Intelligent temperature and humidity control method applied to process oven

Abstract

The invention discloses an intelligent temperature and humidity control method applied to a process oven. The method comprises the following steps: analyzing future temperature and humidity predicted values in real time by adopting a dynamic prediction module; and the predicted value is simultaneously input to a dew point risk avoidance module and an MPC optimization module. The current dew point temperature is calculated through the dew point risk avoiding module, dew point risk judgment is conducted in combination with the material surface temperature monitored by the infrared sensor, and when it is detected that the surface temperature is close to a dew point threshold value, an instruction is sent to the gradient air supply control module. And the MPC optimization module performs rolling optimization on the air supply parameters through an objective function based on the predicted value. Therefore, the fuzzy-PID module is helped to dynamically adjust PID parameters, and accurate execution signals are output to a heater and a fan. The gradient air supply control module integrates the dew point risk instruction and the PID control signal, the oven is divided into a plurality of independent areas to execute differential regulation and control finally, the control precision of intelligent regulation of the operation state of the process oven is increased, and energy consumption is remarkably reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation control, more particularly, to a temperature and humidity intelligent regulation system and method applied to a process oven. BACKGROUND

[0002] In modern industrial production, temperature and humidity control is a key link in many process flows, especially in the fields of food processing, chemical production, and lithium battery pole piece drying, accurate temperature and humidity control is crucial to ensure product quality and improve production efficiency. Whether it is the operation state adjustment of the process oven or the temperature and humidity control of the process oven, the core goal is to achieve accurate regulation of environmental parameters through effective technical means to meet the strict requirements of specific process scenarios and ensure the stability of the production process and the high quality of the products.

[0003] Currently, traditional temperature and humidity control technology mainly relies on PID control algorithm, although it can achieve temperature and humidity adjustment to a certain extent, but there are still many shortcomings in practical application. For example, in the field of oven temperature and humidity control, the temperature fluctuation range of traditional PID control is large, usually ≥ ± 1.5 °C, and the humidity deviation is as high as ≥ ± 5% RH, which is difficult to meet the needs of high-precision process scenarios. In addition, this control method lacks self-adaptive ability to changes in material characteristics, and fixed air supply parameters result in high unit energy consumption, reaching 0.85 kW·h / kg, which not only increases production costs, but also causes great burden to the environment. At the same time, the existing technology cannot avoid the risk of dew point temperature in real time, resulting in a high process defect rate of ≥ 8%, which seriously affects the product pass rate and production efficiency.

[0004] In addition, with the continuous development of industrial automation technology, traditional temperature and humidity control technology gradually exposes its limitations. For example, low control precision leads to problems such as condensate accumulation and material fault, which seriously affect product quality and production efficiency; high energy consumption greatly increases production costs, which is not conducive to the sustainable development of enterprises; lack of dynamic optimization capability cannot avoid the risk of dew point temperature in real time, further increasing the process defect rate. The existence of these problems makes it difficult for existing technology to meet the needs of modern industrial production for high-precision, low-energy, and dynamic optimization of temperature and humidity control, and an advanced and intelligent solution is needed to overcome these shortcomings and achieve upgrading and breakthroughs in temperature and humidity control technology. SUMMARY

[0005] In view of at least one defect or improvement demand of the prior art, the present application provides a temperature and humidity intelligent regulation system and method applied to a process oven, which is used to solve the defects of the prior art and realize high-precision, anti-interference personnel activity state recognition without invading privacy.

[0006] To achieve the above object, according to the first aspect of the present application, a kind of temperature and humidity intelligent regulation system applied to process oven is provided, comprising: dynamic prediction module, dew point risk avoidance module, MPC optimization module, fuzzy-PID adaptive control module and gradient air supply control module; Dynamic prediction module, for collecting current environmental temperature parameters, temperature and humidity dynamic prediction model is constructed by LSTM algorithm, and temperature and humidity prediction value is output and sent to dew point risk avoidance module; Dew point risk avoidance module, for monitoring material surface real-time temperature, and judging whether the temperature and humidity prediction value is greater than material surface real-time temperature, if yes, gradient air supply instruction is sent to the gradient air supply control module, otherwise the temperature and humidity prediction value is sent to the MPC optimization module; MPC optimization module, for minimizing temperature deviation, humidity deviation and total energy consumption based on the temperature and humidity prediction value, using model predictive control to carry out rolling optimization, and obtaining temperature control optimization instruction; Fuzzy-PID adaptive control module, for dynamically adjusting PID parameters according to the temperature control optimization instruction and real-time temperature and humidity error and error change rate, obtaining heater power control signal and / or fan speed control signal through preset fuzzy rule; Gradient air supply control module, for dividing the oven into multiple independent air supply areas according to the gradient air supply instruction and heater power control signal and / or fan speed control signal, adjusting the air speed and air supply temperature of each area independently through variable frequency fan and segmented heater, and realizing intelligent adjustment of process oven running state.

