JP6610988B2 - Chemical plant control device and operation support method - Google Patents

Description

  The present invention relates to a technology for supporting operation of a chemical plant, and more particularly to a control device and an operation support method for supporting line control processing of an operator who handles a production line of a chemical plant.

Conventionally, chemical plants that produce chemical substances such as monomers such as ethylene and styrene (hereinafter referred to as “monomers”), or polymer polymers represented by nylon and elastomers (hereinafter referred to as “polymers”). In many cases, a small number of operators manage a production line through which a large amount of product flows.
The quality of chemical substances produced in such a production line may be greatly affected by the skill, knowledge or experience of the operator handling the production line. For example, an operator who is not an expert makes a judgment and performs an operation. In addition to not satisfying the required manufacturing quality, in the worst case, it may cause accidents and disasters.
In addition, since there are many processes in a chemical plant that cause a thermochemical reaction in a container, it is difficult to visually grasp the internal state of the production line. For this reason, sensors for detecting internal states (for example, detected values such as temperature and pressure) are provided at various locations on the production line.

In such a production line, it is necessary for an operator (operator) to check a large amount of data by various sensors as described above to grasp the trend and perform appropriate processing. Very advanced knowledge and skills are required, such as the ability to guess internal conditions and chemical formulas, as well as the ability to make appropriate decisions based on past experience.
In addition, it is difficult not only to transfer the chemical plant operation know-how and skills possessed by skilled operators, but it is a major issue not only in the field of chemical plants but also in field work. There is a need for technology that can support the operation of chemical plants, including technical transfer and education.

  As a technology for supporting operation in a production line, an operation support device for supporting the operation of a manufacturing process in which a continuous product (steel plate by continuous casting) continuously flows, the manufacturing process being the flow of the above product Dividing into multiple stages in the direction, create a product attribute state transition model and equipment state transition model for each stage, and between each product attribute state and equipment state in each stage First state transition diagram creation that sets the initial state for each stage using the discrete model created by defining the influence relationship and creates the state transition diagram of each stage indicating the state that can be reached from the initial state And a second state transition diagram creating unit that creates a state transition diagram of the entire manufacturing process using the state transition diagrams of all stages created by the first state transition diagram creating unit. Knowledge Are (see Patent Document 1).

  In addition, as a technology for realizing prediction operations in chemical plants, a mirror model that visualizes continuous state quantities including data that is not measured by acquiring data online and simulating intermediate values, and dynamic Using a mirror plant that includes an identification model that estimates the performance parameters of plant equipment by data reconciliation with compensation, and an analysis model that predicts the state of the near future plant and searches for the optimal operating state, vinyl acetate A method for predicting an internal state of a plant and detecting an abnormal state in a monomer manufacturing process is known (see Non-Patent Document 1).

JP 2012-146269 A

Emiko Hatsuya and Minoru Nakaya, “Design of Mirror Plant HMI for Realizing Predictive Operation and Application to Vinyl Acetate Monomer Process”, Human Factors, Japan Plant Human Factor Society, February 28, 2014, Vol. No. 18 2

According to the operation support apparatus disclosed in Patent Document 1, a manufacturing process in which a product continuously flows is divided into a plurality of stages, discrete modeling is performed to create a state transition diagram of each stage, Support the operation of the manufacturing process by creating a state transition diagram of the entire manufacturing process using the state transition diagram, such as avoiding operational troubles including the operator's operational know-how and guiding actions when operational troubles occur There is an effect that can be.
However, the operation support apparatus described in Patent Document 1 is an operation support technology applied to a continuous casting process of steel, which is a metal material, and is a control that focuses only on the state change of the steel in the continuous casting process. However, it cannot be directly applied to a manufacturing process of a chemical plant that causes a chemical reaction such as a polymerization reaction between chemical substances.
On the other hand, the prediction operation technique using the mirror plant disclosed in Non-Patent Document 1 can be applied to a chemical plant, but only illustrates the case of a monomer plant having a relatively simple chemical reaction formula. In a polymer plant having a complicated reaction state, it is difficult to construct a prediction model for a chemical reaction because molecular orientation, end groups, and the like affect the polymerization reaction.
Therefore, in the conventional polymer production process in a chemical plant, the quality of the manufactured product cannot be confirmed unless analysis is performed after completion of the chemical reaction.
Because of this, there is an operation support technology that can predict the quality of products inside chemical plants that cause complex and difficult-to-predict chemical reactions in real time, and can provide operational support such as displaying operational guidance during operation. It has been demanded.

  Accordingly, an object of the present invention is to provide a control device and an operation support method in a chemical plant that can support the operation of the chemical plant by modeling the manufacturing process of the chemical plant and performing a prediction simulation of the quality of the manufactured product. Is to provide.

  In order to achieve the above object, the control device for a chemical plant according to the present invention refers to a physical chemical calculation model with respect to a detection value of a sensor provided in the chemical plant, and detects a chemical substance to be manufactured. A reaction control unit that calculates a calculated physical chemical amount at the time, refers to a process management chart with respect to the calculated physical chemical amount at the time of detection of the sensor, and determines guidance information for the operation support, and detection of the sensor A display unit for displaying a value, a calculated physical chemical amount at the time of detection of the sensor, and the guidance information.

  Further, the operation support method for a chemical plant according to the present invention refers to a calculated physical chemistry at the time of sensor detection of a manufactured chemical substance with reference to a physical chemical calculation model with respect to a detection value of a sensor provided in the chemical plant. Calculate the amount, refer to the process control chart for the calculated physical chemical amount of the chemical substance at the time of the sensor detection, determine the guidance information for the operation support, the detection value of the sensor, at the time of the sensor detection The calculated physical chemical amount and the guidance information are displayed to the operator.

According to the control apparatus and operation support method for a chemical plant of the present invention, the operation of the chemical plant can be supported by modeling the manufacturing process of the chemical plant and performing a prediction simulation of the quality of the manufactured product.
In addition, the control device and operation support method of the chemical plant of the present invention are not related to the operator's knowledge, experience, skill, etc. Guidance information such as a recovery operation can be notified to the operator, and an abnormality can be predicted and prevented in advance.

It is the schematic of the manufacturing process model which modeled the manufacturing process to which the control apparatus and operation support method of the chemical plant of this invention are applied. It is a schematic diagram of the chemical plant to which the control device and operation support method of the chemical plant of the present invention are applied. It is a block diagram which shows the structure of the control apparatus which performs the operation support method of the chemical plant of this invention. It is a flowchart which shows the polymer manufacturing process performed with the polymer manufacturing apparatus shown in FIG. 2, Comprising: Fig.4 (a) shows the order of the whole manufacturing process, FIG.4 (b) is the superposition | polymerization process in Fig.4 (a). The steps to be performed are further subdivided. It is a figure which shows the process control chart which shows the specific operation | movement in the operation support method of the chemical plant by Example 1 of this invention. It is a figure which shows the specific operation | movement of the reaction control part in the virtual step subroutine by Example 1 of this invention. It is a figure which shows the specific operation | movement of the reaction control part in the prediction step subroutine by Example 1 of this invention. It is a figure which shows the specific operation | movement of the reaction control part in the virtual step subroutine by Example 2 of this invention. It is a figure which shows the specific operation | movement of the reaction control part in the prediction step subroutine by Example 2 of this invention. It is an example of the monitor screen of the display part in the control apparatus of the chemical plant of this invention.

The chemical plant control device and operation support method of the present invention are applied to assist a control operation or control command input operation of an operator who controls a production line or a production device in a chemical plant production process.
Here, the “chemical plant” in the present application includes a series of apparatuses for performing a process of mixing and reacting a plurality of chemical substances and generating a substance after the reaction.
The “manufacturing apparatus” in the present application means a reaction apparatus for mixing and reacting a plurality of chemical substances, a supply system for supplying the plurality of chemical substances to the reaction apparatus, and a reaction from the reaction apparatus. It includes a recovery system for recovering the subsequent product and a control system for controlling these operations.

In the following examples, polymers are exemplified as chemical substances manufactured in a chemical plant applied to the present invention. The “polymer” here is not particularly limited as long as it is obtained by polymerizing a plurality of chemical substances, but as an embodiment of the present invention, the following (1) The case of a polymer obtained by a polymerization reaction defined by the physicochemical formula of the formula) will be described.

