Electronic device for controlling an electrical energy converter with piezoelectric element(s), associated control method and electrical energy conversion system

Abstract

Electronic device for controlling an electrical energy converter with piezoelectric element(s), associated control method and electrical energy conversion system The present invention relates to a control of an electrical energy converter comprising at least one elementary piezoelectric converter, such that a control cycle equal to s+1 resonance periods of the piezoelectric assemblies, s being an integer greater than or equal to zero, comprises at least two phases at substantially constant voltage per half-control cycle, a first half-control cycle comprising a first phase (II) at substantially constant voltage of first voltage value and first duration and a second phase (IV) at substantially constant voltage of second voltage value and second duration,a second control half-cycle comprising a third phase (VI) at substantially constant voltage of value opposite to the first voltage value and of duration substantially equal to the first duration and a fourth phase (VIII) at substantially constant voltage of value opposite to the second voltage value and of duration substantially equal to the second duration. Figure for the abstract: Figure 3,

Description

Title of the invention: Electronic device for controlling an electrical energy converter with piezoelectric element(s), associated control method and electrical energy conversion system

[0001] The present invention relates to an electronic device for controlling an electrical energy converter capable of converting an input voltage into at least one output voltage and an electrical energy conversion system.

[0002] The invention also relates to a method for controlling an electrical energy converter capable of converting an input voltage into at least one output voltage and an associated electronic electrical energy conversion system.

[0003] The invention lies in the field of electronic electrical energy conversion systems, in particular those comprising a piezoelectric element, in particular voltage conversion systems, i.e. direct-direct conversion systems, also called DC-DC conversion systems (from the English Direct Current - Direct Current), and alternating-direct conversion systems, also called AC-DC conversion systems (from the English Alternating Current - Direct Current).

[0004] Various topologies of electrical energy converter with piezoelectric elements are known.

[0005] For example, electrical energy converters with piezoelectric element(s) are described in documents FR 3 064 850 B1, FR 3 086 471 B1 and FR 3 130 096 AL

[0006] Electrical energy converters with piezoelectric element(s) are known, comprising one or more piezoelectric assemblies, each connected between midpoints of branches of a switching bridge, each of the branches of the switching bridge comprising switches connected together at the midpoint of the branch.

[0007] In known systems, the control device is configured to control, during a respective resonance cycle of the piezoelectric assemblies, a switching of each of the switches to alternate phases at substantially constant voltage across the terminals of the piezoelectric assembly or assemblies and phases at substantially constant charge across the terminals of the piezoelectric assemblies. The control is configured to carry out a given number of phases at substantially constant voltage (or voltage steps), while respecting operating constraints of the piezoelectric converter as a function of its topology, for example zero voltage switching (ZVS), and optimization of power transfer between input and output.

[0008] In particular, control methods with three voltage levels are known, requiring the implementation of several regulation loops to comply with the desired operating constraints.

[0009] The invention aims to present a less complex, and consequently less expensive, device and method for controlling the converter, while retaining the performance of the electrical energy converter, more particularly suitable for control in voltage step-down mode.

[0010] For this purpose, the invention relates to an electronic device for controlling an electrical energy converter, the electrical energy converter comprising at least one elementary piezoelectric converter, the or each elementary piezoelectric converter comprising at least one primary switching branch and at least one secondary switching branch, the primary switching branch being connected between two primary terminals for applying an input voltage of the elementary piezoelectric converter, and comprising at least two primary switches connected in series and connected together at a primary midpoint, the secondary switching branch being connected between two secondary terminals for supplying an output voltage of the elementary piezoelectric converter, and comprising at least two secondary switches connected in series and connected together at a secondary midpoint,the elementary piezoelectric converter comprising a piezoelectric assembly connected between the primary midpoint and the secondary midpoint, the electronic control device being configured to control in the or each elementary piezoelectric converter, a switching of the primary and secondary switches to alternate phases at substantially constant voltage at the terminals of the piezoelectric assembly and phases at substantially constant charge at the terminals of the piezoelectric assembly. This electronic control device is configured to, for a control cycle equal to s+1 resonance periods of the piezoelectric assembly, s being an integer greater than or equal to zero, control at least two phases at substantially constant voltage per half-control cycle,a first control half-cycle comprising a first phase at substantially constant voltage of first voltage value and first duration and a second phase at substantially constant voltage of second voltage value and second duration, a second control half-cycle comprising a third phase at substantially constant voltage of value opposite to the first voltage value and of duration substantially equal to the first duration and a fourth phase at substantially constant voltage of value opposite to the , second voltage value and duration substantially equal to the second duration.

[0011] Advantageously, a balancing of the electrical energy transfers per half-cycle of control is obtained, which makes it possible to reduce the amplitude of energy oscillation within the piezoelectric assemblies, and consequently to improve the electrical energy transfer performance of the converter. Advantageously, the invention applies to any topology of electrical energy converter using at least one piezoelectric resonator.

[0012] According to other advantageous aspects of the invention, the electronic device for controlling an electrical energy converter comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations.

[0013] One of the voltage values ​​among the first voltage value and the second voltage value is equal to the difference or the opposite of the difference between an amplitude of the input dynamic voltage and an amplitude of the output dynamic voltage, the amplitude of the input dynamic voltage being equal to half the input differential voltage, the amplitude of the output dynamic voltage being equal to half the output differential voltage, and the other of the first voltage value and the second voltage value is equal to the sum or the opposite of the sum of said amplitude of the input dynamic voltage and said amplitude of the output dynamic voltage.

[0014] This device is configured to control sequences of phases at substantially constant voltage and at substantially constant load per half-control cycle, the phases of the second half-control cycle being obtained by translation and inversion of sign of the phases of the first half-control cycle.

[0015] It is configured to control the primary switches to realize said first, second, third and fourth phases at constant voltage.

[0016] This device is configured to control one of the primary switches in the closed position for part of the control cycle and to control the other of said primary switches in the closed position with a shift of half a control cycle, each of the primary switches being controlled to close for the same contact duration.

[0017] It comprises a block for estimating a dead time duration between the opening command of one of the primary switches and the closing command of the other of said primary switches, said primary switches being open during the dead time duration.

[0018] The estimation block is configured to determine said dead time duration as a function of a control cycle frequency, the input voltage and the output voltage of the elementary piezoelectric converter.

[0019] The device is configured to regulate an output voltage or current or power by controlling the control cycle frequency and / or a control phase shift of the secondary switches.

[0020] It is configured to control, when the input voltage of the elementary piezoelectric converter is greater than the output voltage of the elementary piezoelectric converter, said dead time duration with a switching tolerance margin.

[0021] It is configured to further control one of the secondary switches in the closed position during said first duration, and to control another of said secondary switches in the closed position during said second duration.

[0022] The invention also relates to an electronic electrical energy conversion system comprising an electrical energy converter, the electrical energy converter comprising at least one elementary piezoelectric converter, the or each elementary piezoelectric converter comprising at least one primary switching branch and at least one secondary switching branch, the primary switching branch being connected between two primary terminals for applying an input voltage of the elementary piezoelectric converter, and comprising at least two primary switches connected in series and connected together at a primary midpoint, the secondary switching branch being connected between two secondary terminals for supplying an output voltage of the elementary piezoelectric converter, and comprising at least two secondary switches connected in series and connected together at a secondary midpoint,the elementary piezoelectric converter comprising a piezoelectric assembly connected between the primary midpoint and the secondary midpoint, and an electronic device for controlling said elementary piezoelectric converter as briefly described above.

[0023] The invention also relates to a method for controlling an electrical energy converter capable of converting an input voltage into at least one output voltage implemented by an electronic control device as described above, comprising, for a control cycle equal to s+1 resonance periods of the piezoelectric assemblies, s being an integer greater than or equal to zero, the control of at least two phases at substantially constant voltage per half-control cycle, a first half-control cycle comprising a first phase at substantially constant voltage of first voltage value and first duration and a second phase at substantially constant voltage of second voltage value and second duration, a second half-control cycle comprising a third phase at substantially constant voltage of value opposite to the first voltage value and of duration substantially equal to the first duration and a fourth ... substantially constant voltage of value opposite to the second voltage value and of duration substantially equal to the second duration

[0024] According to other advantageous aspects of the invention, the control method comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations:

[0025] The method comprises a determination, for each half-control cycle, of a switching angle in the closed position of the switches of one of the first switches.

