Controlling three-phase inductive loads in partial load operation by reducing inverter switching losses.

Abstract

A method (100) for operating an inverter (1) wherein the inverter selectively connects three AC phases (U, V, W) of an inductive load (4) to the positive or negative terminal of a DC voltage (2) provided at the input terminal (1a) of the inverter (1) by manipulating switching elements (3a-3f) arranged in three half-bridges (5a-5c), wherein the switching state (110) of the switching elements (3a-3f) is changed by rotating space vector modulation, wherein, additionally, when the space vectors (6a-6f) having the same switching state of the corresponding switching elements (3a, 3d; 3b, 3e; 3c, 3f) of at least one half-bridge (5a-5c) are modulated, such half-bridges (5a-5c) remain completely off (120).

Description

Technical Field

[0001] The present invention relates to operating a three-phase inductive load on an inverter, wherein the effective current is pre-given by pulse width modulation. Background Technology

[0002] In electric drive systems that convert electrical energy into mechanical energy, three-phase or multi-phase AC motors are preferred due to their high efficiency and the elimination of the need for easily worn brushes compared to DC motors. Such drive systems are used, for example, in elevators, pumps, and in the powertrain systems of electrically driven vehicles. However, electrical energy sources that can be carried in vehicles (such as batteries or fuel cells) typically provide DC voltage. For industrial applications, three-phase AC current is mostly used, which is rectified using a rectifier.

[0003] To convert DC voltage into three-phase or multi-phase AC voltage, an inverter is used.

[0004] In traction operation, the requirements for torque and speed are constantly changing. Thus, higher speeds are needed for acceleration, and higher torque is needed to maintain speed on inclines. Inverters can change not only the frequency of the alternating current but also its effective current intensity over a wide range. However, converting this into alternating current via the rapid switching of the power semiconductors in the inverter consumes energy. DE 10 2004 027 629 A1 discloses a method for pulse width modulation that can save some switching losses when generating a sinusoidal three-phase AC voltage through an inverter. Summary of the Invention

[0005] Within the scope of this invention, a method for operating an inverter has been developed, wherein the three AC phases of an inductive load (e.g., a synchronous machine) are optionally connected to the positive or negative terminal of a DC voltage provided at the input of the inverter by manipulating switching elements arranged in three half-bridges. The switching elements can be configured, for example, as power semiconductors. Here, the switching states of the switching elements are changed by rotational space vector modulation. In space vector representation, the current distribution on the three phases is characterized on the one hand by the magnitude, real or imaginary part of the space vector, and on the other hand by the angle of the space vector relative to the coordinate axes. In this case, "cycle" specifically means that a new, composite space vector is always generated again by modulation, and its angle continuously changes in a consistent direction of rotation.

[0006] Since each half-bridge can only connect its assigned AC phase to either the positive or negative terminal of the DC voltage, the static switching state of the switching elements can only represent a maximum of 2.3 = 8 different space vectors. These space vectors alone are insufficient to drive, for example, a motor to rotate. Therefore, several additional space vectors are needed, located between the adjustable space vectors. Within the scope of space vector modulation, additional space vectors are generated, respectively located between these adjustable space vectors, by alternately applying adjustable space vectors (which are adjacent in the positive or negative direction of the angle or in the zero state) using pulse width modulation. When modulated between space vectors having the same switching state with respect to at least one half-bridge having corresponding switching elements, such a half-bridge now remains completely off.

[0007] In conventional space vector modulation, only the following switching states are permitted: in which each half-bridge connects its associated phase either to the positive or negative terminal of a DC voltage. Therefore, in each half-bridge, exactly one switching element is on and the other is off. A switching state where both switching elements of a half-bridge are off does not occur. The introduction of such a switching state alters the respective space vectors and thus causes the time variation of the current flowing through the AC phase in space vector modulation to deviate from a sinusoidal shape. Correspondingly (Im Gegenzug), all switching processes on a completely off half-bridge are eliminated. It has been recognized that, especially when manipulating inductive loads, this deviation from a sinusoidal shape does not impair the function applied by the load in many cases. Therefore, the time sequence of the switching process can be "lossily compressed" in the described manner. This situation is somewhat analogous to the compression of image, audio, and video data, where the shift from lossless to lossy compression methods increases compression efficiency to new orders of magnitude.

[0008] In a particularly advantageous configuration, the half-bridge remains completely off within the following consecutive sectors in the space vector diagram: in said consecutive sectors, the switching states of the switching elements belonging to at least one half-bridge remain identical. Here, "consecutive" should be understood in the positive or negative direction of the space vector angle.

