perform fuzzy inference, outputting the piezoelectric actuator's output force command. A filtering stage ensures smooth output and system stability. Upon receiving the output force command, the piezoelectric actuator dynamically adjusts the preload of the smart spring bolt to maintain the real-time preload. Tracking the target preload ensures that the smart spring maintains optimal vibration reduction performance under varying excitation conditions;

[0061] S6.3, rigidity is maintained during small vibrations;

[0062] When steps S5 and S4 determine that the condition is a small vibration condition, the controller forces an output. This ensures that the intelligent spring remains fully engaged, corresponding to the initial preload, to avoid structural loosening or decreased maneuverability due to frequent adjustments during small vibrations.

[0063] The intelligent spring control system in this case includes:

[0064] The data acquisition module is used to acquire real-time data of the current operation of the smart spring, including real-time preload, preload deviation, excitation force amplitude, piezoelectric actuator output force, and smart spring support displacement response.

[0065] The input / output signal processing module is used to embed a smart spring fuzzy controller, which includes a preset fuzzy logic controller. The module performs fuzzification processing on the collected smart spring data input to the fuzzy logic controller to obtain fuzzy control quantities. The module then defuzzifies the fuzzy control quantities to generate piezoelectric actuator output force adjustment commands, so that the smart spring can adjust the preload force in real time based on changes in external excitation, thereby switching the working state and adjusting the real-time vibration reduction performance.

[0066] The piezoelectric actuator control module is used to control the output force of the piezoelectric actuator according to the output instructions of the input / output signal processing module.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] I. This invention uses dynamic simulation to set the working parameters as boundary conditions to obtain the dynamic characteristics and frequency response function of the intelligent spring, and then obtains the transfer function model that can represent the dynamic characteristics of the intelligent spring for control, providing an accurate model for subsequent control simulation; in the simulation of different excitations and preloads, the influence law of the vibration reduction performance of the intelligent spring is obtained, and the control objective of the intelligent spring control strategy is determined.

[0069] Second, this invention determines the domain of discourse for the intelligent spring control input and output under varying operating conditions, formulates fuzzy rules applicable to the intelligent spring's operating conditions, and through fuzzy logic control and defuzzification processing, the intelligent spring piezoelectric controller can track changes in external excitation in a timely and accurate manner to control the output force, enabling the intelligent spring to maintain optimal vibration reduction performance in real time under varying excitation conditions.

[0070] Third, this invention determines the magnitude of the vibration intensity of the intelligent spring by real-time monitoring of the output waveform of the intelligent spring support displacement, providing a basis for adjusting the preload and ensuring that the intelligent spring maintains the stability of its vibration reduction performance.

[0071] In summary, the control strategy of this invention enables the intelligent spring to respond in real time to changes in excitation amplitude, achieving rapid and stable vibration suppression control under varying excitation conditions. It effectively ensures that the intelligent spring is in the optimal state of vibration reduction performance under different excitation conditions, providing a solution for nonlinear control of intelligent springs, saving energy, and maintaining optimal vibration reduction performance. Attached Figure Description

[0072] Figure 1 This is a diagram illustrating the intelligent spring control strategy method in this case;

[0073] Figure 2 This is a diagram of the intelligent spring control system in this case;

[0074] Figure 3(a) is a perspective view of the smart spring, and Figure 3(b) is an axial sectional view of the smart spring.

[0075] In the diagram: 1 is the upper hanging lug; 2 is the upper cap; 3 is the wedge-shaped retaining ring; 4 is the outer sleeve; 5 is the main spring; 6 is the lower hanging lug; 7 is the preload bolt; 8 is the wedge-shaped locking block; 9 is the piezoelectric actuator; 10 is the inner mandrel; 11 is the bolt for the piezoelectric actuator; 12 is the bolt; 13 is the nut. Detailed Implementation

[0076] To clearly illustrate the technical features of this patent, the following detailed description is provided through specific embodiments and in conjunction with the accompanying drawings.

[0077] Reference Figure 1 The first embodiment of the present invention provides an intelligent spring control strategy, including the following steps:

[0078] S1. Import the intelligent spring model into Ansys software for dynamic simulation to obtain its frequency response function under different working states and different excitations. Based on the vibration reduction performance law, obtain the control target and set the intermediate state as the control target. In the intermediate state, the preload of the intelligent spring reaches 0.5 times the preload required for the fully engaged state.

