JPS62260593A - Output voltage detection system for pwm inverter - Google Patents

Abstract

PURPOSE:To enable accurate detection of an inverter output voltage with much pulsation by computing an inverter output voltage vector from a voltage of DC voltage source and a switch state variable. CONSTITUTION:A voltage detector 7 is inserted between DC side positive bus 1a and negative bus 1b. A switch state variable of voltage outputted by an inverter 2 this time is kept in mind. Based on this switch state, a switch state variable table is referred to so that a switch state variable may be read out. Then, said variable and a voltage of DC voltage source 1 detected by the voltage detector 7 are used to obtain both d, q axial components of an inverter 2 output voltage which are to be detected actually.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an improvement in an alternating current voltage detection method when controlling the voltage and magnetic flux of a three-phase induction motor using a high-speed digitally controlled PWM inverter.

[Conventional technology]

Three-phase bridge PWM consisting of high-speed switching elements
In a system that supplies power to a three-phase induction motor using an inverter, the method of calculating the motor 1 ignition flux by integrating the motor terminal voltage, that is, the inverter output voltage, and performing follow-up control to the magnetic flux command value is described in the following paper, etc. Disclosed.

1) Institute of Electrical Engineers of Japan study group material RM-84-76 (Sho 59-9
) "High-speed torque control method for induction motors based on new theory 1" 2) Transactions of the Institute of Electrical Engineers of Japan 6O-R61 (Sho 6O-6) "Magnetic flux control type real-time processing PW using a one-chip microcomputer
M Control'' These papers detect the motor input voltage and integrate it within the control circuit, which is the motor magnetic flux. That is, it is a so-called magnetic flux calculation type control system, and the greatest goal is to match the calculated magnetic flux and the actual magnetic flux as much as possible.

The conventional technology related to the magnetic flux calculation method will be explained below.

Fig. 2 is a main circuit diagram of an example of a conventional PWM inverter, in which a DC voltage source 1 is connected to a 3-phase induction motor 3 via a stopper 1jlla + a negative bus 1b, and a 3-phase PWM inverter 2.
to supply power. The control circuit 4 receives commands and detected information 4
2 to generate an energization signal 41 for the switching element of the PWM inverter 2.

The PWM inverter 2 consists of six arms each consisting of transistors Q1 to Q6 and diodes D1 to D6 connected in antiparallel, and these transistors can be replaced with other high-speed switching elements such as GTO or SI thyristor. .

In addition, 2a * 2b, 2c are PWM inverter 2
These are output terminals for AC phases a, b, and c, and power is supplied from each output terminal to the three-phase induction motor 3 via current detectors 5a, 5b + 5c, and voltage detectors 6a, 6b + from each output terminal. 6c is Y-connected so that each phase current and each phase voltage can be detected.

In such a system 7, the most direct way to calculate the motor flux is to integrate the voltage across the terminals of the three-phase induction motor 3.

That is, as an example, the main points of the arithmetic expression detailed in the paper 1) are as follows.

If the primary terminal voltage and current of a three-phase squirrel cage rust conduction motor are V and il, respectively, and the secondary current is 12, then the voltage equation is given by the symbols MS, 11, l! is a vector representation of the quantity transformed on the direct axis and horizontal axis, that is, d and q two axes,
For example, vl represents the d-axis component as vl, 1, and the q-axis component as Vl.
q, then Vl=vt(1+jVIQ”°=””°−”°°°”°
It is indicated by °゛■, and il and i2 are similarly defined.

Note that 0 on the left side of equation (2) represents the case where both the d and q axis components are O, and in the case of a cage multi-rotor, the secondary voltage is thus 0.

The constants in equation (■) are R1; primary winding resistance Lll; primary inductance R2; secondary winding resistance L22; secondary inductance M; mutual inductance - m is the rotational angular velocity, p is the differential operator, and j is the vector product. represent.

