An adaptive control method and system for a CLLC converter

Abstract

The present application relates to the field of power electronics, especially to a kind of adaptive control method and system for CLLC converter, by real-time acquisition of the input and output signal of the converter, system identification is carried out using Kalman filter, and the control parameters of PID controller are automatically adjusted.First, by dynamic monitoring and analysis to the input and output of system, the simplified difference equation of system is obtained;Then, based on the system identification result, the parameters of PID controller are adjusted using pole placement method, and the real-time correction of controller is realized.At the same time, the present application also uses polynomial fitting method to estimate the current value of resonant capacitor, further enhances the adaptability of control system.The control strategy of the present application is based on the change of the working condition of the converter, and the control parameters can be continuously optimized in the whole use cycle, to ensure that the converter can still maintain stable and efficient operating performance under high intensity use, which helps to improve the operation reliability and maintenance efficiency of equipment.

Description

Technical Field

[0001] This invention relates to the field of power electronics technology, and in particular to an adaptive control method and system for CLLC converters. Background Technology

[0002] CLLC converters are high-efficiency power electronic converters widely used in power conversion, variable frequency speed control, and power transmission. During operation, the resonant capacitor plays a crucial role, influencing the converter's frequency response, stability, and dynamic performance. However, with prolonged use, the resonant capacitor endures high voltage and current stresses and is affected by material aging, dielectric losses, and environmental factors (such as temperature and humidity), leading to a gradual decrease in capacitance.

[0003] Capacitor decay directly affects converter performance, particularly changes in resonant frequency and operating point, thereby reducing system stability and dynamic response. After the capacitor value decays to a certain extent, the originally designed controller parameters may fail to effectively control the system, or even lead to system instability. Therefore, maintaining converter stability and efficient operation under capacitor decay has become a pressing technical problem. Traditional PID control methods are widely used in converter control, but they lack adaptability and typically assume that system parameters remain constant. In practical applications, as the capacitor decays, fixed PID control parameters cannot cope with changes in system parameters, leading to a decrease in control accuracy and response speed. To overcome this problem, some studies have proposed adaptive control-based methods, aiming to adjust controller parameters in real time to adapt to system changes. However, these methods still face challenges such as accurately estimating capacitor decay, updating controller parameters in real time, and maintaining high identification accuracy in high-noise environments.

[0004] Current technology has not yet provided an adaptive control scheme that can both estimate capacitor value changes in real time and efficiently update control parameters. Summary of the Invention

[0005] In view of this, the purpose of this invention is to propose an adaptive control method and system for CLLC converters to solve the problems of existing methods being unable to accurately estimate capacitor attenuation, update controller parameters in real time, and maintain high identification accuracy in high-noise environments.

[0006] To achieve the above objectives, this invention provides an adaptive control method for a CLLC converter. The CLLC converter employs PID control, and the adaptive control method includes the following steps:

[0007] The Kalman filter algorithm is used to identify the CLLC converter and estimate the state of the CLLC converter in real time.

[0008] Based on the estimated CLLC converter state, the parameters of the PID controller are dynamically adjusted using the pole placement method.

[0009] This paper analyzes the impact of capacitor attenuation on the transfer function and difference equation of the Kalman filter algorithm.

[0010] During the operation of the CLLC converter, the capacitance value is estimated in real time, and the control strategy is updated based on the dynamic changes in capacitance decay.

[0011] Preferably, the Kalman filter algorithm is dynamically updated based on the state-space model.

[0012] Preferably, the PID controller is an incremental PID controller.

[0013] Preferably, the parameters of the PID controller are dynamically adjusted using the pole placement method, including:

[0014] Discretize the incremental PID controller into a discrete incremental PID controller;

[0015] By combining the z-domain transfer functions of the CLLC converter and the discrete incremental PID controller, a closed-loop system consisting of the CLLC converter and the discrete incremental PID controller is obtained.

[0016] The poles of the closed-loop system are configured using the pole placement method.

[0017] The parameters of the discrete incremental PID controller are updated using the method of undetermined coefficients.

[0018] Preferably, the capacitance value is estimated using a fifth-order polynomial fitting algorithm to estimate the actual capacitance values ​​of the high-voltage and low-voltage side capacitors.

