SU1325421A1 - Continuous weighing system - Google Patents

Abstract

The invention relates to a weight-dosing technique and can be used in the production of mixtures of a given composition in the metallurgical, chemical, feed mills and in other branches of the national economy. The aim of the invention is to increase the accuracy of maintaining a predetermined ratio of the performance of weighing weighers of continuous action. The goal is achieved by the fact that the system of non-linear weight dosing, which contains the aggregator of total capacity and n control channels, each of which contains a dosing capacity setting unit and a continuous weighing dosing unit, introduces a dosing unit settings correction unit, an adder and a unit for determining the statistical characteristics of the actual system performance. , and a block for determining the statistical characteristics of the actual production is entered into each control loop. nosti dispenser block and dividing. 2 Il. with b sl te 1CHE ate

Description

13

The invention relates to weighing technology and can be used in the production of mixtures of a given composition in metallurgists, chemical, feed milling industry and in other sectors of the national economy.

The purpose of the invention is to increase the accuracy of maintaining the predetermined ratio of the NIN capacities of continuous weighing weighers.

The essence of the invention is that a given amount of performance. q (t) to each (i-th) metering unit, depending on the determined proportions of the components N and on the total performance of the dosing system q (t), taking into account the statistical characteristics of the deviation (t) and Aq (t) of the actual capacities of the dispensers q , (t) and dosing systems (y (t) from given values of q (t) and q (t).

Deviations of actual performance are due to many factors (inconsistency of moisture of the material being metered, its density, characteristics of the outflow of material from the supply containers, etc.). Therefore, q. (T) and q (t) can be represented as stationary random processes:

q. (t) qj (t) + Aq. (t), q (t)

 q (t) + Aq (t),

the cross section of which has a normal distribution law with mathematical expectation m and standard deviation 6. Then the actual ratios of the dispenser capacities can be described by a system of n two-dimensional probability densities of random variables (t) and q (t).

We have

f Jq, (t). q (t) - .. X

Г 1 (t) - m, -

() L v

.) - four

-BTB, -Y--

1 1, n (1 where -, m. - mean square

deviation and expectation of i-ro dispenser performance,

nd-- (- average quadratic r.,. I

deviation of the total performance of the dosing system;

Pi T / V is the correlation coefficient of the i-ro metering capacity and the total system capacity and metering, m ZH m. -mathematical expectation of the total performance of the dosing system.

The mathematical expectation of the relation r. Q (t) / q (t) for small G J can be written as

five

about 5

0

five

H 1 i

(2)

where the distribution function of the performance ratio is

with

f, (r.) 5 q (t) f Jr, qg (t), q (t) x

0

о о X clq (t) - (t) .qj. (t), q (t) x

X). (3)

In this case, ra, - r. and the displacement ((m, - r °) increases with an increase in the standard quadratic deviations. Moreover, m,. r ° at and t,. t ° at g. 0.5.

An increase in the accuracy of maintaining a given ratio of metering capacities is achieved due to the fact that, in addition to the actual capacities, their statistical characteristics (mathematical expectations and standard deviations) are also measured, taking into account, using formulas (2) and (3) mathematical expectations of the actual ratios of the performance of the dispensers and their deviations from the specified values, and then form the dispensers of the task of performance, offset by the magnitude of these deviations eny. For this purpose, an adder and a unit for determining the statistical characteristics of the actual performance of the system are entered into the dosing system, and a unit for determining the statistical characteristics of the actual capacity of the dispenser and a division unit are entered into each control loop.

Fig. 1 shows a functional diagram of a continuous weight dosing system; in Fig. 2, functional

The digital version of the system is based on a programmable universal electronics controller K1-20, designed to control five dispensers.

The system consists of a setpoint adjuster 1 of total performance, a unit 2 for adjusting setpoints of dispenser capacity, an adder 3 and a unit D for determining the statistical characteristics of the actual system performance, and also n control channels, each of which contains a unit 5 for the performance of a dispenser, a weight dispenser 6 continuous operation, block 7 division and block 8 of determining the statistical characteristics of the actual performance of the doser.

