Three-phase converter, aircraft and associated control method

Abstract

Three-phase converter, aircraft and associated control method The present invention relates to a three-phase converter (12) comprising: an input (13) for an input current; an output (15) for a direct output current, of output voltage (Vdc); a rectifier (30), comprising three branches (32, 34, 36) connected in parallel to the output, each branch comprising a plurality of switches (51, 52, 53, 54, 55, 56); a control unit (80) of the rectifier, comprising: an output voltage regulation module, capable of determining a three-phase control voltage; and a control module configured to control the switches, as a function of the three-phase control voltage. The control unit comprises a protection module configured to determine a characteristic quantity of the input current. The control module is configured to control the rectifier in a passive mode, and in an active mode. Figure for abstract: Figure 2

Description

Title of the invention: Three-phase converter, aircraft and associated control method

[0001] The present invention relates to a three-phase converter, for an aircraft, the converter comprising: - an input for a three-phase alternating input current defined by a three-phase input voltage; - an output for a continuous output current, defined by an output voltage; - a rectifier, comprising three branches connected in parallel to the output, each branch comprising a plurality of switches, a plurality of diodes and a midpoint of each branch being connected to a phase of the input; - a rectifier control unit, comprising: • an intensity sensor, configured to measure input intensities, measured at the input; • a voltage sensor, configured to measure input voltages, measured at the input; • an output voltage regulation module, capable of determining a three-phase control voltage, applied between the midpoint of each branch and a neutral point; and • a control module configured to control the switches, depending on the three-phase control voltage.

[0002] It is known to use converters to convert an alternating and three-phase network electric current from a generator driven by the propulsion of the aircraft into direct output current to supply electrical loads in the aircraft. These converters must be robust to transient states of the network, in particular sudden variations in the frequency or the network voltage.

[0003] The converters generally used are so-called "passive" converters, that is to say that these converters only include elements that do not need to be controlled, such as transformers and diodes. These passive converters are robust, particularly in the event of transient variations in the network current, however, they occupy a large volume, are heavy and generate a high level of harmonic distortion. In addition, it is not possible to regulate the value of the output voltage, which depends solely on the network voltage.

[0004] So-called "active" converters, i.e. converters which include elements which can be controlled, for example transistors, also exist. Active converters make it possible to reduce the volume, weight and harmonic distortion rate compared to passive converters, as well as to regulate the value of the output voltage. However, these converters are more complex and less reliable than their passive equivalents, in particular because their control requires current and voltage measurements as well as a digital processing unit. Furthermore, active converters may not be very robust to transient states of the network.

[0005] Indeed, in order to regulate the output voltage and / or the input current, the control voltage applied by the regulation module is controlled, via the control of the switches. However, the regulation of the output voltage and / or the input current is not instantaneous. Thus, in the event of a sudden variation in frequency or network voltage, the regulation module regulates the output voltage and the input current too slowly compared to the variations, which risks generating an electrical fault such as an overvoltage or an overcurrent, and damaging the converter, or even other equipment connected upstream or downstream of the converter.

[0006] The aim of the invention is then to propose a converter which is robust to variations in network frequency.

[0007] For this purpose, according to the invention, the control unit comprises a protection module configured to determine a characteristic quantity of the input current, and the control module is configured to control the rectifier in a passive mode in which each switch is controlled by the control module to be electrically insulating, when the characteristic quantity of the input current is greater than or equal to a fault threshold, and to control the rectifier in an active mode in which the switches are controlled by the control module as a function of the three-phase control voltage, when the characteristic quantity is less than the fault threshold.

[0008] Thanks to the invention, it is possible to combine the advantages of so-called active converters and so-called passive converters. Indeed, in the event of significant transient variations in the network voltage or the network frequency leading to potential instability of the active control of the rectifier, the latter is controlled in the passive mode, in which it behaves like a passive rectifier, which makes it possible to avoid electrical faults such as overvoltages or overcurrents which could damage the converter or other equipment of the aircraft.

[0009] When the network returns to a stable state and / or the converter can be safely controlled, the rectifier is controlled in active mode, thus allowing it to benefit again from its advantages compared to a passive converter.