“Dibutyl oxalate (DBO)” + “Nonanediamine (NDA)”
→ “Polyoxamide” + “Butanol” (1)

<Definition of terms>
Terms used in this specification are defined as follows.
・ Detection value
It means the instantaneous value of the output of various sensors provided in the manufacturing apparatus of the chemical plant, for example, the temperature in the polymerization tank, the pressure, the current value of the fin for stirring the polymer, the temperature of the substance flowing in the supply pipe, Pressure, flow rate, etc. are included.
・ Calculated physical chemical amount
This means a numerical value indicating the physical or chemical properties of the polymer calculated by the reaction control unit described later, and is calculated by applying the detected value to the following physicochemical calculation model or temporal continuous model. . For example, the temperature, viscosity, degree of polymerization, etc. of the polymer are included.
・ Physical and chemical state
It means a change in the physical or chemical state of the polymer that is not continuous in time, such as the phase change of the polymer or the active state of the end groups.
・ Production data
Started production of detected values obtained from various sensors when the same or similar polymer production process was carried out in the chemical plant in the past, calculated physical chemical amount of polymer, operation history including operator's operation details, etc. It means what is stored in the storage unit described later along with the time from.

・ Physical chemical calculation model
Includes the physicochemical formula of the polymer defined by the above formula (1), the relational expression between the detected value of the temperature sensor in the polymerization tank and the polymer temperature, or the relational expression between the fin torque value and the degree of polymerization of the polymer. It is used when calculating the calculated physical chemical quantity at the time of sensor detection for the polymer in the polymerization tank from the detection values of various sensors provided in the chemical plant.
・ "Time continuous model"
The definition formulas included in the above physicochemical calculation model and the reaction formulas that cannot be expressed by simple physicochemical models (for example, side reactions that cannot be expressed by the above formula (1) or the influence of the molecular weight distribution of the polymer), etc. When calculating the calculated physical stoichiometry after a predetermined time (for example, t seconds) from the calculated physical stoichiometry at the time of the sensor detection, or the physical chemical calculation model described above is included. It is used when constructing a prediction formula for a reaction formula that cannot be expressed by.
・ "Discrete state model"
It means an arithmetic expression or a calculation model representing the physical chemical state of the polymer, and is applied to the calculated physical stoichiometry after the sensor detection or after the predetermined time, thereby detecting the physical chemical state of the polymer at the time of sensor detection or after the predetermined time. It is used when discriminating. In order to describe such non-continuous discrete state changes, a Petri net modeling technique is used.

・ "Process control chart"
When the control device of the present invention controls a manufacturing process in a chemical plant, it is used to determine the timing of switching between each step based on the detected value or the calculated physical stoichiometry, And a flowchart describing the contents of the operation.
・ Guidance information
The control device of the present invention displays information to the operator, and is information that gives an indication of the next operation of the polymer production process, for example, the current process, the next process to be performed, and the timing for switching the process. Or recovery operation when trouble occurs.
・ "System model"
It is expressed as a flow of the manufacturing process in the chemical plant, grasps at which stage the polymer manufacturing process is currently located, selects and discriminates the above guidance information according to the stage, and provides it to the operator. Used for.
・ "Operation history model"
In the manufacturing process carried out in the past in the chemical plant, the history of the operation actually input by operating the input device of the control device described later by the operator, and the operational behavior (e.g. , “Matters that are extracted by interviews and video shootings such as“ confirm that the sensor value is within the normal operating range before valve operation ”and are not described in the operation manual of the device) And is used to select and discriminate the guidance information referring to the history of past operations according to the calculated physical chemical quantity and physical chemical state.

<Outline of manufacturing process model>
FIG. 1 shows an outline of a manufacturing process model obtained by modeling a manufacturing process to which a chemical plant control apparatus and an operation support method of the present invention are applied.
As shown in FIG. 1, a manufacturing process model 1 of a manufacturing process executed in a chemical plant includes a physicochemical calculation model 2, a temporal continuous model 3, a discrete state model 4, and a system model defined as described above. 5 and an operation history model 6.
In this manufacturing process model, the flow of the manufacturing process (current progress status) is grasped by the system model 5, and the physicochemical calculation model 2 and the temporal continuity with respect to the detected value of the sensor according to the process of the current manufacturing process. By referring to the model 3, the calculated physical chemical quantity of the chemical substance in the plant is calculated, and the physical and chemical state of the chemical substance in the plant is discriminated by referring to the discrete model for the calculation result.
Then, by referring to the operation history model 6 with respect to the calculated physical chemistry amount and the determined physicochemical state, the past operation operation can be performed according to the physical amount or physicochemical state of the chemical substance in the plant. Based guidance information can be extracted.
By using such a manufacturing process model, it is possible to predict the state of chemical substances inside the chemical plant and to provide operators with suggestions on the operation, so provide information on recovery operations when trouble occurs. Can do.
In addition, the above manufacturing process model can be used to simulate operation under virtual conditions by giving a detection value of a virtual sensor other than during actual operation. It can also be used to educate people.

<Outline of polymer production equipment>
FIG. 2 shows an outline of a polymer production apparatus in a chemical plant as an example of a case where the chemical plant control device and operation support method of the present invention are applied to “polymer” of typical chemical substances.
The polymer manufacturing apparatus to which the present invention is applied is a raw material as shown in FIG. 1 when applied to the manufacturing process based on the physicochemical formula shown by the above formula (1) as an example of typical application examples. A reactor 10 for mixing and reacting dibutyl oxalate (DBO) and nonanediamine (NDA), a supply system 20 for supplying dibutyl oxalate, nonanediamine and nitrogen gas (N ₂ ) to the reactor 10, respectively, and the reactor 10 And a recovery system 30 for recovering the product after the reaction.

The reaction apparatus 10 includes a polymerization tank 11 in which a raw material chemical substance is subjected to a polymerization reaction, a heat exchanger 12 for controlling the temperature of a medium for raising or lowering the temperature of the inside of the polymerization tank 11, and the above-described medium. And a circulation pump 13 that circulates between the tank 11 and the heat exchanger 12.
The polymerization tank 11 is provided with fins 11a for stirring a plurality of chemical substances therein, and the introduced plurality of chemical substances (dibutyl oxalate and nonanediamine) are polymerized while being stirred inside the polymerization tank 11 under temperature control. react.
Moreover, the polymerization tank 11, the heat exchanger 12, and the circulation pump 13 are connected by piping through which the above medium flows.
Sensors such as a thermometer and a pressure gauge (not shown) are attached to the components constituting the reaction apparatus 10, and detection values at the time of sensor detection (real time) in each component are detected to be described later. It is configured to send information to the reaction control unit.

Although not shown in FIG. 2, the polymerization tank 11 has a structure including a flow path through which the medium can flow between the inner wall surface and the wall surface.
For example, when the temperature inside the polymerization tank 11 is raised, the medium heated by the heat exchanger 12 is sent to the polymerization tank 11 by the circulation pump 13, and the inside of the polymerization tank 11 is desired while the medium flows through the flow path. Heat to the temperature of.
On the other hand, when the temperature inside the polymerization tank 11 is lowered, the medium cooled by the heat exchanger 12 is sent to the polymerization tank 11 by the circulation pump 13, and the inside of the polymerization tank 11 is moved to a desired state while the medium flows through the flow path. Cool to temperature.
And control of the temperature inside the superposition | polymerization tank 11 is performed by feedback control etc. by the above-mentioned sensor and the reaction control part mentioned later, for example.
Further, from the motor that rotates the fins 11a, for example, the rotational speed and torque of the motor are detected.

The supply system 20 includes a first supply system 21 that supplies dibutyl oxalate (DBO) to the polymerization tank 11 of the reactor 10, and a second supply system 22 that supplies nonanediamine (NDA) to the polymerization tank 11 of the reactor 10. And a third supply system 23 for supplying nitrogen gas (N ₂ ) to the polymerization tank 11 of the reaction apparatus 10.
The first supply system 21 includes a DBO supply source 21a, a reservoir 21b for temporarily storing dibutyl oxalate before being supplied to the polymerization tank 11, and a feed pump 21c for sending the DBO stored in the reservoir 21b. And a valve 21d for adjusting the DBO supply amount, and the DBO supply source 21a, the reservoir 21b, the feed pump 21c, the valve 21d, and the polymerization tank 11 are connected to each other by supply pipes.
The second supply system 22 has the same configuration as that of the first supply system 21, but the description and illustration thereof are omitted because they are redundant.
The third supply system 23 includes a nitrogen gas supply source 23a such as a cylinder, and a valve 23d for adjusting the supply amount of the nitrogen gas. The nitrogen gas supply source 23a, the valve 23d, and the polymerization tank 11 are , Each connected by a supply pipe.
In addition, a sensor such as a thermometer or a pressure gauge (not shown) is attached to each component constituting the supply system 20, and a detection value at the time of sensor detection (real time) in each component is detected to be described later. It is configured to send information to the reaction control unit.

