Chemical production-oriented chemical reaction kettle temperature control method and system

Abstract

The invention relates to the technical field of non-electrical variable control, in particular to a chemical reaction kettle temperature control method and system for chemical production. The method comprises the following steps: at each moment, predicting the predicted temperature of the next moment according to the actual temperature of each experienced moment; determining a prediction error of each moment in the prediction time domain according to the experienced actual temperature and the predicted temperature of each moment; determining system adjusting parameters according to the change condition of the actual temperature at each experienced moment; solving the objective function to obtain a temperature control parameter value at the current moment; the target function comprises a prediction error item and a control item; the prediction error item is determined based on the prediction error and the prediction temperature of each moment in the prediction time domain; the control item is determined based on a system adjustment parameter and a temperature control parameter to be solved at each moment in a control time domain; and performing proportional integral differential control based on the temperature control parameter value at the current moment. By adopting the method, the accuracy of chemical reaction kettle temperature control can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of non-electric variable control, and in particular to a temperature control method and system for a chemical reactor for chemical production. Background Art

[0002] In chemical production, reactor temperature control is a critical step in ensuring product quality, production safety, and process efficiency. Many chemical reactions are extremely sensitive to temperature. Excessively high temperatures can lead to runaway reactions, sudden pressure increases, and even explosions or hazardous gas leaks. Certain reactions (such as polymerization and nitration) are highly exothermic, and if heat is not dissipated promptly, thermal runaway accidents can occur. Temperature directly affects reaction rate, selectivity, and conversion rate. Temperature fluctuations can lead to increased side reactions, decreased product purity, and even the production of substandard products. Chemical reactor temperature control is not only crucial to production safety, product quality, and economic efficiency, but also a key technology for achieving intelligent and green production.

[0003] Traditional temperature control methods rely primarily on manual adjustments or simple PID control (proportional-integral-derivative control). These methods monitor the temperature signal within the reactor and adjust the flow rate or power of the heating or cooling medium to achieve temperature control. However, chemical reactions often exhibit nonlinearity, time-varying characteristics, and significant hysteresis. This makes traditional PID control, particularly in complex reactions or large-scale production, difficult to adapt to dynamic operating conditions and prone to overshoot, oscillation, and response lag. This leads to inaccurate temperature control, impacting reaction selectivity and yield. Summary of the Invention

[0004] In order to solve the technical problem of inaccurate temperature control of chemical reactors, the purpose of the present invention is to provide a chemical reactor temperature control method and system for chemical production. The technical solutions adopted are as follows: The present invention provides a temperature control method for a chemical reactor for chemical production, the method comprising: At each moment, the predicted temperature at the next moment is predicted based on the actual temperature at each moment that has passed; Determining the prediction error for each moment in a prediction time domain based on the actual temperature and the predicted temperature at each moment that has passed; the prediction time domain includes the past moment and the next moment; Determine the system adjustment parameters based on the actual temperature changes at each moment; Solving an objective function to obtain a temperature control parameter value at a current moment; the objective function includes a prediction error term and a control term; the prediction error term is determined based on the prediction error and the predicted temperature at each moment in the prediction time domain; the control term is determined based on the system adjustment parameter and the temperature control parameter to be solved at each moment in the control time domain; Proportional-integral-derivative control is performed based on the temperature control parameter value at the current moment.

[0005] According to the temperature control method for a chemical reactor in chemical production provided by the present invention, determining the prediction error at each moment in the prediction time domain based on the actual temperature and the predicted temperature at each moment that has been experienced includes: Determining a prediction error coefficient for each time instant based on the actual temperature and the predicted temperature at each time instant; Performing function fitting based on the prediction error coefficients at each experienced moment to obtain a prediction error function; The prediction error at each moment in the prediction time domain is determined according to the prediction error function.

[0006] According to the temperature control method for a chemical reactor for chemical production provided by the present invention, at each moment, based on the actual temperature at each moment, the predicted temperature at the next moment is predicted, including: Starting from the initial time period, at each moment, the predicted temperature at the next moment is predicted based on the actual temperature at each moment that has passed; The step of determining the prediction error coefficient at each experienced moment based on the actual temperature and the predicted temperature at each experienced moment includes: Determining the temperature difference between the actual temperature and the predicted temperature at each time point that has passed except the initial time period, and the time difference between each time point that has passed except the initial time period and the initial time period; A prediction error coefficient for each time point that has elapsed except for the initial time period is determined based on the temperature difference and the time difference for each time point that has elapsed except for the initial time period.

[0007] According to the chemical reactor temperature control method for chemical production provided by the present invention, determining the prediction error coefficient of each time point that has passed except the initial time period based on the temperature difference and the time difference at each time point that has passed except the initial time period includes: For each time instant that has passed except the initial time period, a prediction error coefficient at the current time instant is determined according to the product of the temperature difference at the time instant and the time difference.

