CN117270483B

CN117270483B - Full-flow dynamic optimization control method and device for chemical production device and electronic equipment

Info

Publication number: CN117270483B
Application number: CN202311562473.6A
Authority: CN
Inventors: 娄海川; 袁美晨; 王皖慧; 管振国; 刘伟; 陆海琛; 陈俊伟
Original assignee: Zhongkong Technology Co ltd
Current assignee: Zhongkong Technology Co ltd
Priority date: 2023-11-22
Filing date: 2023-11-22
Publication date: 2024-04-12
Anticipated expiration: 2043-11-22
Also published as: CN117270483A

Abstract

The application discloses a full-flow dynamic optimization control method and device for a chemical production device and electronic equipment. Comprising the following steps: acquiring a target parameter set of a chemical production device in the production process; constructing a nonlinear steady-state model of the chemical production device based on the target parameter set, and determining a nonlinear gain matrix of the controlled variable parameter based on the model; constructing a dynamic linear prediction matrix according to the operation variable parameters and the controlled variable parameters, and correcting the dynamic linear prediction matrix through the nonlinear gain matrix to obtain a full-flow nonlinear dynamic optimization model; and (3) performing online data setting on the target parameter set, combining an optimization target and optimization constraint conditions, solving the full-flow nonlinear dynamic optimization model in real time to obtain a dynamic optimization result, and transmitting the result to an advanced controller for execution. The method and the device solve the technical problems that the time for calculating the optimization process variable of the chemical production device by the related real-time optimization technology based on the steady-state mechanism model is too long and is not easy to converge.

Description

Full-flow dynamic optimization control method and device for chemical production device and electronic equipment

Technical Field

The application relates to the technical field of chemical production, in particular to a full-flow dynamic optimization control method and device for a chemical production device and electronic equipment.

Background

At present, the online optimization technology in the chemical industry has obtained more successful application examples, and most of the examples of the industrial application can bring a certain economic benefit improvement to enterprises. With the popularization of advanced control (Advanced Process Control, APC), enterprises are not required to meet the stable operation of single devices (such as rectification columns), but hope to dig greater economic benefit profit to improve market competitiveness on the premise of ensuring safe and stable operation of single devices or multiple devices, and therefore, urgent demands are made for real-time optimization/dynamic optimization.

A related Real-time online optimization (RTO) system is generally built on an advanced controller, which calculates an overall optimization objective function to obtain an overall optimization objective value, and transmits the overall optimization objective value to the advanced controller, and the advanced controller adjusts the chemical production device to the optimization objective value according to the calculated overall optimization objective value. Meanwhile, as the actual industrial process application is more real-time optimization based on a steady-state mechanism model, the industrial process application needs the process to be in a steady state or approximately steady state, and the problems of long calculation time and easy non-convergence exist. Therefore, for a typical chemical production device such as an ethylene glycol unit, due to the characteristics of nonlinearity, strong coupling property and high energy consumption, the existing real-time optimization technology cannot realize better process control on the ethylene glycol unit.

In view of the above problems, no effective solution has been proposed at present.

Disclosure of Invention

The embodiment of the application provides a full-flow dynamic optimization control method, a full-flow dynamic optimization control device and electronic equipment for a chemical production device, and aims to at least solve the technical problems that the time for calculating an optimization process variable of the chemical production device by a real-time optimization technology based on a steady-state mechanism model is too long and the optimization process variable is not easy to converge.

According to an aspect of the embodiment of the application, a full-flow dynamic optimization control method for a chemical production device is provided, which comprises the following steps: obtaining a target parameter set of the chemical production device in the production process, wherein the target parameter set at least comprises: operating variable parameters, controlled variable parameters and disturbance variable parameters; constructing a nonlinear steady-state model of the chemical production device based on the target parameter set, and determining a nonlinear gain matrix of the controlled variable parameter based on the nonlinear steady-state model; constructing a dynamic linear prediction model according to the operation variable parameters and the controlled variable parameters, and correcting the dynamic linear prediction model through a nonlinear gain matrix of the controlled variable parameters to obtain a full-flow nonlinear dynamic optimization model; performing online data setting processing on the target parameter set, solving the full-flow nonlinear dynamic optimization model in real time based on the processed target parameter set under the limitation of an optimization target and an optimization constraint condition to obtain a dynamic optimization result, and transmitting the dynamic optimization result to an advanced controller for execution.

Optionally, the target parameter set further comprises: the product property parameter, wherein, obtain the target parameter set of chemical production device in the production process, include: obtaining product property parameters produced by the chemical production device from an online analysis database, wherein the product property parameters comprise at least one of the following: product density, product content, moisture content, acidity; acquiring operation variable parameters and controlled variable parameters of the chemical production device from a real-time database, wherein the operation variable parameters comprise at least one of the following: the controlled variable parameters comprise at least one of the following: purity, reaction rate, pH, liquid level, and material concentration.

Optionally, constructing a nonlinear steady-state model of the chemical production device based on the target parameter set, and determining a nonlinear gain matrix of the controlled variable parameter based on the nonlinear steady-state model, including: determining a preset artificial intelligence learning model, wherein the artificial intelligence learning model comprises at least one of the following: a deep learning model, a neural network model, and a BOOST model; inputting the operation variable parameters and the disturbance variable parameters as input features into an artificial intelligence learning model to obtain a prediction result of artificial intelligence learning output; adjusting model parameters of the artificial intelligent learning model based on the prediction result and the controlled variable parameters to obtain a nonlinear steady-state model of the chemical production device; and performing bias guide on the operation variable parameters through a nonlinear steady-state model to obtain a nonlinear gain matrix of the controlled variable parameters.

Optionally, constructing a dynamic linear prediction model according to the operation variable parameter and the controlled variable parameter includes: determining transfer functions of the operation variable parameters and the controlled variable parameters, and identifying the transfer functions to obtain a dynamic prediction matrix; and establishing a dynamic linear prediction model based on the dynamic prediction matrix.

Optionally, performing online data setting processing on the target parameter set, including performing a preprocessing operation on the target parameter set, where the preprocessing operation includes at least one of: consistency detection, filling of missing values and normalization treatment; and carrying out online data setting processing and model parameter estimation on the preprocessed target parameter set, wherein the model parameter estimation is used for estimating model parameters of a dynamic material balance model, and the dynamic material balance model is used for balancing the material relation between input variables and output variables of the chemical production device.

