Intelligent temperature control method for preparation reaction process of oligomeric anti-swelling ammonium salt

Abstract

The invention relates to the technical field of industrial data acquisition and advanced process control, and discloses an intelligent temperature control method for an oligomeric anti-swelling ammonium salt preparation reaction process. The method comprises the following steps: identifying a reaction stage switching critical window based on a multi-point temperature field and thermodynamic parameters; a fifth-order polynomial nonlinear temperature setting track with continuous first-order and second-order derivatives is generated in the window; in combination with multi-area temperature measurement weighted average feedback and reaction heat release rate feedforward, the jacket medium flow is dynamically adjusted through feedforward-feedback compound control. The system comprises a multi-source data acquisition unit, a stage judgment unit, a track generation unit, a feedback control unit, a feed-forward compensation unit and an execution driving unit. The thermal disturbance is effectively inhibited, the standard deviation of molecular weight distribution of the product is reduced by more than 40%, and the batch consistency is remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of industrial data acquisition and advanced process control technology, specifically relating to an intelligent temperature control method for the preparation reaction process of oligomeric anti-swelling ammonium salts. Background Technology

[0002] In the fields of fine chemicals and oilfield chemistry, oligomeric anti-swelling ammonium salts, as an important class of shale inhibitors, are widely used in drilling fluid systems to stabilize the wellbore and prevent clay hydration and swelling. Their synthesis typically employs a multi-step batch reaction process, including initiation, polymerization, and ripening stages, each with strict and differentiated temperature control requirements. Temperature, as a core parameter affecting reaction rate, molecular chain growth, and product structural uniformity, directly determines the performance indicators of the final product, especially the narrowness of the molecular weight distribution and the regularity of repeating units.

[0003] The preparation process of oligomeric anti-swelling ammonium salts often involves switching from a significantly exothermic polymerization phase to a maturation phase in thermodynamic equilibrium. This switching process requires the reaction system temperature to be rapidly adjusted from a higher setpoint to a lower maintenance value, or vice versa. However, traditional temperature control strategies generally employ a step-type setpoint change method, that is, directly assigning the target temperature to the constant value of the target stage at the instant of stage switching. This abrupt command causes a violent response in the heating / cooling system, resulting in instantaneous unevenness of the temperature field of the medium inside the reactor and generating significant thermal stress shocks.

[0004] Neither the classic PID-based controller nor the improved scheme incorporating feedforward compensation adequately considers the dynamic continuity requirement of the temperature trajectory during stage switching. Step changes in the setpoint cause localized overcooling or overheating in the reactor wall and central region, disrupting the thermodynamic homogeneity of the reaction system. This leads to side reactions or fluctuations in chain termination / transfer rates, ultimately resulting in a significant broadening of the product's molecular weight distribution and poor batch-to-batch repeatability. Especially in large-scale production, the thermal inertia effect is further amplified, making these problems even more pronounced and severely restricting the stable preparation and industrial application of high-performance oligomeric anti-swelling ammonium salts. Summary of the Invention

[0005] This invention provides an intelligent temperature control method for the preparation reaction process of oligomeric anti-swelling ammonium salts. It generates a smooth and continuous temperature setting trajectory by constructing a dynamic identification model of the reaction stage and a multi-scale temperature response prediction model, combined with real-time sensing data of the temperature field at multiple points in the reactor, thermodynamic state parameters of the reactants, and stage switching instructions. This trajectory adopts a nonlinear ramp transition strategy at the stage switching boundary to avoid abrupt changes in the temperature set point. Furthermore, it drives the jacket medium flow regulation mechanism through a feedforward-feedback composite control structure to achieve precise dynamic compensation for the heat input of the reaction system, thereby suppressing local overcooling or overheating and ensuring the narrowing of the molecular weight distribution of the product and batch consistency.

