JPH0243285A - Method for control of reactor - Google Patents

Abstract

PURPOSE:To efficiently control an operation to thereby extend a tune-up cycle by predetermining a specified load allowance function and allotting a load according to the control factor data for each reactor in a reaction series involving a plurality of reactors. CONSTITUTION:Preparation is made for a load allowance function which is a function showing the operating conditions of each reactor in a reaction series involving a plurality of reactors in terms of the degree of approximation to a preset control limit using factors for controlling the operation of each reactor as variables. In operation, a degree of load allowance, i.e., a load allowance function value, is calculated by a computer from the control factor data for each reactor; according to a requirement for changing the quantity of a conversion reaction product to be manufactured, a required amount of change in series load, which corresponds to a required amount of change in the necessary quantity of the conversion reaction product to be manufactured, is allotted to each reactor depending upon said degree of load allowance, thus giving a required amount of change in reactor load, i.e., a required amount of change in load for each reactor; the set values of control factors for each reactor are changed so as to give a necessary quantity of the conversion reaction product to be manufactured, which is a quantity of the conversion reaction product to be manufactured corresponding to said required amount of change in reactor load; and the above-mentioned procedures are repeated to control the reactors.

Description

[Detailed description of the invention] [Industrial field of application] /1. Reactor for efficiently operating a reactor series consisting of a plurality of reactors such as a thermal cracking furnace for chlorinated hydrocarbons such as 2-dichloroethane Concerning a control method.

[Conventional technology]

Chemical plants that perform conversion reactions in a reactor series consisting of multiple reactors usually shut down each reactor in sequence and periodically while ensuring a constant production amount of the desired product for the entire series. Tune-up is practiced.

Such types of chemical plants include, for example, hydrocarbon pyrolysis plants, chlorinated hydrocarbon pyrolysis plants, and the like. The conventional operating conditions will be explained below, mainly taking a hydrocarbon pyrolysis plant as an example.

Thermal decomposition of gaseous hydrocarbons such as ethane, propane, and putane, and liquid hydrocarbons such as naphtha, gas oil, kerosene, light oil, and heavy oil in the presence of steam (steam cranking)
Methods for producing olefinic hydrocarbons such as ethylene and propylene are well known. In the above pyrolysis, carbonaceous material (coking) occurs in the cracking tube (cheap) of the pyrolysis furnace and is deposited on the inner wall of the tube, so the cracking furnace must be stopped at regular intervals for decoking. ing. Normally, industrial olefin production plants have multiple furnaces, for example, 10 or more cracking furnaces, and while ensuring a constant amount of olefin production for the entire plant,
Decoking treatment is carried out periodically for each cracking furnace. In this case, the operator should take into consideration the current operating conditions of each cracking furnace during decoking process and adjust the amount of olefin produced by each cracking furnace to suit the amount of olefin produced in the cracking furnace during the decoking process. This is handled by allocating different types of fuel and changing operating conditions (feedstock hydrocarbon feed rate, thermal decomposition temperature, etc.) accordingly.

[Problem to be solved by the invention]

However, with the above-mentioned method of allocating the load to each reactor, in which the amount of olefin produced in each cracking furnace is roughly allocated based on the operator's experience, the deterioration state of each reactor cannot be fully grasped, and therefore each Since it is not possible to accurately and precisely judge how much room there is in the reactor, it may put more load than necessary on a reactor that does not have room, so it is necessary to tune up the reactor, such as shortening the decoking cycle of the cracking furnace. This was not a simple matter, but there were problems such as variations in the deterioration state of each reactor.

[Means to solve the problem]

In view of the above-mentioned state of the prior art, the present inventors have completed the present invention as a result of intensive study on a method to accurately grasp the deterioration state of each reactor and operate each reactor with appropriate load distribution. .

