JPH04259815A - Measured value processing method - Google Patents

Abstract

PURPOSE:To eliminate the influence of disturbance by ranking the maximum width of predicted measurements and using the physical quantity corresponding to the maximum rank frequency of the rank frequencies corresponding to the measurements continuously obtained in a short time as the measurement of the short time. CONSTITUTION:The thickness (d) of a noodle band 1 can be calculated when pulsative rays of light are projected upon the noodle band 1 from optical sensors 4a and 4b provided on both sides of the band 1 at distances x1 and x2 and the distances x1 and x2 are found from the time lags between the emitting time of the rays of light and receiving time of the reflected rays of light from the band 1. However, because of disturbance, namely, the oscillation of the band 1 produced when the band 1 is cut into pieces of a fixed length by means of a cutter 5, the thickness (d) largely fluctuates at points of time t2, t2, and t3 at nearly regular time intervals. Therefore, the maximum width of predicted measurements are classified into an appropriate number of ranks and the physical quantity corresponding to the rank having the maximum frequency is used as the measurement of a short time by counting the frequency of the ranks corresponding to the measurements continuously obtained in the short time. Therefore, the influence of the disturbance to the measurements during the short time can be eliminated by finding the frequency distribution at every rank of a plural pieces of continuously obtained measurements.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for processing measured values of physical quantities obtained from a system subjected to disturbances that disturb its steady state.

0002

BACKGROUND OF THE INVENTION Several systems are known that measure physical quantities (temperature, speed, thickness, etc.) of a system during operation and process the measured values to control the operating state of the system.

[0003] Such systems often exhibit measured values that far exceed predicted measured values due to some external influence. In such a case, if a measured value obtained by averaging these protruding measured values is used as a measured value for determining the amount to be manipulated, the amount to be controlled may not be appropriate and accurate control may not be possible. Therefore, it is preferable to perform processing that is not affected by this protruding measurement value.

FIG. 1 shows an example of such a system, which is a noodle sheet thickness control device used for manufacturing noodle sheets such as udon and soba noodles.

In the figure, 1 is a noodle strip, 2 is a pair of rolling rollers for adjusting the thickness of the noodle strip 1, 3 is a guide roller that guides the noodle strip 1 whose thickness has been adjusted, and 4a and 4b are guide rollers. Optical sensors for thickness measurement arranged near each side of the noodle strip 1, 5
6 is a cutter for cutting the noodle strips 1 into predetermined lengths, and 6 is a cutting blade roll for making noodle strings 1' from the cut noodle strips.

7 is a controller that calculates the thickness value of the noodle strip 1 from the outputs of the optical sensors 4a and 4b and outputs the difference from the thickness setting value; 8 is a controller that responds to the thickness adjustment signal output from the controller 7; 9 is a screw rotated by the motor 8, and the thickness adjusting roller 2 approaches or separates by a distance corresponding to the amount of rotation of the screw 9 to adjust the thickness of the noodle strip. ing.

Next, the principle of measuring the thickness of a noodle strip will be explained with reference to FIG.

As shown in the figure, optical sensors 4a and 4b are arranged at distances x1 and x2 on both sides of the noodle strip 1 (thickness d), and the optical sensors 4a and 4b are directed toward the noodle strip 1. When the pulsed light is emitted and the reflected light is received, the distances x1 and x2 are
can be found. Therefore, both optical sensors 4a and 4b
When the distance between the noodle strips is y, the thickness d of the noodle strip can be calculated from the following formula.

[0009]d=y-(x1+x2)

0010

Problem to be Solved by the Invention In the thickness control device shown in FIG. However, since the noodle strip 1 is cut at a constant length by the cutter 5, the shaking of the noodle strip during cutting causes a disturbance, which is calculated from the outputs of the optical sensors 4a and 4b. The thickness d is influenced by disturbances at almost constant timing and varies greatly at time points t1, t2, and t3, as shown in FIG. For this reason, the controller 7 needs to perform signal processing so that the measured value of the thickness when subjected to the disturbance is not used for thickness control by the roller 2, but a processing circuit for this is required, and the If the timing of reception becomes irregular, signal processing becomes difficult.

The present invention has been made in view of the above-mentioned points, and it is an object of the present invention to eliminate the influence of disturbances from measured values obtained from a system in which disturbances that disturb the steady state are irregularly applied.

0012

[Means for Solving the Problems] In order to achieve the above object, in the present invention, when measuring a physical quantity that changes over time, the maximum width of expected measured values is ranked appropriately and shortened. The frequency of the rank corresponding to the measured values obtained continuously over time was counted, and the physical quantity corresponding to the rank having the maximum frequency was determined as the measured value for this time.

[0013]

[Operation] By determining the frequency distribution for each rank of a plurality of continuously obtained measured values, it is possible to eliminate the influence of short-term disturbances on the measured values.

[0014]

[Example] Below is a fourth explanation of the measured value processing according to the present invention.
This will be explained with reference to the figures.

As an example, processing of measured values obtained by measuring the thickness of noodle strips will be described, but it goes without saying that the present invention can also be applied to processing of measured values that include other disturbances of the same type.