[0008] According to the temperature control method for a chemical reactor in chemical production provided by the present invention, determining the system adjustment parameters according to the actual temperature changes at each time point includes: Determine the slope between data points corresponding to actual temperatures at adjacent moments in the past; Clustering the slopes to obtain a number of clusters; For each chemical reaction stage corresponding to the cluster, determining the temperature change degree of the chemical reaction stage corresponding to the cluster according to the slope and temperature in the chemical reaction stage; The system adjustment parameters are determined according to the temperature change degree of the chemical reaction stage corresponding to each cluster.

[0009] According to the chemical reactor temperature control method for chemical production provided by the present invention, the temperature change degree of the chemical reaction stage corresponding to each cluster is determined according to the slope and temperature in the chemical reaction stage, including: For each chemical reaction stage corresponding to the cluster, determining the degree of dispersion and central tendency of the slope in the chemical reaction stage, and the temperature span in the chemical reaction stage; The temperature variation degree of the chemical reaction stage corresponding to the cluster is determined according to the dispersion degree, the central tendency and the temperature span.

[0010] According to the chemical reactor temperature control method for chemical production provided by the present invention, determining the temperature change degree of the chemical reaction stage corresponding to the cluster according to the discrete degree, the central tendency and the temperature span includes: The temperature variation degree of the chemical reaction stage corresponding to the cluster is determined according to the product of the dispersion degree, the central tendency and the temperature span.

[0011] According to the chemical reactor temperature control method for chemical production provided by the present invention, determining the system adjustment parameters according to the temperature change degree of the chemical reaction stage corresponding to each cluster includes: The system adjustment parameter is determined according to the average value of the temperature change degree of the chemical reaction stage corresponding to each cluster.

[0012] According to the chemical reactor temperature control method for chemical production provided by the present invention, solving the objective function to obtain the temperature control parameter value at the current moment includes: Solving the objective function to obtain the temperature control parameter value at each moment in the control time domain, and extracting the temperature control parameter value at the current moment from the temperature control parameter value at each moment in the control time domain; the control time domain includes the experienced moments.

[0013] The present invention provides a chemical reactor temperature control system for chemical production. The system includes a memory and a processor; the memory is used to store executable program code; the processor is used to call and run the executable program code from the memory to implement the chemical reactor temperature control method for chemical production provided by the present invention.

[0014] The present invention has the following beneficial effects: at each moment, based on the actual temperature of each moment that has been experienced, the predicted temperature of the next moment is predicted, and based on the actual temperature of each moment that has been experienced and the predicted temperature, the prediction error of each moment in the prediction time domain is determined, and based on the change of the actual temperature of each moment that has been experienced, the system adjustment parameter is determined, and then the prediction error term is determined based on the prediction error and the predicted temperature of each moment in the prediction time domain, and the control term is determined based on the system adjustment parameter and the temperature control parameter to be solved at each moment in the control time domain, and the objective function is constructed and solved based on the prediction error term and the control term to obtain the temperature control parameter value at the current moment, and proportional integral differential control is performed based on the temperature control parameter value at the current moment, so as to compensate for the large lag effect in advance by predicting the future temperature trend, and perform temperature error correction in real time, thereby transforming the passive correction of the traditional PID control method into active intervention, and improving the accuracy of temperature control of the chemical reactor. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A schematic flow chart of a temperature control method for a chemical reactor in chemical production provided by one embodiment of the present invention; Figure 2 The overall control process of the temperature control method of a chemical reactor for chemical production provided by one embodiment of the present invention; Figure 3 A schematic diagram of a flow chart for determining prediction errors at each moment in a prediction time domain provided by one embodiment of the present invention; Figure 4 A schematic diagram of a flow chart for determining system adjustment parameters provided by one embodiment of the present invention; Figure 5 This is an architectural diagram of a chemical reactor temperature control system for chemical production provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a method and system for controlling the temperature of a chemical reactor in chemical production, including its specific implementation, structure, features, and effectiveness. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The specific scheme of the temperature control method and system for a chemical reactor for chemical production provided by the present invention is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a schematic flow chart of a temperature control method for a chemical reactor in chemical production provided by one embodiment of the present invention, comprising the following steps: Step 101: At each moment, based on the actual temperatures at each moment, predict the predicted temperature at the next moment.

[0021] The experienced moments include historical moments and the current moment. The duration between each moment can be set according to the actual situation, for example, one moment per second.

[0022] In one embodiment, a temperature sensor may be installed in the chemical reactor to collect the real-time actual temperature of the chemical reactor.

[0023] In one embodiment, multiple temperature sensors can be installed at different locations within the chemical reactor to avoid measurement errors caused by localized temperature variations. For example, sensors can be installed at the top, middle, and bottom of the reactor. Temperature data collected by multiple temperature sensors can be integrated using methods such as weighted averaging or Kalman filtering to improve measurement reliability.

