RU2046809C1 - Method of operation of simple polyester polymerization process - Google Patents

Abstract

FIELD: chemical industry. SUBSTANCE: method involves determination of actual reactor loading and actual partial pressure of propylene oxide vapor. Consumption of propylene oxide is regulated by definite partial pressure of propylene oxide vapor. Consumption of cooling water is regulated by deviation of measured reaction temperature from assigned value with correction by definite actual reactor loading. EFFECT: improved method of operation. 3 dwg

Description

 The invention relates to the automation of polymerization processes in chemical engineering industries, in particular in the production of simple polyester resins, and can be used in chemical, petrochemical and other industries.

A known method of producing polyethers by copolymerization of propylene oxide and ethylene at an elevated temperature carried out in two stages [1]
In this method, the supply of ethylene oxide is carried out without taking into account the partial pressure in the reactor, which does not allow to reduce the supply time, and therefore, to achieve maximum plant performance. In addition, the lack of regulation of the supply of cooling water to the heat exchanger depending on the load of the reactor does not allow to accurately maintain the desired reaction temperature, which reduces the quality of the target product.

The closest to the proposed technical essence and the achieved result is a method of controlling the polymerization of polyethers in batch type polymerization reactors by measuring the temperature of the reaction mass in the reactor and heat exchanger, measuring the flow rate of propylene oxide, controlling the flow rate of oxide depending on the pressure and temperature in the reactor and regulating the flow of cooling water in the heat exchanger [2]
However, the total pressure in the reactor, in addition to the partial pressure of the oxide vapor (which ultimately determines the heat generated), is significantly affected by the partial pressure of nitrogen, which depends on the temperature and load of the reactor. This makes it difficult to control the flow of oxide in order to achieve the highest possible heat dissipation and, therefore, the maximum feed rate. At the same time, the known method for regulating the flow of water does not take into account the reactor load measured during the process, which does not allow achieving the necessary stabilization accuracy t of the reaction mass, which affects the quality of the product.

 The technical result of the invention is to increase productivity by reducing the time of polymerization, improving the quality of the finished product.

 This is achieved by the fact that in the method for controlling the polymerization of polyethers in batch type polymerization reactors, the current load of the reactor and the current partial vapor pressure of propylene oxide are determined, the flow rate of the oxide is controlled by a specific partial vapor pressure of the oxide, the flow rate of cooling water is controlled by the deviation of the measured reaction temperature from set value with correction for a certain current reactor load.

 The combination of new features in combination with the known ones gives this technical solution new properties, which include determining the current reactor load and the current partial vapor pressure of propylene oxide (ethylene), regulating the oxide flow rate at a specific partial oxide vapor pressure, and regulating the cooling water flow rate by deviating the measured reaction temperature from a given value with correction for a certain current reactor load, which provide an increase in product quality and an increase in productivity by reducing the time of polymerization.

 In FIG. 1 shows a schematic diagram of a polymerization process control system that implements this method; in FIG. 2 and 3 graphs of comparative analysis.

 The polymerization process is carried out in a batch polymer reactor 1 with an external heat exchanger 2 by gradually adding to the preloaded prepolymer a predetermined amount of propylene oxide (or ethylene oxide, depending on the stage and grade of polyester, hereinafter referred to as oxide). As a result of the exothermic polymerization reaction in the presence of an inert gas (nitrogen), a high molecular weight polymerizate is formed. The heat of reaction is removed by forced heat transfer in the external heat exchanger 2. The process is carried out subject to restrictions on the total pressure in the reactor 1 and on the temperature of the reaction mass in the reactor and the heat exchanger. Violation of temperature restrictions leads to a decrease in product quality.

 The control system implementing this method includes a sensor 3 for the total pressure in the reactor, a sensor 4 for the temperature of the reaction mass in the reactor, a sensor 5 for the current flow rate of the oxide, a sensor 6 for the temperature of the reaction mass in the heat exchanger, a regulator 7 for the flow of oxide, and a regulator 8 for the flow of cooling water. The diagram also shows the functional blocks implemented in the control computer 9: a block 10 for determining the current reactor load, a block 11 for determining the current partial pressure of oxide vapor, a block 12 for determining the oxide flow rate, and a block for determining the cooling water flow rate 13.

