A medicament reaction monitoring method and system based on constant temperature control

Abstract

The application relates to the technical field of temperature control, and particularly discloses a medicament reaction monitoring method and system based on constant-temperature control, which comprises the following steps: generating a temperature change rate distribution gradient field in the interior of a reaction container and predicting a thermal diffusion trend vector diagram of a next control period according to a detection point data set with space-time characteristics; determining a feedforward decoupling control component based on local extreme points in the thermal diffusion trend vector diagram and real-time temperature values, and taking a difference between a global average temperature and a set target temperature as a main deviation; integrating and adaptively coupling the main deviation to generate a feedback adjustment reference quantity according to a temperature field uniformity index, and preventing overshoot fusion of the feedforward decoupling control component and the feedback adjustment reference quantity to generate a comprehensive driving instruction; and the application effectively solves the problems of temperature response lag, overshoot oscillation and poor adaptability to temperature field distribution changes, and improves the temperature control precision and stability of a medicament reaction process.

Description

Technical Field

[0001] This invention relates to the field of temperature control technology, and in particular to a method and system for monitoring pharmaceutical reactions based on constant temperature control. Background Technology

[0002] In the field of pharmaceutical reaction monitoring, isothermal control is a core element in ensuring the safety of chemical reaction processes and the stability of product quality. Multiple temperature sensors are typically installed inside the reaction vessel to collect temperature change signals of reactants during synthesis, crystallization, fermentation, and other processes. Based on the collected data, heating or cooling actuators are driven to maintain the reaction system within a preset target temperature range. Existing isothermal control methods mainly rely on PID control or conventional feedback control, adjusting output power by detecting the deviation between the global average temperature and the set target temperature. However, these methods only focus on the current temperature amplitude, ignoring the dynamic propagation characteristics of temperature in space and time. They struggle to predict the direction of heat diffusion and convergence trends within the reaction vessel, causing control adjustments to lag behind the actual physical process of temperature changes, resulting in response delays and localized overheating or overcooling.

[0003] Existing isothermal control methods lack the ability to adapt to the non-uniformity of temperature field distribution at the feedback control strategy level. They struggle to dynamically adjust the intensity of the integral action based on temperature differences at different spatial locations within the reaction vessel, leading to a significant exacerbation of the integral accumulation effect when the temperature field uniformity is poor. This easily triggers temperature overshoot and long-period oscillations. Furthermore, traditional methods lack an overshoot prevention arbitration mechanism based on the energy accumulation of temperature fluctuations when integrating feedforward and feedback control. This makes it difficult to balance suppressing overshoot and maintaining response speed, resulting in frequent actuator switching or control output saturation, further deteriorating temperature control quality. Therefore, there is an urgent need to develop an isothermal control and monitoring method that can predict heat diffusion trends, decouple feedforward control components, and perform adaptive integral fusion based on temperature field uniformity. This would address the problems of response lag, overshoot oscillations, and poor adaptability to changes in temperature field distribution in existing methods, thereby improving the accuracy and stability of temperature control in pharmaceutical reaction processes. Summary of the Invention

[0004] This invention provides a method and system for monitoring pharmaceutical reactions based on constant temperature control, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for monitoring pharmaceutical reactions based on isothermal control, comprising: K1: Based on the detection point dataset with spatiotemporal features, generate the temperature change rate distribution gradient field inside the reaction vessel and predict the thermal diffusion trend vector diagram for the next control cycle. K2: Based on the local extreme points in the thermal diffusion trend vector diagram and the real-time temperature values, determine the feedforward decoupling control component, and take the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation. K3: Based on the temperature field uniformity index inside the reaction vessel, the main deviation is integrally adaptively coupled to generate a feedback adjustment reference quantity, and the feedforward decoupling control component and the feedback adjustment reference quantity are fused together to generate a comprehensive drive command.

[0006] In a preferred embodiment, the detection point dataset with spatiotemporal features includes: Real-time temperature signals at the locations of each sensor are collected synchronously at a fixed sampling period; Based on the real-time temperature signal, the spatial coordinates of each location point, the temperature value at the current moment, and the temperature change rate at the current moment are associated and stored to obtain a detection point dataset with spatiotemporal features.

[0007] In a preferred embodiment, generating the temperature change rate distribution gradient field inside the reaction vessel includes: Based on the temperature change rate and spatial coordinates in the detection point dataset, determine the temperature change rate gradient vector of the detection point; Using the temperature change rate gradient vector of all detection points as known nodes, radial difference is performed on the spatial points inside the reaction vessel to obtain the temperature change rate gradient vector at the spatial points. The temperature change rate gradient vector is arranged continuously according to spatial coordinates to obtain the temperature change rate distribution gradient field.

[0008] In a preferred embodiment, the thermal diffusion trend vector diagram for predicting the next control cycle includes: The reverse direction of the temperature change rate gradient vector in spatial coordinates is taken as the heat diffusion direction vector at each point. The thermal diffusion direction vectors at all spatial points are smoothed, and the magnitude of the smoothed thermal diffusion direction vectors is used as the weight of the thermal diffusion rate. All the smoothed thermal diffusion direction vectors and their corresponding thermal diffusion rate weights are arranged according to spatial coordinates to obtain the thermal diffusion trend vector diagram for the next control cycle.

