CN-122018456-A - Environment-friendly continuous flow high-pressure reaction method and system based on supercritical fluid

CN122018456ACN 122018456 ACN122018456 ACN 122018456ACN-122018456-A

Abstract

The invention relates to the technical field of industry, and discloses an environment-friendly continuous flow high-pressure reaction method based on supercritical fluid. The system realizes the full-flow automatic management from data acquisition to control execution, and completes the core control tasks of accurate perception of process state, accurate prediction of future trend, optimal decision of control strategy, coordinated configuration of execution action and the like. The system can accurately grasp complex physical property changes and reaction dynamic characteristics of the supercritical fluid through the combination of model prediction and real-time calibration, and effectively inhibit the influence of various disturbances on the process stability through the cooperation of feedforward compensation and feedback regulation.

Inventors

ZHANG BINTIAN
YAN ZHIYONG
HU YING
LI XINGJIE
YIN ANLIN
ZHANG KUIHUA
Ye Junning
ZHAN JIANCHAO
LI YING

Assignees

嘉兴大学

Dates

Publication Date: 20260512
Application Date: 20260113

Claims (10)

1. An environment-friendly continuous flow high-pressure reaction method based on supercritical fluid is characterized by comprising the following specific steps: S1, distributed data acquisition, namely respectively arranging a pressure sensor, a temperature sensor, a flowmeter and a densimeter at a feed end, a reaction section and an outlet section of a reaction device, wherein the sensors acquire data in a high-frequency mode, transmit the data to a data processing unit after filtering processing, and calculate local fluid density and phase state information through a soft measurement algorithm; s2, establishing a simplified model, namely extracting features of a three-dimensional flow field simulation result which is finished offline, constructing a simplified fluid mechanics model containing main flow characteristics, combining the model with a supercritical fluid state equation to calculate real-time physical parameters, and coupling a reaction rate calculation module to enable single prediction calculation to meet real-time control requirements; s3, model calibration synchronization, namely driving the simplified model to operate by taking real-time measured process parameters as input, setting an allowable deviation range of a predicted value and an actual measured value, automatically adjusting parameters such as a heat transfer coefficient, a resistance coefficient, catalyst activity and the like in the model when the parameters exceed the allowable deviation range, and acquiring the optimal estimation of the system operation state through a data fusion technology; S4, multivariable optimization control, namely taking the calibrated model as a deduction tool of a future state, setting a prediction time window, calculating an optimal adjustment scheme of the opening degree of a back pressure valve, the rotating speed of a feed pump and the power of a heater on the premise of meeting constraints such as pressure fluctuation, flow deviation and temperature change rate, and executing a first-step control action after solving the optimization problem; S5, a disturbance compensation mechanism is used for calculating the influence of the observable feed fluctuation or composition change on the system in advance according to the response characteristic of the model and generating a compensation signal, and meanwhile, a feedback adjustment function based on deviation is kept to process uncertain factors, and the two adjustment signals are adaptively weighted according to the current working condition and then act on an executing mechanism together; S6, the coordination management of the executors is that the action capability boundaries of all the executors are checked before the control instructions are issued, when one of the executors reaches the adjustment limit, part of control tasks are distributed to other available equipment, the control actions are reasonably distributed according to the accumulated workload of each executor, and safety limit is set to prevent operation from crossing the limit; s7, experience accumulation application, namely establishing a historical operation record library covering different raw material characteristics, catalyst states and environmental conditions, inducing past working conditions into a plurality of typical categories, identifying the category to which the past working conditions belong in operation through feature comparison, adopting a corresponding mature control scheme, and continuously bringing new successful experiences into the record library to perfect a control strategy; S8, predicting the equipment state, namely tracking the change trend of the model parameters, establishing a prediction relation of equipment performance decline, judging the development progress of conditions such as catalyst activity decline, sealing element aging or heat transfer surface pollution, and the like, and when the expected target level is reached in a short period, finely adjusting the operation parameters in advance to compensate and prompting proper maintenance time to an operator.
