JPS62233071A - Pulse controller for inverter - Google Patents

Abstract

PURPOSE:To prevent noise and torque ripples by causing respective inverter constituent elements to generate ignition-extinction pulses in a manner of distrib uting a given time in proportion to peak values of three-phase sine waves. CONSTITUTION:A current-type inverter converts a DC power 1 into AC through a DC reactor 4 and feeds a load induction moror 7. A controller of this inverter is furnished with one-chip microcomputer 10 performing a pulse pattern control and is further composed of input/output 101, 106, ROM 103, RAM 104, ALU 105, pulse pattern(event)-setting register 107, holding register 109, associative memory 110, comparator 112, excution controller 113 and others. Accordingly, numerical values, on which peak values of three sine waves with 120 deg. phase difference are to be retrieved, are obtained at every given time and an element, of which ignition is to be extinguished, is determined within said time. Also, a pulse width data-computing part operates to obtain data for dividing a given inproportion to peak values of three since waves according to a phase information.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an inverter pulse control device, and more particularly to an improvement of an inverter pulse control device suitable for converting output voltage and current into sine waves.

[Conventional technology]

6 semiconductor switches (GTO or diode with reverse blocking function and GTO or transistor series port M) 51,5
Figure 2 shows a current source inverter in which 2, 53, 54, 55, 56, DC reactor 4, and output end capacitor 6 are connected.
1 Output waveform analysis of a sine wave output current source inverter" (by Shigeta Ueda and 3 others), the pulse control method is shown in "Analysis of the output waveform of a sine wave output current source inverter" (Shota Ueda et al.). A common method is to compare the values and obtain a pulse pattern based on the magnitude relationship. However, in this case, a carrier wave generation circuit, a modulated wave generation circuit, a comparator circuit, etc. are required, which makes the circuit complicated.If these pulse control devices are constructed with analog circuits, variations in pulse width due to temperature, aging, etc. may become a problem. Become. On the other hand, from this point of view, attempts have been made to construct digital circuits using microcomputers, etc., but in this case, if a method using carrier waves and modulated waves is adopted, the microcomputer is constantly constrained to compare W waves. This may cause problems such as the computer becoming too crowded and almost no other processing possible.

Furthermore, converting the output waveform into a sine wave over a wide frequency range is a problem that is almost impossible due to the configuration of the modulated wave generating circuit.

[Problem that the invention seeks to solve]

Although the above-mentioned conventional technology has been devised to convert the output waveform into a sine wave, its level of perfection is still insufficient.For example, when an induction motor is connected as an inverter load, harmonics This caused problems in the form of motor noise, torque ripple, etc.

An object of the present invention is to provide means for generating a pulse pattern so that the output of an inverter can be made into a sine wave.

[Means for solving problems]

The above object is achieved by generating extinction pulses in each component of the inverter so as to divide a predetermined time in proportion to the peak value of the three-phase sine wave.

[Effect]

The means for determining the phase operates to determine at predetermined time intervals numerical values that are the basis for retrieving the peak values of the three sine waves having a phase difference of 120 degrees. The means for determining the extinguishing element operates to determine the element to be extinguished within at least a predetermined period of time in accordance with the phase information.

Pulse width data calculation operates to obtain data that proportionally divides a predetermined time into peak values of three sine waves according to phase information. Pulse pattern generation operates to distribute a pulse pattern to the corresponding element according to the element to be turned off and the proportionally divided pulse width data.

Due to these effects, the output waveform of the inverter can be made into a sine wave, so noise and torque ripple will not be generated even when an electric motor is used as the load.

〔Example〕

An embodiment of the present invention will be described below with reference to FIG.

1 is a DC power supply, 2 is a frequency command input terminal, 3 is a phase command input terminal, and 4 is a DC reactor.

51 to 56 are six transistors that constitute a current source inverter, 6 is a capacitor for overvoltage suppression, 7 is an induction motor as an example of a load, 8 is an extinction/ignition signal line, and 10 is a one-chip microcomputer that performs pulse pattern control. It is. Its interior includes a command input port 101, an internal bus 102. ROM103 that stores programs, pulse width data tables, etc.
, a RAM 104 used as temporary storage and a register; an ALU 105 that executes calculations; and an event setting register 107 that sets an event to output a predetermined pulse pattern (event) to an output port 106. a time setting register 108 for setting the time when to enable this event;
Holding register 1 that concatenates and holds the contents of both setting registers
09. Associative memory 110 in which several sets of setting data set in holding registers are stored. A comparison unit 112 compares the timer 111 with the set time contents in the associative memory 110 and generates an output when they are equal, and an execution controller 113 receives a trigger from the comparison unit 112 and controls the output of the set event to the output port 106. It consists of etc.

Next, the operation in this configuration will be explained using a flowchart. FIGS. 3 and 4 show flowcharts of two major processing systems.