[0009] In a preferred embodiment, determining the feedforward decoupling control component based on the local extrema points in the thermal diffusion trend vector diagram and the real-time temperature values ​​includes: Extract all heat diffusion direction vectors pointing to the local highest temperature point in the real-time temperature value from the heat diffusion trend vector diagram, and use the vector sum of all heat diffusion direction vectors as the heat convergence vector; Extract all heat diffusion direction vectors that leave the local minimum temperature point in the real-time temperature value from the heat diffusion trend vector diagram, and use the vector sum of all heat diffusion direction vectors as the heat loss vector; Based on the heat convergence vector and the heat loss vector, the cooling trend suppression amount and the heating trend suppression amount are determined respectively; Multiply the cooling trend suppression amount by a preset first temperature difference value to obtain a cooling pre-compensation component, and multiply the heating trend suppression amount by a preset second temperature difference value to obtain a heating pre-compensation component. The cooling pre-compensation component and the heating pre-compensation component are combined into a feedforward decoupling control component.

[0010] In a preferred embodiment, using the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation includes: Extract the real-time temperature values ​​of all locations from the detection point dataset with spatiotemporal features; The global average temperature is obtained by averaging the real-time temperature values ​​at all locations. The difference between the global average temperature and the target temperature inside the reaction vessel is used as the main deviation.

[0011] In a preferred embodiment, the temperature field uniformity index inside the reaction vessel includes: The absolute deviation value of each detection point is obtained by subtracting the real-time temperature value of each detection point from the global average temperature and then taking the absolute value. The largest absolute deviation value among the detection points is used as the temperature field uniformity index inside the reaction vessel.

[0012] In a preferred embodiment, the step of integrally adaptively coupling the main deviation to generate a feedback adjustment reference quantity includes: The temperature field uniformity index is used as an adjustment factor to dynamically adjust the integral velocity adjustment coefficient. A nonlinear dead zone is set for the main deviation, and it is determined whether the main deviation falls within the dead zone range; Based on the dead zone determination result and the integral speed adjustment coefficient, the main deviation is conditionally integrated and accumulated to generate the integral accumulation amount. The main deviation is multiplied by a preset proportional coefficient and then added to the integral accumulation to obtain the feedback adjustment reference value.

[0013] In a preferred embodiment, the step of fusing the feedforward decoupling control component and the feedback adjustment reference quantity to prevent overshoot and generate a comprehensive drive command includes: The cumulative energy value of temperature fluctuation is obtained by squaring each major deviation and then summing the results. The formula for calculating the cumulative energy value of temperature fluctuations is:

[0014] in, This is the cumulative energy value of the temperature fluctuation. The preset cumulative number of periods, For the first The main deviation of each control cycle; Based on the accumulated energy value of the temperature fluctuation, determine whether to enter the overshoot suppression mode; In normal operating mode, the positive value of the feedback adjustment reference quantity is added to the heating pre-compensation component, and the negative value is added to the cooling pre-compensation component to obtain the preliminary fusion quantity; In the overshoot suppression mode, the corresponding components in the feedback adjustment reference quantity and the feedforward decoupling control component are multiplied by the suppression coefficient and then fused to obtain the preliminary fused quantity. Based on the upper and lower output limits inside the reaction vessel, the output amplitude of the initial fusion amount is limited. If the initial fusion amount is greater than the upper limit of the output, the upper limit of the output will be used as the final output value. If the initial fusion amount is less than the lower limit of the output, the lower limit of the output will be used as the final output value. If the initial fusion amount is between the two, it will be used directly as the final output value.

[0015] To address the above problems, the present invention also provides a pharmaceutical reaction monitoring system based on constant temperature control, the system comprising: The thermal diffusion trend prediction module is used to generate a gradient field of temperature change rate distribution inside the reaction vessel based on a dataset of detection points with spatiotemporal characteristics, and to predict the thermal diffusion trend vector map for the next control cycle. The feedforward decoupling module is used to determine the feedforward decoupling control component based on the local extreme points in the thermal diffusion trend vector diagram and the real-time temperature value, and to take the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation. The adaptive integral fusion module is used to perform integral adaptive coupling on the main deviation based on the temperature field uniformity index inside the reaction vessel, generate a feedback adjustment reference quantity, and perform anti-overshoot fusion on the feedforward decoupling control component and the feedback adjustment reference quantity to generate a comprehensive drive command.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention constructs a temperature change rate distribution gradient field and predicts a thermal diffusion trend vector diagram to identify the accumulation and loss paths of heat inside the reaction vessel in advance. Based on this, a feedforward decoupled control signal containing heating pre-compensation and cooling pre-compensation components is generated. This mechanism allows control commands to intervene before actual temperature deviations occur, transforming the response mode from passive feedback to active prediction, significantly shortening the lag time, and solving the regulation delay problem caused by thermal inertia. It is particularly suitable for pharmaceutical reactions with violent exothermic or rapid endothermic reactions.

