Model-free predictive control method for dual active bridge converters considering current stress

Abstract

The present invention discloses a model-free predictive control method for a dual active bridge converter considering current stress, which belongs to the field of power electronic control. Firstly, a voltage prediction hyperlocal model of the dual active bridge converter under extended phase-shift modulation is constructed, which effectively reduces the dependence on model parameters. Secondly, an adaptive linear extended state observer is designed, the observer gain is adjusted in real time, and the dynamic part of the voltage prediction hyperlocal model is compensated in advance by adaptively following the observation error, a cost function is constructed, and the functional relationship between the optimal control inner and outer shift ratios (#imgabs0#) is solved online. Finally, a Lagrange multiplier method under extended phase-shift modulation is proposed to solve the optimal inner shift ratio when the optimal current stress is obtained, and the optimal inner and outer shift ratios are obtained, so as to realize the model-free predictive voltage control of the dual active bridge converter, which not only reduces the inductor current stress, but also ensures the excellent dynamics and robustness of the system.

Description

Technical Field

[0001] The invention belongs to the field of power electronic control, and in particular relates to a model-free predictive control method for a dual active bridge converter taking current stress into consideration. Background Art

[0002] In today's industrial field, the dual active bridge (DAB) converter is one of the key topologies in the field of modern power electronics. It adopts a dual active design concept to effectively overcome the unidirectional power flow limitation in traditional power conversion systems. It is suitable for a variety of application scenarios and has a wide range of applications. However, under complex working conditions, the dual active bridge converter is affected by factors such as the strong nonlinear characteristics of the high-frequency isolation transformer, temperature changes in actual work, and sensor accuracy, making it difficult to establish an accurate physical model of the system. The traditional model predictive control strategy uses a large signal modeling method and is more dependent on an accurate system model. However, the large signal reduced-order model control accuracy of the dual active bridge converter is poor, so the control performance of the model predictive control algorithm is reduced when the parameters are mismatched. At the same time, the design of the traditional extended state observer is mainly for feedback compensation. Under nonlinear external interference, it is often limited by the fixed matrix gain value and it is difficult to make timely dynamic adjustments. At the same time, due to the inherent delay in the digital system, the feedback result of the observer will always lag behind the change of external interference, further resulting in limited dynamic performance.

[0003] In addition, the traditional single phase-shift modulation strategy has only one phase-shift angle, which can only achieve energy transmission. When the voltage is mismatched, the return power and current stress of the DAB system increase, and the system transmission efficiency decreases. The use of extended phase-shift modulation can effectively reduce the return power and current stress of DAB and improve the system transmission efficiency under voltage mismatch conditions. However, the extended phase-shift has widened the phase shift in the bridge, with a total of two phase-shift angles. The combination of different phase-shift angles makes the control system more complicated, and it is difficult to select the optimal inner and outer phase-shift angles. Summary of the invention

[0004] The purpose of the present invention is to propose a model-free predictive control method for a dual active bridge converter taking current stress into consideration, which can effectively improve the dynamics and robustness of the DAB system, while reducing current stress, realizing multi-objective optimization control of the DAB converter and significantly improving the control performance of the system under complex working conditions.

[0005] The present invention provides a model-free predictive control method for a dual active bridge converter considering current stress, comprising the following steps:

[0006] Step S1, constructing a dual active bridge converter voltage prediction super-local model under extended phase shift modulation, so as to reduce the dependence of the traditional model predictive control algorithm on the model parameters;

[0007] Step S2, design an adaptive linear extended state observer, fit and predict the voltage error at the next moment by using the recursive least square method with a forgetting factor, adjust the observer gain in real time, adaptively follow the observation error to advance compensate the dynamic part of the voltage prediction super-local model, construct a cost function, solve the functional relationship between the optimal control inner and outer shifts online, and realize model-free predictive control;

[0008] Step S3, design an extended Lagrange multiplier method under phase-shift modulation, solve the optimal inner shift ratio when the optimal current stress is obtained, and substitute it into the functional relationship between the optimal control inner and outer shift ratios obtained by the model-free predictive control algorithm to obtain the optimal inner and outer shift ratios, and use it as the output of the predictive control to achieve multi-objective optimization control of the model-free predictive voltage of the dual active bridge converter.

