RU2427774C1 - Method to dry paste-like materials in roll-and-tape dryer with variation of plate conveyor movement speed - Google Patents

Abstract

FIELD: construction.

SUBSTANCE: method to dry paste-like materials in a roll-and-tape dryer includes formation of dried material, its serial movement through drying chambers in process of counterflow-cross-like movement of a drying agent, removal of which is carried out in cyclic mode, which includes saturation of the drying agent and removal of the drying agent, besides, duration of removal and stop is identified depending on the specified moisture content of the drying agent. At the same time, depending on speed of plate conveyor movement, time of dried material stay is identified in n and n+1 chambers of the drying plant, in which the main or surface moisture is removed, and material moisture is assessed online in n and n+1 chambers with multiple sensors along neuron networks, compliance of calculated material moisture values with the rated ranges is verified, depending on which numerical estimates are defined for measures of trust to achieve the required material moisture at the drying plant output, and on the basis of produced values of material moisture in the specified chambers and numerical estimates of trust measures a fuzzy inference is produced, which defines the value to change the plate conveyor movement speed to achieve the required quality of material at the output of the drying plant.

EFFECT: achievement of the required material quality at the drying plant output with optimal efficiency of drying process.

5 dwg, 7 tbl

Description

The invention relates to drying equipment, and in particular to a method for drying pasty materials, and can be used in chemical and related industries.

Known methods of drying pasty materials in drying plants [1] containing a roller system, plate and scraper conveyors, steam heaters, agitator and recirculation fans, where the briquettes of the dried material are dried on a plate conveyor with continuous removal of the spent drying agent.

The disadvantages of the methods are significant energy consumption for the drying process of 37-45 GJ / t of dry product, low productivity of 30-50 kg of dry product / h, removal of the drying agent with low humidity 0,0175-0,023 kg / kg at a high temperature of 60-75 ° C .

There are also known methods of drying on tape dryers [2], in which the drying is carried out by air or its mixture with flue gases at a temperature of the mixture of 70-170 ° C. Air circulation is carried out by TsAGI axial fans. Air passes through the material from top to bottom and presses the material to be dried to the plate conveyor.

Their disadvantage is the unevenness of drying along the thickness of the layer. The best results for uniformity of drying are provided by methods where multi-tier belt dryers are used, in which the material is mixed when transferring from one belt to another. In these methods, the heat consumption is from 5000 to 7500 kJ / kg of moisture, and electricity is up to 10 kWh per 1 ton of evaporated moisture.

Closest to the invention relates to a method for drying pasty materials in a rolling mill dryer with a cyclic mode of removal of the drying agent [RF patent No. 2338974, F26B 17/28, application 2006137249/06, 10.20.2006, published November 20, 2008], including molding the material to be dried, sequential movement through the drying chambers during countercurrent cross-movement of the drying agent, removal of the drying agent and its partial reuse, removal of the drying agent is carried out in a cyclic mode, including saturation of the drying agent Gent and the removal of the drying agent, and the duration of the removal and stopping is determined depending on the specified moisture content of the drying agent, and the moisture content of the drying agent is controlled by its temperature, while the moisture content of the drying agent varies within 0.033-0.051 kg / kg and the drying process goes into self-mode when the system itself sets the consumption of the drying agent depending on the moisture content of the drying agent and the moisture content of the dried material.

The advantage of this method is to reduce energy consumption and the intensification of the drying process.

The disadvantage of this method is that it does not take into account the residence time of the material in the chambers of the drying installation, depending on the speed of movement of the plate conveyor and affecting the quality of the material at the outlet of the drying installation.

An object of the invention is to achieve the required quality of the material at the outlet of the drying installation with optimal performance of the drying process.