The recovery system 30 includes a first recovery system 31 that takes out the produced polymer from the polymerization tank 11 of the reaction apparatus 10, and a second recovery system 32 that recovers butanol generated by the polymerization reaction from the polymerization tank 11.
The first recovery system 31 includes a valve 31a that opens and closes an outlet that allows the polymer produced from the polymerization tank 11 to flow out, and a pelletizer (granulation) that molds the polymer that flows out of the valve 31a into pellets of a desired size. Machine) 31b, and the polymerization tank 11, the valve 31a and the pelletizer 31b are connected to each other by piping.
The second recovery system 32 includes a valve 32a for opening and closing a flow path for allowing butanol produced as a by-product during the polymerization reaction of the polymer to flow out of the polymerization tank 11, and a condenser (recovery) for liquefying butanol gas that has passed through the valve 32a. Water tank) 32b and a tank 32c for storing and recovering the butanol liquid liquefied by the condenser 32b, and the polymerization tank 11, the valve 32a, the condenser 32b and the tank 32c are connected by piping.
Further, the valve 32a provided in the second recovery system 32 is opened by the instruction or adjustment of the operator, and in addition to the function of recovering butanol during the polymerization reaction of the polymer, the polymerization tank 11 is generated by the generation of butanol. It also has a function as a safety valve that automatically opens at the predetermined value (for example, 0.5 MPa) so that the internal pressure of the valve does not become larger than the predetermined value.
In addition, sensors such as a thermometer and a pressure gauge (not shown) are attached to the components constituting the recovery system 30, and detection values at the time of sensor detection (real time) in each component are detected to be described later. It is configured to send information to the reaction control unit.

<Outline of control device>
FIG. 3 is a block diagram showing the configuration of an example of a typical control apparatus that executes the operation support method for a chemical plant of the present invention.
As shown in FIG. 3, the control device 100 of the present invention includes, as a typical aspect, past manufacturing data of a polymer manufacturing device, a detected value detected by the sensor, and a polymer calculated by an arithmetic processing unit described later. A storage unit 110 that stores and stores data such as a calculated physical chemical amount, and a reaction control unit 120 that calculates the calculated physical chemical amount of the polymer based on the detection value of the sensor and controls the operation of the polymer production apparatus. And a display unit 130 for displaying the detected value of the sensor, the calculated physical chemical amount of the polymer, and the operation support information.

The storage unit 110 is connected to the reaction control unit 120 to exchange electric signals, and the detection values obtained from various sensors when the same or similar polymer is manufactured in the past by the polymer manufacturing apparatus of the chemical plant, Data such as a calculated physical chemical amount of a polymer calculated by an arithmetic processing unit, which will be described later, or an operation history including an operation content of an operator is stored and stored together with the time from the start of production.
Here, in the polymer production process using the polymer production apparatus of the present invention, the data stored and stored in the storage unit 110 is accumulated every minute unit time in the arithmetic processing in the arithmetic processing unit described later. It is configured. When it is desired to suppress the data capacity of the storage unit 110, the data for each minute unit time may be accumulated as thinned data as data for each predetermined interval.
In this way, past information including manufacturing data in the current manufacturing process, operation history executed by the operator, and the like is accumulated in the storage unit 110 as so-called know-how.

As an example, the reaction control unit 120 includes a main control unit 121 and an arithmetic processing unit 122 as shown in FIG.
The main control unit 121 exchanges command signals, data, and the like between the storage unit 110 and the display unit 130, and performs a calculation operation of the calculated physical stoichiometry of the polymer using the process management routine in the arithmetic processing unit 122. A command signal is issued as follows.
The main control unit 121 is also connected to the polymer production apparatus, sends a drive or stop command to each component of the polymer production apparatus, receives an output from a sensor attached to each component, and A function of sending the detected value to the storage unit 110, the arithmetic processing unit 122, and the display unit 130 is provided.

The arithmetic processing unit 122 receives a command from the reaction control unit 140 and performs a main processing unit 122a that performs a main arithmetic processing operation, a virtual processing unit 122b that executes a virtual step subroutine that will be described later, and a prediction step subroutine that is also described later. And a prediction calculation unit 122c to be executed.
The main processing unit 122a contains a program for executing the main arithmetic processing operation, and the virtual arithmetic unit 122b and the prediction arithmetic unit 122c are sub-executions executed during the main arithmetic processing operation of the main processing unit 122a. An arithmetic operation (subroutine) is executed.
Here, FIG. 3 illustrates the case where the arithmetic processing unit 122 includes the main processing unit 122a, the virtual arithmetic unit 122b, and the prediction arithmetic unit 122c as separate processing units. You may comprise suitably combining as one process part.
The program built in the main processing unit 122a includes a physicochemical calculation model used in a subroutine executed by the virtual calculation unit 122b and the prediction calculation unit 122c, a temporal continuous model, a discrete state model, including.

The display unit 130 is connected to the reaction control unit 120. For example, in response to a display command from the reaction control unit 120, an operator who operates the polymer production apparatus operates various detection values and calculated physical stoichiometry of the polymer. It includes a monitor that displays information and an input device for the operator to actually perform the operation.
As the operation information displayed to the operator, the detected value (real time) detected from each sensor, the calculated physical chemical amount of the polymer calculated by the arithmetic processing unit 122 and its predicted value, or the operator Guidance information including the suggestion of the next action and recovery operation when trouble occurs.
As the input device, various input means such as a keyboard and a button type can be adopted. Further, a touch panel type display unit in which a monitor and an input device are integrated may be used.

<Outline of polymer production process>
Below, the outline | summary is shown about an example of the polymer manufacturing process using the control apparatus of the polymer manufacturing apparatus of this invention.
FIG. 4 is a flowchart showing a polymer manufacturing process executed by the polymer manufacturing apparatus, FIG. 4 (a) shows the order of the entire manufacturing process, and FIG. 4 (b) is a polymerization step in FIG. 4 (a). The process performed is further subdivided.

The polymer production process for producing the polymer by the polymerization reaction defined by the physicochemical formula of the above formula (1) using the polymer production apparatus shown in FIG. 2 is NDA (nonanediamine) input as shown in FIG. It consists of process S1, polymerization process S2, and granulation process S3.
The NDA charging step S1 is a step of charging NDA into the polymerization tank in a predetermined molar amount out of the two chemical substances to be polymerized, and using the second supply system 22 shown in FIG. A predetermined amount of NDA is supplied into the polymerization tank 11.
In the polymerization step S2, DBO (dibutyl oxalate) is charged into the polymerization tank 11 in which NDA is previously charged and the temperature and pressure are controlled, and a polymerization reaction as shown in the above formula (1) is caused to generate a polymer. Details of the process will be described later.
The granulation step S3 is a step in which the polymer produced after the reaction is formed into pellets (granular bodies) of a desired size by a pelletizer (granulator) 31b shown in FIG. It becomes a product as spherical particles of about 3 mm.

As shown in FIG. 4B, the polymerization step S2 includes an NDA melting step S21, a mixing / reaction step S22, and a standing step S23.
The NDA melting step S21 is a step of heating and stirring so that the NDA charged in advance is in a molten state (temperature and viscosity) suitable for the reaction before supplying DBO into the polymerization vessel 11. Is heated under temperature control, and when the NDA reaches a predetermined stirring start temperature (for example, 50 ° C.), the fin 11a is rotationally driven to perform stirring. And if the temperature of NDA reaches | attains predetermined | prescribed polymerization start temperature (for example, 150 degreeC), it will transfer to mixing and reaction process S22.

The mixing / reaction step S22 is a step in which DBO is supplied to the polymerization tank 11 and mixed with the melted NDA to cause the polymerization reaction described in the above formula (1), and the first supply to the polymerization tank 11 is performed. DBO is supplied from the system 21 and heating and stirring in the polymerization tank 11 are continued until the polymerization reaction between DBO and NDA sufficiently proceeds. At this time, butanol generated as a by-product of the polymerization reaction is sucked out of the polymerization tank 11 and recovered.
And if reaction fully advances in the superposition | polymerization tank 11, and the inside will fall to about atmospheric pressure, it will transfer to stationary process S23.

The stationary step S23 is a step for removing the gas (bubbles) entrained inside the polymer by a polymerization reaction or stirring before the granulation step S3 for collecting the produced polymer. While stirring by the fins 11a in the tank 11 is stopped, the inside of the tank is made airtight, and high-pressure nitrogen (N ₂ ) is supplied to the inside, thereby applying air pressure from the surface of the polymer to discharge bubbles inside the polymer.
And after allowing predetermined time to pass by stationary process S23, it transfers to granulation process S3.

<Example 1>
The chemical plant control device and the operation support method of the present invention can be applied in each step shown in FIGS. 4A and 4B. Hereinafter, as a specific example thereof, mixing in the polymerization step S2 is performed. -The case where it applies to reaction process S22 is demonstrated.
That is, before implementing the following specific example, in the polymer manufacturing apparatus shown in FIG. 2, NDA is introduced into the polymerization tank 11 in advance using the first supply system 21.
Then, the melted NDA is stirred with the fins 11a while heating the inside of the polymerization tank 11 until the NDA inside the polymerization tank 11 is in a physicochemical state suitable for the polymerization reaction with DBO.
The operator who confirms the end of these operations instructs the start of the mixing / reaction process using the input device of the control device.
Then, the main control unit 121 of the reaction control unit 120 shown in FIG. 3 issues a signal instructing the arithmetic processing unit 122 to start the mixing / reaction process, and the main processing unit 122a of the arithmetic processing unit 122 that has received this signal Then, a program for performing the main arithmetic processing operation of the mixing / reaction process is read and the processing is started.