[0017] 2. This invention introduces the maximum absolute deviation as an index of temperature field uniformity, dynamically adjusts the integral speed, and avoids excessive accumulation of integrals due to local temperature differences. Simultaneously, it establishes an overshoot suppression mechanism based on the accumulated sum of squared principal deviations, automatically attenuating the gain control according to temperature fluctuation energy and limiting the output amplitude. This solves the problems of integral saturation and parameter tuning difficulties, effectively suppresses overshoot and oscillation, and improves control stability and product yield under non-uniform temperature fields. Attached Figure Description

[0018] Figure 1 This is a schematic flowchart of a drug reaction monitoring method based on isothermal control, provided in an embodiment of the present invention. Figure 2 A functional block diagram of a drug reaction monitoring system based on constant temperature control provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] This application provides a method for monitoring pharmaceutical reactions based on isothermal control. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for monitoring pharmaceutical reactions based on isothermal control can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0021] Reference Figure 1The diagram shown is a schematic flowchart of a drug reaction monitoring method based on isothermal control according to an embodiment of the present invention. In this embodiment, the drug reaction monitoring method based on isothermal control includes: K1: Based on the detection point dataset with spatiotemporal features, generate the temperature change rate distribution gradient field inside the reaction vessel and predict the thermal diffusion trend vector diagram for the next control cycle. In this embodiment of the invention, the detection point dataset with spatiotemporal features includes: Real-time temperature signals at the locations of each sensor are collected synchronously at a fixed sampling period; Based on the real-time temperature signal, the spatial coordinates of each location point, the temperature value at the current moment, and the temperature change rate at the current moment are associated and stored to obtain a detection point dataset with spatiotemporal features.

[0022] The temperature change rate distribution gradient field inside the generating reaction vessel includes: Based on the temperature change rate and spatial coordinates in the detection point dataset, determine the temperature change rate gradient vector of the detection point; Using the temperature change rate gradient vector of all detection points as known nodes, radial difference is performed on the spatial points inside the reaction vessel to obtain the temperature change rate gradient vector at the spatial points. The temperature change rate gradient vector is arranged continuously according to spatial coordinates to obtain the temperature change rate distribution gradient field.

[0023] The vector diagram predicting the thermal diffusion trend for the next control cycle includes: The reverse direction of the temperature change rate gradient vector in spatial coordinates is taken as the heat diffusion direction vector at each point. The thermal diffusion direction vectors at all spatial points are smoothed, and the magnitude of the smoothed thermal diffusion direction vectors is used as the weight of the thermal diffusion rate. All the smoothed thermal diffusion direction vectors and their corresponding thermal diffusion rate weights are arranged according to spatial coordinates to obtain the thermal diffusion trend vector diagram for the next control cycle.

[0024] Multiple temperature sensors are pre-embedded at different heights and radial positions inside the reaction vessel. All sensors are activated simultaneously with the same fixed sampling period, and each sensor outputs a real-time temperature signal at each sampling time. For each sensor location, the control system subtracts the temperature value of the previous sampling period from the temperature value measured in the current sampling period, and then divides this difference by the duration of the fixed sampling period to calculate the rate of temperature change at that location during that time period.

[0025] The three pieces of information—the pre-calibrated spatial coordinates of the location point inside the reaction vessel, the real-time temperature value measured in the current sampling period, and the temperature change rate just calculated—are linked and stored in the same record. The collection of records for all sensor locations forms a detection point dataset with spatiotemporal characteristics.

[0026] The temperature change rate and spatial coordinates of each detection point are extracted one by one from the detection point dataset with spatiotemporal features. For each detection point, the neighboring detection points within a predetermined distance range are searched. The temperature change rate of the detection point is subtracted from the temperature change rate of the neighboring detection points to obtain the difference. Then, this difference is divided by the spatial straight-line distance between the two detection points to obtain the numerical value of the rate of temperature change in that direction.

[0027] These rapid and slow temperature values ​​in all directions are combined into spatial vectors according to their respective directions. The resulting vector is the temperature change rate gradient vector at the detection point. This gradient vector indicates the spatial direction in which the rate of temperature change increases most rapidly from that point. Taking the temperature change rate gradient vectors of all detection points as known nodes, for any spatial point inside the reaction vessel that is not the location of the sensor, find the nearest several detection points around that point. The gradient vectors of these detection points are used as a basis for a weighted summation, where the weights are determined by the distance between the spatial point and each detection point. The result of the weighted summation is the temperature change rate gradient vector at that spatial point.

[0028] The temperature change rate gradient vector directly calculated at each detection point and the temperature change rate gradient vector obtained by weighted summation at each non-detection point are continuously filled into a grid covering the entire three-dimensional space inside the reaction vessel according to their spatial coordinates inside the reaction vessel. This ensures that each cell in the grid has a temperature change rate gradient vector, thus generating a complete temperature change rate distribution gradient field.

[0029] For each spatial point in the temperature change rate distribution gradient field, extract the temperature change rate gradient vector at that point, and reverse the direction of this vector completely. That is, take the opposite direction from the direction that originally pointed to the direction of the fastest increase in temperature change rate as the heat diffusion direction vector at that point, because heat always diffuses from the high temperature region to the low temperature region, and the opposite direction of the fastest increase in temperature change rate is the dominant direction of heat diffusion.

