JP2001226680A - Method of automatically controlling drawing amount of cut in topper and automatic controller for drawing amount - Google Patents

Abstract

PROBLEM TO BE SOLVED: To carry out automatic controlling of the drawing amounts of cuts so as to properly keep properties of the each cut when a factor causing a variation in the properties of the cuts is added. SOLUTION: This method of automatically controlling the drawing amounts of the cuts in a topper is for automatically controlling the drawing amount of the each cut so as to satisfy the properties of the each cut when the factor causing the variation in the properties of the cuts is added in the topper fractionating crude petroleum into various kinds of cuts containing at least naphtha, kerosene, a light cut and a heavy cut, and comprises making a model of the drawing amounts in which the ideal control value of the drawing amounts is modeled corresponding to the variation in the properties of the naphtha, the kerosene and the light cut, determining a target control value of the drawing amounts in the each cut by a fuzzy inference from the difference between the modeled drawing amounts in the model of the drawing amounts and actual extracting amounts and correcting the target control value of the drawing amount of the light cut by a fuzzy rule based on the drawing amount of the heavy cut.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a crude oil from naphtha,
In a topper for fractionating kerosene, light gas oil, heavy gas oil, and the like, when a factor that changes the properties of each fraction such as switching of the type of the crude oil is added, the extraction amount for each fraction ( The present invention relates to a method and an apparatus for automatically controlling a withdrawal amount at the time of switching crude oil in a topper, which can automatically control a withdrawal amount while satisfying the properties of each fraction.

[0002]

2. Description of the Related Art Crude oil is obtained from LP gas (hereinafter, referred to as LPG), naphtha (hereinafter, referred to as FRN), kerosene (hereinafter, referred to as KERO), light oil by a normal pressure distillation apparatus called a topper. It is fractionated into fractions such as light oil (hereinafter referred to as LGO) and heavy light oil (hereinafter referred to as HGO). FIG. 13 is a schematic diagram showing the configuration of the topper. The topper comprises a distillation column 1 for distilling the crude oil supplied from the crude oil tank A or the crude oil tank B, and a stripper 2 for purging light gas oil from the mixed solution obtained by fractional distillation in the distillation column 1.
And a crude oil heating device (not shown) and a desalination tank. Then, LPG is fractionated in the distillation column 1, and FRN, KERO, LGO, and HGO are fractionated in the stripper 2.

[0003] Since the amount of each fraction extracted at the cut point (distillation temperature) of the crude oil differs for each type, the distillation column 1 is switched by switching the tank switching valve 3 or the tank switching valve 4. When the type of crude oil to be supplied (crude oil type) is switched, it is necessary to control the amount of each fraction extracted from the start of switching to the completion of switching in accordance with the supply ratio of crude oil supplied from tanks A and B. is there.

[0004] Here, the FRN is an endpoint (E
P), KERO is 95% distillation temperature, LGO is 90%
At the distillation temperature, HGO controls the extraction amount with the aim of color properties (for example, see JIS2254). When the type of crude oil is switched, the amount of extraction is controlled to the amount of extraction corresponding to the type of crude oil in order to keep the properties of FRN constant, but this affects the amount of extraction of KERO. In addition, if the extraction amount is controlled to the extraction amount corresponding to the type of crude oil in order to keep the properties of KERO constant, the extraction amount of LGO will be affected by this.

[0005] As described above, if an attempt is made to control the amount of extraction of one fraction to a constant value in accordance with the type of crude oil in order to keep the properties of one fraction constant, the effect spreads to the amount of extraction of the other fraction. Therefore, it is necessary to control the withdrawal amount for each fraction in accordance with the type of crude oil switched in consideration of these effects. Furthermore, FRN, KERO, LG
If the extraction amount is controlled while performing the property control in the order of O, the fluctuation in the extraction amount of these fractions will eventually be HGO
However, in order to keep the properties of HGO constant, while considering the relationship with the amount of LGO extracted,
The amount of LGO and HGO withdrawn must be delicately controlled. Such a task is difficult with conventional proportional differential and integral (PID) control, and at present it relies on the intuition and experience of a skilled person.

[0006] However, manual work is inefficient. In addition, it takes time and money to train skilled workers, and there is a problem that the properties of the fractions are apt to vary from worker to worker and the quality varies.

