SU646351A1 - Correlation function determining device - Google Patents

Description

one

The invention relates to the field of automatic control and can be used in the study and design of pulse systems.

Methods and devices are known for determining the output correlation of the output signal of a dione and an electrically controlled and impulsive automatic control system using shaping filters and analog models that allow determining the correlation function in the established mode or, in the unsetting mode I. .

The task of determining a more complete statistical characteristic — the correction function of the output signal of the subsequent system, which characterizes both the established and unsteady modes of its operation and is called the full correlation function, has now been solved only for continuous systems.

The closest technical solution to this invention is

A device for determining the correlation function, containing two chains, each of. which are connected from one of the following connected clamps, delay block and multiplier block, the first chain lock input is connected to the filter input and the first input of the device, the second lock clip input is connected to the filter output, and 1 multiplier blocks of both chains are connected respectively to the first and second block inputs addition, the other inputs of the blocks of the multiplication of the first and second chains are connected by the L outputs of the latches of the second and first chains 2l, respectively.

However, the use of this device to obtain the full brenadionic function of a pulsed system is impossible, since pulsed systems have

essential features in comparison with continuous ones. This applies not only to the distinction of elements characteristic only of impulse systems (keys.

restoring elements) but also, mainly, to the replacement of operations and HTIning with its discrete analogue by summation. .

, , h,,

The purpose of the present invention is to expand the functionality due to the complete function of the output signal of the pulse system, which is obtained fully to the transmitter, which makes it possible to analyze its behavior in the transition K) M and steady state.

The supplied ufejtb is achieved by introducing an additional multiplication unit and an amplifier whose output is connected to the third output of the adder, whose output is the output of the device, the amplifier input is connected to the output of the additional multiplication unit, the first and second inputs of which are connected respectively to BXOjPiOM and the output of the delay block of the first chain, the third in the Additional input of the multiplier block, is the second input of the device ,.

In order to obtain a voltage proportional to the total correlation functions of the system, both from the delay time and from the current time (from the moment of switching on and the system), the latter is formed by summing the sum of the three components.

The first component is proportional to the B – J product of the output voltage of the system under study when a very short pulse is applied to its input and a signal is applied backwards after a linear transformation with a discrete (| yurii filter in accordance with the type of correlation functions of the input action. The second component is receive as a product of the delayed signal from the output of the system and the non-delayed output signal of the discrete shaping filter. The third component is proportional to the output of the dispersion of the input si Nala, who was detained and undelayed

signals from the system output.

The principle of determining the full correlation function of Cc 2 T; tjl T; With T iy the best-selling signal of the pulse system is based on the calculation of a one-time sum, which can be reduced to the general expression KGPT, friT, t}.

For the correlation function of the output signal:

gN + e G

Ky gT; NT, T., (N.

-8lT, T wt {n-EiT,

T w3tnT., ETJw (n-th) T.

, 6T2 is the weighting function of the system under study;

eB et - | wPi-k) t,

Where Wg is the weighting function of the equivalent Cysteragl formed by the system under study and a discrete shaping filter; .. ;; f -. ; , ..

WojjiT is the weighting function of the Discrete shaping filter;

To the initial value of the input signal correlation function (dysperse);

f - key closure period of an unsecured system}

P, W 0,1,2, ...;

Z; N hrs (n, m);

, TT - discrete points in time; . ;; :

8 t - delay time; V T is the time elapsed from the monaHTa of the system;

 - offset parameter

The drawing shows the block diagram of the proposed device; TWA.

The device contains the system under investigation (or its model) 1, filter 2., clamps 3 and 4, blocks 5 and in delays, first and second blocks of multiplication 7.8 and additional block of multiplication 9, amplifier 10, block of addition 11, adder 12 .

The discrete formative filter2 and the system 1 under study form an equivalent system with a transfer function, t. The inclusion of a discrete shaping filter 2 after the system under study 1 is possible, since their parameter is constant. Catchers 3 and 4 n; LEADERS are rebuilding elements.

At the time T 0, the input of the system under study is given a signal in the form of the following 6-function

, 0 Jl at-pT 0;

LP at n m J o,

representing a pulse of single amplitude and infinitely small duration. In practice, such a signal is replaced by a pulse of a single amplitude1 and of a very short duration as compared with the period T of the key closure. The expression for the 6-function describes the correlation function of the discrete white 1.um. At the output of system 1 and filter 2 receive, respectively, the weight functions, 6 Tj and

NVaEn T. e.

After passing the signals, cooiv

Corresponding to these functions, through blocks 5 and 6 of delay, having the same delay time equal to BT, at the outputs of blocks 7 and 8 receive the first two components, which are equal to the total correlation function.

To obtain the third component, the two inputs of block 9 are connected to the outputs of latch 4 and block 5 of the delay, and the third is given a signal equal to the dispersion. Since one third of the component comes under the sums with a negative sign, the signal from the output of the Q block is fed to the amplifier 10 with the gain K 1.

All three components are fed to the addition unit 11 and then to the adder 12, and during the course of the adder 12, in the form of a step function, a cross section of the full correlation function is obtained at a fixed value of the delay V.T.

To obtain the output of the voltage adder 12. System 1, proportional to the total correlation at the moment of key closure (not shown), system 1 keys and latches 3 and 4 must operate synchronously and in phase. If it is necessary to obtain the values of the full correction function between the moments of the key closure of the system 1, Cho creates a fixed 3Ha4smife offset parameter (1,2,3, ...), with i 1, Keys of the latches 3 and 4 at the same time the closure; simhrotyu with a system key 1, but, not in phase, but with an -J T

In the absence of a suitable adder, the corresponding results at the output of block 11 are measured (fixed) in each clock cycle and then summed up 5 manually or with the help of the C.B.M.

By varying the delay time P T from zero to infinity and the symmetry of the complete correlation function relative to the current time axis M T O, the entire envelope surface is obtained by the full correlation of the function. Obviously, for a given, it is necessary that AND + 1 include models.

When E - O get a dispersion in

5 unsteady mode.

For sufficiently large N values, an established correlation function is obtained.

The analysis of the behavior of the correlation function in both the steady state and the unsteady mode makes it possible to better solve the problems of synthesis. . (choice of command and control), which

It is especially important for proktirovatshi systems, 5 Invariant to external factors, and, in general, systems that have high requirements for accuracy of regulation.

thirty

Form u l a It 3 about b re shadow

A device for determining a correlation function comprising two chains, each of which consists of a successive latch clamp, a delay block, and a multiplication block, the latch input of the first string is connected to the filter input and is the first

the input of the device, the input of the latch of the second chain is connected to the output of the filter, the outputs of the blocks of the utilozhenn both chains are connected respectively to the first and second inputs of the block

addition, other inputs of multipliers of the first and second chains are connected to the outputs of the latches of the second and first chains, respectively, characterized in that, in order to expand the functionality of the device by obtaining a complete correlating function of the output signal of the pulse system, an additional block of multiplications is introduced into it; and the amplifier whose Bbixoa is connected to the third input of the block, the output koTbpbi o is the output of the device, the input of the amplifier is connected. . ;:.;: ... to the output of the downstream multiplication unit, the first n inputs of which are connected to the input to the output of the delay unit of the first hop, the third input of the terminal block is the second input of ycTpoftcTBa. 5i Sources of information taken into account during the examination 1. P. Krutko. Statistical dynamics of impulse systems. M., Soviefeoe Radio, 1963, p. 271-279. 2. Authors certificate of the USSR M 19954S, cl. q 0.6 G, 7/52, 1966.

KfCflT mTj 6inT.fa