FIG. 5 is a diagram showing a process management chart showing specific operations in the mixing / reaction process of the operation support method for the polymer production apparatus of the present invention.
As shown in FIG. 5, in the mixing / reaction process, first, a detection value is acquired from a sensor attached to each component of the polymer production apparatus connected to the reaction control unit 120 (step S221).
At this time, in the mixing / reaction process, for example, the temperature and pressure in the polymerization tank 11, the temperature of the medium circulating between the heat exchanger 12 and the current of the motor that rotates the fins 11 a are detected values at the time of sensor detection. The value and the number of revolutions, the flow rate of DBO in the piping of the first supply system 21, the elapsed time since the start of the mixing / reaction process, and the like can be mentioned.

Subsequently, the main processing unit 122a causes the virtual calculation unit 122b to execute a virtual step subroutine S222.
The virtual step subroutine S222 refers to the physicochemical calculation model with respect to the detection value of the sensor acquired in step S221, and the virtual calculation unit 122b calculates the calculated physicochemical amount of the polymer at the time of sensor detection (for example, after the polymerization reaction). The temperature and viscosity of the polyoxamide are calculated.
At this time, if the polymer to be produced cannot be represented by the physicochemical formula as in the above formula (1), an approximate physicochemical calculation model is constructed by combining the past production data and the temporal continuous model. In addition, by referring to the approximate physicochemical calculation model, it is also possible to calculate the calculated physical chemistry amount of the polymer at the time of sensor detection.

Next, the main processing unit 122a causes the prediction calculation unit 122c to execute a prediction step subroutine S223.
The prediction step subroutine S223 refers to the temporal continuity model with respect to the calculated physical stoichiometry of the polymer at the time of sensor detection calculated in step S222, and the predictive calculation unit 122c calculates the calculated physical stoichiometry of the polymer after a predetermined time ( For example, the temperature and viscosity of the polyoxamide are calculated.
The prediction calculation unit 122c uses the above-described physicochemical calculation model and the temporal continuity model to calculate the temporal calculation time for calculating the calculated physicochemical quantity of a continuous polymer after a minute unit time (for example, 0.1 second) from the time of sensor detection. Discrete state model discrimination that determines the physicochemical state of the polymer, such as the phase change of the substance and the active state of the terminal group, by referring to the discrete state model for the continuous model calculation and the calculated physical stoichiometry of the continuous polymer Until the integrated value of the minute unit time reaches a predetermined time (for example, 5.0 seconds), and the final continuous polymer of the continuous unit when the integrated value of the minute unit time reaches the predetermined time. The calculated physical chemical quantity is output as the calculated physical chemical quantity of the polymer after the predetermined time.

Subsequently, the main processing unit 122a performs the detection value of the sensor acquired in step S222, the calculated physical chemical amount of the polymer at the time of sensor detection calculated in the virtual step subroutine S222, and the prediction step subroutine S223. The calculated physical chemical amount of the polymer after the predetermined time is sent to the main control unit 121 (step S224).
Receiving the data from the main processing unit 122a in step S224, the main control unit 121 outputs a command signal to display these data on the display unit 130. Upon receiving this command signal, the display unit 130 outputs the data such as the detected value of the sensor of the polymer production apparatus and the calculated physical chemical quantity of the polymer sent to the monitor.
Then, the operator who sees the display on the display unit 130 continues the heating and stirring in the polymerization tank 11 under the same conditions, or the heating conditions or the stirring conditions (for example, the number of rotations of the fins 11a) according to the operator's judgment. Adjust as appropriate.
Further, the main control unit 121 transmits the sent data and operation history to the storage unit 110, and stores and accumulates them as new manufacturing data.

Subsequently, the main processing unit 122a determines whether or not the feed amount of DBO supplied to the polymerization tank 11 has reached a predetermined value (step S225).
In this case, for example, the feed amount is calculated by integrating the flow rate value of DBO detected by a sensor provided in the piping of the first supply system 21 in FIG. 2, and the determination is made using the feed amount. be able to.
Here, when the determination result in step S225 is “Yes”, the main processing unit 122a outputs a DBO supply stop command signal to the reaction control unit 140 (step S226).
The main control unit 121 that has received the DBO supply stop command signal output in step S226 outputs a command signal to display on the display unit 130 that a predetermined amount of DBO has been supplied into the polymerization tank 11. To do. Upon receiving this command signal, the display unit 130 displays guidance information indicating that a predetermined amount of DBO has been supplied into the polymerization tank 11 on the monitor.

The operator who sees this display turns off the feed pump 21c of the first supply system 21 using the input device, and stops the supply of DBO. By such an operation, it is possible to cause the NDA previously charged in the polymerization tank 11 to react with the supplied DBO without excess or deficiency.
On the other hand, when the determination result in step S225 is “No”, the main processing unit 122a determines that the DBO supply amount has not reached the predetermined amount, and proceeds to step S227.

Subsequently, the main processing unit 122a determines whether or not the internal pressure of the polymerization tank 11 is lower than normal pressure (atmospheric pressure) (step S227).
As shown in the above formula (1), butanol is generated as a gas as a by-product in addition to the polymer as a product by the polymerization reaction of DBO and NDA. On the other hand, at the stage where the mixing / reaction process is started, the polymerization tank 11 is not in communication with anything other than the supply pipe of the first supply system 21, so that the generated butanol gas enters the polymerization tank 11 as the polymerization reaction proceeds. Full pressure increases the internal pressure.
Therefore, when the determination result in step S227 is “No”, the main processing unit 122a outputs a command signal for recovering butanol to the main control unit 121 (step S228).

Upon receiving the butanol recovery command signal output in step S228, the main control unit 121 outputs a command signal to display on the display unit 130 that the valve 32a of the second recovery system 32 is opened to recover butanol gas. To do. Upon receiving this command signal, the display unit 130 displays guidance information indicating that the operation of collecting butanol should be executed on the monitor.
The operator who sees this display turns on the valve 32a of the second recovery system 32 using the input device, and recovers the butanol gas in the polymerization tank 11.

Subsequently, the main processing unit 122a determines whether or not the polymerization reaction has ended (step S229).
Whether or not the polymerization reaction between NDA and DBO is completed is determined by, for example, using the fact that when both of them react without excess or deficiency, the polymer reaches the predetermined viscosity, and the polymer at the time of sensor detection calculated in the virtual step subroutine S222 is used. It is discriminated based on whether or not the viscosity of the liquid has increased to a predetermined value.
And when the determination result in step S229 is “No”, the main processing unit 122a determines that the polymerization reaction is in progress in the polymerization tank 11, and returns to the start of this process control chart and again from S221 to S229. Repeat the operation.

On the other hand, when the determination result in step S228 is “Yes”, the main processing unit 122a completes the polymerization reaction of NDA and DBO in the polymerization tank 11 and collects all the butanol in the polymerization tank 11. A command signal is output to the main control unit 121 so that the mixing / reaction process is completed and the process proceeds to the stationary process.
The main control unit 121 that has received a command signal for shifting to the stationary process from the main processing unit 122a outputs a command signal to display on the display unit 130 that all the polymerization reactions in the polymerization tank 11 have been completed.
Receiving this command signal, the display unit 130 displays guidance information on the monitor that all the polymerization reactions in the polymerization tank 11 have been completed. The operator who sees this display turns off both the switch for supplying DBO from the first supply system 21 and the switch for the valve for recovering butanol in the second recovery system 32 using the input device, and the polymerization tank. 11, the switch for supplying high-pressure nitrogen gas from the third supply system 23 is turned ON.

FIG. 6 is a diagram showing a specific operation of the virtual operation unit 122b in the virtual step subroutine S222 shown in FIG.
As illustrated in FIG. 6, the virtual calculation unit 122b acquires (takes over) the detection value of the sensor from the main processing unit 122a (step SS11).
And while extracting the physicochemical formula (the said (1) formula), heat transfer calculation formula, etc. of the polymer contained in the physicochemical calculation model of the program which the main process part 122a incorporates, the past manufacture data from the memory | storage part 110 are extracted. Data is read when the mixing / reaction process is performed in (Step SS12).
Here, the past production data includes, for example, an appropriate range for the temperature and viscosity of the polymer in the mixing / reaction process, an average value thereof, and the like with the time from the start of production.
Further, as described above, when the polymer to be produced cannot be represented by a physicochemical formula as in the above formula (1), approximate physicochemical calculation is performed by combining the past production data and the temporal continuous model. By constructing a model and referring to the approximate physicochemical calculation model, it is also possible to calculate the calculated physical chemistry amount of the polymer at the time of sensor detection.