[0007]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and is intended to change the properties of each fraction fractionated by a topper, such as switching of crude oil types. If any factors are added, by automatically controlling the extraction amount to maintain the properties of each fraction properly, work efficiency, further improvement of safety, cost reduction and quality stabilization are achieved. It is an object of the present invention to provide a method and an apparatus for automatically controlling a withdrawal amount of a fraction in a topper.

[0008]

In order to achieve the above-mentioned object, the present invention relates to a skilled worker in a case where a factor which changes the properties of each fraction is added, for example, switching of crude oil type. Is expressed by a rule using a membership function, and the extraction amount is automatically controlled by fuzzy inference based on this rule. In particular, when controlling the LGO extraction amount, the HGO extraction amount is monitored, and the control target value of the LGO extraction amount is corrected by fuzzy inference based on the HGO extraction amount.

More specifically, in the topper for fractionating crude oil into various fractions including at least naphtha, kerosene, light gas oil and heavy gas oil, as in the first aspect of the present invention, A method for automatically controlling a withdrawal amount of a fraction in a topper, which automatically controls a withdrawal amount for each of the fractions so as to satisfy properties of each of the fractions when a factor giving a change is added. In accordance with the change in the properties of the fraction, a withdrawal amount model is created by modeling the ideal control value of the withdrawal amount, and the difference between the model withdrawal amount of the withdrawal amount model and the actual withdrawal amount is calculated by fuzzy inference. A control target value of a withdrawal amount is determined for each of the fractions, and a control target value of a withdrawal amount of a predetermined fraction of the fractions is set in the topper from below the predetermined fraction. There As a method of correcting by which other fraction withdrawal amount based fuzzy rules.

Further, as in the second aspect of the present invention, a method may be adopted in which the control of the extraction amount of the predetermined fraction is performed by a fuzzy rule based on a change in the property of the other fraction. Preferably, the predetermined fraction is light gas oil and the other fraction is heavy gas oil. These methods, as described in claim 4 or claim 5,
This may be performed when changing the type of crude oil or when changing the amount of withdrawal of a certain fraction. In addition, as described in claim 6, it is preferable that safety control restrictions are added to the control target value.

An automatic apparatus for controlling the amount of withdrawal of a fraction in a topper according to the present invention for performing the above method is provided in claim 7.
As described in the above, in a topper for fractionating crude oil into various fractions including at least naphtha, kerosene, light gas oil and heavy gas oil, if a factor that changes the properties of the fraction is added, the fraction A withdrawal amount automatic control device for a topper in a topper that automatically controls the withdrawal amount of each fraction so as to satisfy the properties of each of the fractions, wherein an ideal amount of withdrawal corresponding to a change in the properties of the fractions A control target setting unit that creates a extraction amount model in which the control value is modeled, and determines a control target value for each of the fractions by fuzzy inference from a difference between a model extraction amount of the extraction amount model and an actual extraction amount. And, based on the control target value determined by the control target setting unit, an individual control unit that controls the extraction amount of each fraction, and detects the actual extraction amount of each fraction. A withdrawal amount detection unit to be transmitted to the control target setting unit, and a control target value of a withdrawal amount of a predetermined fraction among the fractions, wherein the topper distills from below the predetermined fraction in the topper. And a correction unit that corrects with a fuzzy rule based on the amount of the extracted fraction.

[0012] As described in claim 8, the predetermined fraction may be light diesel and the other fraction may be heavy diesel. Also, in order to further secure the safety of the system, a safety countermeasure unit is provided which outputs the control target value to the individual control unit in consideration of a control constraint on the safety of the system, as described in claim 9. Good.

According to a tenth aspect of the present invention, there is provided a start / stop instructing unit for instructing start / stop of the control target setting unit corresponding to at least the naphtha, kerosene and light gas oil, and the extraction amount is arbitrarily determined. May be configured to be able to perform the control. For example, when switching the type of crude oil as described in claim 11, the start / stop instruction unit may be operated to start / stop the control target setting unit.

