Polyurethane raw material preheating control method and system

Abstract

The present application relates to the technical field of industrial production control, and more particularly to a polyurethane raw material preheating control method and system. The method comprises the following steps: by acquiring tank temperature, ambient temperature, batch temperature rise, tank structure and raw material liquid data, a thermal resistance-thermal capacity series-parallel model is constructed to represent the system thermal inertia characteristics, then the anti-initial value drift section, the main temperature rise section and the disturbance absorption section are identified in sections, and the power is backstepped based on the segmented data to dynamically generate the heating power trajectory; the heating trajectory is re-planned in combination with real-time change data, thereby realizing accurate modeling and adaptive control of the temperature rise process, and effectively improving the temperature control accuracy and energy efficiency response capability of the preheating system under different raw material batches and different working conditions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial production control, and in particular to a polyurethane raw material preheating control method and system. BACKGROUND

[0002] In the production process of polyurethane materials, preheating treatment of raw materials is a key link to ensure the stability of subsequent mixing, reaction and foaming quality. The traditional preheating control method mainly relies on fixed temperature set point combined with simple PID feedback control. When facing the differences in thermal inertia of different batches of materials, environmental temperature fluctuations and nonlinear heating characteristics, problems such as overshoot, reaction delay or energy consumption deviation often occur, and the temperature control accuracy and energy efficiency regulation cannot be considered.

[0003] In the scene of complex tank structure and large difference in specific heat of raw materials, the traditional strategy is difficult to accurately describe the thermal response characteristics of the system, resulting in mismatch between power setting and temperature trajectory, reducing the consistency of production rhythm and the thermal efficiency of equipment. At the same time, once the raw material is replenished, the environment temperature drops or the actuator performance fluctuates, the system is difficult to adjust the preheating strategy, and there is a certain robustness bottleneck. SUMMARY

[0004] To solve the above technical problems, the present application provides a polyurethane raw material preheating control method and system to solve at least one of the above technical problems.

[0005] The present application provides a polyurethane raw material preheating control method, comprising the following steps:

[0006] Step S1: Obtain tank temperature data, environmental temperature data, batch temperature rise data, tank data and raw material liquid data; construct a thermal resistance-thermal capacity series-parallel model according to the tank temperature data, the environmental temperature data, the batch temperature rise data, the tank data and the raw material liquid data, and obtain the thermal resistance-thermal capacity series-parallel model;

[0007] Step S2: Anti-initial value drift section processing is performed on the thermal resistance-thermal capacity series-parallel model to obtain anti-initial value drift section data; main heating section processing is performed on the thermal resistance-thermal capacity series-parallel model to obtain main heating section data; and disturbance absorption section processing is performed on the thermal resistance-thermal capacity series-parallel model to obtain disturbance absorption section data;

[0008] Step S3: According to the anti-initial value drift section data, the main heating section data and the disturbance absorption section data, the power is backstepped to obtain heating power data;

[0009] Step S4: Obtain real-time change data, and change the trajectory according to the heating power data and the real-time change data to obtain heating change data.

[0010] In the application, by constructing a thermal resistance-thermal capacity series-parallel model of fusing tank temperature, ambient temperature, historical batch temperature rise data and physical structure parameters, the thermal inertia characteristics in the raw material preheating process can be accurately represented, and the fitting accuracy of temperature rise response is improved. The initial temperature uncertainty and system hysteresis effect are effectively compensated by adopting anti-initial value drift section processing, and the stability of the model in the preheating start-up stage is enhanced; through main temperature rise segment modeling and disturbance absorption segment processing, fine control of the target temperature rise trajectory and dynamic adaptation ability to the intermediate disturbance factors are realized. Based on the power backstepping of the segmented trajectory, the heating power data matching the physical process can be obtained, and the cooperative optimization of energy consumption control and temperature rise time limit is ensured; combined with real-time change data for trajectory change, the rapid response ability of the system to external disturbance and production dynamic adjustment is improved, so that the high-precision and high-robustness intelligent control of the polyurethane raw material preheating process is realized.

[0011] Preferably, the thermal resistance-thermal capacity series-parallel model is constructed as follows:

[0012] According to the tank data and the raw material liquid data, a thermal capacity-thermal resistance network model is constructed to obtain the thermal capacity-thermal resistance network model;

[0013] According to the tank temperature data, the raw material liquid data and the liquid temperature data, the thermal capacity-thermal resistance network model data is processed to obtain the state space model data;

[0014] The tank temperature data, the ambient temperature data and the batch temperature rise data are labeled to the state space model data to obtain the thermal resistance-thermal capacity series-parallel model.

[0015] In the application, by constructing the thermal capacity-thermal resistance network model, the tank structure parameters and the thermal physical property data of the raw material liquid can be modeled in series-parallel connection, the structural analysis of heat conduction path and heat storage capacity is realized, and the physical expression ability of heat exchange process is improved. The state space model processing is combined with the tank temperature and liquid temperature data, so that the model has dynamic response characteristics and time sequence prediction ability, which is helpful to capture the evolution law of each node state in the temperature rise process. Through batch temperature rise data and ambient temperature data, the state space model is labeled and trained, which not only enhances the adaptability of the model to different running scenes, but also improves the identification accuracy of thermal inertia change trend, so as to establish a thermal resistance-thermal capacity series-parallel model with physical constraint, dynamic modeling ability and scene adaptation ability, which provides a high-quality modeling basis for segmented control and power planning.

[0016] Preferably, the anti-initial value drift section processing is as follows:

[0017] According to the thermal resistance-thermal capacity series-parallel model, initial drift data is obtained;

[0018] The initial drift data is subjected to temperature rise rate deviation activation determination to obtain drift activation data;

[0019] The drift activation data is subjected to anti-initial value drift section processing to obtain preliminary section data;

[0020] The preliminary section data is subjected to slow rise section dynamic feedback correction to obtain anti-initial value drift section data.

[0021] In the present application, the initial drift data is subjected to temperature rise rate deviation activation determination to obtain drift activation data, which can identify the response drift phenomenon caused by original temperature measurement deviation, tank thermal inertia lag or unstable heat exchange in the starting stage, and ensure that the model can reflect the real heating state. Through temperature rise rate deviation activation determination, the section with drift intensity exceeding the normal change range can be accurately identified, and a differentiated compensation strategy is triggered. Through anti-initial value drift section processing, the temperature rise trajectory of the activated interval is reconstructed, the cumulative effect of the early deviation of the model is eliminated, and the wrong trajectory is avoided for power back calculation. Through slow rise section dynamic feedback correction, the model output tends to be stable transition, effectively improving the response stability and reliability of power planning in the early stage of the whole preheating process, and improving the accuracy and system robustness of the whole temperature control process.

[0022] Preferably, the temperature rise rate deviation activation determination is specifically:

[0023] The initial drift data is subjected to rate deviation curve construction to obtain rate deviation curve data;

[0024] Slope activation data is obtained by judging the slope according to the rate deviation curve data;

[0025] Drift activation data is obtained by triggering the threshold surface according to the activation data.

[0026] In the present application, the initial drift data is subjected to rate deviation curve construction, which can dynamically extract the actual rate change trend in the temperature rise process, and quantize the difference with the theoretical temperature rise trajectory, so as to accurately represent the response deviation behavior in the initial stage. Based on the rate deviation curve, the system can identify abnormal steep points or slow and flat sections in the rate change in real time through slope activation judgment, and then realize early perception of potential drift trend. Compared with the traditional fixed threshold judgment method, the method can construct dynamic activation conditions by combining multiple features such as slope amplitude and duration, realize adaptive triggering of temperature rise deviation mode, and effectively avoid misjudgment or omission. The above strategy enhances the sensitivity and discrimination accuracy of the system to the temperature control deviation in the starting stage, provides high reliability and high resolution input basis for drift section modeling and feedback correction, and improves the adaptability of the whole temperature control system to initial state abnormalities and the stability of the temperature rise trajectory.

[0027] Preferably, the anti-initial value drift section processing is specifically:

[0028] The drift activation data and the preset prior thermal inertia data are differentially compared to obtain drift error data;

[0029] The drift error data is subjected to disturbance intensity integral calculation to obtain disturbance intensity integral data;

[0030] The drift activation data is divided into drift risk intervals according to the disturbance intensity integral data to obtain drift risk interval data;

[0031] The drift activation data is subjected to anti-drift correction according to the drift risk interval data to obtain preliminary section data.

[0032] In the present application, by differentially comparing the drift activation data with the preset prior thermal inertia data, a personalized drift error curve conforming to the current device structure characteristics and the thermal properties of raw materials can be constructed to realize accurate quantification of the initial stage deviation amplitude and direction. Through disturbance intensity integral calculation, not only the instantaneous extreme value of the drift error is considered, but also the cumulative influence of its duration on system stability is integrated to obtain a more comprehensive drift strength evaluation index. Through drift risk interval division, different levels of deviation states can be dynamically segmented and managed, and the key sections that need to be corrected can be clearly defined to avoid misadjustment of the normal response interval. Based on the risk interval, anti-drift correction is performed to ensure that the temperature rise trajectory has smoothness and consistency in the starting stage, providing reliable initial data support for energy efficiency control and trajectory planning in the heating section. The present application effectively improves the adaptability of the model response under non-ideal initial conditions and enhances the robustness and control accuracy of the temperature control system under the background of thermal inertia.

[0033] Preferably, the main heating section processing is specifically:

[0034] The thermal resistance-thermal capacity series-parallel model is subjected to main heating section boundary determination to obtain main heating section boundary data;

[0035] The thermal resistance-thermal capacity series-parallel model is subjected to shortest heating time backstepping according to the main heating section boundary data to obtain shortest heating time data;

[0036] The thermal resistance-thermal capacity series-parallel model is subjected to constant temperature rise rate slope calculation according to the shortest heating time data to obtain constant temperature rise rate slope data;

[0037] The main heating section function is generated according to the constant temperature rise rate slope to obtain main heating section data.

