Mechanical resonance suppression method for servo control system

Abstract

The invention relates to the technical field of motor servo control, and discloses a servo control system mechanical resonance suppression method, which comprises the following steps: establishing a dynamic model of a servo system; creating a digital twinborn model based on the dynamic model, and dynamically updating the digital twinborn model through real-time data; performing frequency domain analysis based on the real-time data, predicting whether the system enters a resonance state, and generating a prediction result; and automatically adjusting the control gain of the servo system according to the prediction result. According to the method, the digital twinborn technology is combined with the servo system dynamic model, the digital twinborn model is continuously corrected through real-time sensor data, then the effects of real-time monitoring and dynamic updating are achieved, the state of the virtual model is always synchronous with a physical system, and therefore the real-time monitoring and dynamic updating effect is achieved through real-time data feedback. The digital twinning technology enables the system to perceive changes in real time and make adjustments, and the adaptability and accuracy of the system are greatly improved.

Description

Technical Field

[0001] The invention relates to the technical field of motor servo control, and in particular to a method for suppressing mechanical resonance of a servo control system. Background Art

[0002] Servo system is an automatic control system, usually used to accurately control dynamic variables such as angle, position, speed and acceleration. Servo system is widely used in robots, automated production lines, power tools, spacecraft control, aircraft systems, CNC machine tools and other occasions that require high-precision control. Its core feature is the ability to automatically adjust the input signal and adjust the output to keep the system running in the desired state.

[0003] However, existing servo system control schemes generally rely on traditional static models, which usually assume that the system operates in a stable working state and ignore the impact of environmental changes on system performance. As actual working conditions and environmental conditions continue to change, traditional static models face the challenge of being difficult to adapt effectively. Due to the dynamic changes of the system, static models cannot reflect the real-time status in time, resulting in a lag in prediction and control, which ultimately affects the accuracy and stability of the system.

[0004] In addition, traditional control methods usually rely on empirical rules or fixed gain strategies based on set parameters, which makes the system slow to respond to complex or changing loads. For example, under different working conditions, load changes may cause system performance fluctuations, and traditional methods cannot automatically adjust the control strategy to cope with these fluctuations. As a result, the flexibility and adaptability of the system are greatly reduced, and it cannot cope with changes in complex dynamic environments.

[0005] Moreover, many current technologies cannot fully utilize real-time feedback data for adjustment. Although some systems have introduced sensors for monitoring, they are unable to respond quickly and make dynamic adjustments due to the lack of effective real-time data analysis and modeling. This lag not only affects the real-time response speed of the system, but also limits the system's ability to adapt in a changing environment. Summary of the invention

[0006] In view of the deficiencies of the prior art, the present invention provides a method for suppressing mechanical resonance of a servo control system, which solves the problems of lag, poor flexibility and insufficient adaptability of the control scheme in the prior art.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for suppressing mechanical resonance of a servo control system, comprising the following steps:

[0008] Establish a dynamic model of the servo system to describe the inertia, damping and stiffness characteristics of the servo system;

[0009] Based on the dynamic model, a digital twin model is created and dynamically updated through real-time data;

[0010] Perform frequency domain analysis based on real-time data to identify the natural frequency, and predict whether the system will enter a resonant state by comparing the relationship between the external excitation frequency and the natural frequency, and generate prediction results;

[0011] According to the prediction results, the control gain of the servo system is automatically adjusted, and the control gain is dynamically adjusted based on real-time feedback to avoid the influence of resonance on system performance.

[0012] Preferably, the step of establishing the kinetic model comprises:

[0013] Inertia description: Describe the mass and inertia characteristics of system components through mass matrix;

[0014] Damping description: The friction and resistance effects of the system are represented by the damping matrix;

[0015] Stiffness description: The stiffness of the system is expressed by the stiffness matrix, which reflects the degree of deformation under the action of external forces;

[0016] Global dynamic equations: Combined with mass, damping, and stiffness matrices, they describe the overall response of the system to external excitations.