2. The method for environment-friendly continuous flow high-pressure reaction based on supercritical fluid according to claim 1, wherein in the step S1, the specific steps of distributed data acquisition are as follows: S1.1, arranging a pressure transmitter and a flowmeter at a feed end of a reaction device, arranging temperature measuring points at a reaction section along the flowing direction, arranging a densimeter and a pressure sensor at an outlet section, acquiring process parameters by the sensor in a continuous sampling mode, transmitting the process parameters to a data processing unit through an industrial communication network, and performing digital filtering processing on a transmission signal to eliminate interference and vibration influence; S1.2, the data processing unit calculates flow resistance according to the pressure measured value and the pipeline parameter, estimates the heat transfer parameter according to the temperature distribution and the heat flow density, obtains the fluid density and the viscosity by using the pressure value and the temperature value through physical correlation or table lookup, organizes the measured data and the calculated parameter into a data sequence with a time mark and transmits the data sequence to the model calculating unit, and establishes a rolling window to store recent data.
3. The method for environment-friendly continuous flow high-pressure reaction based on supercritical fluid according to claim 2, wherein in the step S2, the specific steps for simplifying the model establishment are as follows: s2.1, performing modal decomposition on three-dimensional flow field simulation data, extracting main spatial modes and time evolution coefficients, reconstructing a simplified hydrodynamic model based on the main modes, retaining speed distribution and pressure drop characteristics, reducing calculation dimensions, converting a three-dimensional grid into a one-dimensional pipeline model and a quasi-two-dimensional section model, and controlling the number of degrees of freedom through a modal cutoff threshold; S2.2, establishing a data interface between the simplified fluid mechanics model and a supercritical fluid state equation, calculating a fluid density value and a fluid viscosity value according to a temperature value and a pressure value at each calculation moment, updating the fluid density value and the fluid viscosity value to a constitutive relation of a flow field model, embedding a reaction dynamics module, calculating a reaction rate according to a local temperature and a concentration, and introducing the reaction rate into the equation as a source term to complete the circulation calculation of a flow field, physical parameters and the reaction rate.
4. The method for environment-friendly continuous flow high-pressure reaction based on supercritical fluid according to claim 3, wherein in the step S3, the model calibration synchronization comprises the following specific steps: S3.1, inputting the real-time parameters obtained in the step S1 into the simplified model established in the step S2 to drive the simplified model to perform calculation, outputting predicted values of pressure, temperature and flow of each key position by the model, comparing the predicted values with corresponding sensor measured values to calculate deviation amounts, setting deviation thresholds for different parameters, judging that the model is mismatched with the actual state when the deviation of the measuring points exceeds the thresholds in a plurality of continuous periods, and recording deviation characteristics; S3.2, starting a parameter identification program according to the deviation characteristics of the step S3.1, analyzing deviation amplitude and position to determine correction parameter types, adjusting heat transfer coefficients when temperature deviation is dominant, adjusting resistance coefficients when pressure deviation is dominant, adjusting catalyst activity coefficients when conversion rate deviation is dominant, solving a parameter value which enables the square sum of deviation to be minimum by adopting a least square method or a gradient optimization algorithm, updating a model, generating optimal estimated values of system pressure, temperature and flow by applying Kalman filtering fusion of the updated prediction result and measured data, and transmitting the optimal estimated values to the step S4.
5. The method for continuous flow high pressure reaction based on supercritical fluid according to claim 4, wherein in the step S4, the multivariable optimization control comprises the following specific steps: S4.1, taking the state estimation value obtained in the step S3 as an initial condition, utilizing the calibrated model deduction system to cover main dynamic response characteristics by setting the length of a prediction window, setting the opening degree of a back pressure valve, the rotating speed of a pump and the heating power as adjustable operation variables and setting the variation amplitude and the rate limit of the back pressure valve, setting constraint conditions for pressure fluctuation, flow deviation and the temperature variation rate, expressing in an inequality group form and establishing a constraint optimization mathematical model; And S4.2, establishing a weighted objective function for comprehensively reflecting the deviation of the controlled variable and the change of the operating variable, solving an operating variable adjusting track which is the smallest in the objective function under the constraint condition of the step S4.1 by adopting a sequence quadratic programming algorithm or a gradient optimization algorithm, wherein the track comprises valve opening, pump rotating speed and heating power value at each moment, extracting a first group of values as control instructions, and sending the control instructions to an execution device, and repeatedly executing the prediction and optimization process in the next period by taking the state estimated value updated in the step S3 as an initial condition.
