JPS62161061A - Method for measuring electric power - Google Patents

Abstract

PURPOSE:To enable highly accurate measurement, by alternately sampling input voltage proportional to voltage and input voltage proportional to a current by an A/D converter. CONSTITUTION:A current input terminal 1, a voltage input terminal 2, an analogue multiplexer 3 and an A/D converter 4 etc. are provided. The input of a voltage proportional signal is alternately sampled from the terminal 1 at every definite time to be stored in a memory 6 through CPU5. Thereafter, the voltage input signal taken in from the terminal 2 is regarded as a sinusoidal wave and subjected to numerical analysis to calculate the max. value thereof. By this method, the voltage of the current input signal inputted from the terminal 1 at a measuring point is interpolated from the sinusoidal wave curve calculated by numerical integration and calculated. The result obtained by multiplying two numerical values is again subjected to numerical integration to make it possible to measure electric power with high accuracy.

Description

[Detailed description of the invention] Industrial applications The present invention relates to a method for measuring power.

Conventional Technology Analog instruments have long been used to measure electrical quantities such as current, voltage, and power. Moving coil type for voltage and current,
There are various types such as movable iron piece type, rectifier type, and amperometric type.
Both utilize the fact that torque is generated in the coil by placing the coil in a magnetic field and passing a current through it. Since this torque is proportional to the square of the current flowing through the coil, the electric value can be measured by measuring the deflection of the coil. The current dynamometer type uses the fact that current flows through both a fixed coil and a moving coil, and the torque is proportional to the product of the two, and is generally used as an analog wattmeter.

On the other hand, in recent years, computerization has progressed and electrical measuring instruments have also become digital. The voltmeter is the most famous digital meter, and its basic principle is to convert DC voltage into A/D and display it. In the case of alternating current, it is first full-wave rectified, then integrated, and the average value is A/D converted. In the case of current, it can be measured in the same way by using the voltage generated when it is passed through a resistor. However, in the case of electric power, there is no good method, and some measure it using the square proportional portion of the base-collector characteristic curve of a transistor.

Problems to be Solved by the Invention However, in conventional methods, for example, in the case of an analog type, the principle is mostly for direct current, and for alternating current, the average value is indicated. In the case of alternating current, a wide range of effective values are used for electrical measurements, and the only type that can truly be used for this purpose is the moving iron type. , is not suitable for distorted waves containing high-frequency components.In terms of accuracy, these are only meaningful at the commercial frequency of sinusoidal alternating current.

On the other hand, in the case of digital type, conventionally it is average value type, so
It was multiplied by a certain coefficient to display the effective value. Again, since the frequency is assumed to be a commercial frequency, measurement of distorted waves and high frequencies will cause errors.

Recently, voltage and ammeter meters with effective value indications, as well as high-frequency wattmeters, have become available. However, the frequency of these is limited to about 2KHz, and the price ranges from 600,000 yen to 100,000 yen.
It is extremely expensive at 1,000,000 yen. Even if the fundamental frequency is a commercial frequency, its waveform may be significantly distorted depending on the object to be measured, and a measuring instrument is required to have high frequency characteristics. Moreover, even though it is a digital method, it is only the display part.
Since the part that obtains the DC voltage proportional to the power is an analog type, problems such as offset and temperature characteristics become more troublesome as the precision increases.

In addition to the above-mentioned measurement problems, there are interface problems when processing the above-mentioned data with a computer or the like. For such purposes, the measuring instrument can output digital values directly, use a general-purpose interface such as GP-IB, or output a proportional analog DC voltage. There is. To connect these to a computer, GP-IB, parallel interface, analog interface, etc. are used. There is no big problem with this method if it is basically static data, but when measuring a large number of dynamic data, GP-IB and parallel methods are used due to responsiveness and wiring problems. is difficult. In this case, an analog output method is often adopted, but in the case of this interface, analog-to-digital conversion is required on the computer side, and in principle, analog DC voltage is being measured. These problems would be solved if the A/D converter could directly measure alternating current.

