Method for controlling a reversible DC-DC converter.

Abstract

A method (1) for controlling a reversible DC-DC LLC converter adapted for charging and discharging a battery, and when said converter is operating in discharge mode, said method (1) implements a verification step (41, 42) of the converter's operating point during which it is verified whether the converter is operating in a frequency saturation zone and whether the converter is operating in a predefined operating zone. If the converter is operating in a frequency saturation zone and if the converter is operating in a predefined operating zone, the method implements pulse-width modulation control; otherwise, the method implements pulse-frequency modulation control. Figure 4

Description

Title of the invention: Method for controlling a reversible DC-DC converter.

[0001] The present invention relates to a DC-DC converter for a bidirectional electric storage battery charger and its control method.

[0002] A conventional electric accumulator battery charger is unidirectional in the sense that it only allows the electric accumulators to be recharged from an external electrical supply network, which is commonly called the charging direction or forward direction.

[0003] Such a unidirectional electric accumulator battery charger generally comprises a power factor correction stage, also known by its English abbreviation PFC, and a direct-direct conversion stage, more generally called a DC-DC stage.

[0004] However, it is useful for accumulator chargers to also be able to deliver the accumulated electricity to an external electrical network, as a current source, or replace a network and operate as a voltage source to which loads are connected; we then speak of bidirectional chargers and in the context of motor vehicles used V2X, for Vehicle to everything. The supply of current by the electric accumulator battery to the external network is called the discharge direction, or even indirect direction.

[0005] Bidirectional chargers are known in particular, such as that of document FR3014260 A1, which describes a resonant DCDC charger of the LC series assembly type. However, such an assembly does not allow the type of energy conversion to be changed because its gain is always less than 1.

[0006] A bidirectional (or reversible) charger for high power density applications is also known, as shown in [Fig.l] of the prior art, which implements a full-bridge resonant LLC type DC-DC converter.

[0007] A full-bridge resonant LLC converter 10 according to [Fig.l], comprises a switching full bridge 11 generating a square signal or current exciting an LLC circuit 12, composed of a series capacitor Cr and two inductors, a series inductor Lr and an inductor Lm in parallel with the primary winding of a transformer 13. The LLC circuit 12 then produces a resonant sinusoidal current in the transformer 13 which is rectified by the rectifier bridge 14, then transmitted to the battery 16. The voltage referenced VBArest equal to the voltage across the battery 16. The converter 10 is connected at the input (in the charging mode) to a DC bus of voltage Vdc represented by a capacitor in [Fig.l]. In other words, in charge mode, the complete bridge 11 is powered by this continuous bus of voltage Vdc. The complete bridge 11 comprises four bidirectional switching cells SI to S4, the rectifier bridge 14 (in charge mode) is also a complete bridge comprising four bidirectional switching cells S5 to S8.

[0008] Generally in reversible chargers known from the prior art, the regulation frequencies of the DCDC, in charge mode and in discharge mode, are limited between substantially 60kHz and 200kHz. However, these values ​​are given as an example and depend in particular on the values ​​of the components.

[0009] However, in discharge mode, at high VBat battery voltage and low power, the regulation towards the desired gain, Vdc / VBAT which is generally less than 0.9, causes a divergence of the DCDC regulation frequency, towards switching frequencies higher than 200 kHz. This causes a sharp drop in the charger's efficiency in discharge mode.

[0010] The control of a DCDC LLC converter, the structure of which is shown [Fig.l], is usually based on switching frequency control strategies.

[0011] In particular, document FR3083929 discloses a command consisting of calculating the cutting frequency by inversion of the gain starting from an equivalent model based on the approach of the first harmonic.

[0012] However, this strategy has a fairly significant saturation zone of the switching frequency at 200kHz, which causes a limitation of the regulation of the DC bus voltage.

