Dynamic heat transfer temperature control method for reaction kettle heating furnace

Abstract

The invention discloses a dynamic heat transfer temperature control method for a reaction kettle heating furnace, and relates to the technical field of temperature monitoring, and the method comprises the steps: S1, placing a target reaction kettle and a target heating furnace in a temperature field, and marking a first heat conduction region and a second heat conduction region; s2, marking the size of an occupied space block, and constructing a hot area evaluation strategy; s3, marking out a first-order overheat area; calculating a third heat conduction value by using a time step prediction method; and S4, a temperature gradient vector field is calculated, a second-order overheat area is generated, and temperature overheat adjustment is conducted on the second-order overheat area. Quantitative modeling is carried out on the flowing state of heat between the reaction kettle and the heating furnace, the estimated value of temperature rise change is obtained, the heat transfer direction and strength are recognized, space positioning and early warning of the imminent temperature overheating area are achieved, space look-ahead recognition of potential local overtemperature can be enhanced, and the accuracy and the reliability of the system are improved. And the judgment blind area of the existing system for the temperature rise which is not yet generated but is about to generate is compensated.

Description

Technical Field

[0001] This invention relates to the field of temperature monitoring technology, and more specifically to a dynamic heat transfer temperature control method for a reactor heating furnace. Background Technology

[0002] In pressure reactor heating systems, external circulating heaters are the core equipment for maintaining high-temperature and high-pressure reactions. The conventional workflow in existing technologies is as follows: high-temperature heat transfer oil is heated to a set temperature (typically 200-350°C) in an external heater, such as an electric or gas-fired heat transfer oil heater. After being pressurized by a high-temperature circulating pump, it is transported to the reactor jacket or internal coils through insulated metal pipes. The heat transfer medium flows within the reactor's heat exchange structure, and heat is conducted to the internal materials through the reactor wall. The cooled heat transfer oil returns to the heater for reheating, forming a closed-loop cycle. Existing temperature control systems typically adjust the heater outlet temperature or the opening of the heat transfer medium regulating valve based on the deviation of data from the reactor's internal temperature sensors, using an adjustment algorithm to achieve temperature regulation.

[0003] Current control systems typically rely on single-loop feedback control or acquire temperature feedback and perform temperature compensation at certain frequencies in a time-series format. When encountering violent and complex reactions in the reactor, the temperature changes become even more complex, making it difficult for existing temperature control systems to dynamically assess the coupling relationship between the heat transfer medium flow rate and temperature decay. This results in untimely monitoring and feedback of dynamic heat transfer in various parts of the heating furnace, as well as untimely prediction of impending temperature changes in each heated area receiving dynamic heat transfer. Consequently, the overall temperature control accuracy during reactor heating decreases, leading to lag in temperature control response and even process runaway. Summary of the Invention

[0004] This invention provides a dynamic heat transfer temperature control method for reactor heating furnaces, which solves the problem that existing temperature monitoring technologies are inaccurate in assessing potential temperature changes caused by dynamic heat transfer during heating when dynamically monitoring and adjusting the temperature of reactor heating furnaces with complex reaction changes, resulting in poor accuracy of furnace temperature monitoring.

[0005] This invention is achieved through the following technical solution: A method for dynamic heat transfer temperature control in a reactor heating furnace, the method comprising: Step S1: Construct a temperature field in a spatial coordinate system, divide the spatial region of the temperature field into a number of spatial blocks along the coordinate axes, place the target reactor and the target heating furnace in the temperature field, mark the first heat conduction zone representing local overheating in the form of spatial blocks for the target reactor, and mark the second heat conduction zone representing local overheating in the form of spatial blocks for the target heating furnace. Step S2: Mark the size of the spatial block occupied by the first thermal conduction zone and the second thermal conduction zone along the spatial block in the temperature field, and mark the heat energy values ​​in the first thermal conduction zone and the second thermal conduction zone as the first thermal conduction value and the second thermal conduction value respectively, and construct a thermal zone evaluation strategy based on the first thermal conduction value and the second thermal conduction value. Step S3: Mark the superheated space blocks in the temperature field that indicate potential dynamic heat transfer in the first thermal conductivity zone using the thermal zone assessment strategy, and mark all superheated space blocks as first-order superheated zones; use the time step prediction method to predict and calculate the thermal energy trend value of each first-order superheated zone based on the first thermal conductivity value and the second thermal conductivity value, and mark it as the third thermal conductivity value. Step S4: Calculate the temperature gradient vector field of the corresponding local overheating zone using the third thermal conductivity value in each first-order overheating zone. Use the location pointed to by the temperature gradient vector field to generate a second-order overheating zone with the same spatial size as the first-order overheating zone and representing the end of potential dynamic heat transfer. Perform temperature overheating adjustment on the target reactor part covered by the second-order overheating zone.