[0038] In the present application, the boundary of the main heating section is determined by the thermal resistance-thermal capacity series-parallel model, which can determine the start and end conditions of the system entering the steady state heating interval, avoid the interference of initial fluctuation and end disturbance on the temperature control strategy, and improve the stage accuracy of trajectory planning. The shortest heating time is deduced by the boundary condition, which not only realizes the heating limit efficiency evaluation combined with the thermal inertia of the tank and the heat capacity of the raw material, but also provides time constraints for power distribution. Based on the shortest heating time, the constant temperature rise rate slope is calculated, which can ensure the dynamic balance of power output and thermal response, and avoid the risk of thermal shock or overheating caused by sudden change of power. The main heating section function is generated by the slope parameter driving, which realizes the conversion of the temperature control trajectory from physical modeling to function expression, and makes the heating process have the characteristics of adjustable, predictable and schedulable. The response efficiency and energy consumption control ability of the system to the key heating stage are enhanced, which provides strong model support for the efficient and stable preheating of polyurethane raw materials within a specific time window.

[0039] Preferably, the disturbance absorption section processing is specifically:

[0040] The absorption section data is obtained by performing absorption section triggering judgment on the thermal resistance-thermal capacity series-parallel model;

[0041] The disturbance absorption section time window data is obtained by determining the disturbance absorption section time window according to the absorption section data;

[0042] The disturbance absorption section data is obtained by performing slow convergence structure processing on the thermal resistance-thermal capacity series-parallel model according to the disturbance absorption section time window data.

[0043] In the present application, the absorption section triggering judgment is performed on the thermal resistance-thermal capacity series-parallel model, which can identify the response deviation opportunity caused by environmental fluctuations, batch differences or external power disturbance in the heating process, and realize the dynamic perception and boundary judgment of the disturbance event. The time window of the disturbance absorption section is determined by the absorption section data, which not only considers the start and end time points of the disturbance, but also integrates the hysteresis and buffering capacity of the system thermal response, so that the disturbance processing process has time adaptability and response redundancy. The slow convergence structure processing is used to perform stage regulation on the thermal capacity-thermal resistance coupled channel, which can effectively suppress the slope mutation or trajectory nonlinear fluctuation caused by disturbance, and make the temperature rise process gradually return to the preset trajectory. The recovery ability of the model to non-steady state disturbance is strengthened, and the robustness and fault tolerance of the trajectory planning under real working conditions are improved, which is helpful to realize the rapid absorption of local disturbance and the continuous maintenance of global stability in the preheating process of polyurethane raw materials.

[0044] Preferably, step S3 is specifically:

[0045] The trajectory splicing data is obtained by splicing the anti-initial value drift section data, the main heating section data and the disturbance absorption section data;

[0046] The temperature derivative data is obtained by performing temperature derivative calculation on the trajectory splicing data.

[0047] The heating power data is obtained by performing power backstepping according to the temperature derivative data and the thermal resistance-thermal capacity series-parallel model.

[0048] In the present application, by splicing the data of the anti-initial value drift section, the main temperature rise section and the disturbance absorption section, a complete, smooth and physically consistent temperature rise target trajectory can be constructed on the basis of ensuring the continuity of temperature control in each stage, effectively avoiding the power fluctuation or thermal inertia mismatch problem caused by the disconnection between sections. The first-order dynamic extraction of the spliced trajectory is performed by using the temperature derivative calculation method, which can represent the temperature change rate to provide a time series signal reflecting the intensity of the temperature rise demand. The power backstepping is performed in combination with the derivative information and the thermal resistance-thermal capacity series-parallel model, which not only realizes the dynamic mapping from the expected temperature behavior to the physical heating input, but also adjusts the power output curve according to the actual thermal response capacity of the system, ensuring that the temperature rise process meets the efficiency requirements and avoids the risk of energy overload and thermal shock. The present application improves the interpretability and execution consistency of the system for the temperature rise behavior, enhances the matching accuracy between the power planning and the actual thermal inertia, and provides a solid support for high-performance temperature control strategies.

[0049] Preferably, step S4 is specifically:

[0050] Obtaining real-time change data;

[0051] Performing trajectory change trigger condition judgment according to the real-time change data to obtain trajectory change data;

[0052] Performing trajectory remaining section cutting according to the trajectory change data and the heating power data to obtain trajectory remaining section data;

[0053] Performing change window generation according to the trajectory remaining section data to obtain change window data;

[0054] Performing trajectory function re-planning according to the change window data to obtain heating change data.

[0055] In the present application, by acquiring real-time change data, the dynamic demand changes caused by environmental condition fluctuations, raw material state changes or upper control strategy adjustments can be perceived in a timely manner, ensuring that the system has the ability to respond to external inputs in real time. Through trajectory change triggering condition judgment, it can be accurately determined whether the trajectory needs to be reconstructed based on temperature deviation, rate jump or energy consumption threshold, etc., avoiding frequent intervention in the stable operation of the system due to minor disturbances. The trajectory remaining segment trimming mechanism can perform phased reorganization on the current heating process, and only adjust the unfinished part, preserving the continuity and stability of the existing temperature rise data. A change window is generated, and combined with the current power state and thermal response model, the trajectory function is re-planned, which can make the system quickly switch to a new temperature rise target trajectory, ensuring the dynamic consistency between energy consumption control, temperature rise rhythm and target scheduling. The adaptability and robustness of the temperature rise process to non-ideal working conditions are enhanced, and the intelligent control capability and practicality of the whole temperature control system in the application scene are improved.

[0056] Preferably, the present application also provides a polyurethane raw material preheating control system for executing the polyurethane raw material preheating control method as described above, which comprises:

[0057] A thermal resistance-thermal capacity modeling module is configured to acquire tank temperature data, environment temperature data, batch temperature rise data, tank data and raw material liquid data; construct a thermal resistance-thermal capacity series-parallel model according to the tank temperature data, the environment temperature data, the batch temperature rise data, the tank data and the raw material liquid data; and obtain the thermal resistance-thermal capacity series-parallel model.

[0058] A segmented response identification and processing module is configured to perform anti-initial value drift section processing on the thermal resistance-thermal capacity series-parallel model to obtain anti-initial value drift section data; perform main temperature rise section processing on the thermal resistance-thermal capacity series-parallel model to obtain main temperature rise section data; and perform disturbance absorption section processing on the thermal resistance-thermal capacity series-parallel model to obtain disturbance absorption section data.

[0059] A reverse power solving module is configured to perform power backstepping according to the anti-initial value drift section data, the main temperature rise section data and the disturbance absorption section data to obtain heating power data.

[0060] A heating trajectory change control module is configured to acquire real-time change data, and perform trajectory change according to the heating power data and the real-time change data to obtain heating change data.

[0061] The beneficial effects of the present application are that: by constructing a thermal resistance-thermal capacity series-parallel model, multi-source fusion modeling of the tank structure parameters, raw material thermal properties, environmental temperature and historical batch temperature rise behavior is realized, which can accurately depict the thermal inertia response characteristics of each stage in the raw material preheating process, and provides a high fitting degree physical basis for dynamic temperature control. On this basis, the anti-initial value drift processing, main heating section modeling and disturbance absorption mechanism are introduced in stages, which not only can effectively suppress the influence of temperature drift in the starting stage on the control accuracy of the subsequent stage, but also through slope back propagation and boundary reconstruction, the stability and timeliness of the heating process are ensured. At the same time, the disturbance absorption section processing can realize slow convergence adjustment under the condition that the tank is disturbed or the raw material properties fluctuate, which enhances the adaptability of the system to non-ideal working conditions. By uniformly splicing the above trajectory sections and calculating the temperature derivative, combined with the heat model, the high-precision power output sequence conforming to the thermal dynamic characteristics of the system can be obtained, which ensures the reasonable distribution of energy consumption. Through real-time trajectory change, the control target reconstruction and power adjustment under the sudden scene are supported, the robustness and intelligent response ability of the temperature control system are improved, and a closed-loop optimization control system is formed. BRIEF DESCRIPTION OF DRAWINGS

[0062] Other characteristics, objects and advantages of the present application will become more apparent from the following detailed description of non-restrictive embodiments made with reference to the attached drawings:

[0063] Figure 1 A step flow chart of a polyurethane raw material preheating control method of an embodiment is shown;

[0064] Figure 2 A step flow chart of a thermal resistance-thermal capacity series-parallel model construction method of an embodiment is shown;

[0065] Figure 3 A step flow chart of an anti-initial value drift section processing method of an embodiment is shown;

[0066] Figure 4 A step flow chart of a heating power back propagation method of an embodiment is shown;

[0067] Figure 5 A step flow chart of a heating trajectory change method of an embodiment is shown. DETAILED DESCRIPTION

[0068] The technical method of the present application will be described in detail below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0069] Furthermore, the drawings are merely schematic and may not be drawn to scale. A software form can be employed to implement the functional entities or the functional entities can be implemented in one or more hardware modules or integrated circuits, or the functional entities can be implemented in different network and / or processor methods and / or microcontroller methods.

[0070] It should be understood that, although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of the example embodiments. The term "and / or" as used herein encompasses any and all combinations of one or more of the associated associated items.