[0010] Since transient states causing sudden variations in the frequency or network voltage are rare and of short duration, the time during which the converter is controlled in the passive mode is short compared to the time during which the converter is controlled in the active mode. The invention makes it possible to benefit mainly from the advantages of active converters, namely the low harmonic distortion rate, the possibility of regulating the output voltage and better efficiency.

[0011] According to other advantageous aspects of the invention, the converter comprises one or more of the following characteristics, taken individually or in all technically possible combinations:

[0012] - The control unit comprises a phase-locked loop, configured to estimating an estimated angle, corresponding to an angle between a reference phase, and an input voltage vector obtained from the input voltages, and the conversion module further configured to convert the input voltages into a transformed voltage formed from a direct voltage and a quadratic voltage, from the estimated angle, such that the quadratic voltage is zero, and the characteristic quantity is representative of a quality of the angle estimation.

[0013] - The characteristic quantity is an absolute value of the quadratic voltage.

[0014] - The fault threshold is less than or equal to 30% of an effective voltage value nominal network voltage, preferably less than or equal to 25% of the effective value of the nominal network voltage.

[0015] - When the rectifier is in passive mode and the characteristic quantity of the input current is lower than the fault threshold, the control module is configured to control the rectifier in the active mode if the characteristic quantity remains lower than the fault threshold for a safety period.

[0016] - When the characteristic quantity is greater than or equal to the fault threshold, the control module is configured to control the rectifier in passive mode if the characteristic quantity is greater than or equal to the fault threshold for a confirmation duration.

[0017] - When the control module controls the rectifier in the active mode from passive mode, an output voltage setpoint in the form of a continuous voltage ramp is imposed by the control unit.

[0018] - The safety duration is less than or equal to 10ms.

[0019] - The confirmation duration is less than or equal to 100 ps.

[0020] The invention also relates to an aircraft comprising a propulsion engine, an electrical machine mechanically connected to the engine, and a converter as described above, electrically connected to the electrical machine.

[0021] The invention also relates to a method for controlling a converter as described above, the method comprising at least the following steps:

[0022] - determine the characteristic magnitude of the input current by the module of protection;

[0023] - when the characteristic magnitude of the input current is greater than or equal to at the fault threshold, control the rectifier in passive mode, by the control module; and

[0024] - when the characteristic magnitude of the input current is lower than the threshold of default, control the rectifier in active mode, by the control module.

[0025] 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: - [Fig.l] [Fig.l] is a schematic representation of an electrical circuit of an aircraft according to the invention; - [Fig.2] [Fig.2] is an electrical diagram of a converter according to the invention; - [Fig.3] [Fig.3] is a schematic representation of a control unit included in the converter; - [Fig.4] [Fig.4] is a vector diagram of a three-phase voltage at the input of the converter; - [Fig.5] [Fig.5] is a graphical representation of characteristic signals as a function of time; and - [Fig.6] [Fig.6] is a flowchart of a control method according to the invention.

[0026] [Fig.l] is a schematic representation of an electrical circuit of an aircraft 1. The aircraft 1 is for example an airplane, piloted by a pilot on board the aircraft 1.

[0027] The aircraft 1 comprises a propulsion engine 3 of the aircraft 1, which is for example a turbojet. The engine 3 is mechanically connected to an electrical machine 5 for converting a mechanical torque of the engine 3 into electrical current. The electrical machine 5 is for example an alternator, set in motion by the torque produced by the engine 3, in order to produce the electrical current. The electrical current produced by the electrical machine 5 is alternating and three-phase, with a voltage called the network voltage Vr. An effective value of a nominal network voltage is for example equal to 115 Vrms.

[0028] The output of the electrical machine 5 is electrically connected to a power supply unit 6 to supply it with the produced electrical current. In the following, unless explicitly mentioned, the term "connected" means electrically connected.

[0029] Alternatively, the aircraft comprises several motors 3 and several electrical machines 5, each connected to a motor 3. All of the motors 3 and electrical machines 5 constitute a network of the aircraft.

[0030] The power supply unit 6 is configured to power an actuator 8 from the electric current supplied by the electric machine 5.