[0026] When said output voltage is lower than the input voltage, the determination of a switching angle is carried out with a tolerance margin, following a dead time duration during which the first switches are open, the method comprising an estimation of said tolerance margin as a function of a frequency of the control cycle, of the input voltage and of the output voltage.

[0027] The invention will appear more clearly on reading the description which follows, given solely by way of non-limiting example, and made with reference to the drawings in which:

[0028] [Fig-1] [Fig.l] is a schematic representation of a piezo converter elementary electric and an electric circuit corresponding to a piezoelectric element;

[0029] [Fig.2] [Fig.2] is a functional representation of an electronic system of electrical conversion;

[0030] [Fig.3] [Fig.3] is a representation of the voltage evolution curves at terminals of the piezoelectric assembly of an elementary piezoelectric converter, and control diagrams of the associated switches, in a voltage step-down type configuration;

[0031] [Fig.4] [Fig.4] is a representation of the voltage evolution curves at terminals of the piezoelectric assembly of an elementary piezoelectric converter, and control diagrams of the associated switches, in a voltage booster type configuration;

[0032] [Fig.5] [Fig.5] is a representation of the voltage evolution curves and the control diagrams of switches in a voltage step-down configuration over a control cycle comprising several resonance periods of the piezoelectric assembly;

[0033] [Fig.6] [Fig.6] is a representation of the control diagrams of the inter primary breakers in a step-down configuration;

[0034] [Fig.7] [Fig.7] is a block diagram of the functional blocks of an electronic device control electronics implementing a method for controlling an electrical energy converter;

[0035] [Fig.8] [Fig.8] is an exemplary embodiment of an energy converter electric comprising an elementary piezoelectric converter;

[0036] [Fig.9] [Fig.9] is a representation of the voltage evolution curves at terminals of the piezoelectric assembly of an elementary piezoelectric converter, and control diagrams of the associated switches of the electrical energy converter of [Fig.8].

[0037] [Fig.l] illustrates an electrical energy converter 2 called an elementary piezoelectric converter, as well as an electrical circuit modeling the operation of a piezoelectric element.

[0038] The control device and method of the invention described below apply to electrical energy converters comprising one or more elementary piezoelectric converters, regardless of the topology of the electrical energy converters.

[0039] In particular, the electronic control device is configured to control the or each of the elementary piezoelectric converters of an electrical energy converter.

[0040] The elementary piezoelectric converter 2 illustrated in [Fig.l] comprises a switching branch called primary switching branch 4a and a switching branch called secondary switching branch 4b.

[0041] The primary switching branch 4a comprises a first primary terminal 6a and a second primary terminal 8a, and a primary midpoint 10a.

[0042] Vxai denotes the lowest potential and Vxah the highest potential across the terminals of the primary switching branch 4a. In the example of [Fig.l], Vxai is the potential of the first primary terminal 6a and Vxah is the potential of the second primary terminal 8a. In the event of a potential reversal, for example in the event of the presence of an alternating voltage with a change of polarity, it is then always possible, at any time, to position Vxah and Vxai such that Vxah is greater than Vxab. The voltage applied between the primary terminals 6a, 8a is called the primary voltage.

[0043] The secondary switching branch 4b comprises a first secondary terminal 6b and a second secondary terminal 8b, and a secondary midpoint 10b.

[0044] VxM denotes the lowest potential and Vxbh the highest potential across the terminals of the secondary switching branch 4b. In the example of [Fig.l], Vxbi is the potential of the first secondary terminal 6b and Vxbh is the potential of the second secondary terminal 8b. In the event of a potential reversal, for example in the event of the presence of an alternating voltage with a change of polarity, it is then always possible, at any time, to position Vxbh and Vxbi such that Vxbh is greater than Vxbb. The voltage applied between the secondary terminals 6b, 8b is called the secondary voltage.

[0045] The primary switching branch 4a comprises two arms, respectively a first primary arm 14a connected between the first primary terminal 6a and the primary midpoint 10a, and a second primary arm 16a connected between the second primary terminal 8a and the primary midpoint 10a.

[0046] Each of the first and second primary arms comprises a primary switch 18a, 20a, also denoted Kxd and Kxah.

[0047] The secondary switching branch 4b comprises two arms, respectively a first secondary arm 14b connected between the first secondary terminal 6b and the secondary midpoint 10b, and a second secondary arm 16b connected between the second secondary terminal 8b and the secondary midpoint 10b.

[0048] Each of the first and second secondary arms comprises a secondary switch 18b, 20b, also denoted KxM and Kxbh.

[0049] Each switch of the elementary piezoelectric converter 2, namely each of the switches 18a, 18b, 20a, 20b, is preferably a unidirectional voltage switch, and comprises for example a transistor, or a diode, or even a transistor and a diode in antiparallel, not shown.

[0050] The switch 18a, 18b, 20a, 20b is preferably made up of the transistor, or the diode, or the transistor and the diode in antiparallel.

[0051] The use of a bidirectional voltage switch is required for the primary arms 14a and 16a if the primary voltage is likely to change sign, for example if it is an alternating voltage with a change of polarity. Similarly, the use of a bidirectional voltage switch is required for the secondary arms 14b and 16b if the secondary voltage is likely to change sign, for example if it is an alternating voltage with a change of polarity.

[0052] Alternatively, the switch 18a, 18b, 20a, 20b comprises an association of several transistors, and is preferably made up of such an association of several transistors.

[0053] As a further variant, the switch 18a, 18b, 20a, 20b comprises a mechanical switch, such as a MEMS (MicroElectroMechanical System) microswitch.

[0054] The transistor is, for example, an insulated gate field effect transistor, also called MOSFET (from the English Metal Oxide Semiconductor Field Effect Transistor). Alternatively, the transistor is a bipolar transistor; an insulated gate bipolar transistor, also called IGBT (from the English Insulated Gate Bipolar Transistor); a silicon (Si) based transistor, a GaN (from the English Gallium Nitride) based transistor; a silicon carbide (SiC) based transistor, or a diamond based transistor, or even a thyristor.

[0055] A piezoelectric assembly 22 is connected between the primary midpoint 10a and the secondary midpoint 10b.

[0056] The piezoelectric assembly 22 comprises one or more piezoelectric element(s) 15, each composed of a piezoelectric material having an associated resonant frequency.

[0057] As known per se, the mechanical oscillation of the piezoelectric elements 15 is approximately sinusoidal over respective resonance cycles, also called resonance periods. When several piezoelectric elements are present, the total mechanical deformation of the piezoelectric elements 15 is the sum of elementary mechanical deformations of each of the piezoelectric elements 15.

[0058] An increase or decrease in the energy stored over a period leads respectively to an increase or decrease in the amplitude of mechanical oscillation.

[0059] The voltage Vpx at the terminals of the piezoelectric assembly 22 is dependent on the states of the primary switches Kxah, Kxai, and secondary switches Kxbh, Kxbi, capable of being controlled to alternate phases at substantially constant voltage and phases at substantially constant load.

[0060] By substantially constant charge is meant an exchange of a charge with the outside which is less than 30% of the charge which would have been exchanged with the outside if the voltage had been kept constant. In other words, by substantially constant charge is meant a variation in charge less than 30% of the charge which would have been exchanged with the outside of the piezoelectric assembly 22 if the voltage across the terminals of the piezoelectric assembly 22 had been kept constant over the time period considered.

[0061] By substantially open electrical circuit, we mean a circuit in which a possible leakage current leads to a variation in charge of the piezoelectric assembly 22 of less than 30% of the charge which would have been exchanged with the exterior of the piezoelectric assembly 22 if the voltage across the terminals of the piezoelectric assembly 22 had been kept constant over the time period considered.