Subsequently, the virtual calculation unit 122b refers to the expression extracted from the physical chemical calculation model for the acquired sensor detection value, and calculates the calculated physical chemical amount of the polymer at the time of sensor detection (for example, the temperature and viscosity of the polyoxamide). Is calculated (step SS13).
Here, when calculating the temperature of the polymer at the time of sensor detection, it can be obtained by substituting detection values necessary for an unsteady heat conduction equation including, for example, the initial temperature of the polymer and the temperature of the medium to be heated.
When calculating the viscosity of the polymer at the time of sensor detection, for example, the torque of a motor that rotates the fins 11a is detected, and the viscosity can be obtained from the relational expression between the torque of the motor and the viscosity of the polymer.

Subsequently, the virtual calculation unit 122b determines whether or not the calculated physical chemical amount of the polymer at the time of sensor detection calculated in Step SS13 is included in an appropriate range of past manufacturing data (Step SS14).
The determination at this time can be performed by, for example, determining whether the calculated polymer temperature exceeds the upper limit (first threshold value) of the appropriate temperature of the polymer before polymerization.
If it is determined in step SS14 that the calculated physical chemical amount of the polymer at the time of sensor detection calculated in step SS13 is within the first threshold (that is, “Yes”), the process proceeds to the next step SS15 as it is.
On the other hand, when the determination in step SS14 is “No”, it is determined that the polymerization state of the polymer in the polymerization tank 11 is abnormal, and the process proceeds to step SS16.
Here, the case where the calculated physical chemical amount of the polymer at the time of sensor detection is within the first threshold is exemplified as the determination method of step SS14, but the first threshold is a predetermined value having an upper limit value and a lower limit value. The range may be determined based on whether or not it is within the predetermined range.

Subsequently, the virtual calculation unit 122b determines whether the calculated physical chemical amount of the polymer at the time of the sensor detection is different from the past manufacturing data with respect to the elapsed time of the mixing / reaction process. (Step SS15).
As described above, the storage unit 110 in the control device 100 of the polymer production apparatus of the present invention stores, as past production data, for example, a calculated physical chemical amount (temperature and viscosity) of a polymer when an appropriate mixing / reaction process is performed. Data indicating the relationship between time and time is also stored.
Therefore, in step SS15, for example, the calculated physical stoichiometry of the polymer at the time of the sensor detection calculated with respect to the past appropriate calculated physical stoichiometry of manufacturing data in the elapsed time acquired in step SS11 is a predetermined variation (for example, standard It is determined whether it is within the range of (deviation).
When it is determined that the calculated physical chemical amount of the polymer at the time of sensor detection is within a predetermined variation range (that is, “Yes”), the virtual calculation unit 122b calculates the calculated physical chemistry of the polymer at the time of sensor detection. The amount is output to the main processing unit 122a and the routine is terminated.
On the other hand, when the determination in step SS15 is “No”, it is determined that the polymerization state of the polymer in the polymerization tank 11 is abnormal, and the process proceeds to step SS16.

As described above, when the determinations at step SS14 and step SS15 are “No”, respectively, the type and numerical value of the calculated physical stoichiometric amount of the polymer at the time of detecting the sensor determined to be abnormal are specified (step SS16). The type and numerical value of the calculated physical chemical amount of the polymer at the time of detection of the sensor are sent to the reaction control unit 140, and a warning command signal for instructing display of a warning is also output (step SS17).
Receiving the warning command signal, the main control unit 121 outputs a command signal to display on the display unit 130 that the calculated physical chemical amount of the polymer in the polymerization tank 11 is abnormal.
Upon receiving this command signal, the display unit 130 displays the calculated physical chemical amount of the polymer when the sensor in the polymerization tank 11 is detected on the monitor, and the calculated physical chemical amount of the polymer when the sensor is detected is abnormal. Displays warning information. The operator who sees this display stops the operation of the polymer production apparatus using the input device, and takes measures such as reviewing the heating conditions and confirming whether or not the apparatus has failed.

FIG. 7 is a diagram showing a specific operation of the prediction calculation unit 122c in the prediction step subroutine S223 shown in FIG.
As shown in FIG. 7, the prediction calculation unit 122c acquires (takes over) the calculated physical chemical amount of the polymer at the time of sensor detection from the main processing unit 122a (step SS21).
The definition of the function of the physicochemical formula and the heat transfer formula of the polymer included in the physicochemical calculation model of the program built in the main processing unit 122a, and the calculated physicochemical quantity and time of the polymer included in the temporal continuous model In addition to extracting a formula or a relational expression indicating the calculated physical chemical amount of the polymer included in the discrete state model and the phase state of the polymer, the data when the mixing / reaction process in the past manufacturing data is performed from the storage unit 110 Read (step SS22).
Here, the past production data includes, for example, an appropriate range for the temperature and viscosity of the polymer in the mixing / reaction process, an average value thereof, and the like from the start of production.

Subsequently, the prediction calculation unit 122c resets the variable t for predicting the calculated physical chemical amount of the polymer after a predetermined time t (step SS23), and then adds the minute unit time Δt to the variable t (step SS23). S24).
Here, for example, 0.1 second is adopted as the minute unit time Δt, but a minute interval time or the like can be adopted according to the performance of the processing device (computer or the like) to be calculated.

Subsequently, the prediction calculation unit 122c refers to an equation extracted from the temporal continuous model for the calculated physical chemistry amount of the polymer at the time of sensor detection, and calculates the calculated physicochemical amount of the polymer at time t described above ( For example, the temperature and viscosity of the polyoxamide are calculated (step SS25).
Here, when calculating the temperature of the polymer at time t, it can be obtained by calculating a spatial approximation method such as a finite element method in consideration of a change in a minute unit time, for example.
Further, when calculating the viscosity of the polymer at time t, for example, the torque of the motor that rotates the fin 11a is detected, and the viscosity is obtained from a relational expression using the time between the torque of the motor and the viscosity of the polymer as a variable. You can also.
Next, the prediction calculation unit 122c refers to the discrete state model with respect to the calculated physical stoichiometry of the polymer at time t calculated in step SS25, and determines the phase state of the substance at the time t, the active state of the terminal group, and the like. The physicochemical state of the polymer is determined (step SS26).

Subsequently, the prediction calculation unit 122c determines whether or not the current time t coincides with a predetermined time (target value) set in advance (step SS27).
Here, when the determination in step SS27 is “Yes”, it is determined that the calculated physical chemical amount of the polymer after the predetermined time has been calculated, and the process proceeds to the next step SS28.
On the other hand, when the determination in step SS27 is “No”, the process returns to step SS24 and the operations in steps SS24 to SS26 are repeated again.

Subsequently, the prediction calculation unit 122c determines whether or not the calculated physical chemical amount of the polymer at the time t calculated in Step SS25 is included in an appropriate range of past manufacturing data (Step SS28).
The determination at this time can be performed by, for example, determining whether the calculated polymer temperature exceeds the upper limit (second threshold value) of the appropriate temperature of the polymer before polymerization.
If it is determined in step SS28 that the calculated physical stoichiometry of the polymer at time t calculated in step SS25 is within the second threshold (that is, “Yes”), the process proceeds to the next step SS29 as it is.
On the other hand, when the determination in step SS28 is "No", it is determined that the polymerization state of the polymer in the polymerization tank 11 is abnormal, and the process proceeds to step SS30.
Here, the case where the calculated physical chemical amount of the polymer at the calculated time t is within the second threshold is exemplified as the determination method of Step SS28, but the second threshold is a predetermined value having an upper limit value and a lower limit value. The range may be determined based on whether it is within the predetermined range.
Further, when the calculated physical and chemical amounts of the determined polymers are the same (for example, temperature, viscosity, etc.), the second threshold value may be the same as the first threshold value shown in FIG.

Subsequently, the virtual calculation unit 122b determines whether the calculated physical chemical amount of the polymer at the time t is different from the past data with respect to the elapsed time of the mixing / reaction process (Step S1). SS29).
As described above, the storage unit 110 in the control device 100 of the polymer production apparatus of the present invention stores, as past production data, for example, a calculated physical chemical amount (temperature and viscosity) of a polymer when an appropriate mixing / reaction process is performed. Data indicating the relationship between time and time is also stored.
Therefore, in step SS29, for example, the calculated physical stoichiometry of the polymer at the calculated time t with respect to the calculated physical stoichiometry of the past appropriate production data in the elapsed time acquired in step SS21 has a predetermined variation (for example, standard deviation). ) Is determined.
When the calculated physical stoichiometry of the polymer at the calculated time t is determined to be within a predetermined variation range (that is, “Yes”), the prediction calculation unit 122c calculates the calculated physical stoichiometric amount of the polymer at the time t. The calculated physical chemical amount of the polymer after a predetermined time is output to the main processing unit 122a and the routine is terminated.
On the other hand, when the determination in step SS29 is “No”, it is determined that the polymerization state of the polymer in the polymerization tank 11 is abnormal, and the process proceeds to step SS30.
When the calculated physical and chemical amounts of the discriminated polymers are the same (for example, temperature and viscosity), the standard deviation may be the same as that shown in FIG.