[0006] Current control systems struggle to dynamically assess the coupling relationship between heat transfer fluid flow rate and temperature decay when encountering intense and complex reactions in the reactor. This results in delayed monitoring and feedback of dynamic heat transfer in various parts of the furnace, as well as untimely prediction of impending temperature changes in each heated area receiving dynamic heat transfer. Consequently, the overall temperature control accuracy during reactor heating decreases, leading to lag in temperature control response and even process malfunction. Therefore, this invention provides a dynamic heat transfer temperature control method for reactor furnaces. This method addresses the problem of inaccurate assessment of potential temperature changes caused by dynamic heat transfer during heating in reactor furnaces with complex reaction conditions, resulting in poor temperature monitoring accuracy.

[0007] Furthermore, the process of marking the first and second thermally conductive areas includes: The process of starting up the target reactor and the target heater is defined as one working cycle. The heating rate threshold and the upper temperature limit threshold are set independently for each spatial block of the target reactor and the target heater. The heating rate and the upper temperature limit of the target reactor and the target heater are monitored during each working cycle. When the heating rate reaches the heating rate threshold or the upper temperature limit reaches the upper temperature limit threshold, it is determined that there is local overheating in the corresponding spatial block. All the spaces within the target reactor that are locally overheated are labeled as the first heat conduction zone by merging adjacent spaces; all the spaces within the heating furnace that are locally overheated are labeled as the second heat conduction zone by merging adjacent spaces.

[0008] Furthermore, the hot zone assessment strategy includes: Set an overheating space threshold. The first thermally conductive zone whose heat energy value reaches the overheating space threshold is marked as a first-order overheating zone. The first and second thermally conductive zones that are in adjacent contact are marked as the overheating zone to be detected by merging adjacent space blocks. The combined heat energy value in the overheating zone to be detected is calculated by weighted summation of the first and second thermally conductive values. When the combined heat energy value reaches the overheating space threshold, the space block of the currently detected first and second thermally conductive zones is determined to be an overheating space block, and the current overheating zone to be detected is determined to be a first-order overheating zone.

[0009] Furthermore, the calculation process for the third thermal conductivity value includes: Let T represent the real-time temperature value of the temperature field, and t represent the time step; let the first thermal conductivity be denoted as Q1, the second thermal conductivity as Q2, and the third thermal conductivity as Q3; let the average material density of the reactor be denoted as ρ, and let the isobaric specific heat capacity inside the reactor be c. ρ Let the ordinal number of the spatial block be i, and the volume of the spatial block be V; let m represent the thermal energy parameter. The formula for calculating the third thermal conductivity value Q3 is as follows: , Where, Q3 in the formula (i) T represents the third thermal conductivity value Q3 of the i-th spatial block, where T is the thermal conductivity value Q3 of the i-th spatial block. i (t+1) It represents the real-time temperature value predicted by the i-th spatial block at time step t+1.

[0010] Furthermore, the real-time temperature value T predicted at the (t+1)th time step i (t+1) The acquisition process includes: Obtain the real-time temperature value T of the target spatial block at the current time step. i t The system monitors and collects reactant flow rates and stirring speed values ​​of the target reactor after changes in reactant concentration. Within the current time step, it calculates reactant heat load using the reactant flow rate and mechanical heat load using the stirring speed. The reactant heat load and mechanical heat load values ​​are then added together to update the real-time temperature value. This updated real-time temperature value is set as the predicted real-time temperature T for the (t+1)th time step. i (t+1) .

[0011] Furthermore, the process of calculating the temperature gradient vector field to generate the second-order superheated region includes: In each first-order superheated region, a third thermal conductivity matrix is ​​constructed, and an initial gradient vector representing the heat transfer direction is calculated based on the third thermal conductivity matrix. The fluid simulation rate and eddy diffusion coefficient of the target reactor are generated using CFD simulation of the flow velocity field. The initial gradient vector is corrected using the fluid simulation rate and eddy diffusion coefficient and a modified gradient vector is generated. The coordinates of the spatial blocks in each first-order superheated region are labeled as first-order block coordinates. The first-order block coordinates are integrated along the direction of the modified gradient vector to represent the heat flow path update. The integrated values ​​are represented as the second-order spatial coordinates of each spatial block after the heat flow path update is completed. The temperature field spatial region composed of each spatial block corresponding to the second-order spatial coordinates is labeled as the second-order superheated region.

[0012] Furthermore, the initial gradient vector is calculated using the central difference method, which includes: For each spatial block in the spatial coordinate system, calculate the x, y, and z gradient components respectively. Let the difference in the third thermal conductivity value between the block and the adjacent spatial blocks be represented as the block temperature difference. The x-gradient component is calculated by taking the block temperature difference in the x-axis direction and dividing it by twice the step size in the x-axis direction. The y-gradient component is calculated by taking the block temperature difference in the y-axis direction and dividing it by twice the step size in the y-axis direction. The z-gradient component is calculated by taking the block temperature difference in the z-axis direction and dividing it by twice the z-axis step size. The x-gradient components, y-gradient components, and z-gradient components are combined to form the initial gradient vector.