Optionally, performing online data tuning processing and model parameter estimation on the preprocessed target parameter set, including: performing online data setting processing on the preprocessed target parameter set, and constructing a dynamic material balance model based on the processed target parameter set; determining a dynamic material balance constraint, wherein the dynamic material balance constraint comprises at least one of: material balance equation, energy balance equation, device connection equation, physical property calculation equation and molecular normalization equation; based on dynamic material balance constraint conditions, establishing a target setting function by taking the minimum model error of a dynamic material balance model as a target; and solving the target setting function by adopting a nonlinear optimizer to obtain model parameters of the dynamic material balance model.

Optionally, based on the processed target parameter set, under the limitation of an optimization target and an optimization constraint condition, solving the full-flow nonlinear dynamic optimization model in real time to obtain a dynamic optimization result, including: determining an optimization constraint condition, wherein the optimization constraint condition comprises at least one of the following: dynamic material balance constraint conditions, first upper and lower limits of operation variable parameters, second upper and lower limits of first increment of operation variable, third upper and lower limits of controlled variable parameters, fourth upper and lower limits of second increment of controlled variable, fifth upper and lower limits of corrected controlled variable parameters and dynamic material balance constraint conditions; determining a relaxation variable corresponding to the third upper and lower limits of the controlled variable parameter and the corrected fifth upper and lower limits of the controlled variable parameter, and expanding the third upper and lower limits and the fifth upper and lower limits based on the relaxation variable; determining an optimization target, and determining an optimization objective function of the full-flow nonlinear dynamic optimization model based on the processed target parameter set under the limitation of the optimization target, wherein the optimization target comprises at least one of the following: the yield is maximized and the consumption is minimized; and solving an optimization objective function by using a nonlinear programming solver in a feasible region defined by the adjusted optimization constraint condition to obtain a dynamic optimization result, wherein the dynamic optimization result comprises at least one of the following components: a first target dynamic running track and a first target steady state value of the operating variable parameter, and a second target dynamic running track and a second target steady state value of the controlled variable parameter.

Optionally, issuing the dynamic optimization result to the advanced controller for execution, including: and controlling the operation variable parameters to be adjusted to a first target steady-state value according to the first target dynamic running track through the advanced controller, and controlling the controlled variable parameters to be adjusted to a second target steady-state value according to the second target dynamic running track.

According to another aspect of the embodiment of the present application, there is also provided a full-process dynamic optimization control device for a chemical production device, including: the acquisition module is used for acquiring a target parameter set of the chemical production device in the production process, wherein the target parameter set at least comprises: operating variable parameters, controlled variable parameters and disturbance variable parameters; the first construction module is used for constructing a nonlinear steady-state model of the chemical production device based on the target parameter set and determining a nonlinear gain matrix of the controlled variable parameter based on the nonlinear steady-state model; the second construction module is used for constructing a dynamic linear prediction model according to the operation variable parameters and the controlled variable parameters, and correcting the dynamic linear prediction model through a nonlinear gain matrix of the controlled variable parameters to obtain a full-flow nonlinear dynamic optimization model; and the optimization control module is used for carrying out online data setting processing on the target parameter set, solving the full-flow nonlinear dynamic optimization model in real time under the limitation of an optimization target and an optimization constraint condition based on the processed target parameter set to obtain a dynamic optimization result, and transmitting the dynamic optimization result to the advanced controller for execution.

According to another aspect of the embodiments of the present application, there is also provided an electronic device including: the system comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the whole-flow dynamic optimization control method of the chemical production device through the computer program.

In this embodiment of the application, a target parameter set of a chemical production device in a production process is obtained, where the target parameter set at least includes: operating variable parameters, controlled variable parameters and disturbance variable parameters; constructing a nonlinear steady-state model of the chemical production device based on the target parameter set, and determining a nonlinear gain matrix of the controlled variable parameter based on the nonlinear steady-state model; constructing a dynamic linear prediction model according to the operation variable parameters and the controlled variable parameters, and correcting the dynamic linear prediction model through a nonlinear gain matrix of the controlled variable parameters to obtain a full-flow nonlinear dynamic optimization model; and carrying out online data setting processing on the target parameter set, solving the full-flow nonlinear dynamic optimization model in real time based on the processed target parameter set under the limitation of an optimization target and an optimization constraint condition to obtain a dynamic optimization result, and transmitting the dynamic optimization result to an advanced controller for execution.

In the technical scheme, the optimization solution is calculated through the full-flow nonlinear dynamic optimization model of the Wiener structure based on the combination of the nonlinear steady-state model and the dynamic linear prediction model, so that the solving speed is higher, the optimization period is shorter, and the technical problems that the time for calculating the optimization process variable of the chemical production device by the related real-time optimization technology based on the steady-state mechanism model is too long and is not easy to converge are solved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:

FIG. 1 is a flow diagram of an alternative overall flow dynamic optimization control method according to an embodiment of the present application;

FIG. 2 is a schematic diagram of an alternative coal-to-ethylene glycol unit plant overall flow dynamic optimization control in accordance with an embodiment of the present application;

FIG. 3 is a schematic diagram of an alternative coal-to-ethylene glycol unit plant overall process dynamic optimization system in accordance with an embodiment of the present application;

FIG. 4 is a schematic diagram of an alternative coal-to-ethylene glycol unit apparatus overall flow dynamic optimization in accordance with an embodiment of the present application;

Fig. 5 is a schematic structural diagram of an alternative full-flow dynamic optimization control device according to an embodiment of the present application.

Detailed Description

In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

In addition, the related information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for presentation, analyzed data, etc.) referred to in this application are information and data authorized by the user or sufficiently authorized by the parties. For example, an interface is provided between the system and the relevant user or institution, before acquiring the relevant information, the system needs to send an acquisition request to the user or institution through the interface, and acquire the relevant information after receiving the consent information fed back by the user or institution.

For better understanding of the embodiments of the present application, technical terms related in the embodiments of the present application are explained below:

advanced process control a control method and technique for optimizing and improving an industrial process. Advanced mathematical models, optimization algorithms and control strategies are adopted, and real-time data acquisition and analysis are combined, so that the accurate control and optimization of the industrial process are realized. In general, the technology is widely applied to the fields of complex industrial processes, such as chemical industry, petroleum, pharmacy, electric power and the like. The method can carry out joint control on a plurality of parameters, and realizes optimized production scheduling and process management.

