Classifications

• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
  • H02M7/00—Conversion of ac power input into dc power output; Conversion of dc power input into ac power output
  • H02M7/42—Conversion of dc power input into ac power output without possibility of reversal
  • H02M7/44—Conversion of dc power input into ac power output without possibility of reversal by static converters
  • H02M7/48—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode
  • H02M7/53—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal
  • H02M7/537—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only, e.g. single switched pulse inverters
  • H02M7/5387—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only, e.g. single switched pulse inverters in a bridge configuration
  • H02M7/53871—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only, e.g. single switched pulse inverters in a bridge configuration with automatic control of output voltage or current

Definitions

• the present invention relates to a current control circuit in an electrical control member or the voltage across said electrical control member.
• a particularly interesting application of the invention lies in the field of current control of electrical machines with variable inductance such as the actuators used for so-called electromagnetic valves ("camless" system in English) in motor vehicles.
• Electromagnetic actuation of the valves requires electrical energy. This electrical energy is taken from the vehicle's on-board network.
• the performance of the electromagnetic valve system involves the minimization of the electrical energy consumption on the on-board system. In fact, the power available at the output of the crankshaft is equal to the total power developed by the heat engine minus the power required for the proper functioning of its auxiliaries (actuation of the valves, drives of the water and oil pumps, etc.).
• H-power bridges also called “quadrant” bridges (or “full-bridge chopper”) single-phase or polyphase depending on the structure of the machine.
• quadrant bridge means a bridge control that works on all quadrants of the voltage-current characteristic.
• the control electronics control the current to a setpoint by applying to the power bridges, a PWM type control (Pulse Width Modulation) fixed frequency.
• PWM type control Pulse Width Modulation
• a source of DC voltage 9 (it may be for example a battery, a DC-DC power converter DCDC or a DC ACDC) connected to the first and second terminals 5 and 6 of the bridge 1,
• an electrical control member 10 (such as an inductive load with variable inductance) making it possible to control an actuator and connected between the third and fourth terminals 7 and 8 of the bridge circuit 1,
• the bridge 1 thus comprises two arms B1 and B2 respectively formed by the switches Ci and C 3 in series and the switches C 2 and C 4 in series.
• the power switches Ci to C 4 may be metal oxide semiconductor field effect transistor (MOSFET) transistors or IGBT transistors (Insulated Gate Bipolar Transistor).
• MOSFET metal oxide semiconductor field effect transistor
• IGBT Insulated Gate Bipolar Transistor
• the valve supervisor calculates a voltage setpoint V * that must be applied to the terminals of the magnetic circuit.
• the pulse width modulation strategy translates the voltage setpoint into the closing / opening order of the electronic switches of the power electronics (here to the numbers of four). Power electronics applies the controls of the MLI strategy while respecting its own constraints (time management dead, on-call time, etc.).
• the determination of the temporal evolutions of the switching functions Sc 1 and Sc 2 is done by comparing the value of the normalized voltage setpoint v * with a single triangular carrier V p (t) of frequency fiuiu-
• the triangular function can be any taking values between a minimum value V pm j n and maximum V pma ⁇ .
• the value of the normalized target voltage v * is then: v * ⁇ F pma ⁇ - ⁇ min ( y * _ U ⁇ + ⁇
• FIG. 2 graphically represents the determination of the switching functions Sci (t) and Sc 2 (Q.
• the first curve represents the evolution as a function of time of the normalized target voltage v * and of the triangular carrier V p (t).
• the second curve representing the evolution of Sci as a function of time
• the third curve complementary to the second, representing Sc 2 as a function of time.
• the fourth curve represents the evolution of the voltage U ac t across the terminals of the magnetic circuit 10 which varies from + UD C to -U D c and is equal to an average V * over a switching period f M ij. Since the voltage U act can only take two distinct values, it is called a two-state PWM strategy.