[0024] In one embodiment, the temperature sensor can be a thermocouple or a resistance temperature detector (RTD). For example, a Pt100 or Pt1000 RTD can be used. Thermocouples are based on the Seebeck effect: when two dissimilar metal conductors change temperature, a potential difference is generated. The temperature is inferred by measuring the voltage. Thermocouples are characterized by a wide measurement range (-200°C to 1800°C), fast response, and low cost. However, they have low accuracy (±1°C to 2°C), require cold junction compensation, and are susceptible to electromagnetic interference. Thermocouples are suitable for high-temperature, rapidly changing reactions, such as polymerization or cracking reactions. RTDs measure temperature by exploiting the temperature-dependent resistance of metals. They are characterized by high accuracy (±0.1°C to 0.5°C), good stability, and excellent linearity. However, they have a narrow measurement range (-200°C to 600°C), slow response, and higher cost than thermocouples. RTDs are suitable for precision temperature control at medium and low temperatures, such as in the pharmaceutical and fine chemical industries.

[0025] In one embodiment, the temperature sensor can transmit the collected temperature data to a programmable logic controller (PLC) or distributed control system (DCS) for filtering and linearization compensation. For example, the linearization compensation can be thermocouple cold-junction compensation. The PLC or DCS can then transmit the filtered and linearized temperature data to a SCADA (Supervisory Control and Data Acquisition) system or a cloud platform. The SCADA or cloud platform can then execute the chemical reactor temperature control method for chemical production provided herein to implement real-time monitoring, historical data storage, and alarm functions.

[0026] In one embodiment, temperature data can be transmitted using analog or digital signals. Analog signals can use 4-20 mA current signals, which are highly resistant to interference and suitable for industrial sites. Digital signals can use RS485, HART, or wireless transmission. For example, wireless transmission can use LoRa (a long-range, low-power radio technology standard) or ZigBee (a short-range, low-rate wireless communication technology).

[0027] In one embodiment, starting from an initial time period, at each moment, the predicted temperature at the next moment can be predicted based on the actual temperature at each moment. The initial time period is a time period of a preset duration at the beginning of the chemical reaction. For example, the preset duration can be 10 seconds.

[0028] In one embodiment, a time series prediction model can be used to predict the temperature at the next moment based on the actual temperature at each moment. For example, the time series prediction model includes ARIMA (Autoregressive Integrated Moving Average).

[0029] Step 102: Determine the prediction error at each moment in the prediction time domain based on the actual temperature at each moment that has passed and the predicted temperature; the prediction time domain includes the moment that has passed and the next moment.

[0030] In one embodiment, a prediction error function is determined based on the actual temperature and the predicted temperature at each time instant, and the prediction error at each time instant in the prediction time domain is determined based on the prediction error function.

[0031] In one embodiment, a prediction error function can be determined based on the actual temperature and the predicted temperature at each time point after the initial time period, and the prediction error at each time point in the prediction time domain can be determined based on the prediction error function. The prediction time domain includes the time point after the initial time period and the next time point.

[0032] Step 103: determining the system adjustment parameters according to the actual temperature changes at each moment.

[0033] Understandably, because the temperature change rate within the reactor is not constant, rapid temperature changes can pose serious safety risks, especially for highly exothermic reactions, which can lead to thermal runaway and sudden pressure increases, or even explosions. Furthermore, this can cause product quality to decline, such as increased side reactions, uneven molecular weight distribution, or catalyst deactivation. When the temperature changes too slowly, not only can the reaction cycle be significantly prolonged, reducing production efficiency, but it can also lead to incomplete reactions, accumulation of intermediates, or crystal defects. Therefore, it is necessary to analyze the temperature changes in different reaction processes to determine the system adjustment parameters.

[0034] In one embodiment, the slope between data points corresponding to actual temperatures at adjacent moments in the past is determined, and the system adjustment parameter at the current moment is determined based on the slope.

[0035] Step 104, solve the objective function to obtain the temperature control parameter value at the current moment; the objective function includes a prediction error term and a control term; the prediction error term is determined based on the prediction error and the predicted temperature at each moment in the prediction time domain; the control term is determined based on the system adjustment parameter and the temperature control parameter to be solved at each moment in the control time domain.

[0036] The temperature control parameter is a parameter used to control the temperature. For example, the temperature control parameter can be the opening of a steam valve or the heating power. The temperature control parameter value refers to the specific value of the temperature control parameter.

[0037] Model predictive control (MPC) is an advanced control strategy based on dynamic models, rolling optimization, and feedback correction. It is widely used in fields such as chemical engineering, energy, and process control. Its core concept is to predict the system's behavior over a period of time using a dynamic model and solve a finite-time optimal control problem within each control cycle to determine the optimal control action at the current moment. The core process consists of three steps: First, based on the current system state and a mathematical model, the output response is predicted for multiple time steps into the future, such as the temperature trajectory of a reactor. Then, a control sequence is calculated through rolling optimization to minimize an objective function, such as the weighted sum of temperature tracking error and the change in the controlled variable, while satisfying process constraints, such as temperature safety limits and valve opening ranges. Finally, the optimized first controlled variable is executed and, in the next cycle, feedback correction is performed based on real-time measurements to compensate for model errors and external disturbances.