[0007] In an optional embodiment, further comprising: data-driven optimization module, for analyzing historical process data to divide typical operating mode, and using reinforcement learning algorithm to optimize MPC weight matrix in MPC optimization module, and outputting optimization parameters to MPC optimization module.

[0008] In an optional embodiment, the reinforcement learning algorithm uses Q-learning algorithm, and the reward function is specifically: R=10×(1-∣ΔT∣) + 5×(1-∣ΔRH∣)-0.3×Ptotal; Wherein, R is reward value, P_{total} is total energy consumption, ΔT is temperature deviation absolute value (the difference between actual temperature and set temperature), ΔRH is humidity deviation absolute value (the difference between actual humidity and set humidity).

[0009] In an optional embodiment, the data-driven optimization module uses K-means algorithm to cluster historical process data, divides three operating modes of high condensation risk, stable operation and low energy consumption, and presets corresponding control parameter combination.

[0010] In an optional embodiment, the dew point risk avoidance module comprises: a calculation unit configured to calculate the dew point temperature in the oven in real time based on the August-Roche-Magnus formula and monitor the material surface temperature to obtain the real-time material surface temperature; a risk prediction unit configured to determine whether the temperature and humidity prediction value is greater than the real-time material surface temperature, and if so, send the gradient air supply instruction to the gradient air supply control module, otherwise send the temperature and humidity prediction value to the MPC optimization module.

[0011] In an optional embodiment, the dew point risk avoidance module monitors the material surface temperature in real time through an infrared sensor array to obtain the real-time material surface temperature.

[0012] In an optional embodiment, the dynamic prediction module comprises: a construction unit configured to establish a temperature and humidity dynamic prediction model through an LSTM algorithm; a prediction unit configured to input the current environmental temperature parameters into the temperature and humidity dynamic prediction model, output the temperature and humidity prediction value, and send the temperature and humidity prediction value to the dew point risk avoidance module, wherein the environmental temperature parameters include temperature and humidity, heating power, air supply amount, and environmental parameters.

[0013] In an optional embodiment, the dew point risk avoidance module is further configured to locate the low-temperature area and trigger local heating compensation when it is monitored that the material surface temperature in the local area is lower than the preset dew point temperature range, and the compensation power is 50-100 W.

[0014] In an optional embodiment, the zoning strategy of the gradient air supply control module is specifically: high-temperature and high-speed air supply is adopted for high-moisture materials in the inlet area, medium-temperature and medium-speed air supply is adopted for part of the dry materials in the middle area, and low-temperature and low-speed air supply is adopted for nearly dry materials in the outlet area.

[0015] According to a second aspect of the present application, an intelligent process oven operation state adjustment method is also provided, and the method comprises: collecting current environmental temperature parameters, constructing a temperature and humidity dynamic prediction model through an LSTM algorithm, and outputting a temperature and humidity prediction value; monitoring the real-time material surface temperature, determining whether the temperature and humidity prediction value is greater than the real-time material surface temperature, and if so, generating a gradient air supply instruction, otherwise performing MPC optimization on the temperature and humidity prediction value; the MPC optimization is a rolling optimization using model predictive control based on the temperature and humidity prediction value to minimize temperature deviation, humidity deviation, and total energy consumption to obtain a temperature control optimization instruction; According to the temperature control optimization instruction and the real-time temperature and humidity error and error change rate, the PID parameter is dynamically adjusted through a preset fuzzy rule to obtain a heater power control signal and / or a fan rotating speed control signal. According to the gradient air supply instruction and the heater power control signal and / or the fan rotating speed control signal, the oven is divided into a plurality of independent air supply areas, the air speed and the air supply temperature of each area are independently adjusted through the variable frequency fan and the segmented heater, and intelligent adjustment of the process oven running state is realized.