[0009] It has been recognized that in a spatial vector diagram, there exists a sequence of first, second, and third spatial vectors in either the positive or negative direction of the angle. This allows for a change from the first spatial vector to the second spatial vector by altering the switching state of the first half-bridge, and a subsequent change from the second half-bridge to the third spatial vector. Therefore, within the sector enclosed by the first and third spatial vectors, progress is generally characterized by changes in the switching states of only two half-bridges. The switching state of the third half-bridge thus adds no further information, and no significant information is lost when the third half-bridge remains completely off.

[0010] In another advantageous configuration, space vector modulation occurs only between so-called active space vectors. An active space vector represents a state in which a voltage other than zero is applied between at least two of the AC phases. Furthermore, there exists a so-called zero-voltage space vector, which represents a state in which all three AC phases are short-circuited. These zero-voltage space vectors are also required if a sinusoidal rotating field is to be generated by space vector modulation. However, since the precise sinusoidal deviation of the generated AC current from the half-bridge pair is already "done" within the scope of this method, the switching process required to apply the zero-voltage space vectors can also be omitted.

[0011] In a particularly advantageous configuration, space vector modulation is controlled and / or regulated to achieve the following objective: a three-phase alternating current with a pre-given effective value flows through an inductive load, which commutates in the sense of a rotating field. Especially in such applications, the precise time variation of the current within the period of the alternating current is not important, as long as the effective value corresponds to the pre-given value. Therefore, the "modification" of this time variation is not crucial. "Forming a sinusoidal shape is not necessary for the technical function of inductive loads; it only takes time for the switching process."

[0012] Of particular advantage, the inductive load can be an electric motor. To excite and maintain the rotation of the motor, it is only necessary to generate a combined magnetic force between the stator and rotor at any given moment, which continues to drive the rotation. This does not necessarily presuppose that the excitation of the motor's magnetic coils has a precisely sinusoidal curve over time.

[0013] In another particularly advantageous configuration, the engine operates in a star connection. This means that the three windings of the engine, powered by three AC phases, share a common star point. This star point is not grounded.

[0014] In this configuration, for example, current flowing from the positive terminal of the DC voltage through one half-bridge of the inverter through the motor windings can only reach the other half-bridges of the inverter through the other two windings and ultimately return to the negative terminal of the DC voltage to close the circuit. Completely shutting off the half-bridge has the effect that only one half-bridge, instead of two, is available to draw back the negative terminal of the DC voltage. However, this means that the current no longer needs to overcome the parallel connection of the other two windings on the path from the positive terminal to the negative terminal, but rather a specific winding within those windings. Since each winding has unavoidable impedance, the current must overcome a higher overall impedance.

[0015] This results in a smaller voltage drop across each winding, and therefore less electrical power being converted in the engine overall. Simultaneously, the voltage drop across the inverter's switching elements is also smaller. If these switching elements are power semiconductors (e.g., transistors), this reduces the conduction losses of these semiconductors (which also contribute to switching losses). Conduction losses depend not only on the voltage drop across the switching elements but also on the current allowed to pass through each element.

[0016] Conversely, reduced power consumption means the engine can provide less mechanical power. Under full load operation, the engine's capacity will therefore be reduced. However, this method is specifically designed for partial load and no-load operation of the engine and other inductive loads.

[0017] It has been recognized that, especially during partial load operation or even no-load operation of the motor, the proportion of switching losses in the total losses increases significantly: if the required effective current is reduced to one-twentieth, the switching losses are reduced to only one-half according to existing technology. In no-load operation, this can thus reduce the inverter efficiency to below 25%. Here, a significant improvement is now achieved at the cost of a more or less significant deviation of the time-varying current of the three-phase AC current from a sinusoidal shape. However, the current ripple is unaffected because the PWM frequency remains fixed. However, especially during partial load and no-load operation, the reduction in maximum motor power output due to increased resistance and possibly additionally due to the deviation from a sinusoidal shape is not significant. More importantly, there is an improvement in inverter efficiency.

[0018] Many drive systems typically operate at maximum speed for only a very small percentage of their runtime. Therefore, partial load operation and no-load operation together constitute the largest portion of the operating time for an electric powertrain system. Using the proposed method, the system efficiency is significantly improved during partial load operation.

[0019] Therefore, significant energy is saved by reducing switching losses, which is particularly advantageous in applications involving vehicle drive powertrain systems. Since vehicles typically carry limited energy reserves, the energy savings are also reflected in increased effective range.

[0020] Therefore, it is advantageous to select the following operating phase for the inductive load: in which the inductive load operates at a maximum of 20%, preferably a maximum of 10%, and very particularly preferably a maximum of 5% of its rated power.

[0021] In particular, the method can be implemented entirely or partially by a computer. Therefore, the invention also relates to a computer program having machine-readable instructions that, when executed on one or more computers, cause the computers to perform the method. In this sense, control devices for vehicles and embedded systems for technical devices, which are also capable of implementing machine-readable instructions, should also be considered as computers.