[0071] Referring to Figures 1 to 5 The application provides a polyurethane raw material preheating control method, comprising the following steps:

[0072] Step S1: obtaining tank temperature data, environment temperature data, batch temperature rise data, tank data and raw material liquid data; constructing a thermal resistance-thermal capacity series-parallel model according to the tank temperature data, the environment temperature data, the batch temperature rise data, the tank data and the raw material liquid data, to obtain the thermal resistance-thermal capacity series-parallel model;

[0073] Specifically, the system collects multi-source thermophysical data related to the polyurethane raw material heating process, including data of tank body temperature change over time; ambient temperature data; initial temperature data of raw material liquid; historical temperature rise curve data of multiple batches; tank structure parameters, including tank wall thickness, thermal conductivity, tank heat capacity; heat capacity parameters and thermal conductivity of raw material liquid, etc. According to the heat transfer path in the tank heating process, the tank wall layer, the raw material liquid layer and the outer layer in contact with the air are layered modeled according to the heat flux direction, and an equivalent thermal resistance and heat capacity series-parallel network structure is established, including equivalent thermal resistance calculation (i.e. the product of layer thickness divided by thermal conductivity and heat exchange area), equivalent heat capacity (i.e. mass multiplied by specific heat capacity), which constitutes a heat capacity-thermal resistance series-parallel combination network in space, and truly reflects the thermal inertia characteristics of the tank and the raw material liquid in the dynamic heating process. Based on the heat balance equation, the above thermal network structure is mathematically modeled and converted into a standard state space equation form: the rate of change of each temperature node (i.e. the derivative of temperature with respect to time) is determined by the system matrix of the heat capacity-thermal resistance structure and the heating power input. Formally, it can be expressed as temperature change rate = system state matrix x current temperature state + control matrix x heating power input, where the system state matrix is determined by the topological structure of the thermal resistance-thermal capacity network; the heating power input represents the output behavior of the actual heater. In order to make the thermal resistance-thermal capacity state space model constructed have good actual representativeness, the historical temperature rise response curves of multiple batches are used as supervision data to fit and train the model or calibrate the parameters. By minimizing the mean square error between the model output temperature and the historical measured temperature rise curve, the model is adaptively adjusted to restore the thermal response characteristics of the tank-raw material system.

[0074] Step S2: Anti-initial value drift section processing is performed on the thermal resistance-thermal capacity series-parallel model to obtain anti-initial value drift section data; main temperature rise section processing is performed on the thermal resistance-thermal capacity series-parallel model to obtain main temperature rise section data; disturbance absorption section processing is performed on the thermal resistance-thermal capacity series-parallel model to obtain disturbance absorption section data;

[0075] Specifically, the system compares the slope change of the actual temperature rise curve (based on actual measured temperature) and the model predicted curve (obtained by multiple regression fitting of the curve in the model), calculates the real-time temperature rise rate deviation. When the absolute value of the difference between the actual temperature rise rate and the model predicted rate per unit time exceeds the preset threshold, it is considered that the initial value drift occurs. The system integrates the rate deviation in the initial period of temperature rise to obtain the disturbance intensity integral. According to the integral value, the system divides the drift level into high, medium and low three levels through clustering calculation or preset threshold judgment, and determines the corresponding response correction strength. In order to prevent the system from being dynamically deviated due to the too fast heating in the initial heating process, the system applies a slope limiting strategy to the initial heating trajectory, that is, compares the heating slope calculated by the model with the set maximum starting slope, and takes the smaller one as the actual heating speed to obtain the anti-initial value drift section data. This processing helps to prevent the thermal response deviation caused by large initial temperature difference or model error in the starting stage, and ensures smooth transition of the system in the initial heating period. By limiting the actual heating slope, the over-responding or overshooting phenomenon of the temperature control system is avoided.

[0076] After identifying the end point of the anti-drift section (i.e. the rate returns to stable), the system sets the starting point of the main heating section; and the end point is set as 95% of the target temperature of the current batch, leaving a control space for the disturbance absorption section. According to the current temperature and the target temperature, and the maximum allowed heating slope, the theoretical shortest heating time of this stage is estimated: shortest heating time = (target temperature - current temperature) / maximum allowed heating slope. The main heating section adopts constant slope for heating control, and the trajectory is expressed as current temperature = current temperature + constant heating slope × (current time - main heating section starting time). The control of the main heating section aims to ensure the linear controllability of the heating process, so that the system maintains an efficient and stable energy input mode in the main heating stage, while leaving a buffer space for subsequent disturbance absorption, improving the overall temperature control accuracy and robustness.

[0077] The system detects the temperature curvature anomaly caused by external disturbances (such as liquid level fluctuation, external feed, sudden change of thermal load, etc.) in real time during the main heating section. If the second derivative of the heating curve (i.e. the acceleration of temperature change) exceeds the set preset threshold, the disturbance duration interval is set and the disturbance absorption is triggered, including absorption time window setting and slow convergence structure processing. According to the disturbance duration (obtained based on the disturbance duration interval) and the system thermal inertia constant (representing the thermal response delay time of the system, which can be estimated by state space model or step response), the system sets the minimum maintenance time window of the disturbance absorption section: wherein is the minimum maintenance time window of the disturbance absorption section, is the maximum function, is the actual detected disturbance duration, is an empirical adjustment coefficient, ensuring that the absorption window is sufficient to cover the disturbance process, is the system thermal inertia constant, i.e., the product of the system equivalent thermal resistance and the system equivalent thermal capacity. In the disturbance absorption segment, the system performs curvature suppression on the original temperature rise slope, and uses a smoothing function to adjust the current slope to be consistent with the target slope of the main temperature rise segment, thereby achieving flexible regression of the temperature rise trajectory, so as to prevent control oscillation or system overshoot. The disturbance absorption segment is used to cope with external disturbances or temperature fluctuations that occur during the temperature rise process, so as to avoid power rebound or control system oscillation caused by severe disturbance. The system transitions smoothly back to the main trajectory through a slow convergence mechanism, thereby enhancing the robustness and temperature control flexibility of the system.

[0078] Step S3: According to the anti-initial value drift segment data, the main temperature rise segment data and the disturbance absorption segment data, the power is backstepped to obtain the heating power data;

[0079] Specifically, the system time-aligns and logically combines the aforementioned three types of characteristic temperature rise segments (including replacing the main temperature rise segment data with the disturbance absorption segment data), and constructs a target temperature rise trajectory function. The temperature change data of the anti-initial value drift segment, the main temperature rise segment and the disturbance absorption segment are spliced according to the time sequence to form a continuous complete temperature rise function. After obtaining the complete temperature trajectory, the system uses numerical differentiation methods (such as forward difference, central difference, etc.) or analytical derivative methods to solve the first-order derivative of the temperature curve, and obtains the real-time temperature rise rate. The system uses the following energy balance equation to backstep the power according to the constructed thermal resistance-thermal capacity series-parallel model and the temperature derivative data: , wherein represents the heating power at time t, represents the system equivalent thermal capacity, including the total thermal capacity value of each thermal capacity unit after combining the tank material, raw material liquid, etc., represents the actual temperature of the system at time, represents the ambient temperature at time, represents each thermal resistance unit involved in the heat conduction process from the system to the outside. The backstepping model includes two core thermal terms: one is the thermal capacity term , which represents the internal energy input required for the system temperature rise; the other is the thermal loss term , which represents the power loss compensation due to heat conduction. The sum of the two is the actual heating power value required for the system to rise to the target trajectory.

[0080] Step S4: Obtain real-time change data, and perform trajectory change according to the heating power data and the real-time change data to obtain heating change data.

[0081] Specifically, the system continuously collects and monitors real-time change data of the environment and the device during the heating process, and adjusts and re-plans the current heating trajectory when the preset change triggering condition is met, so as to realize the stability and target orientation of the system operation. The system continuously monitors various dynamic disturbances or state change events during operation, including but not limited to external heating source interruption; liquid replenishment of the raw material tank body causes liquid heat capacity mutation; sudden drop or rise of ambient temperature; actual output power of the heater is inconsistent with the planned power, etc. These events constitute real-time change data for determining whether the original planned trajectory needs to be adjusted. The system sets multi-dimensional triggering threshold conditions to determine whether the current temperature rising process deviates from the established trajectory. The triggering conditions include that the absolute deviation of the current measured temperature and the target temperature exceeds the set tolerance; the actual output power of the current heater is inconsistent with the original planned power. When any condition is met, the system triggers the trajectory change process. Once the trajectory change is triggered, the system deletes all the planned segments after the current time in the original temperature trajectory to avoid subsequent control relying on outdated prediction. The system sets an update window interval wherein is the time length dynamically set according to the system response inertia, disturbance duration or controller adjustment capability. The window serves as the definition interval of the new trajectory function. Within the newly generated change window, the system constructs a new temperature control trajectory, which meets the following two requirements: the first end of the trajectory needs to smoothly connect with the current temperature state to avoid derivative mutation; the end of the trajectory should lead to a new target temperature or modify the planned point under the target curve. The trajectory function uses a smooth interpolation function (such as cubic spline), a time optimal control function or an optimization solution based on model predictive control. The present application can dynamically respond and reconstruct the heating trajectory when a sudden disturbance or input deviation occurs in the temperature control process, avoiding the continuous execution of unreasonable historical trajectory, effectively improving the response flexibility, temperature rising accuracy and energy consumption control ability of the system, and is particularly suitable for the production process of polyurethane raw materials which are highly sensitive to temperature changes.

[0082] Preferably, wherein the thermal resistance-thermal capacity series-parallel model is constructed as follows:

[0083] Step S11: Constructing a thermal capacity-thermal resistance network model according to the tank body data and the raw material liquid data;

[0084] Specifically, the tank data includes tank wall thickness, material thermal conductivity, heat exchange area (effective heat transfer area of the part of the tank outer wall in contact with the heater), and tank equivalent heat capacity (tank mass multiplied by tank material specific heat capacity), and the raw material liquid data includes liquid density, liquid volume, liquid specific heat capacity, liquid thermal conductivity, liquid layer height, and heat exchange area of the liquid layer in contact with the tank wall. The system performs equivalent thermal resistance calculation, including tank wall layer thermal resistance calculation (wall thickness divided by the product of material thermal conductivity and heat exchange area) and liquid layer thermal resistance calculation (liquid height divided by the product of liquid thermal conductivity and heat exchange area of the liquid layer in contact with the tank wall). The system performs liquid heat capacity calculation: liquid density multiplied by liquid volume multiplied by liquid specific heat capacity. According to the physical topology of the heat transfer path, a series-parallel structure heat capacity-thermal resistance network is constructed, including a series structure: the path from the heater input end to the liquid center is heater → tank wall layer thermal resistance → tank equivalent heat capacity → liquid layer thermal resistance → liquid heat capacity. If the raw material liquid is layered (for example, there is a temperature difference between different material components), the liquid layer can be divided into multiple equivalent heat capacity nodes to form a parallel structure to reflect the uneven heat distribution inside the liquid. The formed heat capacity-thermal resistance network model can be regarded as a graph structure composed of multiple heat capacity nodes and thermal resistance channels.

