Moisture emission analyzing and processing system
By using a dual-closed-loop nested control architecture, the inner loop responds quickly to production changes, while the outer loop optimizes finished product quality. This solves the problem of unstable humidity control and achieves consistent product quality and adaptive optimization of the production process.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-19
- Publication Date
- 2026-03-27
AI Technical Summary
Existing humidity control strategies cannot respond promptly to changes in operating conditions when faced with complex industrial production environments, resulting in unstable humidity within the drying equipment, which affects product quality consistency and production efficiency.
A dual-closed-loop nested control architecture is adopted. The inner loop control unit makes rapid adjustments based on feedforward parameters and process feedback, while the outer loop monitoring unit dynamically optimizes based on finished product quality feedback, thus constructing an adaptive humidity control system.
It achieves rapid and stable control of the drying environment, ensuring the uniformity of final product quality and the self-adaptability of the production process, and reducing reliance on manual intervention.
Smart Images
Figure CN121739726A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of industrial automation control, and particularly relates to a moisture emission analysis and processing system. BACKGROUND
[0002] In modern industrial production, drying process is an indispensable key link, which is widely used in many industries such as papermaking, textile, coating, building materials and food processing. In these processes, accurate and stable control of the humidity inside the drying equipment is of decisive significance to ensure product quality, improve production efficiency and reduce unit energy consumption. The ideal drying environment requires the humidity to be maintained around a stable target value. Too high humidity will lead to insufficient drying, affecting product performance and subsequent processing; too low humidity will cause excessive energy consumption and lead to product over-drying, brittleness and even damage. At present, the humidity control method commonly used in the industry is a single closed-loop feedback control strategy. This strategy installs a humidity sensor in the exhaust duct of the drying equipment to monitor the internal humidity in real time, compares it with a preset humidity target value, and then uses a PID (Proportion-Integral-Derivative) controller to adjust the opening of the exhaust valve according to the deviation between the two, thereby controlling the amount of moisture emission and indirectly controlling the internal humidity.
[0003] However, the inherent limitations of the above-mentioned single closed-loop feedback control strategy are increasingly prominent when dealing with complex industrial production environments. Industrial drying processes usually have large inertia and pure lag dynamic characteristics, which means there is a significant time delay between control execution and state response. When production conditions change, such as the running speed of the production line increases, the initial moisture content or grammage of the material entering the drying equipment fluctuates, these changes will constitute the main disturbance to the drying system. Single feedback control can only start adjusting after the humidity has deviated from the target value, and its response is after the fact. Due to the hysteresis of the system, this adjustment often cannot timely offset the impact of the disturbance, and is prone to cause the humidity inside the drying equipment to produce large overshoot and sustained oscillation, making it difficult to quickly recover to stability. This instability in the process directly leads to uneven drying of the same batch of products, affecting the consistency of product quality.
[0004] In view of the above related art, the prior art also has a fundamental deficiency in ensuring the long-term stability of the quality of the final product. The ultimate goal of the drying process is to ensure that the physical indicators of the finished product meet the standards accurately, and the humidity in the drying equipment is only an intermediate process variable. The humidity target value tracked by the single feedback control system is usually a fixed value set in advance through offline experiments or experience under ideal working conditions. However, in actual continuous production, due to the long-term influence of various unmodeled factors such as differences in physical properties between batches of raw materials, seasonal changes in environmental temperature and humidity, and slight drifts in equipment performance, this fixed humidity target value often deviates from the true optimal process point. This results in systematic overall drift in the residual moisture content of the finished product produced, which cannot meet the quality standards, even if the process humidity is stably controlled at the set value. The prior art lacks a closed-loop optimization mechanism that can automatically correct the target value of the intermediate process variable based on the actual feedback of the quality of the final product, resulting in a serious dependence on experienced operators for periodic manual intervention and calibration of process parameters. This not only responds in a timely manner, but also increases the scrap rate, and limits the automation level and self-adaptability of the production process to different working conditions. SUMMARY
[0005] The purpose of the present application is to provide a moisture emission analysis and processing system, which solves the problem that the existing control scheme lacks a closed-loop optimization mechanism guided by the quality of the final product, resulting in the inability to adaptively adjust the process set point to cope with slow changes such as differences in raw material batches, and thus making it difficult to ensure the long-term stability and batch-to-batch consistency of the quality of the final product.