[0019] Preferably, the fifth-order polynomial fitting algorithm includes an encoding process step, which treats the high-voltage side and low-voltage side capacitors as a whole and performs joint fitting by combining the system difference equation.

[0020] The present invention also provides an adaptive control system for a CLLC converter, wherein the CLLC converter employs PID control, and the control system includes:

[0021] The system state identification module uses the Kalman filter algorithm to estimate the state of the converter in real time;

[0022] The controller parameter update module dynamically adjusts the PID controller parameters based on the system state estimate using the pole placement method.

[0023] The capacitance estimation module is used to estimate the capacitance values ​​of the high-voltage and low-voltage side capacitors in real time, so as to adjust the control strategy according to the capacitance changes.

[0024] An adaptive feedback loop automatically updates controller parameters through a self-tuning PID control strategy.

[0025] Preferably, the feedback loop automatically adjusts the controller parameters based on the current system state estimate after each update cycle.

[0026] The present invention also provides an adaptive PID control method for capacitor attenuation detection and control of CLLC converter. The method uses a Kalman filter algorithm to identify the input and output signals of the converter, estimates the capacitance value change in real time, and adjusts the parameters of the PID controller based on the result.

[0027] Preferably, the adaptive PID control method further includes: when the capacitor decays to a critical value, taking compensation measures or capacitor replacement schemes in advance by estimating the capacitor change trend in advance.

[0028] The beneficial effects of this invention are as follows: This invention effectively addresses the performance degradation of converters caused by capacitor decay. Compared to traditional fixed controller parameter schemes, this invention can automatically adapt to system changes and continuously maintain optimal system performance. Furthermore, the capacitor value estimation algorithm designed in this invention can predict and implement corresponding compensation measures before the capacitor value decays to a critical level, ensuring the reliability and efficient operation of the converter during long-term use.

[0029] The adaptive PID control method and system of the present invention have broad application prospects, and are particularly suitable for systems such as power electronic converters that require long-term stable operation, which can improve the operational reliability of the equipment and extend its service life. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a schematic diagram of the adaptive control system structure for a CLLC converter according to an embodiment of the present invention;

[0032] Figure 2 This invention relates to the basic principle of identification based on a Kalman filter system in its embodiments.

[0033] Figure 3 This is the process of using the Kalman filter algorithm for system identification in an embodiment of the present invention;

[0034] Figure 4 This is a control block diagram of the incremental PID controller in an embodiment of the present invention;

[0035] Figure 5 This is a reference diagram of the closed-loop poles after PID controller parameter correction in an embodiment of the present invention;

[0036] Figure 6 This is a flowchart of the self-calibrating PID control algorithm in an embodiment of the present invention;

[0037] Figure 7 This is a flowchart illustrating the fitting of the resonant capacitor value in an embodiment of the present invention;

[0038] Figure 8 This is a diagram of the decoupling algorithm for resonant capacitor fitting in an embodiment of the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0040] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0041] like Figure 1 The system shown is a structural diagram of the adaptive control method for a CLLC converter implemented in this invention. It is mainly used to address the instability and performance degradation caused by the decay of the resonant capacitor value during long-term use, which leads to a decline in the performance of the system control loop. The self-tuning PID algorithm uses a Kalman filter algorithm as the system parameter identification algorithm. It estimates the converter's state using the switching frequency input to the converter and the converter's output voltage. Then, based on the estimated system state, it updates the control parameters using the pole placement method, thereby achieving the algorithm's adaptability. Specifically, it includes the following steps:

[0042] The Kalman filter algorithm is used to identify the CLLC converter and estimate the state of the CLLC converter in real time.

[0043] Based on the estimated CLLC converter state, the parameters of the PID controller are dynamically adjusted using the pole placement method.

[0044] This paper analyzes the impact of capacitor attenuation on the transfer function and difference equation of the Kalman filter algorithm.

[0045] During the operation of the CLLC converter, the capacitance value is estimated in real time, and the control strategy is updated based on the dynamic changes in capacitance decay.

