JP2759052B2 - Liquid level control device and liquid level control method for urea plant synthesis tube - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and a method for controlling a liquid level in a synthesis pipe of a urea plant using fuzzy inference.

[0002]

_2. Description of the Related Art Urea is converted from ammonia (NH ₃ ) and carbon dioxide (CO ₂ ) at a pressure of 140 to 200 bar, 1 bar.
It is synthesized under the condition of 60 to 200 ° C., and its reaction formula is shown below.

2NH ₃ + CO ₂ → NH ₄ CO ₂ NH ₂ (1) NH ₄ CO ₂ NH ₂ → NH ₂ CONH ₂ + H ₂ O (2) where NH ₄ CO ₂ NH ₂ is carbamate, NH ₂ C
ONH ₂ is urea.

[0004] In order to stably operate the entire urea process, it is important to keep urea synthesis conditions such as synthesis pressure and synthesis tube top temperature constant. Also, since the synthesis pressure moves in conjunction with the synthesis tube liquid level, the synthesis tube liquid level control is also an important operating factor. Therefore, the synthesis tube is designed to be larger than necessary in consideration of the fluctuation range of the liquid level of the synthesis tube.

Conventionally, various processes for urea synthesis have been proposed. One example of a flow diagram of a urea synthesis system is shown in FIG. The urea synthesis system includes a synthesis tube 21, a stripper 22,
Carbamate condensers 23, 24, scrapper 2
It is composed of 5 or the like organically.

There are various operating factors that cause fluctuations in the liquid level 27 of the synthesis tube. Among them, the main ones are (a) NH ₃ flow rate, (b) recovered urea liquid flow rate to the scrapper, and (c) recovered urea liquid to the stripper. Liquid flow rate, (d) CO ₂ flow rate,
(E) synthesis pipe liquid level, (f) stripper supply steam pressure, (g) synthesis pressure, and the like.

[0007] The liquid level of the synthesis tube is adjusted by combining the conditions of these operating factors. These multiple operating factors that affect the liquid level of the synthesis tube are complicatedly intertwined.To achieve optimal automatic control, it is necessary to simultaneously understand the effects of many operating factors and set conditions even if disturbance occurs. is necessary.

Conventionally, as one of the automatic control methods, there is a control method by PID (proportional, integral, differential) operation (hereinafter referred to as PID control). The number of inputs and outputs of the control target in PID control is basically one input and one output. When the operating conditions are changed by the steam pressure supplied to the stripper, the generated steam pressure, etc., when the operating load is changed, and the combined pressure fluctuates. When the disturbance that causes the disturbance cannot be dealt with at all, the synthesis pipe outlet valve 26 shown in FIG. 2 is installed, and a skilled driver constantly monitors the operation state of the synthesis system and changes the liquid level of the synthesis pipe according to the situation change. The opening of the synthesis pipe outlet valve 26 was manually adjusted so that 27 was kept within the set upper and lower limits. As described above, it is necessary to set the opening of the synthesis pipe outlet valve 26 in consideration of a plurality of operating factors, which places a great burden on a skilled driver having abundant operation experience.

As one means for solving such a problem, one example of a method for improving general PID control is disclosed in Japanese Patent Application Laid-Open No. Sho 62-135902. Japanese Patent Application Laid-Open No. Sho 62-135902 discloses a method for optimally controlling a process in a process control system, in which a relation between a response pattern and an operation amount is stored in a fuzzy adjustment rule table based on experience and knowledge of a skilled operator. In summary, when performing adjustment in response to a disturbance, an adjustment method is selected and inferred from this table to optimally control the process (Japanese Patent No. 4 orchid, second row to first row).
7 lines). That is, if response patterns corresponding to various types of disturbances and adjustment methods corresponding to these response patterns are prepared for PID adjustment, and a response pattern due to disturbances or the like occurs in the process control system. In this method, an adjustment method corresponding to the response pattern is selected and inferred based on the data, and the process control system is adjusted according to the result. However, since it is based on the PID control, it is not possible to cope with sudden disturbances and fluctuations.