FIG. 3 shows an event calculation processing program F 100 for determining an event, that is, a pulse pattern, to be generated at the output port 106.
0 is a schematic flowchart. First, FIloo reads the frequency command ω1-1 and the phase command θfist from the input port 101. Of course, this ωl and θ fist are also one-chip microcontrollers.
When calculating inside 0, port reading is not necessary. This frequency command ω1 is integrated every fixed time Δtl and added to 0 phase commands to process 6 total phases.
Among the six modes in which the 0-order electrical angle 3601 found in 1200 is divided into 60 #s, which mode should the pulse pattern be output for the total phase 0 found this time?

In other words, the output event is determined by the F1300 process depending on the number of 0 copies. Note that the relationship between θT and the six modes will be explained in detail later. Finally, the pulse pattern is changed during the interrupt interval Atz, and the process of determining the time until the change is made by referring to the data table with 0 units is performed in F1400. Through this processing, the event details and event change times to be set in the two registers are determined.

Next, FIG. 4 shows a process F2000 in which the two previously determined items are set in the associative memory for output boat control. First, the F 2100 determines whether the event settings and time settings required for the six transistors have been completed. If NO, the F 2200 performs the corresponding event settings, and the F 23
The event change time is set at 00 and the process ends.

Next, FIG. 5 shows how these two processes F 1000 and F 2000 are activated over time.

The event setting process F2000 is activated in synchronization with a timer interrupt that occurs every Δtl. On the other hand, event calculation processing F 1
000 is activated by a second timer interrupt that occurs prior to the timer interrupt, and completes the event calculation process before F 2000 is activated. The reason why the event calculation process F1000 is completed immediately before the event calculation process F2000 is to enable the latest data to be used by the event calculation process F2000. Of course, if a dead time element corresponding to the timer interrupt interval can be included, F1000 may be performed following F2000. In that case, the time required for interrupt determination becomes shorter, so the interrupt interval Δt1 can be set shorter, and the converter can operate at a higher frequency.

In the present invention, once the predetermined event and time settings are completed, the associative memory section within the microcomputer takes over control of the output port, thereby freeing the main processor section from output processing.

Next, the determination of the pulse pattern in process F1300 will be explained using FIG. In the case of inverter control, the three-phase AC output has an electrical angle of 360′ as shown in FIG.
It is obtained by dividing into 6 sets of modes M5 to MB for each 0@ and repeating this process. Then, six sets of modes M1 to Me having an interval of 6o0 may be selected by the overall phase θT, and the pulses may be distributed.

The flowchart is shown in FIG.

As can be seen from FIG. 7, the waveforms of each mode of M 1 '' M s are as follows, considering the difference in phase and the difference in positive and negative polarity.

The waveforms are exactly the same. That is, the Mz mode is the Mr mode with the positive and negative reversed, the M8 mode has the same waveform as the M1 mode, but has a different phase, and the M4~
Similarly to Mx and Ms, Mg also has the same waveform as the M1 mode waveform. Therefore, the PWM pulse pattern creation procedure in each mode is the same as the PWM pulse pattern creation procedure in the ML mode.
The transistor group to be turned on or turned off may be selected based on the overall phase θT.

Next, regarding the pulse pattern creation method in M1 mode, once the 0 phase θ is determined as explained in Figure 8, in order to make the output a sine wave, the pulse width of each phase must be distributed to the ratio of the peak values of the sine waveform. Good. That is.

When the interrupt interval is Δtl, the pulse width TI of the U phase is T I =Δt 1 ・sinθt
--(1) The pulse width T2 of the W phase is T z =Δjx・5in(θt −240”)
--(2) The pulse width T8 of the ■ phase is T8 child T1+Tx...
...(3) may be controlled.

The processing speed can be improved by making the calculations of these equations (1) to (3) into a table as shown in FIG. 9, and calculating each of them using the phase zero.

The pulse pattern creation method shown in FIG. 8 described above is also applicable to the cases of modes M2 to Mg, but it is necessary to change the transistors to be turned on or off. The combination of transistors for each mode is shown in Figure 10. Therefore, θ↑
If this is known, the transistor whose mode should be turned off and turned off can be specified, and the width of the turned off turn can be determined by the method shown in FIG.

The above example shows an example of a current source inverter, and since the inverter operates as a simple switch, it has the advantage of not requiring data table processing.

When applied to a voltage source inverter, it is necessary to modulate the amplitude after searching the table, but in this case, it can be obtained by multiplying the values in the table in Figure 9 by the conduction factor γ, and a similar method can be used. can be applied.

FIG. 11 shows examples of operating modes and boat output signals 851 to 856 given to transistors 51 to 56. The reason for the variation in electrical angle in the modes is that the timer interrupt interval Δt1 is asynchronous with respect to the frequency command ω1. This occurs due to certain factors, and in order to eliminate this, it is possible to control Δtz to make it variable according to ω16. Now, a flowchart of the event setting process is specifically shown in FIG. 12, taking the first part of mode 1 in this figure as an example. In Fig. 4, a loop configuration was used for the sake of general explanation, but in reality, the first
As shown in Figure 2, the processing is performed in series. This flowchart starts from time to in FIG. 11 to to+Δt1.
This shows the event setting process for one timer interrupt period up to. First, when an interrupt occurs at to, the transistor (55) that always fires in this mode 1 at F2410 and the transistor (55) that fires until the first event occurs. 51) Two sets of event sets and time sets are performed for each transistor so that the firing signal is generated immediately.