[0009] The step S1 is specifically:

[0010] Step S11: using the state average modeling method to obtain the first-order average value equivalent model of the reduced-order dual active bridge converter, and establishing the dual active bridge converter voltage model:

[0011] Extended phase shift modulation includes two working modes, namely mode (a) and mode (b) According to the topological structure of the dual active bridge converter and the working waveform of the extended phase-shift modulation, the inductor current values ​​of each node in the two modes are obtained respectively:

[0012]

[0013] In the formula, is the inductor current at time i, k is the voltage transfer ratio, and the transformer ratio is n:1;

[0014] According to the working waveform of the dual active bridge converter under extended phase shift modulation, the first half of the switching cycle is symmetrical with the second half of the switching cycle. The average power of the half cycle is solved, so the transmission power of the dual active bridge converter in one cycle under extended phase shift modulation is for:

[0015]

[0016] in, is the output voltage of the primary side of the converter;

[0017] The maximum transmission power of the dual active bridge converter under extended phase shift modulation is defined as:

[0018] (4)

[0019] Taking the maximum transmission power as the reference value, the transmission power of the two modes under extended phase shift modulation is normalized to obtain the per-unit transmission power of the two modes:

[0020]

[0021] By optimal control shift ratio function relationship D EPS The transmission power in equation (3) is Simplification is performed, including inward shift compared to D1 and outward shift compared to D2. The specific simplification is as follows:

[0022]

[0023] That is, the shift ratio function relationship between the two modes is D EPS for:

[0024]

[0025] According to the law of conservation of power, the equivalent output current of the primary side of the dual active bridge is:

[0026] (8)

[0027] in, U i is the DC voltage at the input terminal, U o is the DC voltage at the output end, L d is the primary auxiliary inductor, R is the output side load resistance, C o is the output side voltage stabilizing capacitor, T s For a switching cycle, D EPS is the shift ratio functional relationship; D 1 is the ratio of the inner shift under the extended phase shift modulation; D 2 is the external shift comparison under extended phase shift modulation;

[0028] Therefore, Kirchhoff's current law is written for node N to establish the voltage model of the dual active bridge converter:

[0029] (9)

[0030] Step S12: constructing a voltage prediction super-local model for the dual active bridge converter:

[0031] The first-order hyperlocal model of the system designed according to the voltage model of the dual active bridge converter is:

[0032]

[0033] Among them, for , represents the controller gain of the system; for , represents the control force input of the system; the output voltage U o The differential value of for , represents the state change of the system, and is the output voltage under EPS modulation U o ;

[0034] Transform the voltage model of the dual active bridge converter into a voltage prediction hyperlocal model:

[0035]

[0036] in, is the controller gain of the system, for ; In order to expand the unknown disturbance terms that are not modeled under phase-shift modulation, To expand the lumped disturbance term under phase-shift modulation, the lumped disturbance term is the dynamic part in the super-local model;

[0037] The forward Euler method is used to discretize equation (11) to obtain the dual active bridge converter under extended phase shift modulation: Hyperlocal model for output voltage prediction with time discretization:

[0038] (12)

[0039] in, for k Output voltage sampling value at the moment, for Output voltage prediction value at the moment; for k The aggregate disturbance at time, for k The shift phase at the moment is a function of for k System controller gain at time t;

[0040] Step S13: In order to compensate for the delay in the calculation of the digital processor, a two-step prediction method is used to design a dual active bridge converter. The discrete output voltage prediction hyperlocal model at time instant:

[0041] (13)

[0042] in, for Output voltage sampling value at the moment, for Output voltage prediction value at the moment; for The aggregate disturbance at time, for The shift phase at the moment is a function of for The system controller gain at time t.

[0043] The step S2 is specifically:

[0044] Step S21, the equation of the recursive least square method including the forgetting factor is expressed as:

[0045]

[0046] in, Input and output information vector for the system; is the parameter vector of the system; For the forgetting factor; is the correction term, ; P (*) is the covariance matrix, and its initial value is , is a second-order identity matrix, p =10 5 ; Indicates voltage observation value and the actual voltage The error between

[0047] Designing observer adaptive reinforcement factor for:

[0048]

[0049] An adaptive linear extended state observer under extended phase-shift modulation is designed to perform advance compensation for the dynamic part in the hyperlocal model. The system state equation in formula (10) is:

[0050]

[0051] in, ;

[0052] Step S22: Design the adaptive linear extended state observer as follows:

[0053]

[0054] in, is the observation value of the observer under EPS modulation; , is the gain vector of the observer under EPS modulation; is the observer adaptive enhancement factor under EPS modulation;

[0055] Note the output voltage differential value for , represents the differential value of the system state quantity, , The state quantities The observed values ​​and the observed differential values;

[0056] In order to make the observer stable, the roots of the characteristic equation should all fall on the negative half axis, and the gain coefficient of the observer is designed as:

[0057]

[0058] in, is the bandwidth of the adaptive linear extended state observer;

[0059] According to the Euler forward method and the two-step prediction design method, equation (17) is discretized as:

[0060]

[0061] in, ;

[0062] After simplifying equation (19), the output voltage prediction equation after adaptive linear extended state observer compensation under converter EPS modulation is obtained as follows:

[0063]

[0064] The total disturbance observation value of the system is calculated , to replace the actual value , thereby compensating the voltage prediction hyperlocal model in advance;