As described above, when the determinations in step SS28 and step SS29 respectively become “No”, the type and numerical value of the calculated physical stoichiometry of the polymer determined to be abnormal are specified (step SS30), and the polymer The type and numerical value of the calculated physical chemical amount are sent to the main control unit 121, and a warning command signal for instructing display of a warning is also output (step SS31).
The main control unit 121 that has received the warning command signal outputs a command signal to display on the display unit 130 that the calculated physical chemical amount of the polymer in the polymerization tank 11 after a predetermined time is predicted to be abnormal.
Upon receiving this signal, the display unit 130 displays the calculated physical stoichiometry of the polymer after the predetermined time in the polymerization tank 11 on the monitor and means that the calculated physical stoichiometry after the predetermined time is predicted to be abnormal. Display warning information.
The operator who sees this display confirms whether the detected value of the sensor at the time of detecting the sensor displayed on the display unit 130 is normal, and if necessary, operates the polymer manufacturing apparatus using the input device. Is stopped and the heating conditions are reviewed and the device is checked for failure.

In the control device 100 of the present invention, the main control unit 121 of the reaction control unit 120 causes the arithmetic processing unit 122 to execute the above-described operation, so that the calculated physical properties of the polymer at the time of sensor detection in the mixing / reaction process and after a predetermined time. By calculating the chemical amount, it is possible to predict the physicochemical state of the polymer inside the polymerization tank, which was difficult in the past.
In addition, by providing a function of determining abnormal states with respect to past production data when calculating the calculated physical stoichiometry of the polymer at the time of the sensor detection and after a predetermined time, the polymerization tank 11 can be used during the mixing / reaction process. Even when an abnormal polymerization reaction occurs inside, it is possible to detect an abnormal event in the polymer production apparatus or to provide warning information including an abnormal event after a predetermined time to the operator.

<Example 2>
Hereinafter, as a second specific example of the control device and the operation support method of the chemical plant of the present invention, a case where a judgment referring to the operation history model is added in the virtual step subroutine and the prediction step subroutine of the mixing / reaction step S22 will be described. To do.
Here, in Example 2, the mixing / reaction process is performed using the same process control chart as shown in FIG. Since most of the operations are the same as the operations executed in the virtual step subroutine and the prediction step subroutine shown in FIGS. 6 and 7, the following description will be focused on the portions added in the second embodiment.

FIG. 8 is a diagram illustrating specific operations in the virtual step subroutine S222 according to the second embodiment of the present invention.
That is, in the second embodiment, as in the first embodiment, the virtual calculation unit 122b acquires (takes over) the detection value of the sensor from the main processing unit 122a, and mixes and reacts in the past manufacturing data from the storage unit 110. Data when the process is performed is read, and a calculated physical chemical amount (for example, temperature and viscosity of polyoxamide) of the polymer at the time of sensor detection is calculated (steps SS11 to SS13).

Subsequently, the virtual calculation unit 122b determines whether the calculated physical chemical amount of the polymer at the time of sensor detection calculated in Step SS13 is included in an appropriate range of past manufacturing data, and then at the time of sensor detection. It is determined whether or not the calculated physical chemical amount of the polymer in the difference between the past data and the elapsed time of the mixing / reaction process is different (steps SS14 to SS15).
When it is determined that the calculated physical chemical amount of the polymer at the time of sensor detection is within a predetermined variation range (that is, “Yes”), the virtual calculation unit 122b calculates the calculated physical chemistry of the polymer at the time of sensor detection. The amount is output to the main processing unit 122a and the routine is terminated.
On the other hand, when the determination in step SS14 or SS15 is “No”, it is determined that the polymerization reaction in the polymerization tank 11 is in an abnormal state, and the process proceeds to step SS16.

As described above, when the determinations in step SS14 and step SS15 are “No”, the virtual calculation unit 122b specifies the type and numerical value of the calculated physical stoichiometry of the polymer determined to be abnormal (step SS16). ).
Thereafter, the virtual calculation unit 122b refers to the operation history model, and from the operation history by the past operator corresponding to the type and value of the calculated physical stoichiometry of the polymer at the time of detecting the sensor determined to be abnormal, An operation for recovering from the currently occurring abnormal state is selected and determined as guidance information (step SS17).
In addition, when an abnormal numerical value that can be determined that recovery cannot be performed by referring to the operation history model is calculated, guidance information indicating that there is no recovery operation may be determined.
Then, the virtual calculation unit 122b sends the type and numerical value of the calculated physical chemical amount of the polymer and the guidance information to the main control unit 121 and outputs a warning command signal for instructing display of a warning (Step SS18). ).

Upon receiving the warning command signal, the main control unit 121 outputs a command signal to display on the display unit 130 that the polymerization reaction in the polymerization tank 11 is in an abnormal state and guidance information for recovering the abnormal state. Then the routine ends.
Upon receiving this command signal, the display unit 130 displays the calculated physical chemical amount and guidance information of the polymer in the polymerization tank 11 on the monitor, and warning information that the polymerization reaction in the polymerization tank 11 is in an abnormal state. indicate. The operator who sees this display executes an operation input to the polymer production apparatus based on the guidance information using the input device, and performs a recovery operation of an abnormal reaction state.
As mentioned above, if an abnormal value is calculated that can be determined that recovery is not possible by referring to the operation history model, only the warning information is displayed to the operator, and the polymer production equipment is stopped Can be encouraged.

FIG. 9 is a diagram showing specific operations in the prediction step subroutine S223 according to the second embodiment of the present invention.
That is, in Example 2, the prediction calculation unit 122c acquires (takes over) the calculated physical chemical amount of the polymer at the time of sensor detection from the main processing unit 122a (step SS21), and mixes the past manufacturing data from the storage unit 110. Data when the reaction process is performed is read (steps SS21 to SS22).

Subsequently, the prediction calculation unit 122c resets the variable t for predicting the calculated physical chemical amount of the polymer after the lapse of the predetermined time t, and then adds the minute unit time Δt to the variable t (Steps SS23 to SS24). .
Here, as in the case of the first embodiment, for example, 0.1 second is adopted as the minute unit time Δt. However, the minute unit time Δt may be further reduced depending on the performance of the processing device (computer or the like). It can be adopted.

Subsequently, as in the case of Example 1, the prediction calculation unit 122c refers to the formula extracted from the temporal continuous model with respect to the calculated physical stoichiometry of the polymer at the time of sensor detection, and the time described above The calculated physical chemical quantity of the polymer at t (for example, the temperature and viscosity of polyoxamide) is calculated (step SS25).
Next, the prediction calculation unit 122c refers to the discrete state model with respect to the calculated physical stoichiometry of the polymer at time t calculated in step SS25, and determines the phase state of the substance at the time t, the active state of the terminal group, and the like. The physicochemical state of the polymer is determined (step SS26).

Subsequently, the prediction calculation unit 122c determines whether or not the calculated physical chemical amount of the polymer at the time t calculated in Step SS25 is included in an appropriate range of past manufacturing data (Step SS27), and thereafter It is determined whether or not the calculated physical stoichiometry of the polymer at time t is different from the past data with respect to the elapsed time of the mixing / reaction process (step SS28).
When it is determined that the calculated physical chemical amount of the polymer at the calculated time t is within a predetermined variation range (that is, when both are determined to be “Yes” in steps SS27 and SS28), the prediction calculation unit 122c is Then, it is determined whether or not the current time t coincides with a predetermined time (target value) set in advance (step SS29).
Here, when the determination in step SS29 is “Yes”, the prediction calculation unit 122c sets the calculated physical chemical amount of the polymer at the time t as the calculated physical chemical amount of the polymer after a predetermined time to the main processing unit 122a. Output and end the routine.
On the other hand, when the determination in step SS29 is “No”, the process returns to step SS24 and the operations of step SS24 to step SS29 are repeated again.

If either of the determinations in step SS27 or SS28 is “No”, it is determined that the polymerization reaction in the polymerization tank 11 is in an abnormal state, and the process proceeds to step SS30. In step SS30, the prediction calculation unit 122c specifies the type and numerical value of the calculated physical chemical amount of the polymer determined to be abnormal.
Thereafter, the prediction calculation unit 122c refers to the operation history model, and from the operation history by the past operator corresponding to the type and value of the calculated physical stoichiometry of the polymer at the time t determined as the abnormality, t An avoidance operation for preventing an abnormal state predicted to occur after a second (that is, after a predetermined time) is selected and determined as guidance information (step SS31).
In addition, when an abnormal numerical value that can be determined that an abnormal state cannot be avoided by referring to the operation history model is calculated, guidance information indicating that there is no avoidance operation may be determined.
Then, the prediction calculation unit 122c sends the type and numerical value of the calculated physical chemical amount of the polymer at the time t and the guidance information to the main control unit 121, and also outputs a warning command signal for instructing display of a warning. (Step SS32).

