A method for improving the robustness of a deadbeat predictive control of a DAB converter

Abstract

A method for improving the robustness of deadbeat predictive control in a DAB converter belongs to the field of converter technology. The method is as follows: Measuring the electrical parameters of a dual active bridge DC-DC converter; designing a superhelical disturbance observer; discretizing the superhelical disturbance observer; and using the output information of the superhelical disturbance observer, namely the predicted value of the output voltage at time k+1 and the predicted value of the system lumped disturbance at time k+1, to calculate the control signal u at time k. * This invention proposes a topology for perturbation observation and compensation based on a perturbation observer. The perturbation observer, employing the superspiral principle, can achieve accurate observation and compensation of perturbations. Compared with traditional perturbation observation algorithms, it improves perturbation compensation performance and reduces chattering.

Description

Technical Field

[0001] This invention relates to a method for improving the robustness of deadbeat predictive control in DAB converters, belonging to the field of converter technology. Background Technology

[0002] The deadbeat output voltage control of a dual active bridge DC-DC converter (DAB) is extremely sensitive to the accuracy of system parameters. In practical applications, with changes in the environment or the aging of components, the values ​​of the inductance of the internal high-frequency transformer and the DC output capacitor of the DAB converter will undergo non-negligible changes. These changes may lead to a deterioration in the performance of the control system. In addition, potential errors in the measurement and manufacturing processes can further affect the accuracy of system parameters, such as steady-state errors, deterioration of dynamic performance, and circulating current problems in parallel structures, thereby increasing the difficulty of control.

[0003] In certain extreme cases, inaccurate parameters can lead to system oscillations, severely threatening the stability of the control system and potentially causing safety hazards such as overvoltage and overcurrent. Especially when the excitation capacity of the high-frequency transformer is limited, inaccurate parameters may cause the high-frequency transformer to saturate, resulting in unexpected current stress and potential damage to the converter. Therefore, improving the robustness of deadbeat output voltage control in dual active bridge DC-DC converters is crucial.

[0004] Existing deadbeat output voltage control methods for DAB converters typically employ a combined control strategy of a current stress-optimized model predictive (MPC) controller and a PI control module to improve system robustness. In this strategy, the MPC controller is used only for calculating the outward phase shift, while the PI control module is used for calculating the inward phase shift, and robustness is enhanced by adjusting parameters. However, while this method improves system robustness to some extent, the introduction of the PI control module may affect transient control performance, limiting the full realization of the inherent fast dynamic advantages of the MPC controller.

[0005] Another improvement is to use parameter identification techniques to estimate transformer leakage inductance online. However, this method is only applicable to single-phase-shift modulation strategies and cannot be applied to dual-phase-shift modulation strategies (DPS) with more degrees of control freedom. Furthermore, this method does not adequately consider compensation for output capacitor parameter mismatch, limiting its effectiveness in practical applications.

[0006] In recent years, model-free predictive control methods have also been applied to the control of DAB converters. Although this method has certain advantages in theory, its computational burden is too large to be effectively deployed in microprocessors, thus limiting its practical applications. Summary of the Invention

[0007] To address the problems existing in the background art, the present invention provides a method for improving the robustness of deadbeat predictive control in DAB converters.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: a method for improving the robustness of deadbeat predictive control in a DAB converter, the method comprising the following steps:

[0009] S1: Measure the electrical parameters of the dual active bridge DC-DC converter;

[0010] S2: Design a superhelical perturbation observer;

[0011] S201: Input control signal u, the control signal u being related to the output power P n The relationship is u = 0.5P n ;

[0012] S202: Obtain the load current I through a current sensor. load ;

[0013] S203: The output voltage U is obtained by measuring the voltage sensor. o ;

[0014] S204: Design of a superhelical perturbation observer:

[0015]

[0016] In formula (1):

[0017] The predicted value represents the output voltage;

[0018] This represents the output voltage error;

[0019] k1 and k2 represent the observer gain that needs to be adjusted;

[0020] b0 represents a constant;

[0021] sgn(*) represents a symbolic function;

[0022] C o Represents the output capacitor;

[0023] The predicted value representing the lumped disturbance of the system;

[0024] t represents time.

[0025] S3: Discretize the superhelical perturbation observer;

[0026] S301: Predicted value of output voltage Initial values ​​and predicted values ​​of system lumped disturbances The initial values ​​are all set to 0;

[0027] S302: Set the control cycle to T and the control frequency to f. s In each control cycle, the output voltage U at time t = k is acquired. o (k) and load current I load (k);

[0028] S303: Calculate the output voltage at time k+1 and the predicted value of the system's lumped disturbance, i.e., the expression for the discretized superspiral disturbance observer is shown below:

[0029]

[0030] S4: Utilizing the output information of the superhelical perturbation observer, i.e., the predicted value of the output voltage at time k+1. The predicted value of the system lumped disturbance at time k+1 Calculate the control signal u at time k * :

[0031]

[0032] In formula (3):

[0033] f s Represents the control frequency;

[0034] b0 represents a constant;

[0035] U ref Represents the given voltage;

[0036] I load (k) represents the load current at time k;

[0037] C o Represents the output capacitor;

[0038] S5: Calculate and constrain the phase shift angle.