[0017] Preferably, the real-time data includes displacement, velocity and acceleration.

[0018] Preferably, the digital twin model updates the system status in real time based on a numerical integration method.

[0019] Preferably, the numerical integration method adopts the Runge-Kutta method to simulate the dynamic behavior of the system.

[0020] Preferably, the frequency domain analysis uses fast Fourier transform to analyze the real-time data to extract the frequency spectrum characteristics of the system and further identify the resonant frequency.

[0021] Preferably, the prediction result determines the main natural frequencies of the system through modal analysis, the natural frequencies are calculated from the mass matrix and the stiffness matrix, and the modal analysis is used to identify multiple natural frequencies in the system.

[0022] Preferably, the control gain is adjusted by a PID controller, and the proportional gain, integral gain and differential gain of the PID controller are automatically adjusted according to the real-time resonance prediction information.

[0023] Preferably, the control gain is performed based on a model reference adaptive control algorithm, and the MRAC algorithm automatically adjusts the gain according to the real-time system status to ensure that the system can respond quickly and operate stably when resonance occurs.

[0024] Preferably, the gain adjustment is automatically reduced when resonance occurs, and automatically increased after the system returns to a normal state, so as to maintain the accuracy and response speed of the system.

[0025] The present invention provides a method for suppressing mechanical resonance of a servo control system. It has the following beneficial effects:

[0026] 1. The present invention combines digital twin technology with the servo system dynamics model, and continuously corrects the digital twin model through real-time sensor data, thereby achieving real-time monitoring and dynamic updating effects, so that the state of the virtual model is always synchronized with the physical system. The difference from the static model in the prior art is that the traditional method cannot cope with the dynamic changes of the system caused by environmental changes, resulting in a lag in prediction and control. Therefore, through the feedback of real-time data, the digital twin technology enables the system to perceive changes and make adjustments in real time, greatly improving the adaptability and accuracy of the system.

[0027] 2. The present invention can accurately predict whether the system enters a resonant state through frequency domain analysis and resonance prediction, and through fast Fourier transform, the system analyzes the difference between the external excitation frequency and the natural frequency in real time, thereby identifying potential resonance risks. When the external excitation frequency is close to the natural frequency, the model sends out an alarm signal in advance to avoid the occurrence of resonance. Compared with the early warning method based on experience or simple monitoring in traditional technology, the present invention can accurately predict before the resonance occurs through the combination of real-time frequency domain data and mathematical models, which significantly improves the safety and accuracy of the system.

[0028] 3. The present invention automatically adjusts the gain of the PID controller through a gain adjustment strategy based on real-time feedback. Therefore, when the occurrence of a resonant state is predicted, the system will automatically reduce the proportional gain, integral gain and differential gain to reduce the system response amplitude and prevent excessive oscillation. Unlike the traditional control method using fixed gain in the prior art, the present invention can flexibly adjust the gain according to real-time dynamic changes to ensure that the system remains stable even when resonance occurs. It not only performs well under static conditions, but can also self-adjust under dynamic disturbances, thereby enhancing the robustness and anti-interference ability of the system.

[0029] 4. The present invention combines Lyapunov stability analysis and adaptive gain adjustment to ensure the stable operation of the servo system in a complex environment. The system uses Lyapunov function analysis to adjust the gain in real time, ensuring that the system can maintain stability when subjected to external disturbances or resonance. Compared with the prior art that simply relies on gain adjustment or traditional PID control methods, the present invention combines stability analysis with dynamic gain adjustment to ensure the accuracy and stability of the servo system under non-ideal conditions. Therefore, it not only improves the system's adaptability to parameter changes, but also significantly reduces the risk of system loss of control due to resonance or external disturbances. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments 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.

[0032] In order to better understand the present invention, the above contents are described in detail below in conjunction with specific embodiments.