[0085] Step S12: State space model processing is performed on the heat capacity-thermal resistance network model data according to the tank temperature data and the liquid temperature data in the raw material liquid data, to obtain state space model data;

[0086] Specifically, based on the heat capacity-thermal resistance network model, the state space dynamic response model of the system is constructed by collecting tank temperature and liquid temperature data. The raw material liquid data also includes liquid temperature data, specifically the system collects the following time series temperature data: tank outer wall temperature data: obtained by the tank sensor in real time; liquid temperature data: single-point thermometer measurement value or multi-measurement point average value. The state variable of the system is set as a temperature node set. According to the heat conduction law and the energy conservation principle, each layer node in the heat capacity-thermal resistance network model is constructed as a differential equation group of temperature dynamic change wherein is a state variable vector, is a time variable, is a system matrix (state transition matrix), is a control input matrix, For input power (heating power). The system dynamics can be modeled as state equations in the following form: the system state rate of change is determined by the system matrix, the control input matrix, the current state (i.e. state variable) and the heating input power; the system matrix is determined by the thermal resistance and the thermal capacity parameters, which is specifically embodied as the first row: the derivative of the tank layer temperature is affected by its own thermal capacity and thermal resistance, and is affected by the liquid layer back heat flow; the second row: the derivative of the liquid layer temperature is affected by the coupling of the tank layer heat transfer and the liquid itself heat dissipation; the control input matrix represents that the heating input only acts on the tank heat capacity node. Thus, a complete continuous-time state space model is established to describe the thermal inertia dynamics of the entire heat transfer process. The state space model data includes the system matrix, the control input matrix and the state variable.

[0087] ;

[0088] wherein is the system matrix, is the thermal resistance between the tank and the liquid, is the tank thermal capacity, is the tank thermal capacity, is the liquid thermal capacity, is the control input matrix.

[0089] Step S13: labeling the tank temperature data, the environment temperature data and the batch temperature rise data to the state space model data to obtain a thermal resistance-thermal capacity series-parallel model.

[0090] Specifically, the state space model is supervised fitted and labeled reinforced with historical batch temperature control data, and a thermal resistance-thermal capacity series-parallel model that can be used for temperature rise segment identification and power planning is output. The system collects temperature control data of multiple historical batches as the target output for supervised fitting, including but not limited to the corresponding tank temperature change sequence of each batch; the ambient temperature change sequence within the corresponding time; the actual heating power input data during the temperature rise process of the batch. The data of each batch is in the form of a three-tuple as follows: tank temperature sequence, ambient temperature sequence, power input sequence. The above sample set is input into the state space model to supervise the temperature prediction ability and parameter fitting accuracy of the model. The system sets the fitting loss function: for the tank temperature sequence predicted by the model, the difference integral is performed with the corresponding measured temperature; the total loss function is formed by summing all batch data, and the index is the weighted average mean square error; the goal is to minimize the loss. For the system parameter matrix in the state space model (such as thermal resistance, thermal capacity, coupling coefficient, etc.), the following optimization methods are used: least squares fitting method; intelligent search optimization strategies such as genetic algorithm; posterior distribution updating methods such as Bayesian estimation. In the above process, the parameters are reasonably fine-tuned under the premise of maintaining thermal physical constraints, so that the model has stable prediction ability under different environments and initial states of the tank. After multiple batch supervised learning and error control, if the model output shows good temperature rise trajectory prediction ability on all samples, the current state space model can be upgraded to a thermal resistance-thermal capacity series-parallel model with engineering adaptability.

[0091] Preferably, the anti-initial drift section processing is specifically:

[0092] Step S21: initial drift data is obtained by initial drift discrimination according to the thermal resistance-thermal capacity series-parallel model;

[0093] Specifically, the thermal resistance-thermal capacity series-parallel model constructed is used to predict the temperature of the target tank, and a model output temperature sequence is obtained. Combined with the real-time measured temperature data of the tank, a temperature deviation function between the two is calculated. The system sets a fixed detection time interval to capture the deviation trend in the initial heating period. Preferably, the detection interval is the first 2 minutes after the system heating starts. In the above detection interval, the average value of the deviation function is calculated by integral mean calculation, that is, the average value of the deviation function in the set time period, which reflects the systematic error between the model and the actual response. If the absolute value of the calculated average deviation exceeds the preset deviation tolerance threshold, the system determines that there is an initial drift problem in the current batch. Once the drift is activated and determined to be true, the system automatically extracts the data of the time period with larger deviation as the input data for subsequent temperature rise trajectory optimization and anti-drift control, and marks it as initial drift data.

[0094] Step S22: drift activation data is obtained by temperature rise rate deviation activation discrimination on the initial drift data;

[0095] Specifically, based on the difference between the measured temperature change rate and the model predicted temperature change rate, a rate deviation curve is constructed and an activation threshold is determined. Within the identified initial drift time section, the system calculates the instantaneous change rate of the measured temperature (i.e. the temperature rise rate) respectively; the instantaneous change rate of the model predicted temperature. The system constructs a rate deviation function of the difference between the two. The system analyzes the above rate deviation curve, sets a rate activation threshold and a cumulative time threshold, i.e. if the cumulative value of the time when the rate deviation exceeds the threshold value exceeds the preset threshold time, it means that there is a significant rate drift response in the current stage. Once the above activation condition is met, the system determines that the section is a drift activation section, and outputs the data of the corresponding time period as drift activation data.

[0096] Step S23: Anti-initial value drift section processing is performed on the drift activation data to obtain preliminary section data;

[0097] Specifically, the system constructs a linear temperature rise trajectory with a controlled slope based on the starting time and starting temperature of the drift activation time period to replace the section with large rate fluctuations in the original trajectory. The control function can be expressed as: wherein is the corrected trajectory temperature function, is the starting temperature, is the temperature rise control slope, satisfying the constraint wherein is the upper limit of the system nominal temperature rise rate, used to limit the temperature rise speed and prevent thermal jump phenomenon, is the current time, is the starting temperature of the trajectory. The system replaces the drift activation time period in the original temperature trajectory according to the above control function to generate preliminary trajectory sections for subsequent power backstepping. This process is equivalent to dynamically correcting the original temperature rise path with obvious initial disturbance to achieve smooth transition of temperature rise. When the heat capacity of the raw material liquid fluctuates, the system selects the second derivative of temperature change as a constraint to suppress the curvature mutation of the temperature rise trajectory and prevent nonlinear acceleration effect caused by sudden change in heat capacity. The limit is expressed as: wherein is the temperature function, is the time variable, is a preset smoothing threshold parameter. In this way, the system can effectively correct the abnormal temperature rise trend caused by initial value disturbance and form a stable, controllable and structurally stable preliminary section temperature rise data.

[0098] Step S24: Dynamic feedback correction of the preliminary section data is performed to obtain anti-initial value drift section data.

[0099] Specifically, the system continuously monitors the actual temperature rise rate at the current time and compares it with the temperature rise rate of the slow start trajectory generated by the model. The actual temperature rate is measured by the sensor; the expected temperature rate is obtained by preliminary slow rise trajectory calculation. If it is found that the actual rate is significantly lower than the expected rate, indicating that there is a delay in the heating system or environmental heat dissipation effect, the system will automatically adjust the target temperature rise rate to avoid the risk of thermal hysteresis or overshoot caused by excessive temperature control instructions. The system sets a time decay for the current temperature rise slope parameter, that is, dynamically adjusts the original constant slope parameter in the temperature trajectory to a time-decaying expression: wherein is the adjusted constant slope parameter, is the initial set slope, is the exponential function, is the slope decay factor for controlling the decay speed, is the current time, is the starting time of the slow rise section. On the basis of applying the dynamic slope correction mechanism, the system uses a sliding time window method to locally smooth and continuously optimize the current temperature trajectory. This processing method combines the temperature change trend of the previous period and the connection requirements of the target trajectory, and corrects the average slope of the trajectory within the window to avoid sudden changes or discontinuities in the curve. The system outputs the anti-initial value drift section data after dynamic feedback optimization.

[0100] Preferably, wherein the temperature rise rate deviation activation judgment is specifically:

[0101] Constructing a rate offset curve on the initial drift data to obtain rate offset curve data;

[0102] Specifically, the system extracts the following temperature sequences from the initial drift period: the measured tank temperature sequence, that is, the real-time monitored temperature change data; the model predicted temperature sequence, that is, the temperature rise trend data output by the thermal resistance-thermal capacity model. The system estimates the temperature rise first derivative using a sliding time window method. Set the time length of the sliding window (such as = 10 seconds), for each time t, calculate the first derivative of the temperature rise at the time t and the previous window endpoint The average temperature rise rate between two adjacent time points is calculated; the actual temperature rise rate at the time is calculated based on the measured temperature data; and the model temperature rise rate at the same time is calculated based on the model predicted temperature data. The method realizes numerical approximation of the temperature rise slope by dividing the temperature difference between adjacent time points by the time interval, and takes into account real-time performance and noise resistance. The system calculates the difference between the actual temperature rise rate and the model temperature rise rate at each time, obtains the temperature rise rate offset value, and constructs a complete rate offset curve in the form of a time series. If the actual temperature rise is faster than the model, the offset value is positive; if the actual temperature rise is slower than the model, the offset value is also positive; and if it is completely consistent, the offset value is zero. The system outputs a complete rate offset curve in the form of a time-varying offset value sequence.