Upon receiving the warning command signal, the main control unit 121 predicts that the calculated physical chemical amount of the polymer after a predetermined time in the polymerization tank 11 is abnormal on the display unit 130, and guidance for avoiding the abnormal state. A command signal is output to display information, and the routine is terminated.
Upon receiving the command signal, the display unit 130 displays the calculated physical chemical amount and guidance information of the polymer after the predetermined time in the polymerization tank 11 on the monitor, and predicts that the calculated physical chemical amount after the predetermined time is abnormal. Displays warning information that means
The operator who sees this display checks whether there is an abnormality in the detection value at the time of sensor detection displayed on the display unit 130 and, if necessary, uses the input device to produce a polymer based on the guidance information. An operation input to the device is executed to avoid an abnormal reaction state.
In addition, as described above, when an abnormal numerical value that can be determined to be unavoidable by referring to the operation history model is calculated, only warning information is displayed to the operator and an emergency stop of the polymer production apparatus is urged. You can also

The control device 100 of the present invention can predict the physicochemical state of the polymer inside the polymerization tank, which has been difficult in the past, as shown in Example 1, and calculates the polymer calculated with reference to the operation history model. It is possible to provide the operator with suggestions of operations according to the physical and chemical amounts.
And when performing the calculation of the calculated physical stoichiometry of the polymer at the time of the sensor detection and after a predetermined time, the abnormal state with respect to the past production data is determined, and the recovery operation to the operator referring to the operation history model is performed. By providing a function for selecting and discriminating as guidance information, even if an abnormal polymerization reaction or the like occurs in the polymerization tank 11 during the mixing / reaction process, damage such as failure or breakage of the polymer production apparatus can be prevented. In addition to minimizing it, it is possible to predict the state after a predetermined time and prevent anomalies in advance.

FIG. 10 shows an example of a monitor screen of the display unit 130 when the mixing / reaction process is performed by the chemical plant control device and operation support method of the present invention.
As shown in FIG. 10, the monitor of the display unit 130 has a display unit 131 for providing information such as various data of the polymer production apparatus to the operator.
In addition, the display unit 131 includes a monitoring screen 132 for displaying a detection value from a sensor provided in the entire polymer production apparatus, and a polymerization tank display screen for displaying a calculation value such as a calculated physical chemical amount of the polymer in the polymerization tank. 133, a quality prediction screen 134 for displaying a predicted value of the quality of the polymer produced in the polymerization tank, and an operation guidance screen 135 for providing information such as guidance information and warning information to the operator.

The monitoring screen 132 displays, for example, a schematic diagram showing a connection state of each component of the polymer manufacturing apparatus, and numerical values of detection values output from various sensors attached to the respective components.
At this time, by displaying the detected value determined to be abnormal in the virtual step subroutine in red, for example, it is possible to make it easy for the operator to identify which component is abnormal together with the warning display. .

The in-polymerization tank display screen 133 displays the calculated physical stoichiometry of the polymer that changes with the polymerization reaction occurring in the polymerization tank 11.
At this time, the detection value from the sensor on the display screen 132 in the polymerization tank, the calculated physical chemical amount of the polymer at the time of sensor detection calculated in the virtual step subroutine, and the like are displayed. By displaying the detected value or calculated physical chemical quantity determined to be abnormal in the virtual step subroutine in red, for example, it is easy for the operator to identify which component has an abnormality along with a warning display Can do.

The quality prediction screen 134 displays a predicted value of a change in the calculated physical chemical amount of the polymer due to a chemical reaction or the like in the polymerization tank 11 in a graph or the like.
For example, regarding the calculated physical chemical quantity such as the degree of polymerization of the polymer in the polymerization reaction, the horizontal axis represents time, and the vertical axis represents the calculated physical chemical quantity of the polymer. The calculated physical chemical amount is displayed in a graph.
At this time, for example, if the threshold value for determining the calculated physical chemical amount of the polymer on the vertical axis is indicated by a horizontal line, the operator sees the predicted value after a predetermined time and determines the abnormality in the polymerization tank 11 in advance. To be able to predict.
Further, a function of switching the calculated physical chemical amount of the polymer to be displayed by operating an input button (not shown) or the like may be provided.

As described above, the operation guidance screen 135 is configured to display guidance information such as process switching and operation start timing based on the calculated physical chemistry amount of the calculated polymer or its predicted value to notify the operator. Has been.
Further, the operation guidance screen 135 displays a warning of an abnormal detected value or calculated physical chemical amount to the operator when the detected value detected by the virtual step subroutine and the predicted step subroutine or the calculated calculated physical chemical amount is determined to be abnormal. It also has a function to display and notify information, its contents, or recovery operations.
At this time, when displaying the guidance information and the warning information, the effect of informing the operator may be improved by sounding a buzzer sound or the like together.

With the above-described configuration and embodiments, the chemical plant control device and the operation support method according to the present invention are described above with reference to the physicochemical calculation model for the detection value of the sensor provided in the chemical plant. Calculate the calculated physical stoichiometry of the polymer at the time of sensor detection, refer to the process management chart for the calculated physical stoichiometry of the polymer at the time of sensor detection, and provide guidance information for supporting operation of the polymer manufacturing apparatus. It discriminate | determines and implements the operation | movement which displays the said detected value, the calculated physical chemical amount of the polymer at the time of the said sensor detection, and the said guidance information with respect to an operator.
With such a configuration, it is possible to accurately obtain the calculated physical chemical amount of the polymer that changes due to the chemical reaction of the polymer generated in the polymerization tank and its predicted value, and it is abnormal due to comparison with past production data. The reaction state can be predicted and notified to the operator.

In Example 1 and Example 2 above, if the target value for the predetermined time is set to be sufficiently long, that is, set to a time after the completion of the normal polymerization reaction, the final product is determined based on the current polymerization reaction status. The calculated physicochemical amount of the polymer and the physicochemical state of the polymer can be predicted.
As a result, when the quality of the final product can be predicted to be poor during operation of the chemical plant, the operator may perform a recovery operation based on the guidance information, or may stop the operation as soon as it is determined that recovery is impossible. It is possible to predict the state inside the polymerization tank during the reaction, which has been difficult in the past.

As described above, according to the control device and the operation support method of the chemical plant of the present invention, the manufacturing process is modeled and the function of predicting the state of the product in the reactor is provided, so that Regardless of experience, skill, etc., it is possible to notify the operator of the timing for executing various operations of the polymer production device at an appropriate timing, and it is possible to predict and prevent an abnormality from occurring in the device. Become.
In addition, according to the control device and operation support method of the chemical plant of the present invention, it is possible to perform an operation simulation under virtual conditions by applying the above process model, and it can be used for training inexperienced operators. Become.

The present invention is not limited to the first and second embodiments described above, and various modifications can be made.
For example, in Example 1 and Example 2 described above, the case where the apparatus and method of the present invention are applied to the mixing / reaction process related to polymer production shown in FIG. 3 has been described. For example, the granulation of FIG. The above-described physicochemical calculation model, temporal continuous model, and discrete state model are also applied to the process and the NDA melting process or stationary process in FIG. It is possible.

In the first embodiment and the second embodiment, the case where the step of calculating the calculated physical stoichiometry of the polymer at the time of sensor detection is performed from the detection value of the sensor in the virtual step subroutine is illustrated. A step of discriminating the physicochemical state of the polymer at the time of sensor detection may be added by referring to the discrete state model with respect to the calculated physical chemistry amount of the polymer at the time of sensor detection.
Further, in Example 1 and Example 2 above, in the virtual step subroutine or the prediction step subroutine, when it is determined that the calculated physical chemical amount of the polymer at the time of the calculated sensor detection or after a predetermined time is an abnormal value, Although the case where the recovery operation and the avoidance operation of the abnormality are discriminated by referring to the operation history model has been illustrated, even if the calculated physical stoichiometry within the normal range is obtained by these subroutines, the operation history model described above is used. By referencing, it is possible to add a step of extracting guidance information on operations for obtaining a product of higher quality performed in the past and making suggestions to the operator.
Furthermore, in the said Example 1 and Example 2, although the aspect which notifies an operator by displaying the guidance information, warning information, etc. regarding operation of a polymer manufacturing apparatus on a display part was illustrated, as needed, an operator Instead of prompting the operation with the input device, the reaction control unit can be configured to automatically perform the process switching or the stop operation in the event of an abnormality.