DCS (Distributed Control System ) is a computer technology system for automation control of industrial production processes, comprising the following parts: a Control Station (Control Station), also known as a process Control unit (Process Control Unit, PCU), is a core part of the DCS, responsible for executing process Control algorithms and logic controls, typically including Control processors, I/O modules, and operator workstations, etc.; an Operator workstation (Operator Station) provides a human-computer interface for a system Operator, and the Operator can monitor the state of the whole process through a graphic display screen, a keyboard, a mouse and other devices to carry out parameter setting and control command issuing; engineer workstation (Engineering Station): for system engineers, it is mainly used for configuration, programming, maintenance and fault diagnosis of the system; the communication network is used for connecting each control station, each operator workstation, each engineer workstation and each field device to realize data transmission and information sharing; the field devices include sensors, actuators, transmitters, etc. that are directly connected to physical variables (e.g., temperature, pressure, flow, fluid level, etc.) in the industrial process, transmit field data to the DCS system, or receive control commands from the DCS system to adjust process parameters. Therefore, DCS can combine computer technology, control technology and communication technology to achieve monitoring, management and optimization of production processes for large industrial devices such as petroleum refining, chemical, electrical, metallurgical, environmental, pharmaceutical, food, etc.

Example 1

The process for preparing glycol from coal is also called oxalate method, and uses coal as raw material, and adopts the processes of gasification, conversion, purification, separation and purification to obtain CO and CO respectivelyWherein, CO is synthesized and refined to produce oxalate through catalytic coupling, and then is combined with +.>And (3) carrying out hydrogenation reaction and refining to obtain the polyester-grade glycol. The realization device of the coal glycol process, namely a glycol unit device, has nonlinear and strong coupling characteristics and has higher energy consumption. In order to reduce the production energy consumption and optimize and improve the yield of the ethylene glycol, the stable and optimized operation of the ethylene glycol unit device is a core target of enterprise production.

At present, in the foreign oil refining and chemical industry, a process online optimization technology and software have obtained a plurality of successfully applied examples, so that the method is not only used for single-device optimization (such as a rectifying tower) but also used for full-process dynamic optimization, and the examples of the industrial application are said to bring about economic benefit improvement of 3% -20% to enterprises. With the great popularity of advanced process control technologies, enterprise users have not satisfied smooth operation of single devices, but hope to dig greater economic benefit profit to improve market competitiveness on the premise of ensuring safe and stable operation of single devices or multiple devices, and thus, urgent demands are put on real-time optimization/dynamic optimization.

Real-time online optimization systems are typically built on advanced controllers, which typically calculate overall optimization target values from overall optimization objective functions contained in the real-time online optimization system and communicate the overall optimization target values to advanced controllers, which will adjust the production plant to the optimal target values based on these target values. The overall optimization objective of the real-time online optimization system can be yield maximization, yield maximization or economic benefit maximization, etc. Because of the nonlinear and strong coupling characteristics of industrial devices such as ethylene glycol unit devices, when searching an optimization target, the real-time online optimization system can overcome the nonlinearity of the system by adopting the technologies such as professional process calculation packages, process simulation systems, online gain updating and the like.

The actual industrial process is more currently applied to real-time optimization based on a mechanism steady-state model, but the industrial application requires that the production process tends to be steady-state or approximately steady-state, and has the defects of long calculation time and easy non-convergence. In order to dynamically track and optimize a process variable, a related patent (application number: 202110718645.9) relates to a chemical dynamic optimization problem sea-gull optimization method, a system and computer equipment, and provides a chemical dynamic optimization problem mixed sea-gull optimization method, a system and computer equipment. However, the actual industrial field dynamic mechanism model is complex to construct, the solving efficiency is low, the application instantaneity and the application effectiveness cannot be guaranteed, and meanwhile, the patent is not combined with a controller, and only one optimization problem is simply solved. In addition, the patent (application number: 201310419751.2) proposes that a certain amount of simulation data is generated by simulating the mechanism of the industrial ethylene cracking furnace and utilizing an experimental design principle, then modeling is carried out by utilizing a neural network proxy model, and the constructed continuous dynamic optimization model is approximately solved by utilizing a linear segmentation method. However, because the actual field working condition is complex and has large change, the method cannot ensure the reliability of the model and the applicability of the model when the working condition changes, and meanwhile, the agent model based on the neural network is not consistent with the advanced control system model, so that the problems of gain violation, misleading operation and the like are easy to occur.

In order to solve the problem, a solution of a full-flow dynamic optimization control method is provided in the embodiments of the present application, and is described in detail below. It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.

Fig. 1 is a flow chart of an alternative overall-flow dynamic optimization control method for a chemical production device according to an embodiment of the present application, as shown in fig. 1, the method at least includes steps S102-S108, where:

step S102, a target parameter set of the chemical production device in the production process is obtained.

Wherein the target parameter set at least comprises: operating variable parameters, controlled variable parameters, disturbance variable parameters, and product property parameters.

As an optional implementation manner, in the technical solution provided in step S102, the method may further include: obtaining product property parameters produced by the chemical production device from an online analysis database, wherein the product property parameters comprise at least one of the following: product density, product content, moisture content, acidity; acquiring operation variable parameters and controlled variable parameters of the chemical production device from a real-time database, wherein the operation variable parameters comprise at least one of the following: the controlled variable parameters comprise at least one of the following: purity, reaction rate, pH, liquid level, and material concentration.

In this embodiment, when the chemical production apparatus is a coal-based ethylene glycol unit apparatus, an initial parameter set (i.e., a target parameter set) of the ethylene glycol unit apparatus may be acquired from a DCS distributed control system/real-time database, an online analysis database, a price database, etc., through a data integration engine, and stored in a data warehouse. Wherein the initialization parameters include at least one of the following: product properties of the ethylene glycol unit device obtained from the online analytical database, for example: density of polyester grade/technical grade ethylene glycol, ethylene glycol content, density of methanol, distillation range temperature, moisture content, acidity, etc.; the operation conditions of the devices such as the ethylene glycol synthesizer, the rectifying tower, the recovery tower, the dehydration tower, the dealcoholization tower, the product tower, the concentration tower and the like obtained from the DCS distributed control system/the real-time database are as follows: operating variable parameter set: dimethyl oxalate flow, hydrogen flow, synthetic fresh hydrogen/ester ratio, synthetic recycle gas outlet temperature, tower bottom temperature of each tower; controlled variable parameter set: steam flow, reflux flow, tower kettle extraction, tower kettle liquid level, light/heavy dihydric alcohol flow, polyester grade/industrial grade glycol flow, methanol outlet flow and the like.

And step S104, constructing a nonlinear steady-state model of the chemical production device based on the target parameter set, and solving nonlinear gains of the controlled variable parameters based on the nonlinear steady-state model.