[0062] By substantially constant voltage is meant a voltage variation of less than 20%, preferably less than 10%, of the input or output voltage of the converter 10. For example, if the input voltage of the converter 10 is equal to 100V, then the voltage variation during each phase at substantially constant voltage, i.e. on each step at substantially constant voltage, is less than 20% of this voltage, i.e. less than 20V; preferably less than 10% of this voltage, i.e. less than 10V. Each phase at substantially constant voltage is also called a voltage step.

[0063] A modeling of a piezoelectric element 15 in the form of an electrical circuit is illustrated in bubble 25 in [Fig.l].

[0064] The piezoelectric element 15 is known per se, and is typically modeled, close to the resonance mode used, in the form of a capacitor 24 and a resonant branch 26 connected in parallel with the capacitor 24, the capacitor 24 and the resonant branch 26 being connected between a first electrode 28 and a second electrode 30 of the piezoelectric element 15. For example, when the set 22 of piezoelectric elements of the elementary piezoelectric converter is reduced to a piezoelectric element, the first electrode 28 is connected to the primary midpoint 10a and the second electrode 30 is connected to the secondary midpoint 10b.

[0065] The resonant branch 26 is typically an RLC branch formed of a resistor 32, a capacitor 34, and an inductor 36 connected in series. The capacitance Co of the capacitor 24 connected in parallel with the resonant branch 26 is called the parallel capacitance, or blocked capacitance, or reference capacitance. The voltage across the terminals of the piezoelectric element 15 then typically corresponds to the voltage across the terminals of the capacitor 24. In steady state, a substantially sinusoidal current iL flows in the RLC branch of the equivalent model of the piezoelectric element 15.

[0066] The respective values ​​of the resistance R, of the capacitance C of the capacitor 34, and of the inductance L, as well as of the reference capacitance Co define the dimensioning of the piezoelectric element 15.

[0067] [Fig.2] schematically illustrates an electronic energy conversion system electrical 35, configured to provide electrical energy to a load 39 from a source 37.

[0068] The electronic electrical energy conversion system 35 comprises an electrical energy converter 40 and an electronic control device 42 for the electrical energy converter 40.

[0069] The electrical energy converter 40 is configured to convert an input voltage Vin into an output voltage Vout.

[0070] The electrical energy converter 40 comprises a first input terminal 44 and a second input terminal 46, a first output terminal 48 and a second output terminal 50. The potential Vinn applied to the first input terminal 44 is lower than the potential Vinp applied to the second input terminal 46. The potential Voutn supplied to the first output terminal 48 is lower than the potential Voutp applied to the second output terminal 50.

[0071] To simplify the description, in the presence of an input voltage that changes sign, Vinn and Vinp are then redefined at each instant so that Vinn is always less than VinP. Similarly, in the case of supplying an output voltage that changes sign, Voutn and VoutP are then redefined at each instant so that Voutn is always less than Voutp-

[0072] Note that we consider a control cycle frequency (or piloting frequency) at least 10 times higher than the variation frequency of the primary or secondary voltage, advantageously at least 100 times, so that during a control cycle, the input voltage Vin and the output voltage Vout can be considered substantially constant and without change of sign. The electrical energy converter 40 is preferably a DC-DC converter capable of converting a substantially DC electrical energy or voltage into another substantially DC electrical energy or voltage on the scale of a control cycle.

[0073] The electronic electrical energy conversion system 35 is for example a DC-DC conversion system capable of converting a first DC electrical energy or voltage received at the input into a second DC electrical energy or voltage delivered at the output, or an AC-DC conversion system capable of converting an AC electrical energy or voltage received at the input into a DC electrical energy or voltage delivered at the output of the electrical energy conversion system. For example, the electrical energy source 37 is a battery, a solar panel, a voltage from a rectified AC electrical network or a DC power supply bus.

[0074] When the electrical energy conversion system is an AC-DC conversion system, with for example an AC input voltage centered on 0V, the electrical energy conversion system preferably further comprises a voltage rectifier, not shown, connected to the input of the electrical energy converter 40 and capable of rectifying the AC electrical voltage received at the input of the conversion system to deliver a rectified electrical voltage at the input of the converter 40. The voltage rectifier is for example a rectifier bridge, such as a diode bridge.

[0075] The electrical energy converter 40 comprises one or more elementary piezoelectric converter(s) 2, as described with reference to [Fig.l].

[0076] In embodiments, the electrical energy converter 40 comprises several elementary piezoelectric converters 2, for example connected in series via their respective primary switching branches 4a or via their respective secondary switching branches 4b, between two terminals among the first, the second input terminal, the first and the second output terminal.

[0077] Driving the electrical energy converter 40 to lower the voltage between its input and its output is called step-down driving.

[0078] The control of the electrical energy converter 40 to raise the voltage between its input and its output is called control in boost mode.

[0079] The electronic control device 42 and an associated control method will be described below, for control in voltage step-down mode and in voltage step-up mode. voltage of an elementary piezoelectric converter as described with reference to [Fig.l].

[0080] It is clear to a person skilled in the art that the control method applies in a similar manner to control each elementary piezoelectric converter forming part of an electrical energy converter comprising a plurality of elementary piezoelectric converters, after calculating the input voltage to be applied to the terminals of the elementary piezoelectric converter and the output voltage supplied by the elementary piezoelectric converter, as a function of the topology of the electrical energy converter 40, the input voltage Vin and the output voltage Vout.

[0081] The electronic control device 42 is configured to control the electrical energy converter 40, in particular to control the control of the switches 18a, 20a and optionally 18b, 20b of each elementary piezoelectric converter, in order to alternate phases at substantially constant voltage at the terminals of the piezoelectric assembly 22 and phases at substantially constant charge, i.e. in substantially open circuit, at the terminals of the piezoelectric assembly 22.

[0082] The electronic control device 42 is for example produced in the form of an electronic circuit comprising one or more electronic components.

[0083] Alternatively, the electronic control device 42 is produced in the form of a programmable logic component, such as an FPGA (Field Programmable Gate Array) or in the form of an integrated circuit, such as an ASIC (Application Specific Integrated Circuit) or even in the form of a computer, such as a microcontroller, a processor.

[0084] The operation of the four-voltage step control of an elementary piezoelectric converter as described with reference to [Fig.2] will be described below.

[0085] The control method implements periodic angular control.

[0086] It should be noted that it is possible, in an equivalent manner, to express the mechanical resonance period of the piezoelectric assembly as equal to a time period between 0 and T or an angular period between 0 and 2ir.

[0087] Thus, there is a mathematical correspondence between temporal duration and angular duration.

[0088] When two distinct phases are indicated to have the same duration, this means both the same temporal duration and the same angular duration.

[0089] [Fig.3] illustrates graphs of the evolution of the dynamic piezoelectric voltage V Pxdyn at the terminals of the piezoelectric assembly Px of an elementary piezoelectric converter, driven in a voltage step-down mode on a control cycle equal to a resonance period of 2ir of the piezoelectric assembly. The control is carried out at angle 0 evolving over a period between 0 and 2ir radians.

[0090] The notation “x” indicates that generically, the elementary piezoelectric converter considered is the elementary piezoelectric converter number “x” in a given electrical energy converter architecture.

[0091] The dynamic voltage VPxdyn across the terminals of the piezoelectric assembly is equal to the voltage Vpx across the terminals of the piezoelectric assembly from which a common mode differential component of the voltage has been subtracted.

[0092] On the primary switching branch, the primary common mode voltage is calculated by the formula:

[0093] [Math.l] \r_ 'xamc 2

[0094] In other words, the input common mode voltage is equal to the average of the potential Vxai applied to the first primary terminal 6a and the potential Vxah applied to the second primary terminal 8a.

[0095] On the secondary switching branch, the secondary common mode voltage is calculated by the formula:

[0096] [Math.2] V. ; + V , , 1 / __ ' xbir ' xbl 'xhmc 2

[0097] In other words, the output common mode voltage is equal to the average of the potential Vxbifoumi on the first secondary terminal 6b and the potential Vxbh supplied on the second secondary terminal 8b.

