JPS63163282A - Multilevel multiplexing function arithmetic system - Google Patents

Abstract

PURPOSE:To eliminate the need to adjust variance in arithmetic error among respective phases which is generated when an arithmetic circuit is simplified and performs arithmetic independently by performing specific function arithmetic processing for plural analog values xi (i=1-n) which are multiplied by time division and obtaining function values f (x1-xn) which have plural analog values as arguments. CONSTITUTION:A circuit which detects the effective value (RMS) of a three- phase AC signal consists of an analog switch 1, an analog switch opening/closing control part 2, a squaring circuit 3, a low-pass filter 4, and an extractor 5. Then sampled values which are time-divided by a switch 1 indicates the mean value in the period and the circuit 3 squares the sampled values successively. Its arithmetic results are smoothed by the filter 3 to obtain the square mean value of three phases, whose square root is extracted to obtain the square root of the square mean value, i.e. RMS value. Thus, the RMS of three or many phases can be obtained by one single-phase RMS converter and an arithmetic error among the phases is eliminated because of the arithmetic circuit of one system.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention is an analog measurement and control system in which the average value, effective average value, etc. of a plurality of measured values are input by time-division multiplexing in the previous stage. Regarding the method of performing a functional operation on a value and calculating with one arithmetic circuit, it is particularly useful for power measurement, W!
, relates to a function calculation method suitable for force conversion control.

[Conventional technology]

Conventional 1 For example, when measuring the effective value of three-phase voltage and current,
First square the value of each phase, add them together, and then square root (,
r-) was calculated to obtain the effective value average. However, in this method, although the same calculation is performed for each phase, independent multipliers are required for each phase, and after that, it is necessary to perform addition and pass through a square rooter.

There was a problem that the circuit scale became large. On the other hand, as an analog element, there is an effective value calculation element (for example, BurrBrown's RMS-Converter 4340, but it is for single phase use).

[Problem that the invention seeks to solve]

The above conventional technology requires an arithmetic circuit for each set of weights to detect the average effective value of multiphase currents, which poses problems in terms of economy and device size6.In addition, the RMS calculation for single phase What is the effective average value if you just add the outputs of the elements as they are? The problem was that I couldn't do it.

An object of the present invention is to calculate the effective average value RMS of three-phase AC, for example.
(Root Mean 5quare) is calculated using an arithmetic circuit for one phase, thereby significantly reducing the number of circuit elements and eliminating the need to adjust for variations in calculation errors for each phase that would occur if calculations were performed independently. This is intended to enable use in similar calculations other than the effective value calculation of three-phase AC as an example.

[Means for solving problems]

The above object is achieved, for example, when calculating the effective value of a three-phase alternating current, by first time-sharing the measured values of the three phases using an analog switch, and directly inputting the output to a single-phase effective value calculation circuit.

[Effect]

This principle is based on the fact that if the sampling period by time division using an analog switch is sufficiently smaller than the period of the measured signal, the sampling value represents the average value of that period, and the single-phase effective value calculation circuit calculates the square of the sampling value. are calculated sequentially. By smoothing the calculation result using a filter, a two-east average value of the three phases can be obtained, and by square rooting it, the square root of the root mean square value, that is, the RMS value can be obtained. In this way, it is possible to obtain three-phase or multi-phase RMS with - single-phase RMS converters, and at the same time, since it is a single-system arithmetic circuit, there is no arithmetic error between phases.

〔Example〕

An embodiment of the present invention will be described below with reference to FIG. 1st
The figure shows the effective value (
This is an overall block configuration diagram of a circuit for detecting RMS). Analog switch opening/closing control section 2, square circuit 3. Low frequency fill 4. It consists of a flattener 5.

FIG. 2 shows operating waveforms of each part of the circuit of FIG. 1.

First, in FIG. 1, a three-phase AC signal XI.

X2F X8 is connected to its output at equal time intervals by analog switch 1 to obtain a time division multiplexed signal X as shown in FIG. From this X, a square circuit 3 obtains x2, a low-pass filter 4 obtains an average value x2, and a square root circuit 5 obtains an effective value average C゜.

In this case, by setting the analog switch switching frequency sufficiently high relative to the input signal frequency, it is possible to achieve accuracy and responsiveness equivalent to those obtained when calculation is performed using individual circuits.

Figure 3 shows a case where the same principle is applied to other functions and the output of the same analog switch is used in multiple function calculation circuits. For example, the output AV is Xi + X2
It becomes the average value of g xs.

In addition, if X is compared with the reference value Xa by the comparator 6a and the flip-flop 7a is set with the output of the comparator 6a, an output will be generated when it is detected that either of input No. 43 x1, X2 + X8 has become an excessive value. Get 0VIh. Further, by comparing the absolute value (x) of X with the reference value Xb by the comparator 6b by the insulation value circuit 8, an excessive absolute value detection output OV D 2 of the AC signal is obtained.

FIG. 4 shows an example of three-phase AC power detection, in which two analog switches la are used.

1b, each pair of currents 11 g xi
113 voltages ex, ex, es selected synchronously
.

The average power value P can be obtained by calculating the product of e by the multiplier 9 and smoothing the product e-i by the filter 4.

〔Effect of the invention〕

As described above, according to the present invention, a plurality of signals of the same quality can be
It is possible to simplify the calculation circuit for obtaining calculation values used in power control, such as effective values, average values, and sum-of-products values (e.g., electric power), which require processing that includes the same calculation, and has the effect of increasing economic efficiency. . Furthermore, by using the same arithmetic circuit, there is no variation due to errors in circuit constants that occur in individual cases.

In addition, effects such as easier adjustment can be obtained.

[Brief explanation of the drawing]

Figure 1 is an overall block diagram of an embodiment of the present invention, Figure 2 is a diagram showing its operating waveforms, Figure 3, Figure 4

Claims (1)

[Claims] 1. A plurality of analog values x_1 (i=1, 2,..., n) are sequentially switched and time-division multiplexed by an analog switch, and a predetermined functional calculation process is performed on the multiplexed analog values. A multi-value multiplex function calculation method characterized in that the function value f(x_1, x_2, . . . , x_n) is obtained using the plurality of analog values as arguments. 2. The multi-value multiplexing of item 1, wherein the functional calculation process consists of a square calculation process, a filtering process, and a square root calculation process, and the effective value average of the analog values is obtained by the series of calculation processes. conversion function calculation method. 3. The multi-value multiplex function calculation method according to item 1, wherein the function calculation process includes a low-pass filtering process, and a scalar average value of the analog values is obtained by the filtering process. 4. The multivalued multi-value of item 1, characterized in that the function calculation process includes a comparison process of comparing the analog value with a predetermined value, and an excess value detection value of the analog value is obtained by the comparison process. Multiplex function calculation method. 5. The multi-value multiplex function calculation method according to item 1, characterized in that many types of function values are obtained simultaneously by performing a plurality of types of function calculation processing on the multiplexed analog values. 6. Ordered pairs formed by providing a plurality of sets of the plurality of analog values (
x_i, y_i, ..., z_i), i=1, 2, ...,
n are synchronously time-division multiplexed using separate analog switches, and a predetermined function calculation process is performed on the multiplexed sets of analog values to obtain f{g(x_1, y_1,
..., z_1), ..., g(x_i, y_i, ..., z
_i), ..., g (x_n, y_n, ..., z_n)}
The first term is a multi-value multiplexing calculation method characterized by obtaining the following.