The stated technical problem is achieved by the fact that in the method of drying pasty materials in a rolling mill dryer, including forming the material to be dried, it is sequentially moved through the drying chambers during countercurrent-cross movement of the drying agent, the discharge of which is carried out in a cyclic mode, including the saturation of the drying agent and the removal of the drying agent moreover, the duration of the drain and stop is determined depending on the specified moisture content of the drying agent, in contrast to otipa, depending on the speed of movement of the plate conveyor, determine the residence time of the dried material in n and n + 1 chambers of the drying unit, in which the main or surface moisture is removed, in real time the humidity of the material in n and n + 1 chambers is estimated by a variety of sensors on neural networks check that the calculated values of the moisture content of the material fall within the normalized ranges, depending on which numerical estimates of confidence measures to achieve the desired moisture content of the material at the outlet of the drying plant are determined and and on the basis of the obtained values of the moisture content of the material in these chambers and numerical estimates of confidence measures, a fuzzy conclusion is produced that determines the value by which the movement speed of the plate conveyor is changed to achieve the required quality of the material at the outlet of the drying unit.

The essence of the proposed method is illustrated in figure 1 ÷ 5.

) большое влияние оказывает процесс сушки в камерах, в которых снимается основная или поверхностная влага. Изменяя время пребывания материала в них, добиваются требуемого качества материала на выходе сушильной установки, представленной на фигуре 1, при оптимальной производительности процесса сушки. Для этого оценивают влажность материала в указанных камерах множеством датчиков по нейронным сетям. Проверяют попадание рассчитанных значений влажности материала в заданные нормированные диапазоны, в зависимости от которых определяют числовые оценки мер доверия достижения требуемой влажности материала на выходе сушильной установки. На основании полученных значений производят нечеткий вывод, определяющий значение, на которое требуется изменить скорость движения пластинчатого конвейера для достижения требуемого качества материала на выходе сушильной установки, если полученное значение отлично от нуля, на него изменяют скорость движения пластинчатого конвейера.The problem is due to the fact that the final absolute moisture content of the material (

) the drying process in chambers in which the main or surface moisture is removed has a great influence. Changing the residence time of the material in them, they achieve the required quality of the material at the outlet of the drying installation shown in figure 1, with optimal performance of the drying process. To do this, evaluate the moisture content of the material in these chambers by a variety of sensors on neural networks. Check the calculated values of the moisture content of the material in the specified normalized ranges, depending on which the numerical estimates of confidence measures to achieve the desired moisture content of the material at the outlet of the drying installation are determined. Based on the obtained values, a fuzzy conclusion is made that determines the value by which it is necessary to change the speed of the plate conveyor to achieve the required quality of the material at the outlet of the drying unit, if the obtained value is non-zero, the speed of the plate conveyor is changed to it.

We call n and n + 1 chambers the most intense in terms of moisture removal or limiting. In the limiting chambers, basic or surface moisture is removed. In these chambers there is an air intake window that provides the flow of the drying agent into the chamber, the air is heated in a steam heater, the exhaust air is discharged through the exhaust gate by an exhaust fan, which operates in a cyclic mode. For the installation of sensors, it is necessary to determine the numbers of the limiting chambers, since chambers with different serial numbers can be stressed by moisture removal in various rolling mill drying plants (SVL).

The determined serial numbers of the limiting cameras depend on the design features, camera settings, the presence of heating on the rollers, and the total number of cameras (sections) - k in the SVL. To determine the chambers in which the maximum amount of moisture is removed, kinetic curves allow. So, for example, in five-section SVL with unheated rollers (Fig. 2, a), camera number two (n = 2) is considered as the first limiting chamber, and number three (n + 1 = 3) as the second limiting chamber. In four-section SVL with heated rollers (Fig. 2, b), the numbers of the limiting chambers are n = 1, and n + 1 = 2. According to the kinetic curves a, b in FIG. 2, approximately 60% -80% of the initial amount of moisture contained in the feed material is removed in chambers n and n + 1.

To implement the proposed drying method, it is sufficient to install the necessary sensors, which allow determining the change in factors affecting the speed of the drying process, in two limiting chambers n and n + 1. Based on the value of the initial moisture of the material and the information received from the sensors (weighted average temperature and air humidity in the chambers, percentage of the exhaust fan’s working time, presence of agitator, percentage of opening of air intake windows and gates, plate conveyor speed, humidity and air temperature in the workshop) in models (1) and (2) in real time, the moisture content of the material in the limiting chambers is estimated during its drying.