[0006] To achieve the above purpose, the present application is implemented by the following technical scheme: a moisture emission analysis and processing system, comprising: An automatic control valve configured at the exhaust port of the drying equipment; A feedforward parameter acquisition module for acquiring at least one feedforward parameter of the material entering the drying equipment in real time; A process feedback parameter acquisition module for acquiring the actual humidity in the drying equipment in real time; and a finished product quality parameter acquisition module for acquiring the final residual moisture content of the finished product leaving the drying equipment in real time; And a central processing module. The central processing module is electrically connected with the feedforward parameter acquisition module, the process feedback parameter acquisition module, the finished product quality parameter acquisition module and the automatic control valve, respectively.
[0007] The central processing module comprises: An outer loop supervisory unit configured to generate or update a humidity target value based on the final residual moisture content of the finished product and a preset standard finished product moisture content target value; and an inner loop control unit configured to receive the humidity target value generated by the outer loop supervising unit, and generate a final control instruction based on the feedforward parameter, the actual humidity and the humidity target value, to send to the automatic control valve.
[0008] With the above arrangement, the system builds a double closed loop nested control architecture. The outer loop supervising unit dynamically optimizes the core parameter (humidity target value) of process control according to the final product quality; The inner loop control unit then quickly and stably executes the optimized target value based on the feedforward parameter and process feedback. The two work together to enable the system to pre-adjust according to the changes in production conditions, while using process feedback for accurate correction, and finally realize adaptive optimization of process targets through closed-loop feedback of finished product quality, thereby suppressing process disturbances, stabilizing the drying environment, and ensuring the uniformity of the final product quality.
[0009] Preferably, the inner loop control unit is specifically configured to: calculate a basic valve opening degree based on the feedforward parameter; calculate a correction valve opening degree based on the deviation between the actual humidity and the humidity target value; and superimpose the basic valve opening degree and the correction valve opening degree to generate the final control instruction. This composite control method uses the feedforward part to quickly respond to major disturbances, and uses the feedback part to eliminate residual deviations, effectively overcoming the limitations of single control mode.
[0010] In one specific embodiment, the inner loop control unit calculates the predicted moisture generation amount based on a preset moisture generation rate prediction model and the feedforward parameter, and determines the basic valve opening degree based on the predicted moisture generation amount. The moisture generation rate prediction model can be represented by the following functional relationship: Q pred (t) = f(V(t), W(t), T(t)); where Q pred (t) is the predicted moisture generation amount at time t, V(t), W(t) and T(t) are the feedforward parameters at time t.
[0011] Further, the feedforward parameter includes the production line running speed V(t), the wet basis weight of the material W(t) and the working temperature in the drying equipment T(t). The moisture generation rate prediction model is specifically used to represent the functional relationship between the predicted moisture generation amount and the production line running speed, the wet basis weight of the material and the working temperature.
[0012] In one specific embodiment, the inner loop control unit uses a PID controller to determine the opening degree of the correction valve based on the deviation between the actual humidity and the target humidity value. The control law of the PID controller can be expressed as: Among them, A fb (t) represents the corrected valve opening at time t; e H (t) represents the deviation between the actual humidity and the target humidity value at time t; K p K i K d These are the proportional, integral, and derivative coefficients of the PID controller, respectively.
[0013] Preferably, the outer ring monitoring unit is specifically configured to: compare the final residual moisture content of the finished product with the preset standard finished product moisture content target value to obtain the finished product quality deviation; And based on the finished product quality deviation, the humidity target value is adjusted through a preset monitoring and control algorithm to generate an updated humidity target value.
[0014] Furthermore, the supervisory control algorithm is an integral controller, which is used to smoothly adjust the humidity target value based on the cumulative value of the finished product quality deviation. The algorithm of the integral controller can be expressed as: H sp (t)=H sp_initial +K s *∫e MC (t)dτ; Among them, H sp (t) represents the target humidity value updated at time t; H sp_initial This is the initial setting for the humidity target value; e MC (t) represents the finished product quality deviation at time t; K s This is the supervisory control gain coefficient.
[0015] In one specific embodiment, the feedforward parameter acquisition module includes: a speed sensor for acquiring the production line operating speed; A non-contact basis weight detector for obtaining the wet basis weight of materials; And a temperature sensor group for obtaining the operating temperature inside the drying equipment.
[0016] In one specific embodiment, the process feedback parameter acquisition module includes a high-precision humidity sensor installed in the exhaust channel of the drying equipment.