[0031] In the example of [Fig.l], an electric flight control system 9, also called FBW, from the English “Fly By Wire” is also connected to the power supply unit 6. The electric flight control system 9 allows the pilot to steer the aircraft 1. The power supply unit 6 advantageously comprises a filter 11, connected to the electric machine 5. The filter 11 is advantageously an electromagnetic interference filter, also called an EMI filter (from the English Electro-Magnetic Interference), for filtering out high parasitic frequencies possibly present in the electric current supplied by the electric machine 5 or emitted by the converter 12.

[0032] The three-phase electric current at the output of the filter 11 is defined by an instantaneous three-phase input voltage Vabc and an instantaneous three-phase input current Iabc. The three-phase input voltage Vabc is formed by three input voltages Va, Vb and Vc, phase-shifted by 2ir / 3 between them. The input voltages Va, Vb and Vc are each taken between a neutral point and a corresponding phase A, B and C. Similarly, the three-phase input current Iabc is formed by three input currents Ia, L and L, phase-shifted by 2ir / 3 between them respectively for the three phases A, B and C.

[0033] The power supply unit 6 comprises a three-phase converter 12, shown in more detail in [Fig.2].

[0034] The converter 12 is configured to convert the three-phase alternating current produced by the electrical machine 5 into direct current. The converter 12 comprises an input 13, connected to the output of the filter 11, and an output 15, connected in the example of [Fig.l] to an inverter 16 of the power supply unit 6. The three-phase input voltage Vabc is thus present at the input 13. The converter 12 has a direct output voltage denoted Vdc.

[0035] The inverter 16 is connected to the output 15 of the converter 12. It is configured to convert the direct current, defined by the output voltage Vdc supplied by the converter 12 into three-phase alternating current, in order to power the actuator 8.

[0036] A control module 18 of the inverter 16 is connected to the output of the electric flight control system 9 and is configured to control the inverter 16, in particular as a function of an action carried out by the pilot on the electric flight control system 9, in order to power the actuator 8 so that it carries out the action desired by the pilot.

[0037] The converter 12 comprises a power stage 27, advantageously comprising three coils 28, at a rate of one coil 28 per phase A, B and C.

[0038] The power stage 27 further comprises a rectifier 30. The power stage 27 is connected between the input 13 and the output 15. In the example of [Fig.2], the rectifier 30 comprises three branches 32, 34, 36. Each branch 32, 34 and 36 comprises a plurality of switches, in the example of [Fig.2], two switches, connected in series and denoted respectively 51, 52; 53, 54 and 55, 56. By switch, we mean a component controllable in switching. Branch 32 comprises switches 51 and 52, branch 34 comprises switches 53 and 54 and branch 36 comprises switches 55 and 56. Each branch 32, 34 and 36 also comprises a plurality of diodes. In the example of [Fig.2], branches 32, 34 and 36 respectively comprise diodes 61, 62, diodes 63, 64 and diodes 65, 66. In the example of [Fig.2], switches 51 to 56 are each associated with a diode, respectively 61, 62, 63, 64, 65 and 66.As seen in [Fig.2], diodes 61 to 66 are arranged in parallel with switches 51 to 56.

[0039] Each branch 32, 34, 36 is also connected in parallel to the output 15.

[0040] The switches 51 to 56 are for example semiconductor switches, such as field effect transistors, or FETs (from the English Field Effect Transistor) or even insulated gate bipolar transistors or IGBTs (from the English Insulated Gate Bipolar Transistor), connected in parallel respectively with the diodes 61 to 66. In a particularly advantageous manner, each switch 51 to 56 and the associated diode 61 to 66 are implemented in a single component, for example in the form of an insulated gate field effect transistor, also called MOSFET (from the English Metal Oxide Semiconductor Field Effect Transistor). The diodes 61 to 66 are in this case so-called intrinsic diodes, integrated into the MOSFET.

[0041] Each branch 32, 34, 36 comprises a midpoint, respectively 72, 74 and 76, between the switches 51, 52; 53, 54 and 55, 56 constituting it, the midpoints 72, 74 and 76 being respectively connected to one of the three phases A, B, C of the input 13. More precisely, the midpoints 72, 74 and 76 are respectively connected to one of the three phases A, B, C of the input 13 at the output of the corresponding coil 28.