[0006] This invention provides an intelligent temperature control method for the preparation reaction process of oligomeric anti-swelling ammonium salts, comprising:

[0007] Acquire multi-point temperature field data, real-time thermodynamic state parameters of reactants, and current reaction stage identification information within the oligomeric anti-swelling ammonium salt synthesis reactor;

[0008] Based on the current reaction stage identification information and the preset reaction stage timing logic rules, determine whether it is in the critical window for stage switching;

[0009] If the stage switching critical window is in progress, the stage switching temperature setting trajectory generation module is called. Based on the steady-state temperature setting value of the target stage, the steady-state temperature setting value of the current stage, the specific heat capacity of the reactants, the effective heat capacity parameters of the reactor, and the jacket heat transfer coefficient, a nonlinear temperature setting trajectory that is continuously differentiable in the time domain and has a continuous first derivative is generated.

[0010] The nonlinear temperature setting trajectory is used as the reference input of the main controller, and the weighted average of three representative temperature measurement points located in the center region, near-wall region and bottom region of the multi-point temperature field data is used as the controlled variable to form a feedback control loop.

[0011] The estimated value of the reaction exothermic rate in the real-time thermodynamic state parameters of the reactants is extracted and used as a feedforward signal to the feedforward compensator. The feedforward compensator calculates the required jacket cooling or heating medium flow increment based on the pre-identified jacket-vessel heat transfer dynamic model.

[0012] The adjustment amount output by the feedback control loop is algebraically superimposed with the flow increment output by the feedforward compensator to generate the final jacket medium flow control command.

[0013] The jacket medium flow control command is sent to the electric regulating valve actuator to drive the regulating valve opening to change, thereby adjusting the volume flow rate of heat transfer oil or cooling water entering the reactor jacket, and realizing dynamic and precise control of the reaction system temperature.

[0014] In one embodiment of the present invention, the multi-point temperature field data is acquired by a distributed temperature sensing array installed inside the reactor. The distributed temperature sensing array includes no less than nine platinum resistance temperature sensors, of which three sensors are evenly distributed along the reactor axis in the area near the central stirring shaft, three sensors are radially distributed on an annular cross section at a distance of 1 / 3 radius from the reactor wall, and the remaining three sensors are arranged in the material retention area at the conical bottom of the reactor.

[0015] As one embodiment of the present invention, the real-time thermodynamic state parameters of the reactants include the reaction liquid density, viscosity, specific heat capacity, and reaction progress indication parameters. The reaction progress indication parameters are calculated by the ratio of the intensity of characteristic absorption peaks collected by an online infrared spectrometer, specifically the ratio of the absorbance of the amide bond characteristic peak to the characteristic peak of the raw material monomer.

[0016] As one embodiment of the present invention, the stage switching critical window is defined as: within the time interval from 15 minutes before the preset stage switching time to 10 minutes after the switching time, the temperature setting trajectory reconstruction process is initiated.

[0017] In one embodiment of the present invention, the nonlinear temperature setting trajectory is constructed using a fifth-order polynomial function, the expression of which is:

[0018]

[0019] in, For time variables starting from the start of the phase transition, the coefficients are... , , , , , It is uniquely determined by the following boundary conditions:

[0020] exist At time 0, It equals the current steady-state temperature setpoint.

[0021] exist Equal to the total transition time time, It equals the target stage steady-state temperature setpoint;

[0022] exist At time 0, Equals 0;

[0023] exist equal time, Equals 0;

[0024] exist At time 0, Equals 0;

[0025] exist equal time, Equals 0;

[0026] The total switching transition time The viscosity is dynamically adjusted based on the ratio of the current viscosity of the reactant to the preset reference viscosity. When the viscosity ratio is greater than 1.2, The time interval is 30 minutes; when the viscosity ratio is less than or equal to 1.2 and greater than 0.8, The time interval is 20 minutes; when the viscosity ratio is less than or equal to 0.8, The value is 15 minutes.

[0027] In one embodiment of the present invention, the jacket heat transfer coefficient is updated in real time by an online identification module. This online identification module calculates the coefficient using a steady-state heat transfer equation based on the jacket medium inlet temperature, outlet temperature, volumetric flow rate, and weighted average temperature inside the vessel. The calculation formula is as follows:

[0028]

[0029] in, The heat power absorbed or released by the jacket medium is obtained by multiplying the medium's mass flow rate by its specific heat capacity and then by the inlet and outlet temperature difference. This refers to the effective heat transfer area of ​​the reactor jacket. The logarithmic mean temperature difference is calculated from the inlet and outlet temperature difference between the jacket medium and the material inside the vessel.