[0038] In one embodiment, the product of the prediction error and the predicted temperature at each moment in the prediction time domain is accumulated to obtain a prediction error term. The temperature control parameter to be solved at each moment in the control time domain is accumulated, and the accumulated result is multiplied by the system adjustment coefficient at the current moment to obtain a control term.

[0039] In one embodiment, the objective function may be the sum of the prediction error term and the control term. The temperature control parameter at the current moment may be obtained by solving the objective function to minimize its value.

[0040] In one embodiment, the objective function can be expressed as follows: in, represents the objective function. represents the prediction error term. Represents a control item. Indicates the The forecast error at that moment. Indicates the The predicted temperature at that moment. Represents the prediction time domain. Indicates the System adjustment parameters at a certain moment. Indicates the Temperature control parameters at each moment. Indicates the control time domain. Indicates taking the minimum value.

[0041] Reflect the The prediction error at each moment is synchronously controlled with the predicted temperature in the reactor. The larger the prediction error and the predicted temperature, the greater the degree of correction of the temperature in the reactor. In the prediction time domain, the prediction error of the system mainly depends on the changes in the stable stage of the reaction. Therefore, the moment with the minimum error is selected as the output of the objective function. The smaller it is, the smaller the error input is. System adjustment parameters It is used to control the smoothness and response speed of temperature regulation, which mainly depends on the change of temperature. If the temperature changes faster, the corresponding system adjustment parameter will be larger. Conversely, the slower the temperature changes, the smaller the system adjustment parameter will be. Indicates the Temperature control parameters at a certain moment, such as the opening of the steam valve.

[0042] In one embodiment, the prediction time domain includes the time that has passed since the initial time period and the next time period. The control time domain includes the time that has passed since the initial time period. For example, if the initial time period is 10 seconds after the chemical reaction begins, and the current time period is 15 seconds, then the prediction time domain is from 11 to 16 seconds, and the control time domain is from 11 to 15 seconds.

[0043] In one embodiment, the formula for solving the objective function is expressed as follows: in, Represents the target temperature control parameter sequence. Represents the search for the objective function The independent variable for which the minimum value is obtained A collection of .

[0044] Step 105 : Perform proportional-integral-derivative control based on the temperature control parameter value at the current moment.

[0045] Among them, proportional-integral-derivative control (PID control) is a control method that controls the controlled object by forming a deviation between a given value and an actual output value, and linearly combining the deviation into a control quantity according to proportion, integration and differentiation.

[0046] It can be understood that solving the objective function can obtain the temperature control parameter values ​​at each moment in the control time domain. (target temperature control parameter sequence), which contains the temperature control parameter values ​​at the current time t , the temperature control parameter value at the current time t Send it to the actuator for proportional integral differential control, for example, send it to the regulating valve based on the opening value of the steam valve at the current moment Proportional-integral-derivative control is performed to control the steam valve opening, thereby accurately controlling the temperature. Here, m represents the end time of the initial time period, m+1 represents the first time after the initial time period, and t represents the current time. For example, in the above example, solving the objective function yields the temperature control parameter values ​​for each time period from seconds 11 to 15 (the control time domain). Proportional-integral-derivative control is then performed based on the temperature control parameter value at the 15th second (the current time).

[0047] like Figure 2 The figure shows the overall control process, which generates the target temperature control parameter sequence through MPC rolling optimization. , from which the current temperature control parameters are extracted Send it to the PID controller as an implicit target for PID control, and output the change in the temperature control parameter value , realize temperature control, and then feedback the current temperature , continue with MPC rolling optimization.

[0048] The proportional integral differential control formula is as follows: in, Indicates the change in the temperature control parameter value at the tth moment. Represents the proportional gain coefficient. Indicates the integral gain coefficient. Represents the differential gain coefficient. represents the predicted temperature at time t. represents the actual temperature at the tth moment, Represents the temperature deviation at the tth moment, that is . Indicates the number of time periods that have passed up to time t.

[0049] As an indirect target of PID control, the temperature deviation Indirect influence, if the actual temperature If the trajectory deviates from the prediction, the PID Fine-tune the temperature control parameters in real time, for example, fine-tune the opening of the steam valve so that the value of the temperature control parameter approaches .

[0050] The above-mentioned chemical reactor temperature control method for chemical production predicts the predicted temperature of the next moment at each moment based on the actual temperature of each moment that has been experienced, and determines the prediction error of each moment in the prediction time domain based on the actual temperature and the predicted temperature of each moment that has been experienced, determines the system adjustment parameter according to the change of the actual temperature at each moment that has been experienced, and then determines the prediction error term based on the prediction error and the predicted temperature of each moment in the prediction time domain, determines the control term based on the system adjustment parameter and the temperature control parameter to be solved at each moment in the control time domain, constructs the objective function based on the prediction error term and the control term and solves it to obtain the temperature control parameter value at the current moment, and performs proportional integral differential control based on the temperature control parameter value at the current moment, thereby compensating for the large lag effect in advance by predicting the future temperature trend, and performing temperature error correction in real time, thereby transforming the passive correction of the traditional PID control method into active intervention, and improving the accuracy of temperature control of the chemical reactor.