[0079] S2. The frequency response function is transformed by Laplace to obtain the transfer function of the smart spring under different working states, and the equivalent transfer function model of the smart spring is constructed by interpolation.

[0080] S3. Determine the universe of discourse of the intelligent spring control input and control output variables based on the frequency response function and operating conditions, and construct the membership function of the intelligent spring input and output variables;

[0081] S4. Input the control target and input / output variables into the fuzzy logic controller for fuzzification processing, formulate fuzzy control rules suitable for the intelligent spring working condition, and defuzzify the fuzzy control quantity to generate the adjustment command of the piezoelectric actuator output force.

[0082] S5. Real-time monitoring of the preload force, excitation force, and output force waveform of the intelligent spring bolt; using RMS to obtain the excitation amplitude over a short period of time, determining the current working condition type, and obtaining the preload force required for the intelligent spring to achieve optimal vibration reduction performance under the current working condition;

[0083] S6. Based on the real-time monitored intelligent spring bolt preload and the excitation force, a fuzzy rule is selected, and the output force of the piezoelectric actuator is tracked in real time according to the magnitude of the preload required for the intelligent spring to achieve optimal vibration reduction performance, so as to ensure that the vibration reduction performance of the intelligent spring remains optimal under different working conditions.

[0084] In step S1, during the dynamic simulation of the smart spring, the three-dimensional model of the smart spring is first simplified, and material definitions, contact settings, and mesh generation are performed. After contact tool checks, boundary conditions for excitation required under different working conditions are applied. For example, in the application condition of a helicopter rotor pitch linkage, a sinusoidal excitation force with a frequency of 21 Hz and an amplitude of 100 N is applied to the top of the smart spring. The contact relationships between the baffle and the inner core shaft are set to be binding and friction, respectively. The support displacement response of the smart spring in the fully engaged state and the soft spring state is obtained. Based on this, the frequency response function of the smart spring under different working conditions and different operating states can be obtained.

[0085] By applying different bolt preloads, the support displacement response of the smart spring under different preloads can be obtained, thus revealing the influence of preload on the vibration reduction performance of the smart spring.

[0086] In a simulation group with an excitation condition of 21 Hz frequency and 100 N, the friction coefficient was set to 0.1. The minimum preload required for the intelligent spring to reach full engagement was 1000 N. Simulation groups with applied preloads of 0 N, 250 N, 500 N, 750 N, and 1000 N were also set. The simulation results under different preloads showed that the intelligent spring support displacement was smallest when a preload of 500 N was applied compared to the other simulation groups. Further simulations under different friction coefficients and preloads yielded the same vibration reduction performance pattern. It was determined that the preload required for the intelligent spring to achieve an intermediate state with better vibration reduction performance than a soft spring is approximately 0.5 times the preload required to reach full engagement. In this state, the intelligent spring's vibration reduction performance is superior to that of a soft spring, and this is used as the control objective of the control strategy.

[0087] The peak-to-peak values ​​of the displacement response of the smart spring support under different friction coefficients and preloads are shown in the table.

[0088]

[0089] Ultimately, the target control parameter was initially determined to be the preload required for the intelligent spring to reach its intermediate state in real time.

[0090] In step S2, the frequency response function of the intelligent spring is subjected to a Laplace transform to obtain the equivalent transfer function model used for its control simulation.

[0091] In step S1, the frequency response function (FRF) of the intelligent spring model is constructed as follows: ;

[0092] in, It's frequency. It is the displacement amplitude. It is phase. It is the force amplitude. It is an FRF dataset.

[0093] When performing Laplace transforms on the frequency response functions of the smart spring under three typical states, the swept frequency data from the dynamic simulation is first processed. Assuming vibrations below 60 Hz are considered in a helicopter application scenario, the swept frequency interval is set to 60, ranging from 1 Hz to 60 Hz, to obtain the displacement amplitude values ​​of the smart spring support at different frequencies. The MATLAB System Identification Toolbox provides the idfrd object to encapsulate the complex frequency response data, and tfest is used to estimate the continuous-time transfer function model from the idfrd and other data, outputting the identified transfer function.

[0094] The frequency response function is transformed using Laplace transform to obtain the transfer function of the smart spring under different operating states. Interpolation yields an equivalent transfer function model, including:

[0095] By selecting a structural form with second-order zeros and second-order poles, a simple and easily controllable SISO equivalent model is obtained; the transfer function calculation formula is as follows: ;

[0096] In the formula, For the Laplace operator, The parameters of the intelligent spring system are determined by the intelligent spring structure, stiffness, damping, etc. The gain parameter of the intelligent spring system affects whether the output amplifies or attenuates the input. The displacement amplitude is in complex field form. For the complex field form of the force amplitude, This is a transfer function.