That is, the gist of the present invention is that when performing a conversion reaction in a reactor series consisting of a plurality of reactors whose operation is controlled under the set value of the control factor selected, each reaction during operation is The operating state of the reactor should be predicted using the control factors that control the operation of each reactor as variables, and when operating the reactor series, [1] (a) Control factors should be determined for each reactor. (b) Calculate the value of the load margin function (hereinafter referred to as "load margin") by importing the data into a computer; (b) Calculate the value of the load margin function (hereinafter referred to as "load margin"); The amount of change in the production amount of conversion reaction products (hereinafter referred to as ``series load change amount'') is distributed to each reactor according to the size of the load margin, and the amount of load change for each reactor (hereinafter referred to as ``reactor load change amount'') is (hereinafter referred to as the "required conversion reaction product production amount");); repeating the steps; Applies to control methods. Specifically, the method of the present invention can be used, for example, when producing olefins by thermal decomposition of hydrocarbons such as naphtha, or when producing vinyl chloride monomers by thermally decomposing chlorinated hydrocarbons such as 1,2-dichloroethane. In order to operate each pyrolysis furnace in an appropriate condition in a pyrolysis furnace series with multiple pyrolysis furnaces used for It is suitably applied to a furnace control method.

EMBODIMENT OF THE INVENTION Hereinafter, the present invention will be described in more detail with reference to the drawings, taking a method of controlling a hydrocarbon pyrolysis furnace as an example.

In industrial hydrocarbon pyrolysis, the raw material hydrocarbons are hydrocarbons that are gaseous at room temperature, such as ethane, propane, and butane, or liquid at room temperature, such as naphtha, gas oil, kerosene, light oil, and heavy oil. Hydrocarbons are used. Each of these hydrocarbon raw materials may be used alone (single cracking), or a liquid hydrocarbon and a gaseous hydrocarbon may be used together (co-cracking).

FIG. 1 is a schematic diagram showing an example of a pyrolysis furnace to which the method of the present invention is applied. In the figure, / is the pyrolysis furnace body, ko is the heating tube,
3 is a quenching heat exchanger, and 3 is a fuel supply pipe. The heating tube is disposed inside the pyrolysis furnace body 1, and is functionally divided into a preheating section 2' and a pyrolysis section 2'. A heating pipe - one end is connected to a feedstock hydrocarbon supply pipe for introducing feedstock hydrocarbons, such as naphtha, and the other end is connected to an outlet pipe 6 for guiding the thermal decomposition products to the quenching heat exchanger 3. ),
Further, a steam supply pipe S for introducing steam is connected to the preheating section. Liquid hydrocarbon raw materials such as naphtha are usually used at a temperature of 100 to 1.30°C, -05 to 4.0°C.
kg/cr/i.

It is supplied from the supply tube at a pressure of G, and while passing through the preheating section 2' of the heating tube, the temperature is usually raised to ttso-bso'c, and the entire amount is vaporized. The thermal decomposition conditions in the thermal decomposition section 2'' include, for example, a decomposition temperature of 6Sθ to 0°C, a steam ratio (steam/raw material hydrocarbon weight ratio) of 0.1I-1
, 0. The cracked gas generated by the thermal decomposition reaction is usually at a temperature of 710 to 3θ°C and a temperature of 0.7 to
At a pressure of 7.2 kg/ai G, the heat from the heating tube comes out, passes through the outlet piping 6, and enters the quenching heat exchanger 3, where it is normally 1. After being cooled to a temperature of yso-soo°C, it is introduced into a separation and purification system via a discharge pipe g.

In the present invention, the pyrolysis furnace has a plurality of pyrolysis furnaces, each of which is controlled under the set value of a selected control factor, for example, a pyrolysis furnace as shown in FIG. When carrying out a conversion reaction in a reactor series consisting of multiple reactors, such as when producing olefins by thermally decomposing hydrocarbon feedstock in the presence of steam in a furnace series, the above load margin function is calculated in advance. Then, when operating the reactor series, the reactor series is controlled by a series of operations (1) to (4) described above at short time intervals (FIG. 1 schematically shows the control method of the present invention). The details will be explained below using thermal decomposition of hydrocarbons as an example.

Setting of Load Margin Function Conventionally, in the pyrolysis of hydrocarbons, operators have to determine the operating status of each pyrolysis furnace by determining (a) the pyrolysis furnace outlet temperature, (b) the feedstock hydrocarbon supply amount, and (C) the steam supply. (d) quenching heat exchanger outlet temperature, (e) convection section temperature, (f) decomposition tube surface temperature, etc., and if any of these measured values approaches the predetermined upper or lower limits. If so, the furnace would have been operated in such a way as to not increase or reduce the load.