[0065] Step S23: In order to achieve the stability of the DC side output voltage of the dual active bridge converter, according to formula (13): The corresponding cost function for constructing the super-local model for discrete output voltage prediction at each moment is:

[0066] (twenty one)

[0067] Where g represents the cost function, is the output voltage reference value;

[0068] The system lumped disturbance observation value can be obtained by formula (20): , which is used to replace the actual value in formula (13) , so as to pre-compensate the voltage prediction super-local model in equation (13), and then obtain the adaptive gain setting calculation under EPS modulation by minimizing the cost function of equation (21): The functional relationship of the optimal control shift ratio at the moment is:

[0069] (twenty two)

[0070] in, The equation in formula (7) still applies, namely:

[0071]

[0072] The step S3 is specifically:

[0073] Step S31, designing a Lagrangian current stress optimization method under extended phase shift modulation:

[0074] The normalized current stress in these two working modes is:

[0075]

[0076] Among them, the voltage transfer ratio , n represents the transformation ratio of the transformer;

[0077] The Lagrangian current stress optimization method under the extended phase shift modulation is designed as follows:

[0078] (25)

[0079] In the formula, Lagrange multipliers designed for solving the optimal current stress optimization problem under EPS modulation, For a given reference transmission power, is the normalized transmission power, is the normalized current stress;

[0080] In order to solve the minimum current stress, the partial derivative of equation (25) is calculated:

[0081]

[0082] Combining equations (24), (25) and (26), the normalized current stress of the two modes in equation (24) is respectively substituted into equation (25) and the partial derivative of equation (25) is obtained through equation (26), and the working mode (a) can be obtained. and working mode (b) The optimal shift ratio ( ) are combined into:

[0083]

[0084] Therefore, the optimal inward shift is for:

[0085]

[0086] Step S32, combining the optimal inner and outer control shift ratio function relationship obtained by the adaptive linear extended state observer and the cost function design in step S2 and the optimal inner shift ratio obtained in step S31, solving the optimal inner and outer shift ratios, thereby realizing closed-loop control and optimizing current stress:

[0087]

[0088] In the formula, .

[0089] A model-free predictive control system for a dual active bridge converter taking current stress into consideration comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the model-free predictive control method for the dual active bridge converter taking current stress into consideration is implemented.

[0090] A computer storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned model-free predictive control method for a dual active bridge converter considering current stress.

[0091] By adopting the technical solution of the present invention, firstly, a voltage prediction hyperlocal model of a dual active bridge converter under expanded phase shift (EPS) modulation is constructed to replace the traditional a priori physical model, which effectively reduces the dependence of the traditional model predictive control algorithm on model parameters; secondly, an adaptive linear extended state observer is designed, and the recursive least square method with a forgetting factor is used to fit and predict the voltage error at the next moment, and the observer gain is adjusted in real time. The dynamic part of the voltage prediction hyperlocal model is compensated in advance by adaptively following the observation error, and the cost function is constructed to solve the optimal control inner and outer shift ratios online ( ); finally, an extended Lagrange multiplier method under phase-shift modulation is proposed to solve the optimal inner shift ratio when the optimal current stress is achieved, and the optimal inner and outer shift ratios are substituted into the online functional relationship of solving the optimal control inner and outer shift ratios to obtain the optimal inner and outer shift ratios, thus realizing model-free predictive voltage control of the dual active bridge converter, which not only reduces the inductor current stress, but also ensures the excellent dynamics and robustness of the system.

[0092] The present invention proposes an ALESO-MFPVC control strategy for the DAB conversion system, realizes accurate voltage prediction control, and designs an adaptive linear extended state observer to dynamically adjust the fixed gain matrix of the linear extended state observer, eliminates feedback delay and inherent delay in the digital circuit, and further improves the dynamics and robustness of the system; and uses EPS modulation as the control basis, proposes to use the LMM optimization method to solve the optimal inner shift phase ratio of the DAB converter, and combines the optimal shift phase ratio function relationship obtained by the ALESO-MFPVC algorithm to solve the optimal inner and outer shift phase ratios, thereby realizing multi-objective optimization control of the DAB converter.

[0093] The present invention focuses on solving the problems of complex construction of accurate models of dual active bridge converters under complex working conditions, the fixed value of the gain matrix of the traditional linear extended state observer that cannot adapt to the changes in external wide-range interference, and the increase of system power reflux and current stress under voltage mismatch conditions, and the decrease in transmission efficiency. By constructing a voltage prediction super-local model and designing an adaptive linear extended state observer to estimate the lumped disturbance and perform advance compensation, the present invention does not need to rely on accurate models, significantly reduces the parameter dependence of the system, and improves the system's anti-disturbance ability and robustness; by combining the Lagrangian current stress optimization method with the model-free predictive control method under extended phase-shift modulation, the functional relationship between the optimal control internal and external phase shift phases is solved, thereby achieving multi-objective optimization control with high dynamics, strong robustness and low current stress.

