curve 4--reading of speed for the new method, where the reference signals are
introduced for the intermediate moments in time (t1, t2 .
. . tn)
FIG. 4 is a view showing a device for the determination of an input signal
which changes in time in accordance with the present invention.
DESCRIPTION OF PREFERRED EMBODIMENTS
In accordance with the present invention, the determination of an input signal
which changes in time is performed with corresponding conversions for not only
a final point, but also for intermediate points. For this purpose, a device
includes a multi-position switch 1 through which an input signal x(t) and
reference signal Xl, X2, . . . are supplied to several elements connected in
series, in particular a convertor of analog into frequency 2, a key 3, a
counter-integrator 4. The device further has a register unit of reference
signals 5, a memory unit 6, and arithmetic-logical unit 7, a register of
expected integral value of the input signal 8 and a digital comparetor 9.
Using the example of calculating the desired speed of a rocket prior to turning
off the engine, the device operates in the following manner.
Before the beginning of measurements of an input signal, frequencies f1,
f2 . . . of the convertor 2 which correspond to the reference
signals X1, X2, . . . are determined. Then the
arithmetic-logical device determines corresponding reading of accelerations
where
is a frequency of pulses of the convertor at a zero input
signal;
is a coefficient of
conversion of scale factor of the convertor 2. The scale factor is determined
as
where D and D1 are determined as follows:
Also, coefficients of decomposition di of
a function X=P(Y) of reverse characteristic of the input/out of the convertor 2
are determined and introduced into the memory unit 6. The values are determined
by solving the following system of linear equations 
This system of equation is solved through the
determinators
where
At the beginning of the measurements, the
acceleration x(t) is supplied to the input of the convertor 2 and its pulse are
counted by the counter 4. Periodically, from cycle to cycle, the code Nk1
which is written in the pulse counter 4 where K is a number of cycle is written
into the memory unit 6. The arithmetic-logical unit determines the
difference Nk = Nk1
- N0, where N0 is a code code corresponding to the value
of the zero reference signal. By dividing Nk - Nk-1 by b1 the value DW - is determined which is an increment of
speed value. After this the arithmetic unit calculates the values DW2, DW3 and adds them
with the results of the previous calculations stored in the memory unit 6.
Therefore, the initial moments of indication Y of the input signal are determined:
Therefore, knowing the expected initial moments of
input signal M1[y], M2[y], . . . are known, and the
arithmetic logical unit calculates and rights in the memory unit 6 the expected
initial moments of the input signal
Since this and subsequent calculations require time,
it is not possible to obtain the measured values of speed exactly at the time
t=T. However, it is necessary for example in the case when upon reaching of the
given speed a command for turning off of the rocket engine is given. Therefore,
knowing the speed and its increment in "k" and some subsequent
cycles, expected values in a subsequent "k+1" cycle are determined by
extrapolation.
Knowing the initial moments M.sub.i [x], from the linear system of equation
where m=0,1,2 . . . n-1 and T is an expected time of
measurements, Ti is determined by determinators
By calculating the value of Ti, it is
possible to find the value of speed expected in the time T
and corresponding to the value W(T) code
Which is written in the register 8 and compared by the digital comparitor 9
with the content of the counter 4. Coincidence of these two codes which is
confirmed by triggering of the digital comparitor 9 indicates that the speed
reached the value V(T).
Now the above described operations can be repeated for the subsequent cycle.
The duration of the cycle Dtk = tk
- tk-1 is limited from below only by a needed time of calculations.
Knowing the value of speed Nk+1 N Vk with a
high accuracy in neighboring cycles, it is possible to find a cycle average
value of acceleration or input signal.
Thus, with the use of the proposed method, it is possible to use as input not
only acceleration but also the input signals of other physical nature
(mechanical, radiant, magnetic, electrical, thermal, chemical, vibrational,
etc).
It should be emphasized that in the algorithm
the values of reference signals can be arbitrary. Therefore it is possible to
select those values of the reference signals which are available. For example, for
measurement of