• FIG. 3 illustrates the commutations induced by this type of PWM strategy on a four quadrant bridge 11 comprising four switches AH, BH, AL and BL respectively identical to the transistors C2, C1, C4 and C3 of the bridge 1 as represented in FIG. the control of a load 10.
• the bridge 11 has two possible states:
• a first difficulty concerns switching losses.
• a switch changes state (passage from open to closed or passage from closed to open), it is the seat of losses due to the simultaneous presence of current flowing through it and a voltage at its terminals.
• the energy then dissipated depends on the value of the cut current I ac t, the DC voltage UD C and the speed of the switching (these switching times are for example adjusted by the value of the gate resistance of the MOSFET transistors used ).
• cutting period there are two openings and two closures regardless of the direction of the current on one of the two switches of each of the bridge arms. This double switching per switching period of course leads to losses all the more important as the frequency will be high. Note that it is important to reconcile performance and bandwidth.
• the frequency MLI is usually several tens or even hundreds of kiloherts. At these high frequencies, switching losses predominate over other conduction losses. It will be noted that, during the transition from a magnetization state to a demagnetization state, on the four commutations per switching period, two are hard commutations and two are soft commutations: in other words, we start with opening the two transistors initially closed (hard switching) then after a dead time, it closes the two initially open transistors (soft switching). This prevents two switches of the same arm are simultaneously closed. During idle time (before soft switching), the intrinsic diodes (so-called "freewheeling") at the MOSFET switches conduct and thus maintain the MOSFET potential close to zero during smooth switching.
• the breakdown of the voltage generates high frequency harmonics which induce losses in the electrical machines.
• These electrical machines are generally made of magnetic materials conducive to eddy currents (Fer-Si for example). Induction generates a voltage induced in the plates which, according to their resistivity, creates sometimes large eddy currents. Although the sheets are finely cut and isolated from each other, the currents circulating there generate losses by Joule effect.
• the fourth curve of FIG. 2 represents the voltage U ac at the terminals of the load 10 as represented in FIG. 1 with a voltage source 9 having a value U D.
• the control of the bridge is of the type MLI with a duty cycle ⁇ .
• the voltage UDC with voltage drops due to the resistors of the switches, is applied to the load 10.
• the rms value (Uf 0U cauit) of the voltage applied to the load 10 is equal to the voltage U D c (which represents the peak voltage of the voltage U ac t):
• Uf 0U cauit UDc- " -
• control electronics and more particularly its switches generate high frequency common mode currents due to cutting.
• the load generally has a capacitive coupling with respect to the earth.
• Common mode currents are thus generated and they loop through the power supply.
• These high-frequency current loops are responsible for the electromagnetic radiation that may have an impact in terms of compliance with current EMC (Electromagnetic Accounting) standards.
• the present invention aims to provide a "four quadrant" bridge circuit for controlling the current in an electrical control member or the voltage across said control member, said circuit making it possible economically to reduce the losses by switching in the power switches, to reduce the losses related to eddy currents in the magnetic circuit and to overcome the aforementioned problems of EMC.
• the invention proposes a circuit for controlling the current in an electrical control device or the voltage across said electrical control device, said circuit comprising:
• pulse width modulation control means of at least two of said four switches said source of power being connected between a first terminal and a second terminal of said bridge, said electrical control member being connected between a third terminal and a fourth terminal of said bridge, the first switch being connected between said first terminal and said third terminal of said bridge, the second switch being connected between said first terminal and said fourth terminal of said bridge, the third switch being connected between said third terminal and said second terminal of said bridge, the fourth switch being connected between said fourth terminal and said second terminal of said bridge, said circuit control device being characterized in that it has:
• a second state in which said second and third switches are closed and said first and fourth switches are open at least one of the following two states: a third state in which said third and fourth switches are closed and said first and second switches are open, a fourth state in which said first and second switches are closed and said third and fourth switches are open; said pulse width modulation control means authorizing:
• the control circuit can manage three or four states of the bridge "four quadrants" (ie the circuit allows com- ⁇ mutate on three or four different states) controlled by a strategy MLI.