 The control scheme is implemented on the basis of commercially available domestic automation and computer technology.

 This method is as follows.

u_j, (1) где V^фп объем форполимера, м³;
U_j расход окиси на j-м такте, м³.According to the values of oxide consumption measured by the sensor 5 with a given discreteness of control of this consumption (1-2 min) in block 10, the current reactor load V _i is determined at each discrete moment i
V _i = V ^fp +

u _j , (1) where V ^{fp is the} volume of the prepolymer, m ³ ;
U _j consumption of oxide on the j-th cycle, m ³ .

Then, in block 11, according to the values of the total pressure P _i and the temperature in the reactor T measured by the sensors 3 and 4 $\genfrac{}{}{0ex}{}{\text{p}}{\text{i}}$ determine the value of the selected partial vapor pressure of the oxide P ^ok _i
P ^ok _i = P _i -P ^N _i , (2)
P ^N _i = (P _o ˙ V ^g _o / T ^p _o ) / (V ^g _i / T $\genfrac{}{}{0ex}{}{\text{p}}{\text{i}}$ + P ˙K ˙Vi ₎ , (3)
V $\genfrac{}{}{0ex}{}{\text{g}}{\text{i}}$ V ^total -V _i , (4) where P _{i is the} total pressure in the reactor, atm;
P ^N _{i the} partial pressure of nitrogen, atm;
V ^g _{i the} volume of the gaseous phase in the reactor, m ³ ;
T ^p _{i the} absolute temperature of the reaction mass in the reactor, deg;
V ^total total volume of the reactor and heat exchanger, m ³ ;
P _{about the} pressure in the reactor before loading the oxide, atm;
V ^g _{about the} initial volume of the gaseous phase, m ³ equal to
V ^r _o = V ^total -V ^fp , (5)
T ^r _{about the} temperature in the reactor before loading the oxide, deg;
R is the universal gas constant;
To the coefficient of solubility of nitrogen in an alkaline polymerizate (the value of RK, 1 / deg, is determined experimentally).

 Then, in block 12, it determines the current oxide consumption, which ensures the minimum total time for oxide supply, subject to the restrictions on the partial pressure of oxide vapor.

 It is known that the partial vapor pressure of an oxide characterizes the concentration of unreacted oxide in the polymerizate.

 The indicated concentration determines the polymerization reaction rate and, consequently, the reaction heat generated. The oxide consumption that is optimal in terms of the total supply time corresponds to the maximum consumption, subject to restrictions on heat removal, and therefore on the allocated vapor pressure of the oxide.

 This flow rate can be determined, for example, as follows.

(U_i+j), U_i+j-1,) прогнозируемое парциальное давление паров окиси в будущий момент i+j, зависящее от текущей и предыдущих подач окиси, а

максимально допустимое парциальное давление паров окиси, ограниченное возможностями теплосъема. Определяют максимально возможное U_i, отвечающее условиям.Let U _{i be the} desired current oxide consumption,

(U _{i + j} ), U _{i + j-1} ,) the predicted partial pressure of the oxide vapor at the future moment i + j, depending on the current and previous oxide supplies, and

maximum permissible partial vapor pressure of oxide, limited by heat removal capabilities. Determine the maximum possible U _i that meets the conditions.

(0,0, u_i, u_i-1,) ≅

, j 1,2,j, (6)
0 ≅ u_i≅

, (7) где

максимально допустимый расход окиси.

(0,0, u _i , u _i-1 ,) ≅

, j 1,2, j, (6)
0 ≅ u _i ≅

, (7) where

maximum allowable oxide consumption.

Here, future oxide feeds U _{i + 1} , U _{i + j} = 0, and previous U _i-1 , U _{i-2 are} measured.