[0042] Advantageously, the power stage 27 comprises a capacitor 79, connected in parallel with the branches 32, 34 and 36 and to the output 15.

[0043] The converter 12 also comprises a control unit 80 for the rectifier 30, shown in detail in [Fig. 3]. The control unit 80 is advantageously formed by a digital processing unit. The digital processing unit is for example a microcontroller, an FPGA (Field Programmable Gate Array), or an integrated circuit, such as an ASIC (Application Specific Integrated Circuit).

[0044] The control unit 80 comprises a voltage sensor 82, a current sensor 83, and an output voltage sensor 84 which respectively measure the input voltages Va, Vb and Vc, in other words, the three-phase input voltage Vabc, at the input 13 of the converter 12, the currents Ia, Ib and L, in other words the three-phase current Iabc, at the input 13 and the output voltage Vdc at the output 15.

[0045] The control unit 80 comprises a conversion module 85, connected to the voltage 82 and current 83 sensors. The conversion module 85 more precisely comprises three conversion sub-modules 85a, 85b and 85c, visible in [Fig. 3]. The conversion sub-modules 85a and 85b are respectively configured to convert the three-phase input voltage Vabc into a transformed voltage Vdq and the three-phase current Iabc into a transformed current Idq. The conversion sub-module 85c is configured to convert transformed quantities into three-phase quantities, as will be detailed later.

[0046] The transformed voltage Vdq is formed from a direct voltage Vd and a quadratic voltage Vq. The transformed voltages Vd and Vq are expressed in a rotating frame called the dq frame, the dq frame having a first axis called the direct axis, and a second axis called the quadratic axis, perpendicular to the direct axis. The same applies to the transformed current Idq. Advantageously, the conversion of the three-phase input voltage Vabc into transformed voltage Vdq and of the three-phase current Iabc into transformed current Idq is carried out by Park transform.

[0047] In order to carry out the Park transform, the three-phase input voltage Vabc is represented in a reference frame whose fixed axes offset from each other by 2ir / 3 correspond to phases A, B, C in the form of an input voltage vector y, represented in Figure 4. The projection of the input voltage vector y on axes A, B and C is respectively equal to the input voltages Va, Vb and Vc. Thus, the input voltage vector y is a rotating vector, forming an angle 0 with phase A, chosen as reference. The angle 0 is equal to the phase eut, co being the pulsation of the network.

[0048] The direct axis is chosen to be aligned with the input voltage vector y, that is, the direct axis and the A axis are offset by the angle 0.

[0049] To perform the Park transform, it is necessary to know the angle 0. For this purpose, advantageously, the control unit 80 comprises a phase-locked loop 87, also called PLL (from the English “Phase Locked Loop”), connected to the voltage sensor 82. Advantageously, the PLL 87 integrates a proportional-integral corrector, or PI. The phase-locked loop 87 is configured to estimate the angle 0 in the form of an estimated angle 0est and the pulsation co in the form of an estimated pulsation coest from the input voltages Va, Vb and Vc measured by the voltage sensor 82 and is connected to the conversion module 85 to provide it with the angle estimated 0est. The transformed current Idq and the transformed voltage Vdq are respectively calculated by the conversion module 85 from the three-phase input current Iabc and the three-phase input voltage Vabc, and from the estimated angle 0est.

[0050] A module 91 for regulating the output voltage Vdc and the three-phase input current Iabc is connected to the conversion module 85 and to the phase-locked loop 87. The regulation module 91 is configured to determine a transformed control voltage V*dq, so that the control unit 80 determines a three-phase control voltage V*abc, formed of control voltages V*a, V*b and V*c specific to each phase, and regulates the input currents Ia, Ib, Ic and the output voltage Vdc. For each branch 32, 34, 36, the control voltage V*a, V*b and V*c is the voltage between their respective midpoints 72, 74 and 76 and a neutral point.

[0051] The regulation module 91 receives as input a voltage setpoint V*dc. The voltage setpoint V*dc is determined according to the needs of the application, and is for example equal to 340V. In the example of Figures 2 and 3, taking into account the topology of the power stage 27, the voltage setpoint V*dc is greater than the maximum peak voltage between phases of the network voltage Vr.

