Temperature control method and system for closed-loop controlled chemical reaction kettle

Abstract

The application provides a closed-loop control chemical reaction kettle temperature control method and system, relates to the technical field of temperature control, and comprises the following steps: dynamically arranging a temperature sensing unit according to stage sensing driving reaction kettle information of a chemical reaction kettle; collecting signals of a reaction kettle monitoring point in real time; transmitting real-time data to an adaptive decision discrimination channel, and performing combined determination based on a temperature deviation calculation subchannel and a control decision generation subchannel in the adaptive decision discrimination channel; adjusting a heating unit, a cooling unit and a heat exchange medium flow of the chemical reaction kettle according to stage regulation instructions, and collecting adjustment signals; inputting reaction temperature feedback of the adjustment signals into the adaptive decision discrimination channel, updating control output, and forming a closed-loop temperature control loop. The application can solve the technical problem of lacking cross-stage regulation in the prior art, achieve the adaptive closed-loop temperature control target, and improve the whole-process temperature control precision.

Description

Technical Field

[0001] This application relates to the field of temperature control technology, and in particular to a closed-loop control method and system for temperature control of chemical reactors. Background Technology

[0002] With the increasing demands for safety, stability, and product consistency in reaction processes in fields such as fine chemicals and new material synthesis, chemical reactors, as core reaction equipment, directly affect reaction rates, reaction selectivity, and final product quality through their temperature control. In particular, in multi-stage, strongly exothermic, or significantly thermally inertial reaction processes, the spatiotemporal distribution of the temperature field and its dynamic evolution characteristics place higher demands on reaction process control.

[0003] Currently, existing temperature control technologies for chemical reactors mostly employ fixed-point temperature sensing and traditional control strategies based on a single feedback loop. Typically, only a small number of temperature sensors are deployed at local locations on the reactor wall or inside the reactor, using the overall average temperature or a local representative temperature as the basis for control. This lacks a systematic characterization of the temperature distribution differences at different time stages of the reaction process and within the reactor. Control strategies are mostly executed independently within a single stage, making it difficult to effectively predict and compensate for the lag effects of previous stage control behavior in subsequent stages, and failing to accurately reflect the true distribution of the temperature field inside the reactor at different reaction stages.

[0004] In summary, existing technologies suffer from a lack of collaborative sensing and cross-stage regulation mechanisms for the time evolution of the reactor and changes in the spatial temperature field during temperature control. This results in temperature control decisions failing to fully reflect the reaction heat hysteresis effect and the impact of stage switching, further affecting the temperature control accuracy and operational stability of chemical reactors in complex reaction processes. Summary of the Invention

[0005] The purpose of this application is to provide a closed-loop control method and system for temperature control of chemical reactors, in order to solve the technical problem in the prior art that the temperature control process lacks a collaborative perception and cross-stage regulation mechanism for the time evolution stage and spatial temperature field changes of the reactor, which leads to the inability of temperature control decisions to fully reflect the reaction heat hysteresis effect and the impact of stage switching, and further affects the temperature control accuracy and operational stability of chemical reactors in complex reaction processes.

[0006] In view of the above problems, this application provides a closed-loop control method and system for temperature control of chemical reactors.

[0007] Firstly, this application provides a closed-loop control method for temperature control of a chemical reactor, implemented through a closed-loop control system. The method includes: dynamically deploying temperature sensing units driven by staged perception based on reactor information to construct reactor monitoring points; receiving real-time signals from the reactor monitoring points to establish real-time data; establishing an adaptive decision-making channel, transmitting the real-time data to the adaptive decision-making channel, and combining and determining a staged control command based on a temperature deviation calculation sub-channel and a control decision generation sub-channel within the adaptive decision-making channel; adjusting the flow rates of the heating unit, cooling unit, and heat exchange medium of the chemical reactor according to the staged control command, and collecting adjustment signals; and inputting the reaction temperature feedback based on the adjustment signals into the adaptive decision-making channel to update the control output, thus forming a closed-loop temperature control loop.

[0008] Preferably, the closed-loop control method for temperature control of a chemical reactor further includes: analyzing the monitoring points of the chemical reactor in the time and space dimensions based on the reactor information to determine the time dimension monitoring points and the space dimension monitoring points; and matching the time dimension monitoring points and the space dimension monitoring points to determine the reactor monitoring points.

[0009] Preferably, the closed-loop control method for temperature control of a chemical reactor further includes: extracting near-wall points, center points inside the reactor, bottom points, and near-liquid surface points based on the reactor structure information in the reactor information; configuring temperature sensing units for the near-wall points, center points inside the reactor, bottom points, and near-liquid surface points to obtain near-wall measuring points, center points inside the reactor, bottom points, and near-liquid surface measuring points; and combining the near-wall measuring points, center points inside the reactor, bottom points, and near-liquid surface measuring points to obtain the spatial dimension monitoring points.

[0010] Preferably, the closed-loop control method for temperature control of a chemical reactor further includes: dividing the reactor state information in the reactor information into stages based on the reaction thermal characteristics to determine the reaction thermal characteristic stages, wherein the reaction thermal characteristic stages include a start-up heating stage, a main reaction stage, a post-reaction stage, and a termination cooling stage; dividing the reactor state information into stages based on process actions to determine the process action characteristic stages, wherein the process action stages include an initial solvent heating stage, a main reactant feeding stage, a batch feeding stage, and a feeding end stabilization stage; and dividing the reactor state information into stages based on the temperature field spatial characteristics to determine the temperature field spatial characteristic stages. The temperature field spatial characteristic stage includes a temperature rise stage, a main reaction stage, a dead zone stage, and a cooling stage. Based on a preset importance weight, the reaction thermal characteristic stage, the process action characteristic stage, and the temperature field spatial characteristic stage are weighted and calculated to obtain a time evolution stage, which includes a main control stage, an auxiliary stage, a constraint stage, and a monitoring stage. Temperature sensing units are configured for the main control stage, the auxiliary stage, the constraint stage, and the monitoring stage to obtain the time dimension monitoring points, which include time dimension main control measurement points, time dimension auxiliary measurement points, time dimension constraint measurement points, and time dimension monitoring points.

[0011] Preferably, the closed-loop control method for temperature control of a chemical reactor further includes: performing time evolution sensing based on the time evolution stage, calling the time dimension monitoring points according to the sensing time stage to obtain real-time time monitoring points; and determining the reactor monitoring points by matching the real-time time monitoring points with the spatial dimension monitoring points.

[0012] Preferably, the closed-loop control method for temperature control of a chemical reactor further includes: calculating the temperature deviation between the real-time data and the target temperature based on the temperature deviation calculation subchannel to obtain the temperature deviation and the temperature deviation change trend; and generating the stage control command for the time evolution stage based on the control decision generation subchannel according to the temperature deviation using a closed-loop control mechanism, wherein the closed-loop control mechanism includes proportional regulation, integral regulation and derivative regulation.