According to the method and apparatus of the present invention, even when a change that affects the properties of a fraction is made, such as a change in the type of crude oil, the change from the start to the completion of the change. Can be automatically controlled by fuzzy inference. In particular, by controlling the extraction amount of light gas oil in accordance with the extraction amount fluctuation of heavy gas oil that absorbs the fluctuation amount of naphtha, kerosene and light gas oil, the optimum extraction amount control can be performed automatically.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a method and an apparatus for automatically controlling the amount of withdrawal of a fraction according to the present invention will be described below in detail with reference to the drawings. In this embodiment, the fraction fractionated from crude oil includes FRN, KERO, LGO
And HGO are included. FIG. 1 is a block diagram for explaining a system configuration of the automatic apparatus for controlling the amount of withdrawn fraction according to the present invention.

[Explanation of Overall Configuration of Apparatus] The automatic withdrawal amount control device includes control units 20, 30, and 40 for controlling the respective withdrawal amounts of FRN, KERO and LGO, and HGO.
Control unit 50 for controlling the amount of unloading, system start / stop instructing unit 10 for starting / stopping control units 20, 30, 40 by operator's operation or automatically, and control units 20, 30, 40. , 50 by the command of the flow path 26, 3
Valves 24, 34, 44 for opening and closing 6, 46, 56,
54, and extraction amount detectors 25, 35, 45, and 55 for detecting the extraction amounts of the respective fractions (FRN, KERO, LGO, and HGO) flowing through the flow paths 26, 36, 46, and 56.

The control units 20, 30, 40, and 50 include control target setting units 21, 31, 41, and 51 for setting a control target value of a withdrawal amount for each fraction, and control target setting units 2 and 31, respectively.
In addition to the control target values set by 1, 31, 41, and 51, taking into account the hardware constraints (control constraints) of the system,
Safety countermeasure unit 2 that determines the extraction amount to be actually controlled
2, 32, 42, 52, and the safety measures 22, 22,
On the basis of the output results of the valves 42, 52, the valves 24, 34, 4
PID (Proportional Differential / Integral) control units 23, 33, 43, 53 as individual control units for opening and closing the 4, 54.

The actual control of the withdrawal amount is performed by opening and closing the valves 24, 34, 44, 54 in accordance with commands output from the PID control units 23, 33, 43, 53. The extraction amount of each fraction by opening and closing the valves 24, 34, 44, 54 is determined by an extraction amount detector 25, 35, 45, 5.
5, and the detection result is transmitted to the control target setting units 21, 31, 41, 51. In the control target setting units 21, 31, 41, 51, the extraction amount detector 25,
The control is performed based on the difference between the actual extraction amount detected by 35, 45, and 55 and the model extraction amount determined according to the supply ratio of the type of crude oil supplied from tanks A and B (see FIG. 13). Whether the target value should be increased, decreased, or kept as it is is determined by fuzzy inference based on fuzzy rules described later.

[Description of Configuration of Control Target Setting Unit] FIG. 2
2A shows the configuration of the control target setting unit of KERO, and FIG.
FIG. 2B shows an extraction amount model representing an ideal change rate of the extraction amount when the crude oil type is switched, and FIG. 2C shows the configuration of the LGO control target setting unit. In FIG. 2A, KERO
Although only the control target setting unit 31 of FIG. 1 is illustrated, the control target setting unit 21 of the extraction amount of KERO is also the same, so that the illustration thereof is omitted. In FIG. 2B, the upper graph shows a change graph of the supply amount of the crude oil from the tank A and the tank B at the time of switching the crude oil, and the lower graph shows an ideal control value of the extraction amount based on the supply amount in the upper graph. 3 shows a model of an extracted amount.

A control target setting unit 31 for controlling the extraction amount of KERO includes an extraction amount model creation unit 31a for creating an extraction amount model as shown in FIG.
And a fuzzy inference unit 31b for obtaining a control target value of the KERO extraction amount by fuzzy theory so as to follow the extraction amount model. In addition, the control target setting unit 41 for controlling the LGO extraction amount is configured as shown in FIG.
As shown in (c), the extraction amount model creation unit 4 similar to the extraction amount model creation unit 31a of the control target setting unit 31
1a and a fuzzy inference unit 41b similar to the fuzzy inference unit 31b, and a correction unit 41c for correcting the control target value obtained by the fuzzy inference unit 41b by fuzzy inference based on the extraction amount of HGO. .

The configuration of the control target setting section 51 (see FIG. 1) for controlling the amount of HGO extraction is not specifically shown, but is basically the same as that of the KERO control target setting section 31. However, the extraction amount of HGO is FRN, KERO
In addition, since the variation of the extraction amount of LGO is absorbed and fluctuated constantly, it is difficult to create the extraction amount model. Therefore, no extraction amount model creating unit is provided. That is, the extraction amount control of the HGO is performed only by fuzzy inference described later.