Models are neural networks that have a multilayer perceptron architecture, trained on the error back propagation algorithm. The obtained analytical models for determining the normalized values of material moisture in n and n + 1 chambers can be written as:

- углы наклона и сдвиги сигмоидальных активационных функций нейронов скрытого и выходного слоев нейронной сети;

,

и

,

- весовые коэффициенты связей нейронов скрытого и выходного слоев нейронной сети n и n+1 камер соответственно; N_n - число нейронов в скрытом слое сети для расчета влажности материала в n камере сушильной установки, N_n+1 - число нейронов в скрытом слое сети для расчета влагосодержания материала в n+1 камере сушильной установки; нормализованные значения:

- влажности материала;

- начальной влажности материала,

- управления работой вытяжного вентилятора;

- управления скоростью движения пластинчатого конвейера,

- наличия или отсутствия ворошителя;

- температуры и влажности окружающего воздуха, здесь о обозначает влияние факторов с этим индексом на все камеры сушильной установки;

,

,

,

- управления воздухозаборными окнами и сбросными шиберами;

,

,

,

- температуры и влажности сушильного агента в n и n+1 камерах соответственно. Значения влажности материала φ_n и φ_n+1 в лимитирующих камерах n и n+1 получают, денормализовав полученные значения

и

.where n and n + 1 are camera numbers,

- tilt angles and shifts of the sigmoidal activation functions of neurons of the hidden and output layers of the neural network;

,

and

,

- weighting coefficients of neuron connections of the hidden and output layers of the neural network of n and n + 1 cameras, respectively; N _n is the number of neurons in the hidden network layer for calculating the moisture content of the material in the n chamber of the drying unit, N _{n + 1} is the number of neurons in the hidden network layer for calculating the moisture content of the material in the n + 1 chamber of the drying plant; normalized values:

- humidity of the material;

- the initial moisture content of the material,

- control the operation of the exhaust fan;

- control the speed of movement of the plate conveyor,

- the presence or absence of agitator;

- temperature and humidity of the surrounding air, here o denotes the influence of factors with this index on all chambers of the drying installation;

,

,

,

- control of air intake windows and discharge gates;

,

,

,

- temperature and humidity of the drying agent in n and n + 1 chambers, respectively. The material moisture values φ _n and φ _{n + 1} in the limiting chambers n and n + 1 are obtained by denormalizing the obtained values

and

.

The obtained analytical dependences make it possible to estimate the moisture content of the material in n and n + 1 chambers of the drying unit during its drying with a relative error not exceeding 4%.

). Учитывая маленькое абсолютное значение (<1%) влажности материала на выходе сушильной установки -

, оценка ее по аналитическим моделям вида (1), (2) нежелательна. Поэтому используя накопленный опыт экспертов в виде нормированных диапазонов (

,

,

,

,

,

, где верхние индексы: х - влажность хорошая; уд - влажность удовлетворительная; пл - влажность плохая, а нижние индексы соответствуют номеру камеры) и числовых оценок двух свидетельств m₁(W) и m₂(D), методом Демпстера - Шафера определяется новое подмножество гипотез V, т.е. m₃(V). Значение w₃(V) рассчитывается по формуле:In the method of drying pasty materials with a change in the speed of movement of the plate conveyor to generate control actions, the determination of numerical estimates of confidence measures for the required material moisture at the outlet of the drying installation m (

) Given the small absolute value (<1%) of the moisture content of the material at the outlet of the drying unit -

, its assessment by analytical models of the form (1), (2) is undesirable. Therefore, using the accumulated experience of experts in the form of normalized ranges (

,

,

,

,

,

where the superscripts are: x - humidity is good; beats - satisfactory humidity; pl - the humidity is bad, and the subscripts correspond to the camera number) and numerical estimates of the two evidence m ₁ (W) and m ₂ (D), using the Dempster-Schafer method, a new subset of hypotheses V is determined, i.e. m ₃ (V). The value of w ₃ (V) is calculated by the formula:

where W is a subset of hypotheses based on the moisture content of the material φ _n ; D is a subset of hypotheses based on the material moisture value φ _{n + 1} . The values of m ₃ (V) calculated for each normalized range are numerical estimates of confidence building measures to achieve the desired moisture content of the material at the SVL outlet.