In the method of the present invention, the operational status of each reactor, which had been evaluated based on the operator's subjective judgment, is changed to the above-mentioned methods (a) to (f) that control the operation of each reactor. The control factors are used as variables and are understood by converting them into functions as the degree of approach to a preset control limit. The above control limits are determined as the limits of the range of operating conditions that allow safe and rational operation based on the physical constraints of each reactor and operating experience. However, in some cases, it may be set for a combination of more than k control factors (for example, a sum, a product, etc.). The above function, which expresses the operating state of each reactor as the degree of approach to such a control limit, can therefore be seen as expressing how much further load can be applied to each reactor. can.

Therefore, this function will be called the "load margin function."

The shape of the load margin function is not particularly limited as long as it appropriately reflects the degree to which the operating state of each reactor approaches the control limit.

Typically, the deviation (
or its function) as a variable, the purpose can be sufficiently achieved. As this latter function, for example, the product or sum of the deviations (or their functions), or a mixed form thereof can be used. The specific embodiment of the control method according to the method of the present invention is subject to modification depending on thermal theory and the form of the load margin function used.

As mentioned above, the load margin function can be set using various methods. An example of thermal decomposition of hydrocarbons using a fuzzy model (a model based on fuzzy theory in mathematics) will be described.

Examples of control factors used as variables to control the operation of the pyrolysis furnace include the following.

Control factor Symbol Pyrolysis furnace outlet temperature x1 Feedstock hydrocarbon feed rate or pulp opening for feeding it x2 Steam feed rate or pulp opening for feeding quenching heat exchanger outlet temperature x4 Convection section temperature
x5 cracking tube surface temperature
x6 driving days,
i'7 The above-mentioned -1 to x6 control factors are, for example, in FIG. The supply valve opening measuring device/1, the quenching heat exchanger outlet temperature measuring device 12, the convection part temperature measuring device 1
3 and cracking tube surface temperature measuring device/lI.

A function is defined that reflects the degree to which each of the above variables approaches a preset control limit. Each of the obtained functions is called a "membership function".

The values evaluated by each membership function are combined as follows to obtain a load margin function.

Load margin = fC, 9x (xt), g2 (x2), gs
("3), 94 ("4), gs(xs), gs(r
6) 1,97(x7)] In order to improve the accuracy of the load margin function, it is preferable to use more variables (control factors) and therefore a membership function, but it is also possible to simplify measurement and computer control. The number of membership functions can also be reduced based on comparison with control accuracy. From this point of view, for example, jcl, x2, and x are control factors.
s can be used. In that case, it will be as follows.

Generally, the shape of the load margin function f is influenced by the direction in which the control factor is operated, so it is preferable to set it for each direction.

Although the selection of control factors used to change the operating conditions is not particularly limited, the necessary changes in the operating conditions can usually be made using two control factors: the pyrolysis furnace outlet temperature and the feedstock hydrocarbon feed rate.

In this case, the operating directions of these control factors are as follows (1) to (
It is classified as pattern IV).

(1) Pyrolysis furnace outlet temperature: Increase feedstock hydrocarbon supply amount: Increase (n) Pyrolysis furnace exit temperature: Decrease feedstock hydrocarbon supply amount: Decrease (If) Pyrolysis furnace exit temperature: Increase feedstock hydrocarbon supply amount 2 Reduction (■) Pyrolysis furnace outlet temperature: Lower feedstock hydrocarbon feed rate: Increase In general, as the temperature of the pyrolysis furnace increases, olefins with shorter carbon chains are more likely to be produced, and as the feedstock hydrocarbon feed rate increases, olefins Production volume will increase overall.

Therefore, there is a request to change the production amount of olefin in the pyrolysis furnace series, i.e., for example, to change (increase or decrease) the olefin production amount itself, or to change the composition of the produced olefin (for example, to increase the production ratio of ethylene, etc.). ), the operating direction of the pyrolysis furnace outlet temperature and raw material hydrocarbon supply amount is selected from the above-mentioned buttons (1) to (IV), so the corresponding load margin function and do.