[0013] The control frequency range, which essentially depends on the elements of the resonant body, the power required, the value of the voltage Vdc of the DC bus capacitors and the battery voltage VBAt, is then sometimes not sufficient to reach all the operating points.

[0014] The literature includes many “classical” control laws, such as sliding mode control and the Proportional-Integral-Derivative regulator known as the PID regulator.

[0015] These control laws are often proposed on the basis of a large or small signal model with strategies based either on:

[0016] - Frequency variation, also called PFM from the English Puise Frequency Modulation, or

[0017] - The variation of the duty cycle of the MOSFET transistor control signals, called PWM from the English Pulse Width Modulation.

[0018] In the literature, the strategies proposed for reversible converters have battery voltage regulation, where the battery voltage varies over a reduced range, for example between 12 and 60 V.

[0019] We know in particular the document FR3096847 which presents a structure of the reversible DCDC-LLC converter with the addition of an additional switched inductance in order to make a DCDC converter symmetrical in both load directions.

[0020] However, a switched inductance generates an undesirable manufacturing cost and increase in size.

[0021] To control this switched inductance structure, this document also discloses a control method implementing regulation of the continuous bus, called DC Bus, known as Burst mode by hysteresis on the frequency and continuous bus.

[0022] However, the control solution proposed by this prior art has limits in terms of efficiency and performance, particularly in V2X operation.

[0023] Also, there is a need for an improved control method that does not require modification of the technical structure of the DCDC LLC converter.

[0024] To this end, a method is proposed for controlling a reversible DC-DC LLC converter, comprising a DC bus, a primary H-bridge and a secondary H-bridge of four MOSFET transistors each, connected to an electric accumulator battery of an electric motor vehicle, said DC-DC converter being adapted to operate in charging and discharging of said battery, and when said converter operates in discharging, a step of verifying the operating point of the converter is implemented during which it is verified whether the converter operates in a frequency saturation zone and whether the converter operates in a predefined operating zone.

[0025] In this application, "secondary" means the part of the converter located between the transformer and the battery, and "primary" means the part of the converter located between the transformer and the DC bus supplying the battery in charging mode.

[0026] If the converter operates in a frequency saturation zone and if the converter operates in a predefined operating zone, the method implements a control step by pulse width modulation depending on the battery voltage, a reference voltage of the DC bus, the power of the converter and a switching frequency of the converter,

[0027] otherwise the method implements a control step by modulation of the pulse frequency as a function of the battery voltage, a reference voltage of the DC bus, and the power of the converter.

[0028] Thus, it is possible to ensure that optimal operation of the converter is obtained even in areas which are usual saturation areas for the pulse frequency modulation solution known from the prior art.

[0029] Advantageously and in a non-limiting manner, said regulation of the control comprises an inversion of the gain of the converter.

[0030] The gain G is by definition a function of the system parameters and the frequency. Gain inversion means calculating the frequency as a function of the gain and the system parameters.

[0031] Advantageously and in a non-limiting manner, said control step comprises a Proportional-Integral regulation, a function of the DC bus voltage and a DC bus reference voltage. Thus, optimal regulation of the control can be ensured regardless of the strategy employed.

[0032] Advantageously and in a non-limiting manner, said frequency saturation zone corresponds to an operating point of the converter, as a function of the battery voltage and the power of the converter, in which the switching frequency saturates at a maximum value.

[0033] Advantageously and in a non-limiting manner, said predefined operating zone is defined by pairs of battery voltage (VBat) and converter power (P) where:

[0034] = (Vb^ P) such that ; 340 V < < 350 V and 1800 W <P< 3000 W ou 350 V < Vbat < 370 V and 2800 W <P< 3300 W ou 370 V < < 390 V et 3700 W < P < 4300 W ou 390 V < Vhat < 410 V and 4700 W < P < 5300 W or 410 V< Vb <a<430 Vet 5000 W<P<6000 W