[0052] The regulation module 91 is connected to the output voltage sensor 84 to receive the output voltage Vdc, to the conversion module 85, more precisely to the conversion sub-module 85b to receive the transformed intensity Idq, and to the phase-locked loop 87 to receive the estimated pulsation coest. The regulation module 91 is configured to determine a transformed control voltage V*dq from the voltage setpoint V*dc, the output voltage Vdc, the transformed intensity Idq provided by the conversion module 85, and the estimated pulsation coest.

[0053] The regulation module 91 comprises the conversion sub-module 85c, which receives the transformed control voltage V*dq supplied by the regulation module 91 to convert it into a three-phase control voltage V*abc. This operation is advantageously carried out by inverse Park transform. For reasons of readability, the conversion sub-module 85c is shown in [Fig.3] as being connected to the regulation module 91.

[0054] The converter 12 also comprises a control module 96, connected to the regulation module 91, more precisely, to the conversion sub-module 85c which is included in the regulation module 91.

[0055] The control module 96 is configured to receive the three-phase control voltage V*abc and to control the switches 51 to 56 in an on state or in a blocked state.

[0056] The control module 96 has two operating modes, namely an active mode and a passive mode. The choice of active mode or passive mode is made according to a validity signal 0_valid, described later.

[0057] In the active mode, the control module 96 controls the switches 51 to 56 in pulse width modulation, or PWM (from the English Pulse Width Modulation). For this, the control module 96 translates the three-phase control voltage V*abc received from the conversion sub-module 85c into a PWM command. Advantageously, the translation of the three-phase control voltage V*abc into PWM command is carried out by the space vector method, or SVPWM (from the English Space Vector Pulse Width Modulation). Thus, the control voltage V*a, V*b and V*c is applied to the respective midpoint 72, 74 and 76 of each branch 32, 34, 36, which makes it possible to regulate the three-phase input current Iabc and therefore the output voltage Vdc.

[0058] In the passive mode, all the switches 51 to 56 are controlled by the control module 96 in the off state. When the switches 51 to 56 are all controlled in the off state, the current can only flow in the diodes 61 to 66. The rectifier 30 is then equivalent to a six-diode rectifier bridge. The DC output voltage Vdc at output 15 is then equal to a free DC voltage Vb which depends on the network voltage Vr, and cannot be modified as long as the rectifier 30 is in the passive mode. In the example of Figures 3 and 5, taking into account the topology of the power stage 27, the free voltage Vi is less than or equal to the maximum peak voltage between phases of the network voltage Vr. In other words, the voltage setpoint V*dc is greater than the free voltage Vb

[0059] The control unit 80 further comprises a protection module 98, connected to the conversion module 85 and to the control module 96 for choosing the operating mode of the rectifier 30.

[0060] The protection module 98 receives the transformed voltage Vdq from the conversion module 85a. The protection module 98 is configured to determine a characteristic quantity E of the input current and to compare it to a fault threshold Eth. The characteristic quantity E is representative of a quality of the estimation of the angle 0. In the example of the figures, the characteristic quantity E is an absolute value IVql of the quadratic voltage Vq, obtained following the conversion of the three-phase input voltage Vabc. The fault threshold Eth is advantageously less than 30% of an effective value of the nominal network voltage. In particular, the fault threshold Eth is less than or equal to 25% of the effective value of the nominal network voltage, here, equal to 25% of the effective value of the nominal network voltage. Alternatively, the characteristic quantity E is another characteristic quantity of the input current.

[0061] When the characteristic magnitude E of the input current is lower than the fault threshold Eth, here when the absolute value IVql of the quadratic voltage Vq is lower than the fault threshold E*, this means that the estimation of the angle 0 is of good quality, that is to say, in the example of [Fig.3], that the estimated angle 0est and the angle 0 are substantially equal. The protection module 98 emits a validity signal 0_valid in a high state, received by the control module 96, which controls the rectifier 30 in the active mode.