[0033] Please see attached Figure 1 , an embodiment of the present invention provides a method for suppressing mechanical resonance of a servo control system, comprising the following steps:

[0034] Establish a dynamic model of the servo system to describe the inertia, damping and stiffness characteristics of the servo system;

[0035] In this embodiment, a dynamic model of the servo system is established to describe the inertia, damping and stiffness characteristics of the servo system, thereby providing a theoretical basis for subsequent resonance prediction, digital twin modeling and gain adjustment. Specifically, the dynamic model of the servo system describes the physical properties of the system such as inertia, damping and stiffness by combining the mass matrix, damping matrix and stiffness matrix, and these properties jointly determine the response behavior of the system to external excitation.

[0036] The steps to establish the servo system dynamics model are as follows:

[0037] First, the dynamic behavior of the servo system can be expressed by the following overall dynamic equation;

[0038] The overall dynamic equation is described by inertia, damping and stiffness:

[0039] in:

[0040] Inertia Description:

[0041] The inertia of the servo system is described by the mass matrix M. Inertia reflects the system's ability to respond to external forces. The elements of the inertia matrix correspond to the mass of the servo system components and describe the inertial characteristics of each degree of freedom. For a single-degree-of-freedom system, inertia can be described by the total mass. For a multi-degree-of-freedom system, the mass matrix takes into account the mass and inertia of each component in the system through a weighted summation.

[0042] In general, the mass matrix is ​​a diagonal matrix that represents the mass of each degree of freedom of the system. For complex multi-degree-of-freedom systems, the mass matrix not only considers the mass of a single degree of freedom, but also includes the mass coupling relationship between the degrees of freedom.

[0043] Specifically, the mass matrix M is defined as follows:

[0044]

[0045] Among them, m 1 ,m 2 ,…,m n is the mass of each degree of freedom, reflecting the inertial characteristics of each component in the system.

[0046] Damping Description:

[0047] The damping characteristics of the system are represented by the damping matrix C. Damping is the system's ability to attenuate vibrations and is usually proportional to the system's speed. In a servo system, factors such as friction and air resistance all play a damping role, helping the system reduce vibrations and restore balance. The damping coefficient is related to the speed of the system, so the elements in the damping matrix are also proportional to the speed of the system's movement.

[0048] The damping matrix C is also usually expressed as a diagonal matrix, which represents the damping coefficient of each degree of freedom of the system. For some complex systems, the damping matrix may also contain coupling damping terms between degrees of freedom. In general, the damping matrix of the system is as follows:

[0049]

[0050] Among them, c 1 ,c 2 ,…,c n is the damping coefficient of each degree of freedom, which represents the friction and energy loss characteristics of the system.

[0051] Stiffness Description:

[0052] The stiffness of the system is described by the stiffness matrix K. Stiffness reflects the system's ability to resist deformation. The greater the stiffness, the smaller the system deformation. For a servo system, the stiffness mainly comes from the load of the motor, the components of the transmission system, and any components with elastic properties. The stiffness matrix of the system not only reflects the independent stiffness of each degree of freedom, but may also contain coupling stiffness terms between degrees of freedom, especially in multi-degree-of-freedom systems.

[0053] The stiffness matrix K is defined as:

[0054]

[0055] Among them, k 1 ,k 2 ,…,k 1n ,k 2n ,k n is the stiffness coefficient of each degree of freedom.

[0056] Therefore, the comprehensive kinetic equation can be expressed as:

[0057]

[0058] Where M is the mass matrix, which describes the mass and inertia characteristics of each component of the servo system. is the acceleration of the system, reflecting the dynamic response of the system, C is the damping matrix, describing the friction and energy consumption of the system, is the velocity of the system, reflecting the motion of the system, K is the stiffness matrix, describing the rigidity characteristics of the system, x(t) is the displacement of the system, indicating the deformation of the system, and F(t) is the external excitation force, indicating the control input or other external disturbances.