[0017] In one embodiment, the finished product quality parameter acquisition module includes a non-contacting moisture content detector installed downstream of the outlet of the drying device.
[0018] The present application provides a moisture emission analysis processing system, which has the following advantages: 1. The present application sets an inner ring control unit, and combines predictive control based on feedforward parameters with feedback control based on actual humidity, so that the system can make preliminary adjustment when production conditions (such as production line speed and material wet basis weight) change, to pre-offset major disturbances, and at the same time eliminate residual deviations by using feedback links. The composite control structure effectively shortens the response time of the system, suppresses the large fluctuations in humidity caused by control lag, and thus maintains the stability of the drying environment.
[0019] 2. The present application sets an outer ring supervision unit, and takes the residual moisture content of the final product as the highest control target, to build a closed-loop optimization path based on the quality of the finished product. When the quality of the finished product drifts for a long time due to unmodeled factors such as batch differences of raw materials and changes in environmental temperature and humidity, the outer ring supervision unit can automatically and continuously correct the humidity target value of the inner ring, so as to drive the entire production process to adaptively adjust towards the direction of ensuring the quality of the final product, and improve the uniformity of product quality between different batches.
[0020] 3. The present application integrates the outer ring supervision unit for process optimization with the inner ring control unit for process stability, and changes the process parameter setting step relying on manual experience in the past into a self-optimizing process that can run autonomously. In the face of different product specifications or production tasks, the system can autonomously seek optimization and stabilize at the corresponding optimal process parameter point, reducing the dependence on manual intervention of operators and enhancing the adaptability of the production process to different working conditions. BRIEF DESCRIPTION OF DRAWINGS
[0021] Figure 1 The figure is a structural block diagram of the central processing system of the present application; Figure 2 The figure is a physical deployment schematic diagram of the drying production system of the present application.
[0022] Among them, 10, feedforward parameter acquisition module; 20, process feedback parameter acquisition module; 30, finished product quality parameter acquisition module; 40, central processing module; 41, outer ring supervision unit; 42, inner ring control unit; 50, automatic control valve arranged at the exhaust port of the drying device. DETAILED DESCRIPTION
[0023] The following will be described in detail in combination with the accompanying Figure 1 - the accompanying Figure 2 The present application will be further described in detail.
[0024] The application provides a moisture emission analysis processing system.
[0025] Referring to the drawings Figure 1 , Figure 1 is a structure block diagram of a moisture emission analysis processing system according to an embodiment of the application. The application provides a moisture emission analysis processing system applied to a drying device, which can include: a feedforward parameter acquisition module 10, a process feedback parameter acquisition module 20, a finished product quality parameter acquisition module 30, a central processing module 40, and an automatic control valve 50 arranged at an exhaust port of the drying device.
[0026] The feedforward parameter acquisition module 10, the process feedback parameter acquisition module 20, and the finished product quality parameter acquisition module 30 are electrically connected to the input end of the central processing module 40, for transmitting real-time collected data signals to the central processing module 40. The output end of the central processing module 40 is electrically connected to the automatic control valve 50, for sending control instructions to the automatic control valve 50.
[0027] The central processing module 40 is logically divided into an outer loop supervisory unit 41 and an inner loop control unit 42 by software or hardware inside the central processing module 40.
[0028] The outer loop supervisory unit 41 is connected to the signal output end of the finished product quality parameter acquisition module 30. The outer loop supervisory unit 41 is configured to receive the final residual moisture content of the finished product collected by the finished product quality parameter acquisition module 30, and compare and calculate the residual moisture content with a preset standard finished product moisture content target value stored in the outer loop supervisory unit 41, so as to generate or periodically update a humidity target value.
[0029] The data input end of the inner loop control unit 42 is connected to the signal output end of the feedforward parameter acquisition module 10 and the process feedback parameter acquisition module 20, and receives the humidity target value generated by the outer loop supervisory unit 41. The inner loop control unit 42 is configured to generate a final control instruction by a preset control algorithm based on the received feedforward parameter, actual humidity, and humidity target value.
[0030] The instruction output end of the inner loop control unit 42 is connected to the automatic control valve 50, for sending the generated final control instruction to the automatic control valve 50, so as to drive the automatic control valve 50 to adjust the valve opening degree, thereby controlling the moisture emission amount of the drying device.
[0031] The system structure forms a double-closed-loop nested control loop. The outer loop supervisory unit 41 constitutes a supervisory loop based on the final product quality, for process target optimization. The inner loop control unit 42 constitutes an execution loop based on production process parameters and real-time feedback, for process rapid stability. The two loops interact and work cooperatively through the internal parameter of the humidity target value.