[0098] The common mode differential component of voltage VPxmc is calculated by:

[0099] [Math.3] T / _ WT / _ Kva / î+Kwrxbl v P Mile ~ V xamc ■ y xbmc “ 2

[0100] The dynamic voltage VPxdyn across the piezoelectric assembly is calculated by:

[0101] [Math.4] Vpxdyn ~ Vpx " V pxmc

[0102] The value of the voltage amplitude across the primary common mode voltage, also called the dynamic input voltage amplitude, is defined by:

[0103] [Math.5] V XV ■ 17. xah xai V xa~ 2

[0104] In other words, the amplitude of the dynamic input voltage Vxa is equal to half the differential input voltage of the elementary piezoelectric converter.

[0105] The value of the voltage amplitude across the secondary common mode voltage, also called the dynamic output voltage amplitude, is defined by:

[0106] [Math.6] V -VT 7 __ xbh xbl vxb~ 2

[0107] In other words, the amplitude of the dynamic output voltage Vxb is equal to half the differential output voltage of the elementary piezoelectric converter.

[0108] By definition, the values ​​of the voltage amplitude Vxa and Vxb are positive.

[0109] In [Fig.3] are represented the sinusoids IL and D corresponding respectively to the normalized current IL between -1 and l (ordinate axis) as a function of the angle over the period [0, 2ir] and the deformation D of the piezoelectric assembly over the same period.

[0110] The electronic control device 42 is configured to alternately control phases at substantially constant voltage (or voltage steps) and phases at substantially constant load, according to an arrangement of the voltage steps illustrated in an example by the graphs in [Fig.3].

[0111] The four combinations of voltage values, which constitute the voltage values ​​of the voltage steps, are respectively:

[0112] [Math.7] va=-vxa+vxh (Kxai and KxW closed, Kxah and Kxbh open)

[0113] Vb = - Vxb (Kxah and Kxbh closed, Kxd and KxW open)

[0114] Vc = Vxa + Vx / j(Kxah and Kxbi closed, Kxd and Kxbh open)

[0115] Vd = -Vxa ~ ^(Kxai and Kxbh closed, Kxah and KxW open)

[0116] As can be seen, the voltage values ​​Va and Vb are of the same value absolute and of opposite signs, and Vc and Vd are of the same absolute value and of opposite signs.

[0117] For the remainder of the description, it is considered that the current IL flowing in the piezoelectric assemblies has the direction indicated by the arrow shown in [Fig.l], and by construction the voltage level of value Va is associated with the positive alternation of the current IL and the voltage level of value Vb with the negative alternation of the current IL

[0118] As illustrated in the graph Gi of [Fig.3], for control in step-down mode, the dynamic voltage VPxdyn at the terminals of the piezoelectric assembly takes, during phases at substantially constant voltage, alternately the voltage values ​​Vb, Vd on a first control half-cycle, corresponding to the first resonance half-cycle (angle from 0 to ji), and the voltage values ​​Va, Vc on a second control half-cycle, corresponding to the second resonance half-cycle (angle from ir to 2ir).

[0119] The graph Gi in [Fig.3] illustrates the control in four phases at substantially constant load and four phases at substantially constant voltage.

[0120] In this step-down configuration, the control cycle, equal to the period resonance in this example, has two half-cycles.

[0121] The first half-cycle comprises two phases I, III at substantially constant charge and two phases II, IV at substantially constant voltage.

[0122] In detail, between 0o=0 and 0b during phase I at substantially constant load, the dynamic voltage VPxdyn at the terminals of the piezoelectric assembly changes from Vc to Vb.

[0123] During phase II at substantially constant voltage, between 0i and 02, the dynamic voltage VPxdyn is substantially equal to Vb.

[0124] During phase III at substantially constant load, between 02 and 02b, the dynamic voltage value VPxdyn evolves between Vb and Vd.

[0125] Finally, during phase IV at substantially constant voltage, between times 02b and 03 =ji, the dynamic voltage VPxdyn is substantially equal to Vd.

[0126] The second half-cycle also comprises two phases V, VII at substantially constant charge and two phases VI, VIII at substantially constant voltage.

[0127] In detail, between 03 and 03b, during phase V at substantially constant load, the dynamic voltage VPxdyn changes from Vd to Va.

[0128] During phase VI at substantially constant voltage, between 03b and 04, the dynamic voltage VPxdyn is substantially equal to Va.

[0129] During phase VII at substantially constant load, between 04 and 05, the dynamic voltage value VPxdyn varies between Va and Vc.

[0130] Finally, during phase VIII at substantially constant voltage, between times 05 and 0 6=2ir, the dynamic voltage VPxdyn is substantially equal to Vc.

[0131] Of course, due to the cyclic repetition, 06=2ir= 0O.

[0132] Furthermore, the respective substantially constant voltage phases II, VI are of substantially the same duration, called the first duration, and the respective voltage values ​​are substantially equal and of opposite signs (Vb, Va). In the example of [Fig.3], the first duration is:

[0133]

[0134] Similarly, the respective substantially constant voltage phases IV, VIII are of substantially the same duration, called the second duration, and the respective voltage values ​​are substantially equal and of opposite signs (Vd, Vc). In the example of [Fig.3], the second duration is:

[0135]

[0136] In other words, the phases of Gi are substantially symmetrical by translation and inversion of sign between the first half-cycle and the second half-cycle of control.

[0137] Graphs G2 and G3 respectively represent the evolution of the potentials Vxar at the primary midpoint and the potential Vxbr at the secondary midpoint, at the terminals of the piezoelectric assembly of the controlled elementary piezoelectric converter.

[0138] The respective potentials Vxar and Vxbr evolve according to the state of opening or closing of the respective switches on the primary side Kxah, Kxa[ and on the secondary side Kxbb, Kxbb

[0139] Thus, the potential Vxarest maintained at Vxah during phases I and II, and passes from Vxah to Vxai during phase III at substantially constant charge, then is maintained at Vxai during phases IV to VI, and passes from Vxa[ to Vxah during phase VII at substantially constant charge, then is maintained at Vxah during phase VIII at substantially constant voltage.

[0140] The potential Vxbr passes from VxM to Vxbh during phase I at substantially constant charge, then is maintained at Vxbh during phases II to IV, and passes from Vxbh to VxW during phase V at substantially constant charge, then is maintained at VxM during phases VI to VIII.

[0141] The control diagrams Comi, Com2, Com3 and Com4 of the primary switches Kxah, Kxa[ and of the switches Kxbh, Kxbi are illustrated over the same control period, from 0 to 2ir.

[0142] By convention, in each control diagram, a high state is representative of the closing state of the corresponding switch and a low state (e.g. 0) is representative of the opening state of the corresponding switch.

[0143] Depending on the scenario, the switches are controlled and held in position on command, which is the case for the Kxah, Kxa[ switches on the primary side, or can switch naturally into the indicated positions, which is the case for the Kxbh, KxW switches when these are implemented by diodes.

[0144] Diode mode conduction is illustrated in [Fig.3] by hatched areas.

[0145] The primary switches Kxah and Kxai are controlled in phase opposition, that is to say that the controls are identical but temporally offset by half a cycle. When one is closed, the other is open and vice versa, except during the periods between 02 and 02b, corresponding to phase III at substantially constant load and between 04 and 05, corresponding to phase VII at substantially constant load.

[0146] In the example of [Fig.3], the primary switch Kxah is held in the closed position between 0o=0 and 02, then between 05 and 06= Oo. In other words, when several control cycles follow one another, the primary switch Kxah is held in the closed position between 05 and 02 of a following control cycle, and in the open position between 02 and 0 5-

[0147] The primary switch Kxa[ is held in the closed position between 02b and 04, and in the open position between 00 and 02b and between 04 and 06=0o. In other words, when several control cycles follow one another, the primary switch Kxa[ is held in the closed position between 02 and 04 of a control cycle, and in the open position between 04 and 02b of the next order cycle.

[0148] Similarly, the secondary switches Kxbh and KxM are closed / open in phase opposition, that is to say that the controls are identical but angularly offset by ji. When one is closed, the other is open and vice versa, except during the periods between 00 and 0b corresponding to phase I at substantially constant load and between 03 and 03b, corresponding to phase V at substantially constant load.