On the other hand, control using fuzzy inference (hereinafter, referred to as control)
There is also a method of controlling a process only by a method called fuzzy control). In this case, the type and number of operation factors are different for each plant, but a rule sentence including many operation factors composed of if and then is considered to include all patterns that can occur in a real plant. Needless to say, it must be created. Therefore, unless the characteristics of the so-called know-how and process related to the operation of each plant are sufficiently understood, it is verified whether the created control system can properly handle steady operation or disturbance until it is installed in an actual factory. Could not.

In addition, when the fuzzy control is adopted, if the types and number of operation factors increase, a complicated rule sentence configuration must be adopted, and therefore tuning becomes complicated. At the same time, besides verifying that fuzzy control is coping properly, even those who designed fuzzy control,
There is also a problem that it is not possible to easily change a rule statement in accordance with a change in an operation condition or an operation mode in accordance with a request of an operation site. For this reason, the simplification of the structure of the rule sentence is being promoted. However, in such a case, it is difficult to adjust the shape (trapezoid or triangle), height, range, etc. of the membership function, and it is necessary to understand the characteristics of the process sufficiently or to use a training simulator that imitates the actual factory movement. If you do not verify the results of the adjustment in advance,
Since control using fuzzy inference may cause disturbances, it cannot be applied to a real factory.

[0012]

As described above, the prior art still has the following problems to be improved.
That is, 1) In PID control, the amount of information that should be the basis for determining a driving operation is too large.
Inability to cope with fluctuations.

2) Even in the PID control assisted by the fuzzy control, since the PID control is basically used, it is impossible to cope with sudden disturbances and fluctuations as described above.

3) In the fuzzy control, it cannot be adopted unless the know-how relating to the operation and the characteristics of the process are sufficiently grasped. If a complicated rule sentence is employed, tuning and verification thereof are very difficult. Further, it is not possible to easily change the operating conditions in accordance with the demands of the operating site and to change the rule text accompanying the changing of the operating mode.

In view of the above-mentioned problems of the prior art, an object of the present invention is to adopt a simplified rule sentence configuration that can make use of the driving sensation, facilitate tuning, and prevent sudden disturbances. The automatic control that can also respond to the situation, it is easy for the driver to understand the situation with the driving sensation at the site, and the change of the driving conditions according to the request of the driving site, the change of the rule statement accompanying the change of the driving mode It is an object of the present invention to provide an automatic control apparatus and method for a liquid level of a urea synthesis tube using fuzzy control which can be easily performed.

[0016]

SUMMARY OF THE INVENTION The present invention relates to a liquid level control device for a synthesis tube in a urea synthesis device including a synthesis tube, a stripper, a carbamate condenser, and a scrapper as main equipment. An input device that inputs the process data from the controller to the fuzzy controller in real time, converts the input process data into a change per unit time, determines the operation mode based on the change in the process data, and fuzzifies the results. A fuzzy controller to be sent to the inference unit, a fuzzy inference unit that performs fuzzy inference based on data from the fuzzy controller and returns the result to the fuzzy controller as a change in the opening of the synthesis pipe outlet valve, and a fuzzy controller for the synthesis pipe outlet valve from the fuzzy controller. Change in opening An output device that periodically outputs the amount to the digital control system, and adjusts the opening of the urea synthesis tube outlet valve by using the change in the opening of the synthesis tube outlet valve periodically output from the fuzzy controller as the set value. It is characterized by doing.

Further, in the fuzzy inference, adjustment of a fuzzy rule and a membership function is performed using a training simulator.