On the other hand, as a definition of magnetic flux, 1 ignition flux ψ1 is now 1 = L
ll il + Mi@・・・・・・・・・
... Old ...... Expand the first line of 0 formula ■ vs = (R11 + pLss ) h + pMi
Substituting 2 formula ■ and rearranging, vl −R41t = p9'l BHHH+B
B...1＋＋＋＋＋＋＋■It can be further determined by integrating both sides.

FIG. 3 is a diagram showing the high-speed switching elements of a PWM inverter replaced with contacts, and the same reference numerals as in FIG. 2 indicate the same parts. The switching elements of the voltage source inverter have two switching contacts 8a, 8b,
It is often expressed using 8c.

Each of the switching contacts Sa, Sb, and Sc may fall toward the positive bus line la or fall toward the negative bus line lb, and never take an intermediate position. The former is in state 1. If the latter is set to state 0, the output state of the inverter is as shown below.

All of the shapes can be represented by a table of constants.

Switch-like parameter table Here, k is a number indicating each switching contact state, and there are only eight of these. Also, vd, vq are ct, q22
The switch state variable expressed in components, the actual d.

The q-axis voltage V id * 'I lq can be expressed as the voltage of the DC voltage source 1. Figure 4 shows the switch state variable shown above.
It represents a voltage vector that steps by 6σ in the time direction as k increases.

Note that k=o and ni=7 are called zero vectors, and in the figure, they coincide with the origin ζ. k=0 and ni=7 are the switching contacts Sa in Fig. 3, which are the outputs of the inverter, respectively.
, Sb, and Sc all fall to the positive generatrix 1a side, or
Although there is a difference in whether it falls to the negative bus line lb side or falls to the negative bus line lb side, the line voltage of the highly conductive motor 3 is all O, and it is in the three-phase short circuit mode. Further, the reference axes of phases a, b, and c correspond to directions of =1°k=3 and k=5, respectively, according to equation (2) described later.

FIG. 5 is a block diagram showing an example of the internal configuration of the control circuit 4.
4a is a microprocessor, 4b is a ROM memory, 4
C is an input/output port, 4d is an address bus, and 4C is a data bus.

The signal 41 output to the input/output port 4C is a signal that drives the bases of the transistors Ql-Q6, and an amplifier circuit is usually installed between the signal 41 and the transistors Q1 to Q6 for both insulation and current amplification. However, it is omitted in this drawing.

The command group for controlling the system is ROM memo +74b
The microprocessor 4a executes the instructions sequentially and performs the necessary calculations and communication with the outside via input/output ports.
) via 4c. Note that the internal RAM of the microprocessor 4a is used for temporary storage of the calculation results.

An example of the command group necessary for control is shown in the flowchart of FIG. Each block will be explained below.

Starting from 401, initial values necessary for the calculation are set in 402, and the calculation is executed from the program point (PI). 403 is a block that inputs information necessary for control from the outside through input/output capo) 4c, and inputs the frequency command f* for the inverter, the motor magnetic flux command ψ↑, and the detected values of inverter output currents la, ib, and ic as shown in FIG. are read from current detectors 5a, 5b, and 5c.

404 is a waiting time Td corresponding to a energization prohibition time to prevent a short circuit between the upper and lower arms due to the storage time of the transistor. An actual base drive signal to the transistor is not generated until a waiting time Td elapses after an energization signal corresponding to an inverter output voltage k based on a calculation result to be described later is output to the input/output port 4C. Waiting time T for base drive signal
A time delay generation circuit for delaying the signal d is provided in a transistor base amplifier circuit (not shown).

The waiting time Td is the inverter output voltage va from the program point P3 outputting to the input/output capacitor 4c.

This is the time until program point P2, where vb and vc are read again from the input/output port 4c, and block 40
Another processing time, such as 3, can be included.

405 is a block that reads inverter output voltages va, vb, vc by voltage detectors 6a, 6b, 6c.