[0149] As can be seen, during phases I, III, V, VII at substantially constant load, only one of the switches Kxah, Kxab Kxbh, KxM is in the closed position, the other three being in the open position.

[0150] During phases II, IV, VI, VIII at substantially constant voltage, one of the primary switches and one of the secondary switches are closed, the other of the primary switches and the other of the secondary switches being open.

[0151] Advantageously, in the step-down configuration, in embodiments in which the primary switches Kxbh, KxW have an intrinsic reverse diode or an additional diode placed in parallel, their conduction can be done naturally with respect to the sign of the current, and consequently the voltage levels making it possible to obtain the potential values ​​Vxbh, VxW at the secondary midpoint are obtained without explicit control. In other words, in such an advantageous configuration, only the primary switches are controlled by the electronic control device 20.

[0152] Alternatively, it is possible to control the secondary switches Kxbh, Kxbi to reduce losses. In this case, in the step-down configuration illustrated in [Fig.3], it would be appropriate to control the switch Kxbh in the closed position at 0B and keep it in the closed position between 0i and 03, then open between 03 and 0ïd'a following control cycle and to control the switch KxM in the closed position at 03b, and keep it in the closed position between 03b and 061'switches KxW being open between 0O and 03b.

[0153] As a variant, when each of the secondary switches comprises a transistor and an antiparallel diode (diode which may be intrinsic to the transistor), the respective antiparallel diodes being able to naturally allow the current to pass when it flows in the opposite direction, the diodes can ensure the first part and / or the last angular part of the conduction phase, and the control signal for maintaining the closed state is applied only to the central part to reduce losses but without requiring great precision in determining the closing and / or opening instant, the angular duration of conduction per diode making it possible to tolerate an error in the angle of application of the opening / closing control.

[0154] Furthermore, advantageously, when each of the primary switches Kxah, Kxa[ comprises a transistor and an antiparallel diode (diode which may be intrinsic to the transistor), the respective antiparallel diodes can naturally let the current flow when it flows in the opposite direction, that is to say between 05 and 2+ for Kxah and between 0 2b and 03 for Kxab During this angular zone of conduction by the reverse diode, it is preferable to apply a closing control order to reduce losses. However, slightly overestimating the closing control angle 05 of Kxah compared to the theoretical angle 05, for example from 0.001 to 0.2 radians, ensures that the closing command does not arrive before the effective angle 05 even if the estimate has a small error in a lower value. For example, applying a command on a mid-angle between 05 and 2ir allows to tolerate an angular error, both negative and positive, of approximately + / -33(2+-05) on the estimate of the closing control angle of Kxah. The same reasoning can be made for the Kxab command which is identical but shifted by jt.

[0155] [Fig.4] illustrates graphs of the evolution of the dynamic piezoelectric voltage V Pxdyn at the terminals of the piezoelectric assembly Px of an elementary piezoelectric converter, driven in a voltage boost mode over a control cycle equal to a resonance period of 2+ of the piezoelectric assembly. The control is carried out at angle 0 evolving over a period between 0 and 2+ radians.

[0156] Note that in the example of [Fig.4], it is considered that the current IL flows in the opposite direction to that illustrated in [Fig.l].

[0157] As illustrated in the graph Hi of [Fig.4], for control in booster mode, the dynamic voltage VPxdyn at the terminals of the piezoelectric assembly takes, during phases at substantially constant voltage, alternately the voltage values ​​Vd, Vb on a first control half-cycle, corresponding to the first resonance half-cycle (angle from 0 to ji), and the voltage values ​​Va, Vc on a second control half-cycle, corresponding to the second resonance half-cycle (angle from + to 2+).

[0158] The graph Hi of [Fig.4] illustrates the control in four phases at substantially constant load and four phases at substantially constant voltage.

[0159] In this elevator type configuration, the control cycle, equal to the resonance period in this example, comprises two half-cycles.

[0160] The first half-cycle comprises two phases F, III' at substantially constant voltage and two phases II', IV' at substantially constant charge.

[0161] In detail, between 0o=0 and 0Ob, during the phase at substantially constant voltage, the dynamic voltage VPxdyn at the terminals of the piezoelectric assembly is equal to Vd.

[0162] During the IF phase at substantially constant load, between 0Ob and 0b the dynamic voltage VPxdyn changes from Vd to Vb.

[0163] During phase III' at substantially constant voltage, between 0i and 02, the dynamic voltage VPxdyn at the terminals of the piezoelectric assembly is equal to Vb.

[0164] Finally, during phase IV' at substantially constant load, between times 02 and 03 =jt, the dynamic voltage value VPxdyn evolves between Vb and Va.

[0165] The second half-cycle also comprises two phases V', VII' at substantially constant voltage and two phases VI', VIII' at substantially constant charge.

[0166] In detail, between 03 and 04, during phase V' at substantially constant voltage, the dynamic voltage VPxdyn is substantially equal to Va.

[0167] During phase VI' at substantially constant load, between 04 and 05, the dynamic voltage VPxdyn changes from Va to Vc.

[0168] During phase VII' at substantially constant voltage, between 05 and 05b, the dynamic voltage value VPxdyn is substantially equal to Vc.

[0169] Finally, during phase VIII' at substantially constant load, between times 05b and 06=2ir, the dynamic voltage VPxdyn goes from Vc to Vd.

[0170] Of course, due to the cyclic repetition, 06=2jr=0o.

[0171] Furthermore, the respective substantially constant voltage phases I', V' are of substantially the same duration, and the respective voltage values ​​are substantially equal and of opposite signs. In other words, ~ ^4 " $3-

[0172] Similarly, the respective substantially constant voltage phases III', VII' are of substantially the same duration, and the respective voltage values ​​are substantially equal and of opposite signs. In other words #2 " ~ $5* " Es In other words, the phases of Hi are symmetrical by translation and sign inversion between the first half-cycle and the second half-cycle of control.

[0174] Graphs H2 and H3 respectively represent the evolution of the potentials Vxbr at the secondary midpoint and the potential Vxar at the primary midpoint, at the terminals of the piezoelectric assembly of the controlled elementary piezoelectric converter.

[0175] The respective potentials Vxar and Vxbr evolve according to the state of opening or closing of the respective switches on the primary side Kxah, Kxai and on the secondary side K-xbh, Kxbb

[0176] Thus, the potential Vxbrest is maintained at Vxbh during phase I', and passes from Vxbh to VxM during phase II' at substantially constant charge, then is maintained at VxW during phases III' to V', and passes from VxW to Vxbh during phase VI' at substantially constant charge, then is maintained at Vxbh during phases VII' and VIII'.

[0177] The potential Vxarest is maintained at Vxa[ during phases I' to III', then passes from Vxa[ to Vxah during phase IV' at substantially constant charge, then is maintained at Vxah during phases V' to VII', and passes from Vxah to Vxa[ during phase VIII'.

[0178] The control diagrams Com'i, Com'2, Com'3 and Com'4 of the primary switches Kxah, Kxa[ and of the switches Kxbh, Kxbi are illustrated over the same control period, from 0 to 2ir.

[0179] By convention, in each control diagram, a high state is representative of the closing state of the corresponding switch and a low state (e.g. 0) is represented indicating the opening state of the corresponding switch.

[0180] Depending on the scenario, the switches are controlled and held in position on command, or can switch naturally into the indicated positions, when they are implemented by diodes.

[0181] Diode mode conduction is illustrated in [Fig.4] by hatched areas.

[0182] The primary switches Kxah and Kxd are controlled in phase opposition, that is to say that the controls are identical but temporally offset by half a cycle. When one is closed, the other is open and vice versa, except during the periods between 02 and 03, corresponding to phase IV' at substantially constant load and between 05b and 06, corresponding to phase VIII' at substantially constant load.

[0183] In the example of [Fig.4], the primary switch Kxah is held in the closed position between 03 and 05b, and in the open position between 0o and 03 and between 05 and 06=0o. In other words, when several control cycles follow one another, the primary switch Kxah is held in the closed position between 03 and 05b of a control cycle, and in the open position between 05b and 03 of the following control cycle.

