JPH0157308B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for processing measured values during range switching. More specifically, the present invention relates to a measured value processing method that alleviates discontinuity in measured values when switching ranges.

For measured values with a wide variation range, measurements are performed while switching ranges. In that case, the error of the measurement circuit differs depending on the range, so
When changing the range, there will be discontinuity in the measured values. Changes in measured values due to this discontinuity may appear larger than the allowable range if the range errors have opposite polarities. It is inconvenient that range switching causes such a change when there is no actual change in the measured value.

An object of the present invention is to provide a measured value processing method that alleviates the discontinuity of measured values when changing ranges.

The present invention includes: a variable range amplification section that increases a signal under test;
An analog/digital converter converts the output signal of this amplifier into a digital signal, and an analog/digital converter that processes the output signal of this analog/digital converter to obtain a measured value and adjusts the gain of the amplifier depending on the signal under test. using a measured value processing device having a measured value processing section, the measured value processing section sets the minimum allowable range in the amplifier section according to the signal under test, and in that range, the measured value is near the range switching point. When the standard value set in the range is exceeded, the normal measured value is obtained by the weighted average of the measured value in that range and the measured value in the range immediately above, and the weighting coefficient for the measured value in that range is calculated. This method of processing measured values at the time of range switching is characterized in that the weighting coefficient is gradually decreased as the measured value increases, and the weighting coefficient for the measured value in the upper range is gradually increased as the measured value increases.

Hereinafter, the present invention will be explained in detail with reference to the drawings.

FIG. 1 is a diagram showing an example of the configuration of an apparatus used to implement the method according to the present invention. In Figure 1,
1 is a variable range amplification section, 2 is an analog/digital conversion section (hereinafter referred to as an A/D conversion section), 3 is a measured value processing section, and 4 is a range switching section.

The amplifying section 1 receives the signal under test Ex, amplifies it in a range determined by the range switching section 4, and supplies the amplified signal to the A/D converting section 2. It is assumed that the amplifier section 1 maintains input/output linearity up to a 5% overrange. The A/D converter 2 converts the amplified signal into a digital signal and provides the digital signal to the measured value processor 3 . The measured value processing unit 3 multiplies this digital signal by a magnification according to the range to obtain a digital measured value representing the value of the signal under test Ex.
This measured value is given to the next stage device (not shown) that utilizes it. The measured value processing section 3 performs A/
It instructs the D conversion section 2 to perform a conversion operation, and also gives a range selection command to the range switching section 4. The functions of the measured value processing section 3 are realized by a microprocessor or computer program.

The measured value processing section 3 performs measured value processing as described below near the range switching point to alleviate discontinuity at the time of range switching. Figure 2 shows a conceptual diagram of range switching. In FIG. 2, the vertical axis is the true value of the signal under measurement Ex, and the horizontal axis is the measured value obtained by the measured value processing section 3. The signal under test Ex and its measured value should originally be equal (dotted chain line),
If there is an error in the amplifier section 1 or the A/D converter section 2, they will not be equal. Since this error differs depending on the range (solid lines a, b), even if the value of the signal under test Ex is the same, when the range is switched, a discontinuous change occurs in the measured value (dotted line). The measured value processing section 3 shifts this range switching as shown by the thick line to alleviate the change in the measured value.

The operation of the measured value processing section 3 is as follows. First, set the range to the maximum and obtain the measured value, then find the minimum range within which this measured value falls within 105%, set it in the amplifier 1, and obtain the measured value in this range. And this measurement value is for example 97, 100, 103
are compared with each reference value, and based on the comparison results, the following weighted average is calculated to obtain the normal measurement value.

That is, (1) When |measured value|≦97, measure four times in the above minimum range and use the average value of the obtained measured values as the normal measured value.

(2) When 97<|Measurement value|≦100, measure three times in the minimum range above and once in the range immediately above it, and take the average value of the obtained measurement values as the normal measurement value. do.

(3) When 100<|measured value|≦103, measure twice in the minimum range above, then measure twice in the range immediately above it, and take the average value of the obtained measured values as the normal measured value. do.

(4) When 103<|Measurement value| Measure once in the above minimum range, measure three times in the range immediately above it, and use the average value of the obtained measurement values as the regular measurement value.