406 converts the voltages va, vb, vc into d, q2
+! This is a block that corrects the voltage drop of the resistor due to the change in the tl+ component, and the d and q components of the voltage effective for generating magnetic flux of the motor are determined by equation (2). This is called the AC true pressure equivalent signal.

Vxd' = Vtd - R11td ......■ VIQ' = Vlq - 几1ixq Block 408 is for calculating the magnetic flux vector f1,
The product of the voltage value according to equation (2) and the calculation cycle time Δt is added to the value of the magnetic flux vector ψ1 in the previous calculation cycle. Δt is a time equivalent to the computation cycle time, and does not need to be equal to the computation time, but it is necessary to have a size proportional to the computation time.

vl is the length of the magnetic flux vector, and is obtained by calculating the square root of the sum of squares of the d, q22 components ψtd and ψIQ of the magnetic flux vector ψl.

Is block 409 a frequency command? From θゝ=2πf*t (t is time)...
・Calculate phase command 0 using ■.

Block 410 calculates the magnetic flux d, q22 components ψsd, ψIq
This is a block that detects the phase θ from the ground, and the calculation of the tuft is based on a function table, but the details thereof are directly related to the essence of the present invention, so a description thereof will be omitted.

The magnetic flux vector ψl is controlled so as to draw a circular locus in the ct, q plane as shown in FIG. In the 6th → * figure, the magnetic flux command vector is represented by vl, and its * length is the magnetic flux command ψ1, d read in the professional printer 403.
The angle formed with the axis is θ calculated in block 409.

The magnetic flux vector ψ1 to be controlled is the magnetic flux command vector i
*, and is controlled so that its tip is always kept within a ring of width Δψ. Here, Δψ is the permissible error range of the magnetic flux ψl.

In step 411, it is determined whether the flag set in the previous calculation cycle is fg=1 or fg=-1.
Or proceed to either 412+1 o 412-412
A is a block that determines whether the length ψ1 of the magnetic flux vector exceeds the upper and lower limits, respectively, for the magnetic flux command 91iζ read in 403, so as to draw a limit cycle as shown in FIG.

When the upper limit of 9 ri is exceeded, the controller 413a sets the voltage control flag to fg = -th ζ and commands magnetization.

When it becomes smaller than the lower limit ψl - Δψ, the controller sets the voltage control flag to fg=l by 413 and commands demagnetization. It remains set during the calculation cycle.

Blocks 414 and 414a are for determining whether phase 0 has overtaken the command phase θ, and if θ>θ, block 415a determines whether the zero vector of 0 or 7 as explained in FIG. is output and the process proceeds to block 416. When the inverter outputs a zero vector, both the phase θ and the length vl of the magnetic flux vector are maintained as they are.

If θ≦θ~, in block 415, the following PW
Search the M signal table and proceed to block 416.

According to the equation (2), the magnetic flux vector ψ1 is obtained by integrating the voltage vector of (vl-R1ft), but since R11t is small, it is almost in the same direction as the inverter output voltage v1. The inverter output voltage v1 can be expressed by the value of k in the switch state variable table, and if this is expressed by Vl (k), then the optimum value s (k) is determined by the value of phase 0 and voltage increase/decrease flag fg. Varies according to the PWM signal table.

As an example, in the case of Fig. 6, 13ge≦0<30°, and the magnetic flux vector? When 1 reaches the outer number of the annular part with width Δψ, the flag fg=1, and the demagnetization component (inward of the circle)
k = 3, Vl(31) is selected, and when the magnetic flux vector ψ1 reaches the inside ζ of the torus, the flag fg =
-1, and = 21 vt(2) is selected to have an increasing component.

The value of k increases by 1 as phase 0 progresses, and the value of k increases by 1 as the phase 0 progresses.
The values are as shown in the WM signal table.