[0184] The primary switch Kxa[ is held in the closed position between 0O and 02, then in the open position between 02 and 06= 0O.

[0185] Similarly, the secondary switches Kxbh and Kxbi are closed / open in phase opposition, that is to say that the controls are identical but angularly offset by jt. When one is closed, the other is open and vice versa, except during the periods between 0Ob and 0b corresponding to the IF phase at substantially constant load and between 04 and 05, corresponding to the VF phase at substantially constant load.

[0186] As can be seen, during phases II', IV', VF, VIII' at substantially constant load, only one of the switches Kxah, Kxab Kxbh, KxW is in the closed position, the other three being in the open position.

[0187] During phases I', III', V', VII' at substantially constant voltage, one of the primary switches and one of the secondary switches are closed, the other of the primary switches and the other of the secondary switches being open.

[0188] [Fig.5] illustrates an elementary piezoelectric converter control in step-down mode spread over three resonance periods between 0 and 6ir.

[0189] In [Fig.5], the current IL normalized in amplitude between -1 and 1 is represented, over 3 resonance periods, as well as the graph GL' of the evolution of the dynamic voltage VPxdyn at the terminals of the piezoelectric assembly when the converter is controlled by control periods equal to the resonance period, which corresponds to the graph Gide la [Fig.3] repeated over three resonance periods, as well as the control diagrams Comi and Com2 of the corresponding primary switches.

[0190] In the lower part of [Fig.5], the GL* graph of the evolution of the dynamic tension VPxdyn at the terminals of the piezoelectric assembly corresponds to control on a control cycle spread over the three resonance periods, and the control diagrams Commet Com2* of the corresponding primary switches.

[0191] The GL* graph comprises, during the first half-cycle (between 0 and 3jt), a phase during which the potential Vxar is fixed at +Vxah, followed by the first half-period of the graph Gi (between 0 and ji) or half-period DPi; a second half-cycle (between 3ir and 6ir), comprising a phase during which the voltage Vxar is fixed at Vxai, followed by the second half-period of the graph Gi (between ir and 2ir) or half-period DP2.

[0192] The control diagram Comi* illustrates the closing (high state) or opening (“low state”) states of the primary switch Kxah and the control diagram Com2* illustrates the closing (high state) or opening (“low state”) states of the switch Kxai to obtain the voltage Vxah or Vxai at the primary midpoint to carry out the control cycle GL*.

[0193] As can be seen, the command to switch each of the primary switches to close is carried out once during the control cycle comprising P=1+s periods (here s=2), and the command to switch the primary switches to open is carried out once during the control cycle.

[0194] Advantageously, the primary switches are less stressed, and consequently are better preserved over control cycles of the type spread over P resonance periods, with P>1.

[0195] This embodiment is particularly advantageous in a strongly step-down type configuration, a configuration in which the gain, i.e. the ratio between the output voltage and the input voltage, is less than ^2. Indeed, during a strongly step-down conversion, for s=0, the transition from +Vxa to -Vxa or from -Vxa to +Vxa in the primary occupies a significant part of each half-period, thereby limiting the duration of the current exchange phases in the secondary and therefore the output power. Furthermore, a good part of the energy supplied by the primary in Va and Vb is given back to the primary in Vc and Vd, inducing losses. By using a control cycle with s>l, the frequency of polarity changes in the primary is reduced, leaving more time for power transfers to the output of the elementary piezoelectric converter and reducing power returns to the input of the elementary piezoelectric converter.

[0196] [Fig.6] illustrates the control diagrams of the primary switches Kxah and K xai to achieve a step-down type configuration described previously.

[0197] Diagram Diag-1 represents the control over spread control cycles, with possible control ranges, and diagram Diag-2 represents an example of effective control.

[0198] Advantageously, on each half-period, one of the primary switches among K xah and Kxd is alternately in the closed position over an angular duration 0on: a first of the primary switches among Kxah and Kxai being open, a second different from the first of the primary switches among Kxai and Kxah being closed, then the second of the primary switches being open and the first of the primary switches being closed. This duration 0on is also called contact duration. Two phases of duration 0 on are separated by an intermediate phase, of angular duration 0dt during which the primary switches are open, the duration 0dt being also called dead time duration.

[0199] The sum of the duration 0on and the duration 0dt is equal to ji(s+1) over a spread control cycle, so it is sufficient to estimate one of the durations 0on or 0dt to deduce the other by a simple arithmetic calculation.

[0200] For example, the dead time duration 0dt is estimated as explained in more detail below.

[0201] Furthermore, advantageously, when each of the primary switches comprises a transistor and an antiparallel diode (diode which may be intrinsic to the transistor), the respective antiparallel diodes being able to naturally allow the current to pass when it flows in the opposite direction, a time tolerance margin is available for controlling the closing command.

[0202] In other words, the closing command is likely to be carried out between [ 0(ll min? Odt-max]*

[0203] The value 0dt min indicates the minimum duration between the opening of one of the os switches (for example os=02 in the example of [Fig.3]) and the closing of the other switch allowing a ZVS closure to be ensured.

[0204] The value 0dt min indicates the maximum duration between the opening of one of the switches at 0 2 and the closing of the other switch.

[0205] With reference to the example in [Fig.3], the difference 0dt max-0dt min is 03-02b or by symmetry 2ir-05

[0206] Diagram Diag-2 illustrates a scenario with a dead time duration 0dt between 0dt min and 0dt max. During 0dt ​​the two switches Kxah and Kxai are controlled open.

[0207] Advantageously, the duration of dead time is estimated by the control device as a function solely of the frequency f of the control cycle of the converter, the input voltage and the output voltage of the elementary piezoelectric converter.

[0208] In particular in the step-down type configuration in which the output voltage is lower than the input voltage, the control method is particularly simplified, only two adjustment parameters 0dt and the control cycle frequency are controlled to carry out the control of the converter from measurements of the respective input and output voltages.

[0209] For example, in one embodiment, values ​​of dead time duration 0dten as a function of the input voltage values ​​and the control cycle frequency values ​​are previously calculated and stored, for example in a table or any other suitable value storage structure of the electronic control device.

[0210] In another embodiment, the minimum dead time duration value 0dt_min is estimated in real time as a function of the input voltage Vxa, the output voltage Vxb and the control cycle frequency of the converter, from a model of the piezoelectric resonator. Furthermore, the voltages Vxa and Vxb can be derived from the input voltages Vin and output voltages Vout of the converter 40 for a defined converter topology. As an optional addition, a maximum estimation error 0 in max is also determined. For example, the maximum estimation error 0eir_max is determined in advance, as a function of operating conditions, for example temperature.

[0211] In another more sophisticated embodiment, a control loop using an observer that estimates the conduction time of the antiparallel diodes of the first controlled switches is implemented, the dead time duration 0dt being adjusted to achieve a desired conduction time. For example, an observer as described in patent application FR 3 124042 is used.

[0212] [Fig.7] is a block diagram of the main functional blocks of an electronic control device, implementing a method for controlling an electrical energy converter according to the invention.

[0213] The electronic control device implements a block 60 for estimating the duration of dead time 0dt.

[0214] Block 60 takes as input the respective input voltages Vxa and output Vxb, and the control cycle frequency.

[0215] Optionally, an observation parameter Vobs representative of at least one voltage at the closing of one of the switches is also provided as input. The observation parameter Vobs is, for example, a voltage difference before the closing of the switch (or ZVS error), or a measurement of the derivative of the voltage at the closing of the switch, or a measurement of a minimum or maximum voltage envelope at the closing of the switch. In other words, the observation parameter Vobs is representative of the ZVS error, which one seeks to cancel in order to achieve a zero voltage closing (or ZVS).

[0216] In one embodiment, the dead time duration estimation block 60 implements a calculation. All or part of the calculations can be performed upstream and stored in a storage structure, for example a table. Alternatively, the calculations implement an interpolation function. The interpolation function is for example a polynomial, a ratio of polynomials, one or more exponentials, one or more nth roots, which involves the parameters Vin, Vout and f as well as pre-calculated parameters to best restore the dead time.