[0016] Overall, compared with the prior art, the above technical solutions conceived by the present application can achieve the following beneficial effects: The dynamic prediction module analyzes the temperature, humidity, heating power and other parameters in the oven in real time through the LSTM algorithm, and outputs the future temperature and humidity prediction value. The prediction value is input into the dew point risk avoidance module and the MPC optimization module at the same time. The current dew point temperature is calculated through the dew point risk avoidance module, and the dew point risk is judged in combination with the material surface temperature monitored by the infrared sensor. When the surface temperature is detected to be close to the dew point threshold, an instruction is immediately sent to the gradient air supply control module. At the same time, the MPC optimization module optimizes the air supply parameters through the target function based on the prediction value, and transmits the optimization instruction to the fuzzy-PID adaptive control module. The fuzzy-PID module dynamically adjusts the PID parameter according to the real-time error and change rate, and outputs accurate execution signals to the heater and the fan. The gradient air supply control module divides the oven into a plurality of independent areas to execute differentiated regulation and control, finally realizes intelligent adjustment of the process oven running state, greatly improves the control precision of the process oven running state, and significantly reduces the energy consumption. At the same time, through the real-time intervention of the dew point module, the condensate generation rate is greatly reduced, and the material surface dewing condition is completely blocked. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Figure 1 One of the structure schematic diagrams of an optional process oven running state intelligent adjustment system provided by the embodiments of the present application; Figure 2 The second structure schematic diagram of an optional process oven running state intelligent adjustment system provided by the embodiments of the present application; Figure 3 The flow schematic diagram of an optional process oven running state intelligent adjustment method provided by the embodiments of the present application. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely intended to explain the present application and are not used to limit the present application. In addition, the technical features involved in the various embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0020] The terms "first", "second", "third", etc. in the specification and claims of the present application and the above-described drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product or device.

[0021] According to one aspect of the embodiments of the present application, a process oven running state intelligent adjustment system is provided. The process oven running state intelligent adjustment system provided by the embodiments of the present application will be described below with reference to the accompanying drawings. Figure 1 The process oven running state intelligent adjustment system provided by the embodiments of the present application is described.

[0022] Figure 1 is one of the structures of an optional process oven running state intelligent adjustment system provided by the embodiments of the present application, as shown in Figure 1 The system includes a dynamic prediction module, a dew point risk avoidance module, an MPC optimization module, a fuzzy-PID adaptive control module and a gradient air supply control module. The dynamic prediction module is configured to collect current environmental temperature parameters, construct a temperature and humidity dynamic prediction model through an LSTM algorithm, output temperature and humidity prediction values and send the temperature and humidity prediction values to the dew point risk avoidance module. The dew point risk avoidance module is configured to monitor real-time temperature of a material surface, determine whether the temperature and humidity prediction values are greater than the real-time temperature of the material surface, and if so, send a gradient air supply instruction to the gradient air supply control module, otherwise send the temperature and humidity prediction values to the MPC optimization module. The MPC optimization module is configured to minimize temperature deviation, humidity deviation and total energy consumption based on the temperature and humidity prediction values, perform rolling optimization through model predictive control to obtain temperature control optimization instructions. The fuzzy-PID adaptive control module is configured to dynamically adjust PID parameters through preset fuzzy rules according to the temperature control optimization instructions and real-time temperature and humidity errors and error change rates to obtain a heater power control signal and / or a fan speed control signal. The gradient air supply control module is used for dividing the oven into multiple independent air supply areas according to the gradient air supply instruction and the heater power control signal and / or the fan rotating speed control signal, independently adjusting the air speed and air supply temperature of each area through the variable frequency fan and the segmented heater, and realizing intelligent adjustment of the process oven running state.

[0023] An optional process oven running state intelligent adjustment system in the embodiments of the present application can be applied to food, chemical industry, lithium battery pole piece drying and other process scenes requiring high-precision temperature and humidity control. The following will be described by taking the application in a lithium battery pole piece drying process oven as an example.

[0024] In an optional embodiment, Figure 2 A structure diagram of an optional process oven running state intelligent adjustment system provided in the embodiments of the present application is shown in FIG. 2, wherein Figure 2 As shown in FIG. 2, the dynamic prediction module collects the temperature and humidity data of the current environment and the temperature distribution data of the material surface in real time through the high-precision temperature and humidity sensor (S1, precision ±0.1°C, humidity precision ±1%RH) and the infrared sensor array (S2, resolution 0.1°C) installed in the oven. These data are input into the temperature and humidity dynamic prediction model based on the LSTM algorithm as input parameters, combined with the heating power (P_heat), the air supply amount (Q_air) and the environmental temperature and humidity (T_env, RH_env). The model is trained by a large amount of historical process data, can learn the dynamic characteristics of the temperature and humidity change in the oven, and thus outputs the temperature and humidity prediction values (T_pred, RH_pred) in the next 10 minutes, for example, and sends them to the dew point risk avoidance module. For example, in a process oven for drying lithium battery pole pieces, the current temperature is 60°C, the humidity is 30%RH, the heating power is 5 kW, the air supply amount is , the environmental temperature is 25°C, the humidity is 50%RH, the model predicts that the temperature will rise to 60.5°C and the humidity will decrease to 29%RH after 10 minutes.