We give an example of determining numerical estimates of confidence building measures, achieving the required moisture content of the material at the output of a five-section SVL. Based on the experts' statement, the moisture content of the material in each limiting chamber (n = 2, n + 1 = 3) is divided into 3 normalized ranges. For each normalized range, experts appointed confidence measures:

<0,5%);- to event A, that the drying process will end well, i.e. at the output we get a material with a moisture content of not more than 0.5% (

<0.5%);

≤1%);- to event B, which means that the drying process will complete satisfactorily, i.e. at the output we get a material with a moisture content of 0.5% to 1% (0.5% ≤

≤1%);

>1%);- to event C, which indicates that the drying process will end poorly, i.e. at the output we get a material with a humidity of more than 1% (

>1%);

- to other elements of group Q of mutually exclusive events.

Confidence-building measures designated by experts are shown in Table 1.

Table 1 Expert Ratings Humidity range,% Range designation Drying ends well m (A) Drying will end satisfactorily m (B) Drying will end poorly m (C) Other possible situations m (Q) Camera n = 2 m ₁ (W) <40

0.65 0.27 0,07 0.01 from 40 to 50

0.4 0.43 0.15 0.02 > 50

0.15 0.32 0.5 0,03 Camera n + 1 = 3 m ₂ (D) <14

0.85 od 0.04 0.01 from 14 to 18

0.3 0.54 0.14 0.02 > 18

0.05 0.22 0.7 0,03

Consider the case when the moisture content of the material in the n chamber φ _n = 41%, and in n + 1 φ _{n + 1} = 8.7%. For this data, the following subsets will participate in the calculation of confidence measures for the required material moisture at the dryer outlet:

m ₁ {A} = 0.4; m ₁ {B} = 0.43; m ₁ {C} = 0.15; m ₁ {Q} = 0.02; and

m ₂ {A} = 0.85; m ₂ {B} = 0.1; m ₂ {C} = 0.04; m ₂ {Q} = 0.01.

The combination of the subsets m ₁ and m ₂ in two chambers without taking into account the normalization coefficient is given in table 2.

table 2 Combining evidence m ₁ and m ₂ m ₁ m ₂ m ₃ m ₁ {A} = 0.4 m ₂ {A} = 0.85 m ₃ {A} = 0.34 m ₁ {B} = 0.43 m ₂ {A} = 0.85 m ₃ {⌀} = 0.3655 m ₁ {C} = 0.15 m ₂ {A} = 0.85 m ₃ {⌀} = 0.1275 m ₁ {Q} = 0.02 m ₂ {A} = 0.85 m ₃ {A} = 0.017 m ₁ {A} = 0.4 m ₂ {B} = 0.1 m ₃ {⌀} = 0.04 m ₁ {B} = 0.43 m ₂ {B} = 0.1 m ₃ {B} = 0,043 m ₁ {C} = 0.15 m ₂ {B} = 0.1 m ₃ {⌀} = 0.015 m ₁ {Q} = 0.02 m ₂ {B} = 0.1 m ₃ {B} = 0.002 m ₁ {A} = 0.4 m ₂ {C} = 0.04 m ₃ {⌀} = 0.016 m ₁ {B} = 0.43 m ₂ {C} = 0.04 m ₃ {⌀} = 0.0172 m ₁ {C} = 0.15 m ₂ {C} = 0.04 m ₃ {C} = 0.006 m ₁ {Q} = 0.02 m ₂ {C} = 0.04 m ₃ {C} = 0,0008 m ₁ {A} = 0.4 m ₂ {Q} = 0.01 m ₃ {A} = 0.004 m ₁ {B} = 0.43 m ₂ {Q} = 0.01 m ₃ {B} = 0.0043 m ₁ {C} = 0.15 m ₂ {Q} = 0.01 m ₃ {C} = 0.0015 m ₁ {Q} = 0.02 m ₂ {Q} = 0.01 m ₃ {Q} = 0,0002