In other words, using the measurement value in the minimum range that allows the measurement value and the measurement value in the range immediately above it,
The average value is obtained by changing the weighting coefficients in a complementary manner in order and as a normal measurement value.

FIG. 3 shows details of the transition of measured values near the range switching point when such measured value processing is performed. In Figure 3, the range is switched from the a range to the b range immediately above it, where the a range is, for example, 0 to 100 mV and includes an error of +0.1%, and the b range is, for example, 0 to 200 mV. It is assumed that this includes an error of -0.1%.

Now, assuming that the true value of the signal under test Ex is 97 mV, the measured value by the a range is 97.1 mV due to an error of +0.1%. Therefore, since 97<|measured value|<100, method (2) above is applied,
A normal measurement value is determined by a weighted average of three measurements in the a range and one measurement in the b range. The measurement value from the b range is 96.8mV due to an error of -0.1%. In addition, b range is 0-200m
Since it is V, an error of -0.1% is -0.2 mV.
Therefore, the measured value based on the weighted average is 97.025 mV. That is, the measured value, which should be 97.1 mV if measured in the a range, is corrected to 97.025 mV. As a result, the locus of the measured values in FIG. 3 shifts from the locus of the a range to a locus a' that is approximately parallel to the locus of -0.075 mV. The measured value of the a range is 97
While the value is greater than mV and less than 100 mV, the weighted average measurement value is on the trajectory a'. That is, according to the weighted average, a measured value that is discontinuously changed by -0.075 mV from the measured value that would be obtained by the a range is obtained. However, since the amount of change is less than 0.1% of the a range, there is no problem.

When the true value of the signal under measurement Ex becomes 100 mV, the measured value of the a range becomes 100.1 mV, so the averaging using the method (3) above is performed. At this time, the measured value of the b range is 99.8 mV, so the weighted average value
It becomes 99.95mV. Since the measured value based on the weighted average up to that point was on the trajectory a', it should have been 100.025 mV for the true value of 100 mV, but due to the change in the weighting coefficient, the value changed by -0.075 mV. In other words, the trajectory of the measured value is further −0.075
The locus a'' is shifted approximately parallel to mV.The amount of change in the measured value at this time is also less than 0.1% of the a range, so there is no problem.

Similarly, each time the measured value in the a range exceeds each reference value, the weight for average calculation is changed, and each time the measured value changes discontinuously by -0.075 mV, and eventually reaches the trajectory of the b range. Arrived and finished changing the range. That is, range switching is performed in such a way that discontinuous changes in measured values are alleviated to within a permissible range. If you have a standard value increments for comparison, you can make it finer.
If the steps for changing the weighting coefficient are made smaller accordingly, the amount of change at the discontinuous point can be further reduced. Further, by setting the locus of transition from the a range to the b range using an appropriate function, continuous transition between ranges can be performed. Note that this corresponds to a case where the weighting coefficient is continuously changed.

As described above, the present invention calculates a normal measured value by the weighted average of the measured value measured in a certain range and the measured value measured in the range immediately above it near the range switching point, and The weighting coefficients are changed in a complementary manner depending on the measured value. Therefore, it is possible to realize a measured value processing method that alleviates discontinuity during measurement when switching ranges.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an example of the configuration of an apparatus used for carrying out the method according to the present invention, and FIGS. 2 and 3 are diagrams illustrating the operation of the apparatus shown in FIG. 1. 1... Amplification section, 2... A/D conversion section, 3... Measured value processing section, 4... Range switching section.

Claims (1)

[Claims] 1. A variable range amplification section that increases the signal under test, an analog-to-digital conversion section that converts the output signal of this amplification section into a digital signal, and a process for processing the output signal of this analog-to-digital conversion section. a measured value processing unit that calculates a measured value based on the measured value and adjusts the gain of the amplification unit according to the signal under test; When the minimum range is set in the amplification section and the measured value in that range exceeds the reference value set near the range switching point, the weighted average of the measured value in that range and the measured value in the range immediately above is used. Therefore, a normal measured value is obtained, and the weighting coefficient for the measured value in that range is gradually decreased as the measured value increases, and the weighting coefficient for the measured value in the upper range is gradually increased as the measured value increases. How to process measured values when switching ranges.