[0025] After receiving the temperature and humidity prediction values, the dew point risk avoidance module monitors the material surface temperature in real time through the infrared sensor array (S2). Based on the August-Roche-Magnus formula, the dew point temperature (T_dp) of the air in the oven is calculated in real time. When the material surface temperature (T_surface) satisfies T_surface≤T_dp+1°C, that is, the material surface temperature is close to or lower than the dew point temperature, the module will trigger the gradient air supply instruction and send it to the gradient air supply control module. For example, the calculated current dew point temperature is 38.5°C, and the material surface temperature of a certain area decreases to 39.3°C, which is close to the dew point temperature. At this time, the module will trigger the gradient air supply control to prevent the generation of condensed water on the material surface.

[0026] After receiving the temperature and humidity prediction values from the dew point risk avoidance module, the MPC optimization module uses model predictive control for rolling optimization with the goal of minimizing temperature deviation (AT), humidity deviation (ARH), and total energy consumption (P_total). This module uses the temperature and humidity prediction values as reference trajectories, combined with the current hot air parameters (such as supply air temperature, air speed, etc.), to generate the optimal combination of hot air parameters by solving a quadratic programming (QP) problem. For example, in the above lithium battery pole piece drying scenario, the optimized supply air temperature is 61°C, the air speed is 2.2 m / s, and the dehumidification instruction is to increase the fresh air ratio to 30% to ensure that the temperature is maintained at 59.8-60.2°C and the humidity is within the range of 28-32% RH, while minimizing energy consumption.

[0027] The fuzzy-PID adaptive control module dynamically adjusts the proportional coefficient (Kp), integral coefficient (Ki), and derivative coefficient (Kd) of the PID controller based on the temperature control optimization instructions generated by the MPC optimization module, as well as the real-time collected temperature and humidity errors (AT, ARH) and error change rates (ΔΔT, ΔΔRH). The fuzzy rule base contains 25 control rules based on error and error change rate, for example, when the temperature error is large and the error change rate is fast, the proportional coefficient is increased to speed up the response, while the integral coefficient is appropriately reduced to avoid overshoot. The adjusted PID parameters are used to control the heater power and fan speed to achieve precise temperature and humidity control. For example, when the temperature rises rapidly, the fuzzy-PID controller automatically reduces the proportional coefficient and increases the derivative coefficient to prevent temperature overshoot and ensure that the temperature is stable within the set range.

[0028] The gradient air supply control module divides the oven into at least three independent air supply zones (Z1-Z3) based on the gradient air supply instructions and the heater power control signal and fan speed control signal output by the fuzzy-PID adaptive control module. Each zone independently adjusts the air supply temperature and air speed according to the material thickness and moisture content. For example, during the lithium battery pole piece drying process, the inlet zone (Z1) uses high-temperature high-speed air supply (62°C, 3 m / s) for high-moisture materials, the middle zone (Z2) uses medium-temperature medium-speed air supply (60°C, 2.5 m / s) for partially dried materials, and the outlet zone (Z3) uses low-temperature low-speed air supply (58°C, 2 m / s) for nearly dry materials, achieving intelligent adjustment of the process oven operating state, ensuring that the temperature deviation in each zone does not exceed 2°C, while optimizing energy utilization and avoiding excessive heating or cooling.

[0029] In a preferred embodiment, the process oven operating state intelligent adjustment system further comprises a data-driven optimization module for analyzing historical process data to divide typical operating mode, and using a reinforcement learning algorithm to optimize the MPC weight matrix in the MPC optimization module, and outputting the optimized parameters to the MPC optimization module. In the present embodiment, Figure 3 An optional process oven operating state intelligent adjustment method provided by the present embodiment is shown in the flowchart. The data-driven optimization module analyzes historical process data, uses a K-means algorithm to perform cluster analysis on the data, divides three typical operating modes of high condensation risk, stable operation and low energy consumption, and presets corresponding control parameter combinations. For example, through cluster analysis, it is found that in the high condensation risk operating mode, the ambient humidity is high and the material moisture content is large, and the system presets the control parameters as air supply temperature 65°C, air speed 3.5 m / s, and heating power 6 kW; in the stable operation operating mode, the environmental conditions are moderate and the material characteristics are stable, and the preset parameters are air supply temperature 60°C, air speed 2.5 m / s, and heating power 5 kW; in the low energy consumption operating mode, the ambient temperature is high and the material dryness is good, and the preset parameters are air supply temperature 55°C, air speed 2 m / s, and heating power 4 kW. At the same time, the module uses a reinforcement learning algorithm (such as a Q-learning algorithm) to optimize the MPC weight matrix in the MPC optimization module, dynamically adjusts the weight coefficients according to the reward function in the system operation process, so as to realize continuous optimization of the control strategy and further improve the control precision and energy efficiency.