The operating direction of the control factor for each pyrolysis furnace may be selected depending on the situation of each pyrolysis furnace, and there is no need for the entire series to have the same pattern.

However, since it is often easier and more rational for the operation of a pyrolysis furnace to distribute changes in the production amount of olefin for the entire pyrolysis furnace to more pyrolysis furnaces, it is usually done at once. It is preferable that the entire series has the same operating direction pattern.

Below, specific examples of setting numbers for each of the patterns (1) to (IV) above will be explained in detail.

Pattern (1) For the variables xI-x6, the limit value that allows safe and rational operation determined from the physical constraints of the cracking furnace and operating experience, that is, the upper limit value, is H/(i), and these variables The point at which it is undesirable to increase the load beyond the normal operating limit (for example, O, rH/
(set from a range of (t) or more and less than H/(1)) is set as H,+(1), then the membership function is defined as, for example, as follows.

t1=i (xt<82(1)) yl(::-[o,/], (1-/~A) The larger 11 is, the more margin there is.

Regarding the number of driving days x7, since the effect of the number of driving days on the driving state is different from other variables, for example, if L2(1) ＞r-Initial stage of operation]
? ,! =1, 0 (1,≦”?<D) ?, =7-α (D≦X,≦D) [End of operation] α [[θ, o, o r ), D: Like the upper limit of the number of operating days can be defined as

For example, the composition of the above membership functions is defined as follows.

f(21,22,'3,'4,?3.26,t?)=
r, −f2− r, −f, −MIN (r4, r
, , fa) (MIN(??6.26) indicates the minimum value among 2 and 26.) Pattern (II) Reasonable operation of the cracking furnace is possible for variables X and x2 Let the lower limit value, that is, the lower limit value for operation, be L / (L), and the point where it is undesirable to lower it any further because it has approached the lower limit value is defined as the normal operation limit value, 21 = / (, xl) L2 (i)) yle Co, /), (t=/ ~J) is defined.

There is no particular lower limit set for X, ~X,
If each variable is close to its upper limit, the load will be reduced more greatly. For example, y1=t (xl(H
2(1)) fIC[1, ko], (1-3 to 6) Define.

Regarding the number of days of operation

for example, ? 7 = / −α (θ ≦ X de 〈
lower iD)? ,=1,0 (75D<,rr<109) 27=/+α (100D≦x,'=o)αC[θ,
O, OS ], D: Upper limit of number of operating days.

For example, the composition of the above membership functions is defined as follows.

t (r, %72% too, ?4, ?3,'6,"I)
=11・2.・1.・27・MAX(P,,?s,f
a) (MAX (r,,t,,?,) is 24,?
, and P6. ) Pattern (1) For variable x1, pattern (1) focusing on the upper limit of operation
As in K, ? 1=/(xl<HJ(i)).

Regarding x2, focusing on the lower limit of operation, similarly to pattern (n), t1=t (,:z:i〉L,z(4))PiC(
0, l], (t=2).

Assume that X, to X are the same as in pattern (1).

The composition of membership functions is also the same as in pattern (■).

f ('I s 92.93.24,?3,'6,'
? ) ””fl・2.・2.・ 2 pieces ・ M
I N(r, , f, , 26) pattern (IV) For variable X, pattern (It
), it is defined as rle(0, l], (i=7) 7l-l (xs) Lco(1)) fLGcO1/], (i=/).

For x2, focus on the upper limit of operation and define it as 11-/(xl(H2(1))?tC[O,/], (1=2), as in turn (1).

Regarding x3 to x7, the same as in pattern (1) is used. The composition of the 0 membership functions is also the same as in pattern (1).

icy+,y□,? 8.24,? 3, r6, r7)”?
,・? 2.2.・? ,・MIN(rF, , ?6)
In the method of the present invention, the reactor series is controlled by a series of operations (1) to (4) below using the load margin function as described above.