As an alternative embodiment, in the technical solution provided in the step S104, the method may include the following steps S1041 to S1044:

step S1041, determining a preset artificial intelligence learning model, where the artificial intelligence learning model includes at least one of the following: a deep learning model, a neural network model, and a BOOST model;

step S1042, the operation variable parameter and the disturbance variable parameter are used as input characteristics to be input into an artificial intelligent learning model to obtain a prediction result of artificial intelligent learning output;

step S1043, adjusting model parameters of the artificial intelligent learning model based on the prediction result and the controlled variable parameters to obtain a nonlinear steady-state model of the chemical production device;

step S1044, obtaining a nonlinear gain matrix of the controlled variable parameter by performing bias derivative on the operation variable parameter through a nonlinear steady-state model.

In the embodiment, a nonlinear steady-state model of the whole flow of the coal-to-ethylene glycol unit device is established based on an artificial intelligence learning model, wherein the expression of the model is as follows:

Wherein,representing a nonlinear steady state model, which may represent a nonlinear steady state model associated with plant throughput, yield or quality, or may represent a conventional process variable relationship model,/->An operating variable indicating when the device is in or tending to stabilize,/->Disturbance variable indicating when the device is in or tending to stabilize, +.>The controlled variable representing the steady state or tending to steady state of the device, such as the device throughput, yield, quality controlled variable or conventional controlled variable, and a represents the model adjustable parameter for online model updating.

It should be noted that, the nonlinear steady-state model based on the artificial intelligence learning model may include the following two modeling types: the first type is a mixed model, namely, the mixed training of an artificial intelligent learning model is carried out by adopting historical data generated by a mechanism model of a coal glycol unit device; and secondly, a data model is obtained by performing model training by adopting historical data of a coal glycol unit device.

After obtaining the nonlinear steady-state model, byAsk for->The bias of (3) can obtain a nonlinear steady-state gain model, and the expression can be written as:

wherein,and representing a nonlinear gain matrix at steady state for compensating the error of the nonlinear steady state model. This is because in a nonlinear steady-state model, nonlinear deviations in the system response, i.e., errors between the model output and the actual output, may occur. By calculating the nonlinear gain matrix, the input to the system can be adjusted according to the characteristics of the error to minimize the error. This may improve the stability and accuracy of the model. In addition, a nonlinear gain matrix is obtained >Then, the gain deviation compensation of the controlled variable parameter can be determined by the following expression:

wherein,gain bias compensation indicative of a controlled variable parameter, < ->A nonlinear gain matrix representing the advanced controller for compensating for errors in the dynamic linear prediction model. Also, this is because in a dynamic linear prediction model, the system response may be affected by external disturbances or model errors, resulting in a deviation between the model output and the actual output, byThe APC linear gain matrix is calculated and the output of the control system can be adjusted based on the characteristics of the error to minimize the error. This can improve the response speed and robustness of the model. The accuracy, stability, response speed and robustness of the full-flow nonlinear dynamic optimization model can be improved by calculating gain deviation compensation based on the nonlinear gain matrix of the advanced controller and the nonlinear gain matrix of the controlled variable parameter in steady state.

And S106, constructing a dynamic linear prediction model according to the operation variable parameters and the controlled variable parameters, and correcting the dynamic linear prediction model through a nonlinear gain matrix of the controlled variable parameters to obtain a full-flow nonlinear dynamic optimization model.

As an alternative embodiment, the above process may include: determining transfer functions of the operation variable parameters and the controlled variable parameters, and identifying the transfer functions to obtain a dynamic prediction matrix; and establishing a dynamic linear prediction model based on the dynamic prediction matrix.

Specifically, a transfer function and a step response coefficient of an operation variable parameter and a controlled variable parameter are obtained based on identification of a control model of an advanced controller of the coal glycol unit device, and a dynamic prediction matrix of the glycol unit device is established according to the transfer function and the step response coefficient, wherein the expression is as follows:

furthermore, a dynamic linear prediction model between a prediction time domain and a control time domain can be constructed according to the dynamic prediction matrix, and the model expression is as follows:

meanwhile, the constraint conditions of the dynamic linear prediction model can be written as:

wherein P represents the prediction time domain andm represents the control time domain and +.>，

Representing the i-th operating variable,represents the j-th controlled variable, +.>Representing p-step offset correction, ">An increment of an operating variable that indicates when the device is at or tending to stabilize.

Nonlinear gain by the controlled variable parameter determined in step S104 Correcting the dynamic linear prediction model to obtain a corrected full-flow nonlinear dynamic optimization model, wherein the full-flow nonlinear dynamic modelThe state optimization model can obtain corrected controlled variable parameters:

wherein,representing the corrected controlled variable parameter.

And S108, performing online data setting processing on the target parameter set, solving the full-flow nonlinear dynamic optimization model in real time under the limitation of an optimization target and an optimization constraint condition based on the processed target parameter set to obtain a dynamic optimization result, and transmitting the dynamic optimization result to an advanced controller for execution.

Wherein, the optimization target can be the highest yield and the lowest energy consumption; such optimization constraints include, but are not limited to: all the upper and lower limits of the operation variable, the upper and lower limits of the increment of the controlled variable, the constraint condition of the dynamic prediction model, the constraint condition of the nonlinear steady-state model and the like.

As an optional implementation manner, in the technical solution provided in step S108, performing online data setting processing on the target parameter set may include: performing a preprocessing operation on the target parameter set, wherein the preprocessing operation comprises at least one of the following steps: consistency detection, filling of missing values and normalization treatment; and carrying out online data setting processing and model parameter estimation on the preprocessed target parameter set, wherein the model parameter estimation is used for estimating model parameters of a dynamic material balance model, and the dynamic material balance model is used for balancing the material relation between input variables and output variables of the chemical production device.

In this embodiment, automatic data consistency monitoring and online preprocessing of data are performed on an initial parameter set (i.e., a target parameter set) stored in a data warehouse to ensure reliability and validity of the data, wherein the data consistency detection means that the initial parameter set of the data warehouse is ensured to exist and be valid through parameter matching so as to prevent the loss of data transmission process variables; such data on-line preprocessing includes, but is not limited to: filling the missing value, removing the outlier point by a 3-delta method, removing noise by exponential smoothing filtering and the like. And then the data consistency detection and the effective data set after the online pretreatment of the data (namely the target parameter set after the pretreatment) are used for carrying out dynamic data setting and model parameter estimation, so that the quality of measured data and the high precision and reliability of an optimized model can be ensured.