6. The method for environment-friendly continuous flow high pressure reaction based on supercritical fluid according to claim 5, wherein in the step S5, the disturbance compensation mechanism comprises the following specific steps: S5.1, monitoring flow and composition data of a feeding end in real time, judging the flow to be measurable disturbance when the flow step change exceeds a threshold value or the composition deviates from a normal range, simulating the influence process of disturbance on pressure, temperature and flow by using the model established in the step S2 and taking the current working condition as a reference, analyzing the peak value, arrival time and attenuation characteristic of an influence track, reversely calculating the valve opening, pump rotating speed and heating power adjustment quantity required for counteracting the influence, and organizing the valve opening, pump rotating speed and heating power adjustment quantity into feedforward compensation signals; And S5.2, maintaining the operation of a feedback regulation loop, collecting actual measurement values of pressure, temperature and flow, calculating the deviation between the actual measurement values and a set value, generating a feedback control signal through a proportional-integral-derivative algorithm according to the deviation and the change rate, setting initial weighting coefficients of feedforward and feedback signals, increasing feedforward weight when the disturbance amplitude is large and the model error is small, increasing feedback weight when unmodeled dynamics or noise is large, adjusting weight according to working condition characteristics and control effects, superposing the weighted and fused comprehensive signal and the prediction instruction of the step S4, and transmitting the superposed comprehensive signal to the step S6.
7. The method for environment-friendly type continuous flow high pressure reaction based on supercritical fluid according to claim 6, wherein in the step S6, the specific steps of the executor coordination management are as follows: S6.1, receiving the comprehensive control signals output in the step S5, inquiring the current states of the actuating mechanisms such as the opening degree of the back pressure valve, the rotating speed of the pump, the power of the heater and the like, calculating the available adjustment margin from each actuating mechanism to the physical limit, starting task redistribution when the required adjustment quantity exceeds the available margin, determining a substitute actuating mechanism combination based on the coupling relation of pressure, flow and temperature, calculating the action quantity of each mechanism of the substitute combination, and redistributing the control task according to the adjustment capacity proportion; S6.2, establishing and updating accumulated action times, running time and response time records of all execution mechanisms, calculating health evaluation indexes, when a plurality of execution mechanisms can meet requirements, preferentially selecting mechanisms with high health and low load, respectively verifying whether position parameters are in a safety interval, whether an adjustment rate exceeds an allowable value and whether cooperative actions cause mutation or not by implementing three-level verification before issuing, limiting amplitude or prolonging execution intervals when risks are detected, issuing a verification-passing instruction to an execution device, and transmitting action records and health state data to the step S7.
8. The method for continuous flow high pressure reaction according to claim 7, wherein in step S7, the empirical accumulation application is as follows: S7.1, recording process parameters, control actions and system response data of different raw material batches, catalyst using stages and environmental conditions, extracting working condition feature vectors comprising average pressure, temperature distribution, flow stability and conversion rate level, dividing a historical working condition into a plurality of typical operation modes by adopting a clustering algorithm, establishing a feature parameter range and a corresponding effective control parameter combination for each mode, and storing the feature parameter range and the corresponding effective control parameter combination in a historical database in a structured form; And S7.2, extracting the feature vector of the current working condition during operation, calculating the distance or similarity between the feature vector and the feature center of each typical mode, selecting the mode with the closest distance or the highest similarity, calling the control parameter of the mode as the initial configuration of the step S4 and the step S5, evaluating performance indexes such as pressure stability, flow fluctuation, temperature deviation and the like after the control period is completed, adding the increment after the working condition data marking mode category into a database when the standard is met, initializing and independently recording the new working condition with the similarity lower than the threshold value by using the relatively optimal mode parameter, and transmitting the updated database and mode information to the step S8.