[0030] In one embodiment of the present invention, the feedback control loop adopts an improved internal model control structure, and its controller transfer function is:

[0031]

[0032] in, A simplified first-order plus pure time-delay model is used for the dynamic process of temperature in the reactor. The filter time constant is 20% of the process time constant.

[0033] As one embodiment of the present invention, the gain coefficient of the feedforward compensator Based on the jacket heat transfer coefficient U and the specific heat capacity of the reactants The product of is dynamically tuned, and the relationship is as follows:

[0034]

[0035] in, As the reference feedforward gain, and These are the nominal heat transfer coefficient and the nominal specific heat capacity, respectively.

[0036] As one embodiment of the present invention, the actuator of the electric regulating valve is an equal percentage flow characteristic regulating valve, whose valve core stroke is linearly related to the logarithm of the medium volume flow rate. The valve diameter is selected according to the maximum heat load requirement of the reactor to ensure that it can still provide a medium flow capacity of not less than 110% of the design value under the maximum temperature difference condition.

[0037] This invention provides an intelligent temperature control system for the preparation reaction process of oligomeric anti-swelling ammonium salts, comprising:

[0038] The multi-source data acquisition unit is used to acquire multi-point temperature field data, real-time thermodynamic state parameters of reactants, and current reaction stage identification information within the oligomeric anti-swelling ammonium salt synthesis reactor.

[0039] The stage switching determination unit is used to determine whether the current reaction stage identification information and the preset reaction stage timing logic rules are in a critical window for stage switching.

[0040] The temperature setting trajectory generation unit is used to calculate and generate a nonlinear temperature setting trajectory that is continuously differentiable in the time domain and has a continuous first derivative when the stage switching critical window is in the stage. This is based on the steady-state temperature setting value of the target stage, the steady-state temperature setting value of the current stage, the specific heat capacity of the reactants, the effective heat capacity parameters of the reactor, and the jacket heat transfer coefficient.

[0041] The feedback control unit is used to take the nonlinear temperature set trajectory as a reference input and the weighted average of three representative temperature measurement points in the multi-point temperature field data as the controlled variable to form a feedback control loop and output the adjustment amount.

[0042] The feedforward compensation unit is used to extract the estimated value of the reaction exothermic rate from the real-time thermodynamic state parameters of the reactants, and calculate the required jacket cooling or heating medium flow increment based on the pre-identified jacket-vessel heat transfer dynamic model.

[0043] The control command fusion unit is used to algebraically superimpose the adjustment amount output by the feedback control unit and the flow increment output by the feedforward compensation unit to generate the final jacket medium flow control command.

[0044] The actuator is used to send the jacket medium flow control command to the electric regulating valve actuator to drive the regulating valve opening to adjust the volumetric flow rate of the heat transfer oil or cooling water entering the reactor jacket.

[0045] As one embodiment of the present invention, the multi-source data acquisition unit includes a distributed temperature sensing array, an online infrared spectrometer, a densitometer, a viscometer, and a reaction stage status register. The distributed temperature sensing array includes no less than nine platinum resistance temperature sensors, which are divided into a central region group, a near-wall region group, and a bottom region group according to their spatial location.

[0046] As one embodiment of the present invention, the temperature setting trajectory generation unit has a built-in fifth-order polynomial trajectory planner, whose coefficient solving module constructs a system of linear equations based on six boundary conditions and solves the polynomial coefficients in real time using the Gaussian elimination method.

[0047] In one embodiment of the present invention, the feedback control unit adopts a digital internal model controller, the discretization form of which is realized by the bilinear transformation method, and the sampling period is set to two seconds.

[0048] In one embodiment of the present invention, the feedforward compensation unit includes a thermal power estimation submodule and a dynamic gain tuning submodule. The former calculates the reaction heat release rate in real time based on material balance and energy balance, while the latter dynamically adjusts the feedforward gain based on the online updated heat transfer coefficient and specific heat capacity.