[0013] Furthermore, the calculation process of the corrected gradient vector includes: Let the corrected gradient vector be denoted as Yc, the initial gradient vector as Yo, the fluid simulation velocity as ε, and the eddy diffusion coefficient as λ. Then, the formula for calculating the corrected gradient vector is: , Where in the formula The vector derivative of the fluid simulation rate is represented by , and the velocity curl is represented by .

[0014] Furthermore, for the spatial block where the axial flow agitator of the target reactor is located, the axial step size of the x-axis, y-axis, and z-axis is twice the radial length of the axial flow agitator.

[0015] Furthermore, when the block temperature difference in the z-axis direction exceeds twice the block temperature difference in both the x-axis and y-axis directions, numerical convergence is performed with respect to the step size in the z-axis direction.

[0016] Compared with the prior art, the present invention has the following advantages and beneficial effects: 1. A temperature field is constructed in a spatial coordinate system. By labeling the heat energy values ​​of the first and second heat conduction zones, the flow state of heat between the reactor and the furnace is quantitatively modeled, thereby realizing the dynamic evaluation of potential heat conduction paths and alleviating the inadequacy of prediction in traditional systems that make it difficult to dynamically evaluate the coupling relationship between the heat transfer medium flow rate and the actual heat transfer. 2. The time-step prediction algorithm is used to predict the thermal energy trend of the first-order superheated zone in advance and obtain the estimated value of temperature rise change. This helps to analyze the time dynamics of temperature change, improve the response to sudden violent reactions and heat changes, and enhance the feedforward regulation capability of the target reactor temperature control system. 3. By constructing a temperature gradient vector field, the direction and intensity of heat migration are identified, and a second-order overheating zone is generated in the spatial block at the end of the heat flow trend. This enables spatial positioning and early warning of the upcoming temperature overheating area, which can enhance the spatial forward-looking identification of potential local overheating and make up for the blind spot of the existing system in judging the temperature rise that "has not yet occurred but is about to occur". Attached Figure Description

[0017] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings: Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0018] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with examples and drawings. The exemplary embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0019] Example 1, as Figure 1 As shown in the figure, this embodiment is a dynamic heat transfer temperature control method for a reactor heating furnace, the method comprising: Step S1: Construct a temperature field in a spatial coordinate system, divide the spatial region of the temperature field into a number of spatial blocks along the coordinate axes, place the target reactor and the target heating furnace in the temperature field, mark the first heat conduction zone representing local overheating in the form of spatial blocks for the target reactor, and mark the second heat conduction zone representing local overheating in the form of spatial blocks for the target heating furnace. Step S2: Mark the size of the spatial block occupied by the first thermal conduction zone and the second thermal conduction zone along the spatial block in the temperature field, and mark the heat energy values ​​in the first thermal conduction zone and the second thermal conduction zone as the first thermal conduction value and the second thermal conduction value respectively, and construct a thermal zone evaluation strategy based on the first thermal conduction value and the second thermal conduction value. Step S3: Mark the superheated space blocks in the temperature field that indicate potential dynamic heat transfer in the first thermal conductivity zone using the thermal zone assessment strategy, and mark all superheated space blocks as first-order superheated zones; use the time step prediction method to predict and calculate the thermal energy trend value of each first-order superheated zone based on the first thermal conductivity value and the second thermal conductivity value, and mark it as the third thermal conductivity value. Step S4: Calculate the temperature gradient vector field of the corresponding local overheating zone using the third thermal conductivity value in each first-order overheating zone. Use the location pointed to by the temperature gradient vector field to generate a second-order overheating zone with the same spatial size as the first-order overheating zone and representing the end of potential dynamic heat transfer. Perform temperature overheating adjustment on the target reactor part covered by the second-order overheating zone.

[0020] The target reactor typically consists of a vessel body, jacket, and internal coils, designed to withstand chemical reactions under high temperature and pressure. A circulating heat transfer medium within the jacket or internal coils heats or cools the materials inside the reactor. Since the reactions are mostly exothermic or endothermic, temperature fluctuations are significant, requiring a high-precision, high-response temperature control system to ensure process stability and product quality. The complex temperature changes during the reaction process make accurate monitoring and control of localized overheating or temperature dead zones difficult. Traditional temperature control often relies on single-point temperature measurement, failing to reflect the overall temperature field dynamics and potentially leading to process instability or safety risks. The target heating furnace typically includes an electric heating thermal oil furnace or a gas-fired thermal oil furnace, providing a stable heat input to the reactor by heating the thermal oil. It is a core component of the external circulation heating system. The thermal oil heat transfer medium is pressurized by a high-temperature circulating pump and transported to the reactor jacket or internal coils for heat exchange. After heat transfer, it returns to the heating furnace for further heating, forming a closed loop. Temperature control of heating furnaces generally relies on outlet temperature or heat medium valve adjustment. The feedback signal of temperature change mainly comes from the temperature sensor inside the reactor. The control method is usually simple and has a lag in response.