[0039] S501: Based on the minimum current stress dual-phase-shift modulation strategy calculation method, the control signal u at time k is... * The phases are modulated into an inner phase shift angle D1 and an outer phase shift angle D2 and then applied to the DAB converter.

[0040] S502: Apply the following constraints to the outputs of the inner phase shift angle D1 and the outer phase shift angle D2 respectively to achieve precise and robust voltage control:

[0041]

[0042] Compared with the prior art, the beneficial effects of the present invention are:

[0043] This invention proposes a topology for disturbance observation and compensation based on a disturbance observer. The disturbance observer, employing the superhelical principle, enables precise observation and compensation of disturbances, improving disturbance compensation performance and reducing chattering compared to traditional disturbance observation algorithms. Furthermore, the observer described in this invention is designed based on the second-order sliding mode principle, hiding the sign function within the integral term, thus fundamentally eliminating chattering and improving voltage control performance and robustness. It effectively eliminates output voltage control errors caused by parameter errors. Attached Figure Description

[0044] Figure 1 This is a control block diagram of the present invention;

[0045] Figure 2 This is a comparison chart of the output voltage control results of the method of the present invention and the conventional method under the condition of 60% transformer inductance error and sudden change in load resistance. Among them: (a) represents the experimental result chart of the conventional deadbeat method, and (b) represents the experimental waveform chart of the present invention. The three waveforms from top to bottom represent the DAB converter port voltage, port output current and high-frequency transformer internal current, respectively. The load resistance switching occurs at the point where the current waveform increases step in the figure.

[0046] Figure 3 This is a comparison of the experimental performance waveforms of the method of the present invention and the conventional method when the output voltage drops from 80V to 60V with a step of 60% transformer inductance error and the load resistance remains at 30Ω. In this diagram, (a) represents the conventional deadbeat method and (b) represents the method of the present invention. Detailed Implementation

[0047] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the invention, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0048] This invention relates to a method for improving the robustness of deadbeat predictive control based on a superspiral disturbance observer, which is used to enhance the robustness of deadbeat output voltage control of a dual active bridge DC-DC converter (DAB) against parameter errors and external disturbances.

[0049] This invention requires the converter to have only one current sensor and one voltage sensor to improve the robustness of deadbeat output voltage control, without the need for additional hardware design, thus reducing hardware costs.

[0050] The algorithm can be programmed and implemented in a digital control chip (DSP), and can be applied to general-purpose dual active bridge DC-DC converter controllers, which has extremely high application and economic value.

[0051] A method for improving the robustness of deadbeat predictive control in a DAB converter according to the present invention includes the following steps:

[0052] S1: Measure the electrical parameters of the dual active bridge DC-DC converter through offline testing. These electrical parameters include the high-frequency transformer leakage inductance L. r0 Output capacitor C o The ratio of the number of turns on the primary and secondary sides of the transformer, n;

[0053] S2: Using the port output voltage and load current of the DAB converter under different operating states, a super-helical disturbance observer is designed based on the voltage and current equations of the DAB converter. The influence of non-ideal factors such as parameter errors and external disturbances during the voltage control process of the DAB converter is observed and suppressed. The super-helical disturbance observer is designed based on the super-helical principle, which can achieve finite-time convergence of observation errors. Moreover, the second-order sliding mode property can significantly suppress the chattering phenomenon caused by the traditional sliding mode structure, achieving the best observation effect.

[0054] S201: Input control signal u, the control signal u being related to the output power P n The relationship is u = 0.5P n ;

[0055] S202: Obtain the load current I through a current sensor. load ;

[0056] S203: The output voltage U is obtained by measuring the voltage sensor. o ;

[0057] S204: Design of a superhelical perturbation observer:

[0058]

[0059] In formula (1):

[0060] The predicted value represents the output voltage;

[0061] This represents the output voltage error;

[0062] k1 and k2 represent the observer gain that needs to be adjusted;

[0063] b0 represents a constant related to the electrical parameters, input voltage, and control frequency of the DAB converter;

[0064] sgn(*) represents a symbolic function;

[0065] C o Represents the output capacitor;

[0066] The predicted value representing the lumped disturbance of the system;

[0067] t represents time.