[Explanation of Extraction Amount Model] The extraction amount model is determined according to the supply amount of crude oil from the tank A and the supply amount of crude oil from the tank B. For example, FIG.
(B) As shown in the upper graph, between the start of the switching of the crude oil type and the completion of the switching, the supply amount of the crude oil from the tank A decreases linearly, and the supply amount of the crude oil from the tank B decreases linearly. In the case where the amount of KERO in the distillation column 3 starts to change and stabilizes, the KERO corresponding to the crude oil type in the tank B is determined based on the amount of KERO extracted from the crude oil type in the tank A. Extraction amount (final extraction amount)
Until then, an extraction amount model that linearly changes as shown in the lower part of FIG. 2B is created by the extraction amount model creation unit 31a. The final withdrawal amount is calculated, for example, from the oil type composition of the crude oil in the tank B using the fraction yield table for each crude oil using FRN to LG.
The amount of distillate up to O can be determined by a weighted average of the composition ratio of the oil type. In addition, the extraction amount model can be modified so as to reduce the extraction amount of KERO according to the amount when the extraction amount of FRN is to be increased, for example, in relation to product demand.

[Description of System Start / Stop Instruction Unit] FR
The control units 20, 30, and 40 for controlling the extraction amounts of N, KERO, and LGO are connected to the system start / stop instruction unit 10 as shown in FIG. The system start / stop instructing unit 10 is operated by an operator or automatically when a factor that changes the properties of the fraction is added, such as switching of the type of crude oil. 3
An instruction to start 0, 40 is output, and if the properties of each fraction are stabilized by controlling the extraction amount, an instruction to stop the control units 20, 30, 40 is output by an operator's operation or automatically. By this stop instruction, the control units 20, 3
When 0 and 40 stop, the valves 24, 34 and 44 for controlling the withdrawal amount are maintained at a constant opening.

In this embodiment, the system start / stop instructing unit 10 includes a control unit 5 for controlling the amount of HGO extraction.
0 is not connected. This means that fluctuations in the withdrawal amounts of FRN, KERO and LGO during normal operation are reduced by H
This is because the control unit 50 needs to be constantly operated irrespective of the instruction of the system start / stop instruction unit 10 because it needs to be absorbed by GO.

The detection result of the extraction amount detector 55 for detecting the extraction amount of HGO is also inputted to the control target setting section 41 of LGO. That is, L
The control target setting unit 41 for controlling the GO extraction amount is H
The LGO control target value is corrected by fuzzy inference in consideration of the GO extraction amount. This makes it possible to automatically control the amount of extraction so as to keep the properties of LGO and HGO constant.

[Description of Control of FRN and KERO Extraction Amounts] Next, control of the FRN and KERO extraction amounts will be described with reference to FIGS. 3 and 4. FIG. FIG. 3 (a)
Is a fuzzy rule table for controlling the FRN extraction amount, and FIG. 3B is a membership function related to the rule table of FIG.

The FRN controls the extraction amount with the end point (EP) as a property target. Therefore, as shown in FIG. 3, whether the EP of the FRN is appropriate for the property target or high is determined based on whether the EP change of the FRN is increasing, not changing, or decreasing. E
The opening and closing of the valve 24 is controlled so as to decrease or increase the P change.

For example, even if the change in the EP of the FRN increases, if the EP is appropriate (OK) with respect to the property target, the opening degree of the valve 24 is maintained and the control is performed so that the withdrawal amount remains unchanged. . Then, the subsequent change is monitored, and when a high EP is shown with respect to the target value, the valve 24 is closed so as to greatly reduce the EP, and the extraction amount is reduced.

FIG. 4A shows a fuzzy rule table for controlling the amount of KERO extraction, and FIG. 4B shows a membership function for the rule table of FIG. KERO controls the extraction amount with the 95% distillation temperature as a property target. Therefore, as shown in FIG.
Whether the 95% distillation temperature change of RO is rising, not changing, or falling,
The opening and closing of the valve 34 is controlled so that the change in the 95% distillation temperature of KERO is reduced or increased depending on whether or not the% distillation temperature is appropriate for the property target.