[0097] The transfer functions for the three operating states obtained from the frequency response function of the smart spring are as follows:

[0098] Soft spring state:

[0099] ;

[0100] Intermediate state:

[0101] ;

[0102] Fully bound state:

[0103] ;

[0104] The effectiveness of the identified transfer function model was verified. Since the transfer function parameters were extracted based on finite element simulation frequency sweep analysis data, the output frequency response characteristics obtained from the Simulink simulation of this model were compared with the finite element simulation frequency sweep response results. Under the excitation conditions of 21 Hz and 100 N, the error between the transfer function model results of the smart spring in three states and the finite element simulation results was less than 20%, which is within the allowable error range. Therefore, the smart spring transfer function identification model based on finite element simulation is effective.

[0105] Then, a continuous mapping between preload and transfer function is established, so that the transfer function of the control simulation object can change continuously with the change of preload. Piecewise linear interpolation is used to construct the equivalent transfer function.

[0106] When the real-time preload is less than the preload required for the intermediate state:

[0107] ;

[0108] in, ;

[0109] When the real-time preload is greater than the preload required for the intermediate state but less than the preload required to achieve the fully engaged state:

[0110] ;

[0111] in, ;

[0112] Ultimately, this forms the continuous control model used for control simulation.

[0113] It should be noted that the interpolated continuous control model is based on the law of intelligent spring vibration damping performance changing with preload obtained in dynamic simulation. The influence curve of preload change on the intelligent spring support displacement response shows a linear relationship. In the continuous control model, it is also indicated that when the preload approaches... At that time, the equivalent model approximates the intermediate state model. When the preload approaches the full engagement threshold At that time, the equivalent model is close to the fully integrated model. .

[0114] In step S3, the continuous control model is input into the fuzzy logic controller for fuzzification processing to obtain fuzzy control quantities. The fuzzy control quantities are then defuzzified to generate piezoelectric actuator output force adjustment commands, so that the intelligent spring piezoelectric actuator can adjust according to the adjustment commands.

[0115] Specifically, the control input and output variables of the continuous control model are input into a preset fuzzy logic controller for fuzzification processing to obtain fuzzy control quantities, including:

[0116] Determine the fuzzy set of control input and output variable parameters;

[0117] Calculate the membership degree of the control input / output variables to the fuzzy set; wherein the membership degree calculation formula is as follows:

[0118]

[0119] in, Indicates membership degree; This represents the normalized target control parameters; a, b, c, d, e, and f are all predefined constants.

[0120] Subsequently, based on the membership degree and combined with preset fuzzy rules, the fuzzy control quantity is obtained.

[0121] It is worth noting that determining the fuzzy set of input and output variables for intelligent spring control is the foundation of fuzzy logic control.

[0122] For example, regarding the preload deviation of the input variable in intelligent spring control, five fuzzy sets can be defined: "negative small," "negative large," "zero range," "positive small," and "positive large." Specifically, the specific range can be fine-tuned based on the simulation results of intelligent spring control. If the preload deviation range is -850 to 450N, then -850 to -320N can be classified as "negative large," -520 to 0N as "negative small," -90 to 90N as "zero range," 0 to 380N as "positive small," and 250 to 450N as "positive large." Setting sets that include overlapping regions is one of the characteristics of fuzzy logic.

[0123] For example, for the fuzzy set "Zhengda", the coordinates of the four vertices of the trapezoidal function can be set to (250,0), (380,1), (450,1), and (450,0). When the preload deviation is 300N, its membership degree to the "Zhengda" set can be calculated to be approximately 0.38 using the formula. This method is simple and intuitive, and suitable for real-time control systems.

[0124] Derivation of fuzzy control quantities based on membership degrees and fuzzy rules is a crucial step. During fuzzification, the fuzzy rule base of the control system plays a key role. The rule base consists of a series of fuzzy control rules based on operating conditions. For example, if the vibration intensity is "large" and the preload deviation is "negatively small," then the output force of the piezoelectric actuator is "very large"; if the vibration intensity is "medium" and the preload deviation is "zero range," then the output force of the piezoelectric actuator is "medium." These rules define how to adjust the output control quantity based on the input data. In actual operation, the control system processes the fuzzified input data according to these rule bases to obtain the fuzzy control quantity; this process provides the basis for the generation of subsequent output control commands.