 Condition (6) must be fulfilled for j future cycles (in practice, it is enough to take j = 2, since at future zero flows the pressure of the oxide vapor will begin to drop sharply).

h_k·u_i-k+φ_i, (8) где h_k, K=0,1, N весовые коэффициенты линейной модели;
φ_i случайная составляющая.Prediction of the partial pressure of oxide vapor is carried out using a specially defined pressure model, for example, linear
P _i =

h _k · u _ik + φ _i , (8) where h _k , K = 0,1, N are the weighting coefficients of the linear model;
φ _{i is a} random component.

=

h_k+j·u_i-k+

, (9) где

вычисляется стандартными методами прогнозирования случайных процессов.The predicted vapor pressure of the oxide, taking into account formula (6), takes the form

=

h _{k + j} · u _{ik +}

, (9) where

calculated by standard methods for predicting random processes.

. (10)
С учетом ограничений (7) текущий расход окиси определяют
u_i= max{0, min{u $\genfrac{}{}{0ex}{}{\text{1}}{\text{i}}$ , u $\genfrac{}{}{0ex}{}{\text{2}}{\text{i}}$ , u $\genfrac{}{}{0ex}{}{\text{j}}{\text{i}}$ ,

} (11)
Выбираемый согласно (11) расход окиси U_i используется в качестве уставки в регуляторе расхода окиси. В следующий дискретный момент i+1 процедура расчета расхода окиси повторяется на базе новых измерений с помощью датчиков 3, 4 и 5.Considering (6) in the form of strict equalities, we obtain for each j
u $\genfrac{}{}{0ex}{}{\text{j}}{\text{i}}$

. (10)
Given the restrictions (7), the current oxide consumption is determined
u _i = max {0, min {u $\genfrac{}{}{0ex}{}{\text{1}}{\text{i}}$ , u $\genfrac{}{}{0ex}{}{\text{2}}{\text{i}}$ , u $\genfrac{}{}{0ex}{}{\text{j}}{\text{i}}$ ,

} (eleven)
The oxide consumption U _i selected according to (11) is used as the set point in the oxide flow controller. At the next discrete moment i + 1, the procedure for calculating the oxide consumption is repeated on the basis of new measurements using sensors 3, 4, and 5.

Качество стабилизации температуры определяет качество готового продукта и может быть достигнуто только при учете меняющейся вместе с загрузкой реактора 1 инерционности охлаждаемой реакционной массы. Эта цель достигается, например, с помощью реализуемого в ЭВМ 9 цифрового ПИД-регулятора с настройкой, зависящей от определяемой в блоке 10 текущей загрузки реактора V_i (K)
W_i= A(V_i)·(T $\genfrac{}{}{0ex}{}{\text{т}}{\text{i}}$ ^о-

)+B(V_i)(T $\genfrac{}{}{0ex}{}{\text{т}}{\text{i}}$ ^о-T $\genfrac{}{}{0ex}{}{\text{то}}{\text{i-}}$ ₁)+C(V_i)

(T $\genfrac{}{}{0ex}{}{\text{то}}{\text{i-}}$ _j-T), (12) где К глубина интегрирования;
A(V_i), B(V_i), C(V_i) настройки ПИД-регулятора, зависящие от загрузки реактора, например, линейно
A(V_i)=aV_i B(V_i)=bV_i, C(V_i)=cV_i.
С ростом инерционности реакционной массы значения настроечных коэффициентов регулятора увеличиваются. Определяемый блоком 12 расход воды W_i используется в регуляторе 8 расхода воды в качестве уставки. В следующий дискретный момент i+1 процедура определения расхода повторяется на базе новых измерений.In block 13 with a given discreteness of control of the cooling water supply (15 s), the current water flow rate W _i is determined in order to maintain the temperature of the reaction mixture in the heat exchanger T _i ^then at the level required by the technological regulations

The quality of temperature stabilization determines the quality of the finished product and can be achieved only by taking into account the inertia of the cooled reaction mass changing with the loading of reactor 1. This goal is achieved, for example, by using a digital PID controller implemented in computer 9 with a setting that depends on the current load of the reactor V _i (K) determined in block 10
W _i = A (V _i ) $\genfrac{}{}{0ex}{}{\text{t}}{\text{i}}$ ^about -