For example, even if the change in the 95% distillation temperature of KERO increases, if the 95% distillation temperature is appropriate (OK) with respect to the property target, the opening degree of the valve 34 is maintained and the amount of extraction is maintained. Is controlled to remain as it is. Then, the subsequent change is monitored, and when a high 95% distilling temperature is shown with respect to the target value, the valve 34 is set so as to greatly reduce this.
Close to reduce the withdrawal amount.

In the flowchart of FIG.
And a control procedure of the extraction amount of KERO. As initial settings, the composition ratio of the crude oil to be switched, the final withdrawal amount of each fraction, the switching start time and end time of the crude oil tank, the leaving time after the withdrawal amount is stabilized, and the like are set in advance by the control unit 2.
0 and 30 are input. The operator operates the system start / stop instructing unit 10 before switching the type of crude oil,
A start instruction is output to the control units 20 and 30 (step S1). Thereafter, switching of the crude oil type is started (step S2).

The control units 20 and 30 calculate the time when the properties of the fraction in the distillation column 1 start to change from the crude oil processing amount automatically input from an analyzer (not shown) and the amount of oil retained in the pipe. (Step S4). Then, after this time has elapsed, control of the extraction amount is started (step S5).
In addition, the extraction model creation units 21a and 31a of the control target value setting units 21 and 31 create an extraction amount model according to the pre-input crude oil type after switching (step S3). This extraction amount model can be created so as to increase (or decrease) the required extraction amount of the FRN fraction, for example, in relation to product demand.

The control target value setting units 21 and 31 calculate the fuzzy rule tables and the membership function from FIGS. 3 and 4 based on the current extraction amount detected by the extraction amount detection units 25 and 35 and the extraction amount model created earlier. Fuzzy inference is performed on the basis of (steps S6 and S7), and a control target value is set (step S8). Safety measures 22, 32
Calculates the difference between the control target value and the actual extraction amount (step S9). If it is determined that the difference exceeds the allowable range of the control constraint (step S10), the upper limit of the control constraint is set to the control target. Set as a value (step S1
1). If it is determined that the difference is within the allowable range of the control constraint (step S10), the control target value obtained by fuzzy inference is set as it is as a control target value (step S12). When the required extraction amount of the FRN is changed (step S13), based on the changed amount,
Modify the extraction amount model of KERO (Step 1
4).

The PID control units 23 and 33 operate the valves 24 and 24 in accordance with the control target values determined as described above.
34 is controlled to open and close. If the extraction amount is stabilized by repeating the above procedure, a stop instruction is output from the system start / stop instruction unit 10 after a certain period of time,
The process ends (step S15). If the extraction amount is not stable, the processing of steps S6 to S14 is repeated.

[Explanation of LGO extraction amount control]
The control of the GO extraction amount will be described with reference to FIGS. Since the LGO controls the extraction amount with the 90% distillation temperature as a property target, the LGO 90% distillation temperature changes depending on whether the 90% distillation temperature change is increasing, not changing, or falling. 90% of LGO when% distillation temperature is appropriate for property target and high
Valve 4 to reduce or increase the distillation temperature change
4 is controlled to open and close. In controlling the amount of LGO extraction, a fuzzy control rule table and membership function related to the HGO extraction amount shown in FIG. 7 are used in addition to the fuzzy control rule table and membership function of FIG.

The fuzzy control rule table and the membership function shown in FIG. 7 indicate whether the HGO overflash amount is rising, not changing, or falling, and is appropriate for the HGO extraction amount. The opening and closing of the valve 44 is controlled so as to reduce or increase the amount of HGO overflash between the case and the high case.
That is, the opening / closing amount of the valve 44 is obtained by correcting the control target value obtained from the fuzzy control rule table and the membership function in FIG. 6 with the correction value obtained from the fuzzy control rule table and the membership function in FIG. Is determined by Here, the amount of HGO overflash refers to the amount of vaporized HGO existing above the HGO extraction port being cooled and falling as a liquid, and is monitored by a detector (not shown). Give information.

For example, in the case where the 90% distillation temperature change of LGO is unchanged, if the 90% distillation temperature of LGO is appropriate (OK), the opening degree of the valve 44 is maintained, and the extraction amount is reduced. Usually, a control target value is set so as to remain as it is. However, when the HGO extraction amount is small and the overflash amount is small, the HGO extraction amount thereafter decreases, and the HGO extraction amount may be smaller than the minimum value of the HGO extraction amount calculated due to device restrictions or the like. Therefore, based on the LGO control rule table of FIG. 7, the valve 44 is closed to greatly reduce the amount of LGO withdrawn.