[0035] The invention also relates to a device for controlling a reversible DC-DC converter comprising an H-bridge at the primary and an H-bridge at the secondary of four MOSFET transistors each connected to an electric motor vehicle battery, said DC-DC converter being adapted to operate in charging and discharging of said battery, and

[0036] Said device comprising means for implementing, when said converter is operating in discharge, a step of verifying the operating point of the converter during which it is verified whether the converter is operating in a frequency saturation zone and whether the converter is operating in a predefined operating zone;

[0037] the device comprising means for controlling the DC-DC converter adapted to implement, if the converter operates in a frequency saturation zone and if the converter operates in a predefined operating zone, a pulse width modulation control, and otherwise to implement a pulse frequency modulation control.

[0038] The invention also relates to an electrical system comprising a battery of electrical accumulators, a reversible DC-DC converter, and a control device as described previously.

[0039] The invention also relates to an electric motor vehicle comprising an electrical system as described above.

[0040] Other features and advantages of the invention will emerge from reading the description given below of a particular embodiment of the invention, given for informational purposes but not as a limitation, with reference to the appended drawings in which:

[0041] [Fig-1] is a schematic representation of a DC-DC converter known from the prior art;

[0042] [Fig.2] is a schematic view of the frequency, battery voltage and power ratio during discharge operation of a DC-DC converter;

[0043] [Fig.3] is a schematic representation of an equivalent model of the converter continuous-continuous of [Fig.l];

[0044] [Fig.4] is a schematic view of the method according to the invention;

[0045] [Fig.5a] is a schematic view of the PFM control according to the invention;

[0046] [Fig.5b] is a schematic view of the PWM control according to the invention;

[0047] [Fig.6] is a graph representing the normalized gain as a function of the ratio cyclic in PWM.

[0048] [Fig.7] is a graph representing an example of the evolution of the DC bus of the converter according to a first example of PWM control;

[0049] [Fig.8] is a graph representing the evolution of the duty cycle of the PWM control of the converter according to the first example;

[0050] [Fig.9] is a graph representing an example of the evolution of the DC bus of the converter according to a second example of PWM control;

[0051] [Fig. 10] is a graph representing the evolution of the duty cycle of the PWM control of the converter according to the second example;

[0052] [Fig. 11] is a graph comparing in the saturated zone and in the predetermined zone according to the invention for the PWM control, the evolution of the DC bus according to the PWM control and according to the PFM control.

[0053] The invention relates to a hybrid method for implementing control by pulse width modulation, called Pulse Width Modulation, frequently abbreviated PWM in order to avoid in the saturation zone the usual limitations of the pulse frequency modulation strategy, also called Pulse Frequency Modulation, PFM.

[0054] The invention here implements the PFM strategy except in the saturation zone as explained below.

[0055] The method according to the invention firstly comprises a modeling of the system in V2X mode, Vehicle to Everything, by dynamic modeling of the DCDC-LLC converter according to the first harmonic approach.

[0056] Thus, the objective of the invention is to control in discharge mode, V2X, the DCDC LLC of [Fig.l],

[0057] In the literature, large-signal models are known in particular from scientific publications:

[0058] - L. Yao, D. Li, and L. Liu, “An improved large signal model of full-bridge LLC converter,” PLOS ONE, vol. 13, no. 10, E. Wang, Ed., e0205904, Oct. 2018; and

[0059] - Z. Fang, J. Wang, S. Duan, K. Liu, and T. Cai, “Control of an LLC Resonant Converter Using Load Feedback Linearization,” IEEE Transactions on Power Electronics, vol. 33, no. 1, pp. 887-898, Jan. 2018

[0060] Such a large-signal model is based on the extension of the description function, called extending describing function, abbreviated EDF, which consists of approximating the non-linear terms in the dynamic equations of the DCDC LLC converter by first harmonic sinusoidal components.

[0061] This model provides sufficient dynamic information in the event of significant transient signal disturbance.