[0032] The feedforward parameter acquisition module 10, in this embodiment, is specifically composed of a group of sensors deployed at different positions of the production line. The feedforward parameter acquisition module includes a speed sensor for acquiring the running speed of the production line, which can be an incremental rotary encoder installed at the shaft end of the main drive roller of the drying equipment production line for outputting pulse signals proportional to the production line speed. The feedforward parameter acquisition module also includes a non-contact basis weight detector, such as a beta-ray or near-infrared detector, installed at a position after the material is impregnated with glue and before entering the drying equipment inlet for online measurement of the wet basis weight of the material per unit area. The feedforward parameter acquisition module further includes a temperature sensor group composed of multiple PT100 platinum thermal resistors distributed at different temperature zones inside the drying equipment for measuring the real-time temperature of each zone and obtaining an overall working temperature of the drying equipment by calculating the weighted average value thereof.
[0033] The process feedback parameter acquisition module 20 is specifically a high-precision humidity sensor. In this embodiment, the feedforward parameter acquisition module is a high-temperature-resistant capacitive humidity sensor installed at the center position of a straight pipe section of the total exhaust duct of the drying equipment, which can collect the fully mixed representative exhaust humidity. The sensor is used to measure the actual humidity value inside the drying equipment in real time and continuously and convert the value into a standard electrical signal output.
[0034] The finished product quality parameter acquisition module 30 is specifically a non-contact finished product moisture content detector. In this embodiment, the detector is a near-infrared (NIR) moisture analyzer installed at the end of the production line at a position after the finished product leaves the drying equipment and passes through a cooling zone. This installation position ensures the relative stability of the surface temperature of the measurement object, thereby obtaining an accurate final residual moisture content of the finished product. The detector continuously scans and measures the finished product on the production line.
[0035] The automatic control valve 50 is specifically an electric or pneumatic proportional regulating valve with a precise positioner. The automatic control valve is installed at the total exhaust port of the drying equipment, and the valve opening degree thereof can be continuously and linearly adjusted according to the input control signal (such as a 4-20 mA current signal). The valve positioner functions to receive instructions from the central processing module 40 and accurately drive the valve to the specified opening degree to achieve precise control over the moisture discharge amount.
[0036] The central processing module 40 can have an industrial programmable logic controller (PLC) or an embedded industrial personal computer (IPC) as the hardware platform. In this embodiment, the aforementioned outer ring supervision unit 41 and inner ring control unit 42 are not independent physical hardware, but functional units realized through preset program logic running on the processor of the PLC or industrial personal computer. The two functional units exchange data through internal data registers or variables.
[0037] The inner loop control unit 42 operates in a high frequency control cycle (e.g., once per second). In one control cycle, it first receives the current production line running speed V(t), the material wet basis weight W(t), and the drying equipment working temperature T(t) from the feedforward parameter acquisition module 10. Based on these feedforward parameters, the inner loop control unit 42 calculates the predicted moisture generation amount Q pred (t) by a pre-set moisture generation rate prediction model. In this embodiment, the prediction model can be a multivariate linear regression model, whose specific function form is as follows: Q pred (t) = k1*V(t) + k2*W(t) + k3*T(t) + C; where k1, k2, k3 are the regression coefficients of the model, respectively representing the sensitivity of the moisture generation amount to the production line running speed, the material wet basis weight, and the working temperature; and C is a constant bias term. These coefficients and the constant term are fixed values obtained by offline analysis and fitting of historical production data.
[0038] Specifically, the calculation process of the above formula implicitly unifies the dimensions in physical meaning. In the internal calculation of each single term, the unit of the input variable is absorbed and converted by the unit carried by its corresponding coefficient k. For example, if the unit of the predicted moisture generation amount Q pred (t) is kg / h, then the coefficient k1 implicitly has a unit conversion relationship of [(kg / h) / (m / min)], so that the result unit of the k1*V(t) term is converted to kg / h; the coefficient k2 implicitly has a unit conversion relationship of [(kg / h) / (kg / m 2 )], so that the result unit of the k2*W(t) term is converted to kg / h; and the coefficient k3 implicitly has a unit conversion relationship of [kg / (h°C)], so that the result unit of the k3*T(t) term is converted to kg / h. Only when all single terms in the formula are converted to the same mass flow unit (such as kg / h), does the addition operation between them have physical meaning. In engineering practice, when multivariate linear regression analysis is performed based on historical production data containing a large amount of V, W, T, and corresponding actual moisture emission amount Q, the calculated k1, k2, k3 values naturally contain the above unit conversion relationship, thereby ensuring the physical consistency of the model.