• the MLI strategy proposed is said to be tri-state because the voltage VL applied across the load (electrical control member) is based on three levels: UDC, UDC and UDC (UDC UDC refers to the voltage delivered by the source of power). food).
• UDC UDC refers to the voltage delivered by the source of power). food.
• the introduction of a third state saves hard switching (and soft switching).
• the circuit according to the invention allows a transition from the "magnetization" state to a state called "magnetization”.
• the rms value is very different between a conventional four-quadrant bridge control that switches the two-state bridge and the proposed control according to the invention that switches the bridge between three or four states.
• the three or four state control applies less voltage to control the same current, ie, to provide the same average voltage.
• the actuator controlled by the electrical member has a parasitic capacitance relative to the chassis. The breakdown of the voltage generates high frequency harmonics at the control electronics that return through the earth. These are common mode disturbances. EMC standards limit this level of noise.
• the voltage generated by the 3-state control generates fewer harmonics than the two-state one. A reduction of 6dB common mode currents is thus achieved.
• the system according to the invention may also have one or more of the features below, considered individually or in any technically possible combination.
• control means by modulation of pulse widths allow:
• a first so-called negative half-wave phase comprising transitions from said first state to said third state and transitions from said third state to said first state
• said first and third switches forming a first arm, said amplitude arm, switching at a frequency said clipping and said second and fourth transistors forming a second arm, said sign arm, being respectively closed and open
• a second so-called positive half-wave phase comprising transitions from said second state to said fourth state and transitions of said fourth state to said first state
• said first and third switches switching to said so-called switching frequency and said second and fourth switches being respectively open and closed
• switching said second and fourth switches of said sign arm ensuring the transition between said halfwaves and positive to a f a frequency lower than said switching frequency.
• control circuit comprises a shunt resistor connected in series between said electrical control member and said fourth terminal.
• control circuit preferably comprises an operational amplifier, the terminals of said shunt resistor forming the inverting and non-inverting inputs of said operational amplifier.
• control circuit according to the invention comprises:
• control means comprise:
• means for passing between the positive and the negative half cycles implement the following state machine: when said bridge is in its positive half cycle and the duty cycle ⁇ is canceled, the machine detects a change of state and passes alternately negative. o When said bridge is in its negative half cycle and cyclic ratio ⁇ is equal to 100%, the machine detects a change of state and switches to positive alternation.
• said power source delivers a DC + DC voltage
• said control circuit comprising:
• control means for applying to said electrical control unit a mean voltage equal to (CC-I) XUDC during said negative half-cycle.
• said control means comprise:
• means for generating a pulse width modulation signal having a duty ratio ⁇ means for converting said pulse width modulation signal having a cyclic ratio ⁇ to a pulse width modulation signal having a duty cycle -l ⁇ ;
• said power source delivers a DC + DC voltage
• said control circuit comprising means for applying to said electrical control device a mean voltage equal to the product (2 ⁇ -1) ⁇ D C during said positive half cycle and during said negative alternation.
• said means for converting said pulse width modulation signal having a duty cycle ⁇ into a Pulse width modulation signal having a duty cycle or '
• 2ûr-1] comprises:
• said means for effecting the subtraction between a duty ratio signal equal to 50% and said duty cycle signal ⁇ are means performing an exclusive OR logic function having on its two inputs respectively said duty cycle signal equal to 50% and said duty cycle signal ⁇ .
• said means for doubling the signal obtained by said subtraction comprise at least one counter.
• control means are included in a programmable logic circuit.
• said switches are MOSFET transistors.
• the present invention also relates to a use of the control circuit according to the invention for an electrical member formed by an inductive load with variable inductance.
• the electrical member is included in an actuator provided with an actuated part, said electrical member controlling said actuated part in displacement.