[0103] Slope activation data is obtained according to the rate offset curve data;

[0104] Specifically, the system sets a slope deviation threshold, for example, 1.5°C / min. If the following conditions are met: wherein is the monitoring start time, is the monitoring end time, is an indicator function, is a temperature rise derivative deviation function, is a slope deviation threshold, is a time variable, is an activation duration threshold (e.g., 30 seconds). The time period that meets the conditions constitutes the slope activation data interval.

[0105] Drift activation data is obtained according to the activation data.

[0106] Specifically, a three-dimensional threshold surface is constructed wherein is time, is temperature rise derivative deviation, is current temperature. The threshold surface can be obtained by empirical data fitting, historical simulation optimization or control expert parameter tuning, and represents the drift risk intensity evaluation result under the joint action of the above three parameters. In the embodiments, the threshold surface can be modeled by a polynomial function, for example, a weighted sum of the square term of the temperature rise derivative deviation, the first term of the temperature value, the first term of the time and the constant term, which represents the contribution strength of different influence factors to the drift risk. wherein is a disturbance intensity evaluation function value, is a temperature rise derivative deviation, is a derivative deviation weighting coefficient, is a temperature weighting coefficient, is a time weighting coefficient, The bias term (constant term) represents the base value or starting threshold value of the disturbance intensity score, which is used to adjust the overall level. The system performs threshold value determination on the three-dimensional feature vector at each time point: if the threshold surface output value is greater than the preset drift activation threshold value, it is determined that there is a significant drift risk at this time, triggering the anti-drift mechanism; the threshold value can be adjusted according to experience, adapting to different heating power scenarios or batch differences of raw materials. To avoid misjudgment or mutation caused by "critical value effect", the system can set fuzzy judgment for the threshold trigger area: for the judgment points close to the threshold value (such as floating up and down by a certain percentage, for example, plus or minus 10%), a fuzzy weight factor is set; the factor represents the intensity or confidence of the current drift activation, with a value range of 0 to 1; it is dynamically adjusted according to the activation degree to enhance the robustness and response smoothness of the system. The system integrates all time periods that meet the threshold surface trigger conditions into a drift activation data segment.

[0107] Preferably, the anti-initial value drift section processing is specifically:

[0108] The drift activation data and the preset prior thermal inertia data are compared to obtain drift error data;

[0109] Specifically, in order to identify whether there is a significant deviation from the normal thermal inertia behavior in the current temperature rise process, the system compares the current temperature rise rate with the prior thermal inertia model in the drift activation section. Based on the measured tank temperature data, the system calculates the temperature change rate per unit time in the drift activation section. This rate can be calculated by taking the first derivative of the temperature time series at each time or using the sliding window difference quotient method, reflecting the current actual temperature rise speed, denoted as the real-time temperature rise rate. The system pre-constructs and stores a set of standard thermal inertia response data, referred to as the prior thermal inertia model. This model can be obtained based on the stable operation data statistics in the historical multiple batches of temperature rise processes, or established through experimental calibration, representing the reference temperature rise rate under standard conditions for the same structure and the same raw material properties, denoted as the reference temperature rise rate. The system calculates the difference between the current actual temperature rise rate and the reference temperature rise rate to form the drift error data in time series form. This error at each time point represents the deviation of the current system temperature rise behavior from the ideal thermal inertia response, which can be used for subsequent disturbance intensity evaluation, drift risk quantitative analysis or compensation control strategy generation. The output drift error data is a sequence of error signals that change with time, which is an important input basis for drift intensity integral calculation and risk interval division.

[0110] The drift error data is subjected to disturbance intensity integral calculation to obtain disturbance intensity integral data;

[0111] Specifically, the disturbance intensity is defined as the absolute area of the error curve, representing the cumulative deviation intensity: wherein a disturbance intensity integral data, a start time of a drift activation time period, an end time of the drift activation time period, a drift error function, a time variable. The integral process can be implemented by numerical methods such as trapezoidal integration, Simpson's rule or midpoint method, etc. numerical approximation algorithm, to ensure that the total amount of disturbance can be accurately obtained even if the sampling time is short. The output is a single value type disturbance intensity integral index.

[0112] According to the disturbance intensity integral data, the drift activation data is divided into drift risk interval to obtain drift risk interval data;

[0113] Specifically, the system divides the drift activation data into risk levels based on the disturbance intensity integral index to realize dynamic evaluation and response adjustment of the temperature control abnormal trend. The system sets the classification threshold of the disturbance intensity integral value to divide the disturbance degree in the drift activation time period into different risk levels: low risk interval: if the disturbance intensity integral in the corresponding time period is less than the first threshold, it is judged as low risk; medium risk interval: if the disturbance intensity integral value is between the first threshold and the second threshold, it is judged as medium risk; high risk interval: if the disturbance intensity integral value is greater than or equal to the second threshold, it is judged as high risk. The first threshold and the second threshold can be set based on historical data analysis, field experience or dynamic adaptive strategy, which has engineering adjustability, and the second threshold is greater than the first threshold. The system divides the drift activation time period into multiple fixed length subintervals (such as every 30 seconds), and calculates the corresponding local disturbance intensity integral on each time period by applying a sliding window mechanism. According to the above classification rules, each subinterval is labeled with the corresponding risk level label. The output form is a "time period-risk level" mapping table, which is used to indicate the risk state of each time slice, for example, t1 is the start time of the time period, time period [t1, t1+30 seconds]: medium risk; time period [t1+30 seconds, t1+60 seconds]: high risk; time period [t1+60 seconds, t1+90 seconds]: low risk; …… and so on.

[0114] According to the drift risk interval data, the drift activation data is corrected to obtain preliminary section data.

[0115] Specifically, the system dynamically corrects the temperature rise trajectory in the corresponding drift activation section based on the aforementioned drift risk interval data, thereby generating preliminary trajectory section data after anti-drift processing. According to the risk level of the current drift activation section, the temperature rise trajectory is implemented with corresponding correction control strategy to realize hierarchical response and control the temperature control deviation caused by overshoot and thermal inertia of the system:

[0116] As shown in Table 1: risk level control strategy table

[0117]

[0118] For the medium-high risk interval, the system introduces a dynamic trajectory correction function to generate the temperature change curve. Taking the high-risk interval as an example, an exponential temperature rise function with a slow start characteristic can be used as a substitute trajectory: wherein is the corrected temperature function value, is the corrected starting temperature, is the temperature increment amplitude, is the base of natural logarithm, is the slow rise adjustment coefficient, is the current time, is the corrected starting time. The corrected trajectory segment is output as the preliminary section data, replacing the original active interval.

[0119] Preferably, wherein the main temperature rise section processing is specifically:

[0120] The main temperature rise section boundary is determined for the thermal resistance-thermal capacity series-parallel model to obtain the main temperature rise section boundary data;

[0121] Specifically, the system receives the following model and parameters as input: thermal resistance-thermal capacity series-parallel model, target temperature value set by the system (manual input or pre-input), temperature rise starting temperature value (obtained by collecting the sensor pre-set on the terminal), maximum applicable heating power value (manual input or pre-input). The starting time (main temperature rise section start) is the time point after the completion of the initial value drift section processing. The end time (main temperature rise section termination) is triggered when any of the following conditions is met: the real-time temperature reaches η times of the target temperature, wherein η is a pre-set proportion coefficient, the value range is 0.90 to 0.98, and is used to reserve a disturbance absorption window; the current temperature rise speed (i.e. the first derivative of temperature) drops below the specified rate threshold, indicating that the system thermal inertia response tends to be stable and the temperature rise power weakens. The system simulates the temperature rise trajectory based on the thermal resistance-thermal capacity series-parallel model under the input maximum power to obtain the fitting curve of temperature change with time; combined with the derivative trend and absolute temperature value of the simulated temperature curve, the starting time and termination time of the main temperature rise section are automatically identified according to the above boundary definition conditions; output structured time interval data for labeling the time window range of the main temperature rise section. The system outputs the boundary time data of the main temperature rise section, i.e. time pair.

[0122] According to the main temperature rise section boundary data, the shortest temperature rise time is backstepped for the thermal resistance-thermal capacity series-parallel model to obtain the shortest temperature rise time data;

[0123] Specifically, based on the thermal resistance-thermal capacity series-parallel model, the theoretical shortest temperature rising time required for the system to rise from the current starting temperature to the target temperature is inversely calculated under the constraint condition of limiting the maximum heating power, which is used to set the energy efficiency baseline or the optimization target function. According to the starting temperature and the target temperature recorded in the boundary of the main temperature rising section, and in combination with the current maximum heating power upper limit, the system determines that the temperature rising target is to solve the shortest time required for the temperature to rise from the current starting temperature to the target temperature without exceeding the maximum heating power. The thermal resistance-thermal capacity series-parallel structure model is converted into a first-order or second-order linear differential equation form, so that it has state space solvability; the power input function is set as a unit step function, that is, the system continuously heats at the maximum power since time zero; the numerical simulation method (such as the fourth-order Runge-Kutta method) is used to integrate and calculate the response trajectory of the temperature over time; in the simulation process, the time point at which the temperature curve first meets the condition that the current temperature is greater than or equal to the target temperature is determined in real time, and the time is defined as the shortest temperature rising time. The system outputs the theoretical shortest temperature rising time, which is a single scalar variable, reflecting the shortest time required for the system to reach the target temperature under the current model structure and power constraint.

[0124] According to the shortest temperature rising time data, the constant temperature rising rate slope of the thermal resistance-thermal capacity series-parallel model is calculated to obtain constant temperature rising rate slope data;

[0125] Specifically, based on the theoretical shortest temperature rising time obtained by the foregoing solving, the constant temperature rising rate slope required for the system to realize the temperature rising from the starting temperature to the target temperature under ideal conditions is calculated, and the upper and lower limit conditions are modified to obtain the constant rate parameter that can be used for the construction of the temperature control trajectory. According to the difference between the starting temperature and the target temperature of the main temperature rising section, and the theoretical shortest temperature rising time, the system calculates the constant temperature rising rate slope under ideal conditions. The slope represents the temperature amplitude that the system should improve per unit time, and constitutes the basic reference parameter of the subsequent temperature rising control function. In order to adapt to the control ability and safety limit of the actual system, the system modifies the upper and lower limits of the constant temperature rising rate slope, which specifically includes that if the system is preset with a maximum temperature rising rate limit value, that is, the maximum temperature change speed allowed per unit time, the calculated constant slope needs to be compared with the upper limit, and the smaller one is taken as the current effective temperature rising slope; if the system has a minimum rate requirement for temperature rising response (such as ensuring a certain thermal driving efficiency or time constraint), the slope is compared with the minimum rate lower limit, and the larger one is taken as the temperature rising slope. The output slope parameter is a scalar variable, representing the constant temperature rising rate slope required for the temperature to rise to the target temperature in the shortest time, and considering the upper and lower limits of the temperature rising rate allowed by the system.