[0030] The process involves iterating through the heat diffusion direction vectors at all spatial points. When a significant abrupt change is found between the heat diffusion direction vector at a given point and its neighboring points, the original vector at that point is replaced with the average value of the heat diffusion direction vectors from its multiple neighboring points. This ensures that the heat diffusion direction vectors within the entire reaction vessel change continuously in space without directional jumps. This process is called smoothing the heat diffusion direction vectors at all spatial points. For each smoothed heat diffusion direction vector, its magnitude is calculated. This magnitude is calculated by squaring the vector's components along the three spatial axes, summing them, and then taking the square root. This magnitude value is used as the weight of the heat diffusion rate at that spatial point; a larger magnitude indicates a stronger driving force for heat diffusion at that point.

[0031] Each smoothed thermal diffusion direction vector is bound together with its corresponding thermal diffusion rate weight, and arranged in the order of the spatial coordinates of these spatial points inside the reaction vessel. This is then filled into a three-dimensional grid diagram corresponding to the temperature change rate distribution gradient field. Each cell in this grid diagram contains a thermal diffusion direction vector representing the direction of heat diffusion and a thermal diffusion rate weight representing the strength of diffusion ability. This complete three-dimensional grid diagram is the thermal diffusion trend vector diagram for the next control cycle.

[0032] The beneficial effects are as follows: Based on the temperature change rate and spatial coordinates of the detection point dataset, the temperature change rate gradient vector of each detection point is determined. This vector is then used for radial interpolation of spatial points inside the reaction vessel to complete the gradient vectors at non-detection points. Finally, all gradient vectors are continuously arranged according to spatial coordinates to generate a complete temperature change rate distribution gradient field. Furthermore, the inverse direction of this gradient vector is used as the heat diffusion direction vector. After smoothing all direction vectors, their magnitudes are used as weights for the heat diffusion rate. Arranged according to spatial coordinates, a heat diffusion trend vector map for the next control cycle is obtained. This series of operations enables temperature control to utilize spatial gradient information to anticipate the direction and speed of heat diffusion inside the reaction vessel, significantly improving the ability to predict temperature changes. This represents a leap from passive response to active prediction, effectively shortening the control lag time and providing accurate spatiotemporal variation data for subsequent feedforward compensation.

[0033] K2: Based on the local extreme points in the thermal diffusion trend vector diagram and the real-time temperature values, determine the feedforward decoupling control component, and take the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation. In this embodiment of the invention, determining the feedforward decoupling control component based on the local extreme points in the thermal diffusion trend vector diagram and the real-time temperature value includes: Extract all heat diffusion direction vectors pointing to the local highest temperature point in the real-time temperature value from the heat diffusion trend vector diagram, and use the vector sum of all heat diffusion direction vectors as the heat convergence vector; Extract all heat diffusion direction vectors that leave the local minimum temperature point in the real-time temperature value from the heat diffusion trend vector diagram, and use the vector sum of all heat diffusion direction vectors as the heat loss vector; Based on the heat convergence vector and the heat loss vector, the cooling trend suppression amount and the heating trend suppression amount are determined respectively; Multiply the cooling trend suppression amount by a preset first temperature difference value to obtain a cooling pre-compensation component, and multiply the heating trend suppression amount by a preset second temperature difference value to obtain a heating pre-compensation component. The cooling pre-compensation component and the heating pre-compensation component are combined into a feedforward decoupling control component.

[0034] The step of using the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation includes: Extract the real-time temperature values ​​of all locations from the detection point dataset with spatiotemporal features; The global average temperature is obtained by averaging the real-time temperature values ​​at all locations. The difference between the global average temperature and the target temperature inside the reaction vessel is used as the main deviation.

[0035] Find the spatial point with the highest real-time temperature from the heat diffusion trend vector map and mark it as the local temperature maximum point. Then, search all heat diffusion direction vectors in the heat diffusion trend vector map and filter out those vectors whose direction directly points to the local temperature maximum point. Add these filtered heat diffusion direction vectors according to the vector addition rule, that is, add the three axial components of each vector separately. The resulting composite vector is the heat convergence vector. The magnitude and direction of this vector represent the overall trend of heat converging towards the local temperature maximum point.

[0036] Find the spatial point with the lowest real-time temperature from the heat diffusion trend vector map and mark it as the local minimum temperature point. Then, search all heat diffusion direction vectors in the heat diffusion trend vector map and filter out those vectors whose direction leaves the local minimum temperature point, i.e., those pointing outward from the point. Accumulate these filtered heat diffusion direction vectors according to the vector addition rule, that is, add the three axial components of each vector separately. The resulting composite vector is the heat loss vector. The magnitude and direction of this vector represent the overall trend of heat loss from the local minimum temperature point to the outside.

[0037] Multiplying the magnitude of the heat accumulation vector (i.e., the square root of the sum of the squares of its three spatial axial components) by a pre-defined first transformation constant yields the cooling trend suppression amount, which reflects the intensity of the cooling effect required at the local temperature peak. Simultaneously, multiplying the magnitude of the heat loss vector by a pre-defined second transformation constant yields the heating trend suppression amount, which reflects the intensity of the heating effect required at the local temperature minimum.