• said actuator is an electromagnetic valve actuator of a motor vehicle.
• FIG. 1 is a simplified schematic representation of the electronic structure of a four-quadrant bridge illustrating the state of the art
• FIG. 2 graphically represents the determination of the switching functions Sc 1 (t) and Sc 2 (t) of the switches of a four-quadrant bridge as represented in FIG. 1
• FIG. 3 illustrates the switches induced by a PWM strategy according to the state of the art on a four-quadrant bridge;
• FIG. 4 illustrates the states of a control circuit according to a first embodiment of the invention
• FIGS. 5 and 6 illustrate the evolution as a function of time of the voltage applied across a load such as the load of FIG. 4 and the current in this load, respectively in the case of a PWM control according to FIG. state of the art using two states and in the case of a control circuit according to the invention using three states;
• FIG. 7 represents a circuit making it possible to carry out the direct measurement of the charging current
• FIG. 8 illustrates the states of a control circuit according to a second embodiment of the invention
• FIG. 9 illustrates the states of a control circuit according to a third embodiment of the invention incorporating a shunt resistor
• FIG. 10 illustrates the evolution of the measured current and the charging voltage as a function of time for a circuit as shown in FIG. 9;
• FIG. 11 shows a load pump circuit of a switch used in a control circuit according to the invention
• FIG. 12 represents a control loop used for a control circuit according to the prior art
• FIG. 13 illustrates the construction of a pulse width modulated signal used constructed according to the control loop of FIG. 12;
• FIG. 14 represents the average voltage seen by the load in the case of the control loop of FIG. 12;
• FIG. 15 illustrates a first embodiment of a regulation loop used in a four-state control circuit according to the invention;
• FIG. 16 represents a state machine for implementing the regulation loop according to FIG. 15;
• FIG. 17 represents the average voltage seen by the load in the case of the control loop of FIG. 15;
• FIGS. 18 to 21 show the measured current, the current setpoint, the PWM signal, the sign signal and the voltage across the load as a function of time in different configurations of the control loop according to FIG. 15.
• FIG. 22 illustrates a second embodiment of a regulation loop used in a four-state control circuit according to the invention
• FIG. 23 represents an embodiment of the subtraction means used in the regulation loop of FIG. 22;
• FIG. 24 represents an embodiment of the doubling means used in the regulation loop of FIG. 22;
• FIG. 25 represents a state machine for implementing the regulation loop according to FIG. 22;
• FIG. 26 represents the average voltage seen by the load in the case of the control loop of FIG. 22.
• the common elements bear the same reference numbers.
• FIG. 4 illustrates the three states of a control circuit 100 according to the invention.
• the circuit 100 comprises:
• DC voltage source 109 providing a voltage + U D c, for example a battery or a DC-DC power converter DCDC (or a continuous AC power converter
• ACDC connected to the first and second terminals 105 and 106 of the bridge 1, an electrical control member 110 such as an inductive load for controlling an actuator and connected between the third and fourth terminals 107 and 108,
• a third switch AL connected between the second and third terminals 106 and 107
• a fourth switch BL connected between the second and fourth terminals 106 and 108.
• Power switches AH, BH, AL and BL are, for example, MOSFET transistors. Each transistor has a diode mounted antiparallel (present by construction in the case of MOS-FET transistors).
• the electrical control member 110 is here a variable inductance for controlling an actuator (electromagnet) of electromagnetic valves.
• the variable inductance is of course perfect and has a resistive portion.
• the current in the electrical control member 110 allows for example to control the opening and closing of the valves (via pallets that hold the valves in the open or closed position).
• the position of the valves is defined by a $ setpoint corresponding to a setpoint current.
• FIG. 4 Three states (respectively, of magnetization, freewheeling and demagnetization) are represented in FIG. 4: the magnetization state corresponding to the case where the transistors BH and
• FIGS. 5 and 6 illustrate the evolution of the voltage VL applied across a load such as the load 110 of FIG. 4 and the current IL in this load, respectively in the case of a PWM control according to the state of the technique using two states and in the case of a control circuit according to the invention using three states.