9. The method for environment-friendly type continuous flow high pressure reaction based on supercritical fluid according to claim 8, wherein in the step S8, the device state pre-judging specific steps are as follows: S8.1, continuously monitoring a model parameter history sequence obtained in the step S3, extracting a time evolution curve of a heat transfer coefficient, a resistance coefficient and a catalyst activity coefficient, calculating the change rate and fluctuation amplitude of each parameter by adopting a sliding window method, comparing the change rate with a normal range, judging surface pollution when the heat transfer coefficient continuously drops by a super threshold value, judging channel blockage when the resistance coefficient continuously rises, judging inactivation when the catalyst activity continuously decays, and establishing a mapping model of parameter change trend and equipment performance decay state by combining the history data of the step S7; S8.2, predicting the development track of each parameter in a future time period based on the mapping model in the step S8.1, starting precompensation when predicting that a certain parameter reaches an off-level in a plurality of periods, increasing heating power in advance for heat transfer coefficient decrease, adjusting pump rotation speed or valve opening for resistance coefficient increase, adjusting reaction temperature or residence time for catalyst activity decrease, superposing compensation quantity to a control instruction in the step S4, generating a maintenance suggestion according to parameter degradation degree and residual time, sending a cleaning suggestion when the pollution degree is predicted to reach a cleaning threshold, sending a replacement suggestion when the activity is predicted to be reduced to a replacement standard, and outputting state evaluation and maintenance suggestion to an interaction interface.
10. The environment-friendly continuous flow high-pressure reaction system based on the supercritical fluid is characterized by being used for executing the environment-friendly continuous flow high-pressure reaction method based on the supercritical fluid, which is disclosed in any one of claims 1-9, and comprises a reaction device module, a data acquisition module, a model calculation module, an optimization control module, an execution regulation module, a historical data module, a state prejudgment module and a man-machine interaction module; the reaction device module is used for bearing the supercritical fluid reaction process, the feed end receives the raw materials, the catalyst in the reaction section completes chemical conversion, the output section outputs products, and the integral bearing capacity accords with the supercritical operation condition; The data acquisition module is used for monitoring the state of the reaction process in real time, acquiring pressure, temperature, flow and density parameters through sensors arranged at each key position, and transmitting data to the calculation unit after filtering treatment to eliminate interference; The model calculation module is used for predicting the running state of the system, calculating real-time physical parameters by adopting a simplified hydrodynamic model and combining a state equation, obtaining the reaction rate by coupling reaction dynamics, and automatically correcting internal parameters through a calibration function to enable a prediction result to be consistent with the actual running; The optimization control module is used for generating an optimal control scheme, deducing future evolution trend of the system based on a calibration model, calculating a coordinated regulation scheme of the back pressure valve, the feed pump and the heater on the premise of meeting safety constraint, and fusing a plurality of control strategies of prediction, compensation and feedback; the execution regulation module is used for implementing control instructions, driving the back pressure valve to regulate pressure, the feed pump to regulate flow and the heater to regulate temperature according to the instructions of the optimization control module, monitoring the running state and accumulated load of each mechanism and feeding back the execution condition to the control link; The historical data module is used for storing operation experience and identifying the type of working condition, classifying and recording operation data of different raw materials, catalyst states and environmental conditions, establishing a typical working condition mode library, quickly matching the current state during operation, calling a verified control scheme and continuously absorbing a new experience-enriched knowledge library; The state pre-judging module is used for predicting the performance change trend of the equipment, monitoring the long-term evolution rule of model parameters, judging performance degradation symptoms such as catalyst attenuation, heat transfer efficiency reduction, flow resistance increase and the like, predicting the development trend, starting compensation adjustment in advance, and generating maintenance opportunity suggestions; The man-machine interaction module is used for connecting an operator with the control system, displaying current operation parameters, control actions, equipment health assessment and maintenance prompts, receiving process setting and mode selection, and supporting visual management and necessary manual intervention of the system.