[0022] Similarly, the present invention also relates to a machine-readable data carrier and / or downloadable product having a computer program. A downloadable product is a digital product capable of being transmitted over a data network, i.e., a digital product downloadable by a user of the data network, which may, for example, be sold in an online store for immediate download.

[0023] In addition, computers may be equipped with computer programs, machine-readable data carriers, or downloadable products. Attached Figure Description

[0024] Hereinafter, in conjunction with the description of preferred embodiments of the invention, additional measures to improve the invention are shown in more detail with the aid of the accompanying drawings.

[0025] The attached diagram shows:

[0026] Figure 1 : An embodiment of method 100;

[0027] Figure 2 A diagram illustrating space vector modulation;

[0028] Figure 3 : An exemplary application of method 100 on an inverter 1 having a motor as an inductive load 4. Detailed Implementation

[0029] Figure 1 The illustration depicts an embodiment of a method 100 for operating inverter 1. Inverter 1 itself is... Figure 3This is shown in more detail below. In step 110, the switching states of the switching elements 3a-3f of the inverter 1 are changed by rotating space vector modulation. According to step 120, the half-bridges 5a-5c, which are given the same pre-defined switching states by the space vectors 6a-6f used in the modulation, are kept completely off.

[0030] In this embodiment, the motor is selected as the inductive load 4 in step 105. According to block 105a, the motor is operated in a star connection with the ungrounded star point 8.

[0031] In step 106, the partial load operation phase 4' of the inductive load 4 is selected as the operation phase in which the half-bridges 5a-5c remain completely off and the space vector modulation of the rotation is modified in this respect.

[0032] The space vector modulation configuration can be illustrated in more detail within box 110.

[0033] According to block 111, space vector modulation is performed specifically only between active space vectors 6a-6f. In contrast to a so-called zero-voltage space vector, an active space vector represents a state in which a voltage other than zero is applied between at least two of the AC phases U, V, and W. By not using any zero-voltage space vector, common-mode current is advantageously reduced.

[0034] According to box 112, space vector modulation can be directed, in particular, by corresponding control and / or regulation, to a three-phase alternating current 4a with a pre-given effective value flowing through an inductive load 4, which is commutated in the sense of a rotating field.

[0035] According to box 121, within consecutive sectors 7 in the spatial vector diagram (in which the switching states of switching elements 3a, 3d; 3b, 3e; 3c, 3f associated with at least one half-bridge 5a-5c remain the same), such half-bridge 5a-5c can remain completely off.

[0036] Figure 2 The diagram illustrates space vector modulation on a space vector map. Six active space vectors, 6a-6f, are marked. The coordinate axis table represents the real and imaginary parts of the complex AC voltage U*. Table 1 shows the switching states of the half-bridges 5a-5c and their switching elements 3a-3f.

[0037]

[0038] Table 1: Space Vector Modulation (without disabling half-bridge 5a-5c)

[0039] At angle The space vectors 6a, 6b, and 6c, which follow each other in the positive rotation direction, have the following common feature: the two switching elements 3c and 3f of half-bridge 5c remain in the same switching state. Therefore, space vectors 6a and 6c enclose a sector 7a in which the switching state of half-bridge 5c is constant. If half-bridge 5c is now kept completely off within this sector 7 (marked by a thick box in Table 1) (i.e., the two switching elements 3c and 3f are continuously "off"), then two switching processes are omitted: when transitioning from space vector 6c to space vector 6d, switching element 3f does not need to be off because it is already off. Similarly, when transitioning from space vector 6f to space vector 6a, switching element 3f does not need to be on because it should remain off.

[0040] Similarly, space vectors 6c and 6e enclose the second sector 7b, in which the switching state of the first half-bridge 5a is constant. Within this sector 7b, marked by the dotted line border in Table 1, the first half-bridge 5a can remain completely off.

[0041] Space vectors 6e and 6a enclose the third sector 7c, in which the switching state of the second half-bridge 5b is constant. Within this sector 7c, marked by a dashed border in Table 1, the second half-bridge 5b can remain completely off.

[0042] These changes dictate that the final time-varying process of the AC current 4a driven by the inductive load 4 will deviate significantly from a sinusoidal shape. For a sinusoidal output current, only a space vector located on circle K in the space vector diagram can be used. This will require not only active switches for all three half-bridges 5a-5c, but also the addition of a zero-voltage space vector. Consequently, more switching processes will be needed overall for the switching elements 3a-3f.