[0021] In this embodiment, the entire temperature field is divided into uniformly sized spatial grid units based on a Cartesian coordinate system, essentially forming a three-dimensional discrete grid structure. Each spatial block represents a voxel unit in space, used to collect or simulate the temperature value at that location. The first thermally conductive zone refers to all areas within the target reactor that may experience localized high temperatures due to intense reactions, concentrated heat transfer, etc. The second thermally conductive zone refers to all areas within the target furnace that may experience overheating due to furnace structure problems, concentrated power of localized heating elements, etc. There may be multiple first and second thermally conductive zones; the terms "first" and "second" are used for classification. The first-order overheating zone refers to the spatial region within the target reactor that exhibits potentially abnormal thermal energy changes under the combined influence of the first and second thermally conductive zones; that is, the overheated spatial blocks in the first-order overheating zone are all taken from the first thermally conductive zone. The second thermally conductive zone is used as an auxiliary influencing condition to determine which areas of the first thermally conductive zone will become overheated spatial blocks. The overheated space block refers to the space block in the first thermal conduction zone that is experiencing abnormal temperature rise. That is, the overheated space block can be a complete first thermal conduction zone or it can be represented by a combination of partial space blocks from multiple first thermal conduction zones. The time-step prediction method refers to a numerical method that discretizes the continuous heat conduction process into continuous time steps, performing state updates and trend predictions at each time step. By using the current first and second thermal conduction values ​​of the reactor and furnace as inputs, and employing a time-stepping approach, the thermal energy change trend of each first-order overheated zone at future moments is predicted, and this trend result is defined as the third thermal conduction value of the first-order overheated zone. In specific implementations, the Crank-Nicolson method or the finite element method can be used for calculation. Based on the third thermal conduction value representing the future thermal energy development trend within each first-order overheated zone, a vector field of its temperature gradient change is constructed. Then, in the direction where heat may conduct and diffuse, the next region that may experience local overheating, i.e., the second-order overheated zone, is predicted and marked. The purpose of using the third thermal conduction value is to determine the direction in which heat will expand in the future and how fast it will spread. The temperature gradient represents the rate and direction of temperature change in space. The temperature gradient vector field is a vector field describing the trend of temperature change in space, representing the direction vector within each spatial block, and indicating the direction and magnitude of heat flow. In specific implementations, the finite difference method can be used for calculation. The second-order superheated zone is used to pre-identify the next location where an abnormal temperature rise may occur, preparing for pre-regulation of the target reactor temperature.

[0022] Furthermore, as a feasible implementation method, the process of marking the first thermally conductive area and the second thermally conductive area includes: The process of starting up the target reactor and the target heater is defined as one working cycle. The heating rate threshold and the upper temperature limit threshold are set independently for each spatial block of the target reactor and the target heater. The heating rate and the upper temperature limit of the target reactor and the target heater are monitored during each working cycle. When the heating rate reaches the heating rate threshold or the upper temperature limit reaches the upper temperature limit threshold, it is determined that there is local overheating in the corresponding spatial block. All the spaces within the target reactor that are locally overheated are labeled as the first heat conduction zone by merging adjacent spaces; all the spaces within the heating furnace that are locally overheated are labeled as the second heat conduction zone by merging adjacent spaces.

[0023] The defined work cycle specifies the timeframe for monitoring and assessment. Within each work cycle, each spatial block of the reactor and heating furnace is independently monitored and evaluated. The heating rate threshold is used to detect the rate of temperature change, reflecting the dynamic characteristics of heat accumulation and transfer. The upper temperature limit threshold is used to detect whether the temperature has reached the maximum safe temperature allowed by the equipment or process. The heating rate threshold and upper temperature limit threshold can be set based on historical data or experience in specific implementations; they are set independently for each spatial block to reflect the differences in thermal sensitivity and design parameters of different parts. Within each work cycle, real-time monitoring is performed to check whether the heating rate of the spatial block reaches or exceeds the corresponding threshold and whether the temperature of the spatial block reaches or exceeds the corresponding upper temperature limit. If either condition is met, the block is determined to have localized overheating. This prevents thermal stress caused by excessively rapid heating and safety risks caused by exceeding temperature limits. All detected locally overheated spatial blocks are merged based on spatial adjacency to avoid fragmented assessments. The merged area better reflects the continuity of actual heat accumulation and diffusion, which is beneficial for subsequent dynamic prediction and adjustment. By employing both periodic and spatially independent determination methods, the system's adaptability to complex thermal conditions in reactors and heating furnaces is improved. Simultaneously, merging overheated blocks avoids misjudgments caused by point anomalies, enhancing the continuity and physical rationality of hot zone determination, which is beneficial for subsequent analysis of thermal gradients and heat flow directions.