The online data setting processing and model parameter estimation are to utilize the relations of material balance or energy balance and the like of the ethylene glycol unit device in the production process, perform data setting on an original measurement data set (namely a target data set) on the basis of the original measurement data set, and estimate the non-measurable variable and the model adjustable parameter in the production process, so that the square sum of deviation between the measurement data set and the original measurement data set after the setting is required to be minimum, the deviation brought by the obvious error of an actual measuring instrument to the model is overcome, and the set data set can better maintain the material balance relation.

Optionally, performing dynamic data setting and model parameter estimation on the preprocessed target parameter set may include the following procedures: firstly, carrying out online data setting treatment on a target parameter set after pretreatment, and constructing a dynamic material balance model based on the target parameter set after treatment; next, a dynamic material balance constraint is determined, wherein the dynamic material balance constraint includes, but is not limited to: material balance equation, energy balance equation, device connection equation, physical property calculation equation and molecular normalization equation; then, based on dynamic material balance constraint conditions, a target setting function is established by taking the minimum model error of a dynamic material balance model as a target; and finally, solving the target setting function by adopting a nonlinear optimizer to obtain model parameters of the dynamic material balance model.

Specifically, the set of operating variables is denoted asThe controlled variable set is denoted +.>Thus, the dynamic physical balance model can be written as:

wherein, the aboveThe dynamic simplified mechanism model of the ethylene glycol unit device is represented and consists of a material balance equation (used for describing the balance relation between the inlet and outlet flow rates of materials in a chemical production device so as to ensure that the total amount of the materials in the device does not change), an energy balance equation (used for describing the balance relation between the inlet and outlet flow rates of the energy in the chemical production device so as to ensure that the total amount of the energy in the device does not change), a connection equation (used for describing the material flow relation between the chemical production device and comprising feeding, discharging and circulating flow so as to ensure the material flow balance between the devices), a physical property calculation equation (used for calculating physical property parameters of the materials in the chemical production device, such as density, viscosity and the like so as to calculate other equations), a molecular normalization equation (used for describing the change relation of the material components in the chemical production device and knowing the distribution of different components in the device) and the like. / >Model-adjustable parameters representing a simplified mechanism model, < +.>Represents the set of measurable variables after tuning, < ->Representing the estimated non-measurable variable parameter.

Dynamic material balance constraints include, but are not limited to, the following:

wherein, the above、/>、/>、/>、/>、/>、/>、/>The upper limit and the lower limit of the controlled variable, the upper limit and the lower limit of the controlled variable increment and the upper limit and the lower limit of the controlled variable increment are respectively expressed.

Based on dynamic material balance constraint conditions, a target setting function of dynamic data setting and parameter estimation is established by taking the minimum model error of a dynamic material balance model as a target, and the expression is as follows:

wherein,representing the minimization of the target setting function +.>、/>All of which represent the weights after the setting,representing the variance of the set variable, +.>Representing the actual measurable variable, +.>Representing control increment variable,/->Representing a dynamic gain matrix.

Then a nonlinear optimizer is applied to solve the target setting function to obtain the adjustable model parameters of the dynamic material balance modelAn undesireable variable after setting +.>And the set measurable variable +.>。

Further, an optimization constraint is determined, wherein the optimization constraint includes at least one of: dynamic material balance constraint conditions, first upper and lower limits of operation variable parameters, second upper and lower limits of first increment of operation variable, third upper and lower limits of controlled variable parameters, fourth upper and lower limits of second increment of controlled variable, fifth upper and lower limits of corrected controlled variable parameters and dynamic material balance constraint conditions; determining a relaxation variable corresponding to the third upper and lower limits of the controlled variable parameter and the corrected fifth upper and lower limits of the controlled variable parameter, and expanding the third upper and lower limits and the fifth upper and lower limits based on the relaxation variable; determining an optimization target, and determining an optimization objective function of the full-flow nonlinear dynamic optimization model based on the processed target parameter set under the limitation of the optimization target, wherein the optimization target comprises at least one of the following: the yield is maximized and the consumption is minimized; and solving an optimization objective function by using a nonlinear programming solver in a feasible region defined by the adjusted optimization constraint condition to obtain a dynamic optimization result, wherein the dynamic optimization result comprises at least one of the following components: a first target dynamic running track and a first target steady state value of the operating variable parameter, and a second target dynamic running track and a second target steady state value of the controlled variable parameter.

In the embodiment of the application, a set target data set is input into a full-flow nonlinear dynamic optimization model of an ethylene glycol unit device, ethylene glycol yield maximization and steam consumption minimization are used as optimization targets, and simultaneously, optimal operation parameter settings such as dimethyl oxalate flow, hydrogen flow, synthetic fresh hydrogen/ester ratio, synthetic recycle gas outlet temperature, each tower kettle temperature, steam flow, reflux flow, tower kettle extraction amount, tower kettle liquid level, light/heavy dihydric alcohol flow, polyester grade/industrial grade ethylene glycol flow, methanol outlet flow product quality requirements and the like are considered, and optimal operation parameters such as optimal synthetic fresh hydrogen/ester ratio, each rectifying tower reflux ratio, tower kettle temperature and the like are obtained on line in real time in a feasible range determined by the optimization constraint conditions by using a nonlinear programming solver.

Specifically, with the objective of maximizing ethylene glycol yield and minimizing steam consumption, the following optimization objective function was determined:

wherein,、/>price coefficients (which can be positive or negative) representing the operating variable and the controlled variable, respectively>、/>Weighting factors representing the operating variable parameter and the controlled variable parameter, respectively, +. >Steady state values representing operating variable parameters +.>、/>Indicating the tracking set point.

In addition, the optimization constraints include, but are not limited to, the following:

dynamic material balance constraint conditions;

upper and lower limits of the operating variable parameters:

upper and lower limits of controlled variable parameters:

upper and lower limits of the manipulated variable parameter delta:

upper and lower limits of controlled variable parameter increment:

；

upper and lower limits of the corrected controlled variable parameter:

constraint conditions of dynamic prediction model:

/>

constraint conditions of nonlinear steady-state model:

meanwhile, in order to ensure feasible optimization solution, the upper limit and the lower limit of the controlled variable parameter are calculated、/>Introducing relaxation variables +.>Wherein, the original upper and lower limit constraint can be converted into equality constraint by introducing a relaxation variable, so that the upper and lower limit constraint condition is better processed, namely the mathematical description of the problem is simplified, and the problem is convenient to understand and solve. Thus, the upper and lower limits of the adjusted controlled variable parameter delta and the upper and lower limits of the corrected controlled variable parameter can be written as:

finally, non-linearity is adoptedThe planning solver is used for solving the overall process optimization objective function and constraint conditions of the coal glycol unit device process by applying the IPOPT solver to obtain the optimal dynamic running track of the operation variable parameters and the controlled variable parameters And->And an optimal steady state value +.>And->The IPOPT solver gradually improves the value of the optimization variable in an iterative mode and calculates the value of the optimization objective function.