[0018]

With respect to the process data of the main operating factor affecting the fluctuation of the liquid level of the synthesis tube, the amount of change per unit time is obtained, and the operation state is determined from the magnitude of the change, and then the operation state is determined based on the operation state. A fuzzy inference is performed to infer an optimal amount of change in the opening of the synthesis pipe outlet valve, and the amount of change in the opening of the synthesis pipe is set as a set value to control the opening of the urea synthesis pipe outlet valve.

Further, the results of steady operation, load-up operation, load-down operation, and operation with disturbance during steady-state operation of the synthesis plant are confirmed by using a training simulation, and the fuzzy inference is applied to the fuzzy inference. Adjustments can be made.

[0020]

An embodiment of the present invention will be described. FIG.
FIG. 2 is a block diagram illustrating a configuration of an example of the present invention. The figure shows the digital control system 2, the fuzzy controller 3, and the fuzzy inference unit 4, and shows the data exchange between the digital control system 2, the data collection system 10 incorporated in the fuzzy controller 3, and the fuzzy inference unit 4. .

The fuzzy control system 5 comprises a fuzzy controller 3 and a fuzzy inference unit 4. The fuzzy controller 3 exchanges data between the digital control system 2 and the fuzzy controller 3, converts the process data received from the digital control system 2 into a variation per unit time, and determines the operation mode from the variation of the process data. Is installed, the fuzzy inference unit 4 includes a membership function in which the degree of influence of a driving factor is quantified, and a fuzzy rule in which an empirical rule is used as a control rule using the membership function. A fuzzy data unit 11 and a fuzzy inference processing unit 12 for performing fuzzy inference based on data from the fuzzy controller and inferring an optimal opening of the synthesis pipe outlet valve are installed.

Next, the operation will be described along the signal flow with reference to FIG. Here, the numbers of the items correspond to the signs a to f in FIG. 1, respectively.

(A) Process data input Various signals 6 received from the on-site instrument 1 of the urea synthesizer and various signals 1 received from the digital control system 2
3 is sent to the fuzzy controller input device 8 which is an input device of the fuzzy controller 3 at a constant period. The following process data was selected as the main factor affecting the control of the liquid level in the synthesis tube.・ NH ₃ flow rate ・ Recovered urea liquid flow rate to the scrapper ・ Recovered urea liquid flow rate to the stripper ・ CO ₂ flow rate ・ Synthetic tube liquid level ・ Synthesis pressure control valve opening ・ Stripper supply steam pressure control valve opening (b) Fuzzy controller The amount of change in the process data is mainly used for controlling the liquid level of the synthesis tube. Therefore, the digital control system 2 and the exchange of data between the on-site instrument 1 and the fuzzy controller 3 and the digital control system 2 and A data collection system 10 is installed which converts process data received from the on-site instrument 1 into an amount of change per unit time and determines an operation mode from the amount of change in process data.

The data collection system 10 converts the process data from the on-site instrument 1 and the digital control system 2 into a variation at a fixed period in the fuzzy controller 3 and sends it to the fuzzy inference unit 4 via the control unit 7. .

Further, since the adjustment range of the valve opening to be controlled changes depending on whether the operation mode is in a steady state or in an unsteady state such as load-up, NH 4 is one of the factors affecting the control of the liquid level in the synthesis tube. ₃ flow rate, or the scraper and collected urea solution flow rate and CO ₂ flow rate such the material flow variation in the size of the stripper operation mode of steady state, it is determined whether the non-steady state, the judgment of the operation mode and the results Convert to a signal called a flag. The calculation cycle of the fuzzy controller 3 is determined based on the data collection cycle (a). Preferably, they are executed in the same cycle.