In the flowchart of FIG.
2 shows a control procedure of the extraction amount. Note that the same steps as those in the flowchart of FIG. 5 are denoted by the same reference numerals, and detailed description of the steps will be omitted. When controlling the LGO extraction amount, the HGO extraction amount is detected (step S21), and a correction value is determined by fuzzy inference based on the fuzzy control rule table and the membership function of FIG. 7 (step S22) (step S22). S2
3). Then, using this correction value, the control target value obtained by fuzzy inference based on the fuzzy control rule table and the membership function in FIG. 6 is corrected (step S24). The safety measure unit 42 calculates a difference between the corrected control target value and the actual extraction amount (step S25),
It is determined whether this difference is within the range of the control constraint (step S8). As described in the flowchart of FIG. 5, when the required extraction amount of KERO is intentionally changed (step S13), the LGO is extracted based on the change amount.
Is corrected (step 14).

[Explanation of Control of HGO Extraction Amount]
The control of the extraction amount of HGO will be described with reference to FIG. The extraction amount of the HGO is controlled with the target of the color property and the overflash amount. For this reason, the fuzzy control rule table of the HGO contains H
From the degree of increase or decrease of the color change of the GO and the amount of overflash, the degree of increase or decrease of the color change of the HGO and the amount of overflash are determined when the color of the HGO is low, appropriate, and high with respect to the target. The opening and closing of the valve 54 is controlled so as to be lowered or raised.

For example, when the color property change of the HGO is increasing and the color of the HGO is higher than the property target, the control target value is set so as to greatly reduce the color change of the HGO. If the overflash amount is large in such a case, the control target value is corrected so that the overflash amount is slightly increased.

As described above, the extraction amount of HGO has a large variation to absorb the variation in the extraction amount of FRN, KERO, and LGO, and it is not possible to create an extraction amount model. Therefore, the control of the extraction amount of HGO is performed only by fuzzy inference for the color change and the overflash amount. In the flowchart of FIG.
The control procedure of the GO extraction amount will be described. The control target value setting unit 51 performs fuzzy inference from the current extraction amount detected by the extraction amount detection unit 55 based on the fuzzy rule table and the membership function in FIG. 9 (step S).
6, S7), a control target value is set (step S8).

The safety countermeasure unit 52 obtains a difference between the control target value and the actual extraction amount (step S9). When it is determined that the difference exceeds the allowable range of the control constraint (step S10), The upper limit of the control constraint is set as a control target value (step S11). If it is determined that the difference is within the allowable range of the control constraint (step S10), the control target value obtained by fuzzy inference is set as it is as a control target value (step S12). The process shown in the flowchart of FIG. 10 is always executed irrespective of the presence or absence of the switching of the crude oil type, and ends when a stop command is output due to a stop of the operation of the system or the like (step S16).

[Second Embodiment] Next, a second embodiment of the present invention will be described with reference to FIG. FIG.
FIG. 2 is a system configuration diagram according to a second embodiment of a withdrawal amount automatic control device of the present invention. In the embodiment shown in FIG. 11, in addition to the control units 20, 30, and 40, the control unit 50 is also connected to the system activation / instruction unit 10 ', and
A system for starting and stopping the control unit 50 is configured by the instruction unit 10 '. In this case, according to the situation, only the control unit 50 does not instruct the stop, and the control units 20, 30, 4
It is preferable that zero stop control can be performed.

[Third Embodiment] In a third embodiment of the present invention, a system configuration similar to the system configuration shown in FIG.
Control is performed in the same manner as O and LGO. The control of the extraction amount of HGO in this case is the same as the control of the LGO of the embodiment described first. That is, FRN, K
Fluctuations in the extraction amounts of ERO, LGO, and HGO are absorbed by residual oil (see FIG. 13) that distills from below, for example, the control of the extraction amount of HGO requires the use of the residual oil. Is performed by fuzzy inference while monitoring the properties of

Fourth Embodiment In the first to third embodiments, switching of the type of crude oil has been described as an example of a factor that changes the properties of each fraction. It is also applicable under other circumstances where the properties of each fraction may change. For example, the present invention can be applied to a case where the extraction amount of FRN is changed in relation to product demand. In this case, FIG.
As shown in the flowchart of FIG.
The extraction amount of N may be changed, and an extraction amount model for changing the extraction amount may be created in step S3. In the KERO flowchart,
It is determined in step S13 whether or not the FRN extraction amount has been changed, and if there has been a change, the process proceeds to step S14.
Then, the extraction amount model of KERO is corrected. Other steps are the same as the steps denoted by the same reference numerals in the flowcharts of FIGS. 5 and 8.