[0030] In a preferred embodiment, the reinforcement learning algorithm uses a Q-learning algorithm, and the reward function is specifically: R=10×(1-∣ΔT∣) + 5×(1-∣ΔRH∣)-0.3×Ptotal; wherein R is the reward value, P_{total} is the total energy consumption, ΔT is the temperature deviation absolute value (the difference between the actual temperature and the set temperature), and ΔRH is the humidity deviation absolute value (the difference between the actual humidity and the set humidity).

[0031] In this embodiment, the reinforcement learning algorithm adopts the Q-learning algorithm, and the reward function is specifically: R=10x(1- |AT| )+5x(1- |ARH| )-0.3xPtotal. Wherein, Ptotal is the total energy consumption, AT is the temperature deviation, ARH is the humidity deviation. During the system operation, the Q-learning algorithm selects the optimal action (i.e. adjusts the weight coefficient in the MPC weight matrix) according to the current state (including temperature and humidity deviation and total energy consumption), and evaluates the effect of the action through the reward function. For example, when the system realizes a temperature deviation of 0.1°C, a humidity deviation of 2%RH, and a total energy consumption of 5 kWh in a certain iteration, the calculated reward value is: R=10x(1-0.1)+5x(1-0.02)-0.3x5=13.4. The system will adjust the weight matrix of MPC according to this reward value, and gradually optimize the control strategy to maximize the long-term reward, so as to reduce the energy consumption while ensuring the control accuracy.

[0032] In a preferred embodiment, the data-driven optimization module adopts the K-means algorithm to cluster the historical process data, divides the high condensation risk, stable operation and low energy consumption three working condition modes, and presets the corresponding control parameter combination.

[0033] In this embodiment, the data-driven optimization module adopts the K-means algorithm to cluster the historical process data, divides the high condensation risk, stable operation and low energy consumption three working condition modes. For example, through the analysis of a large amount of historical data, the system identifies the following three typical working conditions: high condensation risk working condition, when the environmental humidity is high, the material moisture content is large, and the condensate water is easy to form in the oven; stable operation working condition, moderate environmental conditions, stable material characteristics, and smooth system operation; low energy consumption working condition, high environmental temperature, good material dryness, and low energy consumption. For each working condition mode, the system presets the corresponding control parameter combination, so as to quickly adjust the control strategy under different working conditions, improve the system response speed and energy efficiency. For example, in the high condensation risk working condition, the preset control parameters are air supply temperature 65°C, air speed 3.5 m / s, heating power 6 kW; in the stable operation working condition, the preset parameters are air supply temperature 60°C, air speed 2.5 m / s, heating power 5 kW; in the low energy consumption working condition, the preset parameters are air supply temperature 55°C, air speed 2 m / s, heating power 4 kW. Through this data-driven optimization method, the system can better adapt to different working condition changes, improve the control accuracy and energy utilization efficiency.

[0034] In a preferred embodiment, the dew point risk avoidance module comprises: The calculation unit is configured to calculate the dew point temperature in the oven in real time based on the August-Roche-Magnus formula, and monitor the material surface temperature to obtain the real-time material surface temperature. a risk prediction unit configured to determine whether the temperature and humidity prediction value is greater than the real-time material surface temperature, and if so, send a gradient air supply instruction to the gradient air supply control module, or else send the temperature and humidity prediction value to the MPC optimization module.

[0035] In this embodiment, the dew point risk avoidance module includes a calculation unit and a risk prediction unit. The calculation unit calculates the dew point temperature (T dp) of the air in the oven in real time based on the August-Roche-Magnus formula. At the same time, the material surface real-time temperature (T surface) is obtained by monitoring the material surface temperature in real time through the infrared sensor array (S2). The risk prediction unit determines whether the temperature and humidity prediction value is greater than the real-time material surface temperature. If T surface≤T dp+1°C, it indicates that the material surface temperature is close to or lower than the dew point temperature, and there is a risk of condensate formation. At this time, the module sends a gradient air supply instruction to the gradient air control module. Otherwise, if the material surface temperature is higher than the dew point temperature plus 1 degree Celsius, it indicates that the risk of condensate formation under the current working condition is low, and the module sends the temperature and humidity prediction value to the MPC optimization module for subsequent energy consumption optimization and accurate temperature and humidity control. For example, in actual operation, when the calculated dew point temperature is 38.5 degrees Celsius, and the infrared sensor monitors that the material surface temperature in a certain area is 39.3 degrees Celsius, which is only slightly higher than the dew point temperature, the risk prediction unit determines that there is a certain condensation risk, and triggers the gradient air supply control to increase the air supply temperature and air speed in that area to prevent the formation of condensate.