[0059] Therefore, from this equation, it can be seen that when an external excitation F(t) is applied to the system, the system will produce corresponding displacement, velocity and acceleration according to its mass, damping and stiffness characteristics. This model provides the basis for digital twin modeling and resonance prediction of servo systems.

[0060] Based on the dynamic model, a digital twin model is created and dynamically updated through real-time data;

[0061] In this embodiment, a digital twin model is used to model and simulate the system. The digital twin model is a virtual copy of the physical system that can run synchronously with the physical system through real-time data. When creating a digital twin model, the system is modeled based on physical properties such as mass matrix, damping matrix and stiffness matrix based on the servo system dynamics model established above. Through digital twins, the dynamic behavior of the servo system can be simulated in real time, and the model can be dynamically updated according to real-time data, so that the control system can make the best decision based on the latest status.

[0062] Specifically, the steps to create a digital twin model and dynamically update it are as follows:

[0063] First, the creation of the digital twin model is based on the dynamic model of the servo system. Through the combination of mass matrix, damping matrix and stiffness matrix, the inertia, damping and stiffness characteristics of the servo system are converted into a digital simulation model. Specifically, the process of establishing the digital twin model includes the following aspects:

[0064] Dynamic model foundation: In the previous step, the dynamic equations of the servo system have been established, which combine the physical properties of inertia, damping and stiffness. The digital twin model establishes a virtual system model based on these physical properties. Through this model, the dynamic behavior of the servo system can be simulated in real time, thereby obtaining virtual data consistent with the actual system state.

[0065] Real-time data input and dynamic update: In order for the digital twin model to accurately reflect the real-time status of the system, it is necessary to receive sensor data from the system in real time, including but not limited to displacement, velocity, acceleration, etc. Through these real-time data, the digital twin model can continuously update its own status, so that the output of the model is always consistent with the physical system.

[0066] Specifically, the digital twin model is calculated and updated in real time through numerical integration methods. In some embodiments, the state of the system is discretized and updated using the Runge-Kutta method. This method can efficiently simulate the dynamic response of the system by numerically solving the dynamic equations.

[0067] The state variables of the system include displacement, velocity, and acceleration. The digital twin model calculates the changes of these state variables by using numerical integration methods and compares them with the actual data of the physical system.

[0068] Numerical integration and dynamic simulation:

[0069] Specifically, the digital twin model uses numerical integration methods for dynamic simulation to achieve real-time prediction and update of system behavior. In one possible implementation, the Runge-Kutta method is used for dynamic simulation. Specifically, the Runge-Kutta method discretizes the dynamic equations of the servo system and converts the continuous state of the system into a discrete state for numerical calculation. This method can provide high calculation accuracy and can effectively handle nonlinear characteristics in the system.

[0070] When applied to servo systems, the digital twin model continuously calculates and updates the system status through the Runge-Kutta method. Each calculation result will be compared with the actual system data, and the model will be adjusted accordingly to ensure that the model is always consistent with the physical system.

[0071] Real-time data acquisition and synchronization:

[0072] The accuracy and real-time performance of the digital twin model depends on the data provided by various sensors in the system. These sensors include but are not limited to accelerometers, position sensors, and velocity sensors, which are responsible for collecting dynamic data of the servo system in real time. Specifically, the collection and synchronization of real-time data needs to meet the following requirements:

[0073] Data acquisition: The system’s displacement, velocity, acceleration and other status data are acquired in real time through sensors. The acquired data will directly affect the update speed and accuracy of the digital twin model.

[0074] Data synchronization: To ensure that the digital twin model is synchronized with the actual system, the real-time and consistency of the data must be guaranteed. Real-time data will be input into the model to update the current state of the system, ensuring that the feedback loop between the digital twin model and the physical system is always closed.