[0046] in Figure 2 The diagram illustrates the basic principle of the Kalman filter algorithm. This algorithm continuously updates the system state estimate by combining the system's predicted and actual measured values, aiming to minimize the covariance of the estimation error. During the iterative process of the Kalman filter, the state matrix represents the current state of the system, and this matrix is ​​composed of the system's state variables. In each iteration, the system uses the optimal state estimate from the previous update as the initial state estimate for the current update. The optimal estimate is obtained through Kalman gain weighting.

[0047] based on Figure 2 The basic principle of the Kalman filtering algorithm is shown. Figure 3 The flowchart illustrating the system identification process applied in this invention is shown. The Kalman filter algorithm is dynamically updated based on a state-space model, and its transfer function can be approximated by a second-order difference equation. The coefficients of the discrete difference equation are: a 1 , a 2 , b 0 , b 1 By inputting real-time data of the switching frequency and output voltage into the iterative process, the Kalman filter continuously adjusts these coefficients, gradually bringing them closer to the linearized discrete form at the operating point. Specifically, the model parameters in the Kalman filter are first optimized using existing data, and the optimized parameters are written into the control program. The input and output responses of the controlled system are continuously collected, and the system estimation results are dynamically updated.

[0048] Figure 4 The control block diagram for a PID controller controlling a CLLC system is shown below. The linearized mathematical model of the controlled object is as follows: (1)

[0049] (1-1)

[0050] (1-2)

[0051] In this invention, the PID controller is an incremental PID controller, which is discretized and mapped to a discrete incremental PID controller. That is, the controlled target is designed with a discrete incremental PID controller as follows:

[0052] (2)

[0053] The correspondence between the parameters in formula (1) and the parameters of the PID controller is as follows: (2-1)

[0054] (2-2)

[0055] (2-3)

[0056] By simultaneously solving the z-domain transfer functions of the controlled object and the controller, we obtain the expression for the closed-loop transfer function of the control system, which consists of a CLLC converter and a discrete incremental PID controller:

[0057] (2-4)

[0058] (2-5)

[0059] (2-6)

[0060] (2-7)

[0061] The characteristic equation of the final closed-loop system is shown in equation (2-7). In practical applications, one can choose... n T =2, and then the coefficients of the ideal closed-loop characteristic polynomial can be directly determined by the continuous second-order time closed-loop characteristic polynomial.

[0062] Figure 5 This demonstrates the relationship between the pole distribution and damping ratio of a second-order system. The closed-loop pole design of a discrete PID controller can refer to the pole distribution of an ideal second-order system. In this case, the poles of the second-order system can be expressed as... The discrete-domain poles are mapped to ,in Therefore, the design process for the closed-loop system only requires designing the damping coefficient and the oscillation angular frequency. When the real part of the pole of a continuous system is negative, the corresponding pole in the discrete domain will also fall inside the unit circle. Typically, the second-order optimal model sets the damping ratio ξ = 0.707. For ω n The sampling frequency is generally selected based on comprehensive sampling, and empirically it is designed to be 0.1 < T s ω n < 1. Examples selected in this specification ω n =20000.

[0063] After determining the pole distribution of the closed-loop system, the parameters of the PID controller are updated using the method of undetermined coefficients. The specific update process is as follows:

[0064] (3)

[0065] The coefficients of the same powers on both sides of equation (2-7) remain consistent. Based on this, the system of equations extracted is shown in equation (3). It is further rewritten in matrix form as shown in equation (3-1):

[0066] (3-1)

[0067] The updated PID controller parameters can be obtained directly by solving formula (3-1).

[0068] The coefficient matrix in formula (3-1) must be a non-singular matrix to guarantee that the equation has a solution. The conditions for the coefficient matrix to have a solution are shown in formulas (3-2) and (3-3):

[0069] (3-2)

[0070] (3-3)

[0071] In real-world systems, due to limitations in data precision and magnitude, it's extremely rare for the determinant of the coefficient matrix to be zero. Considering the possibility of no solution, Figure 6 The flowchart of the self-tuning PID algorithm is shown. During system operation, an update cycle is manually set. When the set cycle is reached, the adaptive control update program is initiated to update the PID controller parameters, as shown in the flowchart. Since cases where the equation has no solution are extremely rare, when such a case occurs, the controller parameters can be temporarily kept unchanged until the next update cycle arrives, at which point the controller parameters are updated uniformly based on the identification and calculation results. Therefore, this method fully considers the issue of unsolvable equations when setting the update cycle. The update cycle can be set shorter to reduce the impact of a missed update, or a longer update cycle can be used, with an additional shorter update time added when an unsolvable equation is detected to correct the controller parameters.