[0217] In another embodiment, the dead time duration estimation block 60 implements a regulation based on the observation parameter Vobs, for example a regulation loop to cancel the ZVS error.

[0218] In another embodiment, the dead time duration estimation block 60 implements a combination of calculation and regulation, the calculation providing a range of the dead time duration and the regulation refining the dead time duration to tend towards a minimal ZVS error.

[0219] The dead time duration 0dtest supplied to a command generator block 62, which controls the primary switches Kxah, Kxai and optionally the secondary switches Kxbhet Kxbb

[0220] Block 62 also receives as input the control cycle frequency f and optionally, a phase shift q>. This involves shifting the opening of Kxbh beyond 03 and Kx bi beyond 2ir, which naturally also shifts their closing times to ensure their closing at zero voltage (ZVS mode). This shift / phase shift of the conduction phases of Kxbh and Kxbi compared to the case of a simple rectifier type operation induces a reactive power in the secondary, that is to say that a part of the power supplied to the secondary is returned to the piezoelectric assembly. We note q> the phase shift or offset between the effective opening angle and the opening angle that we would have with a simple rectifier operation. In other words, the phase shift q> represents the angular difference between the instant of change of sign of the current and the effective opening of one of the secondary switches.If diodes are used in the secondary, this phase shift q> is considered zero.

[0221] Furthermore, for the same control cycle frequency f, it is possible to modulate the output power by acting on the value of q>.

[0222] This action on q> makes it possible, for example, to avoid certain control cycle frequencies f for which the converter efficiency is degraded. Indeed, the piezoelectric assembly may exhibit spurious modes at certain frequencies which lead to an increase in losses and a change in the behavior of the piezoelectric assembly.

[0223] The commands of the primary switches, and optionally secondary switches, are sent to the electrical energy converter 40, which provides an output voltage Vout and an output current Lut-

[0224] The electrical parameter to be regulated, for example the output voltage, the output current or the output power is transmitted as input to a comparison block 64, which calculates a difference between the electrical parameter to be regulated and a corresponding setpoint parameter.

[0225] The difference between the electrical parameter to be regulated obtained at the output of the electrical energy converter 40 and the corresponding setpoint parameter is transmitted to the input of a correction block 66, which calculates the value of the control cycle frequency f and / or a phase shift value q>.

[0226] From a single error measurement, it is easier to control only one parameter, so, in the case of controlling both f and q> it is advantageous to go through an intermediate parameter, for example [3 then to predefine f and q> according to this common parameter. Thus the corrector corrects this parameter [3 then f and q> are calculated from a function which links them to [3.

[0227] The control cycle frequency impacts the current IL and ultimately the electrical parameter that one seeks to regulate. A lowering of the control cycle frequency allows a reduction in the electrical impedance of the piezoelectric resonator and an increase in the current IL and consequently allows an increase in the electrical parameter to be regulated. Conversely, an increase in the control cycle frequency allows an increase in the impedance of the piezoelectric resonator and a reduction in the output current and consequently allows a reduction in the electrical parameter to be regulated.

[0228] Similarly, the phase q> makes it possible to adjust the reactive power at the secondary and ultimately to adjust the output current and consequently the electrical parameter to be regulated.

[0229] The correction block 66 can therefore act on f or on q> or on both at the same time to regulate the electrical parameter to be regulated.

[0230] Frequency stop values, respectively a first frequency stop fmin and a second frequency stop fmax, may be used to ensure operation on a particular resonance mode of the piezoelectric assembly, further to ensure operation between the resonance and antiresonance frequency of the piezoelectric assembly around the selected resonance mode.

[0231] Likewise phase stop values, respectively a first phase stop <pmin et une deuxième butée de phase <pmax peuvent être utilisée pour garantir la stabilité de fonctionnement.

[0232] By way of non-limiting example, [Fig.8] illustrates a topology of an electrical energy converter 400 comprising an elementary piezoelectric converter as described with reference to [Fig.2], and [Fig.9] a control cycle in voltage step-down mode over a resonance period of the converter 400.

[0233] The electrical converter 400 is configured to convert an input voltage Vin into an output voltage Vout, the voltage Vin being greater than Vout.

[0234] The voltage Vin is applied between two input terminals of potential Vinn and Vinp, and the voltage Vout is supplied between output terminals of respective potentials Voutn and Voutp.

[0235] The converter 400 comprises an elementary piezoelectric converter 2, with a piezoelectric assembly Pb comprising a primary switching branch connected between one of the input terminals of potential Vinp and one of the output terminals of potential Voutp. According to the notations used above: Viah=Vinp and Viai=Voutp.

[0236] The secondary switching branch of the elementary piezoelectric converter is connected between one of the input terminals of potential Vinn and one of the output terminals of potential Voutn. According to the notations used above: Vibh=Voutp and Vibi=Vinn.

[0237] A primary capacitor 52a, of capacitance Cia, is connected in parallel with the primary switching branch, and a capacitor 52b, of capacitance Clb, is connected in parallel with the secondary switching branch.

[0238] The voltage across the piezoelectric element Pi is denoted VP1, between the potentials Viar of the primary midpoint and Vibr of the secondary midpoint.

[0239] [Fig.9] illustrates a four-step voltage step-down mode control cycle of the electrical energy converter 400 of [Fig.8], applying the control method described above.

[0240] [Fig.9] is analogous to [Fig.3] and represents respectively the sinusoids IL and D corresponding respectively to the current ILnormalized between -1 and l (ordinate axis) as a function of the angle over the period [0, 2ir] and the deformation D of the piezoelectric assembly over the same period; the graph GiO of the evolution of the dynamic voltage VPid yn at the terminals of the piezoelectric assembly; the graph Gi2 of the corresponding potentials and Vibr, and the control diagrams Comi, Com2, Com3 and Com4 of the primary switches Kiah, Kiai and of the secondary switches Kibh, KiW.

[0241] The amplitudes Vla and Vlb of the dynamic variations of the voltages and Vibr are:

[0242] [Math. 8] T 7 \al ln^yes V Ifl “ 2 “ 2

[0243] T 7 00^ ihl _ JV vp? ~ 2 — 2 — 2 v out

[0244] In step-down mode control with four voltage levels, as illustrated in [Fig.9], but also in step-up mode (not shown), the dynamic voltage across the piezoelectric assembly varies between the respective voltage values:

[0245] [Math.9] y, - y, -7„ = 17. - yf ' b ' la ' 1b 2 ' m ' yes

[0246] 7--7,-7,--17. 'dr la r 1b 2' w

[0247] 7 - _ 7, + 7„ - .17. + 7 t ' av la ' v 1b 2 v ma

[0248] 7,-7 +7 -17. 'cre the head? rm

[0249] Advantageously, the control cycle illustrated in [Fig.9] allows a great simplification of the control, as described above with reference to Figures 6 and 7.

[0250] Advantageously, the electronic control device and the associated control method achieve a balancing of the electrical energy transfers per half-cycle of control, which makes it possible to reduce the amplitude of energy oscillation within the piezoelectric assembly(s), and consequently to improve the electrical energy transfer performance of the converter.

[0251] In [Fig.8], we have represented at the output of the converter a voltage source, such as a battery. In practice any electrical load can be connected at the output, for example an electronic device, a motor, a resistive load, a DC BUS... In addition, to stabilize the output voltage, advantageously an electrical capacity can be added in parallel with Vout (in practice this is quite essential).

[0252] Similarly, at the input of the converter, we have represented a voltage source of the battery or cell type, but any other source of electrical energy can be used, for example a photovoltaic panel, an AC or DC electrical network, a rectified sine voltage, etc. In addition, to stabilize the input voltage, advantageously an electrical capacity can be added in parallel with Vin.

