Multi-Sensor-Based Monitoring Method for Straw Drying Status

Abstract

The present invention discloses a method for monitoring the drying state of straw based on multi-sensors, specifically relating to the field of monitoring and control of the drying state of straw, and is used to solve the problems of low drying efficiency, inconsistent quality and inaccurate control caused by uneven bulk density and poor air permeability. By modeling density and permeability, the drying bottleneck areas are accurately identified; using the regionalized drying demand model, the drying parameters of each region are accurately predicted; combined with intelligent optimization algorithms and real-time feedback control, the dynamic adjustment and precise matching of the drying process are realized; thereby significantly improving the efficiency and uniformity of straw drying, avoiding the problems of insufficient or excessive drying caused by density and permeability differences, and at the same time effectively reducing the energy consumption and safety risks during the drying process, ensuring the consistency of drying quality and the reliability of equipment operation.

Description

Technical Field

[0001] The present invention relates to the field of straw drying state monitoring and control. More specifically, the present invention relates to a method for monitoring the drying state of straw based on multi-sensors. Background Art

[0002] Currently, straw drying mainly relies on traditional drying equipment and methods, such as natural drying, mechanical drying, and hot air drying. These methods usually rely on unified drying parameters (such as temperature, humidity, and air flow rate) to process the entire straw pile. However, in order to ensure drying efficiency and product quality, some advanced technologies have introduced sensor monitoring systems, mainly monitoring the overall temperature, humidity, and moisture content. These sensors are usually distributed at different positions of the drying equipment, and the drying conditions are adjusted through real-time data collection to ensure that the entire straw pile achieves the expected drying effect. In addition, some intelligent drying systems also incorporate automatic control algorithms that can dynamically adjust the drying parameters according to the sensor data to improve the efficiency and consistency of the drying process.

[0003] However, the existing technologies have significant deficiencies in dealing with the problems of uneven straw stacking density and distribution. First, traditional sensor monitoring systems mostly focus on the overall environmental parameters and lack accurate measurement of the internal density distribution of straw, resulting in an inability to comprehensively understand the drying state of each region inside the straw pile. Second, the existing systems usually adopt unified drying parameters, ignoring the need for different drying strategies due to differences in filling density in different regions, resulting in over-drying or under-drying in some regions and unable to meet the consistent requirements of the demand side for drying uniformity and quality. In addition, the existing technologies are relatively simple in data processing and analysis and cannot fully utilize the multi-sensor data to deeply analyze and model the overall structural density of the straw. These deficiencies not only affect drying efficiency and product quality but also may increase energy consumption and operating costs.

[0004] To solve the above problems, a technical solution is provided now. Summary of the Invention

[0005] To overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a multi-sensor-based straw drying state monitoring method. The multi-sensor-based straw drying state monitoring method can accurately construct a three-dimensional density distribution model of straw accumulation, and comprehensively master the density and airflow characteristics in the drying environment by combining dynamic air permeability analysis. Further, by integrating the initial moisture content and environmental parameters, a regional drying demand model is established, and the target drying curve for each region is generated to achieve accurate prediction and demand matching of the drying process. Advanced global optimization algorithms, such as genetic algorithms or particle swarm optimization, are used to dynamically adjust the temperature and airflow velocity in each region to ensure the optimal configuration of drying parameters. At the same time, the intelligent control module allocates and executes the optimal drying settings in real time to ensure the efficiency and uniformity of the drying process. This method effectively solves the problems of low drying efficiency and inconsistent quality caused by uneven straw accumulation density in traditional drying technologies, and significantly improves the overall efficiency of the drying process and the product quality, so as to solve the problems proposed in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The multi-sensor-based straw drying state monitoring method includes the steps of:

[0008] S1. Collect data from density sensors and airflow sensors, and generate a three-dimensional density distribution model of straw accumulation through multi-point interpolation and data fusion to locate high-density regions and low-density regions.

[0009] S2. Inject a fixed amount of gas and monitor the airflow response data, and obtain the permeability distribution map through dynamic modeling by monitoring the airflow response data.

[0010] S3. Combine the density distribution, permeability characteristics and initial moisture content data to establish a regional drying demand model and generate the target drying curve for each region.

[0011] S4. Adjust the corresponding drying parameters based on the target drying curve to achieve regional adaptive control.

[0012] In a preferred embodiment, step S1 includes the following contents:

[0013] S1.1. Uniformly arrange matrix density sensors and airflow sensors on the inner wall of the drying equipment.

[0014] S1.2. Use the multi-point measurement data of the density sensors to establish an initial density distribution model on the surface of the straw pile through an interpolation algorithm; adopt the inverse distance weighted interpolation method to assign weights to the density values of each measurement point according to its distance to the target point to generate the density distribution of the two-dimensional plane.

[0015] S1.3. Correct the initial density distribution model using the pressure data of the airflow sensor; the pressure difference reflects the resistance on the ventilation path, and the non-linear relationship between the resistance and the local bulk density is quantified through modeling; the density distribution correction logic uses the pressure difference data to enhance the density weight in the ventilation-difficult areas and weaken the interference in the low-resistance areas through exponential transformation.

[0016] S1.4. Based on the density interpolation model and the airflow correction data, use the constrained optimization algorithm to generate a three-dimensional density distribution model; the optimization objective function balances the difference between the initial density model and the airflow correction data; through iterative solution, finally obtain the density distribution values of each spatial point and construct a three-dimensional structure model of the straw stack.