The thickness data of the noodle strip is obtained by processing the outputs of the optical sensors 4a and 4b at regular intervals in the system shown in FIG.

For processing, first prepare an array D (N, M) with M length and N width (F-1), and set variables of the array to X, Y
(However, 0<X≦N, 0<Y≦M). Here, M is divided into an appropriate number considering the measured value and expected variations. This M is a number related to measurement accuracy. N is a number obtained by dividing the total number of measurements, and is determined by taking into account the period of data fluctuation. The ordinal number j(1
≦j≦N) is A=N・k+ from the total number of measurements A
j (where k is the number of repetition periods and is an integer including 0). Next, to initialize the thickness data, each M×N grid in the array is set to 0 (F-2).

Now, N pieces of thickness data obtained at regular intervals after the start of measurement are sequentially stored in a register in the controller 7 (F-3). Now, the j-th thickness data is expressed as Sk(j
), and among the variables X and Y for this Sk(j), the value of variable Y is determined as follows (F-4). If the maximum predicted thickness data due to disturbance is Sk(MAX), the Y value can be determined by the following formula. Note that INT[B] is an operation to obtain the integer part of the value B. The position on the array D(j, Yj) is determined from the Y value thus determined and the X representing the measurement order. For example, the value obtained from the above formula using the first thickness data Sk(1) is Y1
Then, the position (1, Y1) on the array is determined (F
-5). And put 1 in this position. This calculation is repeated until the Nth thickness data (F-6).

Now, if N=10 and M=8, then the array D(
10, 8) becomes a square as shown in FIG. 5, and "1" is filled in the square corresponding to the Y value obtained in steps (F-4) to (F-6).

Next, when the Yj values of N pieces of thickness data for j=1 to N are found and the squares in the D array table shown in FIG. 5 are filled, it is time to For the columns (k=1 to M), values for j=1 to N are totaled, and the obtained values are stored in a separately prepared D' array as shown in FIG. 5 (F-7). That is, if D'(k) is found for k=1 to 8, it will be arranged on the right side of the D array above. This D' array represents a frequency distribution.

From the D' array thus created, k that gives the maximum frequency is determined (F-9). In the above example, D′
The value of k that gives the maximum frequency of the array "4" is "4". Therefore, YMAX = 4 (F-9) and thickness data Sk
is obtained from the following formula (F-10).

Sk=(YMAX/M)×expected maximum value This thickness data Sk is plotted (F-11).

[0023] After that, increase j by 1 (F-12) j
Similarly, the D' array is found for the thickness data from =2 to 11, the value of k that gives the maximum frequency is found, and the thickness data Sk
Calculate and plot. In addition, the D' array includes the thickness data of j=2 to 11 mentioned above and the thickness data of a=3 to 12.
The thickness data up to j=1 to 10 are arranged in a D' arrangement. In this way, the thickness of the noodle strip can be obtained from a series of plots. The accuracy of the measured values can be increased not only by increasing the measurement interval but also by increasing the value of M that divides the variation width.

FIG. 6(a) shows the thickness of the noodle strip obtained by the treatment according to the present invention in comparison with that in the conventional case without treatment (b). It can be clearly seen that it is almost unaffected by disturbances, whereas it is affected by disturbances when no processing is performed.

[0025] The above embodiment relates to the thickness control of the noodle strip, but this is just an example, and the present invention can be applied to any system in which measured values are affected by irregular disturbances. Can be applied.

[0026]

As explained above, in the present invention, a physical quantity that changes over time is measured, the ratio of each measured value to the expected maximum value of change in the physical quantity is determined, and the ratio is determined to correspond to the expected maximum value. Among the predetermined number of ranks ranked in advance, the frequency of the ranks corresponding to the above ratio is increased by a unit amount for each measured value, and the frequency distribution for each rank is calculated for multiple consecutively obtained measured values, and the maximum Since the physical quantity corresponding to the rank having the frequency is determined as the measured value, the influence of irregular disturbances on the measured value can be easily eliminated only by signal processing.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the main parts of a noodle strip thickness control device as an example of a system to which a measured value processing method according to the present invention is applied.

FIG. 2 is a diagram illustrating the principle of measuring the thickness of a noodle strip.

FIG. 3 is a diagram showing the measured thickness of a noodle strip.

FIG. 4 is a flowchart of a measurement value processing method according to the invention.

FIG. 5 is an arrangement table showing the arrangement of data used in the present invention.

FIG. 6(a) shows the thickness of a noodle strip treated by the method of the present invention, and FIG. 6(b) shows the thickness of a conventional noodle strip.

[Explanation of symbols]

1 Noodle belt 2 Thickness adjustment roller 3 Guide roller 4a, 4b Optical sensor 5 Cutter 6 Cutting blade roller 7 Controller 8 Motor

Claims (1)

[Claims] Claim 1: When measuring a physical quantity that changes over time, the maximum range of expected measured values is ranked into an appropriate number, and the frequency of ranks corresponding to measured values obtained continuously in a short period of time. 1. A measured value processing method characterized by counting the number of times and determining the physical quantity corresponding to the rank having the maximum frequency as the measured value for this time.