[0068] S3: Discretize the superspiral perturbation observer;

[0069] S301: Predicted value of output voltage Initial values ​​and predicted values ​​of system lumped disturbances The initial values ​​are all set to 0;

[0070] S302: Set the control cycle to T and the control frequency to f. s In each control cycle, the output voltage U at time t = k is acquired. o (k) and load current I load (k);

[0071] S303: Calculate the output voltage at time k+1 and the predicted value of the system's lumped disturbance, i.e., the expression for the discretized superspiral disturbance observer is shown below:

[0072]

[0073] S4: Consider generating a control signal for one-cycle forward prediction delay compensation. Perform a one-step forward prediction on the control signal u to compensate for the delay in the digital control system. Utilize the output information of the superspiral disturbance observer, i.e., the predicted value of the output voltage at time k+1. The predicted value of the system lumped disturbance at time k+1 Calculate the control signal u at time k * :

[0074]

[0075] In formula (3):

[0076] f s Represents the control frequency;

[0077] b0 represents a constant related to the electrical parameters, input voltage, and control frequency of the DAB converter;

[0078] U ref Represents the given voltage;

[0079] I load (k) represents the load current at time k;

[0080] C oRepresents the output capacitor;

[0081] S5: Calculate and constrain the phase shift angle.

[0082] S501: To achieve control of the DAB converter, based on the existing minimum current stress dual phase-shift modulation strategy calculation method, the control signal u at time k is... * The phases are modulated into an inner phase shift angle D1 and an outer phase shift angle D2 and then applied to the DAB converter. Generally speaking, the actual values ​​of the electrical parameters of the DAB converter during operation may deviate from the above measured values.

[0083] S502: To prevent damage to the converter, the following constraints should be applied to the outputs of the inner phase shift angle D1 and the outer phase shift angle D2 respectively to limit the output range and achieve precise and robust voltage control:

[0084]

[0085] This invention is applied to a dual active bridge DC-DC converter topology, employing a dual phase-shift strategy (DPS) for control, i.e., changing the phase shift angle between different arms within the two H-bridges and between two different H-bridges. The disturbance observer of this invention is designed based on the discretized current-voltage equations of the dual active bridge DC-DC converter. The required information is the current current and voltage of the DAB converter, and the converter parameters in the expression use the same values ​​as those in the controller. The disturbance observer is designed using the superhelical principle in second-order sliding mode theory. The output of the disturbance observer is fed into the main controller to achieve disturbance compensation.

[0086] As attached Figure 3 As shown, while the traditional deadbeat-free method offers faster dynamic performance, it suffers from a significant steady-state error in the output voltage under this condition, and this error is positively correlated with the output power, which is detrimental to practical applications. In contrast, the experimental results of this invention completely eliminate this error, exhibiting superior steady-state and dynamic tracking performance, and significantly improving the system's robustness.

[0087] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0088] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for improving the robustness of deadbeat predictive control in a DAB converter, characterized in that: The method includes the following steps: S1: Measure the electrical parameters of the dual active bridge DC-DC converter; S2: Design a superhelical perturbation observer; S3: Discretize the superspiral perturbation observer; S4: Utilizing the output information of the superhelical perturbation observer, i.e., the predicted value of the output voltage at time k+1. The predicted value of the system lumped disturbance at time k+1 Calculate the control signal u at time k * ; In formula (3): f s Represents the control frequency; b0 represents a constant; U ref Represents the given voltage; I load (k) represents the load current at time k; C o Represents the output capacitor; S5: Calculate and constrain the phase shift angle.

2. The method for improving the robustness of deadbeat predictive control in a DAB converter according to claim 1, characterized in that: S2 includes the following steps: S201: Input control signal u, the control signal u being related to the output power P n The relationship is u = 0.5P n ; S202: Obtain the load current I through a current sensor. load ; S203: The output voltage U is obtained by measuring the voltage sensor. o ; S204: Design of a superhelical perturbation observer: In formula (1): The predicted value represents the output voltage; This represents the output voltage error; k1 and k2 represent the observer gain that needs to be adjusted; b0 represents a constant; sgn(*) represents a symbolic function; C o Represents the output capacitor; The predicted value representing the lumped disturbance of the system; t represents time.

3. The method for improving the robustness of deadbeat predictive control in a DAB converter according to claim 2, characterized in that: S3 includes the following steps: S301: Predicted value of output voltage Initial values ​​and predicted values ​​of system lumped disturbances The initial values ​​are all set to 0; S302: Set the control cycle to T and the control frequency to f. s In each control cycle, the output voltage U at time t = k is acquired. o (k) and load current I load (k); S303: Calculate the output voltage at time k+1 and the predicted value of the system's lumped disturbance, i.e., the expression for the discretized superspiral disturbance observer is shown below:

4. The method for improving the robustness of deadbeat predictive control of a DAB converter according to claim 3, characterized in that: S5 includes the following steps: S501: Based on the minimum current stress dual-phase-shift modulation strategy calculation method, the control signal u at time k is... * The phases are modulated into an inner phase shift angle D1 and an outer phase shift angle D2 and then applied to the DAB converter. S502: Apply the following constraints to the outputs of the inner phase shift angle D1 and the outer phase shift angle D2 respectively to achieve precise and robust voltage control:

Images