[0024] Furthermore, as a feasible implementation method, the hot zone assessment strategy includes: Set an overheating space threshold. The first thermally conductive zone whose heat energy value reaches the overheating space threshold is marked as a first-order overheating zone. The first and second thermally conductive zones that are in adjacent contact are marked as the overheating zone to be detected by merging adjacent space blocks. The combined heat energy value in the overheating zone to be detected is calculated by weighted summation of the first and second thermally conductive values. When the combined heat energy value reaches the overheating space threshold, the space block of the currently detected first and second thermally conductive zones is determined to be an overheating space block, and the current overheating zone to be detected is determined to be a first-order overheating zone.

[0025] The overheating space threshold is used for secondary screening of the space region within the target reactor. First-order overheating zones are divided into two categories: the first category is areas where the first heat-conducting zone may experience abnormal temperature rise under the influence of the second heat-conducting zone; the second category is areas where the first heat-conducting zone experiences abnormal temperature rise due to the reactor's own reaction or other factors. Adjacent first and second heat-conducting zones are designated as the overheating zones to be detected by merging adjacent space blocks. This is because closer proximity results in less temperature loss during heat transfer and higher heating control precision, leading to shorter lengths or distances between the reactor and the heating furnace's heat transfer system or pipelines. However, while maintaining a safe distance, this may still generate additional heat transfer and temperature rise. Therefore, in this embodiment, a first-order overheating zone is included. The overheating space threshold can be set based on historical data or experience in specific implementations. The heat energy values ​​of the first and second heat-conducting zones are weighted and summed to synthesize an overall heat energy index. This weighting reflects the different influence weights of the two heat energy components on the overheating determination, flexibly reflecting the heat contribution of each region. This implementation upgrades the thermal energy determination from a single block to a multi-block joint determination, which can more closely reflect the complex actual heat conduction coupling process. By introducing weighted summation, it can differentiate the heat influence of different regions (reactor, heater), and the accurate identification of the first-order superheated zone also provides stable and accurate basic data for subsequent dynamic thermal energy trend analysis based on time step prediction and temperature gradient vector field.

[0026] Example 2: In this example, the calculation process for the third thermal conductivity value includes: Let T represent the real-time temperature value of the temperature field, and t represent the time step; let the first thermal conductivity be denoted as Q1, the second thermal conductivity as Q2, and the third thermal conductivity as Q3; let the average material density of the reactor be denoted as ρ, and let the isobaric specific heat capacity inside the reactor be c. ρ Let the ordinal number of the spatial block be i, and the volume of the spatial block be V; let m represent the thermal energy parameter. The formula for calculating the third thermal conductivity value Q3 is as follows: , Where, Q3 in the formula (i)T represents the third thermal conductivity value Q3 of the i-th spatial block, where T is the thermal conductivity value Q3 of the i-th spatial block. i (t+1) It represents the real-time temperature value predicted by the i-th spatial block at time step t+1.

[0027] Since the spatial blocks are equally divided, the volume of each spatial block is represented by V. The average material density ρ of the reactor refers to the average mass per unit volume of the filling material throughout the entire reactor and its internal space, i.e., the average mass concentration of the material per unit volume, used to represent the overall mass distribution of the reactor. The isobaric specific heat capacity within the reactor refers to the amount of heat required to increase the temperature by 1 degree Celsius per unit mass of material in the target reactor under constant pressure. In the formula... Partially representing the predicted real-time temperature value T of this spatial block i (t+1) The theoretical thermal energy value under the given conditions. In the formula... The part represents the weighted influence of the current thermal conductivity state on the future thermal trend. The thermal energy parameter m represents the degree of thermal inertia. It takes into account the influence of the residual heat of the system at the previous moment on the temperature field change at the next moment. In physical systems with slow heat conduction or thermal hysteresis, it can more accurately reflect the true trend.

[0028] Furthermore, as a feasible implementation, the real-time temperature value T predicted at the (t+1)th time step... i (t +1) The acquisition process includes: Obtain the real-time temperature value T of the target spatial block at the current time step. i t The system monitors and collects reactant flow rates and stirring speed values ​​of the target reactor after changes in reactant concentration. Within the current time step, it calculates reactant heat load using the reactant flow rate and mechanical heat load using the stirring speed. The reactant heat load and mechanical heat load values ​​are then added together to update the real-time temperature value. This updated real-time temperature value is set as the predicted real-time temperature T for the (t+1)th time step. i (t+1) .

[0029] The real-time temperature value T i tThe actual measured temperature of the i-th spatial block at the current time step t represents the initial baseline used for temperature prediction. The reactant flow rate value indicates the intensity of the reaction and is a direct factor affecting changes in heat load; the stirring speed value indicates the stirring intensity of the target reactor, and changes in this intensity will alter the heat convection rate and introduce energy consumption from the mechanical heat load. In a specific implementation, an online component analyzer can be used to monitor and collect the reactant flow rate value and the stirring speed value of the target reactor after changes in reactant concentration. As a feasible implementation, the reactant heat load value is denoted as Ta, the mechanical heat load value is denoted as Tb, and the formula for calculating the updated real-time temperature value can be set as: T i (t +1) =T i t +Ta+Tb. This implementation method can further improve data accuracy by fusing multi-source data of multiple process variables.