Each of the configurations, functions, processing units, processing means, and the like shown in the first and second embodiments may be realized by hardware by designing a part or all of them with, for example, an integrated circuit.
In addition, each configuration, function, and the like shown in the first and second embodiments may be realized by software by a processor interpreting and executing a program that realizes each function.
Furthermore, although the aspect which comprises separately each process part (for example, virtual calculation part and prediction calculation part) in the reaction control part shown in Example 1 and Example 2 was illustrated, they were united and constituted as an integrated system. May be.
In addition, various modifications can be made to the above-described embodiments without departing from the gist of the present invention.

11 Polymerization tank 11a Fin 12 Heat exchanger 21 First supply system 22 Second supply system 23 Third supply system 31 First recovery system 32 Second recovery system 100 Controller 110 Storage unit 120 Reaction control unit 121 Main control unit 122 Operation processing unit 122a Main processing unit 122b Virtual operation unit 122c Prediction operation unit 130 Display unit

Claims (22)

A control device for supporting operation of a chemical plant,
The controller is
Referring to the physical chemical calculation model for the detection value of the sensor provided in the chemical plant, the calculated physical stoichiometry at the time of sensor detection of the chemical substance to be manufactured is calculated, and the calculated physical chemistry at the time of detection of the sensor A reaction control unit for determining guidance information for the operation support with reference to a process management chart with respect to the quantity;
A display unit for displaying a detection value of the sensor, a calculated physical chemical amount at the time of detection of the sensor, and the guidance information;
A control device comprising: The control device further includes a storage unit connected to the reaction control unit,
The storage unit stores past manufacturing data of the chemical substance manufactured in the chemical plant, and manufacturing data newly acquired based on a command of the reaction control unit,
The reaction control unit constructs an approximate physicochemical calculation model by combining the past manufacturing data and a temporal continuous model when calculating the calculated physical chemical quantity at the time of detection of the sensor, and the approximate physicochemical calculation model The control device according to claim 1, wherein a calculated physical chemical amount at the time of detection of the sensor is calculated with reference to FIG. The reaction control unit calculates a calculated physical stoichiometry after a predetermined time with reference to a temporal continuous model for the calculated physical stoichiometry at the time of the sensor detection, and calculates the calculated physical stoichiometry after the predetermined time. With reference to the discrete state model, the physical chemical state of the chemical substance after a predetermined time is determined,
The control device according to claim 2 , wherein the display unit displays a calculated physical chemical amount after the predetermined time and a physical chemical state after the predetermined time. The reaction control unit determines that the calculated physical chemical amount at the time of sensor detection is an abnormal value when the calculated physical chemical amount at the time of sensor detection exceeds a threshold based on past manufacturing data, and the display The control device according to claim 3 , wherein warning information is displayed on the unit. The reaction control unit refers to an operation history model with respect to a calculated physical chemical quantity at the time of detecting the sensor, and adds operation information according to an operation history of a past chemical plant to the guidance information. The control device according to claim 3 or 4 . The process management chart is a flowchart describing the contents of operations in each process when performing a plurality of processes,
Calculating the physico-chemical calculation model is a definition equation, including physical chemical formula of the chemical substance, the temporal continuity model, the definition formula or the like contained in the physical chemistry calculation model, expressed in function of time in the reaction process an expression, the discrete state model is an arithmetic expression or a calculation model that represents the physical-chemical state of the chemical, the operation history model, behavior or the operating user has actually input, operating person carried unconsciously Is a model that memorizes the history of operational actions
6. The physicochemical state is a change in a physical or chemical state of the chemical substance that is not continuous in time, including a phase change of the chemical substance and an active state of a terminal group. The control device described in 1. The calculated physical stoichiometry at the time of detection of the sensor is a numerical value indicating physical or chemical properties of the chemical substance itself, including the temperature, viscosity, and degree of polymerization of the chemical substance. The control device according to any one of the above. A method for supporting the operation of a chemical plant,
Referring to the physicochemical calculation model with respect to the detection value of the sensor provided in the chemical plant, the calculated physical stoichiometry at the time of sensor detection of the manufactured chemical substance is calculated, and the chemical substance at the time of sensor detection is calculated. Refer to the process control chart for the calculated physical chemical amount, determine guidance information for the operation support,
An operation support method for displaying a detection value of the sensor, a calculated physical chemical amount at the time of detection of the sensor, and the guidance information to an operator. When calculating the calculated physical chemical quantity at the time of sensor detection, an approximate physicochemical calculation model is constructed by combining past manufacturing data of the chemical substance manufactured in the chemical plant and a temporal continuous model, and the approximation 9. The operation support method according to claim 8, wherein a calculated physical chemical amount at the time of sensor detection is calculated with reference to a physical chemical calculation model. With reference to the temporal continuous model for the calculated physical stoichiometry at the time of sensor detection, the calculated physical stoichiometry after a predetermined time is calculated, and the discrete state model is calculated for the calculated physical stoichiometry after the predetermined time. Referring to determine the physicochemical state of the chemical substance after a predetermined time,
10. The operation support method according to claim 9 , wherein the calculated physical and chemical amount after the predetermined time and the physical and chemical state after the predetermined time are displayed to an operator. When the calculated physical chemical amount at the time of sensor detection exceeds a threshold based on past manufacturing data, it is determined that the calculated physical chemical amount at the time of sensor detection is an abnormal value and warning information is displayed. The operation support method according to claim 10 . Refers to the operation history model for calculating physicochemical amount at the time of the sensor detection, claim 10 or information of operation corresponding to the operation history of the past of the chemical plant, characterized in that added to the guidance information 11. The operation support method according to 11 . The process management chart is a flowchart describing the contents of operations in each process when performing a plurality of processes,
Calculating the physico-chemical calculation model is a definition equation, including physical chemical formula of the chemical substance, the temporal continuity model, the definition formula or the like contained in the physical chemistry calculation model, expressed in function of time in the reaction process an expression, the discrete state model is an arithmetic expression or a calculation model that represents the physical-chemical state of the chemical, the operation history model, behavior or the operating user has actually input, operating person carried unconsciously Is a model that memorizes the history of operational actions
The physical or chemical state is a change in a physical or chemical state of the chemical substance that is not continuous in time, including a phase change of the chemical substance and an active state of a terminal group. The operation support method described in 1. The calculated physical stoichiometry at the time of sensor detection is a numerical value indicating the physical or chemical properties of the chemical substance itself, including the temperature, viscosity, and degree of polymerization of the chemical substance. The operation support method according to any one of the above. The reaction control unit determines that the calculated physical chemical amount after the predetermined time is an abnormal value when the calculated physical chemical amount after the predetermined time exceeds a threshold based on past manufacturing data, and the display The control device according to claim 3, wherein warning information is displayed on the unit. The reaction control unit adds an operation information corresponding to an operation history of a past chemical plant to the guidance information with reference to an operation history model with respect to the calculated physical chemical quantity after the predetermined time. The control device according to claim 3 or 15. The process management chart is a flowchart describing the contents of operations in each process when performing a plurality of processes,
The physicochemical calculation model is a definition formula including a physicochemical formula of the chemical substance, and the temporal continuous model is an operation in which the definition formula included in the physicochemical calculation model is expressed as a function of time in a reaction process. The discrete state model is an arithmetic expression or calculation model that represents the physical and chemical state of the chemical substance, and the operation history model is an operation that is actually input by the operator or performed unconsciously by the operator. Is a model that memorizes the history of operational actions
The physicochemical state is a change in physical or chemical state of the chemical substance that is not continuous in time, including a phase change of the chemical substance and an active state of a terminal group. The control device described in 1. 16. The calculated physical stoichiometry after the predetermined time is a numerical value indicating physical or chemical properties of the chemical substance itself, including the temperature, viscosity and degree of polymerization of the chemical substance. The control device according to any one of -17. When the calculated physical chemical amount after the predetermined time exceeds a threshold based on past manufacturing data, the calculated physical chemical amount after the predetermined time is determined to be an abnormal value, and warning information is displayed. The operation support method according to claim 10. The operation information corresponding to the past operation history of the chemical plant is added to the guidance information with reference to an operation history model for the calculated physical chemical quantity after the predetermined time. 19. The operation support method according to 19. The process management chart is a flowchart describing the contents of operations in each process when performing a plurality of processes,
The physicochemical calculation model is a definition formula including a physicochemical formula of the chemical substance, and the temporal continuous model is an operation in which the definition formula included in the physicochemical calculation model is expressed as a function of time in a reaction process. The discrete state model is an arithmetic expression or calculation model that represents the physical and chemical state of the chemical substance, and the operation history model is an operation that is actually input by the operator or performed unconsciously by the operator. Is a model that memorizes the history of operational actions
21. The physical or chemical state is a change in physical or chemical state of the chemical substance that is not continuous in time, including a phase change of the chemical substance and an active state of a terminal group. The operation support method described in 1. The calculated physical stoichiometry after the predetermined time is a numerical value indicating physical or chemical properties of the chemical substance itself, including the temperature, viscosity, and degree of polymerization of the chemical substance. The operation support method of any one of -21.

Images