[0036] In a preferred embodiment, the dew point risk avoidance module monitors the material surface temperature in real time through an infrared sensor array to obtain the real-time material surface temperature.

[0037] In this embodiment, the dew point risk avoidance module monitors the material surface temperature in real time through an infrared sensor array (S2). The infrared sensor array is composed of multiple high-precision infrared sensors, which can accurately measure the temperature of different positions on the material surface. These sensors are distributed in different areas of the oven to ensure comprehensive coverage of the material surface. For example, in a large oven, the infrared sensor array may contain dozens of sensors, evenly distributed on the top and sides of the oven, collecting temperature data of the material surface in real time. When it is monitored that the material surface temperature in a local area is lower than the preset dew point temperature range, the system can quickly locate the low-temperature area and trigger local heating compensation. The compensation power is usually set between 50 to 100 watts, and the local heater is used to heat the area to raise the material surface temperature and prevent the accumulation of condensate. This local heating compensation method not only effectively avoids the formation of condensate, but also reduces unnecessary energy waste and improves energy utilization efficiency.

[0038] In a preferred embodiment, the dynamic prediction module comprises a construction unit for establishing a temperature and humidity dynamic prediction model through an LSTM algorithm, and a prediction unit for inputting the current environmental temperature parameters including temperature and humidity, heating power, air supply volume and environmental parameters into the temperature and humidity dynamic prediction model, outputting temperature and humidity prediction values and sending the prediction values to the dew point risk avoidance module.

[0039] In the embodiment, the dynamic prediction module comprises a construction unit and a prediction unit. The construction unit establishes a temperature and humidity dynamic prediction model using a long short-term memory (LSTM) algorithm. LSTM is a special recurrent neural network (RNN) that can effectively process and predict long-term dependencies in sequence data. In the embodiment, the LSTM model is trained through a large amount of historical process data to learn the dynamic characteristics of temperature and humidity changes in the oven. The prediction unit inputs the current environmental temperature parameters (including temperature and humidity, heating power, air supply volume and environmental parameters) into the trained LSTM model, the model outputs temperature and humidity prediction values (T_pred, RH_pred) for the next 10 minutes, and sends the prediction values to the dew point risk avoidance module. For example, in a specific implementation scenario, the current environmental temperature is 60 degrees Celsius, the humidity is 30%, the heating power is 5 kW, the air supply volume is 2 cubic meters per second, the environmental temperature is 25 degrees Celsius, and the humidity is 50%. The model predicts that the temperature will rise to 60.5 degrees Celsius and the humidity will drop to 29% in 10 minutes. This LSTM-based dynamic prediction method can provide accurate reference for subsequent control decisions, ensuring the intelligence and adaptability of the system.

[0040] In a preferred embodiment, the dew point risk avoidance module is further configured to locate a low-temperature area and trigger local heating compensation when it is detected that the surface temperature of the material in the local area is lower than the preset dew point temperature range, and the compensation power is 50-100 W.

[0041] In this embodiment, when the dew point risk avoidance module detects that the local area material surface temperature is lower than the preset dew point temperature range, it can accurately locate the low temperature area and trigger local heating compensation. The compensation power is usually set between 50 to 100 watts, and the local heater is used to heat the area to raise the material surface temperature and prevent the generation of condensed water. For example, in a lithium battery pole piece drying process oven, when the infrared sensor array detects that the material surface temperature of a certain area drops to 39.3 degrees Celsius, close to the calculated dew point temperature of 38.5 degrees Celsius, the system will start a 50 to 100 watt local heater to accurately raise the temperature of the area and prevent the generation of condensed water, while not significantly affecting the temperature balance of other areas. This local heating compensation technology is a supplement and optimization of the dew point risk avoidance module in the core technical solution, which can more accurately control the material surface temperature and prevent local condensed water generation, improving the accuracy and flexibility of temperature control.

[0042] In a preferred embodiment, the zoning strategy of the gradient air supply control module is as follows: the inlet area (Z1) uses high-temperature high-speed air supply (62°C, 3 m / s) for high-moisture materials; the middle area (Z2) uses medium-temperature medium-speed air supply (60°C, 2.5 m / s) for partially dried materials; and the outlet area (Z3) uses low-temperature low-speed air supply (58°C, 2 m / s) for nearly dried materials.