That is, It
171 is generated, and then, as a time set, a predetermined time t4 is added to the current time to, and the result is set in a predetermined register. Since ignition occurs immediately, t- should be chosen to have a sufficiently small value.

As a result, the event and time are set in the associative memory, and a "1" signal is output to 55 and 51 after 1 d has elapsed according to the schedule.

In the F2420, assuming that the operating mode has changed from the previous one due to a sudden change in the phase command θ umbrella, etc., in this mode, a process is performed to confirm the extinguishment of transistors that should be in the extinguished state.The process uses associative memory as in the F2410. However, since the event here is arc extinction, an event is set so that "O" is generated on boats 2, 3, 4, and 6.

Next, F 2430 processes a schedule such that 51 is extinguished at time t o +T x. The event is boat 1
The output is ``'O'', and the time is set to t o + T x.If t- is a certain large value, at this point, multiple events for one output port may set the time within the same timer interrupt. This means that it has been scheduled separately.

Additionally, the F 2440 schedules 53 firings instead of 51 firings.

Note that here, the extinguishing of 51 and the ignition of 53 are at the same time, but in order to prevent overvoltage, the current source inverter wraps the "1" period, and in the voltage source, the T1 period is set to create a non-lap period.
It is also possible to consider changing the time between F 243Q and F 2440.

Further, at the second event occurrence point to+"r1+Tz, a schedule for extinguishing 53 (F2450) and a schedule for igniting 52 (F246G) are successively performed.

In this way, the phase θT is calculated, the transistor to be turned off is determined based on θT, the time for turning off is determined by 0, and finally the transistor to be turned off and its time are paired. The process of creating a schedule is performed every predetermined time period Δt1. This series of processing not only eliminates the problem of the microprocessor being constrained to constant comparison compared to the conventional method of comparing carrier waves and modulated waves, but also achieves the control effect of converting the output waveform into a sine wave. I was able to generate it.

In the above explanation, taking into consideration the vector control of the electric motor, the frequency command number 01 and the phase command θ are used to calculate the overall phase command 0.
Although we have shown an example of calculating ω, we simply calculate the inverter frequency position ω
In the case where only l- needs to be given, the essence of the present invention can be omitted and the 0th position calculated from θth = Σω1*Δt may be used.

In addition, in the schedule processing in the event setting processing F2000, the programmable I1 in the one-chip microcontroller is shown in FIG.
0 function was used, but due to lack of a board or checking the boat output signal, etc., the programmable I10 in the one-chip microcontroller was not used, and the same effect could be obtained by using an external peripheral with the same function. Needless to say, it can be done.

Also, although we have not discussed in detail the changes in the timer interrupt interval Δt1 here, it is also possible to change Tx in proportion to applications where there is a requirement to change the switching frequency due to external factors such as a rise in element temperature. By simply adding a routine that allows Δ to be adjusted to variable machining.

In FIG. 7, when the mode M1 is examined more closely, it can be seen that it is a repetition of the small mode m1 up to 0≦θth≦30@. Therefore, if the process shown in FIG. 13 is performed at the beginning of the process M z '' M s in FIG. 6, the pulse distribution table can control the entire 360'' range with a single angle of 0≦θth≦30°. In this way, processing with a table in a narrow range of 0≦θT≦30° can increase the carrier frequency and make the sine wave approximation more precise.

〔Effect of the invention〕

According to the present invention, since the output waveform of the current source inverter can be made into a sine wave, electromagnetic noise, torque ripple, etc. can be significantly reduced when an induction motor is connected as a load.

Furthermore, since it is only necessary to prepare a pulse width data creation table for 601 electrical degrees or 30 degrees, the number of tables can be reduced.

[Brief explanation of drawings]

FIG. 1 is an overall configuration diagram of the present invention, FIG. 2 is a schematic diagram of a current source inverter to which the present invention is applied, and FIGS. 3 to 13 are diagrams for explaining one embodiment of the present invention. 51-56... Current interrupting new function element, 10... One-chip microcomputer, 107... Event setting register, 10
8... Time setting register, 109... Holding register, 110... Content addressable memory, 106... Output port,
111...Timer, 112...Comparison section, 113...
・Execution controller, F 1000...Event calculation process, F2000...Event setting process, ω14...Frequency command, θ umbrella...Phase command, θT...Comprehensive phase command, T1...th Time until event 1 occurs, Tz + Tz
...Time until the second event occurs, Δt1...Timer interrupt interval.

Claims (1)

[Claims] 1. Means for determining the phase of a three-phase sine wave having a phase difference of 120 degrees in electrical angle; means for determining the inverter component to be turned on and off from the phase value; and a peak value of the three-phase sine wave from the phase value. In the PWM type inverter, the means for determining pulse width data from the phase value is provided with means for determining pulse width data proportional to the phase value, and means for generating a pulse pattern in the element to be turned off and on from the pulse width data. 1. A pulse control device for an inverter, characterized in that a pulse pattern over the entire 360° range is generated by repeatedly using data for 30° or 60° electrical angle.