[0125] Understandably, this fuzzy control method can effectively handle the nonlinear characteristics of intelligent springs, improving their control accuracy and vibration suppression stability. By rationally setting fuzzy sets, selecting appropriate membership functions, and formulating suitable fuzzy rules, intelligent springs can maintain optimal vibration reduction performance in real time under different excitation conditions, reducing the impact of frictional losses.

[0126] Defuzzification is a crucial step in the fuzzy control process. Its purpose is to transform fuzzy control quantities into concrete numerical control commands for execution by intelligent spring-piezoelectric controllers and piezoelectric actuators. A commonly used method for defuzzification is the center-average method, which calculates the output value by taking the weighted average of all control quantities.

[0127] For example, the output of the fuzzy control rule is also a fuzzy set. For instance, the output force "Medium (M)" corresponds to two universe values: 0.6 for 300N and 0.4 for 400N. Similarly, the output force "Large (L)" corresponds to two universe values: 0.3 for 400N and 0.7 for 500N. Each rule has its activation state, determined by the minimum or maximum membership degree of the input / output variables to the fuzzy set. Assuming the above output variable u is activated, the final control command obtained through weighted calculation is 405N. This value is the adjustment command for the output force of the intelligent spring piezoelectric actuator obtained after defuzzification. The numerical control command obtained through defuzzification is transmitted to the piezoelectric actuator, which dynamically adjusts the real-time preload of the intelligent spring according to the command, thereby achieving vibration suppression control under varying excitation.

[0128] In one specific embodiment, after receiving the defuzzified control command, the intelligent spring control system will perform a dynamic adjustment operation on the output force of the corresponding piezoelectric actuator. Specifically, the intelligent spring will adjust the output force of the piezoelectric actuator according to the preload deviation and the vibration intensity of the support displacement, ensuring that the intelligent spring operates under optimal vibration reduction performance under different working conditions. For example, when the external excitation amplitude increases, the piezoelectric actuator increases the real-time preload of the intelligent spring by adjusting the preload deviation.

[0129] It should be noted that the dynamic adjustment of the intelligent spring preload is a real-time process. The intelligent spring adjusts its output variables based on the input variables to maintain optimal vibration damping performance. For example, if the external excitation amplitude changes abruptly, causing insufficient real-time preload to maintain optimal vibration damping performance, the piezoelectric controller and piezoelectric actuator will quickly adjust according to a pre-set fuzzy control strategy, increasing or decreasing the output force of the piezoelectric actuator to prevent the intelligent spring from operating in an unstable state under abrupt changes. The system also provides feedback on the current operating sampling data after each adjustment and further adjusts the control values ​​of the input and output variables based on the feedback information.

[0130] In step S4, the "formulation of fuzzy control rules applicable to the working conditions of the intelligent spring" is preferably implemented in the manner of "working condition discrimination - rule selection - fuzzy reasoning - defuzzification output" to ensure that the intelligent spring can quickly and stably adjust the bolt preload under different excitation amplitude changes, and maintain full engagement under small vibration conditions to meet the maneuverability requirements.

[0131] Specifically, step S4 may include the following sub-steps:

[0132] S4.1 Determine the input and output of the fuzzy controller.

[0133] The control input variable preload deviation determined in step S3 and vibration intensity As input to the fuzzy controller, the preload deviation is denoted as... It can be obtained from the difference between the preload required in the intermediate state and the real-time preload; the vibration intensity is denoted as... The amplitude can be determined by the ratio of the support displacement amplitude to the reference displacement amplitude; the output variable of the fuzzy controller is the piezoelectric actuator output force adjustment command. (or output force increment) This is used for real-time adjustment of the preload of the smart spring bolt.

[0134] S4.2 Establish the constraint relationship between the control target parameters and the output force command.

[0135] When the output force of a piezoelectric actuator is used to adjust bolt preload, the following constraint can be established: the output force command should meet the system's allowable range (e.g., Constraints (Internal), and avoids output over-limit through saturation; at the same time, according to the physical structure of the intelligent spring for adjusting the preload, a mapping relationship can be established between the output force of the piezoelectric actuator and the change of bolt preload, so that the output force command can be directly applied to the preload control quantity.