[0030] Example 3: In this example, the process of calculating the temperature gradient vector field to generate the second-order superheated region includes: In each first-order superheated region, a third thermal conductivity matrix is ​​constructed, and an initial gradient vector representing the heat transfer direction is calculated based on the third thermal conductivity matrix. The fluid simulation rate and eddy diffusion coefficient of the target reactor are generated using CFD simulation of the flow velocity field. The initial gradient vector is corrected using the fluid simulation rate and eddy diffusion coefficient and a modified gradient vector is generated. The coordinates of the spatial blocks in each first-order superheated region are labeled as first-order block coordinates. The first-order block coordinates are integrated along the direction of the modified gradient vector to represent the heat flow path update. The integrated values ​​are represented as the second-order spatial coordinates of each spatial block after the heat flow path update is completed. The temperature field spatial region composed of each spatial block corresponding to the second-order spatial coordinates is labeled as the second-order superheated region.

[0031] A matrix is ​​formed using the third thermal conductivity value of each spatial block within the first-order superheated zone to reflect the thermal energy distribution in that region. The gradient vector of the spatial thermal energy distribution is calculated based on this matrix. The direction of the gradient vector represents the "natural" direction of heat flow, i.e., from high thermal energy areas to low thermal energy areas, reflecting the basic direction of heat transfer. The simulated fluid velocity and eddy diffusion coefficient within the reactor are obtained through CFD simulation of the velocity field. The simulated fluid velocity represents the magnitude of the fluid flow speed at different times; the eddy diffusion coefficient represents the turbulent diffusion capacity in the fluid, affecting the efficiency of heat transfer. The initial gradient vector is corrected using the flow velocity and eddy diffusion coefficient to adjust the heat conduction direction and reflect the influence of fluid flow on heat transfer, making the gradient vector more consistent with the actual heat transfer path. The coordinates of the first-order blocks represent the spatial location of each spatial block within the first-order superheated zone. The coordinates of each block are integrated along the corrected gradient vector direction to simulate the path and distance of heat flow propagation along that direction. The integration result gives new spatial coordinates, which are the second-order spatial coordinates, representing the spatial location of the potential heat flow after transfer, i.e., the downstream region of heat propagation. Integrating along the direction of the modified gradient vector represents the trajectory tracking of future heat transfer paths in three-dimensional space based on the velocity direction and intensity of heat conduction. In practical applications, for example, the first-order block coordinates can be represented as P0, and the second-order spatial coordinates as P1. Then, within the time interval [t, t+Δt], the calculation formula for the second-order spatial coordinates of the current spatial block can be expressed as: In the formula, τ represents a time point, and G(τ) represents the heat transfer correction gradient vector in the current spatial block at time τ. That is, G(τ) is a time-varying vector function used to characterize the direction and intensity of heat transfer at that spatial point. The meaning of the second-order spatial coordinates can be expressed as follows: following the current heat conduction trend, heat will reach the next spatial block location; this location may initially have a normal temperature, but will soon become overheated due to concentrated heat flow. This embodiment, from both thermodynamic and hydrodynamic perspectives, forms a dynamic prediction of the future temperature rise region of the target reactor, improving the foresight and accuracy of the temperature control system.

[0032] Furthermore, as a feasible implementation, the initial gradient vector is calculated using the central difference method, which includes: For each spatial block in the spatial coordinate system, calculate the x, y, and z gradient components respectively. Let the difference in the third thermal conductivity value between the block and the adjacent spatial blocks be represented as the block temperature difference. The x-gradient component is calculated by taking the block temperature difference in the x-axis direction and dividing it by twice the step size in the x-axis direction. The y-gradient component is calculated by taking the block temperature difference in the y-axis direction and dividing it by twice the step size in the y-axis direction. The z-gradient component is calculated by taking the block temperature difference in the z-axis direction and dividing it by twice the z-axis step size. The x-gradient components, y-gradient components, and z-gradient components are combined to form the initial gradient vector.

[0033] The temperature gradient vector of each spatial block is calculated, representing the direction and rate of temperature change in three-dimensional space. This gradient is used for subsequent heat transfer path prediction and dynamic temperature control. The axial step size represents the distance between adjacent spatial blocks in the temperature field. As a feasible implementation, in a specific implementation, the central difference method is expressed as: The x gradient component is represented as: , The y-gradient component is represented as: , The z-gradient component is represented as: , The initial gradient vector is set to L and expressed as: .

[0034] In the formula, Δx, Δy, and Δz represent the step size of the spatial block along the corresponding coordinate axis, i.e., the distance between the center points of adjacent blocks. Compared with forward or backward differencing, the central difference method has higher accuracy, with a second-order error, and is suitable for calculating gradients under uniform grid conditions. By calculating the gradient vector of each spatial block, the direction of spatial heat flow can be captured more accurately.

[0035] Furthermore, the calculation process of the corrected gradient vector includes: Let the corrected gradient vector be denoted as Yc, the initial gradient vector as Yo, the fluid simulation velocity as ε, and the eddy diffusion coefficient as λ. Then, the formula for calculating the corrected gradient vector is: , Where in the formula The vector derivative of the fluid simulation rate is represented by , and the velocity curl is represented by .