[0013] Preferably, the closed-loop control method for temperature control of a chemical reactor further includes: determining whether the real-time data is within a safety margin, wherein the safety margin includes temperature safety margin, pressure safety margin, and material safety margin; if so, determining whether the real-time data is within a process allowable margin; if so, determining whether the real-time data is within a time capability margin; if so, determining whether the real-time data is within a control logic compatibility margin, wherein the control logic compatibility margin characterizes the preset value of interference caused by the coupling effect between control logic and temperature control on non-temperature parameters in subsequent stages of the time evolution stage; if so, predicting the delayed regulation effect of subsequent stages based on the time evolution stage, and performing cross-stage compensation adjustment of the chemical reactor based on the delayed regulation effect through subsequent stage compensation; inputting the reaction temperature of the cross-stage compensation adjustment signal into the adaptive decision discrimination channel, updating the control output, and completing the allowable deviation transmission.

[0014] Preferably, the closed-loop control method for temperature control of a chemical reactor further includes: determining the time interval from the end of the real-time phase to the start of the subsequent phase; acquiring real-time data of the chemical reactor during the real-time phase, retrieving a hysteresis correlation model matching the real-time phase from a pre-stored model library, wherein the hysteresis correlation model is trained based on historical operating data and is used to characterize the influence of the temperature response on the subsequent temperature field of the subsequent phase after the time interval under the real-time operating conditions of the real-time phase, including the heat transfer hysteresis effect and the material reaction accumulation effect; inputting the real-time data into the hysteresis correlation model for extrapolation and calculation, generating a predicted temperature curve for the subsequent phase without additional intervention, using the predicted temperature curve as the delayed control effect, and the predicted temperature curve characterizing the dynamic temperature change after the hysteresis time from the current moment.

[0015] Preferably, the closed-loop control method for temperature control of a chemical reactor further includes: taking the ideal temperature curve of the subsequent stage as the control target, comparing the predicted temperature curve with the ideal temperature curve, and obtaining a compensation target value through the temperature deviation region; based on the compensation target value, combined with the time length and heat exchange capacity of the subsequent stage, calculating the compensation adjustment amount for the subsequent stage; setting a decay strategy for the compensation intensity, obtaining the basic control command for the subsequent stage, converting the compensation adjustment amount from the basic control command into a compensation control command according to the decay strategy, generating a subsequent control command, and completing cross-stage compensation adjustment.

[0016] Secondly, this application also provides a closed-loop control temperature control system for a chemical reactor, used to execute the closed-loop control temperature control method for a chemical reactor as described in the first aspect, comprising: a reactor monitoring point construction module, used to dynamically deploy temperature sensing units driven by stage perception based on reactor information of the chemical reactor to construct reactor monitoring points; a real-time data establishment module, used to receive the collected signals from the reactor monitoring points in real time and establish real-time data; a stage control command establishment module, used to establish an adaptive decision-making channel, transmit the real-time data to the adaptive decision-making channel, and perform a combined judgment based on the temperature deviation calculation sub-channel and the control decision generation sub-channel within the adaptive decision-making channel to establish a stage control command; an adjustment signal acquisition module, used to adjust the heating unit, cooling unit, and heat exchange medium flow rate of the chemical reactor according to the stage control command, and acquire the adjustment signal; and a feedback module, used to input the reaction temperature feedback of the adjustment signal into the adaptive decision-making channel, update the control output, and form a closed-loop temperature control loop.

[0017] The technical solution provided in this application has at least the following technical effects or advantages: by achieving the technical goal of adaptive closed-loop temperature control based on joint sensing of time evolution stage and spatial temperature field, it achieves the technical effect of accurately adjusting temperature changes in stages throughout the reaction process, effectively suppressing temperature lag deviation, and maintaining temperature continuity and control stability between different stages.

[0018] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the closed-loop temperature control method for chemical reactors according to this application.

[0021] Figure 2 This is a schematic diagram of the closed-loop temperature control system for a chemical reactor according to this application.

[0022] Figure labeling: Module 1 for constructing monitoring points in the reactor, Module 2 for establishing real-time data, Module 3 for establishing stage control commands, Module 4 for acquiring control signals, and Module 5 for feedback. Detailed Implementation

[0023] This application provides a closed-loop control method and system for temperature control of chemical reactors, solving the technical problem in existing technologies where the lack of a collaborative sensing and cross-stage regulation mechanism for the temporal evolution stages and spatial temperature field changes in the temperature control process leads to temperature control decisions that cannot fully reflect the effects of reaction heat hysteresis and stage switching, further affecting the temperature control accuracy and operational stability of chemical reactors in complex reaction processes. The application achieves the technical goal of adaptive closed-loop temperature control based on joint sensing of temporal evolution stages and spatial temperature field, enabling precise staged adjustment of temperature changes throughout the entire reaction process, effectively suppressing temperature hysteresis deviations, and maintaining temperature continuity and control stability between different stages.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a closed-loop control method for temperature control of chemical reactors, applicable to closed-loop control systems for chemical reactor temperature control, specifically including the following steps: Based on the reactor information of the chemical reactor, temperature sensing units are dynamically deployed in a phase-sensing manner to construct reactor monitoring points.

[0026] Furthermore, this application also includes: analyzing the monitoring points of the chemical reactor in the time and space dimensions based on the reactor information to determine the time dimension monitoring points and the space dimension monitoring points; and matching the time dimension monitoring points and the space dimension monitoring points to determine the reactor monitoring points.

[0027] Furthermore, this application also includes: extracting near-wall points, center points inside the reactor, bottom points, and near-liquid surface points based on the reactor structure information in the reactor information; configuring temperature sensing units on the near-wall points, center points inside the reactor, bottom points, and near-liquid surface points to obtain near-wall measuring points, center measuring points inside the reactor, bottom measuring points, and near-liquid surface measuring points; and combining the near-wall measuring points, center measuring points inside the reactor, bottom measuring points, and near-liquid surface measuring points to obtain the spatial dimension monitoring points.

[0028] Furthermore, this application also includes: dividing the reactor state information in the reactor information into stages based on the thermal characteristics of the reaction, and determining the thermal characteristic stages, wherein the thermal characteristic stages include the start-up heating stage, the main reaction stage, the post-reaction stage, and the termination cooling stage; dividing the reactor state information into stages based on the process actions, and determining the process action characteristic stages, wherein the process action stages include the initial solvent heating stage, the main reactant feeding stage, the batch feeding stage, and the feeding end stabilization stage; dividing the reactor state information into stages based on the spatial characteristics of the temperature field, and determining the spatial characteristics of the temperature field, wherein the... The temperature field spatial characteristic stages include a temperature rise stage, a main reaction stage, a dead zone stage, and a cooling stage. Based on preset importance weights, the reaction thermal characteristic stage, the process action characteristic stage, and the temperature field spatial characteristic stage are weighted and calculated to obtain a time evolution stage, which includes a main control stage, an auxiliary stage, a constraint stage, and a monitoring stage. Temperature sensing units are configured for the main control stage, the auxiliary stage, the constraint stage, and the monitoring stage to obtain time dimension monitoring points, which include time dimension main control measurement points, time dimension auxiliary measurement points, time dimension constraint measurement points, and time dimension monitoring points.