, то числовые оценки мер доверия, полученные по результатам объединения двух подмножеств, имеют следующие значения: m₃{A}=0,863; m₃{B}=0,119; m₃{C}=0,017; m₃{Q}=0,001. Таким образом, числовые оценки мер доверия: к событию A (процесс сушки завершится хорошо) равна 0,863, т.е. достаточно высока; к событию B (процесс сушки завершится удовлетворительно) - 0,119; к событию C (процесс сушки завершится плохо) равна 0,017; мера доверия к возникновению остальных возможных ситуаций - 0,001. Аналогично рассчитываются числовые оценки мер доверия для всех остальных комбинаций диапазонов влажности материала в n и n+1 камерах. Полученные значения всех возможных комбинаций для пятисекционной СВЛ приведены в таблице 3. В последней графе таблицы приведена влажность материала на выходе сушильной установке

соответствующая приведенным ситуациям, полученная путем лабораторных замеров.Since the denominator in formula (3) is 1 -

, then the numerical estimates of confidence measures obtained by combining the two subsets have the following values: m ₃ {A} = 0.863; m ₃ {B} = 0.119; m ₃ {C} = 0.017; m ₃ {Q} = 0.001. Thus, the numerical estimates of confidence building measures: for event A (the drying process will complete well) is 0.863, i.e. high enough; to event B (the drying process will complete satisfactorily) - 0.119; to event C (the drying process will end poorly) is 0.017; a measure of confidence in the occurrence of other possible situations is 0.001. Similarly, numerical estimates of confidence measures are calculated for all other combinations of material moisture ranges in n and n + 1 chambers. The obtained values of all possible combinations for a five-section SVL are given in table 3. The last column of the table shows the moisture content of the material at the outlet of the drying unit

corresponding to the given situations obtained by laboratory measurements.

Table 3 Numerical estimates of confidence measures for the required moisture content of the material at the outlet of the drying plant No. Situations m (A) m (B) m (C) m (Q)

one (

,

) 0,027 0.178 0.793 0.002 2,34 2 (

,

) 0.123 0.416 0.459 0.002 1.77 3 (

,

) 0.169 0.579 0.25 0.002 1.51 four (

,

) 0.29 0.386 0.322 0.002 1.42 5 (

,

) 0.325 0.61 0,065 0.001 1,08 6 (

,

) 0.555 0.412 0,033 0,0005 0.96 7 (

,

) 0.705 0.174 0.12 0.001 0.64 8 (

,

) 0.862 0.118 0.02 0,0005 0.48 9 (

,

) 0.942 0.051 0.0065 0,0002 0.35

Similarly, numerical estimates of confidence building measures for SVL with a different number of cameras are calculated.

The rapid determination of the numerical evaluation of the confidence measure for the required moisture content of the material at the outlet of the drying installation is carried out according to Table 3. Depending on the normalized ranges, the situation is determined according to which a numerical assessment of the confidence measure for the event is selected - A - the drying process will end well, used in further calculations.

).The drying process is controlled by changing the speed of the plate conveyor using fuzzy logic based on the obtained values of φ _n , φ _{n + 1} , m (

)

- числовые оценки мер доверия к достижению требуемой влажности материала на выходе сушильной установки. Основной выходной лингвистической переменной является U^сл - «скорость движения пластинчатого конвейера». В качестве условных обозначений термов входных лингвистических переменных используются Ф_n, Ф_n+1={ННД, НД, Д, ВД, НВД};

={М, СР, В} (см. таблицы 4-6), здесь ННД - влажность материала намного ниже допустимой, НД - влажность материала ниже допустимой; Д - влажность материала допустимая; ВД - влажность материала выше допустимой; НВД - влажность материала намного выше допустимой. Для их описания использованы трапециевидные функции принадлежности (см. фиг.3а,б,в,г).The following are considered as input linguistic variables: Ф _n - “material humidity at the chamber exit n”; F _{n + 1} - "material moisture at the output n + 1 of the camera" and

- numerical estimates of confidence building measures to achieve the required moisture content of the material at the outlet of the drying plant. The main output linguistic variable is U ^SL - “plate conveyor speed”. As the symbols for the terms of the input linguistic variables are used Ф _n , Ф _{n + 1} = {NI, ND, D, VD, NVD};

= {M, SR, B} (see tables 4-6), here NND - material moisture is much lower than permissible, ND - material moisture is lower than permissible; D - permissible moisture content of the material; VD - material humidity is higher than permissible; NVD - the humidity of the material is much higher than permissible. To describe them, trapezoidal membership functions were used (see Fig. 3a, b, c, d).