The continuous dosing system operates as follows.

The signals of the actual capacities of the metering units 6 are fed to the adder 3, where the total productivity of the metering system is calculated. In blocks 4 for determining statistical characteristics and 8, the mathematical expectations (first output) and the standard deviation (second output) of the actual performance of the metering system 6 are calculated on a sliding time interval. In dividing units 7, the correlation coefficient of the metering capacities 6 and the total performance are calculated systems as the ratio of their standard deviations. The correction unit 2 receives the values of the actual performance from the metering units 6, the specified total system performance from setpoint 1, the specified capacity of the dosers from the task sets 5 and the statistical characteristics of the deviations of the actual values from the values specified from blocks 4 and 8. In the correction unit, the deviation of the mathematical performance ratio

gQ continuous dispenser

from the value set by the blocks 5, and the task is generated to the dispensers 6 shifted by the magnitude of this deviation.

Blocks 4 and 8 of statistical characteristics determination are identical and measure the average value and standard deviation of the input signal at the time slider; variable interval t,. The dosing system is not critical to the duration of this interval, but its value should

to be such as to smooth random fluctuated dispensers performance 6. For real dispensers

„-. . ,,. In this block 4

Blocks 2, 3, 4, and 8 represent its

  lb and 8 are implemented dependencies

conventional computing (.ari (| methical) devices) that can be implemented, in particular, on the basis of single-board microcomputers (Electronics

g

 | - 1 q (t) cit; 6 - f S q (t) -shZa -

m

0

15 0

25 45

HU-80, K1-20, K1-30, etc.). The operation of these blocks can be most fully described by the functioning algorithms, where q is the performance,

t is the ratio of dispenser performance to system performance

Q - the greatest limit of productivity;

n - the number of dispensers in the system;

m is the mathematical expectation; 6 is the standard deviation; 2 - value characterizing

total system performance; i is a value characterizing

i-ro dispenser performance

Ф - fictitious (intermediate, calculated) value; t is the current time.

The adder 3 implements the dependence -q, (t).

q. (t s

14

The algorithm of its work, presented in a formalized language, has the following form:

 0;

 1, q. 0;

one :

q: yaf + chh

1 g p

that

(q.

 q.

transition

p if

on 2), .

i: i + 1;

go to 3.

Thus, at the output of the adder 3, in the established mode, there is a value of the total performance of the continuous weighing system.

Blocks 4 and 8 of statistical characteristics determination are identical and measure the average value and standard deviation of the input signal at the time slider; variable interval t,. The dosing system is not critical to the duration of this interval, but its value should

8 implemented dependencies

g

 | - 1 q (t) cit; 6 - f S q (t)

Approximately with sufficient accuracy, m and 6 can be determined by the following algorithm.

Transition to 2, where K is the coefficient imitating integration over a skew time interval. The value of K is determined once, depending on the actual duration of one cycle tj:.

K-, while K. one . Perhaps a more precise definition of n and 6. At the same time, n is remembered (where n is the last samples of q and the expressions are calculated

h n 1 "

m - tstl (s - (q - m) 2.

n i «4

However, this method requires the storage of an array of considerable dimensionality, which entails an unreasonable increase in the memory capacity of blocks 4 and 8, and does not allow the use of standard single-board microcomputers.

Coordination unit 2 determines the shifting mathematical expectation of the actual ratios of dispenser performance and calculates the values of the coordinating action, which ensures the increased accuracy of maintaining the actual ratios.

The value of the mathematical expectation of the ratio q,. (T) / qg (t) at m

lykh

defined as

m

  . where fp () is the probability density. And

00

f, (r ,.) «| q (t) fjr, q (t), q (t) di-00

- I) 4hq (t. Qk (t) d.

where fo is the two-dimensional probability density of the processes q. (t) and

q5: (t).