[0036] The vector differential is the spatial derivative of the fluid velocity field, used to quantify the local velocity curl and direction of a fluid element, representing the influence of local flow field disturbances on the direction of heat flow; the larger the vector differential value, the more intense the fluid rotation at that point. The eddy diffusion coefficient λ represents the parameter that enhances heat diffusion due to local disturbances. The larger the eddy diffusion coefficient λ value, the stronger the correction effect of turbulence on the direction of heat flow. The product of the vector differential and the eddy diffusion coefficient represents the weighted influence factor of local velocity disturbances on the direction of heat flow migration. The partial representation represents the coefficients used to correct the initial gradient vector Yo. By calculating the corrected gradient vector, the heat flow direction can be predicted more accurately when modeling the temperature field, which helps to identify potential second-order overheating zones in advance and avoid process runaway caused by uneven temperature distribution or local overheating.

[0037] Furthermore, for the spatial block where the axial flow stirrer of the target reactor is located, the axial step size of the x-axis, y-axis, and z-axis is twice the radial length of the axial flow stirrer; when the block temperature difference in the z-axis direction exceeds twice the block temperature difference in the x-axis direction and the block temperature difference in the y-axis direction, numerical convergence is performed for the step size in the z-axis direction.

[0038] Axial flow stirrers generate strong fluid disturbances and mixing through rotation, leading to potentially drastic changes in temperature distribution and gradients around them. The spatial step size in the x, y, and z directions is set to twice the radial length of the axial flow stirrer, indicating that the size of the spatially divided blocks in that region has been enlarged. This is to match the physical dimensions of the stirrer and its scale of influence on fluid flow and heat transfer, avoiding computational complexity or errors caused by overly fine divisions. Using twice the radial length of the stirrer as the spatial step size ensures that the spatial division matches the physical dimensions of the equipment, facilitating accurate capture of the stirrer's influence on the temperature field. When the temperature difference in the z-direction blocks exceeds twice the temperature difference in both the x and y directions, it indicates a very significant temperature change in the z-direction, with strong axial heat flux variations, requiring an increase in the spatial resolution in the z-axis direction. Therefore, the step size in the z-axis direction is adjusted; in practice, this can be implemented by reducing the step size to improve computational accuracy until the gradient results stabilize, ensuring accurate and reliable simulation results.

[0039] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for dynamic heat transfer temperature control in a reactor heating furnace, characterized in that, The method includes: Step S1: Construct a temperature field in a spatial coordinate system, divide the spatial region of the temperature field into a number of spatial blocks along the coordinate axes, place the target reactor and the target heating furnace in the temperature field, mark the first heat conduction zone representing local overheating in the form of spatial blocks for the target reactor, and mark the second heat conduction zone representing local overheating in the form of spatial blocks for the target heating furnace. Step S2: Mark the size of the spatial block occupied by the first thermal conduction zone and the second thermal conduction zone along the spatial block in the temperature field, and mark the heat energy values ​​in the first thermal conduction zone and the second thermal conduction zone as the first thermal conduction value and the second thermal conduction value respectively, and construct a thermal zone evaluation strategy based on the first thermal conduction value and the second thermal conduction value. Step S3: Mark the superheated space blocks in the temperature field that indicate potential dynamic heat transfer in the first thermal conductivity zone using the thermal zone assessment strategy, and mark all superheated space blocks as first-order superheated zones; use the time step prediction method to predict and calculate the thermal energy trend value of each first-order superheated zone based on the first thermal conductivity value and the second thermal conductivity value, and mark it as the third thermal conductivity value. Step S4: Calculate the temperature gradient vector field of the corresponding local overheating zone using the third thermal conductivity value in each first-order overheating zone. Use the location pointed to by the temperature gradient vector field to generate a second-order overheating zone with the same spatial size as the first-order overheating zone and representing the end of potential dynamic heat transfer. Perform temperature overheating adjustment on the target reactor part covered by the second-order overheating zone.

2. The dynamic heat transfer temperature control method for a reactor heating furnace according to claim 1, characterized in that, The process of marking the first and second thermally conductive areas includes: The process of starting up the target reactor and the target heater is defined as one working cycle. The heating rate threshold and the upper temperature limit threshold are set independently for each spatial block of the target reactor and the target heater. The heating rate and the upper temperature limit of the target reactor and the target heater are monitored during each working cycle. When the heating rate reaches the heating rate threshold or the upper temperature limit reaches the upper temperature limit threshold, it is determined that there is local overheating in the corresponding spatial block. All the spaces within the target reactor that are locally overheated are labeled as the first heat conduction zone by merging adjacent spaces; all the spaces within the heating furnace that are locally overheated are labeled as the second heat conduction zone by merging adjacent spaces.