[0253] The frequency f of the control cycle of the converter imposes the frequency of the mechanical movement of the piezoelectric element(s) 15. However, to obtain the desired effect, the piezoelectric element 15 is typically operated between its so-called series resonance frequency ((OS=1 / V(LC) where L and C correspond to the inductance and capacitance of the resonant branch and the so-called parallel resonance frequency of the piezoelectric element 15 (cop=l / '' / (LCCo / (C+Co))), also respectively called resonance frequency and antiresonance frequency of the piezoelectric element 15. The control cycle frequency (or piloting) is therefore between these two frequencies or over a more restricted range to remain within a predefined impedance range, or to avoid parasitic resonance modes.Furthermore, the control cycle frequency f is typically adjusted between these two frequency limits to adjust the amplitude of the current IL and ultimately regulate the output voltage, current or power via a control loop. It is also possible to change, during operation of the converter, the resonance mode on the resonator, for example switching from a thickness resonance mode to a radial or elongation resonance mode, or switching to a higher resonance mode to address different power ranges with improved efficiency.

[0254] The frequency of the control cycle is equal to the pilot frequency f divided by (1+s).

[0255] In steady state, the control cycle is such that: • Over a half-cycle of control, the total power exchanged by a piezoelectric assembly on the bearings at substantially constant voltage is substantially zero apart from losses. Let Qb be the charge exchanged with the bearing Vb and Qd the charge exchanged with the bearing Vd then QbVb+QdVd=0 which amounts to Qd / Q^Vt / Vd, which makes it possible to define the relative duration of the bearings Vb and Vd and consequently to define the angle 02. By translation of a half-cycle of control and inversion of sign, we deduce the angle 04. • Over a control cycle, the total amount of charge exchanged is substantially zero, apart from charge losses. Let Qa be the charge exchanged with the Va level and Qc the charge exchanged with the Vc level, then Qa+Qb+Qc+Qd=0. This condition allows, at the end of a control cycle, to return to the initial voltage. • All switch closures are performed at zero voltage (ZVS switching). This condition then defines 0b 02b, 03b, 05. • Angle 03=ir and angle 06=2ir

[0256] The control cycle frequency then defines the relationship between the periodic control instants and the above control angles.

[0257] Advantageously, thanks to the invention, the control of an electrical energy converter using at least one piezoelectric resonator is facilitated.

Claims

1. Claims Electronic control device (42) of an electrical energy converter (40), the electrical energy converter comprising at least one elementary piezoelectric converter (2), the or each elementary piezoelectric converter (2) comprising at least one primary switching branch (4a) and at least one secondary switching branch (4b), the primary switching branch being connected between two primary terminals for applying an input voltage of the elementary piezoelectric converter (2), and comprising at least two primary switches (Kxab Kxah) connected in series and connected together at a primary midpoint (10a), the secondary switching branch (4b) being connected between two secondary terminals for supplying an output voltage of the elementary piezoelectric converter, and comprising at least two secondary switches (Kxbb Kxbh) connected in series and connected together at a secondary midpoint (10b),the elementary piezoelectric converter (2) comprising a piezoelectric assembly (22) connected between the primary midpoint (10a) and the secondary midpoint (10b), the electronic control device (42) being configured to control in the or in each elementary piezoelectric converter (2), a switching of the primary (Kxab Kxah) and secondary (Kxbb Kxbh) switches to alternate phases at substantially constant voltage at the terminals of the piezoelectric assembly (22) and phases at substantially constant charge at the terminals of the piezoelectric assembly (22), the electronic control device (42) being characterized in that it is configured to, for a control cycle equal to s+1 resonance periods of the piezoelectric assembly, s being an integer greater than or equal to zero, control at least two phases at substantially constant voltage per half-control cycle,a first control half-cycle comprising a first phase at substantially constant voltage of first voltage value and first duration and a second phase at substantially constant voltage of second voltage value and second duration, a second control half-cycle comprising a third phase at substantially constant voltage of value opposite to the first voltage value and of duration substantially equal to the first duration and a fourth phase, at substantially constant voltage of value opposite to the second voltage value and of duration substantially equal to the second duration.

2. The device of claim 1, wherein one of the first voltage value and the second voltage value is equal to the difference (Vxa-Vxb) or the opposite of the difference (-Vxa+Vxb) between an amplitude of the input dynamic voltage (Vxa) and an amplitude of the output dynamic voltage (Vxb), the amplitude of the input dynamic voltage being equal to half the input differential voltage, the amplitude of the output dynamic voltage being equal to half the output differential voltage, and the other of the first voltage value and the second voltage value is equal to the sum (Vxa+Vxb) or the opposite of the sum (-Vxa-Vxb) of said amplitude of the input dynamic voltage (Vxa) and said amplitude of the output dynamic voltage (Vxb).

3. Device according to claim 1 or 2, configured to control sequences of phases at substantially constant voltage and at substantially constant load per half-cycle of control, the phases of the second half-cycle of control being obtained by translation and inversion of sign of the phases of the first half-cycle of control.

4. Device according to one of claims 1 to 3, configured to control the primary switches (Kxab Kxah) to produce said first, second, third and fourth phases at constant voltage.

5. Device according to claim 4, configured to control one of the primary switches (Kxab Kxah) in the closed position over part of the control cycle and to control the other of said primary switches (Kxah> Kxah) in the closed position with an offset of half a control cycle, each of the primary switches being controlled to close for the same contact duration (0on).

6. Device according to claim 5, comprising a block (60) for estimating a dead time duration (0dt) between the opening command of one of the primary switches and the closing command of the other of said primary switches, said primary switches (Kxab Kxah) being open during the dead time duration.

7. Device according to claim 6, wherein said estimation block is configured to determine said dead time duration as a function of a control cycle frequency (f), the input voltage and the output voltage of the elementary piezoelectric converter (2).

8. Device according to claim 5 or 6, configured to regulate a voltage or current or output power by controlling the control cycle frequency and / or a control phase shift of the secondary switches.

9. Device according to one of claims 6 to 8, configured to control, when the input voltage of the elementary piezoelectric converter is greater than the output voltage of the elementary piezoelectric converter, said dead time duration with a switching tolerance margin.

10. Device according to any one of the preceding claims, configured to further control one of the secondary switches (Kxab Kxah) in the closed position during said first duration, and to control another of said secondary switches in the closed position during said second duration.

11. Electronic electrical energy conversion system (35) comprising an electrical energy converter (40), the electrical energy converter comprising at least one elementary piezoelectric converter (2), the or each elementary piezoelectric converter comprising at least one primary switching branch (4a) and at least one secondary switching branch (4b), the primary switching branch (4a) being connected between two primary terminals for applying an input voltage of the elementary piezoelectric converter, and comprising at least two primary switches (Kxab Kxah) connected in series and connected together at a primary midpoint (10a), the secondary switching branch (4b) being connected between two secondary terminals for supplying an output voltage of the elementary piezoelectric converter,and comprising at least two secondary switches (Kxbb Kxbh) connected in series and linked together at a secondary midpoint (10b), the elementary piezoelectric converter (2) comprising a piezoelectric assembly (22) connected between the primary midpoint and the secondary midpoint, and an electronic control device (42) for said elementary piezoelectric converter (2) in accordance with claims 1 to 9.,

12. Method for controlling an electrical energy converter capable of converting an input voltage (Vin) into at least one output voltage (V out) implemented by an electronic control device according to claims 1 to 10, comprising, for a control cycle equal to s+1 resonance periods of the piezoelectric assemblies, s being a integer greater than or equal to zero, the control of at least two phases at substantially constant voltage per half-control cycle, a first half-control cycle comprising a first phase at substantially constant voltage of first voltage value and first duration and a second phase at substantially constant voltage of second voltage value and second duration, a second half-control cycle comprising a third phase at substantially constant voltage of value opposite to the first voltage value and of duration substantially equal to the first duration and a fourth phase at substantially constant voltage of value opposite to the second voltage value and of duration substantially equal to the second duration.

13. Method according to claim 12, comprising a determination, for each half-cycle of control, of a switching angle in closed position of the switches of one of the first switches.

14. A method according to claim 13, wherein, when said output voltage is lower than the input voltage, the determination of a switching angle is carried out with a tolerance margin, following a dead time duration during which the first switches are open, the method comprising an estimation of said tolerance margin as a function of a frequency of the control cycle, the input voltage and the output voltage.

Images