[0051] In one embodiment, see Figure 3 Step 102 determines the prediction error at each moment in the prediction time domain based on the actual temperature and the predicted temperature at each moment, and includes the following steps: Step 1021: Determine the prediction error coefficient at each time point based on the actual temperature and the predicted temperature at each time point.

[0052] Among them, the prediction error coefficient is used to characterize the size of the prediction error.

[0053] Step 1022: Perform function fitting based on the prediction error coefficients at each time instant to obtain a prediction error function.

[0054] In one embodiment, the prediction error coefficients of each moment are combined into a data sequence, and a fitting function is obtained using the least squares fitting algorithm as the prediction error function. .

[0055] Step 1023: Determine the prediction error at each moment in the prediction time domain according to the prediction error function.

[0056] The time value of each moment in the prediction time domain can be substituted into the prediction error function to obtain the prediction error of each moment in the prediction time domain. For example: assuming that the prediction time domain is from the 11th second to the 16th second, the prediction errors of each moment in the prediction time domain are 、 、 、 、 、 .

[0057] In the above embodiment, since the temperature changes differently at different stages, the accuracy of temperature prediction is also different. Therefore, a prediction error function can be obtained by performing function fitting based on the prediction error coefficients of each moment that has been experienced, so that the prediction error at each moment in the prediction time domain can be accurately determined based on the prediction error function.

[0058] In one embodiment, at each moment, predicting the predicted temperature at the next moment based on the actual temperature at each previously experienced moment includes: starting from an initial time period, at each moment, predicting the predicted temperature at the next moment based on the actual temperature at each previously experienced moment. Determining a prediction error coefficient for each previously experienced moment based on the actual temperature and the predicted temperature at each previously experienced moment includes: determining the temperature difference between the actual temperature and the predicted temperature at each previously experienced moment except the initial time period, and the time difference between each previously experienced moment except the initial time period and the initial time period; and determining the prediction error coefficient for each previously experienced moment except the initial time period based on the temperature difference and the time difference at each previously experienced moment except the initial time period.

[0059] In one embodiment, the difference between the actual temperature and the predicted temperature at each time point that has passed except the initial time period may be used as the temperature difference between the actual temperature and the predicted temperature at each time point that has passed except the initial time period.

[0060] In one embodiment, the duration between each time instant that has passed except the initial time period and the end time instant of the initial time period is used as the time difference between each time instant that has passed except the initial time period and the initial time period.

[0061] In one embodiment, the prediction error coefficient of each time instant except the initial time period is positively correlated with the temperature difference of each time instant except the initial time period, and is positively correlated with the time difference of each time instant except the initial time period.

[0062] In the above embodiment, since the greater the difference between the predicted temperature and the actual temperature, and the farther the current moment is from the initial time period, the less accurate the current prediction is, that is, the larger the prediction error coefficient at the current moment, therefore, based on the temperature difference at each moment other than the initial time period and the time difference from the initial time period, the prediction error coefficient at each moment other than the initial time period can be accurately determined.

[0063] In one embodiment, the prediction error coefficient of each moment that has passed except the initial time period is determined based on the temperature difference and time difference of each moment that has passed except the initial time period, including: for each moment that has passed except the initial time period, the prediction error coefficient of the current moment is determined based on the product of the temperature difference at the moment and the time difference.

[0064] in, Indicates the The forecast error coefficient at each moment. Indicates the The actual temperature at a moment. Indicates the The predicted temperature at that moment. Indicates the Temperature difference at each moment. Indicates the The time interval between the initial time period time difference.

[0065] In the above embodiment, the greater the difference between the predicted temperature and the actual temperature, and the farther the current moment is from the initial time period, the less accurate the current prediction is, that is, the larger the prediction error coefficient at the current moment. Therefore, the prediction error coefficient at the current moment can be accurately determined based on the product of the temperature difference at the moment and the time difference between the moment and the initial time period.

[0066] In one embodiment, see Figure 4 Step 103 determines the system adjustment parameters according to the actual temperature changes at each moment, including the following steps: Step 1031 : Determine the slope between the data points corresponding to the actual temperature at adjacent moments in each of the experienced moments.

[0067] Step 1032: cluster the slopes to obtain a number of clusters.

[0068] In one embodiment, a slope sequence is formed by the slopes between data points corresponding to the actual temperatures at adjacent moments in the past. The slope sequence is clustered using a hierarchical clustering algorithm to obtain a plurality of clusters. In one embodiment, the preset clustering level in the hierarchical clustering algorithm can be set to 2, i.e., clustering stops after 2 iterations.

[0069] Step 1033 : For each chemical reaction stage corresponding to each cluster, determine the temperature change degree of the chemical reaction stage corresponding to the cluster according to the slope and temperature in the chemical reaction stage.