[0029] Furthermore, this application also includes: performing time evolution perception based on the time evolution stage, calling the time dimension monitoring points according to the perceived time stage to obtain real-time time monitoring points; and determining the reactor monitoring points by matching the real-time time monitoring points with the spatial dimension monitoring points.

[0030] Specifically, reactor structural information refers to the fundamental information extracted from structural parameters of chemical reactors, such as their geometric configuration, internal cavity distribution, heat exchange interface location, and material filling height, which characterizes the temperature field distribution within the reactor. Based on this structural information, spatial location analysis is performed on representative temperature-sensitive areas within the reactor to determine near-wall points, center points, bottom points, and near-liquid surface points. Near-wall points characterize areas significantly affected by the heat exchange medium near the reactor wall; center points characterize the overall temperature state of the main reaction zone within the reactor; bottom points characterize temperature changes in material deposition areas or areas with lagging heat conduction; and near-liquid surface points characterize temperature changes in the gas-liquid interface or near the liquid surface caused by phase changes, volatilization, or other factors.

[0031] After determining the near-wall points, center points, bottom points, and near-liquid surface points of the vessel, temperature sensing units are configured at each point to transform the spatial locations from geometric points into measurement nodes with real-time temperature acquisition capabilities, thus forming near-wall measuring points, center measuring points, bottom measuring points, and near-liquid surface measuring points. The temperature sensing units are used to continuously or periodically acquire the reaction temperature at the corresponding spatial locations, and the acquired temperature data serves as the raw monitoring data reflecting the thermal state of that area.

[0032] Furthermore, by combining and integrating the near-wall measuring points, the center measuring points inside the vessel, the bottom measuring points, and the near-liquid surface measuring points, a temperature measurement set covering the key spatial areas of the reactor is constructed, forming spatial dimension monitoring points. This allows the spatial dimension monitoring points to collaboratively reflect the overall distribution characteristics and changes of the internal temperature field of the reactor from multiple spatial locations.

[0033] Based on the thermal characteristics of the reaction vessel, the process of dividing the reaction vessel state information into stages refers to analyzing the thermal state of the reaction vessel at different operating moments according to the changing laws of heat generation, absorption, and transfer during the chemical reaction process. The reaction process is then segmented according to different thermal behaviors, such as external heating dominance, reaction exothermic dominance, and residual heat decay, thereby determining the stages of the thermal characteristics of the reaction. Specifically, the start-up heating stage characterizes the process where external heating units provide heat to bring the reaction system to the initial reaction conditions in the early stages of the reaction; the main reaction stage characterizes the process where the reaction rate is high and the reaction exothermic or endothermic effects are significant; the post-reaction stage characterizes the process where the heat of reaction gradually decreases after the main reaction is completed, but the system still needs to maintain the process temperature; and the termination cooling stage characterizes the process where the reaction temperature is reduced by cooling methods after the reaction to meet safety or post-processing requirements.

[0034] Furthermore, dividing the reactor state information into stages based on process actions refers to using manual or automated operational behaviors during the reaction process as the basis for stage identification, dividing the reactor operation process, and thus determining the characteristic stages of the process actions. Specifically, the initial solvent heating stage characterizes the operation process of preheating the solvent or system before adding the main reactant; the main reactant feeding stage characterizes the reaction start-up process triggered by the continuous or instantaneous addition of the main reactant to the reactor; the batch feeding stage characterizes the operation process of controlling the reaction rate and exothermic intensity through multiple or continuous feeding methods; and the post-feeding stabilization stage characterizes the process of the reaction system gradually stabilizing under given conditions after the feeding is completed.

[0035] Furthermore, dividing the reactor state information into stages based on the spatial characteristics of the temperature field refers to identifying the process of the temperature field evolving from uniform to non-uniform or from non-uniform to uniform by analyzing the temperature distribution and changes at different spatial locations within the reactor, thereby determining the stages of the spatial characteristics of the temperature field. Specifically, the temperature rise stage characterizes the process where the overall temperature level of the reactor continuously increases over time and the spatial temperature difference gradually becomes apparent; the main reaction stage characterizes the process where the spatial distribution of the temperature field is highly correlated with the reaction intensity and local temperature changes are significant; the dead zone stage characterizes the process where the regional characteristics of local temperature response lag are dominant due to limitations in stirring efficiency, material distribution, or heat transfer conditions; and the cooling stage characterizes the process where the overall temperature field within the reactor evolves towards a lower temperature through cooling methods.

[0036] Based on preset importance weights, a weighted calculation is performed on the reaction thermal characteristic stage, the process action characteristic stage, and the temperature field spatial characteristic stage. This means that according to the differences in the degree of influence of different stage division dimensions on temperature control decisions, corresponding weight coefficients are assigned to the stage division results, and multi-dimensional stage information is fused and calculated to obtain a time evolution stage that can comprehensively reflect the current reaction state. Among them, the main control stage is used to characterize the stage state that plays a dominant role in temperature control decisions; the auxiliary stage is used to characterize the stage state that supplements and corrects the main control stage; the constraint stage is used to characterize the stage state that imposes restrictions on the temperature regulation amplitude and regulation method; and the monitoring stage is used to characterize the stage state whose main purpose is state observation and risk warning.

[0037] Subsequently, temperature sensing units are configured for the main control stage, auxiliary stage, constraint stage, and monitoring stage. This refers to selecting key temperature measurement locations within the reactor corresponding to each stage and configuring temperature sensing units based on their functional roles in temperature control at different time evolution stages, thereby forming time-dimensional monitoring points. Among them, the time-dimensional main control measurement points provide the primary feedback for closed-loop temperature control, the time-dimensional auxiliary measurement points are used to correct and supplement the data from the main control measurement points, the time-dimensional constraint measurement points are used to monitor critical areas that may cause safety or process risks, and the time-dimensional monitoring points are used to continuously observe the overall temperature change trend.

[0038] Time evolution perception based on time evolution stages refers to the continuous identification and judgment of the reaction process during reactor operation according to the aforementioned determined time evolution stages. Time evolution perception is used to determine the stage attributes of the reaction process in real time, clarifying the current state of the reactor at different stages such as main control, auxiliary, constraint, or monitoring, thereby providing a stage-based basis for subsequent monitoring and control. The time evolution stages characterize the comprehensive operational features of the reaction process as it changes over time.