[0062] When the characteristic magnitude E of the input current is greater than or equal to the fault threshold Eth, this means that the estimation of the angle 0 is of poor quality, that is to say in the example of [Fig.3], that the estimated angle 0est and the angle 0 are significantly different. The protection module 98 emits the validity signal 0_valid in a low state, which is received by the control module 96 and the control module 96 controls the rectifier 30 in the passive mode.

[0063] An example of operation of the converter 12 is shown in [Fig.5], opposite [Fig.6].

[0064] When the network voltage Vr supplied by the electrical machine 5 to the filter 11 is stable, in other words, when the three-phase input voltage Vabc is stable, that is to say of effective value and of constant frequency or varying according to a dynamic range clearly lower than the bandwidth of the phase-locked loop 87, the phase-locked loop 87 correctly estimates the angle 0, that is to say the estimated angle 0est and the angle 0 are substantially equal.

[0065] As illustrated during phase P, when the converter 12 is in a state 102 in which the rectifier 30 is in the active mode, the protection module 98 determines the characteristic quantity E of the input current, by calculating the absolute value IVql of the quadratic voltage Vq. In the example of [Fig.5], during phase P, the quadratic voltage Vq is zero. The protection module 98 compares the characteristic quantity E with the fault threshold Eth during a step 104. Advantageously, step 104 is carried out at each calculation cycle. The absolute value IVql of the quadratic voltage Vq is lower than the fault threshold Eth during phase P. The protection module 98 then emits the validity signal 0_valid in the high state. The control module 96 receives the validity signal 0_valid in the high state and maintains the rectifier 30 in the active mode.

[0066] The control module 96 also receives the three-phase control voltage V*abc from the conversion sub-module 85c. The control of the switches 51 to 56 by the control module 96 as a function of the three-phase control voltage V*abc makes it possible to regulate the three-phase input current Iabc and the direct voltage Vdc which is then equal to the voltage setpoint V*dc.

[0067] In the event of a transient disturbance, corresponding to the instant Q of [Fig. 5], the network voltage Vr supplied by the electrical machine 5 to the converter 12 is suddenly modified. The disturbances are, for example, interruptions in the current, phase losses, sudden variations in the amplitude or frequency of the network voltage Vr.

[0068] In the example of [Fig. 5], a frequency of the network voltage Vr supplied by the electrical machine 5 is halved in a few microseconds. In this case, a frequency of the input voltages Va, Vb, Vc, are also halved in a few microseconds. In [Fig. 5], only the input voltage Va is shown. The phase-locked loop 87 has a non-zero response time to variations in the input voltage Vabc, and more particularly to variations in the frequency of the input voltage Vabc. This response time is for example between 1 and 10 ms. In this case, the estimated angle 0est and the estimated pulsation coest no longer correspond to the angle 0 and the pulsation co respectively, because the response time of the phase-locked loop 87 is significant compared to the speed of variation of the frequency. The error in the estimated angle 0est and the estimated pulsation coest disturbs the regulation module 91 and the conversion module 85.This can cause electrical faults such as overvoltages or overcurrents that can damage the converter 12 or other equipment of the aircraft 1. During this period, the quadratic voltage Vq is non-zero, more precisely, in the example of [Fig.5], the quadratic voltage Vq becomes negative.

[0069] The absolute value IVql of the quadratic voltage Vq is determined to be greater than or equal to the fault threshold Eth by the protection module 98 in step 104.

[0070] The control unit 80 then advantageously executes a step 105, during which the protection module 98 determines whether the absolute value IVql of the quadratic voltage Vq remains greater than or equal to the fault threshold Eth for a confirmation duration Tc. The confirmation duration Tc is for example less than or equal to 100 ps. If the absolute value IVql of the quadratic voltage Vq remains greater than or equal to the fault threshold Eth for the entire confirmation duration Tc, in other words, if the malfunction persists for the entire confirmation duration Tc, the control unit 80 performs a step 108 in which the validity signal 0_valid emitted by the protection module 98 goes to the low state and the rectifier 30 is controlled by the control module 96 in the passive mode from the active mode, corresponding to the instant R in [Fig.5].The direct voltage Vdc at the output of the converter 12 decreases from the voltage setpoint V*dc to the free voltage Vb which is 280 V in [Fig.5].