The selected vs(k) is sent to the input/output capo 4c, and sends an energization signal to each transistor. Thereafter, the program jumps from the program point P3 to the program point P1, and the calculation can be repeated infinitely.

If a high-speed microprocessor is used, an operation cycle time of about 50 to 100 As can be obtained, but of this time, the waiting time Td due to the transistor energization prohibition time occupies about 30 μs.

[Problem that the invention seeks to solve]

When the inverter output voltage changes as a result of the voltage control described above, for example, the a-phase and b-phase remain in their previous states, but when the switch state variable of the C-phase changes from 1 to 0, that is, the transistor Q5 Consider when Q2 switches from conductive to non-conductive and transistor Q2 switches from non-conductive to conductive.

Although the base signal of transistor Q5 can be immediately extinguished, sending the base signal of transistor Q2 in its place must wait at least the time required for transistor Q5 to be completely turned off.

In addition to this, taking into account variations in transistor characteristics, a predetermined margin is added to the waiting time Td, which is a constant energization prohibition time.
It becomes.

Therefore, a normal voltage is not generated at the inverter output until the waiting time Td has elapsed from program point P3 in FIG. This results in a delay in the waiting time Td and a decrease in the inverter output voltage during that period.

In fact, as described above, in order to simplify programming, a time delay generation circuit for delaying the energization signal by the waiting time Td is provided in the base amplifier circuit (not shown) of the transistor.

FIG. 9 is a diagram showing the influence of waiting time Td on the waveform.
(a) is a waveform generated at the input/output capo 4c for C-phase control, and is preferably a waveform that is desired to be generated at the output point 2C of the C-phase. (b) is the waveform that actually occurs at the C-phase output point 2C due to the delay of the transistor, and (C) is the waveform that occurs at the voltage detector 6C, and the DOPT used in the voltage detector 6c is expensive. However, many of them have poor frequency characteristics, and generally have a waveform with a blunted rise as shown in the figure.

Moreover, in digital control using a microprocessor and a subprogram as shown in FIG. 8, even if a voltage with a waveform as shown in FIG. It can only be observed on the ground, so it cannot be accurate information.

[Means for solving problems]

Figure 1 shows a PW using the output voltage detection method according to the present invention.
FIG. 10 is a main circuit diagram of the M inverter, and is a flowchart of an embodiment of the command group of the corresponding control circuit, and the same reference numerals as in FIGS. 2 and 8 indicate the same parts.

Comparing Figure 1 and Figure 2 and mentioning only the differences, PW
Voltage detector 6a on the AC side of M inverter 2.

6b and 6c are removed, and instead a voltage detector 7 is inserted between the positive bus 1a and the negative bus 1b on the DC side. Also, in the flowchart of the instruction group, in Fig. 10, the inverter output voltage vector v1 is calculated from the voltage Vdc of the DC voltage source 1 detected by the voltage detector 7 and the switch state variables yd and vq using equation (2).
Program point P3 to program point 1-PI
A block 421 is provided in the process of returning to , and after the inverter voltage vector v1 stored this time is stored as the previous one, a new program cycle is entered.

[For production]

The operating principle of one embodiment of the present invention will be explained with reference to the flowchart shown in FIG. The switch state variable of the output voltage currently output by the inverter is stored in block 421.

Based on this switch state, the switch state variable yd is determined by referring to the switch state variable yd in block 404a.
, vq are read out, and in block 405a, using the voltage Vdc of the DC voltage source 1 detected by the voltage detector 7 and equation (2), vld, which is the d and q axis components of the inverter output voltage to be actually detected, is determined. .

Ylq can be obtained.

Since there is no sudden fluctuation in the voltage Vdc of DC voltage source 1,
Even a DOPT whose frequency characteristics are not very good can be used sufficiently, and only one is required. In this way, each phase output voltage va of the inverter from the input/output boat 4C,
There is no need to read vb and vc, and since the waiting time Td is already known, there is no need to set the transistor energization prohibition time as the waiting time Td. Therefore, accurate information will be obtained faster.