[0049] In one embodiment of the present invention, the execution drive unit includes a signal isolation module, a current output module and a fault diagnosis module. The current output module outputs a standard current signal of 4 to 20 mA to drive the electric regulating valve. The fault diagnosis module monitors the deviation between the valve position feedback signal and the command signal of the regulating valve in real time. When the absolute value of the deviation exceeds the set threshold for 3 seconds, an alarm is triggered and the system switches to the backup control channel.

[0050] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0051] 1. This invention fundamentally eliminates the thermal stress shock problem caused by sudden changes in temperature setpoint in traditional control methods by introducing a stage switching critical window discrimination mechanism and a nonlinear temperature setpoint trajectory generation strategy.

[0052] 2. The fifth-order polynomial trajectory employed ensures the continuity of the temperature setpoint and its first and second derivatives at the switching boundary, resulting in smooth changes in heat input and effectively preventing instantaneous overcooling or overheating in localized areas within the reactor. Simultaneously, by integrating multi-point temperature field weighted average feedback with feedforward compensation based on the reaction exothermic rate, a feedforward-feedback composite control structure is constructed, significantly improving the system's ability to suppress reaction thermal disturbances.

[0053] 3. The online-updated jacket heat transfer coefficient and dynamically tuned feedforward gain further enhance the adaptability of the control strategy to changes in material properties. Ultimately, this method controls the standard deviation of the product molecular weight distribution to within 5%, which is more than 40% lower than the traditional step temperature control method, significantly improving the batch uniformity and industrial application performance stability of oligomeric anti-swelling ammonium salt products. Attached Figure Description

[0054] Figure 1 This is a schematic diagram of the overall technical architecture of the intelligent temperature control method for the preparation reaction process of oligomeric anti-swelling ammonium salt proposed in this invention;

[0055] Figure 2 This is a schematic diagram of the core principle framework of the stage switching critical window discrimination and nonlinear temperature setting trajectory generation in this invention;

[0056] Figure 3 This is a schematic diagram of the framework of the multi-point temperature field sensing and weighted average feedback control logic in this invention;

[0057] Figure 4 This is a schematic diagram of the feedforward compensation mechanism and dynamic gain tuning principle framework based on reaction exothermic rate estimation in this invention;

[0058] Figure 5 This is a flowchart illustrating the logical flow of the online identification and heat transfer coefficient update of the jacket-vessel heat transfer dynamic model in this invention.

[0059] Figure 6 This is a schematic diagram of the multi-level interaction relationship and data flow of the feedforward-feedback composite control command fusion and execution drive in this invention. Detailed Implementation

[0060] Please refer to Figures 1 to 6 This invention provides an intelligent temperature control method for the preparation reaction process of oligomeric anti-swelling ammonium salts. Its core lies in solving the problem of thermal stress shock caused by abrupt changes in temperature setpoints during reaction stage switching. In the synthesis process of oligomeric anti-swelling ammonium salts, the reaction typically goes through multiple stages such as polymerization and maturation, each with different steady-state temperature setpoint requirements. Traditional control strategies directly switch the temperature setpoint from the current stage target to the next stage target at stage switching moments, causing a step change in the jacket medium flow command, which in turn leads to instantaneous overcooling or overheating in localized areas within the reactor.

[0061] Such thermal disturbances disrupt the thermal equilibrium of the reaction system, inducing side reactions or molecular chain breakage, ultimately leading to a broadening of the product molecular weight distribution and a decrease in batch consistency. To overcome these shortcomings, this invention constructs a composite temperature control mechanism that integrates dynamic identification of the fusion stage, nonlinear trajectory planning, multi-point temperature feedback, and reaction heat feedforward compensation. This ensures that the temperature setpoint trajectory is continuously differentiable in the time domain and that the first derivative is continuous, thereby achieving smooth transition and precise control of the heat input.

[0062] The intelligent temperature control method first acquires multi-point temperature field data, real-time thermodynamic state parameters of the reactants, and current reaction stage identification information within the oligomeric anti-swelling ammonium salt synthesis reactor. The multi-point temperature field data is collected through a distributed temperature sensor array installed inside the reactor, which contains no fewer than nine platinum resistance temperature sensors.