[0039] Furthermore, calling the time dimension monitoring points according to the perceived time stage means that after determining the current time evolution stage, based on the monitoring strategy corresponding to that stage, a set of monitoring points matching the current stage is selected from pre-configured time dimension main control monitoring points, time dimension auxiliary monitoring points, time dimension constraint monitoring points, or time dimension monitoring points to achieve targeted collection of the reaction state under that time stage, thereby forming real-time time monitoring points that can reflect the key characteristics of the current stage. The real-time time monitoring points are used to carry the effective temperature sensing data source under the current stage.

[0040] Furthermore, matching the spatial dimension monitoring points based on real-time time monitoring points refers to associating and mapping real-time time monitoring points with pre-constructed spatial dimension monitoring points. By matching their correspondence in the spatial location and time stage attributes of the reactor, the specific set of spatial measurement points that need to participate in monitoring and control at the current time stage is determined. The spatial dimension monitoring points are used to characterize the temperature acquisition capability at different spatial locations inside the reactor.

[0041] Therefore, determining the reactor monitoring point means forming a set of monitoring points that simultaneously possess both time stage attributes and spatial location attributes after completing the joint matching of time and spatial dimensions. This reactor monitoring point is used to perceive the temperature status of key areas of the reactor in real time during the current operating stage, thereby providing accurate, effective and stage-specific input data for subsequent closed-loop temperature control decisions.

[0042] It receives signals from the monitoring points of the reactor in real time and establishes real-time data.

[0043] Specifically, real-time reception of acquisition signals from reactor monitoring points refers to the continuous acquisition of temperature sampling signals generated at various monitoring locations of the reactor during its operation, through temperature sensing units, signal acquisition units, and communication interfaces installed at the reactor monitoring points. The acquired signals are used to characterize the instantaneous temperature state and its changes inside the reactor or at relevant locations on the reactor body. Real-time reception emphasizes that the acquisition process is synchronized with the reaction process and does not rely on offline sampling or post-processing.

[0044] Furthermore, establishing real-time data refers to timestamping, formatting, and validating the real-time received acquisition signals, and organizing and storing them according to a preset data structure to form a data sequence that can continuously reflect the temperature state of the reactor over time. The real-time data is used as the basic input data for subsequent temperature deviation calculation, trend analysis, and closed-loop control decisions.

[0045] An adaptive decision-making channel is established, and the real-time data is transmitted to the adaptive decision-making channel. Based on the temperature deviation calculation sub-channel and the control decision generation sub-channel within the adaptive decision-making channel, a combined judgment is made to establish a stage control command.

[0046] Furthermore, this application also includes: calculating the temperature deviation between the real-time data and the target temperature based on the temperature deviation calculation sub-channel to obtain the temperature deviation and the temperature deviation change trend; and generating the stage control command for the time evolution stage based on the control decision generation sub-channel according to the temperature deviation using a closed-loop control mechanism, wherein the closed-loop control mechanism includes proportional regulation, integral regulation and derivative regulation.

[0047] Furthermore, this application also includes: determining whether the real-time data is within a safety margin, wherein the safety margin includes temperature safety margin, pressure safety margin, and material safety margin; if so, determining whether the real-time data is within a process allowable margin; if so, determining whether the real-time data is within a time capability margin; if so, determining whether the real-time data is within a control logic compatibility margin, wherein the control logic compatibility margin characterizes the preset value of interference caused by the coupling effect between control logic and temperature control on non-temperature parameters in the subsequent stages of the time evolution stage; if so, predicting the delay control effect of the subsequent stages based on the time evolution stage, and performing cross-stage compensation adjustment of the chemical reactor based on the delay control effect through subsequent stage compensation; inputting the reaction temperature of the cross-stage compensation adjustment signal into the adaptive decision discrimination channel, updating the control output, and completing the allowable deviation transmission.

[0048] Furthermore, this application also includes: determining the time interval from the end of the real-time phase to the start of the subsequent phase; acquiring real-time data of the chemical reactor during the real-time phase, retrieving a hysteresis correlation model matching the real-time phase from a pre-stored model library, wherein the hysteresis correlation model is trained based on historical operating data and is used to characterize the influence of the temperature response on the subsequent temperature field of the subsequent phase after the time interval under the real-time operating conditions of the real-time phase, including the heat transfer hysteresis effect and the material reaction accumulation effect; inputting the real-time data into the hysteresis correlation model for extrapolation and calculation, generating a predicted temperature curve for the subsequent phase without additional intervention, using the predicted temperature curve as the delayed control effect, and the predicted temperature curve characterizing the dynamic temperature change after the hysteresis time from the current moment.

[0049] Furthermore, this application also includes: taking the ideal temperature curve of the subsequent stage as the control target, comparing the predicted temperature curve with the ideal temperature curve, and obtaining a compensation target value through the temperature deviation region; based on the compensation target value, combined with the time length and heat exchange capacity of the subsequent stage, calculating the compensation adjustment amount of the subsequent stage; setting a decay strategy for the compensation intensity, obtaining the basic control command of the subsequent stage, converting the compensation adjustment amount from the basic control command into a compensation control command according to the decay strategy, generating a subsequent control command, and completing cross-stage compensation adjustment.

[0050] Specifically, establishing an adaptive decision-making channel refers to constructing a functional pathway for analyzing, judging, and making decisions based on real-time temperature information. This channel identifies the operational characteristics of different stages of the reaction process and dynamically adjusts the control logic and parameters according to changes in the reactor's current operating conditions. "Adaptive" characterizes the channel's ability to automatically correct decision-making criteria as temperature conditions change, while "decision-making" determines whether control actions are triggered and their intensity. The training process for the adaptive decision-making channel includes constructing a training dataset. This involves collecting historical operational data from the reactor under different reaction types, process parameters, and time evolution stages. This historical data includes temperature data from multiple spatial monitoring points, corresponding timestamps, target temperature setpoints, stage control command records, and actual temperature response results after execution, forming a sample dataset with complete temporal correlation. Preprocessing of the training data includes removing or correcting abnormal temperature values, aligning data from different monitoring points, normalizing temperature data, and uniformly encoding time evolution stage labels to ensure consistency of input data in terms of numerical scale and semantics, thereby improving the stability and convergence of the training process. Subsequently, the preprocessed training data is input into the temperature deviation calculation sub-channel. By comparing historical actual temperatures with corresponding target temperatures, the historical temperature deviation sequence and its changing trend are calculated. The temperature deviation and its changing trend are used as feature inputs to characterize the dynamic behavior of temperature shifts at different response stages, thus providing fundamental features for subsequent control decision learning. Further, during the training process of the control decision generation sub-channel, temperature deviation features, time evolution stage labels, and historical control commands are used as joint inputs. By comparing the temperature response effects before and after the execution of historical control commands, an objective function is constructed with temperature stability, overshoot, settling time, and steady-state error as evaluation indicators. The proportional, integral, and derivative control parameters are iteratively updated with the optimization objective of minimizing temperature control error and control oscillation. During training, a stage constraint mechanism is introduced, allowing different time evolution stages to correspond to different parameter search spaces or weight update strategies. For example, the response weights of proportional and derivative controls are increased in the main control stage, and the ability of integral control to correct steady-state error is enhanced in the auxiliary stage. This ensures that the training results reflect the impact of stage differences on the temperature control strategy. In addition, the stability and generalization ability of the trained control parameter combinations are evaluated by cross-validation or batch training. The parameter sets that are stable under multiple historical conditions and whose control effects meet the preset threshold are used as the initial control parameters of the adaptive decision discrimination channel, thereby avoiding overfitting of parameters to a single condition.Finally, the adaptive decision-making channel, after completing the training, is deployed in the actual reactor operating environment and continues to receive real-time data during online operation. By comparing the actual control effect with the training period target, the control parameters are slightly corrected online, thus forming a closed-loop training and application mechanism that combines offline training and online adaptive updates.