Optionally, when the dynamic optimization result is issued to the advanced controller for execution, the advanced controller can control the operation variable parameter to be adjusted to a first target steady-state value according to the first target dynamic running track, and control the controlled variable parameter to be adjusted to a second target steady-state value according to the second target dynamic running track.

That is, in the process of obtaining the optimal dynamic running track of the operation variable parameter and the controlled variable parameterAndand an optimal steady state value +.>And->Then, the optimal dynamic running track of the operation variable parameter and the controlled variable parameter is +.>And->And an optimal steady state value +.>And->The method is characterized in that the method is used as the set values of the operation variable and the controlled variable of the advanced controller in real time and is automatically downloaded to each advanced controller of the ethylene glycol production device to be executed, and in each dynamic optimization period, the advanced controller can ensure that the ethylene glycol production device is stably and quickly closed to the optimal steady state value along the optimal dynamic running track, so that the online closed loop dynamic real-time optimization of the ethylene glycol production device is realized, the device is enabled to run to achieve the maximum ethylene glycol yield and the minimum steam energy consumption, and the economic benefit of the production of the ethylene glycol unit device is further improved.

As an alternative implementation manner, fig. 2 is a schematic diagram of overall-flow dynamic optimization control of an alternative coal-based ethylene glycol production device according to an embodiment of the present application, and as shown in fig. 2, the whole ethylene glycol unit device includes 11 towers, 11 advanced controllers, and total operating variables and controlled variables are 78. In addition, the objective function includes: minimizing energy consumption, maximizing quality priority and maximizing ethylene glycol/methanol yield; constraints include, but are not limited to: the property of the methanol meets the requirements of DMO feeding quality, the ethylene glycol meets the requirements of superior quality, the upper and lower limits of operating variables, the upper and lower limits of controlled variables and the like. By adopting the full-flow dynamic optimization control method provided by the embodiment of the application, the coal glycol unit device of fig. 2 is subjected to full-flow dynamic optimization according to the optimization system architecture shown in fig. 3 and the optimization principle shown in fig. 4, the average optimization period can be realized for 5 minutes, the glycol unit device is subjected to dynamic optimization, the energy consumption is ensured to be reduced by 2%, the glycol yield is improved by 1%, and obvious economic benefits are obtained.

Based on the scheme defined in the steps S102 to S108, it can be known that, in the embodiment, a nonlinear steady-state model based on a big data model or a hybrid model is adopted and fused with a dynamic linear model of each advanced controller in the chemical production device to form a full-flow dynamic optimization model, so that a strict mechanism model is not required to be established, and optimization can be performed without the production process being in a steady state, that is, the scheme has stronger applicability, and the problems of slow calculation speed and incapability of ensuring convergence are avoided; in addition, the Wiener type nonlinear model (namely the full-flow nonlinear dynamic optimization model) combining the nonlinear steady-state model and the dynamic linear prediction model is adopted in the embodiment of the application, the solving speed is higher, the optimization period is also faster, and meanwhile, the Wiener type nonlinear model and the advanced controller execute at the same frequency, so that the real-time dynamic optimization is realized, and the problem that the optimization and the control cannot be executed simultaneously due to inconsistent models is avoided. And further, the technical problems that the time for calculating the optimized process variable of the chemical production device by the related real-time optimization technology based on the steady-state mechanism model is too long and the chemical production device is not easy to converge are solved.

Example 2

Based on embodiment 1 of the present application, an embodiment of a full-flow dynamic optimization control device for a chemical production device is also provided, and the device executes the full-flow dynamic optimization control method for the chemical production device in the embodiment when running. Fig. 5 is a schematic structural diagram of an alternative overall-process dynamic optimization control device for a chemical production device according to an embodiment of the present application, where, as shown in fig. 5, the overall-process dynamic optimization control device for a chemical production device at least includes an obtaining module 51, a first building module 53, a second building module 55 and an optimization control module 57, where:

the obtaining module 51 is configured to obtain a target parameter set of the chemical production device during a production process, where the target parameter set at least includes: an operating variable parameter, a controlled variable parameter, a disturbance variable parameter.

Optionally, the target parameter set further includes a product property parameter.

As an alternative embodiment, the obtaining module 51 is further configured to obtain, from the online analysis database, a product property parameter produced by the chemical production device, where the product property parameter includes at least one of the following: product density, product content, moisture content, acidity; acquiring operation variable parameters and controlled variable parameters of the chemical production device from a real-time database, wherein the operation variable parameters comprise at least one of the following: the controlled variable parameters comprise at least one of the following: purity, reaction rate, pH, liquid level, and material concentration.

When the chemical production device is a coal glycol unit device, an initial parameter set (namely a target parameter set) of the glycol unit device can be obtained from a DCS distributed control system/real-time database, an online analysis database, a price database and the like through a data integration engine and stored in a data warehouse. Wherein the initialization parameters include at least one of the following: product properties of the ethylene glycol unit device obtained from the online analytical database, for example: density of polyester grade/technical grade ethylene glycol, ethylene glycol content, density of methanol, distillation range temperature, moisture content, acidity, etc.; the operation conditions of the devices such as the ethylene glycol synthesizer, the rectifying tower, the recovery tower, the dehydration tower, the dealcoholization tower, the product tower, the concentration tower and the like obtained from the DCS/real-time database are as follows: operating variable parameter set: dimethyl oxalate flow, hydrogen flow, synthetic fresh hydrogen/ester ratio, synthetic recycle gas outlet temperature, tower bottom temperature of each tower; controlled variable parameter set: steam flow, reflux flow, tower kettle extraction, tower kettle liquid level, light/heavy dihydric alcohol flow, polyester grade/industrial grade glycol flow, methanol outlet flow and the like.

The first construction module 53 is configured to construct a nonlinear steady-state model of the chemical production device based on the target parameter set, and determine a nonlinear gain matrix of the controlled variable parameter based on the nonlinear steady-state model.

Optionally, the first construction module 53 is further configured to determine a preset artificial intelligence learning model, where the artificial intelligence learning model includes at least one of: a deep learning model, a neural network model, and a BOOST model; inputting the operation variable parameters and the disturbance variable parameters as input features into an artificial intelligence learning model to obtain a prediction result of artificial intelligence learning output; adjusting model parameters of the artificial intelligent learning model based on the prediction result and the controlled variable parameters to obtain a nonlinear steady-state model of the chemical production device; and performing bias guide on the operation variable parameters through a nonlinear steady-state model to obtain a nonlinear gain matrix of the controlled variable parameters.