[0070] Among them, the degree of temperature change represents the severity of the temperature change during the chemical reaction stage.

[0071] Understandably, the temperature change rate of a chemical reactor varies during different chemical reaction stages. Initially (the initiation phase), the temperature typically changes slowly, as the activation energy barrier must be overcome, resulting in a significant hysteresis. Once the reaction enters the acceleration phase, the temperature rises exponentially due to chain reactions or autocatalysis. During the stable reaction phase, the system reaches dynamic equilibrium, temperature fluctuations stabilize, and the heat release rate and heat dissipation rate are essentially equal. At the end of the reaction (the decay phase), the temperature must be slowly cooled according to process requirements. This is particularly true for crystallization or polymer reactions, where step-wise cooling can prevent product defects. Therefore, each cluster corresponds to a specific chemical reaction stage.

[0072] Step 1034 : determining system adjustment parameters according to the temperature variation degree of the chemical reaction stage corresponding to each cluster.

[0073] In one embodiment, the system adjustment parameter is positively correlated with the degree of temperature change in the chemical reaction stage corresponding to each cluster.

[0074] In the above embodiment, since the temperature change rate of the chemical reactor is different in different chemical reaction stages, the chemical reaction stages are divided by clustering according to the slope between the actual temperatures at adjacent moments, and then the degree of temperature change of each chemical reaction stage is determined respectively. According to the degree of temperature change of the chemical reaction stage corresponding to each cluster, the system adjustment parameters can be accurately determined.

[0075] In one embodiment, for each chemical reaction stage corresponding to each cluster, the degree of temperature change of the chemical reaction stage corresponding to the cluster is determined based on the slope and temperature in the chemical reaction stage, including: for each chemical reaction stage corresponding to the cluster, determining the degree of dispersion and central tendency of the slope in the chemical reaction stage, and the temperature span in the chemical reaction stage; determining the degree of temperature change of the chemical reaction stage corresponding to the cluster based on the dispersion, central tendency and temperature span.

[0076] In one embodiment, the degree of dispersion may be variance or standard deviation.

[0077] In one embodiment, the central tendency may be a mean or a median, etc.

[0078] In one embodiment, the temperature span may be determined based on the difference between the maximum temperature value and the minimum temperature value.

[0079] In one embodiment, the temperature variation degree in the chemical reaction stage is positively correlated with the dispersion degree of the slope in the chemical reaction stage, is positively correlated with the central tendency of the slope in the chemical reaction stage, and is positively correlated with the temperature span in the chemical reaction stage.

[0080] In the above embodiment, since the greater the degree of discreteness, the higher the temperature change rate, the higher the temperature increase may be at one moment and lower at another. Therefore, the greater the degree of discreteness of the slope in the chemical reaction stage, the greater the degree of temperature change in the chemical reaction stage. The greater the temperature span in the chemical reaction stage, the greater the degree of temperature change in the chemical reaction stage. The mean of the slope in the chemical reaction stage represents the overall speed of the temperature change of the chemical reaction node. The greater the mean, the faster the overall temperature change. Therefore, based on the degree of discreteness and central tendency of the slope in the chemical reaction stage, as well as the temperature span in the chemical reaction stage, the degree of temperature change in the chemical reaction stage corresponding to the cluster can be accurately determined.

[0081] In one embodiment, determining the temperature variation degree of the chemical reaction stage corresponding to the cluster according to the dispersion degree, the central tendency and the temperature span includes determining the temperature variation degree of the chemical reaction stage corresponding to the cluster according to the product of the dispersion degree, the central tendency and the temperature span.

[0082] In the above embodiment, since the temperature variation degree in the chemical reaction stage is positively correlated with the dispersion degree, central tendency, and temperature span of the slope in the chemical reaction stage, the temperature variation degree in the chemical reaction stage corresponding to the cluster can be accurately determined according to the product of the dispersion degree, central tendency, and temperature span.

[0083] In one embodiment, determining the system adjustment parameter according to the temperature variation degree of the chemical reaction stage corresponding to each cluster includes determining the system adjustment parameter according to the average value of the temperature variation degree of the chemical reaction stage corresponding to each cluster.

[0084] In one embodiment, the system adjustment parameter can be determined according to the following formula: in, Indicates system adjustment parameters. Indicates the The variance of the slopes in each chemical reaction stage. Indicates the The maximum temperature during a chemical reaction stage. Indicates the The minimum temperature during a chemical reaction stage. Indicates the The average of the slopes in each chemical reaction stage. Represents the total number of chemical reaction stages, that is, the total number of clusters. Indicates normalization processing.

[0085] In the above embodiment, the system adjustment parameters can be accurately determined according to the average value of the temperature variation degree of the chemical reaction stage corresponding to each cluster.

[0086] In one embodiment, solving the objective function to obtain the temperature control parameter value at the current moment includes: solving the objective function to obtain the temperature control parameter value at each moment in the control time domain, and extracting the temperature control parameter value at the current moment from the temperature control parameter values ​​at each moment in the control time domain; the control time domain includes the moments that have been experienced.