[0126] According to the constant temperature rising rate slope, the main temperature rising section function is generated to obtain the main temperature rising section data.

[0127] Specifically, based on the calculated constant temperature rise rate slope, the temperature trajectory function of the main heating stage is constructed and discretized for actual control execution, so as to obtain the main heating stage temperature setting data. The system can model the main heating stage using the following two types of functions according to different control targets and thermal inertia response characteristics: when the system has stable heating capacity and the thermal inertia is small or has been smoothly transitioned through the previous anti-drift control, a linear temperature function is used to construct the main heating stage trajectory. The temperature function is: the temperature of the main heating stage is equal to the starting temperature plus the product of the effective heating slope and the time difference (current time-starting time). If the system has strong thermal inertia or has high requirements for the stability of the temperature control process, a slow-rising function in exponential form is generated to avoid temperature shock. The specific expression is as follows: the temperature function is equal to the target temperature minus an exponential decay term, which ensures that the heating speed gradually slows down, that is wherein is the current temperature value, is the target temperature, is the initial temperature, is the base of natural logarithm, is the heating response adjustment coefficient, is the current time, is the starting time of the main heating stage. The system performs numerical discretization processing and control instruction conversion on the above temperature control trajectory function, which specifically includes discretizing the continuous trajectory function into temperature set point sequences according to time intervals (for example, every second) to form temperature control frames of a certain length, which can be input to the PID controller or other temperature control execution module; based on the fitting relationship between the temperature trajectory function and the thermal resistance-thermal capacity model, a calculated power reference value can be attached to each set point to assist the power backstepping module to achieve accurate heating control. The output is a complete main heating stage temperature control function data set, which specifically includes temperature set point time sequence, indicating the time node corresponding to each control frame; temperature set point value sequence, indicating the set temperature that the system should reach at each time; and corresponding heating trajectory function analytical expression.

[0128] Preferably, wherein the disturbance absorption stage processing is specifically:

[0129] The thermal resistance-thermal capacity series-parallel model is subjected to absorption stage triggering judgment to obtain absorption stage data;

[0130] Specifically, the input data includes a real-time temperature curve: a real-time temperature change sequence collected from a tank body or system temperature sensor, used to reflect the actual temperature rise behavior of the system under the current operating state; a model-predicted temperature curve: a theoretical temperature rise curve derived based on a thermal resistance-thermal capacity series-parallel model, representing the temperature rise trend under an ideal undisturbed state; a temperature rise rate deviation sequence: the difference between the two in the time derivative level, calculated as the absolute value of the difference between the real-time temperature change rate and the predicted temperature change rate. The system can be determined to currently exist a disturbance event and trigger the absorption segment identification according to any one of the following conditions: if the temperature rise rate deviation continuously exceeds the set threshold and the duration exceeds the preset duration threshold τ seconds, it indicates that the system is subjected to a sudden temperature rise disturbance. For example, the temperature rise derivative deviation is greater than the threshold and the duration is > τ seconds. If the system temperature curve appears a platform period or irregular fluctuations within a period of time, which is manifested as the first derivative after low-pass filtering being lower than the set threshold, and the temperature standard deviation exceeding the set upper limit within the time period, it is determined that the system exists temperature shock. The system uses a sliding window analysis strategy to continuously monitor the temperature change data, and the sliding window time range is set to 20-30 seconds. The system calculates the indicators involved in the above two criteria in each sliding window and conducts joint analysis. The specific fusion strategy can be optimized and adjusted according to the set weight or logical relationship to achieve sensitive identification of different types of disturbances. When any criterion is met, the system outputs the following information as the absorption segment judgment result: absorption segment trigger timestamp, i.e. the starting time when the disturbance occurs and meets the trigger condition, used to mark the starting point of the absorption segment; absorption segment preliminary label information, including but not limited to “sudden temperature rise”, “sudden temperature drop” or “temperature platform” (i.e. the system temperature curve appears a platform period or irregular fluctuations within a period of time) and the like.

[0131] According to the absorption segment data, the disturbance absorption segment time window is determined, and the disturbance absorption segment time window data is obtained;

[0132] Specifically, based on the absorption segment trigger information, the effective duration interval of the disturbance absorption segment on the time axis, i.e. the absorption segment time window, is determined to support the temperature control trajectory switching and disturbance type identification. The system sets a preliminary time window based on the identified disturbance absorption segment trigger time, specifically including a starting time, set as the absorption segment trigger time point determined in the foregoing step, i.e. the time when the system recognizes the temperature rise disturbance signal; an ending time is initially estimated, and by default, it is extended by a fixed time period based on the trigger time, denoted as Δt, set to 60-120 seconds, to cover most of the disturbance absorption period; the foregoing two are combined to estimate the coarse-grained disturbance absorption segment length. The system performs temperature recovery trend judgment and dynamically corrects the absorption segment ending time, specifically as follows: the absolute error between the real-time temperature and the prediction model is calculated, defined as the residual function, i.e. , wherein is the residual, the measured temperature value, the temperature value predicted by the thermal resistance-thermal capacity model. If the system detects that the residual error is less than the safety error threshold for consecutive τ seconds, it is considered that the current disturbance has ended; at this time, the current time is taken as the true end time of the disturbance segment, and the effective time window of the disturbance absorption segment is locked. The system outputs the time window data and attribute label information of the disturbance absorption segment, including the disturbance absorption segment time window: expressed in the form of a time interval, used to clearly define the duration of the disturbance; the disturbance attribute type label: combined with the temperature change pattern and rate characteristics, the absorption segment is assigned one of the following type labels: platform type disturbance: the temperature curve approximately remains constant for a period of time, oscillation type disturbance: the temperature curve shows obvious up and down oscillation fluctuations, slope reversal type disturbance: the temperature derivative direction reverses.

[0133] According to the disturbance absorption segment time window data, the thermal resistance-thermal capacity series-parallel model is processed to obtain the disturbance absorption segment data.

[0134] Specifically, for the time window corresponding to the disturbance absorption segment, a smooth transition temperature control function is constructed to avoid sudden changes in the temperature control trajectory after the disturbance ends, thereby reducing the risk of thermal shock to the system and enhancing the stability and safety of the thermal control system. After the end of the disturbance segment, a temperature buffer transition mechanism is introduced, and the specific goals include avoiding trajectory sudden jumps: not directly returning to the main temperature rise trajectory in the early stage of disturbance recovery, preventing power or temperature slope from appearing drastic fluctuations; introducing a transition smoothing period: delaying convergence to the target temperature through smooth trajectory, alleviating the impact of disturbance on the thermal inertia of the system. For different types of disturbances, the following two types of temperature control functions can be selected to construct the transition trajectory: exponential transition function (suitable for slope mutation type disturbance), after the disturbance trigger time, an exponential function is constructed to gradually converge the temperature to the target temperature: the current temperature is equal to the temperature at the disturbance trigger time plus the target temperature difference multiplied by an exponential decay factor; wherein the current temperature value, the measured temperature at the disturbance trigger time, the target temperature of the main temperature rise segment, the base of the natural logarithm, the slow convergence adjustment coefficient, which is positively related to the disturbance intensity, and is usually adjusted between 0.01 and 0.1, used to control the transition speed, the current time, The disturbance trigger time is the time at which the disturbance is triggered. For temperature plateau or oscillation disturbance types, a piecewise linear function can be used for transition, and the structure is as follows: first stage: the temperature remains constant for 5-20 seconds, which is used for system buffering; second stage: slowly rising at a slow temperature rise slope (usually 50%-80% of the normal temperature rise slope); third stage: a smooth connection curve is constructed between the current temperature and the main temperature rise segment to ensure continuity and slope consistency, and is seamlessly spliced to the main temperature rise track. According to the generated disturbance absorption segment temperature curve, the heating power required at each control time is recalculated; the power change rate constraint is set to limit the power adjustment amplitude per unit time, so as to reduce the equipment response pressure and avoid the energy consumption surge or thermal inertia accumulation caused by "power jump". According to the above strategy, the system generates the following output data: absorption segment temperature trajectory function data: contains the target temperature value corresponding to each time in the absorption segment, that is, the time-temperature pair sequence, which can be directly used as the input of the controller; power correction suggestion: includes the recommended power adjustment value at each sampling time or the limitation condition of the power change slope; absorption segment end flag: clearly indicates the end time of the absorption segment, which is used for splicing with the main temperature rise segment track to ensure the time continuity and physical consistency of the overall track.