The second construction module 55 is configured to construct a dynamic linear prediction model according to the operation variable parameter and the controlled variable parameter, and correct the dynamic linear prediction model through a nonlinear gain matrix of the controlled variable parameter to obtain a full-flow nonlinear dynamic optimization model.

Optionally, the second construction module 55 is further configured to determine transfer functions of the operation variable parameter and the controlled variable parameter, and identify the transfer functions to obtain a dynamic prediction matrix; and establishing a dynamic linear prediction model based on the dynamic prediction matrix.

The optimization control module 57 is configured to perform online data setting processing on the target parameter set, solve the full-flow nonlinear dynamic optimization model in real time under the limitation of the optimization target and the optimization constraint condition based on the processed target parameter set to obtain a dynamic optimization result, and send the dynamic optimization result to the advanced controller for execution.

Optionally, the optimization control module 57 further includes: a data processing unit 571, a solving unit 572, wherein,

a data processing unit 571, configured to perform a preprocessing operation on the target parameter set, where the preprocessing operation includes at least one of: consistency detection, filling of missing values and normalization treatment; and carrying out online data setting processing and model parameter estimation on the preprocessed target parameter set, wherein the model parameter estimation is used for estimating model parameters of a dynamic material balance model, and the dynamic material balance model is used for balancing the material relation between input variables and output variables of the chemical production device.

In addition, the data processing unit 571 is further configured to perform online data setting processing on the preprocessed target parameter set, and construct a dynamic material balance model based on the processed target parameter set; determining a dynamic material balance constraint, wherein the dynamic material balance constraint comprises at least one of: material balance equation, energy balance equation, device connection equation, physical property calculation equation and molecular normalization equation; based on dynamic material balance constraint conditions, establishing a target setting function by taking the minimum model error of a dynamic material balance model as a target; and solving the target setting function by adopting a nonlinear optimizer to obtain model parameters of the dynamic material balance model.

The solving unit 572 is configured to determine an optimization constraint, where the optimization constraint includes at least one of: dynamic material balance constraint conditions, first upper and lower limits of operation variable parameters, second upper and lower limits of first increment of operation variable, third upper and lower limits of controlled variable parameters, fourth upper and lower limits of second increment of controlled variable, fifth upper and lower limits of corrected controlled variable parameters and dynamic material balance constraint conditions; determining a relaxation variable corresponding to the third upper and lower limits of the controlled variable parameter and the corrected fifth upper and lower limits of the controlled variable parameter, and expanding the third upper and lower limits and the fifth upper and lower limits based on the relaxation variable; determining an optimization target, and determining an optimization objective function of the full-flow nonlinear dynamic optimization model based on the processed target parameter set under the limitation of the optimization target, wherein the optimization target comprises at least one of the following: the yield is maximized and the consumption is minimized; and solving an optimization objective function by using a nonlinear programming solver in a feasible region defined by the adjusted optimization constraint condition to obtain a dynamic optimization result, wherein the dynamic optimization result comprises at least one of the following components: a first target dynamic running track and a first target steady state value of the operating variable parameter, and a second target dynamic running track and a second target steady state value of the controlled variable parameter.

Further, when the dynamic optimization result is issued to the advanced controller for execution, the optimization control module 57 may control the operation variable parameter to be adjusted to the first target steady-state value according to the first target dynamic running track by the advanced controller, and control the controlled variable parameter to be adjusted to the second target steady-state value according to the second target dynamic running track.

In the embodiment of the application, the nonlinear steady-state model based on the big data model or the mixed model is adopted to be fused with the dynamic linear model of each advanced controller in the chemical production device to form a full-flow dynamic optimization model, so that a strict mechanism model is not required to be established, optimization can be executed without the production process being in a steady state, namely, the applicability of the scheme is stronger, and the problems that the calculation speed is low and convergence cannot be guaranteed are avoided; in addition, the Wiener type nonlinear model (namely the full-flow nonlinear dynamic optimization model) combining the nonlinear steady-state model and the dynamic linear prediction model is adopted in the embodiment of the application, the solving speed is higher, the optimization period is also faster, and meanwhile, the Wiener type nonlinear model and the advanced controller execute at the same frequency, so that the real-time dynamic optimization is realized, and the problem that the optimization and the control cannot be executed simultaneously due to inconsistent models is avoided. And further, the technical problems that the time for calculating the optimized process variable of the chemical production device by the related real-time optimization technology based on the steady-state mechanism model is too long and the chemical production device is not easy to converge are solved.

Note that each module in the above-described full-flow dynamic optimization control device may be a program module (for example, a set of program instructions for implementing a specific function), or may be a hardware module, and for the latter, it may take the following form, but is not limited thereto: the expression forms of the modules are all a processor, or the functions of the modules are realized by one processor.

Example 3

There is also provided, in accordance with an embodiment of the present application, an electronic device, wherein the electronic device includes one or more processors; and a memory for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement a method for running the programs, wherein the programs are configured to execute the whole-flow dynamic optimization control method of the chemical production device in embodiment 1.

Optionally, the processor is configured to implement the following steps by computer program execution:

step S102, a target parameter set of the chemical production device in the production process is obtained, wherein the target parameter set at least comprises: operating variable parameters, controlled variable parameters and disturbance variable parameters;

Step S104, constructing a nonlinear steady-state model of the chemical production device based on the target parameter set, and determining a nonlinear gain matrix of the controlled variable parameter based on the nonlinear steady-state model;

step S106, a dynamic linear prediction model is constructed according to the operation variable parameters and the controlled variable parameters, and the dynamic linear prediction model is corrected through a nonlinear gain matrix of the controlled variable parameters, so that a full-flow nonlinear dynamic optimization model is obtained;

The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.

In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of units may be a logic function division, and there may be another division manner in actual implementation, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be essentially or a part contributing to the related art or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application and are intended to be comprehended within the scope of the present application.