That is, as shown in FIG. 1, when operating the reactor series, various steps are taken to approach the target conversion reaction product production amount or to change the target conversion reaction product production amount itself. For various reasons, changes in the amount of conversion reaction products produced are required. The amount of change in the production amount of the conversion reaction product required according to the request (for series load change i, [1] (a) calculate the load margin of each reactor using the load margin function, ■ (b ) Determine the distribution of the series load change amount to each reactor (reactor load change amount) according to the size, and ■ Change the set value of the control factor of each reactor to generate a conversion reaction that corresponds to the reactor load change amount. The operation is repeated by changing the amount of product produced (required amount of conversion reaction product produced) and continuing the operation.

[1] (a) Calculating load margin , that is, calculate the load margin.

Depending on the size of the load margin of each reactor determined as above, the amount of change in the production amount of conversion reaction product required in response to a request for change in the production amount of conversion reaction product of the reactor series, i.e. The series load change amount is distributed to each reactor to obtain the load change amount for each reactor (reactor load change amount). The distribution is basically calculated using the following formula (% formula). However, if the accuracy of the load margin function is insufficient and the operational status of a particular reactor is reflected too much or too little, the above formula can be used based on experience. This can also be corrected by introducing an appropriate coefficient.

■Change in the operating conditions of each reactor Change the set values of the control factors for each reactor to the amount of conversion reaction product produced corresponding to the amount of change in reactor load calculated in
Operation of the reactor series is continued with changes to provide the required amount of conversion reaction product produced.

There are no particular restrictions on how to change the set values of the control factors for each reactor. For example, the relationship between the set value and the production amount of the conversion reaction product can be determined empirically from past operating conditions and changed accordingly.

For example, in the case of pyrolysis of hydrocarbons, as mentioned above, as the temperature of the pyrolysis furnace (X) increases, olefins with shorter carbon chains tend to be produced, and the feedstock hydrocarbon supply amount (x2) Since there is a general tendency that the amount of olefin produced increases as a whole when By tabulating the relationship, the values of the control factors for each pyrolysis furnace can be changed accordingly. It is also possible to convert the relationship between the value of such a control factor and the amount of olefin production into a function in advance based on experimental data, and use this function to calculate the value of the corresponding control factor from the required amount of olefin production. .

According to the method of the present invention, when operating a pyrolysis furnace series,
By importing the control factor data for each pyrolysis furnace into the converter and performing the series of operations from 1 to 3 above, the control can be performed as needed until the operating conditions of each pyrolysis furnace are changed. Operating conditions can be easily, appropriately, and quickly changed in order to apply an appropriate load to each pyrolysis furnace. By performing control according to the method of the present invention at fixed time intervals, for example, every 0, S ~ 72 hours, preferably every / ~ g hours, variations in the deterioration state of each pyrolysis furnace can be reduced and more appropriate load distribution can be achieved. can control the pyrolysis furnace series.

In addition, when controlling a chlorinated hydrocarbon pyrolysis furnace series, the control factors for controlling the operation of the pyrolysis furnace include, for example,
Pyrolysis furnace outlet temperature (x,'), raw material chlorinated hydrocarbon (for example, 1,-1 dichloroethane) supply amount or its supply valve opening (x2'), quenching heat exchanger outlet temperature (x,'),
By selecting the convection section temperature (x,'), the cracking tube surface temperature (-,'), and the number of operating days (x,''), the series is By distributing the load change amount to each pyrolysis furnace according to its load margin, it is possible to control the pyrolysis furnace series with appropriate load distribution. I will explain in detail,
The present invention is not limited to the following examples unless it exceeds the gist thereof.

Example / 12 operated using naphtha as feedstock hydrocarbon
Regarding the pyrolysis furnace series consisting of the pyrolysis furnace of the furnace, according to the method of the present invention, by the above-mentioned turn (1), 2□ to v7
The load margin of each pyrolysis furnace at a certain point in time was calculated using the membership function of , and the results shown in Table 1 were obtained.

The amount of change in olefin production amount (series load change amount) determined as the difference between the actual value of olefin production amount and the preset target value was distributed to each pyrolysis furnace based on the load margin calculated above. The results (ratio of furnace load changes (■/Σ■ in the table)) are shown in Figure 3.