(C) Fuzzy inference unit (input signal) The signal converted into the amount of change by the fuzzy controller 3 and the following signal C which is the instantaneous value of the liquid level are sent to the fuzzy inference unit 4 and used for fuzzy inference. Is done.・ Change of NH ₃ flow rate ・ Change of total recovered urea flow rate (sum of recovered urea liquid flow rate to scrapper and stripper) ・ Change of CO ₂ flow rate ・ Change of synthesis pipe liquid level ・ Moment of synthesis pipe liquid level Value (same as the signal of (a)) ・ Change of synthetic pressure control valve opening ・ Change of stripper supply steam pressure control valve opening ・ Operation mode judgment flag Fuzzy inference unit 4 has influence of operating factors A fuzzy data unit 11 comprising a membership function obtained by numerically expressing, a fuzzy rule using an empirical rule as a control rule using the membership function, and a fuzzy controller 3.
A fuzzy inference unit 12 is installed to infer fuzzy inference based on data from the above and to infer an optimum opening of the synthesis pipe outlet valve.

(D) Fuzzy inference unit (output value) The fuzzy inference unit 4 executes fuzzy inference based on the amount of change sent from the fuzzy controller 3 and determines whether the operation is in a steady state or a non-steady state according to the operation mode judgment flag. The fuzzy inference output value d is sent back to the fuzzy controller 3 as the change amount of the synthesis pipe outlet valve opening in the steady state.

(E) Fuzzy controller (output signal) The amount of change in the opening degree of the outlet of the synthesis pipe returned from the fuzzy inference unit 4 to the fuzzy controller 3 is transmitted through the fuzzy controller output device 9 at a constant cycle to the digital control system 2. Is output to

(F) Set value In the digital control system 2, the signal processing 14 is performed at a constant cycle, and the sum of the previous value and the current variation of the valve opening of the synthesis pipe outlet is calculated and set as the valve opening set value. .

The fuzzy data is generally defined by a membership function capable of numerically expressing the degree of influence of a driving factor expressed in an expression close to human sensation, and an empirical rule and know-how are described by using the membership function as “IF-THEN”. It consists of fuzzy rules that can be expressed by control rules in the form of “-”, and was created by the following method.

1) Based on the analysis results of the urea synthesis process development test and the interviews with those who have actually operated the actual plant, the operation factors affecting the liquid level of the urea synthesis tube and the direction of the influence are collected.

This specific example is as follows, for example.

The liquid level rises as each raw material (NH ₃ , CO ₂ , amount of recovered liquid) increases.

When the steam pressure on the stripper is increased, the liquid level rises.

When the generated steam pressure is increased, the liquid level rises.

When the synthetic pressure increases, the liquid level also tends to increase.

The liquid level rises when the temperature at the bottom of the synthesis tube rises.

When the outlet valve of the synthesis tube is opened, the liquid level drops, but after a few minutes the liquid level tends to rise.

2) Quantification of the degree of influence of each operation factor on the liquid level The results of steady operation, load-up operation, load-down operation and operation with disturbance during steady-state operation of the urea synthesis plant are analyzed using a training simulator. By adjusting the fuzzy rule and the membership function, the controllability can be improved without giving any disturbance to the operation at the factory. The training simulator used for this purpose is, for example, registration number P1643-1, program name upots.

As one example of the method, a training simulator is used to change the operating conditions for each of the operating factors affecting the liquid level of the urea synthesis tube collected in 1), and the degree of the influence on the liquid level is determined. Thus, a simplified rule sentence of one rule sentence and one factor is basically determined, and the membership function can be adjusted. Therefore, tuning work on site becomes easy.

3) Application to an actual factory Application to an actual factory has individuality peculiar to the factory because the equipment and instrumentation equipment employed for each factory are different.
In some cases, the membership function needs to be fine-tuned according to its personality. Therefore, it is necessary to confirm, for example, the following items from interviews with skilled operators at the factory and investigation of past operating data.

-Adjustment range of the synthesis pipe outlet valve opening in each plant load and the valve opening in the steady operation-Operation fluctuation width of the raw material supply amount in the steady operation-Operation fluctuation width of the synthesis pipe liquid level and the steady operation area-Load Up / down speed ・ Countermeasures when disturbance occurs According to the above confirmation items, it is possible to further fine-tune the membership function for each operating factor to make it a control method applicable to each actual factory. It is.