[0072] When the capacitance value degrades significantly, it becomes impossible to compensate for the system's performance by adjusting parameters. Therefore, estimating the remaining capacitance is equally crucial while dynamically adjusting control parameters. Accurate fitting to estimate the current state of the capacitor facilitates a reasonable assessment of the overall system condition, especially when the capacitance value approaches a critical value. Predicting trends allows for proactive countermeasures. When the capacitance value degrades to a critical level, existing control strategies alone may not be sufficient to maintain optimal system operation. In such cases, replacing the capacitor or updating system components is necessary to ensure the converter's continued stable operation. The flowchart of the capacitance fitting algorithm is shown below. Figure 7 As shown.

[0073] By processing the input and output variables of the controlled system, the true state of the system can be indirectly reflected. Theoretically, after processing a large amount of input and output data, artificial intelligence algorithms such as neural networks or machine learning can efficiently complete this task. However, for embedded systems, the aforementioned black-box algorithm process is unknown and cannot be directly written into the control program, and it requires significant computational resources, making it unsuitable for this application scenario. Therefore, this invention designs a fitting algorithm for capacitance estimation based on the system identification results of Kalman filtering.

[0074] By inputting the converter's input and output data into the Kalman filter system identification module, the Kalman filter algorithm is used to estimate the system state in real time, providing a basis for subsequent controller parameter adjustments. During system identification, parameters of the difference equations are extracted from the identified system model. a 1 , a 2 , b 0 , b 1 The coefficients of the difference equations directly reflect the current state of the system. Fifth-order polynomial fitting was performed on the capacitance values ​​of the high-voltage and low-voltage capacitors respectively, yielding two different fitting results. These two different results allow for a relatively accurate estimation of the actual capacitance value of the resonant capacitor in the resonant cavity, providing accurate system parameters for subsequent control algorithms, thereby ensuring the stability and efficiency of the converter under different operating conditions.

[0075] However, in actual design, it is not possible to simply fit the capacitance values ​​of the two capacitors separately, because the capacitance values ​​of the two capacitors have a decisive effect on the transfer function of the system. Therefore, there is a strong coupling between the transfer function difference equation and the two capacitors. Thus, when fitting, the two capacitors should be regarded as a whole, which together determines the coefficients of the system's difference equation. Therefore, in order to obtain the state of the two capacitors separately, a decoupling operation must be performed.

[0076] Figure 8 This paper demonstrates a fitting algorithm considering the coupling of capacitors on both sides. The basic idea of ​​this algorithm is to encode the capacitors on the high-voltage and low-voltage sides and analyze them as a whole. The encoded capacitance values ​​are then combined with the system difference equations for fitting. After decoding, the fitted result can accurately estimate the capacitance values ​​of both sides. It is important to note that the weighting of the capacitance values ​​during encoding directly affects the accuracy of the fitting. When the weightings of the capacitors on both sides are similar, the accuracy of the fitting result will have some deviation from the capacitance values ​​of the two capacitors. Conversely, if encoding with significantly different weightings is used for fitting, the capacitance value of the capacitor with the higher weighting will be more accurately estimated, while the capacitance value of the capacitor with the lower weighting will have a larger error. Therefore, the solution proposed in this embodiment is to perform two encodings, adjusting the weights of the capacitors on both sides in the encoding respectively, so that the weighting of each side is reflected in each fitting. Through two independent fittings, a more accurate estimate of the capacitance values ​​of both sides can be obtained.

[0077] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention is limited to these examples; within the framework of the invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity. Any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the scope of protection of the invention.