As described above, in the present invention, the control target value of the extraction amount of a predetermined fraction among the fractions fractionated by the topper is determined based on the extraction amount of the other fraction below this fraction. By correcting with the fuzzy rules, when factors that change the properties of each fraction act,
The stable properties of each fraction can be obtained automatically and quickly.

Although the preferred embodiments of the present invention have been described, the present invention is not limited to the above embodiments. For example, the safety countermeasure process executed in consideration of the control constraints of the system includes the safety countermeasures 22, 32, and 4.
Although the description has been made on the assumption that the control is performed by the control target setting units 21, 32, 42, and 52, the same processing can be performed by the control target setting units 21, 31, 41, and 51.

Further, the above-described safety countermeasure processing is performed by the control target setting units 21, 31, 41, 51 and the safety countermeasures 22, 3.
The safety can be further improved by performing the operations in both of 2, 42 and 52.

Further, the system start / stop instruction unit 10,
Although 10 'has been described as being manually operated by the operator at the time of switching between the types of crude oil, the control units 20, 30, 40, and 40 are automatically operated based on the switching start time or the completion time of switching the type of crude oil which is input in advance. It is also possible to configure so that 50 is started or stopped.

The system start / stop instructing unit 10
Not only at the time of switching the type of crude oil, but also when an abnormality of the system equipment is detected, when a predetermined time (for example, 4 hours) has elapsed from the start of the system, when the extraction amount is stabilized, the hold time of the fraction after the extraction amount is stabilized is fixed time (for example, 30 minutes) May be configured to stop the control units 20, 30, and 40, for example, when elapses. Further, in the flow charts shown in FIGS. 5 and 8, if the amount of extraction of a fraction such as FRN is not intentionally changed, steps S13 and S14 need not be particularly provided. .

[0051]

As described above, according to the present invention, the amount of each fraction to be withdrawn can be automatically controlled, so that the operator's hand is not required, and In addition to enabling efficient and safer control of the amount of each fraction extracted when factors that change the properties of each fraction, such as switching, act, the properties of each fraction And the quality can be improved.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of a fraction extraction amount automatic control device of the present invention.

FIG. 2 is an explanatory diagram of a configuration of a control target setting unit in FIG. 1;
(A) is a block diagram illustrating the configuration of a control target setting unit that controls the amount of KERO extraction, and (b) is a diagram illustrating an example of an extraction amount model. Of the change in the supply amount of crude oil, the lower part is an extraction amount model that models the ideal control value of the extraction amount based on the upper supply amount, and (c) is the LGO
FIG. 4 is a block diagram illustrating a configuration of a control target setting unit that controls a withdrawal amount.

FIG. 3A is a fuzzy rule table for fuzzy control in FRN, and FIG. 3B is a membership function.

4A is a fuzzy rule table for fuzzy control in KERO, and FIG. 4B is a membership function.

FIG. 5 is a flowchart showing a control procedure of a withdrawal amount in FRN and KERO.

FIG. 6A is a fuzzy rule table for fuzzy control in LGO, and FIG. 6B is a membership function.

FIG. 7A is a graph showing LG related to the amount of HGO extracted.
A fuzzy control rule table for fuzzy control of O, where (b) is a membership function.

FIG. 8 is a flowchart showing a control procedure of a withdrawal amount in LGO.

9A is a fuzzy rule table for fuzzy control in HGO, and FIG. 9B is a membership function.

FIG. 10 is a flowchart illustrating a control procedure of a withdrawal amount in the HGO.

FIG. 11 is a system configuration diagram according to a second embodiment of the fraction extraction amount automatic control device of the present invention.

FIG. 12 is a flowchart for explaining a procedure of controlling the extraction amount in FRN and KERO according to the fourth embodiment of the present invention.