• the circuit 200 comprises, in addition to the components already described above with reference to FIG. 4, a shunt resistor 201 in series with the load 110, said shunt resistor having a terminal 203 connected to the terminal 108 of the circuit 200 and to the non-inverting input of an operational amplifier 202 and a terminal 204 connected to the load 110 and to the inverting input of the operational amplifier 202.
• the advantage of putting a current measurement directly on the load when implementing an average current control is that one does not need to reconstruct the current to obtain the average current image.
• the shunt resistor is an inexpensive solution, the use of an operational amplifier accepting high common mode voltages is interesting. It is thus possible to have a measurement referenced with respect to the mass.
• the phase undergoes strong variations of potential with respect to the mass. This is why a high common mode rejection rate ("high CMRR" in English) is necessary in order to have a non-noise measurement at the output of the amplifier. Filtering the measurement can reduce common-mode noise but it also hampers bandwidth or stability. When it is necessary - to have-a great bandwidth and that it is necessary to have fast switching, it may be difficult to have a measure of its own.
• the current measurement is noisy by peaks due to the common mode at each switching of the switches which disturbs the regulation of the current.
• a solution to overcome this problem is to organize the sequence of four possible states (magnetization, demagnetization, freewheel, low freewheel) so that the bridge arm (here the arm B) connected to the current measurement shunt resistor 201 does not switch at the chopping frequency.
• FIG. 8 represents the sequence of the four states of a control circuit 300 formed of components identical to those of the circuit 100 represented in FIG. 4.
• Negative alternation is made by switching only the switches of the arm A AH and AL, the switches of the arm B always remaining in the same state (BH open and BL closed); we thus pass in the case of the negative alternation of a magnetization state (closed AH and open AL) to a low freewheel state (open AH and closed AL), the transition between the two states occurring at the frequency chopper (typically several tens or even hundreds of kHz).
• the frequency chopper typically several tens or even hundreds of kHz.
• arm B (BH closed, BL open) does not switch while arm A switches to the switching frequency.
• the arm B (open BH, BL closed) does not switch while the arm A switches to the switching frequency.
• the arm B does not switch.
• Switching arm B occurs only in case of change of alternation when the sign of the average voltage applied to the load 110 changes. The switching between two alternations occurs at a frequency much lower than the switching frequency and at a frequency lower than the bandwidth of the current regulation, theoretically twice as small as the switching frequency but generally more than ten times smaller than the switching frequency (a few kilohertz in general).
• the amplitude arm (the arm A) for amplitude regulation and the sign arm (the arm B) for the polarization of the voltage applied to the load will be distinguished.
• the positive and negative half-waves of such a circuit 400 are illustrated in FIG. 9.
• the circuit 400 in addition to the components already described above with reference to FIG. 4, comprises a shunt resistor 201 in series with the load 110 with a terminal 203 connected to the terminal connected to the terminal 108 of the circuit 200 and to the non-inverting input of an operational amplifier 202 and a terminal 204 connected to the load 110 and to the inverting input of the operational amplifier 202.
• the arm Has formed by the switches AH and AL is the amplitude arm that switches to the switching frequency and allows to set the absolute value of the current in the load 110.
• the arm B formed by the switches BH and BL is the sign arm and allows transitions between the positive and negative halfwaves.
• FIG. 10 illustrates the operation according to this embodiment: the voltage VL across the load and the current IL measured in the load (via the shunt resistor 201 and the operational amplifier 2-02) are represented in FIG. function of time.
• a positive alternation AP (during which BL leads and BH is closed) is followed by a negative alternation AN (during which BH leads and BL is closed).
• the common-mode disturbance (current peak disturbing the measured current) occurs only at the moment of change of alternation. As indicated above, with a two-state configuration according to the prior art, we would observe a peak at each switching.