[0062] However, the model remains non-linear, which makes the design of the control more complex.

[0063] Also, we first proceed to the calculation of an equivalent model of the LLC resonant converter of [Fig.l] operating in inverted mode, in other words in V2X mode.

[0064] This equivalent circuit is shown [Fig.3], and allows the transfer function of the DCDC LLC circuit to be calculated in V2X mode as follows:

[0065] The transfer function of the DCDC LLC converter is given by:

[0066] [Math.l] c = = R 'r~ w "

[0067] In which in discharge mode, Vof is the output voltage of the resonant circuit, primary side, i.e. across the terminals of the “rectifier” bridge 11, Vif the input voltage of the resonant circuit, secondary side, i.e. battery side, s is the Laplace variable, and R the equivalent resistance calculated as: [Math.2]

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080] With P the power transferred between the input and output of the converter We can then write the transfer function of the gain of the DCDC LLC converter, n being the transformation ratio of the transformer: [Math.3] Therefore, we can obtain the equation for calculating the switching frequency f by inverting the gain ie by solving a degree 4 equation expressing the relationship between the gain of the converter in discharge mode, the switching frequency and the input and output voltages of the converter 10: [Math.4] With w = Zt f Equation [Math 4] allows the feedforward switching frequency (i.e. by anticipation) to be calculated by inversion of the gain, presented in the following equation: [Math.5] The calculation of the switching frequency by gain inversion shows a significant saturation region 20, as shown [Fig.2], where the frequency is saturated at 200 kHz, which is the maximum switching frequency allowed in this embodiment of the invention. In this region, the regulation of the DC bus is lost and the DCDC has a low efficiency. Also in order to be able to regulate the value of the DC bus for operating points in this area, a hybrid control strategy can be implemented by applying a control strategy by pulse width modulation, called Pulse Width Modulation, also abbreviated PWM, by inversion of the gain with a constant switching frequency. However, outside the saturation zone, PFM control is implemented.

[0081] The PWM strategy is implemented based on the variation of the duty cycle.

[0082] In this PWM strategy, the switching frequency f is fixed and the bus voltage DC is regulated by the variable duty cycle D. Applying a first harmonic method, the input voltage of the resonant circuit, in V2X mode, can be written as presented in the following equation:

[0083] [Math.6]

[0084] The output voltage is:

[0085] [Math.7]

[0086] The gain transfer of the DCDC LLC converter for the PWM mode G' is a function of the gain G given by Equation 1 and can be expressed by:

[0087] [Math. 8]

[0088] The gain G' of the converter voltage is regulated by varying the duty cycle rather than increasing the switching frequency as in the case of the PFM strategy. Starting from equation 8, the duty cycle by anticipatory action, called feedforward can be defined as presented in the following equation:

[0089] j p+£f^)r A = 2^arccos|l- x |—-1

[0090] However, in order to implement the hybrid control law between the two PFM and PWM strategies, the operating range of each strategy must be defined.

[0091] Two operating zones are shown in [Fig.2] which can be defined by the following equation:

[0092] [Math. 10] S = (P) such that (VhaP P) E frequency saturation zone 20

[0093] A = (^ P ) such that 340 V < < 350 V and 1800 VF < P < 3000 W or 350 V <Vbat< 370 V et 2800 W < P < 3300 VF ou 370 V <Vbat< 390 Vet 3700 W<P<4300 VF ou 390 V < V^<410 V and 4700 W <P< 5300 W ou 410 V <Vhat< 430 V et 5000 W < P < 6000 VF

[0094] These areas of course depend on various physical parameters of the converter and battery used and must be adapted to each use case.

[0095] Thus, as shown [Fig.4], 5a and 5b, the converter control is hybrid.

[0096] If the couple (Battery Voltage VBAt, Power) is located in a saturation zone 20 (also referenced S) as represented [Fig.2], then we check whether this same couple is included in a predefined operating zone A. If necessary, we implement a strategy by pulse width modulation, in English Pulse Width Modulation, abbreviated PWM.