FIG. 13: From crude oil to naphtha (FRN), kerosene (KER)
FIG. 2 is a schematic diagram illustrating a configuration of a topper for fractionating O), light gas oil (LGO), heavy gas oil (HGO), and the like.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Distillation tower 2 Stripper 3,4 Tank switching valve 10,10 'System start / stop instruction part 20,30,40,50 Control part 21,31,41,51 Control target setting part 21a, 31a, 41a Extraction amount model Generation units 21b, 31b, 41b, 51b Fuzzy inference unit 41c Correction units 22, 32, 42, 52 Safety measures units 23, 33, 43, 53 PID control unit (individual control unit)

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5H004 GA14 GB02 HA02 HB01 HB02 JA03 JA13 JA17 JA22 JB08 KA71 KB02 KB04 KB06 KC27 KD03 KD24 9A001 FF07 HH34 JJ61 KK54

Claims (11)

[Claims] 1. In a topper for fractionating crude oil into various fractions including at least naphtha, kerosene, light gas oil and heavy gas oil, when a factor that changes the properties of the fraction is added, the respective fractions are reduced. A method for automatically controlling the withdrawal amount of a fraction in a topper, wherein the withdrawal amount per minute is automatically controlled to satisfy the properties of each of the fractions, wherein the withdrawal is performed in response to a change in the properties of the fraction. Create an extraction amount model that models the ideal control value of the amount, and determine the control target value of the extraction amount for each of the fractions by fuzzy inference from the difference between the model extraction amount of this extraction amount model and the actual extraction amount. The control target value of the extraction amount of a predetermined fraction of the fraction is set to a fuzzy value based on the extraction amount of another fraction extracted from below the predetermined fraction in the topper. A method for automatically controlling the amount of distillate withdrawal in a topper, wherein 2. The method according to claim 2, wherein the control of the extraction amount of the predetermined fraction is
2. The method according to claim 1, wherein the method is performed using a fuzzy rule based on a change in the properties of the other fraction. 3. The gas turbine according to claim 1, wherein the predetermined fraction is light gas oil, and the other fraction is heavy gas oil.
3. The method for automatically controlling the amount of withdrawal of a fraction in a topper according to item 1. 4. The topper according to claim 1, wherein the automatic control of the extraction amount of the naphtha, kerosene and light diesel oil is performed when the type of crude oil is changed. Automatic extraction amount control method. 5. The method according to claim 1, wherein the automatic control of the extraction amount of the naphtha, kerosene and light gas oil is performed when the extraction amount of a predetermined fraction is changed. Automatic control of the amount of distillate withdrawn in the topper. 6. The method according to claim 1, wherein a safety control constraint is added to the control target value. 7. In a topper for fractionating crude oil into various fractions including at least naphtha, kerosene, light gas oil and heavy gas oil, when a factor that changes the properties of the fraction is added, A withdrawal amount automatic control device for automatically controlling the withdrawal amount of the fraction so as to satisfy the properties of each of the fractions, the ideal control of the withdrawal amount corresponding to the change in the properties of the fractions. A control target setting unit that determines a control target value for each of the fractions by fuzzy inference from a difference between a model extraction amount of the extraction amount model and an actual extraction amount while creating an extraction amount model that models the value. An individual control unit for controlling the extraction amount of each fraction based on the control target value determined by the control target setting unit; A withdrawal amount detection unit to be transmitted to the control target setting unit, and a control target value of a withdrawal amount of a predetermined fraction among the fractions, wherein the topper distills from below the predetermined fraction in the topper. An automatic control unit for extracting a fraction in a topper, comprising: a correction unit that corrects with a fuzzy rule based on a fraction extraction amount. 8. The apparatus according to claim 7, wherein the predetermined fraction is light diesel oil, and the other fraction is heavy diesel oil. 9. A fractioner in the topper according to claim 7, further comprising a safety countermeasure unit which outputs the control target value to the individual control unit in consideration of a safety control constraint. Automatic control of extraction amount. 10. A start / stop instruction unit for instructing start / stop of the control target setting unit corresponding to at least the naphtha, kerosene and light gas oil. An automatic controller for extracting a fraction in the topper described in the above. 11. The fraction in the topper according to claim 5, wherein the start / stop instructing unit instructs start / stop of the control target setting unit when switching the type of crude oil. Automatic control device for the extraction amount.