Claims

1. The full-flow dynamic optimization control method of the chemical production device is characterized by comprising the following steps of:

obtaining a target parameter set of a chemical production device in a production process, wherein the target parameter set at least comprises: operating variable parameters, controlled variable parameters and disturbance variable parameters;

constructing a nonlinear steady-state model of the chemical production device based on the target parameter set, and determining a nonlinear gain matrix of the controlled variable parameter based on the nonlinear steady-state model;

constructing a dynamic linear prediction model according to the operation variable parameter and the controlled variable parameter, and correcting the dynamic linear prediction model through a nonlinear gain matrix of the controlled variable parameter to obtain a full-flow nonlinear dynamic optimization model;

performing online data setting processing on the target parameter set, solving the full-flow nonlinear dynamic optimization model in real time based on the processed target parameter set under the limitation of an optimization target and an optimization constraint condition to obtain a dynamic optimization result, and transmitting the dynamic optimization result to an advanced controller for execution;

Based on the processed target parameter set, under the limitation of the optimization target and the optimization constraint condition, solving the full-flow nonlinear dynamic optimization model in real time to obtain the dynamic optimization result, wherein the method comprises the following steps: determining the optimization constraint condition, wherein the optimization constraint condition comprises at least one of the following: a dynamic material balance constraint condition, a first upper limit and a lower limit of the operation variable parameter, a second upper limit and a lower limit of a first increment of the operation variable, a third upper limit and a lower limit of the controlled variable parameter, a fourth upper limit and a lower limit of a second increment of the controlled variable, and a fifth upper limit and a lower limit of the controlled variable parameter after correction; determining a relaxation variable corresponding to a third upper and lower limit of the controlled variable parameter and a fifth upper and lower limit of the corrected controlled variable parameter, and expanding the third upper and lower limit and the fifth upper and lower limit based on the relaxation variable; determining the optimization target, and determining an optimization objective function of the full-flow nonlinear dynamic optimization model based on the processed target parameter set under the limitation of the optimization target, wherein the optimization target of the optimization objective function comprises at least one of the following: the yield is maximized and the consumption is minimized; and solving the optimization objective function by using a nonlinear programming solver in a feasible region defined by the adjusted optimization constraint condition to obtain the dynamic optimization result, wherein the dynamic optimization result comprises at least one of the following components: the first target dynamic running track and the first target steady state value of the operation variable parameter, and the second target dynamic running track and the second target steady state value of the controlled variable parameter.

2. The method of claim 1, wherein the set of target parameters further comprises: the product property parameter, wherein, obtain the target parameter set of chemical production device in the production process, include:

obtaining the product property parameters produced by the chemical production device from an online analysis database, wherein the product property parameters comprise at least one of the following components: product density, product content, moisture content, acidity;

obtaining the operation variable parameter and the controlled variable parameter of the chemical production device from a real-time database, wherein the operation variable parameter comprises at least one of the following: the method comprises the following steps of feeding flow, discharging flow, feeding pressure, discharging pressure, feeding temperature, discharging temperature and reaction temperature, wherein the controlled variable parameters comprise at least one of the following: purity, reaction rate, pH, liquid level, and material concentration.

3. The method of claim 1, wherein constructing a nonlinear steady-state model of the chemical production plant based on the set of target parameters and determining a nonlinear gain matrix of the controlled variable parameters based on the nonlinear steady-state model comprises:

Determining a preset artificial intelligence learning model, wherein the artificial intelligence learning model comprises at least one of the following: a deep learning model, a neural network model, and a BOOST model;

inputting the operation variable parameters and the disturbance variable parameters as input features to the artificial intelligent learning model to obtain a prediction result of the artificial intelligent learning output;

adjusting model parameters of the artificial intelligent learning model based on the prediction result and the controlled variable parameters to obtain a nonlinear steady-state model of the chemical production device;

and performing bias derivative on the operation variable parameter through the nonlinear steady-state model to obtain a nonlinear gain matrix of the controlled variable parameter.

4. The method of claim 1, wherein constructing a dynamic linear prediction model from the operating variable parameter and the controlled variable parameter comprises:

determining transfer functions of the operation variable parameters and the controlled variable parameters, and identifying the transfer functions to obtain a dynamic prediction matrix;

and establishing the dynamic linear prediction model based on the dynamic prediction matrix.

5. The method of claim 1, wherein performing online data tuning processing on the target parameter set comprises:

Performing a preprocessing operation on the target parameter set, wherein the preprocessing operation comprises at least one of the following steps: consistency detection, filling of missing values and normalization treatment;

and carrying out online data setting processing and model parameter estimation on the target parameter set after pretreatment, wherein the model parameter estimation is used for estimating model parameters of a dynamic material balance model, and the dynamic material balance model is used for balancing the material relation between the input variable and the output variable of the chemical production device.

6. The method of claim 5, wherein performing on-line data tuning and model parameter estimation on the preprocessed target parameter set comprises:

performing online data setting processing on the preprocessed target parameter set, and constructing the dynamic material balance model based on the processed target parameter set;

determining a dynamic material balance constraint, wherein the dynamic material balance constraint comprises at least one of: material balance equation, energy balance equation, device connection equation, physical property calculation equation and molecular normalization equation;

based on the dynamic material balance constraint condition, establishing a target setting function by taking the minimum model error of the dynamic material balance model as a target;

And solving the target setting function by adopting a nonlinear optimizer to obtain model parameters of the dynamic material balance model.

7. The method of claim 1, wherein issuing the dynamic optimization result to an advanced controller for execution comprises:

and controlling the operation variable parameter to be adjusted to the first target steady-state value according to the first target dynamic running track through the advanced controller, and controlling the controlled variable parameter to be adjusted to the second target steady-state value according to the second target dynamic running track.

8. The utility model provides a chemical production device full flow developments optimizing control device which characterized in that includes:

the acquisition module is used for acquiring a target parameter set of the chemical production device in the production process, wherein the target parameter set at least comprises: operating variable parameters, controlled variable parameters and disturbance variable parameters;

the first construction module is used for constructing a nonlinear steady-state model of the chemical production device based on the target parameter set and determining a nonlinear gain matrix of the controlled variable parameter based on the nonlinear steady-state model;

the second construction module is used for constructing a dynamic linear prediction model according to the operation variable parameter and the controlled variable parameter, and correcting the dynamic linear prediction model through a nonlinear gain matrix of the controlled variable parameter to obtain a full-flow nonlinear dynamic optimization model;

The optimization control module is used for carrying out online data setting processing on the target parameter set, solving the full-flow nonlinear dynamic optimization model in real time under the limitation of an optimization target and an optimization constraint condition based on the processed target parameter set to obtain a dynamic optimization result, and sending the dynamic optimization result to an advanced controller for execution;

9. An electronic device, comprising: a memory and a processor for running a program stored in the memory, wherein the program is run to execute the full-flow dynamic optimization control method of the chemical production apparatus according to any one of claims 1 to 7.