The present invention has been made in view of this situation, and its object is to provide a low-cost, high-frequency characteristic, high-precision digital calculation method for measuring power that can be used directly on a computer.

Means for Solving the Problems In order to solve the above problems, the present invention alternately samples input voltage proportional to voltage and input voltage proportional to current using an A/D converter, and stores them once in memory. After storing the voltage input as a sine wave and calculating the maximum value by numerical integration, calculate the voltage input value at the current input sampling point from the phase angle, The result of multiplying by the value of is numerically integrated, and the average value is taken as the power value.

According to Shannon's sampling theorem, if an analog signal is sampled at small intervals Δt, the original signal can be restored using the sampled values. In other words, Δt
This means that when digitally sampled at , signals up to a frequency of □ can be processed by 2Δ. From this, if the sampling interval of the A/D converter is shifted from lQMs, the frequency characteristics of the current and voltage signals will be 25 k H7. Also, since sampling is done alternately, if the signals are multiplied as is, a phase error will occur for the sampling interval, but since the value of one signal at the sampling point of the other signal is determined by calculation, this problem does not occur. . Furthermore, since all digital calculations are performed after A/D conversion, problems with offset and temperature characteristics do not occur. Since the measured values are the results of calculations in the computer, there is no need to newly import them for data processing purposes. Moreover, no special hardware is required for processing other than an A/D converter, and the cost can be reduced.

As described above, according to the present invention, it is possible to perform low-cost power measurement suitable for computer processing with high precision and high frequency characteristics.

EXAMPLES Hereinafter, examples of the present invention will be described in detail with reference to the drawings. FIG. 1 is a main configuration diagram of an embodiment of the present invention. In Fig. 1, 1 is an AC signal input terminal proportional to the current;
2 is an AC signal input terminal proportional to the voltage, 3 is the terminal 1
Multiplexer for analog switching between 7 and 2 terminals, 4
5 is an A/D converter that converts the output of the analog multiplexer into digital; 5 is a CPU 16; and 7 is a high-speed arithmetic unit. A concrete application is a configuration in which an A/D conversion board is added to a personal computer.
Examples of high-speed arithmetic units include Intel's 8087.

The operation of the power measuring method of the present invention configured as described above will be explained below with reference to FIGS. 1 and 2. First, FIG. 2 shows the processing of the CPU 5 in FIG.
A waveform 10 shown in FIG. 3 is a current proportional signal input to terminal 1 in FIG. 1, and a waveform 11 is a voltage proportional signal input to terminal 2 in FIG. These two signals are switched by an analog multiplexer 3 at regular time intervals Δ, and converted into digital values by an A/D converter 4. This converted data is stored in the memory 6, where 8 in FIG. 2 is a data area for the current proportional signal, and 9 is a data area for the voltage proportional signal. In this data area, data at a certain point of the current input signal, for example, 12 points in Figure 2, is stored in 81 at the time of sampling start, and then the input is switched to terminal 2 in Figure 1 by multiplexer 3. , Δ, the point of the voltage input signal, for example, data of 13 is stored in 91, and thereafter, the data of point 14 is stored in 82.
For each waveform signal, data at points for each time twice the data sampling interval Δt is stored in chronological order, such that 15 points become 92 points.