[0063] Three sensors are evenly distributed along the axial direction of the reactor near the central stirring shaft to monitor the temperature evolution in the core reaction zone. Three sensors are radially distributed on an annular cross-section at one-third of the radius from the reactor wall to capture the temperature gradient of the heat transfer boundary layer in the near-wall region. The remaining three sensors are arranged in the material retention area at the conical bottom of the reactor to reflect the heat accumulation in the bottom dead zone. All sensors use a four-wire connection to eliminate the influence of lead resistance, the sampling frequency is set to once per second, and the data is uploaded to the central processing unit in real time via industrial Ethernet.

[0064] The real-time thermodynamic parameters of the reactants include the density, viscosity, specific heat capacity, and reaction progress indicators. Density and viscosity were measured in real-time using an online vibratory density meter and a rotational viscometer, with measurement accuracies of ±0.1 kg / m³ and ±1 mPa·s, respectively.

[0065] The reaction progress indicator parameters are calculated using the ratio of characteristic absorption peak intensities acquired by an online infrared spectrometer. Specifically, this ratio is the ratio of the absorbance of the characteristic peak of the amide bond at 1650 wavenumbers to the absorbance of the characteristic peak of the raw material monomer at 1720 wavenumbers. This ratio increases monotonically with the reaction and can be used as a non-destructive characterization indicator of the reaction conversion rate. The current reaction stage identification information is provided by the stage status register in the reaction control system. This register automatically updates the stage labels according to the preset reaction formula and timing logic, including "initial heating", "polymerization period", "maturation period", and "cooling and discharge".

[0066] After acquiring the aforementioned multi-source data, the system determines whether it is within a critical window for stage switching based on the current reaction stage identifier information and the preset reaction stage timing logic rules. The critical window for stage switching is defined as the time interval from 15 minutes before to 10 minutes after a preset stage switching time. Within this interval, the system initiates a temperature setting trajectory reconstruction process to avoid abrupt changes in the setpoint. The stage switching time is preset by the process formulation; for example, the switch from the polymerization stage to the maturation stage is 120 minutes after the start of the reaction. The system continuously compares the current system time with the preset switching time using a high-precision real-time clock module. When the time difference falls within the range of -15 minutes to +10 minutes, it is determined to be within the critical window for stage switching, and subsequent trajectory generation logic is triggered.

[0067] If the system is in a critical phase transition window, the phase transition temperature setting trajectory generation module is invoked. Based on the steady-state temperature setpoint of the target phase, the steady-state temperature setpoint of the current phase, the specific heat capacity of the reactants, the effective heat capacity parameters of the reactor, and the jacket heat transfer coefficient, a nonlinear temperature setting trajectory that is continuously differentiable in the time domain and has a continuous first derivative is generated. This trajectory is constructed using a fifth-order polynomial function, and its expression is:

[0068]

[0069] in, For time variables starting from the start of the phase transition, the coefficients are... , , , , , It is uniquely determined by the following boundary conditions:

[0070] exist At time 0, It equals the current steady-state temperature setpoint.

[0071] exist Equal to the total transition time time, It equals the target stage steady-state temperature setpoint;

[0072] exist At time 0, Equals 0;

[0073] exist equal time, Equals 0;

[0074] exist At time 0, Equals 0;

[0075] exist equal time, Equals 0;

[0076] The aforementioned boundary conditions ensure that the temperature setpoint, its rate of change, and acceleration are all zero at the start and end points of the trajectory, thus achieving a seamless transition with the preceding and following steady-state phases. The total transition time T_trans is not a fixed value, but is dynamically adjusted based on the ratio of the current viscosity of the reactant to the preset reference viscosity.

[0077] The reference viscosity is set to a typical value reflecting the mid-maturation period under standard operating conditions, such as 800 mPa·s. When the ratio of the real-time viscosity to the reference viscosity is greater than 1.2, it indicates a significant decrease in material flowability and a weakening of thermal conductivity. The value is set to 30 minutes to extend the transition time and reduce thermal stress;

[0078] When the viscosity ratio is less than or equal to 1.2 and greater than 0.8, the material properties are close to the standard state. The value is 20 minutes;

[0079] When the viscosity ratio is less than or equal to 0.8, the material is dilute and heat transfer is rapid. The time interval is set to 15 minutes. This dynamic adjustment mechanism adapts the temperature trajectory to the thermal response characteristics under different reaction states, avoiding performance degradation due to excessively fast or slow transitions.