[0094] Compared with the traditional PI combined with current stress optimization method and traditional model-free predictive control, the present invention, on the one hand, replaces the traditional extended state observer with an adaptive linear extended state observer and combines it with model-free predictive control, which further enhances the system's anti-disturbance capability and improves the system's dynamics and robustness; on the other hand, it takes current stress into consideration and proposes a multi-objective optimization predictive control, which ensures excellent dynamics and robustness while optimizing current stress, thereby meeting performance requirements in different application scenarios.

[0095] The present invention has the following beneficial effects:

[0096] (1) Under the condition of linear parameter changes, the linear extended state observer can effectively observe the lumped disturbance. The existing model-free predictive control strategy combined with the linear extended state observer method can effectively reduce the parameter dependence of the dual active bridge converter and improve the system robustness. However, in the actual operation process, external disturbances such as temperature rise and mechanical vibration may cause nonlinear changes in system parameters such as transformers and inductors. The traditional linear extended state observer is feedback compensation. The gain matrix in the linear extended state observer is a fixed value. Under nonlinear external interference, it is often limited by the fixed matrix gain value and difficult to make timely dynamic adjustments. It cannot adapt to all working conditions. At the same time, the inherent delay of the digital circuit will cause the observer feedback to always have a one-beat delay, and the estimated compensation value always lags behind the external interference changes, further resulting in limited dynamic performance. In order to solve the problem that the fixed-value gain matrix of the traditional linear extended state observer cannot be adaptively adjusted following the observation error under nonlinear interference, the control effect always lags behind the feedback result of the observation error, and the dynamic performance is limited, the present invention designs an adaptive linear extended state observer under extended phase-shift modulation. By adopting the recursive least squares method with forgetting factor to fit and predict the voltage error at the next moment, and adjusting the observer gain in real time, the delay and hysteresis of the feedback compensation of the traditional linear extended state observer can be effectively avoided, and the dynamics and robustness of the system can be further improved.

[0097] (2) When there is a voltage mismatch in traditional single phase-shift modulation, the system's return power and current stress increase, and the system transmission efficiency decreases. Therefore, efficiency optimization needs to be taken into consideration. Based on the expansion of phase-shift modulation, the present invention designs a multi-objective optimization method based on the combination of Lagrange multiplier method and model-free predictive control, which effectively reduces the inductor current stress of the converter when there is a voltage mismatch, while ensuring good dynamics and robustness of the system.

[0098] The present invention has significant advantages in the following aspects and has broad future development prospects:

[0099] 1) Efficient and stable power control: Efficient and stable voltage control is achieved without relying on an accurate model. The designed adaptive linear extended state observer has excellent anti-interference ability, dynamics and robustness. This enables the system to maintain stable operation in the face of complex and changeable working conditions, improving the reliability and durability of the system.

[0100] 2) Flexible adaptation to different application scenarios: The model-free predictive control method considering current stress optimization not only reduces the dependence on accurate model parameters and adapts to complex and changeable working conditions, but also can effectively reduce the current stress of the converter under voltage mismatch conditions, reduce the difficulty of switch tube selection, improve system efficiency, and better meet the different performance requirements in different application scenarios. This feature makes it have a wide range of application potential in different application scenarios.

[0101] 3) Clean energy transformation: As the world pays more attention to and demands for renewable energy, the performance of the dual active bridge converter, as the core component of the wind, solar and energy storage microgrid system, directly affects the energy utilization efficiency and the quality of the power system, providing technical support for the development of renewable energy, promoting energy transformation and reducing carbon emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] Figure 1 is a control logic diagram of the present invention;

[0103] Figure 2 This is a structural diagram of the adaptive linear extended state observer of the present invention;

[0104] Figure 3 It is a topological structure diagram of a dual active bridge in the present invention;

[0105] Figure 4 To expand the phase shift modulation in the present invention in working mode (a) Working waveform diagram;

[0106] Figure 5 To expand the phase shift modulation in the present invention in working mode (b) Working waveform diagram;

[0107] Figure 6 It is the average value equivalent model of the dual active bridge converter after order reduction in the present invention. DETAILED DESCRIPTION

[0108] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0109] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0110] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or scheme described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or schemes. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0111] Embodiment 1

[0112] Embodiment 1 of the present invention, as Figure 1 As shown, a model-free predictive control method for a dual active bridge converter considering current stress is provided, comprising the following steps:

[0113] Step S1, constructing a dual active bridge converter voltage prediction super-local model under extended phase shift modulation to replace the traditional priori physical model, so as to reduce the dependence of the traditional model predictive control algorithm on model parameters;