[0039] After calculating the predicted moisture generation amount Q pred (t), the inner loop control unit 42 converts this moisture generation amount to a basic valve opening A ff (t). This functional relationship is stored in the memory of the central processing module 40, and can be in the form of a look-up table or a polynomial function, which establishes a non-linear correspondence between the exhaust flow rate and the valve opening.
[0040] At the same time, the inner loop control unit 42 receives the current actual moisture content H actual (t) from the process feedback parameter acquisition module 20, and receives the current moisture content target value H sp (t) from the outer loop supervising unit 41. By taking the difference between the two, the moisture content deviation e H (t) is calculated. e H (t) = H sp (t) - H actual (t). This moisture content deviation e H (t) is input into a PID controller for calculating the correction valve opening A fb (t). The control law of the PID controller is implemented in discrete form in the central processing module 40 in this embodiment, and its prototype formula in the continuous time domain is: where K p , K i , and K d are the proportional, integral, and derivative gain coefficients of the PID controller, respectively, and their values are adjusted through on-site process debugging after the system is deployed. The proportional term is used to respond to the current deviation, the integral term is used to eliminate long-term static deviation, and the derivative term is used to suppress the trend of deviation change to prevent overshoot.
[0041] Finally, the inner loop control unit 42 linearly superimposes the calculated basic valve opening A ff (t) and the correction valve opening A fb (t) to generate the final control instruction A final (t): A final (t) = A ff (t) + A fb (t). This final control instruction A final (t) (a value representing the percentage of the valve opening) is converted into a standard control signal and sent to the automatic control valve 50.
[0042] The outer loop supervising unit 41 operates in a low-frequency supervising cycle (e.g., once per minute or per production batch). It first receives the measured, stable final residual moisture content of the finished product MC actual (t) in a period of time from the finished product quality parameter acquisition module 30, and compares it with the internally stored standard finished product moisture content target value MC stdBy comparison, the quality deviation of the finished product, e, is obtained. MC (t): e MC (t)=MC std -MC actual (t); Subsequently, the outer loop monitoring unit 41, through an integral controller, adjusts the humidity target value H based on the cumulative effect of the finished product quality deviation. sp (t) is updated. The update algorithm is as follows: H sp (t)=H sp_initial +K s *∫e MC (t)dτ; Among them, H sp (t) is the updated humidity target value; H sp_initial This is the initial setting for the humidity target value; dτ is an infinitesimal time interval; K s This is the monitoring gain coefficient, set at a relatively small value to ensure that the adjustment process of the humidity target value is smooth and slow. The function of this integral element is that as long as there is a persistent, non-zero deviation in the finished product quality, the humidity target value will be continuously fine-tuned until the deviation is eliminated, i.e., e. MC (t) approaches zero.
[0043] It should be noted that the aforementioned outer-loop monitoring unit 41 utilizes the monotonic correspondence between ambient humidity and finished product moisture content for negative feedback adjustment, rather than relying on a pre-set specific function mapping model between humidity and moisture content. In actual production, fluctuations in the physical properties of materials can cause the mapping relationship between humidity and moisture content to drift, making it difficult for a fixed model to adapt to all operating conditions. This embodiment addresses this by setting a relatively small monitoring gain coefficient K. s This method limits the rate of change of the target humidity value, achieving gradual correction of the target value through the cumulative characteristics of the integral element. This approach avoids significant abrupt changes in the target humidity value due to short-term fluctuations or large deviations in the finished product's moisture content, thus preventing over-drying or product embrittlement caused by sudden drops in ambient humidity. It also ensures the stability of the control process even under conditions where material properties are unknown.
[0044] Updated humidity target value H sp (t) is transmitted to the inner loop control unit 42 as the reference target for its next control cycle.
[0045] Before the system starts up, initial settings are first performed. The operator uses a human-machine interface (HMI) connected to the central processing module 40 to input the initial humidity target value H required for the current production process.sp_initial and the standard finished product moisture content target value MC std Input and store into the internal memory of the central processing module 40.
[0046] After the system enters the steady state operation phase, its workflow consists of two control cycles running in parallel at different time scales.