[0038] A first temperature difference value is preset, which is the difference between the real-time temperature value of the local highest temperature point and the target temperature value set inside the reaction vessel. Before system operation, the value of the first temperature difference value is determined according to the allowable temperature deviation range required for the reagent reaction. The cooling trend suppression amount is multiplied by the first temperature difference value to obtain the cooling pre-compensation component. This cooling pre-compensation component is used to initiate the cooling action in advance in subsequent control to suppress further temperature rise at the local high temperature point.

[0039] A second temperature difference value is preset. This second temperature difference value is the difference between the target temperature value set inside the reaction vessel and the real-time temperature value of the local lowest temperature point. Before the system starts, the value of this second temperature difference value is determined according to the allowable temperature deviation range required for the reagent reaction. The heating trend suppression amount is multiplied by this second temperature difference value to obtain the heating pre-compensation component. This heating pre-compensation component is used to initiate the heating action in advance in subsequent control to prevent further cooling of the local low temperature point.

[0040] The cooling pre-compensation component and the heating pre-compensation component are combined as two independent output channels to form a control quantity that simultaneously contains cooling control information and heating control information. This control quantity is the feedforward decoupling control component, in which the cooling pre-compensation component and the heating pre-compensation component act independently on the subsequent fusion steps.

[0041] The real-time temperature value is extracted from each record in the spatiotemporal detection point dataset, and all extracted real-time temperature values ​​are collected to form a temperature value set. The sum of all temperature values ​​in this set is then calculated, and divided by the number of temperature values ​​in the set. The quotient is the average temperature of all detection points inside the entire reaction vessel, denoted as the global average temperature. Subtracting the pre-set target temperature value inside the reaction vessel from this global average temperature yields the principal deviation. The sign and magnitude of this principal deviation reflect the direction and degree of deviation of the current overall temperature from the target temperature.

[0042] The beneficial effects are as follows: Cooling trend suppression and heating trend suppression are determined based on the heat accumulation vector and heat loss vector, respectively. The cooling trend suppression is multiplied by a preset first temperature difference value to obtain a cooling pre-compensation component, and the heating trend suppression is multiplied by a preset second temperature difference value to obtain a heating pre-compensation component. These two components are combined to form a feedforward decoupling control component. This feedforward decoupling control component can independently apply suppressive control to both local high-temperature and local low-temperature regions simultaneously existing within the reaction vessel, avoiding the problem of mutual interference between cooling and heating regulation in traditional methods. By organically combining feedforward decoupling control based on thermal diffusion trends with main deviation feedback control based on global average temperature, the system can suppress local heat sources in advance before heat causes a significant change in the overall temperature, significantly improving the response speed and anti-interference capability of temperature control.

[0043] K3: Based on the temperature field uniformity index inside the reaction vessel, the main deviation is integrally adaptively coupled to generate a feedback adjustment reference quantity, and the feedforward decoupling control component and the feedback adjustment reference quantity are fused together to generate a comprehensive drive command.

[0044] In this embodiment of the invention, the temperature field uniformity index inside the reaction vessel includes: The absolute deviation value of each detection point is obtained by subtracting the real-time temperature value of each detection point from the global average temperature and then taking the absolute value. The largest absolute deviation value among the detection points is used as the temperature field uniformity index inside the reaction vessel.

[0045] The step of performing integral adaptive coupling on the main deviation to generate a feedback adjustment reference quantity includes: The temperature field uniformity index is used as an adjustment factor to dynamically adjust the integral velocity adjustment coefficient. A nonlinear dead zone is set for the main deviation, and it is determined whether the main deviation falls within the dead zone range; Based on the dead zone determination result and the integral speed adjustment coefficient, the main deviation is conditionally integrated and accumulated to generate the integral accumulation amount. The main deviation is multiplied by a preset proportional coefficient and then added to the integral accumulation to obtain the feedback adjustment reference value.

[0046] The step of fusing the feedforward decoupling control component and the feedback adjustment reference quantity to prevent overshoot and generate a comprehensive drive command includes: The cumulative energy value of temperature fluctuation is obtained by squaring each major deviation and then summing the results. The formula for calculating the cumulative energy value of temperature fluctuations is:

[0047] in, This is the cumulative energy value of the temperature fluctuation. The preset cumulative number of periods, For the first The main deviation of each control cycle; Based on the accumulated energy value of the temperature fluctuation, determine whether to enter the overshoot suppression mode; In normal operating mode, the positive value of the feedback adjustment reference quantity is added to the heating pre-compensation component, and the negative value is added to the cooling pre-compensation component to obtain the preliminary fusion quantity; In the overshoot suppression mode, the corresponding components in the feedback adjustment reference quantity and the feedforward decoupling control component are multiplied by the suppression coefficient and then fused to obtain the preliminary fused quantity. Based on the upper and lower output limits inside the reaction vessel, the output amplitude of the initial fusion amount is limited. If the initial fusion amount is greater than the upper limit of the output, the upper limit of the output will be used as the final output value. If the initial fusion amount is less than the lower limit of the output, the lower limit of the output will be used as the final output value. If the initial fusion amount is between the two, it will be used directly as the final output value.