[0114] Step S11: using the state average modeling method to obtain the first-order average value equivalent model of the reduced-order dual active bridge converter, and establishing the DAB converter voltage model:

[0115] Figure 3 It is a dual active bridge converter topology. ~ It is the primary side full-bridge power switching device, ~ It is the secondary side full-bridge power switch device, ~ All use fully controlled power devices to better achieve soft switching and bidirectional power flow. and are the DC voltages at the input and output of the converter respectively; and They are the AC square wave voltages of the full-bridge nodes on the primary and secondary sides of the transformer respectively; It is the input side support capacitor of the converter; is the output side filter capacitor; It is the auxiliary inductance of the transformer; is the output side load resistance; is the inductor current; the transformer ratio is ; is the converter voltage transfer ratio.

[0116] Figure 4 and Figure 5 These are the two working mode waveforms of the dual active bridge converter under extended phase shift modulation. Compared with the inward shift, Compared with outward migration.

[0117] Extended phase shift modulation includes two working modes, namely mode (a) and mode (b) According to the topological structure of the dual active bridge converter and the working waveform of the extended phase-shift modulation, the inductor current values ​​of each node in the two modes are obtained respectively:

[0118]

[0119] In the formula, is the inductor current at time i, k is the voltage transfer ratio, and the transformer ratio is n:1;

[0120] According to the working waveform of the dual active bridge converter under extended phase shift modulation, the first half of the switching cycle is symmetrical with the second half of the switching cycle, so only the average power of half a cycle needs to be solved. Therefore, the transmission power of the dual active bridge converter in one cycle under extended phase shift modulation is for:

[0121]

[0122] in, is the output voltage of the primary side of the converter;

[0123] The maximum transmission power of the dual active bridge converter under extended phase shift modulation is defined as:

[0124] (4)

[0125] Taking the maximum transmission power as the reference value, the transmission power of the two modes under extended phase shift modulation is normalized to obtain the per-unit transmission power of the two modes:

[0126]

[0127] By optimal control shift ratio function relationship D EPS The transmission power in equation (3) is Simplification is performed, including inward shift compared to D1 and outward shift compared to D2. The specific simplification is as follows:

[0128]

[0129] That is, the shift ratio function relationship between the two modes is D EPS for:

[0130]

[0131] like Figure 6As shown in the figure, it is the first-order average value equivalent model of the dual active bridge converter. As can be seen from the figure, the primary and secondary sides of the high-frequency isolation transformer can be equivalent to two controlled current sources. , , They represent the primary equivalent output current, load current and the current flowing through the output capacitor respectively.

[0132] According to the law of conservation of power, the equivalent output current of the primary side of the dual active bridge is:

[0133] (8)

[0134] in, U i is the DC voltage at the input terminal, U o is the DC voltage at the output end, L d is the primary auxiliary inductor, R is the output side load resistance, C o is the output side voltage stabilizing capacitor, T s For a switching cycle, D EPS is the shift ratio functional relationship; D 1 is the ratio of the inner shift under the extended phase shift modulation; D 2 is the external shift comparison under extended phase shift modulation;

[0135] So Figure 5 Write Kirchhoff's current law in the middle node N column to establish the voltage model of the dual active bridge converter:

[0136] (9)

[0137] Step S12: constructing a voltage prediction super-local model for the dual active bridge converter, as follows:

[0138] The first-order hyperlocal model of the system designed according to the voltage model of the dual active bridge converter is:

[0139]

[0140] Among them, for , represents the controller gain of the system; for , represents the control force input of the system; the output voltage U o The differential value of for , represents the state change of the system, and is the output voltage under EPS modulation U o ;

[0141] Transform the voltage model of the dual active bridge converter into a voltage prediction hyperlocal model:

[0142]

[0143] in, is the controller gain of the system, for ; In order to expand the unknown disturbance terms that are not modeled under phase-shift modulation, To expand the lumped disturbance term under phase-shift modulation, the lumped disturbance term is the dynamic part in the super-local model;

[0144] The forward Euler method is used to discretize equation (11) to obtain the dual active bridge converter under extended phase shift modulation: Hyperlocal model for output voltage prediction with time discretization:

[0145] (12)

[0146] in, for k Output voltage sampling value at the moment, for Output voltage prediction value at the moment; for k The aggregate disturbance at time, for k The shift phase at the moment is a function of for k System controller gain at time t;

[0147] Step S13: In order to compensate for the delay in the calculation of the digital processor, a two-step prediction method is used to design a dual active bridge converter. The discrete output voltage prediction hyperlocal model at time instant:

[0148] (13)

[0149] in, for Output voltage sampling value at the moment, for Output voltage prediction value at the moment; for The aggregate disturbance at time, for The shift phase at the moment is a function of for System controller gain at time t;

