JP2610282B2 - Waveform data storage - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Purpose of the Invention] (Industrial application field) The present invention relates to a waveform data storage device.

(Prior Art) A waveform data storage device that stores a temporal change of data such as a temperature, a flow rate, a voltage, and a current in a storage medium such as a semiconductor memory, a magnetic tape, a magnetic disk, and an optical storage medium is generally a first type.
The configuration as shown in the figure is adopted.

The analog signals from the measuring instrument channels Ch1, Ch2, Ch3, Ch4 are taken into the multiplexer 1, each analog amount is amplified by the amplifier 2, and these analog signals are converted into digital signals by the A / D converter 3, and the input port 4 is input. Then, the digital signal given to the input port 4 is given to the CPU 6 via the bus line 5. Further, a CPU 6, a ROM 7, a RAM 8 are connected to the bus line 5, and a recorder interface 9 and a keyboard / display interface 10 are connected to the bus interface 5. A data recorder 11 and a keyboard / display interface 10 are connected to the recorder interface 9. On the other hand, a display 12 such as a CRT and a keyboard 13 are connected respectively.

In such a general waveform data storage device, the operation of the conventional CPU 6 is as follows.

That is, as shown in the flowchart of FIG. 5, defines a constant sampling period △ t, starting from time t ₀ (step S101), it reads the data x _n at every sampling period △ t (step S102 to S106), RAM 8 (Step S107).

(Problems to be Solved by the Invention) However, in such a conventional waveform data storage device, since data is stored at a predetermined period Δt, the consumption of the storage medium is large, and the efficiency of the storage device is high. There was a problem that it could not be used in general.

Further, a storage device may be considered in which the sampling period is lengthened when the temporal variation of the data is small, and the sampling period is shortened when the temporal variation is equal to or more than a certain value. There is a problem that there is no effect if the value fluctuates greatly.

The present invention has been made in view of such a conventional problem, and has as its object to provide a waveform data storage device that can reduce the consumption of a storage medium.

[Structure of the Invention] (Means for Solving the Problems) A waveform data storage device according to the present invention includes a data input reading unit, a data processing unit, and a data storage unit, and the data processing unit reads the data input. Data that is read periodically from the unit, and the read data is regarded as a time-varying waveform, the waveform is approximated by a broken line, and the coordinates of the obtained broken line are stored in the data storage unit as data. It is.

(Operation) In the waveform data storage device of the present invention, the data input at a fixed sampling period is approximated by a broken line, and one data and several subsequent data are approximated by a single broken line. When data exceeding the permissible error range is input, the data is newly set as a break point of the polygonal line, and another data which is within the permissible error range from the new break point to some data following the broken line. By calculating and approximating the bent line of the inclination and repeating this operation, the waveform data is accurately stored with a small consumption of the storage medium.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram of a general waveform data storage device as described in detail in the description of the conventional example, and the feature of the embodiment of the present invention lies in the program operation inside the CPU 6.

First, the operation of the hardware will be described with reference to the block diagram of FIG. 1. The analog signal input from each of the channels Ch1, Ch2, Ch3, Ch4 to the multiplexer 1 is amplified by the amplifier 2, and the analog signal is digitally converted by the A / D converter 3. It is converted to a signal and input to the input port.

CPU6, based on the program stored in ROM7,
The digital data signal supplied to the input port 4 is fetched every fixed sampling period, and is temporarily stored in the RAM 8.

Then, the CPU 6 calculates data of a series of break points obtained by a polygonal line approximation process described in detail later with respect to the data for each fixed sampling period stored in the RAM 8, and outputs the data to the external data recorder 11 through the recorder interface 9.
To remember.

Various parameters necessary for calculation such as a storage start time, an input reading cycle (sampling cycle), and an allowable error range are given to the CPU 6 through the keyboard / display interface 10 when the operator inputs the information from the display 12 and the keyboard 13, Stored in RAM8.

Next, the data processing operation of the CPU 6 in the waveform data storage device having the above configuration will be described with reference to FIG.

The data read value for each sampling period Δt is given by t = t ₀
X ₁ when x _0, t = t ₁ when, as x _n when ...... t = t _n, the coordinate points in these on the graph _{(t 0, x 0),} (t 1,
x ₁ ), (t ₂ , x ₂ ),... (t _n , x _n ).

Assuming that there is a break point P ₀ (t ₀ , x _p ) at t = t ₀ , the next break point is obtained as follows.

Since the read value at t = t ₁ is x ₁ , the value of the approximate broken line at t = t ₁ is (x ₁ − △ e) as the allowable error Δe.
A range from to (x ₁ + △ e) is allowed. Therefore, the fold line extending to the right from the fold point P ₀ passes through P ₀ and has a slope of (x ₁ + △
e−x _p ) / △ t to (x ₁ − △ e−x _p ) / △ t. That is, in FIG. 2, the inclinations l ₁₁ to l ₁₂ to
What is allowed is up to.

Next, the read value at t = t ₂ is x ₂ (t ₂ , x ₂ )
Similarly, when a permissible broken line is obtained, the inclination straight line l ₂₁
Being within the scope of to l ₂₂ is allowed. Therefore, the range of the broken line in which the error is less than or equal to Δe with respect to both the coordinate points (t ₁ , x ₁ ) and (t ₂ , x ₂ ) is represented by the slope straight lines l _{11 to} l ₁₂ and the slope straight line l
Are those included in both of the range of ₂₁ to l _22, in the second diagram in which the range of l ₂₁ to l _22.