[0047] The first control cycle is the high-frequency process stabilization cycle executed by the inner loop control unit 42. Within one inner loop control period, it performs the following steps: Step one, data acquisition: the inner loop control unit 42 synchronously acquires the current production line running speed V(t), material wet basis weight W(t) and working temperature T(t) from the feedforward parameter acquisition module 10, and acquires the current actual humidity H actual (t) from the process feedback parameter acquisition module 20 through its data input end.
[0048] Step two, target acquisition: the inner loop control unit 42 reads the updated humidity target value H sp (t) provided by the outer loop supervising unit 41 from its internal data area. In the first period after system initialization, this value is H sp_initial .
[0049] Step three, instruction calculation: the inner loop control unit 42 calculates the basic valve opening A ff (t) based on the acquired feedforward parameters, and calculates the correction valve opening A fb (t) based on the deviation between the actual humidity and the humidity target value. Then, the two are superimposed to generate the final control instruction A final (t).
[0050] Step four, instruction execution: the inner loop control unit 42 sends the final control instruction to the automatic control valve 50 through its instruction output end to drive its valve opening to adjust to the corresponding position.
[0051] The above high-frequency process stabilization cycle is repeatedly executed with a short fixed period (e.g. 1 second) to achieve rapid response and stable control of the humidity in the drying equipment.
[0052] The second control cycle is the low-frequency process optimization cycle executed by the outer loop supervising unit 41. Within one outer loop supervising period, it performs the following steps: Step one, data acquisition: the outer loop supervising unit 41 acquires the filtered or averaged finished product final residual moisture content MC actual (t) measured in the past supervising period from the finished product quality parameter acquisition module 30.
[0053] Step two, deviation calculation: outer loop supervising unit 41 compares the collected residual moisture content with the preset standard product moisture content target value, and calculates the product quality deviation e MC (t).
[0054] Step three, target updating: outer loop supervising unit 41 inputs the product quality deviation into the supervising control algorithm inside it, and calculates an updated humidity target value.
[0055] Step four, data writing: outer loop supervising unit 41 writes the updated humidity target value H sp (t) into the internal data area of central processing module 40, for inner loop control unit 42 to call in the subsequent control period. The above-mentioned low-frequency process optimization cycle is repeatedly executed in a longer fixed period (for example, 5 minutes) to achieve smooth and stable optimization of the core process parameters.
[0056] Through the above workflow, inner loop control unit 42 and outer loop supervising unit 41 work in different time scales with humidity target value H sp (t) as the data interface. The inner loop is responsible for quickly suppressing process disturbances and accurately tracking the target on a second time scale, while the outer loop adaptively adjusts the target on a minute time scale according to the feedback of the final product quality, so that the entire system realizes autonomous optimization of the process target on the basis of ensuring process stability.
[0057] Referring to the accompanying drawings Figure 1 , when a specific production task starts, the drying process requirements of a certain specification of decorative paper are set. At this time, through the human-computer interaction interface, the standard product moisture content target value is set to 6.5%, and the initial humidity target value is set to 75.0% relative humidity (RH). These parameters are written into the storage area of central processing module 40. At the same time, the pre-wetting generation rate prediction model and the gain coefficients K p ,K i ,K d of the PID controller have also been loaded into inner loop control unit 42.
[0058] The production line starts, and the material enters the drying equipment at a speed of 100 meters / minute. Feedforward parameter acquisition module 10 sends the collected production line running speed V(t) = 100, material wet basis weight W(t), and drying equipment working temperature T(t) to inner loop control unit 42 in real time. Inner loop control unit 42 calculates the basic valve opening A ff (t) according to the above parameters. At the same time, actual humidity H actual (t) measured by process feedback parameter acquisition module 20 is compared with the current humidity target value H sp (t) (which is 75.0% RH at this time), and the PID controller calculates the corrected valve opening Afb (t) Final control command A final (t) is calculated by A ff (t) is calculated by A fb (t) is calculated by A
[0059] Suppose during operation, the production command requires the production line speed to be increased to 120 m / min. The speed sensor immediately transmits the change V(t) = 120 to the inner loop control unit 42. Before the actual humidity changes significantly due to the increased amount of moisture evaporation, the term V(t) in the humidity generation rate prediction model increases, resulting in a predicted amount of humidity generation Q pred (t) instantaneously increases, and the calculated base valve opening A ff (t) also increases accordingly. This increment is superimposed into the final control command A final (t), so that the opening of the automatic control valve 50 is pre-adjusted to discharge the expected extra humidity. Therefore, the humidity disturbance caused by the production line speed-up is largely offset by the feedforward control, and the actual humidity H actual (t) fluctuates within a small range, which is fine-tuned by the PID feedback loop subsequently.