[0097] It should be noted that when the torque (VBat, P) is in the saturation zone S but not in the operating zone A, the DC bus is not controllable, so the strategy adopted is PSM (Phase Shift Modulation) which consists of using the phase shift between the voltage and the current at the primary of the transformer (in discharge mode) to control the power transfer between the battery and the DC bus.

[0098] When the torque is not in a saturation zone, a PFM control known from the prior art is then implemented.

[0099] The feedforward switching frequency fO, within the framework of the PFM control, is defined by inverting the gain by setting the DC bus voltage equal to the reference voltage 450 V, as presented in equation 5. The feedforward duty cycle D0, implemented within the PWM control, is defined by gain inversion by setting the DC bus voltage equal to the reference voltage 450 V with a fixed switching frequency f, as presented in equation 9. Depending on the operating point, the control is chosen from PWM or PFM.

[0100] We first check 41 whether the operating point belongs or not to the saturation zone S.

[0101] If it does not belong to S, then the PFM command 44 can be used.

[0102] In this case, with reference to [Fig.5a], the switching frequency is calculated feedforward fO.

[0103] Otherwise, we check 42 if the operating point belongs to the operating zone A.

[0104] If the operating point belongs to zone A, the PWM command is applied 45 by calculating the feedforward duty cycle D0, with reference to [Fig.5b]. The variable uO by feedforward action, which is either fO for PFM or DO for PWM; is added 46 to the action u of a Proportional-Integral type regulator 43 to cancel the static error e of the VDC voltage of the DC bus (the constants Kp and Ki are adjustment parameters). It should be noted that when the duty cycle variation is far from 0.5 (outside zone A), the first harmonic theory is no longer valid. The ZVS (Zero Voltage Switching) property may also be lost in this case. However, an acceptable bus regulation error can be achieved, requiring significant corrective action by the regulator.

[0105] In the case of low power in the area where the PWM strategy can be applied, good performance in terms of DC bus voltage regulation can be obtained. The efficiency of the DCDC, in this case, is better than that in the case of the PFM strategy, but it remains insufficient. Efficiency improvement remains a necessary need in this operating area.

[0106] The resonant frequency is defined by:

[0107] [Math. 11]

[0108] For f=fr, we have IGI = 1. The normalized voltage gain of the DCDC LLC converter I G'I is illustrated in [Fig.6] as a function of the duty cycle.

[0109] It is noted, with reference to [Fig.6], that the gain of the PWM control increases with the increase in the duty cycle D, until the duty cycle is around 0.5, then it begins to decrease.

[0110] Depending on the operating point, the frequency set in the PWM strategy has an effect on the regulation of the DC bus. It has been noticed that the 200 kHz frequency does not work in the saturation zone range for very high powers. In this case, the frequency used is the resonance frequency at 100 kHz.

[0111] However, in the methods of the prior art, the resonance frequency is often selected for the PWM method.

[0112] In a motor vehicle, or in any other electrical system, the control method is implemented by an electronic device such as an on-board computer, by a processor, a System On a Chip (SoC) or system embedded on a single integrated circuit, a microcontroller or any suitable device, in particular an FPGA processor (from the English "field-programmable gate array").

[0113] The different steps can be implemented by the same electronic device, or the different steps can be implemented by several different electronic devices.

[0114] The results of experimental simulations, with reference to figures 7 and 8, show first of all that the VDC voltage of the DC bus in PWM control with VBat=350 V and P=2000 W converges relatively quickly (approximately 0.4s) towards the reference voltage VDcref.

[0115] Thus, convergence to the reference can be ensured within an acceptable DC bus voltage error tolerance. We also note that the duty cycle varies around 0.36 in steady state.