[0150] Step S2, design an adaptive linear extended state observer, fit and predict the voltage error at the next moment by using the recursive least square method with forgetting factor, adjust the observer gain in real time, adaptively follow the observation error to compensate the dynamic part of the voltage prediction super-local model in advance, construct the cost function, and solve the optimal control inner and outer shift comparison online ( ) to achieve model-free predictive control:

[0151] Step S21, the equation of the recursive least square method including the forgetting factor is expressed as:

[0152]

[0153] in, Input and output information vector for the system; is the parameter vector of the system; is the forgetting factor, which is between 0.95 and 1; is the correction term, ; P (*) is the covariance matrix, and its initial value is , is a second-order identity matrix, p =10 5 ; Indicates voltage observation value and the actual voltage The error between

[0154] Designing observer adaptive reinforcement factor for:

[0155] ;

[0156] An adaptive linear extended state observer under extended phase-shift modulation is designed to perform advance compensation for the dynamic part in the hyperlocal model. The system state equation in equation (10) can be written as:

[0157]

[0158] in, ;

[0159] Step S22: Figure 2 As shown, the adaptive linear extended state observer is designed as:

[0160]

[0161] in, is the observation value of the observer under EPS modulation; , is the gain vector of the observer under EPS modulation, is the observer adaptive enhancement factor under EPS modulation;

[0162] The output voltage differential value is recorded in the present invention for , represents the differential value of the system state quantity, , The state quantities The observed values ​​and the observed differential values;

[0163] In order to make the observer stable, the roots of the characteristic equation should all fall on the negative half axis, and the gain coefficient of the observer is designed as:

[0164]

[0165] in, is the bandwidth of the adaptive linear extended state observer;

[0166] According to the Euler forward method and the two-step prediction design method, equation (17) is discretized as:

[0167]

[0168] in, ;

[0169] After simplifying equation (19), the output voltage prediction equation after adaptive linear extended state observer compensation under converter EPS modulation is obtained as follows:

[0170]

[0171] The total disturbance observation value of the system is calculated , to replace the actual value , thereby compensating the voltage prediction hyperlocal model in advance;

[0172] Step S23: In order to achieve the stability of the DC side output voltage of the dual active bridge converter, according to formula (13): The corresponding cost function for constructing the super-local model for discrete output voltage prediction at each moment is:

[0173] (twenty one)

[0174] Where g represents the cost function, is the output voltage reference value;

[0175] The system lumped disturbance observation value can be obtained by formula (20): , which is used to replace the actual value in formula (13) , so as to pre-compensate the voltage prediction super-local model in equation (13), and then obtain the adaptive gain setting calculation under EPS modulation by minimizing the cost function of equation (21): The functional relationship of the optimal control shift ratio at the moment is:

[0176] (twenty two)

[0177] in, The equation in formula (7) still applies, namely:

[0178]

[0179] Step S3, design an extended Lagrange multiplier method under phase shift modulation, solve the optimal inner shift ratio when the optimal current stress is obtained, and substitute it into the functional relationship between the optimal control inner and outer shift ratios obtained by the model-free predictive control algorithm to obtain the optimal inner and outer shift ratios, which are used as the output of the predictive control to achieve the multi-objective optimization control of the model-free predictive voltage of the dual active bridge converter:

[0180] Step S31, designing a Lagrangian current stress optimization method under extended phase shift modulation:

[0181] The normalized current stress in these two working modes is:

[0182]

[0183] Among them, the voltage transfer ratio , n represents the transformation ratio of the transformer;

[0184] For a given transmission power condition, find the optimal internal shift ratio Compared with outward migration The inductor current stress This type of problem can be summarized as the problem of solving the optimal solution under the equation constraint. As a commonly used optimization algorithm for solving equation constraints, the Lagrange multiplier method is more suitable for solving the minimum current stress under extended phase shift modulation, and compared with the traditional derivation method, it is faster and less computationally intensive.

[0185] The Lagrangian current stress optimization method under the extended phase shift modulation is designed as follows:

[0186] (25)

[0187] In the formula, Lagrange multipliers designed for solving the optimal current stress optimization problem under EPS modulation, For a given reference transmission power, is the normalized transmission power, is the normalized current stress;

[0188] In order to solve the minimum current stress, the partial derivative of equation (25) is calculated:

[0189]

[0190] Combining equations (24), (25) and (26), the normalized current stress of the two modes in equation (24) is respectively substituted into equation (25) and the partial derivative of equation (25) is obtained through equation (26), and the working mode (a) can be obtained. and working mode (b) The optimal shift ratio ( ) are combined into:

[0191]

[0192] Therefore, the optimal inward shift is for:

[0193]

[0194] Step S32, combining the optimal inner and outer control shift ratio function relationship obtained by the adaptive linear extended state observer and the cost function design in step S2 and the optimal inner shift ratio obtained in step S31, solving the optimal inner and outer shift ratios, thereby realizing closed-loop control and optimizing current stress:

[0195]

[0196] In the formula, .