[0135] Preferably, step S3 is specifically:

[0136] Step S31: performing trajectory splicing according to the anti-initial value drift segment data, the main temperature rise segment data and the disturbance absorption segment data to obtain trajectory splicing data;

[0137] Specifically, after completing the modeling of each stage temperature trajectory, it is uniformly processed and sequentially spliced to generate a complete temperature setting trajectory which is continuous in time, reasonable in physics and derivable in value, for guiding the heating control system to execute. The system receives temperature trajectory data from the following three stages: anti-initial value drift section trajectory data, denoted as the first segment temperature trajectory, corresponding to the time interval from the starting time to the first turning point; main heating section trajectory data, denoted as the second segment temperature trajectory, corresponding to the time interval from the turning point to the completion of the main target temperature zone; disturbance absorption section trajectory data, denoted as the third segment temperature trajectory, corresponding to the time interval from the main heating section to the system entering the stable transition period. Each segment trajectory is a time-temperature sequence and can be described by a function form. The system uses the following strategy for splicing processing: the difference between the end point temperature of each adjacent two segment trajectories and the start point temperature of the next segment should not be greater than 1 degree Celsius to prevent temperature jump; if the end point temperature difference of two segment trajectories is out of limit, a transition trajectory is inserted between them. The transition segment can use linear interpolation or Hermite interpolation method to realize smooth connection of slope and take into account the continuity and derivability. The system uniformly samples the three segment trajectories, samples each segment trajectory at a fixed time interval (such as once per second) to obtain a set of temperature points under the standard time axis; the three sampled trajectories are spliced in time sequence to form a unified time-temperature trajectory sequence. Each sampling point in the sequence contains a sampling time and a corresponding target temperature value. The output is a complete temperature control trajectory with the following characteristics: from the starting time of the anti-initial value drift section to the end time of the disturbance absorption section, the entire trajectory has no time interruption; the temperature change between any adjacent sampling points is smooth, supporting the calculation of first order derivative (temperature rise rate) and second order derivative (thermal inertia); each time point contains a data pair of "time point-target temperature value".

[0138] Step S32: Calculate the temperature derivative of the trajectory splicing data to obtain the temperature derivative data;

[0139] Specifically, the derivative of the complete temperature trajectory generated in the previous stage is calculated to obtain the temperature change rate sequence over time (i.e., the temperature rise rate), which provides the basis for power estimation and control response analysis. For each time point in the temperature trajectory, the first-order derivative value of the temperature at the point is calculated, which represents the temperature rise rate per unit time, and is defined as the derivative of the target temperature with respect to time, i.e., the temperature rise rate = the temperature change value after a certain time point divided by the time interval, which is represented as the first-order derivative of the temperature with respect to time. The sliding difference method (central difference) is used for approximate calculation. For the i-th time point in the trajectory, the temperature derivative is approximately represented as: current point temperature rise rate ≈ temperature difference between adjacent previous and subsequent two time points divided by the total time interval between them. That is, the data of the previous time point and the subsequent time point are used to estimate the derivative of the current time point, thereby obtaining a smooth and relatively accurate temperature change rate. The time step is taken as 1 second, which can also be adjusted according to the actual sampling period of the trajectory data. In order to reduce the high-frequency noise interference caused by sensor errors or sampling fluctuations, one of the following smoothing processing strategies is recommended before and after the difference calculation: using a Savitzky-Golay smoothing filter to process the original temperature sequence or the derivative sequence; or using a sliding median filter to suppress abnormal jumps. The system is processed in the following scenarios: for the disturbance absorption segment, in this segment, in order to retain the disturbance transition characteristics, the sliding window width can be appropriately reduced (such as taking only the previous and subsequent 1 point), and the derivative response sensitivity is improved; for boundary point processing: at the starting point, since there is no previous data point, the forward difference method (using the current point and the next point) is used for estimation; at the end point, since there is no subsequent data point, the backward difference method (using the current point and the previous data point) is used for estimation; to avoid derivative missing or error offset at the trajectory boundary. The system output is a set of temperature rise rate data point sequences with time stamps, each data point including the current time point; and the corresponding temperature rise rate (i.e., the first-order derivative value of the temperature).

[0140] Step S33: According to the temperature derivative data and the thermal resistance-thermal capacity series-parallel model, the heating power data is obtained by power backstepping.

[0141] Specifically, based on the previously obtained temperature derivative data and thermal resistance-thermal capacity series-parallel model parameters, the real-time heating power during the heating process of the tank body is inversely calculated, so as to evaluate the energy consumption level and the execution effect of the heat control strategy. In the thermal resistance-thermal capacity series-parallel network model (mainly in the first order approximation), the expression of the heating power at any time includes two items: one is the temperature rise power consumption due to the thermal capacity, and the other is the heat loss caused by the temperature difference with the environment. The overall composition is as follows: the first item represents the thermal capacity effect: the equivalent heat capacity multiplied by the temperature change rate; the second item represents the heat dissipation effect: the difference between the current tank body temperature and the ambient temperature divided by the equivalent thermal resistance. That is, the total power is jointly determined by the system heat storage response and the environmental heat dissipation load. The input temperature derivative data, that is, the temperature rise rate sequence per unit time; the temperature trajectory data, that is, the tank body temperature at each time; the environmental temperature data, which is collected in real time or obtained through a sensor. The equivalent heat capacity is estimated by factors such as tank body material, liquid volume and specific heat capacity; the equivalent thermal resistance is obtained by historical data fitting, system calibration or thermal simulation. If a multi-node series-parallel structure model is used, it is converted into a state space model, and the relationship between the power input and the state derivative is solved through matrix operation. The system calls the thermal resistance-thermal capacity model constructed in step S1 to extract the required equivalent heat capacity and equivalent thermal resistance parameters. For each time point in the temperature derivative sequence, according to the current derivative value, the real-time temperature and the environmental temperature, the power model is calculated, and the heating power value is generated point by point. If the inverse calculation power value at a certain time point appears non-physical jump (such as sudden increase or decrease or negative value), the system determines that it is a sensor abnormality; external disturbance (such as opening the cover or liquid fluctuation); or model fitting error; such points will be marked with an abnormal label for the trajectory adjustment module to reference; the system sets the upper and lower limit thresholds of the power, and exceeds the range can trigger the alarm mechanism or switch back to the predicted power value. The output is a set of heating power value pairs arranged by time, including the time stamp and the corresponding power value.

[0142] Preferably, step S4 is specifically:

[0143] Step S41: obtaining real-time change data;

[0144] Specifically, the real-time change factors affecting the execution precision of the temperature rising trajectory and the system response behavior during system operation are continuously monitored and data is collected to provide input basis for trajectory dynamic adjustment. The real-time change data collected by the system can include obtaining the deviation value between the current actual temperature of the tank and the set temperature; collecting the external environment temperature and comparing it with the previous period temperature, if the temperature difference exceeds the set threshold, it is judged as an environmental disturbance trigger condition; raw material replenishment or mutation event detection, if the liquid level sensor is detected to have a mutation (such as a sharp increase or a sharp decrease in liquid level), it is judged as a replenishment event, or a transient drop in internal pressure of the tank, which can also be regarded as a raw material disturbance behavior. Real-time collection of heating actuator feedback power, if the feedback power continuously exceeds the set upper threshold (such as excessive load), the system will record it as an abnormal energy consumption event or equipment anomaly. The real-time data sampling period is recommended to be set to 1-5 seconds, the system uses a sliding window caching mechanism to dynamically analyze the trend of the continuous data in the past 10-30 seconds to avoid misjudgment due to transient fluctuations; all kinds of real-time change data will be packaged as structured data sets as input of the dynamic control module. The real-time change data set output by the system includes but is not limited to temperature deviation, temperature error at the current time; environmental temperature difference, current environmental temperature change; replenishment event identification, Boolean flag or liquid level mutation value; power anomaly flag or feedback power value; other disturbance related data fields (such as tank pressure, heater status, etc.).

[0145] Step S42: judging the trajectory change trigger condition according to the real-time change data, obtaining trajectory change data;

[0146] Specifically, based on the real-time collected system running state data, it is judged whether the trigger condition of trajectory adjustment is met to realize dynamic response and trajectory update in the temperature control process. The system judges the following typical disturbance types and sets the corresponding trigger rules:

[0147] As shown in Table 2: Trigger type table of trajectory change

[0148]

[0149] When any of the above trigger conditions is met, the system will generate a trajectory change flag and trigger time information, such as setting the change flag bit to an open state and recording the trigger time as the current system clock time, that is, the trajectory change flag: change flag = open; trigger timestamp: change trigger time = current time. The output trajectory change data structure includes disturbance type fields such as "temperature difference disturbance", "environmental disturbance", "liquid level disturbance", etc.; trigger time field, recording the timestamp of the first time the trigger condition is met; additional information such as trigger threshold, deviation amplitude, power error value, etc.

[0150] Step S43: trajectory remaining segment clipping according to the trajectory change data and the heating power data, obtaining trajectory remaining segment data;

[0151] Specifically, based on the triggered trajectory change information, the original temperature setting trajectory is truncated, and the key features and power trend data of the remaining segment are extracted for input of the reconstruction logic. The system uses the trigger time in the current trajectory change data as the boundary to clip the original temperature setting trajectory: the original temperature setting trajectory is a set of ordered temperature-time pairs, and there is a corresponding heating power setting sequence; the system defines the part of the original setting trajectory where the current time is greater than the trajectory change trigger time as the trajectory remaining segment. The system analyzes the thermodynamic characteristics of the trajectory remaining segment, including analyzing whether the trajectory remaining segment sequence is in the warming state, constant temperature state or has temperature overshoot behavior at the clipping starting point; based on the joint evolution trend of the temperature-time pairs and the heating power setting sequence of the trajectory remaining segment sequence, the warming slope (such as the temperature rise rate of linear warming) of the current control stage is calculated; the power target value or actual predicted power in the current paragraph is extracted to evaluate the actuator control stability and power boundary constraint situation. The system outputs the following data content: temperature remaining trajectory segment: the temperature set point sequence in the original setting trajectory where t> trajectory change trigger time; power remaining trajectory segment: the heating power data matching the above time sequence; trend parameter set: including temperature trend label (warming / constant temperature / overshoot), current control slope value, power estimation information, etc.