[0048] For each detection point, the real-time temperature value is recorded. This real-time temperature is then subtracted from the previously calculated global average temperature. The absolute value of the subtraction is the absolute deviation value for that detection point. This process is repeated for all detection points to obtain the absolute deviation value for each point. The largest absolute deviation value is then selected as the temperature uniformity index within the reaction vessel. A larger value indicates greater temperature differences between points within the reaction vessel, i.e., a more non-uniform temperature field; a smaller value indicates a more uniform temperature distribution.

[0049] The temperature field uniformity index is used as an adjustment factor to dynamically change the integral speed adjustment coefficient. Specifically, a baseline value for the integral speed adjustment coefficient is preset. When the temperature field uniformity index increases, the integral speed adjustment coefficient is increased according to a preset proportional relationship. When the temperature field uniformity index decreases, the integral speed adjustment coefficient is decreased according to the same proportional relationship, so that the integral speed adjustment coefficient and the temperature field uniformity index change in the same direction.

[0050] A dead zone threshold range is pre-defined for the main deviation. This dead zone threshold range includes a positive boundary and a negative boundary, and the absolute values ​​of these two boundaries are equal. The values ​​are set before system operation based on the allowable temperature fluctuation range of the reagent reaction. The current main deviation value is compared with the dead zone threshold range. If the absolute value of the main deviation is less than the absolute value of the dead zone threshold boundary, the main deviation is determined to fall within the dead zone range; otherwise, the main deviation is determined not to fall within the dead zone range.

[0051] When the main deviation is determined to fall within the dead zone, the value in the integral accumulator is cleared to zero, and no integral accumulation operation is performed on the current main deviation. When the main deviation is determined not to fall within the dead zone, the current main deviation is multiplied by the dynamically adjusted integral speed regulation coefficient, and then multiplied by the duration of the control cycle to obtain the current integral increment. This integral increment is then accumulated in the integral accumulator to obtain the current integral accumulation. The current main deviation is multiplied by a preset proportional coefficient to obtain the proportional regulation component. The proportional regulation component is then added to the integral accumulation, and the result is the feedback regulation reference value.

[0052] The accumulated temperature fluctuation energy value is stored in the temperature fluctuation energy accumulation register and is used to quantify the overall fluctuation intensity of the main deviation over multiple past control cycles. The preset number of accumulation cycles is determined in advance based on the dynamic response characteristics of the reagent reaction process and the desired smoothness before the system runs, representing the total number of control cycles involved in the energy accumulation calculation. It is an index number used to distinguish different control cycles in the past, incrementing sequentially from the first cycle to the next. One cycle. The first... The main deviation of a control cycle is the difference between the global average temperature and the set target temperature within that cycle. This difference may be positive or negative, but its squared value is always non-negative.

[0053] First, determine the preset cumulative period number, and then retrieve past data from the historical records. For each control cycle, the principal deviation value is calculated. For each extracted principal deviation value, it is multiplied by itself to obtain the squared principal deviation value for that cycle. Then, starting from the squared principal deviation value of the first control cycle, the values ​​of the second, third, and so on are added sequentially until the fourth cycle. The squared value of the principal deviation for each control cycle, and all The sum of the squared values ​​is the cumulative energy of temperature fluctuation. This value reflects the past... The control system calculates the cumulative energy of temperature deviation from the target within each cycle and then compares this value with a preset energy threshold. When the cumulative energy value of temperature fluctuation exceeds the threshold, it determines that it should enter the overshoot suppression mode, which attenuates the control output to effectively suppress temperature overshoot and oscillation.

[0054] The main deviation of each of the past control cycles is squared, and the results of all squares are summed to obtain the cumulative temperature fluctuation energy value. An energy threshold is preset, which is determined before system operation based on the maximum allowable temperature fluctuation of the reagent reaction. The cumulative temperature fluctuation energy value is compared with the energy threshold. If the cumulative temperature fluctuation energy value is greater than the energy threshold, the system is determined to enter the overshoot suppression mode; otherwise, it is determined to enter the normal operation mode.

[0055] In normal operating mode, check the sign of the feedback adjustment reference value. If the feedback adjustment reference value is positive, add the positive value to the heating pre-compensation component, and set the cooling pre-compensation component to zero to obtain the initial fusion amount. If the feedback adjustment reference value is negative, add the absolute value of the negative value to the cooling pre-compensation component and then take the negative value, and set the heating pre-compensation component to zero to obtain the initial fusion amount.

[0056] In overshoot suppression mode, the preset suppression coefficient is read. This suppression coefficient is a constant greater than zero and less than one. The feedback adjustment reference amount is multiplied by the suppression coefficient to obtain the attenuated feedback amount. The heating pre-compensation component and the cooling pre-compensation component are also multiplied by the suppression coefficient to obtain the attenuated heating pre-compensation component and the attenuated cooling pre-compensation component, respectively. Then, the attenuated feedback amount and the corresponding attenuated pre-compensation component are fused according to the same rules as in normal working mode to obtain the preliminary fused amount.