[0116] In another experimental result, shown in figures 9 and 10, we note that with VBat=420 V and P=5000 W the VDC voltage also converges substantially in 0.4 sec, with a duty cycle varying around 0.38 in steady state.

[0117] With reference to [Fig. 11], where the results of a PFM control and a PWM control are compared for VBat=350 V and P=2000 W, it is noted that the loss of regulation of the DC bus with the PFM strategy can be avoided when using the PWM strategy, where the DC bus voltage error has an acceptable tolerance around 5 V.

Claims

Claims

1. Method for controlling (1) a reversible DC-DC LLC converter (10), comprising a DC bus, a primary H-bridge (11) and a secondary H-bridge (14) of four MOSFET transistors each (S1-S4, S5-S8), connected to a battery (16) of electric accumulators of an electric motor vehicle, said DC-DC converter (10) being adapted to operate in charging and discharging of said battery (16), and when said converter (10) operates in discharging, a verification step (41, 42) of the operating point of the converter is implemented during which it is verified whether the converter operates in a frequency saturation zone (S) and whether the converter operates in a predefined operating zone (A),If the converter (10) operates in a frequency saturation zone (S) and if the converter (10) operates in a predefined operating zone (A), the method implements a control step by pulse width modulation depending on the battery voltage, a reference voltage (VDCref) of the DC bus, the power (P) of the converter (10) and a switching frequency (f) of the converter (10), otherwise the method (1) implements a control step by pulse frequency modulation depending on the battery voltage, a reference voltage (VDCref) of the DC bus, the power of the converter (10); characterized in that if the converter (10) operates in a frequency saturation zone (S) and if the converter (10) operates in a predefined operating zone (A), said control step comprises calculating the duty cycle of the modulated signal by gain inversion.,

2. Method according to claim 1, characterized in that said control step comprises an inversion of the gain of the converter (10).

3. Method according to claim 2, characterized in that said control step comprises a Proportional-Integral regulation (43), a function of the voltage VDc of the DC bus and of a reference voltage (VDCref)

4. Method according to any one of claims 1 to 3, characterized in that said frequency saturation zone corresponds to a operating point of the converter (10), depending on the battery voltage (VBat) and the power (P) of the converter, in which the switching frequency (f) saturates at a maximum value. ^Claim 5] Method according to any one of claims 1 to 4, characterized in that said predefined operating zone (A) is defined by pairs of battery voltage (VBat) and converter power (P) where: ' 340 V < < 350 V and 1800 W < P < 3000 VF or 350 V < < 370 V and 2800 W < P < 3300 VF or A = (Vbat, P) such that 370 V < < 390 V and 3700 W < P < 4300 W or 390 V < < 410 V and 4700 W < P < 5300 VF or t 410 V < Vbat < 430 V and 5000 VF < P < 6000 VF ^Claim 6] Device for controlling a reversible DC-DC converter comprising a primary H-bridge (11) and secondary H-bridge (14) of four MOSFET transistors each (S1-S4, S5-S8) connected to a battery (16) of an electric motor vehicle,said DC-DC converter being adapted to operate in charging and discharging of said battery, and said device comprising means for implementing, when said converter operates in discharging, a step of verifying the operating point of the converter during which it is verified whether the converter operates in a frequency saturation zone (S) and whether the converter operates in a predefined operating zone (A); the device comprising means for controlling the DC-DC converter adapted to implement, if the converter operates in a frequency saturation zone (S) and if the converter operates in a predefined operating zone (A), a control by pulse width modulation,and otherwise to implement a control by modulation of the pulse frequency; characterized in that if the converter (10) operates in a frequency saturation zone (S) and if the converter (10) operates in a predefined operating zone (A), said control step comprises the calculation of the duty cycle of the modulated signal by inversion of the gain., 14

7. An electrical system comprising an electric storage battery, a reversible DC-DC converter, and a control device according to claim 6.

8. An electric motor vehicle comprising an electrical system according to claim 7.

Images