In block 407, the voltage drop caused by the primary current i generated across the primary resistor R1 can be processed into an AC output voltage equivalent signal in the same manner as in FIG. The flux operation of block 408a is an incremental time component. ift (Δt −
By substituting Td ), the decrease in the inverter output voltage due to the influence of the waiting time Td is taken into consideration as shown in FIG. 9(b). This is to set the AC output voltage equivalent signal to 0 during the waiting time Td.

Blocks 406a and 406b in FIG. 10 are used to further bring the effects of the present invention closer to perfection, when there is no change in the energization signal to each transistor output to the input/output port 4c during the instruction cycle.

Since no transistor energization prohibition time occurs, the waiting time Td is set to O to prevent the voltage waveform from decreasing. That is, the inverter voltage vector vt7 in the previous command cycle is compared with the current inverter voltage vector v1 in the program controller 406a, and if there is no change, the waiting time Td is set to 0 in the block 406b, and only the actual transistor energization prohibition time is set. The AC output voltage equivalent signal is O.

〔Effect of the invention〕

In a system that digitally controls the paper pressure or magnetic flux of an induction motor driven by a PWST inverter, it is possible to simplify the voltage detection method and utilize the memory function of digital control to accurately measure the inverter output voltage, which often pulsates. A method suitable for control that enables detection can be provided.

Sumida1: Although the present invention has been described using one firing bundle control of a three-phase squirrel cage induction motor as an example, it can be applied to many control methods that accurately control the AC output voltage of a PWM inverter.

[Brief explanation of drawings]

Figure 1 shows a PW using the output voltage detection method according to the present invention.
The main circuit diagram of the M inverter, Figure 2 is the main circuit diagram of an example of a conventional PWM inverter, Figure 3 is the main circuit diagram showing the high-speed switching elements of the PWM inverter replaced with contacts, and Figure 4 is the switch state variables. A vector diagram illustrating a table,
Figure 5 is a block diagram showing an example of the internal configuration of the control circuit;
The figure is a magnetic flux vector diagram, Figure 7 is a limit cycle diagram, Figure 8 is a flowchart of an example of a group of commands required for conventional control, and Figure 9 shows the effect of waiting time on the output voltage according to the flowchart in Figure 8. The graph shown in FIG. 10 is a flowchart of an example of a group of instructions for a control circuit when the output voltage detection method according to the present invention is used. 1...DC voltage source, 2...PWM inverter, 3...Induction motor, 4...Control circuit, 5a to 5c...Current Detector, 6a-6c
, 7... Voltage detector, Q1 to Q6...
Transistor, D1-D6...Diode, S
a~Sc......Switching contact◎

Claims (3)

[Claims] (1) At least includes a DC voltage source, a three-phase voltage type PWM inverter, a three-phase induction motor, and a DC voltage source voltage detector, and the control circuit includes a microprocessor, ROM memory, input/output ports, and connections between these chips. The PWM inverter includes at least an address bus and a data bus, and controls the voltage supplied from a DC voltage source to a three-phase induction motor and the magnetic flux generated by the motor according to an instruction program stored in a ROM memory in advance.
A PWM inverter output voltage detection method that obtains a PWM inverter AC output voltage equivalent signal by multiplying a switch state variable representing each phase output voltage of the inverter and the DC voltage source voltage detector. (2) The output voltage detection method of a PWM inverter according to claim (1), wherein the energization prohibition time of the switching elements of the PWM inverter is such that the AC output voltage equivalent signal of the PWM inverter is set to 0. (3) Compare the switch state variable representing the output voltage of the PWM inverter with the switch state variable in the previous instruction cycle that is temporarily stored in advance, and only if there is a change in state, the PWM inverter is activated for the energization prohibition time. An output voltage detection method for a PWM inverter according to claim 1, wherein the AC output voltage equivalent signal is set to 0.