3. The dynamic heat transfer temperature control method for a reactor heating furnace according to claim 2, characterized in that, The hot zone assessment strategy includes the following: Set an overheating space threshold. The first thermally conductive zone whose heat energy value reaches the overheating space threshold is marked as a first-order overheating zone. The first and second thermally conductive zones that are in adjacent contact are marked as the overheating zone to be detected by merging adjacent space blocks. The combined heat energy value in the overheating zone to be detected is calculated by weighted summation of the first and second thermally conductive values. When the combined heat energy value reaches the overheating space threshold, the space block of the currently detected first and second thermally conductive zones is determined to be an overheating space block, and the current overheating zone to be detected is determined to be a first-order overheating zone.

4. The dynamic heat transfer temperature control method for a reactor heating furnace according to claim 1, characterized in that, The calculation process for the third thermal conductivity value includes: Let T represent the real-time temperature value of the temperature field, and t represent the time step; let the first thermal conductivity be denoted as Q1, the second thermal conductivity as Q2, and the third thermal conductivity as Q3; let the average material density of the reactor be denoted as ρ, and let the isobaric specific heat capacity inside the reactor be c. ρ Let the ordinal number of the spatial block be i, and the volume of the spatial block be V; let m represent the thermal energy parameter. The formula for calculating the third thermal conductivity value Q3 is as follows: , Where, Q3 in the formula (i) T represents the third thermal conductivity value Q3 of the i-th spatial block, where T is the thermal conductivity value Q3 of the i-th spatial block. i (t+1) It represents the real-time temperature value predicted by the i-th spatial block at time step t+1.

5. The dynamic heat transfer temperature control method for a reactor heating furnace according to claim 4, characterized in that, The real-time temperature value T predicted at the (t+1)th time step i (t+1) The acquisition process includes: Obtain the real-time temperature value T of the target spatial block at the current time step. i t The system monitors and collects reactant flow rates and stirring speed values ​​of the target reactor after changes in reactant concentration. Within the current time step, it calculates reactant heat load using the reactant flow rate and mechanical heat load using the stirring speed. The reactant heat load and mechanical heat load values ​​are then added together to update the real-time temperature value. This updated real-time temperature value is set as the predicted real-time temperature T for the (t+1)th time step. i (t+1) .

6. The dynamic heat transfer temperature control method for a reactor heating furnace according to claim 4, characterized in that, The process of calculating the temperature gradient vector field to generate the second-order superheated region includes: In each first-order superheated region, a third thermal conductivity matrix is ​​constructed, and an initial gradient vector representing the heat transfer direction is calculated based on the third thermal conductivity matrix. The fluid simulation rate and eddy diffusion coefficient of the target reactor are generated using CFD simulation of the flow velocity field. The initial gradient vector is corrected using the fluid simulation rate and eddy diffusion coefficient and a modified gradient vector is generated. The coordinates of the spatial blocks in each first-order superheated region are labeled as first-order block coordinates. The first-order block coordinates are integrated along the direction of the modified gradient vector to represent the heat flow path update. The integrated values ​​are represented as the second-order spatial coordinates of each spatial block after the heat flow path update is completed. The temperature field spatial region composed of each spatial block corresponding to the second-order spatial coordinates is labeled as the second-order superheated region.

7. A dynamic heat transfer temperature control method for a reactor heating furnace according to claim 6, characterized in that, The initial gradient vector is calculated using the central difference method. This process includes: For each spatial block in the spatial coordinate system, calculate the x, y, and z gradient components respectively. Let the difference in the third thermal conductivity value between the block and the adjacent spatial blocks be represented as the block temperature difference. The x-gradient component is calculated by taking the block temperature difference in the x-axis direction and dividing it by twice the step size in the x-axis direction. The y-gradient component is calculated by taking the block temperature difference in the y-axis direction and dividing it by twice the step size in the y-axis direction. The z-gradient component is calculated by taking the block temperature difference in the z-axis direction and dividing it by twice the z-axis step size. The x-gradient components, y-gradient components, and z-gradient components are combined to form the initial gradient vector.

8. A dynamic heat transfer temperature control method for a reactor heating furnace according to claim 6, characterized in that, The calculation process of the corrected gradient vector includes: Let the corrected gradient vector be denoted as Yc, the initial gradient vector as Yo, the fluid simulation velocity as ε, and the eddy diffusion coefficient as λ. Then, the formula for calculating the corrected gradient vector is: , Where in the formula The vector derivative of the fluid simulation rate is represented by , and the velocity curl is represented by .

9. A dynamic heat transfer temperature control method for a reactor heating furnace according to claim 7, characterized in that, For the spatial block where the axial flow agitator of the target reactor is located, the axial step size of the x-axis, y-axis, and z-axis is twice the radial length of the axial flow agitator.

10. A dynamic heat transfer temperature control method for a reactor heating furnace according to claim 7, characterized in that, When the block temperature difference in the z-axis direction exceeds twice the block temperature difference in both the x-axis and y-axis directions, numerical convergence is performed with respect to the step size in the z-axis direction.

Images