Claims

1. An adaptive control method for a CLLC converter, characterized in that, The CLLC converter employs PID control, and the adaptive control method includes the following steps: The Kalman filter algorithm is used to identify the CLLC converter and estimate the state of the CLLC converter in real time. Based on the estimated CLLC converter state, the parameters of the PID controller are dynamically adjusted using the pole placement method. This paper analyzes the impact of capacitor attenuation on the transfer function and difference equation of the Kalman filter algorithm. During the operation of the CLLC converter, the capacitance value is estimated in real time, and the control strategy is updated based on the dynamic changes in capacitance decay. The method of dynamically adjusting the parameters of the PID controller using the pole placement method involves combining the control object and the z-domain transfer function of the controller to obtain a closed-loop system consisting of a CLLC converter and a discrete incremental PID controller, and obtaining an expression for the closed-loop transfer function of the control system. After determining the distribution of poles in the closed-loop system, the parameters of the PID controller are updated using the method of undetermined coefficients. The coefficients of the same powers on both sides of the expression of the closed-loop transfer function of the control system are kept consistent, and the system of equations is extracted based on this. The extracted system of equations is further rewritten into matrix form; The updated parameters of the PID controller are obtained by solving the system of equations in matrix form.

2. The adaptive control method for a CLLC converter according to claim 1, characterized in that, The Kalman filter algorithm is dynamically updated based on a state-space model.

3. The adaptive control method for a CLLC converter according to claim 1 or 2, characterized in that, The PID controller is an incremental PID controller, and in the step of dynamically adjusting the parameters of the PID controller using the pole placement method, the PID controller is a discrete incremental PID controller.

4. The adaptive control method for the CLLC converter according to claim 1, characterized in that, The capacitance value is estimated using a fifth-order polynomial fitting algorithm to estimate the actual capacitance values ​​of the high-voltage and low-voltage side capacitors.

5. The adaptive control method for the CLLC converter according to claim 4, characterized in that, The fifth-order polynomial fitting algorithm includes an encoding process step, which treats the high-voltage side and low-voltage side capacitors as a whole and performs joint fitting by combining the system difference equation.

6. An adaptive control system for a CLLC converter, characterized in that, The CLLC converter employs PID control, and the control system includes: The system state identification module uses the Kalman filter algorithm to estimate the state of the converter in real time; The controller parameter update module dynamically adjusts the PID controller parameters based on the system state estimate using the pole placement method. The capacitance estimation module is used to estimate the capacitance values ​​of the high-voltage and low-voltage side capacitors in real time, so as to adjust the control strategy according to the capacitance changes. An adaptive feedback loop automatically updates controller parameters through a self-tuning PID control strategy; The method of dynamically adjusting the parameters of the PID controller using the pole placement method involves combining the control object and the z-domain transfer function of the controller to obtain a closed-loop system consisting of a CLLC converter and a discrete incremental PID controller, and obtaining an expression for the closed-loop transfer function of the control system. After determining the distribution of poles in the closed-loop system, the parameters of the PID controller are updated using the method of undetermined coefficients. The coefficients of the same powers on both sides of the expression of the closed-loop transfer function of the control system are kept consistent, and the system of equations is extracted based on this. The extracted system of equations is further rewritten into matrix form; The updated parameters of the PID controller are obtained by solving the system of equations in matrix form.

7. The adaptive control system for the CLLC converter according to claim 6, characterized in that, The feedback loop automatically adjusts the controller parameters based on the current system state estimate after each update cycle.

8. An adaptive PID control method for capacitor attenuation detection and control in a CLLC converter, characterized in that, The input and output signals of the converter are identified using the Kalman filter algorithm, and the capacitance value change is estimated in real time. Based on this, the parameters of the PID controller are adjusted. The parameters of the PID controller are dynamically adjusted using the pole placement method. The dynamic adjustment using the pole placement method includes combining the z-domain transfer functions of the controlled object and the controller to obtain a closed-loop system composed of the CLLC converter and the discrete incremental PID controller, and obtaining the expression of the closed-loop transfer function of the control system. After determining the distribution of poles in the closed-loop system, the parameters of the PID controller are updated using the method of undetermined coefficients. The coefficients of the same powers on both sides of the expression of the closed-loop transfer function of the control system are kept consistent, and the system of equations is extracted based on this. The extracted system of equations is further rewritten into matrix form; The updated parameters of the PID controller are obtained by solving the system of equations in matrix form.

9. The adaptive PID control method according to claim 8, characterized in that, The control method further includes: when the capacitor decays to a critical value, taking compensation measures or capacitor replacement plans in advance by estimating the capacitor change trend in advance.

Images