Description

Environment-friendly continuous flow high-pressure reaction method and system based on supercritical fluid Technical Field The invention relates to the technical field of industry, in particular to an environment-friendly continuous flow high-pressure reaction method and system based on supercritical fluid. Background In controlling this subdivision direction during the continuous reaction, supercritical fluid high pressure reaction systems represent a top of control difficulties. The system needs to maintain stable operation under extreme operating conditions, wherein pressure-flow cooperative control is a core link for guaranteeing product quality consistency. The existing digital twin application stays at the off-line simulation or state monitoring level, a multi-physical-field real-time prediction model integrating hydrodynamic calculation, phase thermodynamic description and chemical reaction dynamics is not really realized, and a closed-loop framework for directly converting a prediction result into a control instruction is not yet provided. There are three major essential drawbacks to the control schemes currently in common use in the industry. Firstly, simplifying the model excessively, simplifying a complex three-dimensional flow field into a one-dimensional lumped parameter model, failing to capture uneven speed distribution and local density fluctuation in a pipeline, secondly, delaying physical property calculation, wherein density and viscosity data used by a controller are based on temperature and pressure conditions at the last moment and are not the current real state, thirdly, direct influence of supercritical area pressure on flow measurement is ignored in multivariable decoupling control, false deviation is generated in flowmeter reading when pressure fluctuates, and the controller adjusts the disturbance according to the false deviation instead. This architecture barely maintains operation under steady state conditions, but responds slowly and is prone to oscillations in the face of dynamic disturbances. Therefore, we propose an environment-friendly continuous flow high-pressure reaction method and system based on supercritical fluid, so as to solve the above-mentioned problems. Disclosure of Invention The invention aims to provide an environment-friendly continuous flow high-pressure reaction method and system based on supercritical fluid, which are used for solving the problems that the model provided by the background art is excessively simplified, a complex three-dimensional flow field is simplified into a one-dimensional lumped parameter model, and uneven speed distribution and local density fluctuation in a pipeline cannot be captured. In order to achieve the aim, the invention provides the following technical scheme that the environment-friendly continuous flow high-pressure reaction method based on the supercritical fluid comprises the following specific steps: S1, distributed data acquisition, namely respectively arranging a pressure sensor, a temperature sensor, a flowmeter and a densimeter at a feed end, a reaction section and an outlet section of a reaction device, wherein the sensors acquire data in a high-frequency mode, transmit the data to a data processing unit after filtering processing, and calculate local fluid density and phase state information through a soft measurement algorithm; s2, establishing a simplified model, namely extracting features of a three-dimensional flow field simulation result which is finished offline, constructing a simplified fluid mechanics model containing main flow characteristics, combining the model with a supercritical fluid state equation to calculate real-time physical parameters, and coupling a reaction rate calculation module to enable single prediction calculation to meet real-time control requirements; s3, model calibration synchronization, namely driving the simplified model to operate by taking real-time measured process parameters as input, setting an allowable deviation range of a predicted value and an actual measured value, automatically adjusting parameters such as a heat transfer coefficient, a resistance coefficient, catalyst activity and the like in the model when the parameters exceed the allowable deviation range, and acquiring the optimal estimation of the system operation state through a data fusion technology; S4, multivariable optimization control, namely taking the calibrated model as a deduction tool of a future state, setting a prediction time window, calculating an optimal adjustment scheme of the opening degree of a back pressure valve, the rotating speed of a feed pump and the power of a heater on the premise of meeting constraints such as pressure fluctuation, flow deviation and temperature change rate, and executing a first-step control action after solving the optimization problem; S5, a disturbance compensation mechanism is used for calculating the influence of the observable feed fluctuation or composition