[0075] In some embodiments, real-time data is transmitted to the computing platform through a wireless communication module or a wired communication interface, and the platform processes the data to update the digital twin model. The output of the computing platform will directly affect the control strategy of the system so as to adjust the control parameters of the servo system in real time.

[0076] Role and application of digital twin models:

[0077] The core function of the digital twin model is to provide a virtual environment consistent with the physical system through simulation and dynamic update. With the help of this model, the control system can monitor the status of the servo system in real time and adjust the control strategy based on real-time data. The digital twin model can not only help predict and identify resonance problems, but also judge the stability and performance of the system by comparing simulation data with actual data, thus providing a basis for subsequent gain adjustment control.

[0078] In some embodiments, it is assumed that the servo system includes components such as a motor, a load, and a reducer, and the system is equipped with an acceleration sensor, a position sensor, etc. The digital twin model dynamically simulates the speed of the motor, the force of the load, and the working state of the reducer based on the dynamic equations and real-time data of the servo system. Through real-time updates, the digital twin model can not only accurately describe the behavior of the system in the current state, but also predict the dynamic response of the system in the future.

[0079] For example, under certain working conditions, changes in the motor's speed and load may cause the system to approach a resonant state. The digital twin model can identify this potential risk in advance and send an alarm signal to the control system. At this time, the control system can adjust the control gain based on the information provided by the digital twin model to suppress possible resonance and ensure stable operation of the system.

[0080] Therefore, a digital twin model was created based on the servo system dynamics model and dynamically updated with real-time data. The digital twin model can simulate the dynamic behavior of the system in real time and continuously update according to sensor data to ensure that the virtual model is consistent with the actual physical system. Through digital twin technology, the control system can predict and adjust the behavior of the servo system with the support of real-time data, effectively avoid the occurrence of mechanical resonance, and improve the stability and accuracy of the system.

[0081] Perform frequency domain analysis based on real-time data to identify the natural frequency, and predict whether the system will enter a resonant state by comparing the relationship between the external excitation frequency and the natural frequency, and generate prediction results;

[0082] In this embodiment, frequency domain analysis is used to monitor the dynamic behavior of the servo system, identify the natural frequency, and predict the resonant state of the system. The resonance of the servo system usually occurs when the external excitation frequency is close to or equal to the natural frequency of the system. Through the frequency domain analysis of real-time data, the natural frequency can be accurately identified, and the relationship between the external excitation frequency and the natural frequency can be compared, so as to predict in advance whether the system enters the resonant state. Therefore, the occurrence of mechanical resonance can be avoided by timely adjusting the control strategy through the prediction process.

[0083] Specifically, the steps of performing frequency domain analysis based on real-time data, identifying the natural frequency and predicting the resonant state are as follows:

[0084] First, the dynamic response of the servo system is usually evaluated through frequency domain analysis. Frequency domain analysis can convert the time domain response of the system into frequency components, helping us identify the natural frequency of the system and judge the stability of the system. Through real-time data acquisition and frequency domain analysis, we can identify the response characteristics of the system at different frequencies and compare them with the external excitation frequency to predict whether the system is in a resonant state.

[0085] Frequency Domain Analysis and Natural Frequency Identification

[0086] Frequency domain analysis uses fast Fourier transform (FFT) to analyze real-time data and convert the system's time domain response into a frequency domain representation. By performing Fourier transform on the system output signal, the system's spectrum can be obtained, which contains the amplitude information of each frequency component of the system.

[0087] Specifically, the external excitation frequency is the frequency of the excitation signal introduced by the control system or external interference. By collecting the state data of the servo system such as displacement, velocity and acceleration in real time and performing FFT analysis on these data, the natural frequency of the system can be obtained. The natural frequency is the frequency of the natural vibration of the system when there is no external excitation, which is usually determined by the mass and stiffness of the system.