[0051] Furthermore, the temperature deviation calculation subchannel calculates the temperature deviation between the real-time data and the target temperature. This involves comparing the current actual temperature value reflected in the real-time data with the pre-set target temperature to calculate the difference between the two. The temperature deviation quantifies the degree to which the current reactor temperature deviates from the target state. Obtaining the temperature deviation and its trend involves performing time-series analysis on the temperature deviation over a continuous period, based on the single-moment temperature deviation calculation. This analysis reveals the direction, rate, and stability of the temperature deviation over time. The temperature deviation trend characterizes the dynamic evolution of the temperature control process, providing a basis for subsequent control strategy adjustments.

[0052] Subsequently, based on the control decision generation sub-channel, according to the temperature deviation, the closed-loop control mechanism is used to generate the stage control command for the time evolution stage. This means taking the temperature deviation and its changing trend as feedback input, and combining it with the control requirements of the current time evolution stage to calculate the control amount, thereby generating a stage control command that matches the current stage. The stage control command is used to indicate specific temperature control operations such as heating, cooling, or maintaining.

[0053] The closed-loop control mechanism includes proportional regulation, integral regulation, and derivative regulation. It refers to the control decision generation sub-channel using a composite control mechanism composed of proportional, integral, and derivative regulation. Proportional regulation generates control action in real time based on the current temperature deviation. Integral regulation accumulates and corrects historical temperature deviations to eliminate steady-state errors. Derivative regulation predictively corrects the temperature deviation trend. These mechanisms work together to improve the response speed, stability, and control accuracy of temperature control.

[0054] Determining whether real-time data is within the safety margin involves comparing and analyzing the temperature, pressure, and material state data collected in real-time from the reactor monitoring points with pre-set safety threshold ranges. This confirms that the current operating state has not reached the dangerous range that could lead to equipment damage, reaction runaway, or safety accidents. The safety margin characterizes the reactor's operational redundancy within an acceptable safety range. Specifically, it includes a temperature safety margin to limit abnormal increases or decreases in reaction temperature, a pressure safety margin to prevent overpressure or negative pressure operation of the reactor, and a material safety margin to constrain abnormal material ratios, reactant concentrations, or phase states. This provides basic safety preconditions for subsequent control decisions.

[0055] After confirming that the real-time data is within the safety margin, further determining whether the real-time data is within the process allowable margin means comparing the current operating parameters of the reactor with the allowed operating range in the process document or process model to verify whether the current temperature, heating rate and related process parameters meet the process constraints defined by the reaction route, reaction rate and product quality requirements, thereby ensuring that even under safe conditions, the reaction process is still within an acceptable process control window.

[0056] Under the premise of meeting the process allowance margin requirements, it is further determined whether the real-time data is within the time capability margin. The time capability margin is used to characterize the degree of matching between the response capability of the reactor and its heating, cooling and heat exchange systems to temperature changes and the time constraint in the current time evolution stage. That is, it is determined whether the existing equipment adjustment capability can complete the target temperature adjustment or deviation correction within a predetermined time, thereby avoiding control lag or excessive intervention due to insufficient adjustment capability.

[0057] After confirming that the time capability margin meets the requirements, it is further determined whether the real-time data is within the control logic compatibility margin. The control logic compatibility margin is used to characterize whether the interaction between the temperature control strategy and other control logic is within an acceptable range in the current time evolution stage and its subsequent stages. Specifically, it reflects whether the coupling interference generated by the temperature regulation behavior on non-temperature parameters such as pressure, stirring state, and material reaction rate during the execution process is lower than the preset interference threshold, thereby avoiding chain deviations caused by control logic conflicts.

[0058] When all multiple margin judgments are satisfied, determining the time interval from the end of the real-time stage to the start of the subsequent stage refers to identifying the stage switching node between the current real-time stage and its adjacent subsequent stage based on the division of time evolution stages, and calculating the duration from the end of the real-time stage to the start of the subsequent stage by analyzing the process flow settings, operating cycle time and reaction kinetic characteristics. The time interval is used to quantify the lag period required for the control action in the reactor system to be transmitted to the subsequent stage.

[0059] After determining the time interval, real-time data of the chemical reactor in the real-time stage is acquired, and a hysteresis correlation model matching the real-time stage is retrieved from the pre-stored model library. This means that multi-dimensional real-time data reflecting the operating status of the reactor is collected and organized in the current stage, and a corresponding hysteresis correlation model is selected from the model library based on the process characteristics, operating mode and heat load level of the real-time stage. The hysteresis correlation model is trained based on a large amount of historical operating data and is used to characterize the influence of temperature response on the temperature field distribution and evolution in the subsequent stage after the time interval under the actual operating conditions in the real-time stage. The influence law specifically includes the heat transfer hysteresis effect caused by the thermal inertia of the equipment and the limitation of the heat exchange path, as well as the material reaction accumulation effect caused by the continuous reaction of reactants or the accumulation of intermediate products.

[0060] Subsequently, the real-time data is input into the lag correlation model for extrapolation calculation. This refers to using the lag correlation model to perform mathematical or data-driven calculations and reasoning on the real-time operating state of the current stage, in order to simulate the natural evolution of temperature over time without the application of additional control intervention, thereby generating a predicted temperature curve for the subsequent stage. The predicted temperature curve serves as the effect of the delayed regulation and is used to characterize the dynamic change trend and amplitude characteristics of temperature after the lag time from the current moment.

[0061] After achieving the delayed control effect, the ideal temperature curve of the subsequent stage is used as the control target. This means that for the subsequent stage that is about to be entered in the time evolution stage, an ideal temperature change reference curve is set in advance according to the process requirements, reaction kinetic characteristics and product quality control objectives. This ideal temperature curve is used as the target benchmark for temperature control in the subsequent stage to evaluate the rationality of actual or predicted temperature behavior.

[0062] Furthermore, by comparing the predicted temperature curve with the ideal temperature curve, the compensation target value is obtained through the temperature deviation region. This means comparing and analyzing the predicted temperature curve formed by the delayed regulation effect with the ideal temperature curve on the same time scale. By calculating the degree of deviation of the two curves on the time axis and their cumulative area, the temperature deviation region used to quantify the degree of influence of the temperature deviation is obtained. Based on this temperature deviation region, the compensation target value that needs to be corrected in the subsequent stage is determined. The compensation target value is used to characterize the heat regulation target that needs to be supplemented or reduced to eliminate or reduce the prediction deviation.