Control factor settings and olefin table for each pyrolysis furnace -/ Change the settings of the control factors for each pyrolysis furnace and operate each pyrolysis furnace to provide the necessary olefin production amount commensurate with the amount of loading changes. Continued. Furthermore, the load margin is calculated every 6 hours using patterns (1) to (IV) according to the relationship with the target value, and the operation is continued by controlling the value of the control factor of each pyrolysis furnace. repeated. As a result, in the past, in this furnace series, the load margin of each pyrolysis furnace was not quantitatively calculated, and the amount of series load change was appropriately distributed to each pyrolysis furnace based on the operator's experience. Compared to the case where the operation continued, the load margin of each pyrolysis furnace at a certain point in time was calculated using pattern (n) in the same manner as in Decokin Example Example 1.
The results shown below were obtained.

The result of allocating the change in olefin production, which was determined as the difference between the actual value of olefin production and the preset target value, to each pyrolysis furnace based on the load margin calculated above as the change in train load. (Ratio of furnace load change amount (θ/Σθ in Table 1)) is shown in Figure 9. As in Actual Example 1, each pyrolysis furnace is operated based on this allocated amount and operation is continued. As a result, the decoking period was able to be extended by is-, to% compared to the conventional example, as in Example/.

Table C [Effects of the Invention] According to the method of the present invention, the operation of the reactors can be adjusted more precisely and appropriately according to the operating status of each reactor, so the cycle of tune-ups such as decoking can be extended. .

In addition, since operators can perform operations that were previously performed empirically in a systematic manner, individual differences in operation are eliminated.
and can be simplified.

Furthermore, since it becomes possible to quantitatively evaluate the margin for each reactor's load change, it becomes possible to control the reactors by a computer.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram schematically showing control of a reactor series by the control method of the present invention. FIG. 2 is a schematic diagram showing an example of a pyrolysis furnace for carrying out the method of the present invention. /: pyrolysis furnace body, 2: heating tube, λ': heating tube preheating section,
KO: heating tube pyrolysis section, 3: quenching heat exchanger, lI: feedstock hydrocarbon supply pipe, 5: steam supply pipe, 6: outlet piping, 7
:Fuel supply pipe. Figures 3 and 9 show the ratio of furnace load changes when the train load changes are distributed to 1 pyrolysis furnaces based on the load margin value in Example/and Example 1, respectively. It is a graph.

Claims (7)

[Claims] (1) When carrying out a conversion reaction in a reactor series consisting of a plurality of reactors, each of which is operated under the set value of a selected control factor, the operating state of each reactor in operation must be checked in advance. , obtaining a function (hereinafter referred to as "load margin function") that expresses the control factors that control the operation of each reactor as variables as the degree of approach to a preset control limit; and, [1] (a) For each reactor, data on control factors is input into a computer and the value of the load margin function (hereinafter referred to as "load margin") is calculated; (b) ) In response to a request to change the production amount of conversion reaction products in a reactor series, the amount of change in the production amount of conversion reaction products (hereinafter referred to as "series load change amount") is calculated according to the size of the load margin. Allocate to the reactors and obtain the load change amount for each reactor (hereinafter referred to as "reactor load change amount"), and [2] set the control factor of each reactor to obtain the reactor load change amount. 1. A method for controlling a reactor, comprising: changing the production amount of a conversion reaction product corresponding to (hereinafter referred to as "required production amount of a conversion reaction product"); and repeating the steps. (2) The method for controlling a reactor according to claim 1, wherein the conversion reaction is a thermal decomposition reaction. (3) The method for controlling a reactor according to claim 2, wherein the thermal decomposition reaction is a hydrocarbon thermal decomposition reaction. (4) The method for controlling a reactor according to claim 3, wherein the hydrocarbon includes a liquid hydrocarbon selected from the group consisting of naphtha, gas oil, kerosene, light oil, and heavy oil. . (5) The method for controlling a reactor according to claim 3, wherein the hydrocarbon includes a gaseous hydrocarbon selected from the group consisting of ethane, propane, and butane. (6) The method for controlling a reactor according to claim 2, wherein the thermal decomposition reaction is a thermal decomposition reaction of chlorinated hydrocarbons. (7) The method for controlling a reactor according to claim 6, wherein the chlorinated hydrocarbon contains 1,2-dichloroethane.