[0071] If the absolute value IVql of the quadratic voltage Vq becomes lower than the fault threshold Eth while the confirmation time Tc has not elapsed, the protection module returns to state 102.

[0072] Verifying that the malfunction persists for the confirmation duration Tc before controlling the rectifier 30 in passive mode makes it possible to filter the measurement noise and to control the rectifier 30 in passive mode only when necessary.

[0073] When the rectifier 30 is in passive mode, the protection module 98 again compares the characteristic quantity E to the fault threshold Eth during a step 110.

[0074] As long as the estimated angle 0est does not correspond to the angle 0, the absolute value IVql of the quadratic voltage Vq is greater than or equal to the fault threshold Eth. The protection module 98 performs step 108 again and continues to emit the validity signal 0_valid in the low state and the control module 96 maintains the rectifier 30 in the passive mode.

[0075] After a few milliseconds, corresponding to the response time of the phase-locked loop 87, the latter responds to the new frequency of the three-phase input voltage Vabc, and estimates the estimated angle 0est correctly, that is to say that the estimated angle 0est and the angle 0 correspond. The absolute value IVql of the quadratic voltage Vq becomes lower than the fault threshold Eth, corresponding to the instant S in [Fig. 5]. The protection module 98 then advantageously executes a step 111, during which the protection module 98 determines whether the absolute value IVql of the quadratic voltage Vq remains lower than the fault threshold Eth for a safety duration Ts. The safety duration Ts is here of the order of 10 ms, but alternatively is less than 10 ms.If the safety time Ts has elapsed and the absolute value IVq I of the quadratic voltage Vq has remained below the fault threshold E* throughout the safety time Ts, the control unit 80 performs a step 114 in which the signal 0_valid emitted by the protection module 98 goes to the high state and the rectifier 30 is controlled by the control module 96 in the active mode from the passive mode, corresponding to the instant T in [Fig. 5]. Otherwise, the protection module 98 performs step 108 again and continues to emit the validity signal 0_valid in the low state.

[0076] The safety duration Ts makes it possible to ensure that the phase-locked loop 87 estimates the estimated angle 0est correctly and stably before controlling the rectifier 30 in the active mode.

[0077] In order to avoid instabilities caused by the transition from passive mode to active mode, due to starting under load or to a capacitive current draw, advantageously, the voltage setpoint V*dc is modified in step 116 and takes the form of a continuous voltage ramp at the moment when the rectifier switches to active mode, in other words at time T in [Fig.5]. This voltage ramp consists, for example, of imposing a linearly increasing voltage from the free voltage Vi to 340 V over a few tens of milliseconds. Once the continuous voltage Vdc is equal to 340 V, the voltage setpoint V*dc becomes constant again, for example equal to 340 V, and the converter 12 returns to state 102.

[0078] In a variant not shown, the rectifier 30 is a rectifier with a different topology, for example a three-phase Vienna rectifier.

[0079] In a variant not shown, the aircraft 1 comprises several inverters 16, each inverter being connected to a separate actuator.

[0080] In a variant not shown, the aircraft 1 does not include an electric flight control system 9 or a control module 18 for the inverter 16. In this case, a set of loads is then connected directly to the output 15 of the converter 12, in which case the set of loads is powered by a direct current.

[0081] The converter 12 thus provides the advantages of active rectifiers, namely benefiting mainly from the low harmonic distortion rate, the possibility of regulating the output voltage and better efficiency, while limiting the risks due to instabilities in the network voltage Vr supplied by the electrical machine 5. The converter 12 makes it possible to benefit from a power density greater than that of the passive rectifiers currently used. Thus, the converter 12 is of smaller volume and mass than a passive converter delivering equivalent power. It is therefore more advantageous to use the converter 12 in applications requiring high power, for example greater than 5 kW.