• FIG. 11 illustrates a load pump circuit 500 of a switch used in a control circuit according to the invention and making it possible to maintain a control voltage on transistor BH (only arm B is shown in FIG. 11).
• the circuit 500 comprises:
• an auxiliary power supply 501 delivering a lower voltage (typically of the order of 12 to 15V, these voltage values corresponding to typical values making it possible to control the gate of a MOSFET transistor) to the voltage UD C of the source main power supply;
• a diode 502 whose anode is connected to the auxiliary power supply 501; a capacitor 503 having a first terminal connected to the cathode of the diode 502 and a second terminal connected to the common terminal of the transistors BH and BL;
• a first switch 504 having a first terminal connected to the cathode of the diode 502 and a second terminal connected to the gate of the transistor BH;
• a second switch 505 having a first terminal connected to the second terminal of the first switch 504 and therefore to the gate transistor BH and a second terminal connected to the common terminal of transistors BH and BL.
• the charge of the capacitor 503 is made by closing the transistor BL via the auxiliary power supply 501.
• the charged capacitor 503 makes it possible to deliver a control voltage of the gate of the transistor BH when the switch 504 is closed.
• one solution consists in forcing the closing of the transistor BL of the sign arm to guarantee the charge of the capacitor 503 of the charge pump.
• the loop 600 comprises:
• a PWM signal generator 603 Pulse Width Modulation having a cyclic ratio ⁇ ;
• a four-quadrant bridge 604 for providing a switched voltage to a load (typically an inductive load) 605 whose current measured my form the current to be controlled by said loop 600.
• the control loop 600 of a two-state control of the bridge 604 operates in the following way: a current setpoint l ref is compared with the measurement of the current to be controlled Imes-
• the difference E between the current setpoint l re f and measuring the current to the slave my determined by the subtraction tor 601 is amplified by the amplifier 602 of gain G to provide an amplified difference Ea (the gain G may be a combination of proportional gain, and proceedingsai differentiator).
• the amplified error Ea is compared with a carrier, typically a triangular or sawtooth signal of amplitude A, in order to produce a PWM signal whose pulse width is modulated by the amplified error ( ⁇ is the ratio cyclic MLI signal).
• the switches of the four-quadrant bridge are controlled by this PWM signal (or by the complementary signal PWM signal) by respecting a dead time between the closing and opening of the switches of the same arm.
• a voltage VL cut at the frequency of the carrier is therefore applied to the load 605.
• the average voltage V L applied to the load 605 depends directly on the duty cycle ⁇ .
• the current I L in the load 605 of inductance L (which neglects the line resistance) integrates this voltage.
• the current in the load can thus increase or decrease in counter-reaction to the deviation of the current or the variation of the setpoint.
• FIG. 13 illustrates the construction of a cyclic ratio PWM signal ⁇ constructed according to the .
• the PWM signal is thus constructed from the comparison between a carrier of amplitude A (here a sawtooth signal but which can also be a triangular signal) and the useful signal (modulator) represented by the amplified error Ea.
• the carrier is used to set the switching frequency.
• the PWM signal which forms the control setpoint of the bridge 604 has a positive pulse as long as the modulator is above the carrier and zero otherwise.
• the MLI signal is a fixed frequency signal (carrier signal) whose pulse length depends on the error: it is the duty cycle of the PWM that controls the current in the load. This
• a PWM signal of cyclic ratio ⁇ is applied to the switches AL and BH and the complementary signal of the signal PWM, of cyclic ratio 1- ⁇ , to the switches AH. and BL.
• DC voltage + U dc is applied to the load for a duration aT
• the DC voltage -U dc is applied to the load for a period of time.
• - ⁇ ) r The average voltage
• a first embodiment of a regulation loop 700 enabling the implementation of a four-state control circuit according to the invention.