[0087] Solving the objective function can obtain the temperature control parameter values ​​at each moment in the control time domain , which contains the temperature control parameter value at the current time t , extract the temperature control parameter value at the current time t The data is sent to the actuator for proportional-integral-derivative control. For example, solving the objective function can obtain the temperature control parameter values ​​at each moment from the 11th to the 15th second (the control time domain), and then proportional-integral-derivative control is performed based on the temperature control parameter value at the 15th second (the current moment).

[0088] In the above embodiment, the objective function is solved to obtain the temperature control parameter value at each moment in the control time domain, and the temperature control parameter value at the current moment is extracted from the temperature control parameter value at each moment in the control time domain for proportional integral differential control, which can control the current temperature in real time and accurately.

[0089] See Figure 5 The present invention provides a chemical reactor temperature control system for chemical production, the system including a memory and a processor; the memory is used to store executable program code; the processor is used to call and run the executable program code from the memory to implement the following steps: at each moment, based on the actual temperature of each moment that has been experienced, predict the predicted temperature of the next moment; based on the actual temperature and predicted temperature of each moment that has been experienced, determine the prediction error of each moment in the prediction time domain; the prediction time domain includes the moment that has been experienced and the next moment; based on the change of the actual temperature of each moment that has been experienced, determine the system adjustment parameter; solve the objective function to obtain the temperature control parameter value at the current moment; the objective function includes a prediction error term and a control term; the prediction error term is determined based on the prediction error and the predicted temperature of each moment in the prediction time domain; the control term is determined based on the system adjustment parameter and the temperature control parameter to be solved at each moment in the control time domain; proportional integral differential control is performed based on the temperature control parameter value at the current moment.

[0090] In one embodiment, the prediction error at each moment in the prediction time domain is determined based on the actual temperature and the predicted temperature at each moment, including: determining the prediction error coefficient at each moment in the prediction time domain based on the actual temperature and the predicted temperature at each moment in the prediction time domain; performing function fitting based on the prediction error coefficient at each moment in the prediction time domain to obtain a prediction error function; and determining the prediction error at each moment in the prediction time domain based on the prediction error function.

[0091] In one embodiment, at each moment, the predicted temperature of the next moment is predicted based on the actual temperature of each moment that has been experienced, including: starting from after the initial time period, at each moment, the predicted temperature of the next moment is predicted based on the actual temperature of each moment that has been experienced; based on the actual temperature and the predicted temperature of each moment that has been experienced, the prediction error coefficient of each moment that has been experienced is determined, including: determining the temperature difference between the actual temperature and the predicted temperature of each moment that has been experienced except the initial time period, and the time difference between each moment that has been experienced except the initial time period and the initial time period; based on the temperature difference and time difference of each moment that has been experienced except the initial time period, the prediction error coefficient of each moment that has been experienced except the initial time period is determined.

[0092] In one embodiment, the prediction error coefficient of each moment that has passed except the initial time period is determined based on the temperature difference and time difference of each moment that has passed except the initial time period, including: for each moment that has passed except the initial time period, the prediction error coefficient of the current moment is determined based on the product of the temperature difference at the moment and the time difference.

[0093] In one embodiment, system adjustment parameters are determined based on changes in actual temperature at each moment that has been experienced, including: determining the slope between data points corresponding to the actual temperature at adjacent moments in each moment that has been experienced; clustering each slope to obtain a number of clusters; for each chemical reaction stage corresponding to each cluster, determining the degree of temperature change of the chemical reaction stage corresponding to the cluster based on the slope and temperature in the chemical reaction stage; and determining the system adjustment parameters based on the degree of temperature change of the chemical reaction stage corresponding to each cluster.

[0094] In one embodiment, for each chemical reaction stage corresponding to each cluster, the degree of temperature change of the chemical reaction stage corresponding to the cluster is determined based on the slope and temperature in the chemical reaction stage, including: for each chemical reaction stage corresponding to the cluster, determining the degree of dispersion and central tendency of the slope in the chemical reaction stage, and the temperature span in the chemical reaction stage; determining the degree of temperature change of the chemical reaction stage corresponding to the cluster based on the dispersion, central tendency and temperature span.

[0095] In one embodiment, determining the temperature variation degree of the chemical reaction stage corresponding to the cluster according to the dispersion degree, the central tendency and the temperature span includes determining the temperature variation degree of the chemical reaction stage corresponding to the cluster according to the product of the dispersion degree, the central tendency and the temperature span.

[0096] In one embodiment, determining the system adjustment parameter according to the temperature variation degree of the chemical reaction stage corresponding to each cluster includes determining the system adjustment parameter according to the average value of the temperature variation degree of the chemical reaction stage corresponding to each cluster.

[0097] In one embodiment, solving the objective function to obtain the temperature control parameter value at the current moment includes: solving the objective function to obtain the temperature control parameter value at each moment in the control time domain, and extracting the temperature control parameter value at the current moment from the temperature control parameter values ​​at each moment in the control time domain; the control time domain includes the moments that have been experienced.