Next, when t = t ₃ is similarly considered, the slope is l
₃₁ becomes the range of to l _32. Therefore, these (t ₁ ,
x _1), polygonal line is in the range of l ₂₁ to l ₃₂ shown by hatching in FIG. 2, which allowed for three points _{(t 2, x 2),} (t 3, x 3) Things.

Looking at the next point, (t ₄ , x ₄ ) at t = t ₄ , the permissible ranges l _{41 to} l ₄₂ of the polygonal line at this point are as follows:
No common area is found in the allowable ranges l _{21 to} l ₃₁ for the three points up to that point, and an approximate broken line including the _fourth point (t ₄ , x ₄ ) cannot be drawn.

Therefore, the three points (t ₁ , x ₁ ), (t ₂ , x ₂ ),
(T _3, x ₃₎ common to the polygonal line L ₁ for, from break points P ₀ of the first, third point (t _3, x ₃₎ of inclination in the range l ₂₁ to l
Determined by connecting the intermediate point P ₁ of _32, a point P ₁ in the third point (t _3, x ₃₎ as a new breakpoint (t _3, x _q), the fold line approximation processing The fourth and subsequent points (t ₄ ,
x ₄ ), (t ₅ , x ₅ ),...

The resulting folding points P ₀ (t ₀ , x _p ), P ₁ (t ₃ ,
x _q), P _2, for ... is the slide into stores the coordinate data to the data recorder 11 through the recorder interface Sue over scan 9.

In this way, when the folding points obtained by the folding line approximation are stored as data, the consumption of the storage medium is reduced in each stage as compared with the case where the data is stored at each sampling cycle as shown in FIG. I understand. That is, only six points P ₀ , P ₁ , P ₂ ,... Need to be stored as break points. In the conventional case, however, sampling points S ₀ , S ₁ , S ₂ ,.
You have to remember as many as fifteen.

Figure 4 is shows the CPU6 program flow for executing the above-described embodiment, and waits until reading start time t ₀ (step S1), and when there is a read command, reads the first input x ₀ Then, the coordinate point P ₀ (t ₀ , x ₀ ) is stored (steps S2, S3).

Thereafter, initialization is performed (step S4), and the input value _xn is read at every constant sampling period Δt (steps S5 and S5).
6).

Then a m = 1, if it is necessary to find a new polygonal line for break point x _p, the minimum value of the polygonal line slope for reading values x _n at t _n MIN, the maximum value MAX determined (step
S8), 1 is added to each of m and n, and the process returns to step 5 (steps S9 and S10).

In the step S7, the break points x if already one or more fold lines is the _p has been found a m ≠ 1, the minimum value MIN and maximum value MAX of the polygonal line slope which Dasa last Heading Is calculated (step S11), and the minimum value (x _n − △ e−x _p ) of the inclination of the broken line with respect to the current read value x _n is calculated.
Δt and the maximum value ( _xn + Δe− _xp ) / Δt are the previous minimum value
MIN and the maximum value MAX are compared with each other to obtain new inclination ranges MIN and MAX of the broken line (step S12).

If there is a range in the inclination of the new broken line thus obtained (step S13), the process returns to step S5 to further approximate the broken line for the next data (step S13).
9, S10).

However, the new minimum value M obtained in step S12
If the magnitude is reversed between IN and the maximum value MAX and the next data read value does not enter between them, it is necessary to set a new break point, t _n-1
The midpoint MID ranging fold line slope and the break point P _l in, the new break points P _l of coordinates _{(t n-1, x p} = MID) based on, then come been read data (Steps S14, S15, S10).

In this way, the CPU 6 approximates several pieces of read data by one broken line by the broken line approximation, and when data that cannot be approximated within the allowable error range appears, sets a break point to approximate by a new broken line. determined, performs broken line approximation similarly for subsequent several data, thus it went successively sought corner, the coordinates _{P l (t n, x p} ) of the obtained corner sequentially stored It is.

In the above embodiment, the coordinates (t _n , x _p ) of the break point are
Midpoint between the minimum value MIN and maximum value MAX of the tolerance of the gradient of the approximate polygonal line at the time t _n which has been found in Its corner (MAX + M
Although IN) / 2 is represented by _xp , this is not particularly limited, and any value may be used as long as it is not less than MIN and not more than MAX.

[Effects of the Invention] As described above, according to the present invention, a series of waveform data is approximated by a broken line to find a break point and store the break point. This means that it is sufficient to store a smaller number of data than in the case where the data to be stored is stored one by one.In addition, even if the number of stored data is small, the approximation is within an allowable error range, and the data accuracy is reduced. Therefore, the load on the storage medium can be reduced.

[Brief description of the drawings]

FIG. 1 is a circuit block diagram of one embodiment of the present invention, FIG. 2 is an explanatory diagram for explaining the operation of the above embodiment, FIG. 3 is a graph showing a broken line approximation curve obtained in the above embodiment, FIG. Is a flowchart for explaining the operation of the CPU in the above embodiment.
FIG. 2 is a flowchart for explaining the operation of the conventional example. 1… Multiplexer 2… Amplifier 3… A / D converter 4… Input port 5… Bus line 6… CPU 7… ROM 8… RAM 9… Recorder interface 10… Keyboard and display interface 11 Data recorder 12 Display unit 13 Keyboard

Claims (1)

(57) [Claims] A data input reading unit, a data processing unit,
A data storage unit, wherein the data processing unit periodically reads data from the data input reading unit, captures the read data as a time-varying waveform, performs a linear approximation on this waveform, A data storage unit for storing the coordinates of the turning point as data in a data storage unit.