[0080] The jacket heat transfer coefficient U, a key parameter for trajectory generation, is updated in real time by an online identification module. This module calculates U based on the jacket medium inlet temperature, outlet temperature, volumetric flow rate, and weighted average temperature inside the vessel, using the steady-state heat transfer equation. The calculation formula is as follows:

[0081]

[0082] in, The heat power absorbed or released by the jacket medium is obtained by multiplying the medium's mass flow rate by its specific heat capacity and then by the inlet and outlet temperature difference. This refers to the effective heat transfer area of ​​the reactor jacket. The logarithmic mean temperature difference is calculated from the inlet and outlet temperature difference between the jacketed medium and the material inside the vessel. The mass flow rate of the medium is measured in real time by an electromagnetic flowmeter with an accuracy of ±0.5%.

[0083] Inlet and outlet temperatures are measured by high-precision resistance temperature detectors (RTDs) with a resolution of 0.1 degrees Celsius. The online identification module performs calculations every five minutes and writes the updated U-values ​​to shared memory for use by the trajectory generation module. This mechanism overcomes the limitations of the assumption of a constant heat transfer coefficient in traditional methods and effectively compensates for heat transfer performance drift caused by scaling, medium aging, or changes in flow rate.

[0084] The generated nonlinear temperature setpoint trajectory is used as the reference input for the main controller. Simultaneously, the system uses the weighted average of three representative temperature measurement points located in the reactor's central, near-wall, and bottom regions from the multi-point temperature field data as the controlled variable, forming a feedback control loop.

[0085] The weighting coefficients are set according to the degree of influence of each region on product quality: the central region has a weight of 0.5 because it directly reflects the temperature of the main reaction zone;

[0086] The near-wall region has a weight of 0.3 to suppress side reactions on the wall; the bottom region has a weight of 0.2 to prevent overheating and degradation of the bottom material. Weighted average temperature. ,in, Temperature of the central area measurement point. Temperature at the measurement point near the wall. The temperature is measured at the bottom region. The feedback control loop uses an improved internal model control structure, and its controller transfer function is:

[0087]

[0088] in, A simplified first-order plus pure time-delay model of the dynamic temperature process in the reactor is given by: , For process gain, It is a time constant. The time delay is pure; all three parameters are identified in advance through step response testing and stored in the controller parameter library. The filter time constant is the process time constant. 20% is allocated to suppress high-frequency noise and enhance robustness. The internal model controller is discretized in the digital domain using a bilinear transform method, with a sampling period set to 2 seconds to ensure the real-time performance and stability of control commands.

[0089] Simultaneously, the system extracts the estimated reaction exothermic rate from the real-time thermodynamic state parameters of the reactants and inputs it as a feedforward signal to the feedforward compensator. The reaction exothermic rate is calculated in real-time by a heat power estimation submodule, which establishes a dynamic equation based on material and energy balance. Specifically, the reaction exothermic power is:

[0090]

[0091] in, The density of the reaction solution, For the effective reaction volume, For specific heat capacity, The weighted average rate of change of temperature This refers to the jacket heat exchange power. The result is calculated from the medium flow rate, specific heat capacity, and inlet / outlet temperature difference. By solving this equation in real time, the following can be obtained: The instantaneous estimate. The feedforward compensator, based on the pre-identified jacket-vessel heat transfer dynamic model, will... Converted to the required jacket cooling or heating medium flow rate increment. Feedforward gain coefficient. It is not a fixed value, but depends on the jacket heat transfer coefficient. The product of the reactant's specific heat capacity c_p and the reactant's specific heat capacity is dynamically tuned, and the relationship is as follows:

[0092]

[0093] in, The reference feedforward gain is tuned through simulation or experiment. and These are the nominal heat transfer coefficient and the nominal specific heat capacity, respectively. This dynamic tuning mechanism ensures that the feedforward compensation amount always matches the heat transfer efficiency and material heat capacity under the current operating conditions, avoiding over-compensation or under-compensation of the feedforward due to changes in physical properties.