[0063] After obtaining the compensation target value, the compensation adjustment amount for the subsequent stage is calculated based on the compensation target value and the time length and heat exchange capacity of the subsequent stage. This means that, on the basis of a clear compensation target, the compensation target is allocated and quantified by comprehensively considering the effective time window available for adjustment in the subsequent stage and the maximum and effective heat exchange capacity of the reactor heating, cooling and heat exchange system in that stage, so as to determine the actual compensation adjustment amount that can be implemented in the subsequent stage, so as to ensure that the compensation measures can be executed within the constraints of physical and equipment capabilities.

[0064] Subsequently, the attenuation strategy for the compensation intensity is set. This refers to the pre-design of attenuation rules that gradually weaken the compensation adjustment amount over time or as the stage progresses, in order to address the over-adjustment or control oscillation problems that may be caused during the cross-stage compensation process. This is used to smooth the execution process of compensation control and improve system stability.

[0065] Obtaining the basic control instructions for subsequent stages refers to the standard temperature control instructions generated based on the conventional closed-loop temperature control strategy of subsequent stages without the introduction of cross-stage compensation. The basic control instructions are used to maintain normal temperature control operation in subsequent stages.

[0066] Furthermore, converting the compensation adjustment amount from the basic control command to the compensation control command according to the attenuation strategy means superimposing or modifying the compensation adjustment amount based on the basic control command and the attenuation strategy, so that the compensation effect is gradually and controlledly integrated into the temperature control process in the subsequent stage, thereby forming a compensation control command for actual execution.

[0067] Finally, generating subsequent control commands and completing cross-stage compensation adjustment means using the attenuated compensation control commands as the final temperature control output for the subsequent stage to guide the heating or cooling operation of the reactor in the subsequent stage, thereby achieving effective compensation for the temperature lag effect of the previous stage and ensuring temperature continuity and control stability between different time evolution stages.

[0068] Subsequently, the temperature change caused by the cross-stage compensation adjustment signal is used as feedback input to the adaptive decision discrimination channel. This means that the actual temperature response after compensation adjustment is reintegrated into the control decision analysis process. By comparing the control effects before and after compensation, the control output parameters are updated, thereby achieving the orderly transmission and dynamic correction of allowable deviations at different time evolution stages.

[0069] The heating unit, cooling unit, and heat exchange medium flow rate of the chemical reactor are adjusted according to the stage control instructions, and the adjustment signals are collected.

[0070] Specifically, adjusting the heating unit, cooling unit, and heat exchange medium flow rate of the chemical reactor according to the stage control command means that after the adaptive decision-making channel generates a stage control command corresponding to the current time evolution stage, the control command is parsed into specific execution parameters and applied to the heating unit, cooling unit, and heat exchange medium flow rate adjustment mechanism of the reactor respectively. The heating unit is used to provide heat energy to the reactor, the cooling unit is used to remove excess heat from the reactor, and the heat exchange medium flow rate is used to adjust the amount of medium participating in heat exchange per unit time, thereby achieving fine control of the reactor temperature through multi-execution terminal collaboration.

[0071] Furthermore, the acquisition of adjustment signals during the adjustment process refers to the synchronous monitoring and data acquisition of the actual operating status of each execution unit while the heating, cooling and heat exchange medium flow rate adjustment actions are executed. The adjustment signals include, but are not limited to, heating power output value, cooling intensity set value, actual value of heat exchange medium flow rate and its change process parameters. These adjustment signals are used to reflect the actual implementation of the stage control command at the physical execution level.

[0072] Based on the temperature feedback input of the adjustment signal, the adaptive decision-making channel is updated to update the control output, forming a closed-loop temperature control circuit.

[0073] Specifically, the reaction temperature feedback input adaptive decision-making channel based on the adjustment signal refers to the process of collecting and acquiring the actual reaction temperature change inside the reactor in real time according to the execution state corresponding to the adjustment signal after the heating unit, cooling unit and heat exchange medium flow rate of the chemical reactor are adjusted. The reaction temperature is then used as feedback information and re-inputted into the adaptive decision-making channel. The reaction temperature feedback is used to reflect the temperature response effect of the control command in the actual physical system.

[0074] Furthermore, updating the control output means that after receiving the reaction temperature feedback, the adaptive decision-making channel compares and analyzes the feedback temperature with the target temperature and historical control states, recalculates the temperature deviation and its changing trend, and dynamically corrects the control parameters or regulation strategies accordingly, thereby generating a new control output to adapt to the changes in the current operating conditions of the reactor.

[0075] Ultimately, forming a closed-loop temperature control circuit means that by applying the control output to the reactor execution unit and feeding back the resulting reaction temperature to the decision-making channel, a continuous loop control link is constructed, allowing the temperature regulation process to iterate continuously under the feedback correction mechanism, thereby achieving stable control and improved precision of the chemical reactor temperature.

[0076] In summary, the closed-loop control method for temperature control of chemical reactors provided in this application has the following technical effects: by achieving the technical goal of adaptive closed-loop temperature control based on the joint sensing of time evolution stage and spatial temperature field, it achieves the technical effects of precise adjustment of temperature changes in stages throughout the reaction process, effectively suppressing temperature lag deviation, and maintaining temperature continuity and control stability between different stages.

[0077] Example 2: Based on the same inventive concept as the closed-loop control method for temperature control of a chemical reactor in the foregoing examples, this application also provides a closed-loop control system for temperature control of a chemical reactor. Please refer to the appendix. Figure 2 The system includes: a reactor monitoring point construction module 1, used to dynamically deploy temperature sensing units driven by stage perception based on reactor information of the chemical reactor, and construct reactor monitoring points; a real-time data establishment module 2, used to receive the collected signals from the reactor monitoring points in real time and establish real-time data; a stage control command establishment module 3, used to establish an adaptive decision-making channel, transmit the real-time data to the adaptive decision-making channel, and perform combined judgment based on the temperature deviation calculation sub-channel and control decision generation sub-channel within the adaptive decision-making channel to establish stage control commands; an adjustment signal acquisition module 4, used to adjust the heating unit, cooling unit, and heat exchange medium flow rate of the chemical reactor according to the stage control commands, and acquire adjustment signals; and a feedback module 5, used to input the reaction temperature feedback into the adaptive decision-making channel based on the adjustment signals, update the control output, and form a closed-loop temperature control loop.

[0078] Furthermore, the closed-loop controlled chemical reactor temperature control system is also used for: analyzing the monitoring points of the chemical reactor in the time and space dimensions based on the reactor information, determining the time dimension monitoring points and the space dimension monitoring points; and matching the time dimension monitoring points and the space dimension monitoring points to determine the reactor monitoring points.