[0088] For a servo system, the natural frequency can be calculated from the system's mass matrix M and stiffness matrix K. For example, the natural frequency of a single degree of freedom system can be expressed as:

[0089]

[0090] Among them, ω n is the natural frequency, k is the stiffness of the system, and m is the mass of the system. In a multi-degree-of-freedom system, the calculation of the natural frequency is more complicated, and it is usually necessary to obtain the various modal frequencies of the system through modal analysis.

[0091] Comparison of external excitation frequency with natural frequency

[0092] The spectrum obtained through frequency domain analysis can help us identify the natural frequency of the system. In practical applications, the external excitation frequency will change continuously and may be close to or coincide with some natural frequencies of the system. Generally, resonance occurs when the external excitation frequency is close to the natural frequency of the system. At this time, the amplitude of the system will increase sharply, which may cause system instability or damage.

[0093] In order to determine whether to enter the resonant state, the relationship between the external excitation frequency and the natural frequency is compared in this embodiment. When the difference between the external excitation frequency and a certain natural frequency of the system is less than a certain threshold, the system may enter the resonant state. This threshold can be adjusted according to the damping characteristics of the system. The response of a system with less damping is more likely to enter the resonant state.

[0094] Specifically, when the external excitation frequency f ext and the system's natural frequency f nat When the difference between them is less than a preset tolerance value, the system will be judged to have entered a resonant state. This can be judged by the following relationship:

[0095] |f ext -f nat |≤∈

[0096] Among them, f ext is the external excitation frequency, f nat is the natural frequency, ∈ is the tolerance value, indicating that the difference between the external excitation frequency and the natural frequency is within an acceptable range.

[0097] Generate prediction results

[0098] After performing frequency domain analysis and comparing the external excitation frequency with the natural frequency, the digital twin model generates a prediction based on the calculation results. If the system's external excitation frequency is close to the natural frequency and meets the conditions for resonance to occur, the digital twin model predicts that the system may enter a resonant state and generates a warning signal. Based on the prediction results, the system's control strategy will be adjusted accordingly to avoid excessive oscillation of the system.

[0099] Specifically, the digital twin model continuously monitors real-time data. When it finds that the external excitation frequency is close to the natural frequency, the model will issue an alarm in advance, indicating that the control system may enter a resonant state. Based on this alarm, the control system will automatically adjust the control gain, reduce the amplitude of the excitation force, or change the excitation frequency to prevent the system from entering a resonant state.

[0100] Moreover, the digital twin model can not only predict the resonant state of the system in real time, but also make feedback adjustments based on real-time data. By continuously updating the dynamic behavior of the system, the digital twin model can make timely adjustments when the system state changes, thereby ensuring that the servo system is always in a stable state. At this time, the gain adjustment function of the control system will automatically adjust according to the prediction results to ensure that the system does not enter a resonant state.

[0101] Therefore, by performing frequency domain analysis on the real-time data of the servo system, the natural frequency of the system can be accurately identified and compared with the external excitation frequency to predict whether the system has entered a resonant state. Subsequently, the digital twin model can effectively simulate and predict the dynamic behavior of the system, discover potential resonance risks in advance, and avoid the adverse effects of resonance by adjusting the control strategy in real time. This provides a resonance suppression method for the servo system, greatly improving the stability and accuracy of the system.

[0102] According to the prediction results, the control gain of the servo system is automatically adjusted, and the control gain is dynamically adjusted based on real-time feedback to avoid the impact of resonance on system performance.

[0103] In this embodiment, a dynamic control gain adjustment method based on real-time feedback is adopted. According to the frequency domain analysis and the prediction results of the system resonance, the control gain of the system can be adjusted in real time to cope with the occurrence of the resonance phenomenon and ensure that the system is always in a stable operating state. In this way, the resonance caused when the external excitation frequency is close to the natural frequency of the system is avoided, thereby ensuring the efficient and stable operation of the servo system.