) + B (V _i ) (T $\genfrac{}{}{0ex}{}{\text{t}}{\text{i}}$ ^about -T $\genfrac{}{}{0ex}{}{\text{then}}{\text{i-}}$ ₁ ) + C (V _i )

(T $\genfrac{}{}{0ex}{}{\text{then}}{\text{i-}}$ _j -T), (12) where K is the depth of integration;
A (V _i ), B (V _i ), C (V _i) PID controller settings, depending on the reactor load, for example, linearly
A (V _i ) = aV _i B (V _i ) = bV _i, C (V _i ) = cV _i.
With an increase in the inertia of the reaction mass, the values of the tuning coefficients of the controller increase. The water flow rate W _i determined by block 12 is used in the water flow controller 8 as a set point. At the next discrete moment i + 1, the flow determination procedure is repeated based on new measurements.

 Accounting for the load of the reactor when calculating the flow rate of the unit 13 allows by reliable stabilization of the reaction temperature to improve the quality of the finished product. At the same time, this allows achieving stabilization of the partial pressure of oxide vapor near the maximum permissible level by calculating the consumption of oxide, which leads to a reduction in the polymerization time.

 PRI me R 1 (prototype). Laprom 5003-2B-10 brand polyester is synthesized, prepolymer loading 2070 kg, propylene oxide loading 11500 kg, ethylene oxide loading 2100 kg.

 Pressure control is carried out by changing the oxide feed rate, and the reaction temperature is controlled by changing the flow rate of cooling water.

In the graph of FIG. 2 shows the changes in the total pressure and temperature during the reaction. In steady-stationary process conditions the overall average pressure was 3.5 kgf / cm ^2, an average reaction temperature of 114 ^{° C,} with a maximum value of up to ^about 121 C. The average oxide feeding rate 1.9 ^m3 / h. The line capacity was 1.0 t / h. Prepared finished product with a content of unsaturated compounds (iodine number) J 1.95 g _2/100 g

 PRI me R 2 (by the proposed method). The synthesis of polyester brand Laprom 5003-2B-10 is carried out at downloads of raw materials similar to example 1.

 The oxide supply is controlled by the selected partial oxide vapor pressure, and the cooling water flow is controlled by the reaction temperature with correction from the current reactor load.

In the graph of FIG. 3 shows the total pressure and reaction temperature in Example 2. Under steady-state steady-state process conditions, the total average pressure was 4.0 kgf / m ² , including the allocated partial vapor pressure of the oxide was kept within 2.4-3.0 kgf / cm ² . The average reaction temperature was 118 ^{° C} at the maximum to ^about 119.1 C. An average oxide feeding rate 2.6 ^m3 / h. The line capacity was 1.25 m ³ / h. Prepared finished product with a content of unsaturated compounds (iodine number) J 1.45 g _2/100 g

 From the above examples it is seen that the use of the proposed method for controlling the polymerization process of polyethers allows, by reducing the polymerization stage, to increase the productivity of one line by 25% and to improve the quality of the finished product by reducing the maximum temperature (in terms of iodine number).

 The implementation of the method in the production of polyethers allows to reduce the duration of the polymerization stage by an average of 5% per process unit with a capacity of 7.5 thousand tons of polyester per year.

Claims (1)

 METHOD FOR CONTROL OF THE POLYMERIZATION PROCESS OF SIMPLE POLYESTERS in batch type polymerization reactors by measuring the temperature of the reaction mixture in the reactor and heat exchanger, measuring the total pressure in the reactor, measuring the flow rate of propylene oxide, controlling the flow rate of the oxide depending on the pressure and temperature in the reactor and regulating the flow rate of cooling water in heat exchanger depending on the temperature of the reaction mixture in the heat exchanger, characterized in that determine the current load of the reactor and the current batch noe propylene oxide pressure vapor flow rate is adjusted by a particular oxide partial pressure oxide vapor, the cooling water flow rate is controlled by the deviation of the measured reaction temperature setpoint adjusted in a certain current load of the reactor.

Images