Table 4 Membership functions of the terms of the linguistic variable Ф _n (material moisture at the chamber exit n) Term number Symbols of terms of variable Ф _n Term membership functions one Nnd 1/0 + 1 / 17.5 + 0 / 22.5 2 Nd 0 / 17.5 + 1 / 22.5 + 1 / 27.5 + 0 / 32.5 3 D 0 / 27.5 + 1 / 32.5 + 1 / 37.5 + 0 / 42.5 four VD 0 / 37.5 + 1 / 42.5 + 1 / 47.5 + 0 / 52.5 5 NVD 0 / 47.5 + 1 / 52.5 + 1/100 Table 5 Membership functions of the terms of the linguistic variable Ф _{n + 1} (material moisture at the output n + 1 of the chamber) Term number Symbols of terms of a variable Term membership functions one Nnd 1/0 + 1/5 + 0/7 2 Nd 0/5 + 1/7 + 1/9 + 0/11 3 D 0/9 + 1/11 + 1/13 + 0/15 four VD 0/13 + 1/15 + 1/17 + 0/19 5 NVD 0/17 + 1/19 + 1/100 Table 6

(мера доверия к влажности материала на выходе сушилки)Membership Functions of Terms of a Linguistic Variable

(a measure of confidence in the moisture content of the material at the outlet of the dryer) Term number Symbols of terms of a variable

Term membership functions one N 1/0 + 1 / 0.575 + 0 / 0.675 2 CP 0 / 0.575 + 1 / 0.675 + 1 / 0.725 + 0 / 0.825 3 AT 0 / 0.725 + 1 / 0.825 + 1/1

Table 7 presents the membership functions of the output linguistic variable U ^SL , which can take five values: SPOV - greatly increase the speed of the tape; POV - increase the speed of movement of the plate conveyor; О - leave the speed of the plate conveyor unchanged; PON - reduce the speed of the plate conveyor; SPON - greatly reduce the speed of movement of the plate conveyor, i.e. U ^SL = {SPON, PON, O, POV, SPOV}.

Table 7 The membership functions of the terms of the linguistic variable U ^SL (change in the speed of movement of the plate conveyor) Term number Legend of terms variable U ^cl Term membership functions one SPON 1 / -0.3 + 1 / -0.2 + 0 / -0.17 2 PON 0 / -0.2 + 1 / -0.17 + 1 / -0.07 + 0 / -0.04 3 ABOUT 0 / -0.07 + 1 / -0.04 + 1 / 0.04 + 0 / 0.07 four POV 0 / 0.04 + 1 / 0.07 + 1 / 0.17 + 0 / 0.2 5 SPOV 0 / 0.17 + 1 / 0.2 + 1 / 0.3

A graphical representation of the membership functions of the terms of linguistic variables is shown in Fig.3.

The algorithm for solving the problem of determining the optimal speed of the plate conveyor is performed in the following sequence:

, а также определение значений функций принадлежности их термов.Step 1. Fuzzification of input linguistic variables Ф _n , Ф _{n + 1} and

, as well as determining the values of membership functions of their terms.

Step 2. Fuzzy conclusion, using production rules of the form:

» И «Ф_n+1 есть

» И «

есть М^T», ТО «U^сл есть (U^сл)^T», здесь индекс T обозначает соответствующее значение лингвистической переменной.IF "f _n is

"And" Ф _{n + 1} is

"And"

there is M ^T ”, then“ U ^SL is (U ^SL ) ^T ”, here the index T denotes the corresponding value of the linguistic variable.

Step 3. Accumulation. At this stage, the resulting fuzzy sets for the terms of the output linguistic variable are combined into one fuzzy set.

,

,

}.Step 4. Defuzzification (conversion) of the output linguistic variable into the numerical values of the change in the speed of the tape using three different methods of defuzzification (the center of gravity method; the median method; the first maximum method), i.e. getting multiple values {

,

,

}.