[0079] Furthermore, the closed-loop controlled chemical reactor temperature control system is also used to: extract near-wall points, center points inside the reactor, bottom points, and near-liquid surface points based on the reactor structure information in the reactor information; configure temperature sensing units for the near-wall points, center points inside the reactor, bottom points, and near-liquid surface points to obtain near-wall measuring points, center measuring points inside the reactor, bottom measuring points, and near-liquid surface measuring points; and combine the near-wall measuring points, center measuring points inside the reactor, bottom measuring points, and near-liquid surface measuring points to obtain the spatial dimension monitoring points.

[0080] Furthermore, the closed-loop controlled chemical reactor temperature control system is also used for: dividing the reactor state information in the reactor information into stages based on the reaction thermal characteristics to determine the reaction thermal characteristic stages, wherein the reaction thermal characteristic stages include the start-up heating stage, the main reaction stage, the post-reaction stage, and the termination cooling stage; dividing the reactor state information into stages based on the process actions to determine the process action characteristic stages, wherein the process action stages include the initial solvent heating stage, the main reactant feeding stage, the batch feeding stage, and the feeding end stabilization stage; and dividing the reactor state information into stages based on the temperature field spatial characteristics to determine the temperature field spatial characteristic stages. The temperature field spatial characteristic stage includes a temperature rise stage, a main reaction stage, a dead zone stage, and a cooling stage. Based on a preset importance weight, the reaction thermal characteristic stage, the process action characteristic stage, and the temperature field spatial characteristic stage are weighted and calculated to obtain a time evolution stage, which includes a main control stage, an auxiliary stage, a constraint stage, and a monitoring stage. Temperature sensing units are configured for the main control stage, the auxiliary stage, the constraint stage, and the monitoring stage to obtain the time dimension monitoring points, which include time dimension main control measurement points, time dimension auxiliary measurement points, time dimension constraint measurement points, and time dimension monitoring points.

[0081] Furthermore, the closed-loop controlled chemical reactor temperature control system is also used for: performing time evolution sensing based on the time evolution stage, calling the time dimension monitoring points according to the sensing time stage to obtain real-time time monitoring points; and determining the reactor monitoring points by matching the real-time time monitoring points with the spatial dimension monitoring points.

[0082] Furthermore, the closed-loop controlled chemical reactor temperature control system is also used for: calculating the temperature deviation between the real-time data and the target temperature based on the temperature deviation calculation subchannel to obtain the temperature deviation and the temperature deviation change trend; and generating the stage control command for the time evolution stage based on the control decision generation subchannel according to the temperature deviation using the closed-loop control mechanism, wherein the closed-loop control mechanism includes proportional regulation, integral regulation and derivative regulation.

[0083] Furthermore, the closed-loop controlled chemical reactor temperature control system is also used to: determine whether the real-time data is within a safety margin, wherein the safety margin includes temperature safety margin, pressure safety margin, and material safety margin; if so, determine whether the real-time data is within a process allowable margin; if so, determine whether the real-time data is within a time capability margin; if so, determine whether the real-time data is within a control logic compatibility margin, wherein the control logic compatibility margin characterizes the preset value of interference caused by the coupling effect between control logic and temperature control on non-temperature parameters in the subsequent stages of the time evolution stage; if so, predict the delay control effect of the subsequent stages based on the time evolution stage, and perform cross-stage compensation adjustment of the chemical reactor based on the delay control effect through subsequent stage compensation; input the reaction temperature of the cross-stage compensation adjustment signal into the adaptive decision discrimination channel, update the control output, and complete the allowable deviation transmission.

[0084] Furthermore, the closed-loop controlled chemical reactor temperature control system is also used to: determine the time interval from the end of the real-time stage to the start of the subsequent stage; acquire real-time data of the chemical reactor during the real-time stage, retrieve a hysteresis correlation model matching the real-time stage from a pre-stored model library, wherein the hysteresis correlation model is trained based on historical operating data and is used to characterize the influence of the temperature response on the subsequent temperature field of the subsequent stage after the time interval under the real-time operating conditions of the real-time stage, including the heat transfer hysteresis effect and the material reaction accumulation effect; input the real-time data into the hysteresis correlation model for deduction and calculation, generate a predicted temperature curve for the subsequent stage without additional intervention, and use the predicted temperature curve as the delayed control effect, wherein the predicted temperature curve characterizes the dynamic temperature change after the hysteresis time from the current moment.

[0085] Furthermore, the closed-loop controlled chemical reactor temperature control system is also used for: taking the ideal temperature curve of the subsequent stage as the control target, comparing the predicted temperature curve with the ideal temperature curve, and obtaining a compensation target value through the temperature deviation region; based on the compensation target value, combined with the time length and heat exchange capacity of the subsequent stage, calculating the compensation adjustment amount of the subsequent stage; setting a decay strategy for the compensation intensity, obtaining the basic control command of the subsequent stage, converting the compensation adjustment amount from the basic control command into a compensation control command according to the decay strategy, generating a subsequent control command, and completing cross-stage compensation adjustment.

[0086] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The closed-loop control method and specific examples of the chemical reactor temperature control in Embodiment 1 described above are also applicable to the closed-loop control system of the chemical reactor temperature control in this embodiment. Through the detailed description of the closed-loop control method of the chemical reactor temperature control described above, those skilled in the art can clearly understand the closed-loop control system of the chemical reactor temperature control in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0087] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0088] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A closed-loop control method for temperature control of a chemical reactor, characterized in that, include: Based on the reactor information of the chemical reactor, temperature sensing units are dynamically deployed in a phase-sensing manner to construct reactor monitoring points. Real-time data acquisition signals are received from the monitoring points of the reactor and real-time data is established. An adaptive decision-making channel is established, and the real-time data is transmitted to the adaptive decision-making channel. Based on the temperature deviation calculation sub-channel and the control decision generation sub-channel within the adaptive decision-making channel, a combined judgment is made to establish a stage control command. The heating unit, cooling unit, and heat exchange medium flow rate of the chemical reactor are adjusted according to the stage control instructions, and the adjustment signals are collected. Based on the temperature feedback input of the adjustment signal, the adaptive decision-making channel is updated to update the control output, forming a closed-loop temperature control circuit.

2. The closed-loop control method for temperature control of a chemical reactor as described in claim 1, characterized in that, Based on the reactor information from the chemical reactor, temperature sensing units are dynamically deployed in a phase-sensing manner to construct reactor monitoring points, including: Based on the reactor information, the monitoring points of the chemical reactor are analyzed in both time and space dimensions to determine the time and space dimension monitoring points. The time-dimensional monitoring points and spatial-dimensional monitoring points are matched to determine the monitoring points of the reactor.