[0043] In this embodiment, the zoning strategy of the gradient air supply control module is as follows: the oven is divided into at least three independent air supply areas, and the air supply temperature and speed of each area are independently adjusted according to the characteristics of the materials and the drying stage they are in. For example, in a lithium battery pole piece drying process oven, the inlet area (Z1) uses high-temperature high-speed air supply (62 degrees Celsius, 3 meters per second) for high-moisture materials, the middle area (Z2) uses medium-temperature medium-speed air supply (60 degrees Celsius, 2.5 meters per second) for partially dried materials, and the outlet area (Z3) uses low-temperature low-speed air supply (58 degrees Celsius, 2 meters per second) for nearly dried materials. This zoning air supply strategy can accurately control the temperature and humidity conditions of each area according to the needs of materials at different drying stages, ensuring uniform heating of materials throughout the drying process, while avoiding energy waste and improving energy efficiency. Through this gradient air supply design, the system can better adapt to changes in material characteristics, optimize drying effect, and reduce energy consumption.

[0044] According to a second aspect of the present application, a process oven operation state intelligent adjustment method is also provided, which comprises: Collecting the current environmental temperature parameters, constructing a temperature and humidity dynamic prediction model through an LSTM algorithm, and outputting temperature and humidity prediction values; The real-time temperature of the material surface is monitored, and it is judged whether the temperature and humidity prediction value is greater than the real-time temperature of the material surface; if yes, a gradient air supply instruction is generated, otherwise, the temperature and humidity prediction value is optimized by MPC; the MPC optimization is based on the temperature and humidity prediction value, and the model predictive control is used for rolling optimization to minimize the temperature deviation, humidity deviation and total energy consumption, so as to obtain a temperature control optimization instruction; According to the temperature control optimization instruction and the real-time temperature and humidity error and error change rate, the PID parameter is dynamically adjusted through a preset fuzzy rule to obtain a heater power control signal and / or a fan speed control signal; According to the gradient air supply instruction and the heater power control signal and / or the fan speed control signal, the oven is divided into a plurality of independent air supply areas, the air speed and air supply temperature of each area are independently adjusted by a variable frequency fan and a segmented heater, and intelligent adjustment of the process oven running state is realized.

[0045] Optionally, specific examples in the embodiment can refer to examples described in the above embodiments, and the embodiment will not be described here.

[0046] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all described as a series of action combinations, but those skilled in the art should know that the present application is not limited to the order of the actions described, because according to the present application, some steps can be performed in other order or at the same time. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0047] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0048] In several embodiments provided by the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic. The division of the units is only a logical function division. There can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different parts can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical or other forms.

[0049] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e. may be located in one place, or may also be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present application.

[0050] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0051] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable memory. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a memory and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: a U disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0052] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by a program instructing relevant hardware, and the program can be stored in a computer readable memory, which can include a flash disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, etc.

[0053] The above is only an exemplary embodiment of the present disclosure, and cannot limit the scope of the present disclosure. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure are still within the scope of the present disclosure. Those skilled in the art will easily think of embodiments of the present disclosure after considering the specification and practicing the disclosure herein. The present application is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or conventional techniques in the art that are not described in the present disclosure. The specification and examples are only considered as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

[0054] Any technical features in the above embodiments can be combined, and for brevity, not every possible combination is described, however, any combination of the technical features is considered to be within the scope of the present specification.

[0055] It is apparent that the above-described embodiments are only the preferred embodiments of the present application and not intended to limit the present application, and modifications, equivalent replacements and improvements made within the spirit and principle of the present application are intended to be included in the scope of the present application.

Claims

1. A temperature and humidity intelligent control method applied to a process oven, characterized in that, Comprise: A dynamic prediction module, a dew point risk avoidance module, an MPC optimization module, a fuzzy-PID adaptive control module, and a gradient air supply control module; The dynamic prediction module is used for collecting current environmental temperature parameters, constructing a temperature and humidity dynamic prediction model through an LSTM algorithm, outputting temperature and humidity prediction values, and sending the temperature and humidity prediction values to the dew point risk avoidance module; The dew point risk avoidance module is used for monitoring real-time temperature of a material surface, judging whether the temperature and humidity prediction values are greater than the real-time temperature of the material surface, and if so, sending a gradient air supply instruction to the gradient air supply control module, otherwise, sending the temperature and humidity prediction values to the MPC optimization module; The MPC optimization module is used for minimizing temperature deviation, humidity deviation, and total energy consumption as an objective based on the temperature and humidity prediction values, adopting model predictive control for rolling optimization to obtain temperature control optimization instructions; The fuzzy-PID adaptive control module is used for dynamically adjusting PID parameters through preset fuzzy rules according to the temperature control optimization instructions and real-time temperature and humidity errors and error change rates to obtain a heater power control signal and / or a fan speed control signal; The gradient air supply control module is used for dividing an oven into multiple independent air supply areas according to the gradient air supply instruction and the heater power control signal and / or the fan speed control signal, independently adjusting air speeds and air supply temperatures of the areas through a variable frequency fan and a segmented heater to realize intelligent adjustment of process oven running states.