[0057] The upper and lower output limits are preset, and these values ​​are determined by the maximum drive signal that the heaters and cooling medium valves inside the reaction vessel can withstand. The initial fusion amount is compared with these two limits. If the initial fusion amount is greater than the upper limit, the upper limit is used as the final output value. If the initial fusion amount is less than the lower limit, the lower limit is used as the final output value. If the initial fusion amount falls between the upper and lower limits, it is directly used as the final output value. This final output value is the integrated drive command.

[0058] The beneficial effects are as follows: By dynamically adjusting the integral speed regulation coefficient using the temperature field uniformity index as an adjustment factor, the integral action intensity is adaptively adjusted according to the temperature field uniformity. The more non-uniform the temperature field, the weaker the integral, effectively avoiding excessive accumulation of integral and temperature overshoot caused by excessive local temperature differences. Simultaneously, a non-linear dead zone is set for the main deviation; integral accumulation only occurs when the main deviation exceeds the dead zone range, eliminating integral saturation and frequent actuator movements caused by small deviations. In normal operating mode, the feedback regulation reference quantity and the corresponding pre-compensation component are directly fused. In overshoot suppression mode, all control quantities are multiplied by the suppression coefficient before fusion. Finally, the output of the initially fused quantity is limited to generate a comprehensive drive command. This series of operations enables the control system to actively attenuate the control gain when the temperature fluctuation energy is high and restore normal control strength when the energy is low, significantly suppressing overshoot and oscillation phenomena. It also enhances the adaptive capability to non-uniform temperature fields, ensuring the temperature stability of the reagent reaction process and product consistency.

[0059] like Figure 2The diagram shown is a functional block diagram of a drug reaction monitoring system based on constant temperature control provided in an embodiment of the present invention.

[0060] The pharmaceutical reaction monitoring system 100 based on isothermal control described in this invention can be installed in an electronic device. Depending on the functions implemented, the pharmaceutical reaction monitoring system 100 based on isothermal control may include a thermal diffusion trend prediction module 101, a feedforward decoupling module 102, and an adaptive integral fusion module 103. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0061] In this embodiment, the functions of each module / unit are as follows: The thermal diffusion trend prediction module 101 is used to generate a gradient field of temperature change rate distribution inside the reaction vessel based on a dataset of detection points with spatiotemporal characteristics, and to predict a thermal diffusion trend vector diagram for the next control cycle. The feedforward decoupling module 102 is used to determine the feedforward decoupling control component based on the local extreme points in the thermal diffusion trend vector diagram and the real-time temperature value, and to take the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation. The adaptive integral fusion module 103 is used to perform integral adaptive coupling on the main deviation based on the temperature field uniformity index inside the reaction vessel, generate a feedback adjustment reference quantity, and perform anti-overshoot fusion on the feedforward decoupling control component and the feedback adjustment reference quantity to generate a comprehensive drive command.

[0062] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0063] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0064] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0065] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0066] The embodiments of this application can acquire and process relevant data based on an artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for monitoring pharmaceutical reactions based on isothermal control, characterized in that, The method includes: K1: Based on the detection point dataset with spatiotemporal features, generate the temperature change rate distribution gradient field inside the reaction vessel and predict the thermal diffusion trend vector diagram for the next control cycle. K2: Based on the local extreme points in the thermal diffusion trend vector diagram and the real-time temperature values, determine the feedforward decoupling control component, and take the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation. K3: Based on the temperature field uniformity index inside the reaction vessel, the main deviation is integrally adaptively coupled to generate a feedback adjustment reference quantity, and the feedforward decoupling control component and the feedback adjustment reference quantity are fused together to generate a comprehensive drive command.

2. The method for monitoring pharmaceutical reactions based on isothermal control as described in claim 1, characterized in that, The detection point dataset with spatiotemporal features includes: Real-time temperature signals at the locations of each sensor are collected synchronously at a fixed sampling period; Based on the real-time temperature signal, the spatial coordinates of each location point, the temperature value at the current moment, and the temperature change rate at the current moment are associated and stored to obtain a detection point dataset with spatiotemporal features.

3. The method for monitoring pharmaceutical reactions based on isothermal control as described in claim 1, characterized in that, The temperature change rate distribution gradient field inside the generating reaction vessel includes: Based on the temperature change rate and spatial coordinates in the detection point dataset, determine the temperature change rate gradient vector of the detection point; Using the temperature change rate gradient vector of all detection points as known nodes, radial difference is performed on the spatial points inside the reaction vessel to obtain the temperature change rate gradient vector at the spatial points. The temperature change rate gradient vector is arranged continuously according to spatial coordinates to obtain the temperature change rate distribution gradient field.

4. The method for monitoring pharmaceutical reactions based on isothermal control as described in claim 3, characterized in that, The vector diagram predicting the thermal diffusion trend for the next control cycle includes: The reverse direction of the temperature change rate gradient vector in spatial coordinates is taken as the heat diffusion direction vector at each point. The thermal diffusion direction vectors at all spatial points are smoothed, and the magnitude of the smoothed thermal diffusion direction vectors is used as the weight of the thermal diffusion rate. All the smoothed thermal diffusion direction vectors and their corresponding thermal diffusion rate weights are arranged according to spatial coordinates to obtain the thermal diffusion trend vector diagram for the next control cycle.