Claims

1. Claims Three-phase converter (12), for an aircraft (1), the converter (12) comprising: - an input (13) for a three-phase alternating input current defined by a three-phase input voltage (Vabc); - an output (15) for a continuous output current, defined by an output voltage (Vdc); - a rectifier (30), comprising three branches (32, 34, 36) connected in parallel to the output (15), each branch (32, 34, 36) comprising a plurality of switches (51, 52, 53, 54, 55, 56), a plurality of diodes (61, 62, 63, 64, 65, 66) and a midpoint (72, 74, 76) of each branch (32, 34, 36) being connected to a phase (A, B, C) of the input (13); - a control unit (80) of the rectifier (30), comprising: • an intensity sensor (83), configured to measure input intensities (Ia, Ib, Ic), measured at the input (13); • a voltage sensor (82), configured to measure input voltages (Va, Vb, Vc), measured at the input (13); • a regulation module (91) of the output voltage (Vdc), capable of determining a three-phase control voltage (V*abc), applied between the midpoint (72, 74, 76) of each branch (32, 34, 36) and a neutral point; and • a control module (96) configured to control the switches (51, 52, 53, 54, 55, 56), as a function of the three-phase control voltage (V*abc), characterized in that: - the control unit (80) comprises a protection module (98) configured to determine a characteristic quantity (E) of the input current; and - the control module (96) is configured to control the rectifier (30) in a passive mode in which each switch (51, 52, 53, 54, 55, 56) is controlled by the control module (96) to be electrically insulating, when the characteristic quantity (E) of the input current is greater than or equal to a fault threshold (Eth), and to control the rectifier (30) in an active mode in which the switches (51, 52, 53, 54, 55, 56) are controlled by the control module (96) as a function of the three-phase control voltage (V*abc), when the characteristic quantity (E) is less than the fault threshold (Eth).

2. Converter (12) according to claim 1, wherein the control unit (80) comprises a phase-locked loop (87), configured to estimate an estimated angle (0est), corresponding to an angle (0) between a reference phase, and an input voltage vector (y) obtained from the input voltages (Va, Vb, Vc), and the conversion module (85) further configured to convert the input voltages (Va, Vb, Vc) into a transformed voltage (Vdq) formed of a direct voltage (Vd) and a quadratic voltage (Vq), from the estimated angle (0est), such that the quadratic voltage (Vq) is zero, and the characteristic quantity (E) is representative of a quality of the estimation of the angle (0).

3. Converter (12) according to claim 2, in which the characteristic quantity (E) is an absolute value (1 Vql) of the quadratic voltage (Vq).

4. Converter (12) according to claim 3, wherein the fault threshold (Eth) is less than or equal to 30% of an effective value of the nominal network voltage, preferably less than or equal to 25% of the effective value of the nominal network voltage.

5. Converter (12) according to any one of the preceding claims, wherein when the rectifier (30) is in the passive mode and the characteristic quantity (E) of the input current is lower than the fault threshold (Eth), the control module (96) is configured to control the rectifier (30) in the active mode if the characteristic quantity (E) remains lower than the fault threshold (Eth) for a safety duration (Ts).

6. Converter according to any one of the preceding claims, wherein when the characteristic quantity (E) is greater than or equal to the fault threshold (Eth), the control module (96) is configured to control the rectifier (30) in passive mode if the characteristic quantity (E) is greater than or equal to the fault threshold (Eth) for a confirmation duration (Tc).

7. Converter (12) according to any one of the preceding claims, wherein when the control module (96) controls the rectifier (30) in the active mode from the passive mode, an output voltage setpoint (V*dc) in the form of a continuous voltage ramp is imposed by the control unit (80).

8. Converter according to claim 5, wherein the safety duration (Ts) is less than or equal to 10ms.

9. Converter according to claim 6, wherein the confirmation duration (Tc) is less than or equal to 100 ps.

10. Aircraft (1), comprising a propulsion engine (3), an electrical machine (5) mechanically connected to the engine (3), and a converter (12) according to any one of the preceding claims, electrically connected to the electrical machine (5).

11. Method for controlling a converter (12) according to any one of claims 1 to 9, the method comprising at least the following steps: - determining (102) the characteristic quantity (E) of the input current by the protection module (98); - when the characteristic quantity (E) of the input current is greater than or equal to the fault threshold (Eth), controlling (108) the rectifier (30) in the passive mode, by the control module (96); and - when the characteristic quantity (E) of the input current is less than the fault threshold (Eth), controlling (114) the rectifier (30) in the active mode, by the control module (96).

Images