[0152] Step S44: generating a change window according to the trajectory remaining segment data to obtain change window data;

[0153] Specifically, based on the data characteristics of the remaining segment of the trajectory and the response capability of the control system, an effective time window for trajectory adjustment, i.e., a change window, is determined, which is used for the subsequent deployment range of the starting part of the interpolation trajectory or the correction function. The change window refers to the minimum control time length within which the trajectory change operation can be safely implemented under the current thermal state and power state of the system. This window not only meets the physical feasibility of the system, but also provides sufficient regulation margin to accommodate the trajectory slow-changing function. The system sets the time length of the change window in combination with the following constraint conditions: power change rate limit, to avoid sudden changes in heating power during trajectory reconstruction, it is necessary to ensure that the subsequent power change rate does not exceed the maximum power slope that the system can withstand, i.e., the power change per unit time after trajectory adjustment should not exceed the preset safe upper limit; thermal inertia response delay, considering the physical lag characteristics of the thermal resistance-thermal capacity system, the system estimates the typical thermal response delay time (such as 10-30 seconds) under the current state as a reference for the delayed effect of trajectory change; actuator control cycle limit, the system also needs to combine the minimum control cycle of the temperature control actuator (such as a heater or an electromagnetic valve) to ensure that the control instructions in the trajectory adjustment process can be stably executed at the hardware level. The starting time is set as the time when the trajectory change is triggered; the end time is set as the time when the trajectory change is triggered plus the window length of dynamic adjustment, the value range is 20-60 seconds, depending on the above three constraint conditions. This window segment is the change window. The system will output the following structured data, the change window data contains two fields of starting time and end time, which are used to define the deployment interval of the new function segment that allows the insertion of the trajectory.

[0154] Step S45: trajectory function re-planning according to the change window data to obtain heating change data.

[0155] Specifically, according to the generated trajectory change window, the temperature setting trajectory is functionally reconstructed to realize dynamic adjustment of the thermal control strategy. The reconstruction target is to ensure that the system temperature curve has good continuity and smoothness before and after the change point, avoiding temperature control overshoot or control rupture, so as to maintain the stability of the system thermal inertia response. Within the change window, that is, the starting time is the trajectory change trigger time, and the ending time is the trigger time plus the window length, the system replaces the original set trajectory with the newly fitted temperature function. According to the control target and thermal response characteristics, one of the following three types of trajectory function is used: piecewise linear function (suitable for scenarios that require rapid re-setting of target temperature, with constant temperature rise slope), exponential slow rise function (used for scenarios that need to suppress system inertia or avoid oscillation, with a smooth change trend from slow to fast and then gradually stable), and cubic spline interpolation function (if continuous interpolation transition between the starting temperature and the original set trajectory is required, a cubic spline function can be used to ensure that the fitted curve remains continuous in the first derivative, thereby avoiding temperature curve breakpoints or sudden slope changes). The temperature function after trajectory re-planning consists of three parts: the first segment: before the change trigger, the original set trajectory is used; the second segment: within the change window, the re-planned temperature trajectory function is used; the third segment: after the end of the change window, it is connected to the remaining part of the original trajectory and can be translated or slope corrected according to the adjusted target. The overall trajectory function should meet the continuity and derivability requirements, that is, the starting and ending temperature values of each segment need to be continuous, and a short smooth segment can be inserted at the connection point if necessary. Based on the re-planning result of the temperature trajectory, the system will generate a new power demand trajectory. Using the thermal resistance-thermal capacity model, the required heating power at each time is estimated as follows: first, calculate the first derivative of the reconstructed temperature curve at each time, that is, the temperature rise rate; then, according to the equivalent heat capacity, equivalent thermal resistance and real-time environment temperature, calculate the required power; this power value will be used to control the output of the actuator, and also used to judge the energy consumption trend. The system outputs the following heating change data content: the new set temperature trajectory function, consisting of three segments, including: the original trajectory (before the change trigger point); the re-planned function (during the change window); the adjusted remaining trajectory (after the change window); the synchronously generated heating power data sequence for control execution and energy consumption analysis; function parameter information, such as the slope coefficient of the exponential function and the control point of the spline function, is output for further explanation of the fitting behavior.

[0156] Preferably, the application also provides a polyurethane raw material preheating control system for executing the polyurethane raw material preheating control method as described above, which comprises:

[0157] The thermal resistance-heat capacity modeling module is used to acquire tank temperature data, ambient temperature data, batch temperature rise data, tank data, and raw material liquid data; and to construct a thermal resistance-heat capacity series-parallel model based on the tank temperature data, ambient temperature data, batch temperature rise data, tank data, and raw material liquid data to obtain the thermal resistance-heat capacity series-parallel model.

[0158] The segmented response identification and processing module is used to process the anti-initial-value drift segment of the thermal resistance-thermal-capacity series-parallel model to obtain the anti-initial-value drift segment data; process the main heating segment of the thermal resistance-thermal-capacity series-parallel model to obtain the main heating segment data; and process the disturbance absorption segment of the thermal resistance-thermal-capacity series-parallel model to obtain the disturbance absorption segment data.

[0159] The reverse power calculation module is used to back-calculate the heating power data based on the data of the anti-initial drift section, the main heating section, and the disturbance absorption section.

[0160] The heating trajectory change control module is used to acquire real-time change data and change the trajectory based on the heating power data and the real-time change data to obtain heating change data.

[0161] Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended application documents rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of the equivalents of the application documents be incorporated into the invention.

[0162] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A polyurethane raw material preheating control method characterized by, The method comprises the following steps: Step S1: obtaining tank temperature data, environment temperature data, batch temperature rise data, tank data and raw material liquid data; According to the tank temperature data, the environment temperature data, the batch temperature rise data, the tank data and the raw material liquid data, a thermal resistance-thermal capacity series-parallel model is constructed to obtain the thermal resistance-thermal capacity series-parallel model; Step S2: initial drift data is obtained by performing initial drift discrimination on the thermal resistance-thermal capacity series-parallel model; drift activation data is obtained by performing temperature rise rate deviation activation discrimination on the initial drift data; preliminary section data is obtained by performing anti-initial value drift section processing on the drift activation data; anti-initial value drift section data is obtained by performing slow rise section dynamic feedback correction on the preliminary section data; main temperature rise section boundary data is obtained by performing main temperature rise section boundary determination on the thermal resistance-thermal capacity series-parallel model; the shortest temperature rise time data is obtained by performing shortest temperature rise time backstepping on the thermal resistance-thermal capacity series-parallel model according to the main temperature rise section boundary data; the constant temperature rise rate slope data is obtained by performing constant temperature rise rate slope calculation on the thermal resistance-thermal capacity series-parallel model according to the shortest temperature rise time data; the main temperature rise section data is obtained by performing main temperature rise section function generation according to the constant temperature rise rate slope; the absorption section data is obtained by performing absorption section triggering judgment on the thermal resistance-thermal capacity series-parallel model; the disturbance absorption section time window data is obtained by performing disturbance absorption section time window determination according to the absorption section data; the disturbance absorption section data is obtained by performing slow convergence structure processing on the thermal resistance-thermal capacity series-parallel model according to the disturbance absorption section time window data; Step S3: trajectory splicing data is obtained by performing trajectory splicing according to the anti-initial value drift section data, the main temperature rise section data and the disturbance absorption section data; the temperature derivative data is obtained by performing temperature derivative calculation on the trajectory splicing data; the heating power data is obtained by performing power backstepping according to the temperature derivative data and the thermal resistance-thermal capacity series-parallel model; Step S4: real-time change data is obtained; trajectory change data is obtained by performing trajectory change triggering condition judgment according to the real-time change data; trajectory remaining section data is obtained by performing trajectory remaining section clipping according to the trajectory change data and the heating power data; change window data is obtained by performing change window generation according to the trajectory remaining section data; The trajectory function is re-planned according to the change window data to obtain heating change data.

2. The method of claim 1, wherein, The thermal resistance-thermal capacity series-parallel model is constructed as follows: The thermal capacity-thermal resistance network model is constructed according to the tank data and the raw material liquid data to obtain the thermal capacity-thermal resistance network model; The state space model data is obtained by performing state space model processing on the thermal capacity-thermal resistance network model data according to the tank temperature data and the liquid temperature data in the raw material liquid data; The thermal resistance-thermal capacity series-parallel model is obtained by labeling the state space model data with the tank temperature data, the environment temperature data and the batch temperature rise data.

3. The method of claim 1, wherein, The temperature rise rate deviation activation discrimination is specifically as follows: The rate offset curve data is obtained by performing rate offset curve construction on the initial drift data; The slope activation data is obtained by performing slope activation judgment according to the rate offset curve data; The drift activation data is obtained by performing threshold surface triggering according to the activation data.

4. The method of claim 1, wherein, The anti-initial value drift section processing is specifically: The drift error data is obtained by differentiating and comparing the drift activation data and the preset prior thermal inertia data; The disturbance intensity integral data is obtained by performing disturbance intensity integral calculation on the drift error data; The drift risk interval data is obtained by dividing the drift risk interval of the drift activation data according to the disturbance intensity integral data; The preliminary section data is obtained by performing anti-drift correction on the drift activation data according to the drift risk interval data.

5. A polyurethane feedstock preheating control system, characterized by, A polyurethane raw material preheating control system for executing the polyurethane raw material preheating control method of claim 1 comprises: A thermal resistance-thermal capacity modeling module is configured to acquire tank temperature data, environment temperature data, batch temperature rise data, tank data, and raw material liquid data; and construct a thermal resistance-thermal capacity series-parallel model based on the tank temperature data, the environment temperature data, the batch temperature rise data, the tank data, and the raw material liquid data to obtain the thermal resistance-thermal capacity series-parallel model. A segmented response identification and processing module is configured to perform anti-initial value drift section processing on the thermal resistance-thermal capacity series-parallel model to obtain anti-initial value drift section data; perform main temperature rise section processing on the thermal resistance-thermal capacity series-parallel model to obtain main temperature rise section data; and perform disturbance absorption section processing on the thermal resistance-thermal capacity series-parallel model to obtain disturbance absorption section data. A reverse power solving module is configured to perform power backstepping based on the anti-initial value drift section data, the main temperature rise section data, and the disturbance absorption section data to obtain heating power data. A heating trajectory change control module is configured to acquire real-time change data and perform trajectory change based on the heating power data and the real-time change data to obtain heating change data.

Images