2. The process oven operating condition intelligent adjustment system of claim 1, wherein, Further comprise: A data-driven optimization module is used for analyzing historical process data to divide typical working condition modes, and optimizing an MPC weight matrix in the MPC optimization module by using a reinforcement learning algorithm to output optimization parameters to the MPC optimization module.

3. The process oven operating condition intelligent adjustment system of claim 1, wherein, The reinforcement learning algorithm adopts a Q-learning algorithm, and a reward function is specifically: R=10×(1-∣ΔT∣) + 5×(1-∣ΔRH∣)-0.3×Ptotal; Wherein, R is a reward value, Ptotal is total energy consumption, ΔT is a temperature deviation absolute value, and ΔRH is a humidity deviation absolute value.

4. The temperature and humidity intelligent control system of the process oven according to claim 1, wherein, The data-driven optimization module clusters historical process data by using a K-means algorithm, divides three working condition modes of high condensation risk, stable operation, and low energy consumption, and predefines corresponding control parameter combinations.

5. The process oven's temperature and humidity intelligent control system as claimed in claim 1, wherein, The dew point risk avoidance module comprises: A calculation unit is used for calculating a dew point temperature in an oven in real time based on an August-Roche-Magnus formula, monitoring a material surface temperature to obtain a real-time temperature of the material surface; A risk prediction unit is used for judging whether the temperature and humidity prediction values are greater than the real-time temperature of the material surface, and if so, sending a gradient air supply instruction to the gradient air supply control module, otherwise, sending the temperature and humidity prediction values to the MPC optimization module.

6. The temperature and humidity intelligent control system of the process oven according to claim 5, wherein, The dew point risk avoidance module monitors the material surface temperature in real time through an infrared sensor array to obtain the real-time temperature of the material surface.

7. The temperature and humidity intelligent control system of the process oven according to claim 1, wherein, The dynamic prediction module comprises: A construction unit is used for establishing a temperature and humidity dynamic prediction model through an LSTM algorithm; The prediction unit is configured to input the current environmental temperature parameter into a temperature and humidity dynamic prediction model, output a temperature and humidity prediction value, and send the temperature and humidity prediction value to the dew point risk avoidance module, wherein the environmental temperature parameter includes temperature and humidity, heating power, air supply amount, and environmental parameters.

8. The temperature and humidity intelligent control system of the process oven according to claim 1, wherein, The dew point risk avoidance module is further configured to locate a low-temperature area and trigger local heating compensation when it is monitored that the material surface temperature of a local area is lower than a preset dew point temperature range, and the compensation power is 50-100 W.

9. The temperature and humidity intelligent control system of the process oven according to claim 1, wherein, The zoning strategy of the gradient air supply control module is specifically as follows: The inlet area adopts high-temperature and high-speed air supply for high-moisture materials; the middle area adopts medium-temperature and medium-speed air supply for part-dry materials; and the outlet area adopts low-temperature and low-speed air supply for near-dry materials.

10. A temperature and humidity intelligent control method applied to a process oven, characterized in that, The method comprises: collecting current environmental temperature parameters, constructing a temperature and humidity dynamic prediction model through an LSTM algorithm, and outputting a temperature and humidity prediction value; monitoring the real-time temperature of the material surface, and determining whether the temperature and humidity prediction value is greater than the real-time temperature of the material surface; if yes, a gradient air supply instruction is generated, otherwise, the temperature and humidity prediction value is optimized through MPC; the MPC optimization is a rolling optimization through model predictive control based on the temperature and humidity prediction value, with the objective of minimizing temperature deviation, humidity deviation, and total energy consumption, to obtain a temperature control optimization instruction; according to the temperature control optimization instruction and real-time temperature and humidity error and error change rate, PID parameters are dynamically adjusted through a preset fuzzy rule to obtain a heater power control signal and / or a fan speed control signal; according to the gradient air supply instruction and the heater power control signal and / or the fan speed control signal, the oven is divided into multiple independent air supply areas, the air speed and air supply temperature of each area are independently adjusted through a variable frequency fan and a segmented heater, and intelligent adjustment of the temperature and humidity running state of the process oven is realized.

Images