[0136] S4.3 Build the logic for determining working conditions and switching rules.

[0137] To simultaneously meet the dual objectives of "vibration reduction under large vibrations" and "rigidity maintenance under small vibrations," a setting based on vibration intensity can be established. Working condition discrimination threshold .when The condition is determined to be a low-vibration condition, and the controller output force command is set to zero. This ensures the intelligent spring maintains its fully engaged state corresponding to the initial preload, guaranteeing the rigidity and maneuverability of the variable-pitch tie rod; when When the condition is determined to be vibration reduction control, the fuzzy control rule library is used to adjust the preload, so that the smart spring works near the intermediate state with better vibration reduction performance (for example, the preload required for the intermediate state is 0.5 times the preload required to reach the fully engaged state).

[0138] S4.4 Establish a fuzzy rule base and output control variables.

[0139] When entering the vibration reduction control mode, fuzzy inference is performed to obtain the fuzzy control quantity based on the fuzzy set membership degrees of the two input variables, "preload deviation" and "vibration intensity". The rule base design follows the following basic principles:

[0140] (1) When the vibration intensity is large and the preload deviation is negative (i.e., the real-time preload is too large and the system tends to be fully engaged), a larger piezoelectric actuator output force should be output to reduce the preload and return to the intermediate state.

[0141] (2) When the vibration intensity is large and the preload deviation is positive (i.e., the real-time preload is small and the system tends to be in a soft spring state), a smaller piezoelectric actuator output force or an opposite adjustment amount should be output to increase the preload and return to the intermediate state.

[0142] (3) When the preload deviation is near zero, output a medium or zero adjustment amount to keep the system near the optimal preload.

[0143] By using the above rules and combining them with the membership function settings in step S3, a fuzzy control output suitable for different excitation amplitude changes can be obtained.

[0144] S4.5 Defuzzify to generate output force adjustment instructions.

[0145] The fuzzy control quantity is defuzzified using the center averaging method or the centroid method to obtain the numerical piezoelectric actuator output force adjustment command, which is then transmitted to the piezoelectric actuator control module for execution, thereby completing the real-time adjustment of the bolt preload.

[0146] In step S5, the step of "using RMS to obtain the excitation amplitude over a short period of time, determining the current working condition type, and obtaining the preload required for the smart spring to achieve optimal vibration damping performance under the current working condition" specifically includes:

[0147] S5.1 Real-time data acquisition and sliding window processing.

[0148] The waveforms of the excitation force on the intelligent spring, the support displacement response waveform, the bolt preload, and the output force of the piezoelectric actuator are acquired in real time. The effective value (RMS) of the excitation signal over a short period of time is calculated using the sliding window root mean square method to obtain stable short-time amplitude estimation results. The length of the sliding window can be selected according to the excitation frequency and the system sampling frequency to ensure smooth and reliable amplitude estimation.

[0149] S5.2 Estimate the excitation amplitude using RMS and classify the operating conditions.

[0150] When the excitation signal is approximately sinusoidal, the short-time amplitude can be estimated based on the correspondence between RMS and amplitude (e.g., amplitude ≈ ). ×RMS), and then divide the excitation amplitude into different working condition types such as small, medium and large, for subsequent fuzzy rule base selection and target preload update.

[0151] For example, based on the displacement response waveform of the intelligent spring support, the vibration intensity is calculated using the sliding root mean square method to calculate the current support displacement amplitude; wherein, the formula for calculating the support displacement amplitude is as follows:

[0152] in, The sliding window length is set appropriately to obtain a stable amplitude. This provides the response data for displacement over time within this window length. The displacement amplitude is given within the specified window length. The support displacement amplitude is compared with the reference displacement amplitude to determine the current vibration intensity and operating condition type.

[0153] S5.3 Determine the optimal preload target. Based on the "variation law of vibration reduction performance with preload" obtained in step S1 and the law that "the optimal preload in the intermediate state is about 0.5 times the preload required to reach the fully engaged state", the preload target required for the intelligent spring to achieve the best vibration reduction performance under the current working condition can be determined from the excitation amplitude of the current working condition. The preload deviation is obtained by combining the real-time preload and used for real-time tracking control in step S6.