[0043] The complete shutdown of half-bridges 5a-5c results in only space vectors 6b, 6d, and 6f being able to be switched using the full rated voltage U1. Therefore, pulse width modulation can only be applied between the three space vectors 6b, 6d, and 6f. Consequently, pulse width modulation can only produce a composite space vector that terminates on a doodled connection between the endpoints of space vectors 6b, 6d, and 6f. This corresponds to a reduced voltage U2. This is reflected in the fact that, as described above, the resistance of the motor operating in a star connection becomes greater.

[0044] Figure 3Figure a shows the complete structure consisting of an inverter 1 and a motor as an inductive load 4. The inverter 1 receives a DC voltage 2 at its input terminal 1a and outputs an AC current 4a having phases U, V, and W at its output terminal 1b. Phase U is controlled by a first half-bridge 5a, where it is optionally connected to the positive terminal of the DC voltage 2 via a switching element 3a or to the negative terminal of the DC voltage via a switching element 3d. Phase V is controlled by a second half-bridge 5b, where it is optionally connected to the positive terminal of the DC voltage 2 via a switching element 3b or to the negative terminal of the DC voltage via a switching element 3e. Phase W is controlled by a third half-bridge 5c, where it is connected to the positive terminal of the DC voltage 2 via a switching element 3c or to the negative terminal of the DC voltage 2 via a switching element 3f.

[0045] Each of phases U, V, and W supplies power to one of the windings 11, 12, and 13 of motor 4, wherein these three windings 11, 12, and 13 share a common ungrounded star point 8. Windings 11, 12, and 13 have not only an inductance L but also an ohmic resistor R connected in series with them. u R v Or R w .

[0046] Figure 3 b illustrates the function of these ohmic resistors when the entire half-bridge 5a-5c is not switched off according to the method. For example, if the AC current 4a is guided from the positive terminal of the DC voltage 2, through the first half-bridge 5a, and through the resistor R u The first winding 11 can only pass through a resistor R. v Winding 12 and with resistor R w The parallel connection of winding 13 is led back to the negative terminal.

[0047] Figure 3 c illustrates how the complete disconnection of the third half-bridge 5c affects the total resistance. For the return loop, now only resistor R is present. v The second winding 12 is also available. Therefore, the total resistance becomes larger.

Claims

1. A method (100) for operating an inverter (1) which connects three alternating current phases (U, V, W) of an inductive load (4) each alternatively to the positive or negative pole of a direct current voltage (2) provided at an input (la) of the inverter (1) by manipulation of switching elements (3a-3f) arranged in three half-bridges (5a-5c), wherein, Changing the switching states of the switching elements (3a-3f) by means of rotating space vector modulation, wherein, in addition, at least one half-bridge (5a-5c) having identical switching states of the respective switching elements (3a, 3d; 3b, 3e; 3c, 3f) remains completely switched off when modulating between space vectors (6a-6f) having identical switching states of the respective switching elements (3a, 3d; 3b, 3e; 3c, 3f) with respect to the at least one half-bridge (5a-5c), characterized in that the at least one half-bridge (5a-5c) remains completely switched off in consecutive sectors (7a-7c) in a space vector diagram in which the switching states of the switching elements (3a, 3d; 3b, 3e; 3c, 3f) assigned to the at least one half-bridge (5a-5c) remain identical.

2. The method (100) according to claim 1, characterized in that The space vector modulation is carried out exclusively between active space vectors (6a-6f), which represent states in which a voltage different from zero is applied between at least two of the AC phases (U, V, W).

3. The method (100) according to claim 1 or 2, characterized in that The space vector modulation is controlled and / or regulated to the goal of a three-phase AC current (4a) with a predefined effective value flowing through the inductive load (4), which AC current commutates in the sense of a rotating field.

4. The method (100) according to claim 1 or 2, characterized in that An electric motor is selected as the inductive load (4).

5. The method of claim 4, wherein, The electric motor is operated in a star connection with an ungrounded star point (8).

6. The method (100) according to claim 1 or 2, characterized in that An operating phase (4') of the inductive load (4) is selected in which the inductive load (4) is operated at a maximum of 20% of its rated power.

7. The method (100) according to claim 6, characterized in that In the operating phase, the inductive load (4) is operated at a maximum of 10% of its rated power.

8. The method (100) according to claim 7, characterized by In the operating phase, the inductive load (4) is operated at a maximum of 5% of its rated power.

9. A computer program product comprising machine-readable instructions which, when implemented on a computer or on a plurality of computers, cause the computer or the plurality of computers to implement a method (100) according to any one of claims 1 to 8.

10. A machine-readable data carrier and / or a download product having a computer program product according to claim 9.

11. A computer which is equipped with a computer program product according to claim 9 and / or which is equipped with a machine-readable data carrier and / or a download product according to claim 10.

Images