[0060] Subsequently, a batch of raw paper is replaced on the production line. Due to the difference in fiber structure of this batch of raw paper, its water-holding performance is slightly different from the previous batch. Although the inner loop control unit 42 still accurately maintains the humidity in the drying equipment at 75.0% RH, the finished product quality parameter acquisition module 30 at the end of the production line begins to continuously detect that the final residual moisture content MC actual (t) of the finished product stabilizes at around 6.8%, which is higher than the target value of 6.5%.
[0061] After a complete supervision period (e.g., 5 minutes), the outer loop supervision unit 41 collects and confirms this stable quality deviation. It calculates the finished product quality deviation e MC (t) = 6.5% - 6.8% = -0.3%. Since the deviation is negative, the integral controller in the outer loop supervision unit 41 begins to accumulate this negative deviation value. The result of updating the integral term of the algorithm will be a negative value, resulting in the updated humidity target value H sp (t) being adjusted downward.
[0062] This updated humidity target value 74.896 RH is written to the internal data area of the central processing module 40. In the next inner loop control period, the inner loop control unit 42 will take 74.8% RH as the new control target. Since the target value is lowered, the humidity deviation e H(t) will become negative, the output of the PID controller will drive the automatic control valve 50 to further increase its opening, to reduce the actual humidity inside the oven.
[0063] In the following several supervising cycles, as long as the final residual moisture content of the finished product MC actual (t) is still higher than 6.5%, the outer loop supervising unit 41 will continuously and slowly reduce the humidity target value H sp (t) until the humidity inside the drying equipment is reduced to a new equilibrium point (e.g. 74.5% RH) at which the final residual moisture content of the finished product is stabilized around 6.5%. At this time, the quality deviation of the finished product e MC (t) approaches zero, the integral term output of the outer loop supervising unit 41 no longer changes, and the humidity target value is stabilized at 74.5% RH, a new set value that is adapted to the current material characteristics.
[0064] Through the above description of the scenario, this embodiment demonstrates how the system responds to fast process parameter changes through the inner loop control unit, and responds to slow quality drifts caused by material characteristics, etc. through the outer loop supervising unit, so as to achieve accurate control of moisture discharge and self-adaptive guarantee of the final product quality without human intervention.
[0065] Referring to the accompanying drawings Figure 1 In an alternative embodiment, the moisture generation rate prediction model employed in the inner loop control unit 42 is not limited to the aforementioned multiple linear regression model. The model can be a mechanism model established based on the principles of thermodynamics and mass transfer, which solves a system of differential equations to predict the moisture generation rate by inputting the physical parameters of the material (such as specific heat capacity, moisture diffusion coefficient) and process parameters.
[0066] In another embodiment, the model can be a nonlinear data-driven model, for example, an artificial neural network or a support vector regression model trained on historical production data. The input of the model is the feedforward parameter V(t), W(t), T(t), and the output is the predicted moisture generation Q pred (t), which can capture and characterize more complex nonlinear relationships in the production process.
[0067] Similarly, the supervisory control algorithm used in the outer loop supervising unit 41 to update the humidity target value is not limited to the aforementioned single integral controller. In another embodiment, the algorithm can be a proportional-integral controller. Compared to pure integral control, the added proportional term can adjust the humidity target value H MCThe size of (t) generates an immediate adjustment of the target value proportional to the deviation, so that the convergence speed of the target value is accelerated when a larger quality deviation occurs. In yet another embodiment, the algorithm can be a fuzzy logic controller. The controller takes the product quality deviation and its rate of change as input variables, and infers an adjustment of the target value through a series of pre-set fuzzy rules based on expert experience (e.g. "if the deviation is large and continuously increasing, then decrease the target value by a large amount"). This approach is suitable for handling non-linear or difficult-to-model process optimization.
[0068] For the hardware implementation of the central processing module 40, in addition to the aforementioned PLC or industrial computer, for large-scale application scenarios involving multiple production lines, the functions of the module can be integrated into the controller or engineer station of a distributed control system. In such an architecture, each acquisition module and actuator is connected to the system bus as a field node, while the control logic of the outer and inner loops is calculated and managed uniformly in the central controller. In addition, the functions of the central processing module 40 can also be realized by a combination of an edge computing device and a cloud server, where the inner loop control logic with high real-time requirements is run on the edge computing device on site, while the outer loop supervisory optimization logic requiring a large amount of data analysis and model training is run on the cloud server.