5. The method for monitoring pharmaceutical reactions based on isothermal control as described in claim 1, characterized in that, The step of determining the feedforward decoupling control component based on the local extreme points in the thermal diffusion trend vector diagram and the real-time temperature value includes: Extract all heat diffusion direction vectors pointing to the local highest temperature point in the real-time temperature value from the heat diffusion trend vector diagram, and use the vector sum of all heat diffusion direction vectors as the heat convergence vector; Extract all heat diffusion direction vectors that leave the local minimum temperature point in the real-time temperature value from the heat diffusion trend vector diagram, and use the vector sum of all heat diffusion direction vectors as the heat loss vector; Based on the heat convergence vector and the heat loss vector, the cooling trend suppression amount and the heating trend suppression amount are determined respectively; Multiply the cooling trend suppression amount by a preset first temperature difference value to obtain a cooling pre-compensation component, and multiply the heating trend suppression amount by a preset second temperature difference value to obtain a heating pre-compensation component. The cooling pre-compensation component and the heating pre-compensation component are combined into a feedforward decoupling control component.

6. The method for monitoring pharmaceutical reactions based on isothermal control as described in claim 1, characterized in that, The step of using the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation includes: Extract the real-time temperature values ​​of all locations from the detection point dataset with spatiotemporal features; The global average temperature is obtained by averaging the real-time temperature values ​​at all locations. The difference between the global average temperature and the target temperature inside the reaction vessel is used as the main deviation.

7. The method for monitoring pharmaceutical reactions based on isothermal control as described in claim 6, characterized in that, The temperature field uniformity index inside the reaction vessel includes: The absolute deviation value of each detection point is obtained by subtracting the real-time temperature value of each detection point from the global average temperature and then taking the absolute value. The largest absolute deviation value among the detection points is used as the temperature field uniformity index inside the reaction vessel.

8. The method for monitoring pharmaceutical reactions based on isothermal control as described in claim 1, characterized in that, The step of performing integral adaptive coupling on the main deviation to generate a feedback adjustment reference quantity includes: The temperature field uniformity index is used as an adjustment factor to dynamically adjust the integral velocity adjustment coefficient. A nonlinear dead zone is set for the main deviation, and it is determined whether the main deviation falls within the dead zone range; Based on the dead zone determination result and the integral speed adjustment coefficient, the main deviation is conditionally integrated and accumulated to generate the integral accumulation amount. The main deviation is multiplied by a preset proportional coefficient and then added to the integral accumulation to obtain the feedback adjustment reference value.

9. The method for monitoring pharmaceutical reactions based on isothermal control as described in claim 5, characterized in that, The step of fusing the feedforward decoupling control component and the feedback adjustment reference quantity to prevent overshoot and generate a comprehensive drive command includes: The cumulative energy value of temperature fluctuation is obtained by squaring each major deviation and then summing the results. The formula for calculating the cumulative energy value of temperature fluctuations is: ； in, This is the cumulative energy value of the temperature fluctuation. The preset cumulative number of periods, For the first The main deviation of each control cycle; Based on the accumulated energy value of the temperature fluctuation, determine whether to enter the overshoot suppression mode; In normal operating mode, the positive value of the feedback adjustment reference quantity is added to the heating pre-compensation component, and the negative value is added to the cooling pre-compensation component to obtain the preliminary fusion quantity; In the overshoot suppression mode, the corresponding components in the feedback adjustment reference quantity and the feedforward decoupling control component are multiplied by the suppression coefficient and then fused to obtain the preliminary fused quantity. Based on the upper and lower output limits inside the reaction vessel, the output amplitude of the initial fusion amount is limited. If the initial fusion amount is greater than the upper limit of the output, the upper limit of the output will be used as the final output value. If the initial fusion amount is less than the lower limit of the output, the lower limit of the output will be used as the final output value. If the initial fusion amount is between the two, it will be used directly as the final output value.

10. A pharmaceutical reaction monitoring system based on isothermal control, characterized in that, The system for implementing the isothermal control-based pharmaceutical reaction monitoring method of claim 1 includes: The thermal diffusion trend prediction module is used to generate a gradient field of temperature change rate distribution inside the reaction vessel based on a dataset of detection points with spatiotemporal characteristics, and to predict the thermal diffusion trend vector map for the next control cycle. The feedforward decoupling module is used to determine the feedforward decoupling control component based on the local extreme points in the thermal diffusion trend vector diagram and the real-time temperature value, and to take the difference between the global average temperature inside the reaction vessel and the set target temperature as the main deviation. The adaptive integral fusion module is used to perform integral adaptive coupling on the main deviation based on the temperature field uniformity index inside the reaction vessel, generate a feedback adjustment reference quantity, and perform anti-overshoot fusion on the feedforward decoupling control component and the feedback adjustment reference quantity to generate a comprehensive drive command.

Images