• the regulation loop 700 comprises:
• an amplifier 702 a PWM signal generator 703 (Pulse Width Modulation) having a cyclic ratio ⁇ ; a four-quadrant bridge in its state of positive half-wave 705 or in its state of negative half-wave 706 intended to supply a voltage cut off to a load (generally an inductive load) 707, the measured current of which I form the current to be servocontrolled by said loop 700;
• a load generally an inductive load
• control means 708 for transiting between the positive half-wave 705 and the negative half-wave 706.
• the regulation loop 700 operates in the following manner: a current setpoint l r ⁇ f is compared with the measurement of the current to be slaved to the mes .
• the difference E between the current setpoint and the r ⁇ f current measurement to the slave my determined by the subtractor 701 is amplified by the amplifier 702 of gain G to provide an amplified difference Ea (the gain G may be a combination of proportional, integral and derivative gain).
• the amplified error Ea is compared to a carrier, typically a triangular or sawtooth signal of amplitude A, in order to produce, via the generator 703, a PWM signal whose pulse width is modulated by the amplified error. (We note ⁇ the duty cycle of the PWM signal).
• control of the quadrant bridge is a function of the error between the current setpoint and the current measurement.
• the difference lies in the PWM function, the state of the sign-dependent bridge which is determined from the PWM signal.
• control means 708 for transiting between the positive half-wave 705 and the negative half wave 706 implement the next state machine (illustrated in FIG. 16): when the bridge is in its positive half-wave 705 and when the duty cycle is canceled, the machine detects a change of state and goes alternately negative: the sign arm switches.
• the signal MLI of cyclic ratio ⁇ is applied to the switch AL and the signal " PWM complementary cyclic ratio (1- ⁇ ) to the switch AH both during the positive half-cycle and during the negative alternation.
• the means 708 control the opening of the switch BH and the closing of the switch BL.
• the means 708 control the opening of the switch BL and the closing of the switch BH.
• the means 708 use a sign signal controlling the sign arm B by allowing said arm to switch from a positive half cycle (positive sign signal) to a negative half cycle (one can choose a negative sign signal or a signal of null sign).
• the transfer function of the bridge is calculated according to whether it is in its positive or negative half-cycle:
• V L - (la) U DC during the negative half cycle.
• control circuit comprises: - means for applying a load to the average voltage equal to the product ⁇ xU D c during the positive half cycle, .705; means for applying to the load a mean voltage equal to
• Figures 18 to 21 show the measured current I, the reference current I ref, the duty cycle of PWM signal ⁇ , the sign of signal allowing the sign of B arm to switch from a positive alternation (positive signal signal) towards a negative alternation (negative sign signal) and the voltage VL across the load as a function of time in different configurations.
• E n designates the amplified error in the case of the negative half cycle.
• the negative static error can therefore be greater than the positive static error. For example, if the supply voltage is 49V and 1V is set across the load, the ratio is 50. This leads to discontinuities in the control. Therefore, a recession of the current during the transition between alternations is likely to occur. This important static error can however be compensated for by an increase of the static gain.
• FIG. 22 illustrates a second embodiment of a regulation loop 800 for implementing a four-state control circuit according to the invention and making it possible to overcome this static error problem.
• the regulation loop 800 comprises:
• a PWM signal generator 803 Pulse Width Modulation having a cyclic ratio ⁇ ;
• a converter 809 for converting the signal.
• a four-quadrant bridge in its state of positive half-wave 805 or in its state of negative half-wave 806 intended to supply a voltage. cut to a load (usually an inductive load)
• the control loop 800 operates as follows: a current I ref setpoint is compared with the current measurement to the slave my.
• the difference E between the current setpoint l r ⁇ f and the measurement of the current to be controlled lm is determined by the subtractor 801 is amplified by the gain amplifier 802 G to provide an amplified difference Ea (the gain G can be a combination of proportional, integral and derivative gain sound).