[0069] The electrical connection and data communication between the feedforward parameter acquisition module 10, the process feedback parameter acquisition module 20, the product quality parameter acquisition module 30, the central processing module 40, and the automatic control valve 50 can be achieved through standard industrial fieldbus technology in this embodiment. It can also be achieved through an industrial Ethernet-based communication protocol, such as OPC UA, to ensure the speed, stability, and interoperability of data transmission.
[0070] In addition, the specific sensor types that make up each acquisition module can also be selected according to actual working conditions and measurement accuracy requirements. For example, the non-contact product moisture content detector in the product quality parameter acquisition module 30 can also be a microwave moisture meter in addition to the near-infrared type. The speed sensor in the feedforward parameter acquisition module 10 can also be other types of speed measurement devices, such as a laser Doppler velocimeter. The selection of these different technical implementations does not deviate from the core concept of the double-loop nested self-optimizing control system constructed by the present application.
Claims
1. A moisture emission analysis and treatment system, characterized in that, include: An automatic control valve configured at the exhaust port of the drying equipment; A feedforward parameter acquisition module is used to acquire at least one feedforward parameter of the material entering the drying equipment in real time; A process feedback parameter acquisition module is used to acquire the actual humidity inside the drying equipment in real time; A finished product quality parameter acquisition module is used to acquire the final residual moisture content of the finished product leaving the drying equipment in real time; And a central processing module, which includes: The outer ring monitoring unit is configured to generate an updated humidity target value based on the final residual moisture content of the finished product and a preset standard finished product moisture content target value. The inner loop control unit is configured to receive the humidity target value generated by the outer loop monitoring unit, and generate a final control command based on the feedforward parameters, the actual humidity, and the humidity target value, and send it to the automatic control valve.
2. The moisture emission analysis and treatment system according to claim 1, characterized in that, The inner ring control unit is specifically configured as follows: The basic valve opening is calculated based on the aforementioned feedforward parameters; The corrected valve opening is calculated based on the deviation between the actual humidity and the target humidity value. The basic valve opening degree and the correction valve opening degree are superimposed to generate the final control command.
3. The moisture emission analysis and treatment system according to claim 2, characterized in that, The inner loop control unit uses a preset moisture generation rate prediction model and calculates the predicted moisture generation amount based on the feedforward parameters, and then determines the basic valve opening based on the predicted moisture generation amount.
4. The moisture emission analysis and treatment system according to claim 3, characterized in that, The feedforward parameters include the production line operating speed, the wet basis weight of the material, and the operating temperature inside the drying equipment. The moisture generation rate prediction model is used to characterize the functional relationship between the predicted moisture generation and the production line operating speed, the wet basis weight of the material, and the operating temperature.
5. The moisture emission analysis and treatment system according to claim 2, characterized in that, The inner loop control unit calculates the opening degree of the correction valve based on the deviation between the actual humidity and the target humidity value using a PID controller.
6. The moisture emission analysis and treatment system according to claim 1, characterized in that, The outer ring monitoring unit is specifically configured as follows: The final residual moisture content of the finished product is compared with the preset standard finished product moisture content target value to obtain the finished product quality deviation. And based on the finished product quality deviation, the humidity target value is adjusted through a preset monitoring and control algorithm to generate an updated humidity target value.
7. The moisture emission analysis and treatment system according to claim 6, characterized in that, The supervisory control algorithm is an integral controller, which is used to smoothly adjust the humidity target value based on the cumulative value of the finished product quality deviation.
8. The moisture emission analysis and treatment system according to claim 1, characterized in that, The feedforward parameter acquisition module includes: A speed sensor used to acquire the operating speed of the production line; A non-contact basis weight detector for obtaining the wet basis weight of materials; And a temperature sensor group for obtaining the operating temperature inside the drying equipment.
9. A moisture emission analysis and treatment system according to claim 1, characterized in that, The process feedback parameter acquisition module includes a high-precision humidity sensor installed in the exhaust channel of the drying equipment.
10. A moisture emission analysis and treatment system according to claim 1, characterized in that, The finished product quality parameter acquisition module includes a non-contact finished product moisture content detector installed downstream of the outlet of the drying equipment.