JPS5833530Y2 - Measured value correction method - Google Patents

Description

[Detailed explanation of the idea] The present invention relates to a measured value correction method.

When measuring a physical quantity, it is common to convert the physical quantity into an electrical signal using an appropriate converter, and then measure the signal using an electrical measuring device.

The correspondence between physical quantities and electrical signals is determined by the characteristics of the converter, but few are linear, and most have non-linear correspondence.

When displaying or processing measured values, non-linear values are inconvenient to read or handle, so linearization is performed through correction calculations.

Even in process control devices that utilize intelligence devices such as microprocessors, electrical signals generated by process variable converters are linearized by correction calculations and used for displaying measured values and processing data.

In such process control devices, a plurality of input signals are processed by a common intelligence device and provided to respective displays, etc., so that if the intelligence device fails, the displayed measured values will be inaccurate.

It is desirable that the display of measured values be as accurate as possible so that at least manual backup control can be performed when the intelligence device fails.

In order to meet such demands, conventionally, an intelligence device performs a linearization calculation f for the input signal
←), divide this operation result by the input signal to obtain the modification coefficient α (-f (x)/x), and calculate this modification coefficient α
There is a system in which a multiplier is set to multiply the input signal X, and the multiplication result αx(-f□□□)) is displayed on a display.

In such a device, when the intelligence device fails, α immediately before the failure is retained in the multiplier, so
As long as there is no major change in the input signal X, a value close to the linearized value is displayed.

However, in such a device, the intelligence device must first obtain a linearization value through a correction calculation and then obtain a modification coefficient, which requires two calculations.

Therefore, when the number of input signals is large, the burden on the intelligence device increases, but the intelligence device also has to perform various other data processing, so
It is not desirable to spend too much effort on linearizing the input signal.

Furthermore, since the calculation is performed in two stages, calculation errors in each stage accumulate, and the accuracy of the corrected output becomes inferior.

An object of the present invention is to provide a measured value correction method that reduces the burden on the intelligence device and provides a highly accurate correction output when correcting input signals, including taking measures against failures of the intelligence device.

In this invention, the intelligence device normally performs a correction calculation f(x) such as nearization on the input signal X and outputs the result, and the backup device inputs the correction calculation value f←). A modification value ε of the input signal is determined based on the signal X, and when the intelligence device is abnormal, a value obtained by modifying the input signal It is.

The present invention will be explained below with reference to the drawings.

The figure is a conceptual configuration diagram of an embodiment of the present invention.

In the figure, 1 is an intelligence device, 2 is a back
It is an up device.

In the backup device 2, 22 is a modification value calculator;
23 is a switch, 24 is a modification value holder, 25 is an input signal modifier, and 26 is an output selector.

The input signal
6.

Input signal X and output signal f(x
) is also provided to a modifier value calculator 22, where a modifier value operation, such as subtraction f(x)-x, is performed.

The output signal of the modification value calculator 22 is given to the holder 24 through the switch 23.

The contents of the holder 24 are provided as a modification value ε to a modifier 25, and the input signal X is modified, for example, added.

The modification result is changed to f by switching the output selector 26)
It can be used as an output signal instead of .

The switch 23 and the output selector 26 are switched according to the operating state signal of the intelligence device, and are switched to the A side when the intelligence device 1 is normal and to the B side when it is abnormal.

Therefore, when the intelligence device 1 is normal,
The signal f(x) corrected by the calculation of the intelligence device 1 is output, and the output signal f (x)-x of the modification value calculator 22 is given to the holder 24.

The output signal of modifier 25 is ignored.

When the intelligence device 1 becomes abnormal, the switch 23
Then, the output selector 26 is switched to the B side.

Then, the intelligence device 1 is placed in the holder 24.
A modification value ε0 based on the last output signal during normal operation is held, and this modification value ε.

is added to the input signal X by the modifier 25 and output.

The output signal at this time is X+εo=x+f(xo)−x.

(1).

If there is no difference in the value of input signal X before and after switching, then X+εo = f(xo) (2)
The value after the correction calculation is output.

If the input signal X does not change significantly thereafter, a value that is not much different from the corrected calculated value is output.

In such a device, the correction output performed by the intelligence device 1 is always output, so the calculation in the intelligence device 1 does not have to be performed twice, and there is no decrease in accuracy due to two-step calculation. .

However, at the time of backup, there is a drop in accuracy, but this is not a problem since we do not originally expect much accuracy at the time of backup.

Note that the calculation of the modification value is not limited to subtraction, and may be performed by division as in the past.

In this case, a multiplier is naturally used as the modifier.

As described above, in the present invention, the intelligence device always performs correction calculation f(x) such as linearization on the input signal X and outputs the result, and the backup device performs correction calculation value A modification value ε of the input signal is calculated based on f(X) and the input signal X, and when an error occurs in the intelligence device, a value obtained by modifying the input signal Now output.

Therefore, according to the present invention, it is possible to obtain a measurement value correction method that reduces the burden on the intelligence device and provides a highly accurate correction output.

[Brief explanation of drawings]

The figure is a conceptual configuration diagram of an embodiment of the present invention. 1...Intelligence device, 2...
Backup device, 22...Modification value calculator,
23...Switch, 24...Modifier value holder, 25...Modifier, 26...Selector.

Claims (1)

[Scope of utility model registration request] A calculation means that performs a predetermined calculation f(x) on the input signal a selection means for selecting and outputting the input signal; means for calculating a modification value of the input signal based on the output signal and input signal of the calculation means; a modification value holding means; giving an output signal of the calculation means;
A measured value correction method comprising a switch means for cutting off the signal supply when the calculation means is abnormal, and a modification means for modifying the input signal X with a value ε held by the modification value holding means.