. Для данного интервала производится расчет оптимального значения скорости пластинчатого транспортера (

)^∗, при котором функционал, учитывающий потери от снижения качества конечной продукции и производительности сушильной установки, достигает минимального значения:Step 5. Based on the obtained values, an interval of permissible changes in control actions is formed.

. For this interval, the optimal value of the plate conveyor speed is calculated (

) ^∗ , in which the functional, taking into account losses from a decrease in the quality of the final product and the productivity of the drying installation, reaches a minimum value:

- нормированное значение скорости пластинчатого транспортера.where c ₁ , c ₂ - weighting factors; b ₀ , b ₁ - parameters of the function of loss of quality of the final product; a ₀ , a ₁ - parameters of the function of the loss of productivity of the process,

- the normalized value of the speed of the plate conveyor.

The algorithm of the method for drying pasty materials in a roll dryer with a change in the speed of movement of the plate conveyor includes the following steps:

Step 1. Determining the current speed of the plate conveyor and the residence time of the material in the chambers of the drying installation.

Step 2. Determination of material moisture content φ _n in _n chamber.

Step 3. Determination of the normalized range (see Table 1) and the corresponding set of hypotheses m ₁ (W).

Step 4. Determination of material moisture content φ _{n + 1} in n + 1 chamber.

Step 5. Determination of the normalized range (see Table 1) of the corresponding set of hypotheses m ₂ (D).

Step 6. Calculation by the Dempster-Schafer method of numerical evaluations of confidence building measures to achieve the required quality of the material at the outlet of the drying unit with its current settings.

)^∗, на которое необходимо изменить скорость движения пластинчатого конвейера.Step 7. Fuzzy inference and determination according to the functional (4) of the optimal value (

) ^∗ , by which it is necessary to change the speed of movement of the plate conveyor.

Step 8. Change the speed of the plate conveyor.

Step 9. Go to the first step.

=3,5 м/ч. При этой скорости влажность материала в n камере φ_n=43% принадлежит нормированному диапазону

, а в камере n+1 - φ_n+1=23% и принадлежит нормированному диапазону

. По методу Демпстера - Шафера мера доверия к достижению требуемой влажности на выходе равна 0,123. Осуществим вывод по следующим продукционным правилам:We give an example of determining the speed of movement of a plate conveyor for a five-section SVL. Let the current value of the speed of the tape

= 3.5 m / h. At this speed, the moisture content of the material in the n chamber φ _n = 43% belongs to the normalized range

, and in the chamber n + 1 - φ _{n + 1} = 23% and belongs to the normalized range

. According to the Dempster-Schafer method, the measure of confidence in achieving the desired output humidity is 0.123. We make a conclusion according to the following production rules:

(see figure 4).

)^∗, на которое следует изменить скорость пластинчатого транспортера, определенное согласно функционалу (4), равно - 0,24. Следовательно, скорость пластинчатого конвейера следует понизить, т.е. увеличить время пребывания материала в камерах СВЛ. Новая скорость движения пластинчатого конвейера

будет равна 3,26 м/ч.Optimal value (

) ^∗ , by which the plate conveyor speed should be changed, determined according to the functional (4), is - 0.24. Therefore, the speed of the plate conveyor should be reduced, i.e. increase the residence time of the material in the SVL chambers. New plate conveyor speed

will be equal to 3.26 m / h.

At the next iteration, as a result of a change in the speed of movement of the plate conveyor, the residence time of the material increased, and the moisture content of the material in the chambers changed as follows:

=37% - принадлежит нормированному интервалу

,

=12% - принадлежит нормированному интервалу

(см. табл.1). При этих значениях влажности материала в камерах СВЛ мера доверия к достижению требуемой влажности на выходе m(

)=0,942 (см. табл.3).

= 37% - belongs to the normalized interval

,

= 12% - belongs to the normalized interval

(see table 1). At these moisture values of the material in the SVL chambers, a measure of confidence in achieving the desired humidity at the outlet m (

) = 0.942 (see table 3).

The conclusion is carried out according to the following rule:

,

,

(see figure 5).