For normal random processes which are q, - (t) and qj. (T), we have

  , -) f 1 +. (r, -) X X) (t)

g, ovn 2m j-iBz & + jj qi () 4, 1

where fe (r, -)

isr-

 Shchg1 26 | r. 46.z)

Cauchy law (density distribution ratio of cent

five

0

five

0

five

0

five

0

five

the constituted components of the processes (q, - (t) and q (t);

h (g) Si i-gs l - L ± -igi i- ° il l

(6.2) (6 r - 24 g, - + if) (Г {) is the Laplace function of the argument h (r,).

Thus, t is a function of four parameters: t, -; t „; 6; and 6.

For real dosing systems, mvetO I Qf; m, 1 q ,,; QJ,;

, 1mJ; 6j.4o 0.1 GOE the range of variation of parameters; ; m .; dy; 6 is already known at the design stage of the system and the dependence t. (T ,; 6; bd) can be calculated in advance on a high-power computer (a program is developed for an EC computer) and tabulated in the memory of the coordination unit in the form of a four-input table. Having obtained the values for i l7n from the table, the deviations ui sh are calculated. - r, and then calculate the coordinating effect of r. Taking into account

uk

that, g, - 0, g- can be

 to 1 "In. obtained as r-AI - - vii.

 n th (

Then the performance tasks for dispensers should be formed as follows:

about . / o Kx q. (r, - - i:.) q.

The operation algorithm of the correction unit 2 is as follows: i: 1;

q- -r-;

if ifn, To (, go to 2);

checking the presence of signals t, -; f-; 6j. at the output of blocks 4 and 8, the determination of statistical characteristics 4 and 8,

i: 1; 0;

search table t. (t;

, -; 6s);

iF: аЧз + m (.. - Г {;

if, To (+ 1, transition to 6),

& V: p / p;

i: 1;

q (r DF-) q ° ;.

if i F then (i: i + l), the transition / transition to 11).

Claims (1)

Invention Formula The system of continuous weighing dosing, containing a master of aggregate capacity and n control channels, each of which includes a metering unit for specifying the metering capacity and a weigher metering unit, characterized in that, in order to increase the accuracy of maintaining the specified ratio of metering capacities of non-selective actions , the system contains a series-connected adder, a unit for determining the statistical characteristics of the actual performance of the system, and a unit for adjusting the setpoints metering units, and in each control channel, the serially connected unit for determining the statistical characteristics of the actual metering capacity and the dividing unit are additionally inserted, the output of the aggregate capacity setting unit is connected to the second input of the metering performance correction unit, the third input of which is connected to the second unit output determining the statistical characteristics of the actual performance of the system and with the second input of the dividing unit of each channel The input of the unit for determining the statistical characteristics of the actual dispenser capacity in each control channel is connected to the output of the weighing dispenser of its own channel, with the corresponding input of the adder and with the corresponding fourth input of the correction unit group of the dispenser capacity settings, the output of the dispenser setting task in each control channel to the corresponding 5th input of the group of the correction unit for the settings of the metering performance, the first output of the unit for determining the st The statistical characteristics of the actual capacity of the metering device of each control channel are associated with the corresponding sixth input of the metering unit’s correction unit group, the second output is connected with the corresponding seventh unit of the metering unit’s performance correction unit, the output of the dividing unit of each control channel is connected to the corresponding eighth setpoint correction unit metering performance of the outputs of which are connected to the inputs of the respective weighers all control channels. From 5l. 10mdflJOm 6l5 From blb From Il. 7 from obd 8 Editor N.Egorova Compiled by L.Tsallagova Tehred L.Oleinik Order 3107 / 42FAF 863Subscription VNSHSHI USSR State Committee for inventions and discoveries 113035, Moscow, Zh-35, Raushsk nab., 4/5 Production and printing company, Uzhgorod, st. Project, 4 Proofreader M. Pojo