[0104] Specifically, the steps for automatically adjusting the servo system control gain are as follows:

[0105] First, in the previous step, the relationship between the external excitation frequency and the natural frequency was evaluated in detail through frequency domain analysis, and the system was predicted to be close to the resonance state. When the external excitation frequency is close to or matches the natural frequency of the system, the digital twin model predicts the occurrence of resonance and generates an alarm signal. Based on this prediction result, the system enters the gain adjustment mode and automatically adjusts the control gain.

[0106] Automatic adjustment mechanism for control gain

[0107] Control gain is a key factor affecting the performance of the servo system, especially when the system enters a resonant state, the adjustment of gain is particularly important. When resonance occurs, the response of the system will increase rapidly, which may cause the system to lose control. To avoid this situation, the control gain in the present invention is dynamically adjusted according to real-time feedback information. Specifically, when the possibility of resonance increases, the system will automatically reduce the gain, thereby weakening the response amplitude of the system and avoiding excessive oscillation.

[0108] Generally speaking, the control gain of the servo system includes the proportional gain K p , integral gain K i and differential gain K d These gains determine how quickly and accurately the system responds to errors. When resonance occurs, the control gains are automatically reduced to slow down the system's response and prevent the system from over-oscillating.

[0109] Among them, the control input of the PID controller can be expressed as:

[0110]

[0111] Where u(t) is the output of the controller (i.e., the control signal), which serves as the control input of the system to adjust the behavior of the system; e(t) is the error of the system, which represents the difference between the expected output and the actual output. The formula is e(t) = x ref (t)-x(t), where x ref (t) is the expected output, x(t) is the actual output, ∫e(t)dt represents the integral part of the error, which helps the system eliminate the long-term accumulated error. It represents the derivative part of the error, helps the system respond to the speed of error changes and suppresses rapidly changing errors.

[0112] Therefore, the adjustment of PID gain depends on real-time feedback information, especially when resonance may occur, the gain will be automatically reduced. Specifically, the proportional gain, integral gain and differential gain will be automatically adjusted according to the system status and resonance prediction results.

[0113] Real-time feedback and dynamic adjustment

[0114] As an option, the system dynamically adjusts the control gain through real-time sensor data (such as displacement, velocity, acceleration, etc.). Specifically, the control gain is adjusted based on the following two key factors:

[0115] Resonance prediction results of the system: When the external excitation frequency is close to the natural frequency of the system, the system will predict the possibility of resonance and adjust the control gain accordingly to reduce the response of the system.

[0116] Real-time feedback signal: Through the real-time data obtained by the sensor, the system will monitor the changes of state variables such as displacement and speed in real time, and dynamically adjust the control gain according to the changes of these data.

[0117] Specifically, when the system is detected to enter a resonant state, the control gain will automatically decrease according to the set rules. The gain can be reduced by reducing the proportional gain K p , integral gain K i Or differential gain K d The process of reducing the gain is done gradually to ensure that the system can transition smoothly and avoid excessive oscillation.

[0118] Gain Adjustment and Stability Analysis

[0119] In order to ensure the stability of the system, the present invention also combines the Lyapunov stability theorem to verify the stability of the system under dynamic gain adjustment. The stability of the system is analyzed by constructing the Lyapunov function V(x):

[0120] V(x)=x T Px

[0121] Among them, P is a symmetric positive definite matrix, usually a constant matrix, representing the mass, stiffness, damping and other characteristics of the system, or used for weighted state variables in system stability analysis, and x is the state vector of the system, which represents the state of the system at a certain moment. It is composed of the displacement, velocity, etc. of the system, and x T is the transpose of the state vector x, which ensures the legitimacy of the matrix multiplication.

[0122] In some embodiments, it is assumed that the motor, load, and transmission system in the servo system have a certain natural frequency, and the system's external excitation frequency is close to the natural frequency. In this case, the digital twin model predicts that the system enters a resonant state and issues an alarm signal. The control system automatically adjusts the proportional gain, integral gain, and differential gain based on the prediction results to prevent excessive oscillation of the system by reducing the amplitude of the control input. At this point, the dynamic response of the system will gradually decrease, avoiding excessive oscillation and stabilizing in the desired working state.