3. The closed-loop control method for temperature control of a chemical reactor as described in claim 2, characterized in that, Based on the reactor information, a spatial dimension monitoring point analysis is performed on the chemical reactor to determine the spatial dimension monitoring points, including: Based on the reactor structure information in the reactor information, extract the near wall points, the center point inside the reactor, the bottom point, and the near liquid surface point; Temperature sensing units are configured at the near-wall points, the center point inside the vessel, the bottom point, and the near-liquid surface points to obtain the near-wall measuring points, the center point inside the vessel, the bottom point, and the near-liquid surface measuring points. By combining the measuring points near the vessel wall, the center measuring point inside the vessel, the bottom measuring point, and the near-liquid surface measuring point, the spatial dimension monitoring points are obtained.

4. The closed-loop control method for temperature control of a chemical reactor as described in claim 2, characterized in that, Based on the reactor information, a time-dimensional monitoring point analysis is performed on the chemical reactor to determine the time-dimensional monitoring points, including: Based on the reaction heat characteristics, the reaction vessel state information in the reaction vessel information is divided into stages to determine the reaction heat characteristic stages, wherein the reaction heat characteristic stages include the start-up heating stage, the main reaction stage, the post-reaction stage, and the termination cooling stage. The reactor state information is divided into stages based on the process actions to determine the characteristic stages of the process actions. The process action stages include the initial solvent heating stage, the main reactant feeding stage, the batch feeding stage, and the feeding end stabilization stage. The state information of the reactor is divided into stages based on the spatial characteristics of the temperature field, and the spatial characteristics of the temperature field are determined. The spatial characteristics of the temperature field include the temperature rise stage, the main reaction stage, the dead zone stage, and the cooling stage. Based on preset importance weights, the reaction thermal characteristic stage, the process action characteristic stage, and the temperature field spatial characteristic stage are weighted and calculated to obtain the time evolution stage, wherein the time evolution stage includes a main control stage, an auxiliary stage, a constraint stage, and a monitoring stage. Temperature sensing units are configured for the main control stage, auxiliary stage, constraint stage and monitoring stage to obtain the time dimension monitoring points, wherein the time dimension monitoring points include time dimension main control measurement points, time dimension auxiliary measurement points, time dimension constraint measurement points and time dimension monitoring points.

5. The closed-loop control method for temperature control of a chemical reactor as described in claim 4, characterized in that, Matching the time-dimension monitoring points and the spatial-dimension monitoring points to determine the reactor monitoring points includes: Based on the aforementioned time evolution stage, time evolution perception is performed, and the time dimension monitoring points are called according to the perceived time stage to obtain real-time time monitoring points; The monitoring point of the reactor is determined by matching the real-time time monitoring point with the spatial dimension monitoring point.

6. The closed-loop control method for temperature control of a chemical reactor as described in claim 4, characterized in that, The real-time data is transmitted to the adaptive decision-making channel. Based on the temperature deviation calculation sub-channel and the control decision generation sub-channel within the adaptive decision-making channel, a combined judgment is made to establish a staged control command, including: Based on the temperature deviation calculation sub-channel, the real-time data and the target temperature are calculated to obtain the temperature deviation and the temperature deviation change trend. Based on the control decision generation sub-channel, and according to the temperature deviation, a closed-loop control mechanism is used to generate the stage control command for the time evolution stage, wherein the closed-loop control mechanism includes proportional regulation, integral regulation and derivative regulation.

7. The closed-loop control method for temperature control of a chemical reactor as described in claim 4, characterized in that, Before establishing phased control instructions, the following are included: Determine whether the real-time data is within the safety margin, wherein the safety margin includes temperature safety margin, pressure safety margin and material safety margin; If so, determine whether the real-time data is within the process allowable margin; If so, determine whether the real-time data is within the time capability margin; If so, determine whether the real-time data is within the control logic compatibility margin, wherein the control logic compatibility margin represents the preset value of interference caused by the coupling effect between the control logic and temperature control on non-temperature parameters in the subsequent stages of the time evolution stage. If, based on the time evolution stage, the effect of delayed regulation in subsequent stages is predicted, and based on the effect of delayed regulation, the chemical reactor is adjusted across stages through compensation in subsequent stages. The adaptive decision-making channel is used to respond to the temperature input of the cross-stage compensation and adjustment signal, update the control output, and complete the allowable deviation transmission.

8. The closed-loop control method for temperature control of a chemical reactor as described in claim 7, characterized in that, Predicting the effect of delay regulation in subsequent stages based on the aforementioned time evolution stage includes: Determine the time interval between the end of the real-time phase and the start of the subsequent phase; The real-time data of the chemical reactor in the real-time stage is obtained, and the hysteresis correlation model matching the real-time stage is retrieved from the pre-stored model library. The hysteresis correlation model is trained based on historical operating data and is used to characterize the influence of the temperature response on the subsequent temperature field of the subsequent stage after the time interval under the real-time operating conditions in the real-time stage, including the heat transfer hysteresis effect and the material reaction accumulation effect. The real-time data is input into the lag correlation model for inference and calculation, generating the predicted temperature curve for the subsequent stage without additional intervention. The predicted temperature curve is used as the effect of the delayed regulation, and the predicted temperature curve represents the dynamic temperature change after the lag time from the current moment.

9. The closed-loop control method for temperature control of a chemical reactor as described in claim 8, characterized in that, Based on the delayed control effect, cross-stage compensation adjustment of the chemical reactor is performed through subsequent stage compensation, including: Using the ideal temperature curve of the subsequent stage as the control target, the predicted temperature curve is compared with the ideal temperature curve, and the compensation target value is obtained through the temperature deviation region. Based on the compensation target value, and combined with the time length and heat exchange capacity of the subsequent stages, the compensation adjustment amount for the subsequent stages is calculated. A decay strategy for the compensation intensity is set, the basic control command for the subsequent stage is obtained, the compensation adjustment amount is converted from the basic control command into a compensation control command according to the decay strategy, the subsequent control command is generated, and the cross-stage compensation adjustment is completed.

10. A closed-loop temperature control system for a chemical reactor, characterized in that, The steps for implementing the closed-loop control method for temperature control of a chemical reactor according to any one of claims 1 to 9 include: The reactor monitoring point construction module is used to dynamically deploy temperature sensing units driven by stage perception based on the reactor information of the chemical reactor, and construct reactor monitoring points. The real-time data establishment module is used to receive the collected signals from the monitoring points of the reactor in real time and establish real-time data. A phase control command establishment module is used to establish an adaptive decision-making channel, transmit the real-time data to the adaptive decision-making channel, and perform a combined judgment based on the temperature deviation calculation sub-channel and the control decision generation sub-channel within the adaptive decision-making channel to establish a phase control command; The adjustment signal acquisition module is used to adjust the heating unit, cooling unit, and heat exchange medium flow rate of the chemical reactor according to the stage control command, and to acquire the adjustment signal. The feedback module is used to update the control output based on the temperature feedback input of the adjustment signal to the adaptive decision-making channel, forming a closed-loop temperature control circuit.

Images