[0098] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the scope of the present application, and such modifications and improvements are all within the scope of protection of the present application.

[0100] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0101] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A chemical reactor temperature control method for chemical production, characterized in that: The method comprises: At each moment, the predicted temperature at the next moment is predicted based on the actual temperature at each moment that has passed; Determining the prediction error for each moment in a prediction time domain based on the actual temperature and the predicted temperature at each moment that has passed; the prediction time domain includes the past moment and the next moment; Determine the system adjustment parameters based on the actual temperature changes at each moment; Solving an objective function to obtain a temperature control parameter value at a current moment; the objective function includes a prediction error term and a control term; the prediction error term is determined based on the prediction error and the predicted temperature at each moment in the prediction time domain; the control term is determined based on the system adjustment parameter and the temperature control parameter to be solved at each moment in the control time domain; Proportional-integral-derivative control is performed based on the temperature control parameter value at the current moment.

2. The temperature control method for a chemical reactor for chemical production according to claim 1, characterized in that: The step of determining the prediction error at each moment in the prediction time domain based on the actual temperature and the predicted temperature at each moment that has been experienced includes: Determining a prediction error coefficient for each time instant based on the actual temperature and the predicted temperature at each time instant; Performing function fitting based on the prediction error coefficients at each experienced moment to obtain a prediction error function; The prediction error at each moment in the prediction time domain is determined according to the prediction error function.

3. The temperature control method for a chemical reactor for chemical production according to claim 2, characterized in that: The method of predicting the temperature at the next moment based on the actual temperature at each moment includes: Starting from the initial time period, at each moment, the predicted temperature at the next moment is predicted based on the actual temperature at each moment that has passed; The step of determining the prediction error coefficient at each experienced moment based on the actual temperature and the predicted temperature at each experienced moment includes: Determining the temperature difference between the actual temperature and the predicted temperature at each time point that has passed except the initial time period, and the time difference between each time point that has passed except the initial time period and the initial time period; A prediction error coefficient for each time point that has elapsed except for the initial time period is determined based on the temperature difference and the time difference for each time point that has elapsed except for the initial time period.

4. The temperature control method for a chemical reactor for chemical production according to claim 3, characterized in that: The step of determining a prediction error coefficient at each time point other than the initial time period based on the temperature difference and the time difference at each time point other than the initial time period comprises: For each time instant that has passed except the initial time period, a prediction error coefficient at the current time instant is determined according to the product of the temperature difference at the time instant and the time difference.

5. The temperature control method for chemical reactors in chemical production according to claim 1, characterized in that: The method of determining the system adjustment parameters according to the actual temperature changes at each time point includes: Determine the slope between data points corresponding to actual temperatures at adjacent moments in the past; Clustering the slopes to obtain a number of clusters; For each chemical reaction stage corresponding to the cluster, determining the temperature change degree of the chemical reaction stage corresponding to the cluster according to the slope and temperature in the chemical reaction stage; The system adjustment parameters are determined according to the temperature change degree of the chemical reaction stage corresponding to each cluster.

6. The chemical reactor temperature control method for chemical production according to claim 5, characterized in that: The step of determining, for each chemical reaction stage corresponding to each cluster, a temperature change degree of the chemical reaction stage corresponding to the cluster according to a slope and a temperature in the chemical reaction stage, includes: For each chemical reaction stage corresponding to the cluster, determining the degree of dispersion and central tendency of the slope in the chemical reaction stage, and the temperature span in the chemical reaction stage; The temperature variation degree of the chemical reaction stage corresponding to the cluster is determined according to the dispersion degree, the central tendency and the temperature span.

7. The chemical reactor temperature control method for chemical production according to claim 6, characterized in that: Determining the temperature variation degree of the chemical reaction stage corresponding to the cluster according to the dispersion degree, the central tendency, and the temperature span includes: The temperature variation degree of the chemical reaction stage corresponding to the cluster is determined according to the product of the dispersion degree, the central tendency and the temperature span.

8. The temperature control method for chemical reactors in chemical production according to claim 5, characterized in that: The determining of the system adjustment parameters according to the temperature change degree of the chemical reaction stage corresponding to each cluster includes: The system adjustment parameter is determined according to the average value of the temperature change degree of the chemical reaction stage corresponding to each cluster.

9. The chemical reactor temperature control method for chemical production according to claim 1, characterized in that: Solving the objective function to obtain the temperature control parameter value at the current moment includes: Solving the objective function to obtain the temperature control parameter value at each moment in the control time domain, and extracting the temperature control parameter value at the current moment from the temperature control parameter value at each moment in the control time domain; the control time domain includes the experienced moments.

10. A chemical reactor temperature control system for chemical production, characterized in that: The system includes a memory and a processor; the memory is used to store executable program code; the processor is used to call and run the executable program code from the memory to implement the chemical reactor temperature control method for chemical production according to any one of claims 1 to 9.

Images