After adjusting the fuzzy rule and the membership function using the training simulator in this way, the apparatus is installed in an actual factory, and when a steady operation and a disturbance are given (change of the raw material flow rate, synthetic pressure, stripper supply steam, etc.) ) The synthesis valve outlet valve opening, which is the result of the fuzzy inference, is output to the digital control system as data (offline output) separately from the actual operation of the synthesis tube outlet valve, and the magnitude of the offline output value and the control direction are actual. In cooperation with a trained operator, adjustment of the membership function such as changing the range of the upper stable region of the trapezoid shown in FIG. The controllability can be increased without giving any. After confirming that the results are satisfactory, it is preferable to combine the operation of the synthesis tube outlet valve with the digital control system.

Through the above process, the fuzzy rules and membership functions exemplified below are determined.

1) Fuzzy Rule One variation of NH ₃ flow rate was selected as a typical example based on one rule and one factor. The symbols used in Table 1 are shown below. Table 1. Example of rule statement IF NH ₃ = NL and operation mode = ABNORMAL THEN OUT1 = NL IF NH ₃ = NM and operation mode = ABNORMAL THEN OUT1 = NM ・ ・ ・ ・ ・ ・ IF LEVEL = PL and operation mode = NORMAL THEN OUT2 = PM IF LEVEL = ZR and operation mode = NORMAL THEN OUT2 = ZR ・ ・ ・ ・ ・ ・ ・ NH ₃ : Change amount of NH ₃ flow rate Operation mode: Operation mode judgment flag LEVEL: Synthetic pipe liquid level OUT1: Unsteady state OUT2: Opening change of synthetic pipe outlet valve in steady state OUT2: Opening change of synthetic pipe outlet valve in steady state ABNORMAL: Operating mode is unsteady state NORMAL: Operating mode is steady state NL: Very small NM: Medium small PL: very large PM: moderately large ZR: little change The same is true for other operating factors.

2) Membership Function FIG. 3 shows a typical example of the output of the membership function.

The symbols used in FIG. 3 are shown below.

NL: very small NM: moderately small NS: slightly small ZR: little change PS: slightly large PM: moderately large PL: very large An experimental example of the present invention will be described below.

EXPERIMENTAL EXAMPLE 1 In a urea plant having a daily production of 750 tons,
The synthesis pipe outlet valve 26 shown in FIG. 2 was operated under fuzzy control. The operating conditions are a synthesis pressure of 175 barG and a synthesis temperature of 1.
90 ° C.

The rule sentence shown in Table 1 was used as an example, and the membership function output shown in FIG. 3 was used.
The operation results are shown in FIG. In FIG.

[0051]

[Outside 1] Indicates that disturbance such as changing the NH ₃ flow rate or increasing the synthesis pressure was applied. The vertical axis shows the transition of the liquid level of the synthesis tube. It can be seen that the liquid level can be sufficiently controlled by the fuzzy controller.

Comparative Example The synthesis pipe outlet valve 26 was operated manually under the same conditions as in Experimental Example 1. The operation results are shown in FIG. The vertical axis indicates the transition of the liquid level of the synthesis tube as in FIG.

The operation result by the fuzzy control of FIG. 4 and FIG.
Comparing the results of the manual operation by the fuzzy control, the operation by the fuzzy control is stable with a small fluctuation range of the liquid level.

Experimental Example 2 In Experimental Example 1, the opening of the synthesis pipe outlet valve 26 was fuzzy controlled during the unsteady operation in the process of changing the production volume of the factory from the low load operation to the steady operation. The operation results are shown in FIG. The vertical axis shows the transition of the synthesis valve outlet valve opening and the synthesis solution level. The horizontal axis indicates the transition of time.