[0094] The adjustment amount output by the feedback control loop is algebraically superimposed with the flow increment output by the feedforward compensator to generate the final jacket medium flow control command. This superposition process is completed in the control command fusion unit, whose output... ,in For feedback adjustment, This is the feedforward flow increment. This command is sent to the actuator of the electric control valve.

[0095] The electric regulating valve is an equal percentage flow characteristic regulating valve, whose valve core stroke has a linear relationship with the logarithm of the medium volumetric flow rate, ensuring high sensitivity at small openings and strong flow capacity at large openings. The valve diameter is selected according to the maximum heat load requirements of the reactor, ensuring that it can still provide a medium flow capacity of no less than 110% of the design value under the maximum temperature difference condition, with a safety margin. The actuation drive unit includes a signal isolation module, a current output module, and a fault diagnosis module.

[0096] The current output module converts digital control commands into 4 to 20 mA standard current signals to drive the electric actuator; the fault diagnosis module monitors the deviation between the valve position feedback signal and the command signal of the regulating valve in real time. When the absolute value of the deviation exceeds the set threshold (e.g., 5% of the full stroke) for 3 seconds, it is determined that the valve is stuck or the actuator is faulty, and an audible and visual alarm is immediately triggered. It also automatically switches to the backup control channel, which uses a combination of a fixed proportional valve and a bypass manual valve to maintain basic temperature control functions until manual intervention.

[0097] Throughout the control process, the system continuously monitors product quality indicators. The standard deviation of the product's molecular weight distribution is measured in real time through online sampling and gel permeation chromatography analysis. Experimental data show that, after adopting the intelligent temperature control method described in this invention, the standard deviation is stably controlled within 5%, a reduction of more than 40% compared to the traditional step temperature control method. Temperature field uniformity is also significantly improved, with the maximum temperature difference within the reactor decreasing from 8 degrees Celsius in the traditional method to below 2 degrees Celsius, effectively suppressing cross-linking side reactions caused by local overheating and incomplete polymerization caused by local undercooling. Furthermore, the thermal stress impact energy during stage switching is reduced by 60%, and the reactor liner life is extended by 25%.

[0098] The method described in this invention is not only applicable to the synthesis of oligomeric anti-swelling ammonium salts, but can also be extended to other fine chemical reaction processes requiring multi-stage temperature control, such as epoxy resin curing, polyurethane foaming, and pharmaceutical intermediate synthesis. Its core lies in eliminating setpoint mutations through nonlinear trajectory planning and combining a feedforward-feedback composite structure to actively suppress dynamic thermal disturbances, thereby ensuring reaction safety while improving product quality consistency and the manufacturing level of high-end chemicals.

Claims

1. A method for intelligent temperature control of the process of preparing oligomeric anti-swelling ammonium salt, characterized in that, The method comprises the following steps: acquiring multi-point temperature field data in an oligomerization anti-swelling ammonium salt synthesis reactor, real-time thermodynamic state parameters of a reactant and current reaction stage identification information; judging whether a stage switching critical window is present based on the current reaction stage identification information and a preset reaction stage time sequence logic rule; if the stage switching critical window is present, calling a stage switching temperature setting trajectory generation module, and calculating and generating a nonlinear temperature setting trajectory which is continuous and derivable in time domain and has continuous first-order derivative based on a steady-state temperature setting value of a target stage, a steady-state temperature setting value of a current stage, specific heat capacity of the reactant, effective heat capacity parameters of the reactor and a jacket heat transfer coefficient; using a weighted average value of three representative temperature measuring points in a central region, a near-wall region and a bottom region of the reactor as a controlled variable, and constructing a feedback control loop with the nonlinear temperature setting trajectory as a reference input of a main controller; extracting a reaction heat release rate estimate value in the real-time thermodynamic state parameters of the reactant as a feedforward signal input into a feedforward compensator, and calculating a required jacket cooling or heating medium flow increment based on a pre-identified jacket-reactor heat transfer dynamic model; algebraically superimposing an adjustment amount output by the feedback control loop and the flow increment output by the feedforward compensator to generate a final jacket medium flow control instruction; sending the jacket medium flow control instruction to an electric regulating valve actuator to drive the regulating valve opening to change so as to adjust the volume flow of the heat conducting oil or cooling water entering the reactor jacket.

2. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 1, characterized in that, The multi-point temperature field data are acquired by a distributed temperature sensing array installed inside the reactor, and the distributed temperature sensing array comprises no less than 9 platinum resistance temperature sensors, of which 3 sensors are uniformly distributed near a central stirring shaft along an axial direction, 3 sensors are distributed on an annular section at 1 / 3 radius from the wall along a radial direction, and the remaining 3 sensors are arranged in a material retention area of a conical bottom of the reactor.

3. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 2, characterized in that, The real-time thermodynamic state parameters of the reactant include reaction liquid density, viscosity, specific heat capacity and a reaction progress indication parameter, and the reaction progress indication parameter is calculated based on a characteristic absorption peak intensity ratio acquired by an online infrared spectrometer, specifically, an absorbance ratio of an amide bond characteristic peak to a raw material monomer characteristic peak.

4. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 3, characterized in that, The stage switching critical window is defined as a time interval from 15 minutes before a preset stage switching time to 10 minutes after the switching time, and a temperature setting trajectory reconstruction process is started in the interval.

5. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 4, characterized in that, The nonlinear temperature setting trajectory is constructed by using a five-order polynomial function, and an expression thereof is: wherein is a time variable counted from the start of the phase switch, the coefficient , , , , , is uniquely determined by the following boundary conditions: At At time equal to 0, At time equal to current stage steady state temperature set point; exist Equal to the total transition time time, It equals the target stage steady-state temperature setpoint; At equal to 0, equal to 0; In equal to the moment, equal to 0; At equal to 0, equal to 0; In equal to the moment, equal to 0; The total length of the switching transition According to the ratio of the current viscosity of the reaction material to the preset reference viscosity, the ratio is dynamically adjusted, when the viscosity ratio is greater than 1.2, The value is 30 minutes; when the viscosity ratio is less than or equal to 1.2 and greater than 0.8, The value is 20 minutes; when the viscosity ratio is less than or equal to 0.8, The value is 15 minutes.

6. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 5, characterized in that, The jacket heat transfer coefficient is updated in real time by an online identification module, and the online identification module calculates the jacket heat transfer coefficient by using a steady-state heat transfer equation based on jacket medium inlet temperature, outlet temperature, volume flow and weighted average temperature in the reactor, and a calculation formula thereof is: wherein, is the heat power absorbed or released by the jacket medium, obtained by multiplying the mass flow of the medium by its specific heat capacity and then by the temperature difference between inlet and outlet; is the effective heat transfer area of the reactor jacket; is the logarithmic mean temperature difference, calculated from the temperature difference between inlet and outlet of the jacket medium and the material in the reactor.

7. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 6, characterized in that, The feedback control loop adopts an improved internal model control structure, and a controller transfer function thereof is: wherein, is a simplified first-order plus time delay model for the temperature dynamic process of the reactor, is a filter time constant, which is 20% of the process time constant.

8. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 7, characterized in that, Gain coefficient of the feedforward compensator According to the jacket heat transfer coefficient The product of the specific heat capacity of the reactant material The dynamic setting is carried out, and the relationship is: wherein, is the reference feedforward gain, and are the nominal heat transfer coefficient and the nominal specific heat capacity, respectively.

9. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 8, characterized in that, The electric regulating valve actuator is an equal percentage flow characteristic regulating valve, the valve core stroke has linear relationship with the logarithm of medium volume flow, and the valve diameter is selected according to the maximum heat load demand of the reaction kettle, so that the medium flow capacity of not less than 110% of the design value can be ensured under the maximum temperature difference working condition.

10. The intelligent temperature control method for the preparation reaction process of the oligomeric anti-swelling ammonium salt according to claim 1, characterized in that, The calculation method of the weighted average value of the three representative temperature measurement points is: wherein, Tc is the temperature of the center region temperature probe, Tb is the temperature of the near wall region temperature probe, Tb is the temperature of the bottom region temperature probe.

Images