[0197] The dual active bridge converter of the present invention may be a high-frequency isolated DC / DC converter.

[0198] Embodiment 2

[0199] The professionals should further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented by electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0200] Specifically, each step of the method embodiment in the embodiment of the present application can be completed by the hardware integrated logic circuit and / or software instructions in the processor, and the steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor, or a combination of hardware and software modules in the decoding processor. Optionally, the software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps in the above method embodiment in combination with its hardware.

[0201] Embodiment 2 of the present invention provides a dual active bridge converter model-free predictive control system considering current stress, the system may include: a memory and a processor, the memory is used to store a computer program and transmit the program code to the processor. In other words, the processor can call and run the computer program from the memory to implement the method of embodiment 1 of the present application. For example, the processor can be used to execute the method of embodiment 1 according to the instructions in the computer program.

[0202] In some embodiments of the present application, the processor may include but is not limited to:

[0203] General-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.

[0204] In some embodiments of the present application, the memory includes, but is not limited to, volatile memory and / or non-volatile memory. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (Direct Rambus RAM, DR RAM).

[0205] In some embodiments of the present application, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the method provided by the present application. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the system.

[0206] Embodiment 3

[0207] Embodiment 3 of the present invention further provides a computer storage medium on which a computer program is stored. When the computer program is executed by a computer, the computer is enabled to execute the method of embodiment 1 above.

[0208] When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the first embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integration. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a solid state disk (solid state disk, SSD)), etc.