• the amplified error Ea is compared with a carrier, typically a triangular or sawtooth signal of amplitude A, in order to produce, via the generator 803, a PWM signal whose pulse width is modulated by the amplified error. (We note ⁇ the duty cycle of the PWM signal).
• the cyclic PWM signal ⁇ is converted by the converter
• the converter 809 comprises, for example, means making it possible to perform the subtraction between a 50% duty cycle PWM signal and the cyclic ratio PWM signal ⁇ .
• the PWM signal obtained has a cyclic ratio
• This subtraction can be obtained by an exclusive OR logic function as illustrated in FIG. 23.
• the inputs of the exclusive OR function are respectively the synchronization PWM signal having a ratio of cyclic port of 50% and the MLI signal cyclic ratio ⁇ and we obtain di ⁇
• The. converter 809 further comprises means for
• the resulting signal is a PWM signal whose pulse width is twice as high.
• the duty ratio is therefore 2a ⁇ ⁇ .
• the control means 808 for transiting between the positive half wave 805 and the negative half wave 806 implement the following state machine (illustrated in FIG. 25):
• the machine detects a change of state and switches to a negative alternation: the sign arm switches.
• the machine detects a change of state and switch to positive alternation: the sign arm switches.
• the PWM pulse signal ML ' is applied to the switch AL and the cyclic ratio complementary PWM signal (1- ⁇ 1 ) to the switch AH.
• the signal MLI of duty cycle 1- ⁇ ' is applied to switch AL and the complementary PWM signal cyclic report ⁇ ' to switch AH.
• the means 808 control the opening of the switch BH and the closing of the switch BL.
• the means 808 control the opening of the switch BL and the closing of the switch BH.
• the means 808 use a sign signal controlling the sign arm B by allowing said arm to switch from a positive half cycle (positive sign signal) to a negative half cycle (one can choose a negative sign signal or a signal of null sign).
• the transfer function of the bridge is calculated according to whether it is in its positive or negative half-cycle:
• control circuit comprises means for applying a load to the average voltage equal to the product (2 ⁇ -1) xUoc during the positive half-wave 805 and during the negative half-wave 806.
• Figure 26 shows the transfer function of the four-deck bridge
• this transfer function is identical to that of a control of a known four-quadrant bridge with two states except that a conversion of the duty cycle has been carried out. It is easily understood that by its symmetry, this transfer function makes it possible to solve the problem of the static error.
• the various control means allowing the control strategy are for example, logic means integrated into an FPGA ("Field Programmable Gaste Array").
• the invention has been described in the case of an application to an actuator of electromagnetic valves but it applies to any type of electric machine controlled by an inductive load such as an electric motor for example.
• an inductive load such as an electric motor for example.
• the latter can be mono or three-phase.
• the described embodiment relates to the control of the current in an inductive load but the invention is of course applied to the control of the voltage across the same load.

Landscapes

• Engineering & Computer Science (AREA)
• Power Engineering (AREA)
• Control Of Direct Current Motors (AREA)
• Inverter Devices (AREA)
• Control Of Ac Motors In General (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=39731783

Family Applications (1)

Country Status (7)

Families Citing this family (11)

Family Cites Families (16)

• 2007
  • 2007-12-07 FR FR0708562A patent/FR2924873B1/fr not_active Expired - Fee Related
• 2008
  • 2008-12-04 US US12/745,845 patent/US8576581B2/en not_active Expired - Fee Related
  • 2008-12-04 JP JP2010536504A patent/JP2011507463A/ja active Pending
  • 2008-12-04 KR KR1020107012616A patent/KR20100089872A/ko not_active Application Discontinuation
  • 2008-12-04 CN CN200880119482.5A patent/CN101889387B/zh not_active Expired - Fee Related
  • 2008-12-04 EP EP08872282A patent/EP2227855A1/de not_active Withdrawn
  • 2008-12-04 WO PCT/FR2008/001695 patent/WO2009101292A1/fr active Application Filing

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events