[0154] In step S6, the step of "selecting fuzzy rules based on the real-time monitored bolt preload and the applied excitation force, and tracking and outputting the piezoelectric actuator output force in real time according to the preload required for optimal vibration damping performance" includes: the preload required for the intelligent spring to reach a fully engaged state varies with the external excitation amplitude; the intelligent spring fuzzy control strategy tracks and corrects the control target parameters based on the external excitation amplitude; and the real-time tracking output force of the piezoelectric actuator is used to determine whether the real-time preload of the intelligent spring allows it to be in an intermediate state for vibration damping, ensuring that the intelligent spring maintains optimal vibration damping performance in real time under varying excitation conditions. Specifically:

[0155] S6.1 Rule base selection and parameter update.

[0156] Based on the current working condition type determined in step S5, select the corresponding fuzzy rule subset or adjust the rule weights; at the same time, update the target preload and preload deviation so that the controller can reflect the characteristic of "optimal preload changing with working condition" caused by changes in the amplitude of external excitation.

[0157] S6.2 Real-time closed-loop tracking control.

[0158] The updated preload deviation value and vibration intensity are input into the fuzzy controller to perform fuzzy inference, outputting the piezoelectric actuator's output force command. A filtering stage ensures smooth output and system stability. Upon receiving the output force command, the piezoelectric actuator dynamically adjusts the preload of the smart spring bolt to maintain the real-time preload. By tracking the target preload, the intelligent spring can maintain optimal vibration reduction performance under varying excitation conditions.

[0159] S6.3, rigidity is maintained during small vibrations.

[0160] When step S5 / step S4 determines that the condition is a small vibration condition, the controller forces an output. This ensures that the smart spring remains fully engaged, corresponding to the initial preload, to prevent structural loosening or decreased maneuverability due to frequent adjustments during small vibrations. This strategy aligns with the design principle of this invention: "maintaining full engagement when the external load amplitude is small to ensure rigid maneuverability."

[0161] Furthermore, the control strategy described in this invention is not only applicable to intelligent spring structures using piezoelectric actuators as actuators, but can also be extended to other actuators capable of adjusting preload or equivalent contact pressure. Moreover, the membership function type, number of fuzzy sets, rule base size, and working condition division method of the fuzzy controller can all be adjusted according to actual applications. As long as they do not deviate from the core idea of ​​this invention—"real-time adjustment of preload based on excitation amplitude changes and vibration intensity feedback, and maintaining complete engagement under small vibrations"—they should all fall within the protection scope of this invention.

[0162] There are many specific ways to implement this invention. The above description is only a preferred embodiment of this invention. It should be noted that for those skilled in the art, several improvements can be made without departing from the principle of this invention, and these improvements should also be considered within the scope of protection of this invention.

Claims

1. A smart spring control strategy, characterized in that, Includes the following steps: S1. Import the intelligent spring model into Ansys software for dynamic simulation to obtain its frequency response function under different working states and different excitations. Based on the vibration reduction performance law, obtain the control target and set the intermediate state as the control target. In the intermediate state, the preload of the intelligent spring reaches 0.5 times the preload required for the fully engaged state. S2. The frequency response function is transformed by Laplace to obtain the transfer function of the smart spring under different working states, and the equivalent transfer function model of the smart spring is constructed by interpolation. S3. Determine the universe of discourse of the intelligent spring control input and control output variables based on the frequency response function and operating conditions, and construct the membership function of the intelligent spring input and output variables; S4. Input the control target and input / output variables into the fuzzy logic controller for fuzzification processing, formulate fuzzy control rules suitable for the intelligent spring working condition, and defuzzify the fuzzy control quantity to generate the adjustment command of the piezoelectric actuator output force. The design of fuzzy control rules follows these basic principles: When the vibration intensity is large and the preload deviation is negative, a larger piezoelectric actuator output force should be output to reduce the preload and return it to the intermediate state. When the vibration intensity is large and the preload deviation is positive, a smaller piezoelectric actuator output force or an opposite adjustment amount should be output to increase the preload and return it to the intermediate state. When the preload deviation is near zero, the output is moderate or zero adjustment to keep the system near the optimal preload. By using the above rules and combining them with the membership function settings in step S3, a fuzzy control output suitable for different excitation amplitude changes can be obtained, and then the fuzzy control quantity is defuzzified. S5. Real-time monitoring of the preload force, excitation force, and output force waveform of the intelligent spring bolt; using RMS to obtain the excitation amplitude over a short period of time, determining the current working condition type, and obtaining the preload force required for the intelligent spring to achieve optimal vibration reduction performance under the current working condition; S6. Based on the real-time monitored intelligent spring bolt preload and the excitation force, a fuzzy rule is selected, and the output force of the piezoelectric actuator is tracked in real time according to the magnitude of the preload required for the intelligent spring to achieve optimal vibration reduction performance, so as to ensure that the vibration reduction performance of the intelligent spring remains optimal under different working conditions.