For alternating current power, it is sufficient to calculate the average of the products of double signs in one cycle, so one cycle's worth of data is sufficient. Since the sampled data is calculated using a numerical integration method, it is desirable that the sampling intervals be equal intervals, and this time precision affects the final error. Therefore, a reference clock generated by a crystal oscillator is used, and A/D conversion is performed in synchronization with this reference clock. In addition, the sampling time Δt is 2 K when the fundamental frequency is 501 Tz.
Considering the response up to H7, it follows from Shannon's theorem. In the case of the present invention, two signals are switched and sampled, so the sampling interval is half this, 125M5. In other words, the switching time of analog multiplexer 3 and A/
It is sufficient that the total conversion time of the D converter 4 is less than this. In this way, data sampled alternately at equal intervals of Δ is processed, and AC power is the product of the instantaneous values of current and voltage, averaged over one cycle. To do this, it is necessary to apply voltage and current data at the same time, and when sampling is performed by switching with a multiplexer, the sampling data is
As shown in the waveform in the figure, there is a deviation of Δt. In other words, data 81 and 91.82 stored in memories 8 and 9 in FIG.
and 92 are values temporally shifted by Δt. If they are multiplied in this state, a phase error of Δt will occur, which will affect accuracy. For this reason, it is necessary to convert one of the sampling data to data having the same timing as the other. Since sampling is performed at equal intervals, the value that is intermediate between two consecutive pieces of data is the value that is determined in terms of timing. In other words, in Figure 3, 1
We will find 17 from 3.15. This method includes
Linear interpolation, the value of 17 = (value of 13 + value of 15)/2 is the simplest, but the error increases if the value is near the maximum value or if the waveform is steep. Therefore, the present invention solves this problem by regarding the voltage as a sine wave and interpolating it.

In general, voltage waveforms are less likely to be distorted, but depending on the case, there will be no problem if applied to the side with less distortion.

This method will be explained in a little more detail. First, find the maximum value from the sampling data. This method numerically integrates the square of the sampling data using Simpson's rule, calculates the square root of the average, and sets it as the effective value. This multiplication is the maximum value V
m. The voltage input is regarded as a sinusoidal curve having this maximum value, and v=VMSinθ. On the other hand, the angle corresponding to this time is calculated from the sampling interval Δt. In other words,
50 Hz Next, from the formula obtained above, the angle with respect to the value of the voltage input signal is calculated using an arc sine.

Find this θi for the first value of sampling,
Determine which angle of 360° the first point is. Once the angle of the first point is determined, the angle Δθ before that angle becomes the angle of the 12 points in FIG. The remainder is this angle plus 2Δθ. If you calculate this angle using the formula obtained earlier, it will be 16.17 in Figure 3.
In other words, the voltage value at the measurement point of the current input signal is determined.

From this result, the instantaneous power can be found by multiplying data at the same timing such as 12, 16, 14, and 17, and by numerically integrating this result again using Simpson's rule and averaging the results, the average power is obtained.

Effects of the Invention As described above, according to the present invention, not only can high-precision, high-frequency power be measured without the need for a special circuit, but also the digital calculation method allows for stable measurement with less influence from offsets, etc. This makes it possible to perform accurate measurements. In addition, since it is possible with a simple hardware configuration such as an A/D conversion board in a personal computer, it is very convenient when the measured data is to be processed again by the computer.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the main configuration of an embodiment of the present invention, FIG. 2 is a diagram showing a sampling data memory, and FIG. 3 is a diagram showing input waveforms. 1... Current proportional signal input terminal, 2...
Voltage proportional signal input terminal, 3...Analog multiplexer, 4...A/D converter, 5...
・CPU, 6...Data memory, 7...
・High-speed arithmetic unit, 8...Current signal data memory,
9... Voltage signal data memory.

Claims (2)

[Claims] (1) a terminal for inputting a signal proportional to current, a terminal for inputting a signal proportional to voltage, an analog multiplexer for switching the terminals, and an A/D converter for converting the output of the analog multiplexer into a digital value; A CP that takes in the data of the A/D converter and processes it.
U, a memory for storing the captured data, and a high-speed arithmetic unit for data processing, and the terminal and voltage proportional signal input are alternately sampled at fixed time intervals and stored in the memory via the CPU. After that, the voltage input signal input from the terminal is regarded as a sine wave and numerically integrated to find its maximum value, and from this, the voltage at the measurement point of the current input signal input from the terminal is determined as a sine wave obtained by the numerical integration. A power measurement method that measures power by calculating by interpolating from a curve, and then numerically integrating the result of multiplying the two. (2) The power measurement method according to claim 1, wherein a quadratic curve passing through three consecutive points of the voltage input signal is determined, and from this, the value at the current input signal measurement point is determined by interpolation.