[0209] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A model-free predictive control method for a dual active bridge converter considering current stress, characterized in that The following steps are involved: Step S1, constructing a dual active bridge converter voltage prediction super-local model under extended phase shift modulation, so as to reduce the dependence of the traditional model predictive control algorithm on the model parameters; Step S2, design an adaptive linear extended state observer, fit and predict the voltage error at the next moment by using the recursive least square method with a forgetting factor, adjust the observer gain in real time, adaptively follow the observation error to advance compensate the dynamic part of the voltage prediction super-local model, construct a cost function, solve the functional relationship between the optimal control inner and outer shifts online, and realize model-free predictive control; Step S3, designing an extended Lagrange multiplier method under phase shift modulation, solving the optimal inner shift ratio when the optimal current stress is obtained, and substituting it into the functional relationship between the optimal control inner and outer shift ratios obtained by the model-free predictive control algorithm, obtaining the optimal inner and outer shift ratios, and using it as the output of the predictive control to achieve multi-objective optimization control of the model-free predictive voltage of the dual active bridge converter; The step S1 is specifically: Step S11: using the state average modeling method to obtain the first-order average value equivalent model of the reduced-order dual active bridge converter, and establishing the dual active bridge converter voltage model: Extended phase shift modulation includes two working modes, namely mode (a) and mode (b) ; According to the topology of the dual active bridge converter and the working waveform of the extended phase-shift modulation, the inductor current values ​​of each node in the two modes are obtained respectively: In the formula, is the inductor current at time i, k is the voltage transfer ratio, and the transformer ratio is n:1; According to the working waveform of the dual active bridge converter under extended phase shift modulation, the first half of the switching cycle is symmetrical with the second half of the switching cycle. The average power of the half cycle is solved, so the transmission power of the dual active bridge converter in one cycle under extended phase shift modulation is for: in, is the output voltage of the primary side of the converter; The maximum transmission power of the dual active bridge converter under extended phase shift modulation is defined as: （4） Taking the maximum transmission power as the reference value, the transmission power of the two modes under extended phase shift modulation is normalized to obtain the per-unit transmission power of the two modes: By optimal control shift ratio function relationship D EPS The transmission power in equation (3) is Simplification is performed, including inward shift compared to D1 and outward shift compared to D2. The specific simplification is as follows: That is, the shift ratio function relationship between the two modes is D EPS for: According to the law of conservation of power, the equivalent output current of the primary side of the dual active bridge is: （8） in, U i is the DC voltage at the input terminal, U o is the DC voltage at the output end, L d is the primary auxiliary inductor, R is the output side load resistance, C o is the output side voltage stabilizing capacitor, T s For a switching cycle, D EPS is the shift ratio functional relationship; D 1 is the ratio of the inner shift under the extended phase shift modulation; D 2 is the external shift comparison under extended phase shift modulation; Therefore, Kirchhoff's current law is written for node N to establish the voltage model of the dual active bridge converter: （9） Step S12: constructing a voltage prediction super-local model for the dual active bridge converter: The first-order hyperlocal model of the system designed according to the voltage model of the dual active bridge converter is: Among them, for , represents the controller gain of the system; for , represents the control force input of the system; the output voltage U o The differential value of for , represents the state change of the system, and is the output voltage under EPS modulation U o ; Transform the voltage model of the dual active bridge converter into a voltage prediction hyperlocal model: in, is the controller gain of the system, for ; In order to expand the unknown disturbance terms that are not modeled under phase-shift modulation, To expand the lumped disturbance term under phase-shift modulation, the lumped disturbance term is the dynamic part in the super-local model; The forward Euler method is used to discretize equation (11) to obtain the dual active bridge converter under extended phase shift modulation: Hyperlocal model for output voltage prediction with time discretization: （12） in, for k Output voltage sampling value at the moment, for Output voltage prediction value at the moment; for k The aggregate disturbance at time, for k The shift phase at the moment is a function of for k System controller gain at time t; Step S13: In order to compensate for the delay in the calculation of the digital processor, a two-step prediction method is used to design a dual active bridge converter. The discrete output voltage prediction hyperlocal model at time instant: （13） in, for Output voltage sampling value at the moment, for Output voltage prediction value at the moment; for The aggregate disturbance at time, for The shift phase at the moment is a function of for System controller gain at time t; The step S2 is specifically: Step S21, the equation of the recursive least square method including the forgetting factor is expressed as: in, Input and output information vector for the system; is the parameter vector of the system; For the forgetting factor; is the correction term, ; P (*) is the covariance matrix, and its initial value is , is a second-order identity matrix, p =10 5 ; Indicates voltage observation value and the actual voltage The error between Designing observer adaptive reinforcement factor for: An adaptive linear extended state observer under extended phase-shift modulation is designed to perform advance compensation for the dynamic part in the hyperlocal model. The system state equation in formula (10) is: in, ; Step S22: Design the adaptive linear extended state observer as follows: in, is the observation value of the observer under EPS modulation; , is the gain vector of the observer under EPS modulation; is the observer adaptive enhancement factor under EPS modulation; Note the output voltage differential value for , represents the differential value of the system state quantity, , The state quantities The observed values ​​and the observed differential values; In order to make the observer stable, the roots of the characteristic equation should all fall on the negative half axis, and the gain coefficient of the observer is designed as: in, is the bandwidth of the adaptive linear extended state observer; According to the Euler forward method and the two-step prediction design method, equation (17) is discretized as: in, ; After simplifying equation (19), the output voltage prediction equation after adaptive linear extended state observer compensation under converter EPS modulation is obtained as follows: The total disturbance observation value of the system is calculated , to replace the actual value , thereby compensating the voltage prediction hyperlocal model in advance; Step S23: In order to achieve the stability of the DC side output voltage of the dual active bridge converter, according to formula (13): The corresponding cost function for constructing the super-local model for discrete output voltage prediction at each moment is: （21） Where g represents the cost function, is the output voltage reference value; The system lumped disturbance observation value can be obtained by formula (20): , which is used to replace the actual value in formula (13) , so as to pre-compensate the voltage prediction super-local model in equation (13), and then obtain the adaptive gain setting calculation under EPS modulation by minimizing the cost function of equation (21): The functional relationship of the optimal control shift ratio at the moment is: （22） in, The equation in formula (7) still applies, namely: （23）； The step S3 is specifically: Step S31, designing a Lagrangian current stress optimization method under extended phase shift modulation: The normalized current stress in these two working modes is: Among them, the voltage transfer ratio , n represents the transformation ratio of the transformer; The Lagrangian current stress optimization method under the extended phase shift modulation is designed as follows: （25） In the formula, Lagrange multipliers designed for solving the optimal current stress optimization problem under EPS modulation, For a given reference transmission power, is the normalized transmission power, is the normalized current stress; In order to solve the minimum current stress, the partial derivative of equation (25) is calculated: Combining equations (24), (25) and (26), the normalized current stress of the two modes in equation (24) is respectively substituted into equation (25) and the partial derivative of equation (25) is obtained through equation (26), and the working mode (a) can be obtained. and working mode (b) The optimal shift ratio ( ) are combined into: Therefore, the optimal inward shift is for: Step S32, combining the optimal inner and outer control shift ratio function relationship obtained by the adaptive linear extended state observer and the cost function design in step S2 and the optimal inner shift ratio obtained in step S31, solving the optimal inner and outer shift ratios, thereby realizing closed-loop control and optimizing current stress: In the formula, .

2. A model-free predictive control system for a dual active bridge converter considering current stress, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the model-free predictive control method for a dual active bridge converter considering current stress as claimed in claim 1 is implemented.

3. A computer storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the model-free predictive control method for a dual active bridge converter considering current stress as claimed in claim 1 is implemented.

Images