2. The intelligent spring control strategy according to claim 1, characterized in that, In step S1, the frequency response function (FRF) of the intelligent spring model is constructed as follows: ; In the formula, For displacement amplitude, For force amplitude, For frequency, For phase; Equivalent state response data exported from ANSYS software, including frequency. Displacement amplitude Phase The FRF dataset is obtained by combining the formula. ; Through simulation comparison, it was determined that when the preload of the intelligent spring reaches 0.5 times the preload required for the fully engaged state, the system is in an intermediate working state and its vibration reduction performance is better than that of the soft spring. Therefore, the preload corresponding to this intermediate state is taken as the control target, that is, the intelligent spring is controlled to track the target value in real time.

3. The intelligent spring control strategy according to claim 1, characterized in that, The formula for calculating the transfer function in step S2 is: ; In the formula, For the Laplace operator, The parameters of the intelligent spring system are determined by the intelligent spring structure, stiffness, damping, etc. The gain parameter of the intelligent spring system affects whether the output amplifies or attenuates the input. The displacement amplitude is in complex field form. For the complex field form of the force amplitude, For transfer functions; Establish a continuous mapping between preload and transfer function so that the transfer function of the control simulation object can change continuously with the change of preload. Use piecewise linear interpolation to construct an equivalent transfer function. When the real-time preload is less than the preload required for the intermediate state: ； in, ; When the real-time preload is greater than the preload required for the intermediate state but less than the preload required to achieve the fully engaged state: ； in, ; The continuous control model used in control simulation is constructed; In the above, and Based on real-time monitoring of sensor data The value is obtained by calculation. This refers to the amount of preload required to achieve a fully bonded state. , , These are the transfer functions for the smart spring's soft spring state, intermediate state, and fully engaged state, respectively.

4. The intelligent spring control strategy according to claim 1, characterized in that, In step S3, determining the universe of discourse for the intelligent spring control input and control output variables includes: First, based on the intelligent spring control strategy, the preload of the intelligent spring is adjusted according to the changes in the amplitude of the external load excitation and the amplitude of the support displacement; then, the control input variables are determined to include the preload deviation. and vibration intensity The control output variable is the output force of the piezoelectric actuator. ; The membership function is: ; in, Indicates membership degree; This represents the normalized target control parameters; a, b, c, d, e, and f are all predefined constants. The universe of discourse for the input and output variables of the intelligent spring control is based on the excitation condition of the helicopter rotor pitch linkage. The universe of discourse for the preload deviation is calculated from the preload and preload deviation required for the intelligent spring to be in the fully engaged state and the intermediate state under different excitation amplitudes. Based on the ratio of the intelligent spring in the soft spring state, intermediate state to fully combined state under different working conditions in the dynamic simulation, the vibration intensity of the intermediate state under typical working conditions can be close to 1, which is convenient for regular design; when the system tends to be fully combined, the peak value increases significantly, and the vibration intensity can reach the range of 2 to 3, thereby effectively distinguishing the working conditions of small, medium and large vibration.