In FIG. 6, the operation is switched from the manual operation to the operation by the fuzzy control when the load reaches 50%, the load is increased to 100% as it is, and the synthesis pipe outlet valve is opened when the steady operation is started in that state. The transition of the degree and the synthesis solution level is shown over time. It is understood that the liquid level can be sufficiently controlled by the fuzzy control even during the unsteady operation.

[0056]

The present invention has the following effects for controlling the liquid level of the synthesis pipe of the urea plant by using fuzzy control.

(1) Conventionally, the liquid level control of the synthesis pipe of the urea plant, which has been manually operated, can be automatically performed, and the fluctuation range of the liquid level is extremely small. Therefore, the urea plant synthesis tube does not need to be designed large considering the large liquid level width, and can be designed smaller than before.

(2) Since the degree of influence of the operation factor can be estimated in advance by the training simulator, the rule statement and the membership function can be easily determined, and the rule statement and the membership function can be easily modified.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.

FIG. 2 is a flow chart of a urea synthesis system.

FIG. 3 is a diagram illustrating an example of an output of a membership function.

FIG. 4 is a diagram showing a result of an application example of the present invention during steady operation.

FIG. 5 is a view showing a result of manual control by a synthesis pipe outlet valve.

FIG. 6 is a diagram showing a result of an application example of the present invention during a load-up operation.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 field instrument 2 digital control system (DCS) 3 fuzzy controller 4 fuzzy inference unit 5 fuzzy control system 6 signal received from field instrument 7 control unit 8 fuzzy controller input device 9 fuzzy controller output device 10 data collection system 11 fuzzy data section 12 fuzzy Inference processing unit 13 Signal received from DCS 14 Signal processing in DCS 21 Synthetic pipe 22 Stripper 23, 24 Carbamate condenser 25 Scrapper 26 Synthetic pipe outlet valve 27 Synthetic pipe liquid level

Continuation of the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G05D 9/12 G05B 13/02

Claims (3)

(57) [Claims] 1. A liquid level control device for a synthesis tube in a urea synthesis device mainly including a synthesis tube, a stripper, a carbamate condenser, and a scrapper, wherein process data from an on-site instrument of the urea synthesis device and a digital control system are provided. An input device that inputs to the fuzzy controller in real time, a fuzzy controller that converts the input process data into the amount of change per unit time, judges the operation mode from the amount of change in the process data, and sends the results to the fuzzy inference unit A fuzzy inference unit that infers fuzzy inference based on data from the fuzzy controller and returns the result to the fuzzy controller as a change in the opening of the synthesis pipe outlet valve, and a change in the opening of the synthesis pipe outlet valve from the fuzzy controller. Digital on a regular basis And an output device for outputting to the control system, wherein the opening of the outlet valve of the urea synthesis pipe is adjusted using the amount of change in the opening of the synthesis pipe outlet valve periodically output from the fuzzy controller as a set value. Liquid level control device for urea plant synthesis tube. 2. A liquid level control device for a urea plant synthesis tube according to claim 1, wherein the fuzzy controller exchanges data between the fuzzy controllers of the digital control system and receives process data received from the digital control system per unit time. A data collection system that performs conversion to the change amount and determines the operation mode from the change amount of the process data is installed, and the fuzzy inference unit includes a membership function quantifying the degree of influence of the operation factor and the membership function. A fuzzy data part consisting of fuzzy rules that use empirical rules as control rules and a fuzzy inference processing unit that infers the optimal opening of the synthesis pipe outlet valve based on data from the fuzzy controller and fuzzy inference are installed. Urea plant characterized by the following: Narukan the liquid level controller. 3. The results of steady operation, load-up operation, load-down operation and operation with disturbance during steady-state operation of the urea synthesis plant are confirmed using a training simulator, and the fuzzy rules installed in the fuzzy data section are checked. A liquid level control method for a urea plant synthesis tube, comprising using the liquid level control device for a urea plant synthesis tube according to claim 1 or 2 for adjusting a membership function.