оставлена на прежнем уровне и равна 3,26 м/ч. В результате управления влажность материала на выходе СВЛ

снизилась с 1,77% до 0,35%.Plate conveyor speed

left at the same level and equal to 3.26 m / h. As a result of control, the moisture content of the material at the SVL outlet

decreased from 1.77% to 0.35%.

A method of drying pasty materials in a roller mill with a change in the speed of movement of the plate conveyor has been introduced at Pigment OJSC, Tambov. This made it possible to increase the probability of high-quality products reaching 0.98, and also to increase the productivity of drying processes by 5%. Thus, the claimed technical result is achieved - the achievement of the required quality of the material at the outlet of the drying installation with optimal performance of the drying process.

Information sources

1. Chernobyl II etc. Drying plants of the chemical industry. Kiev, Technique, 1969, p. 124.

2. Lebedev P.D. Heat exchangers, drying and refrigeration units. Textbook for students of technical universities. Vol. 2, rev. M., Energy, 1972, p. 317.

Description of figures:

Figure 1. Roll-type drying unit (SVL):

1 - case, 2 - roller system, 3 - knife mechanism, 4 - plate conveyor (belt), 5 - drying agent removal system, 6 - sliding system for removing drying agent from each chamber, 7 - exhaust fan, 8 - air intake windows to ensure the flow of fresh drying agent, 9 - heaters, 10 - recirculation fans, 11 - sensor of the initial temperature of the drying agent, 12 - sensor of the initial humidity of the drying agent, 13 and 14 - sensors, respectively, of the temperature and humidity of the drying agent in the nth chamber, 15 and 16 - sensors respectively the temperature and humidity of the drying agent in the n + 1 chamber, 17 and 18 are the sensors of the degree of opening of the intake windows in the n-th and n + 1 chambers, 19 and 20 are the sensors of the degree of opening of the gates in the n-th and n + 1 chambers, 21 - sensor for the presence of agitator.

начальная влажность материала; d - длина СВЛ, м.Figure 2. Kinetics of the drying process in SVL: 1 - curve of the drying process,

initial moisture content of the material; d is the length of the SVL, m

Figure 3. The membership functions of the terms of a linguistic variable:

a) f _n (humidity of the material at the outlet n of the chamber);

b) Ф _{n + 1} (material moisture at the outlet n + 1 of the chamber);

(мера доверия к достижению требуемой влажности материала на выходе сушилки);at)

(a measure of confidence in achieving the required moisture content of the material at the outlet of the dryer);

g) ΔU ^SL (change in the speed of movement of the plate conveyor).

Figure 4. The value of the output linguistic variable:

a) "the humidity of the material at the outlet of the n section";

b) "humidity of the material at the output of n + 1 sections";

c) “a measure of confidence in achieving the required moisture content of the material at the outlet of the dryer”;

d) "change in the speed of movement of the plate conveyor."

Figure 5. The value of the output linguistic variable:

a) "the humidity of the material at the outlet of the n section";

b) "humidity of the material at the output of n + 1 sections";

c) “a measure of confidence in achieving the required moisture content of the material at the outlet of the dryer”;

d) "change in the speed of movement of the plate conveyor";

Claims (1)

A method of drying pasty materials in a roll mill dryer, comprising molding the material to be dried, moving it sequentially through the drying chambers during countercurrent cross-movement of the drying agent, the removal of which is carried out in a cyclic mode, including saturation of the drying agent and the removal of the drying agent, the duration of the drain and stop being determined in depending on the specified moisture content of the drying agent, characterized in that, depending on the speed of movement of the plate to the conveyor determines the residence time of the dried material in n and n + 1 chambers of the drying unit, in which the main or surface moisture is removed, real-time assess the moisture content of the material in n and n + 1 chambers with a variety of sensors on neural networks, check the calculated values of the material moisture in normalized ranges, depending on which numerical estimates of confidence building measures to achieve the required moisture content of the material at the outlet of the drying unit and based on the obtained values of the moisture content of the material are determined said chambers and numerical estimates CBMs produce fuzzy inference determining the amount by which the advance speed of the conveyor plate to achieve the desired quality of the material at the outlet of the dryer.

Images