5. The intelligent spring control strategy according to claim 1, characterized in that, Step S4 is as follows: S4.

1. Use the control input and output variables determined in step S3 as the input and output of the fuzzy controller; The control input variable preload deviation determined in step S3 and vibration intensity As input to the fuzzy controller, the preload deviation is denoted as... The preload is obtained from the difference between the preload required for the intermediate state and the real-time preload; the vibration intensity is denoted as... The amplitude can be determined by the ratio of the support displacement amplitude to the reference displacement amplitude; the output variable of the fuzzy controller is the piezoelectric actuator output force adjustment command. or output force increment It is used to adjust the preload of the intelligent spring bolt in real time; S4.2 Establish the constraint relationship between the control target parameters and the output force command; The following constraints are established: the output force command should meet the system's allowable range, and the output should be prevented from exceeding the limit through a saturation circuit; at the same time, based on the physical structure of the intelligent spring for adjusting the preload, a mapping relationship can be established between the piezoelectric actuator output force and the bolt preload change, so that the output force command can directly act on the preload control quantity. S4.3, Construct the logic for determining working conditions and switching rules; To simultaneously meet the dual objectives of "vibration reduction under large vibrations" and "rigidity maintenance under small vibrations," a vibration intensity-based design is established. Working condition discrimination threshold ;when The condition is determined to be a low-vibration condition, and the controller output force command is set to zero. This ensures the intelligent spring maintains its fully engaged state corresponding to the initial preload, guaranteeing the rigidity and maneuverability of the variable-pitch tie rod; when When the condition is determined to be vibration reduction control, the fuzzy control rule base is activated to adjust the preload, so that the intelligent spring works near the intermediate state where the vibration reduction performance is better. S4.4 Establish a fuzzy rule base and output control variables; When entering the vibration reduction control mode, fuzzy inference is performed to obtain the fuzzy control quantity based on the fuzzy set membership degree of the two input variables, preload deviation and vibration intensity. S4.5 Defuzzify to generate output force adjustment instructions; The fuzzy control quantity is defuzzified using the center averaging method or the centroid method to obtain the numerical piezoelectric actuator output force adjustment command, which is then transmitted to the piezoelectric actuator control module for execution, thereby completing the real-time adjustment of the bolt preload.

6. The intelligent spring control strategy according to claim 1, characterized in that, Step S5 is as follows: S5.1 Real-time data acquisition and sliding window processing; The waveforms of the excitation force on the smart spring, the support displacement response waveform, the bolt preload and the output force of the piezoelectric actuator are collected in real time, and the effective value RMS of the excitation signal in a short time is calculated by the sliding window root mean square method to obtain a stable short-time amplitude estimation result. S5.2 Estimate the excitation amplitude using RMS and classify the operating conditions; When the excitation signal is approximately sinusoidal, the short-time amplitude is estimated based on the correspondence between RMS and amplitude, and then the excitation amplitude is divided into different working condition types such as small, medium and large, which are used for subsequent fuzzy rule base selection and target preload update. S5.3 Determine the optimal preload target; Based on the "variation law of vibration reduction performance with preload" and the "optimal preload in the intermediate state is about 0.5 times the preload required to achieve the fully engaged state" obtained in step S1, the preload target required for the smart spring to achieve the best vibration reduction performance under the current working condition is determined by the excitation amplitude of the current working condition, and the preload deviation is obtained by combining the real-time preload, which is used for real-time tracking control in step S6.

7. The intelligent spring control strategy according to claim 1, characterized in that, Step S6 is as follows: S6.1 Rule base selection and parameter update; Based on the current working condition type determined in step S5, select the corresponding fuzzy rule subset or adjust the rule weights; at the same time, update the target preload and preload deviation so that the controller can reflect the characteristic of "optimal preload changing with working condition" caused by changes in the amplitude of external excitation. S6.2 Real-time closed-loop tracking control; The updated preload deviation value and vibration intensity are input into the fuzzy controller to perform fuzzy inference, outputting the piezoelectric actuator output force command, and the output smoothness and system stability are ensured by the filtering stage. Upon receiving the output force command, the piezoelectric actuator dynamically adjusts the preload of the smart spring bolt to achieve real-time preload. Tracking the target preload ensures that the smart spring maintains optimal vibration reduction performance under varying excitation conditions; S6.3, rigidity is maintained during small vibrations; When steps S5 and S4 determine that the condition is a small vibration condition, the controller forces an output. This ensures that the intelligent spring remains fully engaged, corresponding to the initial preload, to avoid structural loosening or decreased maneuverability due to frequent adjustments during small vibrations.

8. An intelligent spring control system for implementing the intelligent spring control strategy of claim 1, characterized in that, include: The data acquisition module is used to acquire real-time data of the current operation of the smart spring, including real-time preload, preload deviation, excitation force amplitude, piezoelectric actuator output force, and smart spring support displacement response. The input / output signal processing module is used to embed a smart spring fuzzy controller, which includes a preset fuzzy logic controller. The module performs fuzzification processing on the collected smart spring data input to the fuzzy logic controller to obtain fuzzy control quantities. The module then defuzzifies the fuzzy control quantities to generate piezoelectric actuator output force adjustment commands, so that the smart spring can adjust the preload force in real time based on changes in external excitation, thereby switching the working state and adjusting the real-time vibration reduction performance. The piezoelectric actuator control module is used to control the output force of the piezoelectric actuator according to the output instructions of the input / output signal processing module.

Images