[0123] For example, suppose the system's external excitation frequency is f extClose to the natural frequency f of the system nat , the control system automatically reduces the gain through feedback information and prediction results. In this way, the response speed of the system is controlled within a reasonable range, thus avoiding instability caused by resonance.

[0124] Therefore, by automatically adjusting the control gain of the servo system, it is ensured that the system can respond in time and reduce the oscillation amplitude of the system when resonance occurs. The dynamic adjustment mechanism of the control gain is based on real-time feedback and resonance prediction results, which can effectively avoid the negative impact of resonance on the performance of the servo system. By adjusting the proportional gain, integral gain and differential gain in real time, the present invention can ensure the stability and accuracy of the system under different working conditions. This method not only improves the robustness of the servo system, but also ensures that the system can continue to operate efficiently under various dynamic conditions.

[0125] 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 method for suppressing mechanical resonance of a servo control system, characterized in that: The following steps are involved: Establish a dynamic model of the servo system to describe the inertia, damping and stiffness characteristics of the servo system; Based on the dynamic model, a digital twin model is created and dynamically updated through real-time data; Perform frequency domain analysis based on real-time data to identify the natural frequency, and predict whether the system will enter a resonant state by comparing the relationship between the external excitation frequency and the natural frequency, and generate prediction results; According to the prediction results, the control gain of the servo system is automatically adjusted, and the control gain is dynamically adjusted based on real-time feedback to avoid the influence of resonance on system performance.

2. A method for suppressing mechanical resonance of a servo control system according to claim 1, characterized in that: The steps of establishing the kinetic model include: Inertia description: Describe the mass and inertia characteristics of system components through mass matrix; Damping description: The friction and resistance effects of the system are represented by the damping matrix; Stiffness description: The stiffness of the system is expressed by the stiffness matrix, which reflects the degree of deformation under the action of external forces; Global dynamic equations: Combined with mass, damping, and stiffness matrices, they describe the overall response of the system to external excitations.

3. A method for suppressing mechanical resonance of a servo control system according to claim 1, characterized in that: The real-time data includes displacement, velocity and acceleration.

4. A method for suppressing mechanical resonance of a servo control system according to claim 1, characterized in that: The digital twin model updates the system status in real time based on a numerical integration method.

5. A method for suppressing mechanical resonance of a servo control system according to claim 4, characterized in that: The numerical integration method adopts the Runge-Kutta method to simulate the dynamic behavior of the system.

6. A method for suppressing mechanical resonance of a servo control system according to claim 1, characterized in that: The frequency domain analysis uses fast Fourier transform to analyze the real-time data to extract the frequency spectrum characteristics of the system and then identify the resonant frequency.

7. A method for suppressing mechanical resonance of a servo control system according to claim 1, characterized in that: The prediction results determine the main natural frequencies of the system through modal analysis, which is calculated from the mass matrix and the stiffness matrix. The modal analysis is used to identify multiple natural frequencies in the system.

8. The method for suppressing mechanical resonance of a servo control system according to claim 1, characterized in that: The control gain is adjusted by a PID controller, and the proportional gain, integral gain and differential gain of the PID controller are automatically adjusted according to the real-time resonance prediction information.

9. A method for suppressing mechanical resonance of a servo control system according to claim 8, characterized in that: The control gain is performed based on a model reference adaptive control algorithm, and the MRAC algorithm automatically adjusts the gain according to the real-time system status to ensure that the system can respond quickly and operate stably when resonance occurs.

10. The method for suppressing mechanical resonance of a servo control system according to claim 1, characterized in that: The gain adjustment automatically decreases when resonance occurs and automatically increases after the system returns to normal state to maintain the accuracy and response speed of the system.

Images