[0017] The process of marking the high-density area and the low-density area is first analyzed based on the local density value of the three-dimensional density distribution model and the density gradient change of its adjacent points to determine the density characteristic range within the area. The specific operation is as follows: according to the optimized three-dimensional density model, extract the distribution range of the global density value, and calculate the upper and lower quantiles of its statistical distribution as the initial reference thresholds for low density and high density; then at each density sampling point, use the density change rate gradient of its adjacent points for local correction. The correction logic for the high-density threshold area is to detect whether the density change gradually stabilizes within a certain range. If it meets the condition, it is marked as a stable high-density area; for the low-density threshold area, judge whether the density value is less than the initial low-density threshold and the local gradient change is significant. When the conditions are met, it is marked as an abnormal low-density area; finally, generate a regional high-low density distribution map through the high-low density marking points.

[0018] In a preferred embodiment, step S2 includes the following contents:

[0019] S2.1. Inject a certain amount of gas at the inlet of the drying equipment, and ensure the uniform airflow flowing into the equipment by controlling the injection flow rate and the initial pressure; the matrix-type airflow sensors arranged on the inner wall of the equipment record the velocity and pressure of the airflow at the same time; the airflow sensors output the corresponding airflow velocity values and pressure values by sensing the airflow response at different points.

[0020] S2.2. Based on the velocity and pressure difference recorded by the airflow sensors, preliminarily estimate the permeability distribution of the straw stack; the permeability of the airflow is closely related to the pressure drop and the flow velocity on the path. By analyzing the flow velocity and pressure difference data at the sensor position points, calculate the local permeability of each sensor point.

[0021] S2.3. Combine the airflow response data collected at multiple moments to construct a dynamic model of airflow diffusion. The diffusion process of airflow is significantly affected by permeability and density distribution. A dynamic diffusion model is used to describe the change of airflow concentration in time and space. The airflow diffusion process is simulated by discrete solution through the finite difference method, and the model is fitted with the measured data of the sensor to adjust the initial permeability distribution and generate a corrected permeability distribution.

[0022] S2.4. The process of marking high-permeability and low-permeability regions is based on the corrected permeability distribution data and is completed by analyzing the permeability gradient and regional characteristics. First, calculate the permeability gradient at each spatial point, that is, the change rate of the permeability value of adjacent points, and screen out the gradient mutation points as potential boundary regions. Conduct a statistical analysis of the permeability values in the regions with gradients below the standard, and take their average value as the global reference permeability threshold. When marking high-permeability regions, screen out the regions with continuous gradients and permeability values higher than the corresponding threshold to ensure unobstructed and uniform ventilation paths. When marking low-permeability regions, screen out the regions with permeability values lower than the corresponding threshold and significant gradient changes, and analyze the airflow retention characteristics in combination with the dynamic diffusion model to confirm whether there is a potential airflow blockage phenomenon. Finally, generate a regional distribution map.

[0023] In a preferred embodiment, step S3 includes the following:

[0024] S3.1. Integrate the obtained three-dimensional density distribution model and the permeability distribution map, and combine the initial moisture content data of the straw stack. In addition, collect environmental parameter data.

[0025] S3.2. Based on the density distribution , permeability distribution and initial moisture content , construct a regional drying demand prediction function to predict the specific drying demand at each spatial point during the drying process. The formula of the drying demand prediction function is: ; where: is the drying demand index of spatial point ; is the drying demand sensitivity parameter, reflecting the non-linear relationship between moisture, density and permeability; is the environmental parameter influence coefficient, adjusting the influence degree of temperature and humidity on the drying demand; is the base of the natural logarithm.

[0026] S3.3. Use machine learning algorithms to optimize the drying demand prediction function.

[0027] S3.4. Based on the optimized regional drying demand model, generate the target drying curve for each region , where Represents time.

[0028] In a preferred embodiment, step S4 includes the following:

[0029] S4.1, Analyze the target drying curves of each region, and extract the required temperature and air flow rate at different time points for each region; Through spatial mapping technology, correspond each region in the three-dimensional space to the specific control unit of the drying equipment, ensuring that each control unit can independently receive and execute the corresponding drying parameters.

[0030] S4.2, Based on the target drying curve, construct an intelligent control algorithm, and adopt a multivariable non-linear optimization method to ensure that the dynamic adjustment of the drying parameters can accurately match the requirements of each region. The specific steps include:

[0031] Establishment of multivariable optimization model: Establish a multivariable optimization model for temperature and air flow rate, and the model form is as follows: ; where: and are the temperature and air flow rate of the current region at time respectively; and are the temperature and air flow rate specified in the target drying curve; is the total drying time.

[0032] Setting of constraint conditions: Based on the physical limitations and safety parameters of the equipment, set the upper and lower limits of temperature and air flow rate: ; ; where, and represent the minimum and maximum drying temperatures allowed at the spatial point ; represents the minimum and maximum air flow rates allowed at the spatial point .

[0033] Application of optimization algorithm: Adopt genetic algorithm or particle swarm optimization global optimization technology to solve the model, obtain the optimal temperature and air flow rate settings of each region at different time points, and distribute the obtained optimal temperature and air flow rate settings to the execution units of the corresponding regions through the control module in real time, and dynamically adjust the operating parameters of the drying equipment to match the target requirements of each region.

[0034] Technical effects and advantages of the straw drying state monitoring method based on multi-sensors of the present invention:

[0035] Through density distribution modeling, permeability analysis, drying demand prediction, and regional adaptive control, the present invention systematically solves problems such as uneven bulk density, airflow blockage, and difficulty in accurately matching drying conditions during the straw drying process. Through refined density and permeability modeling, drying bottleneck regions are accurately identified; using a regional drying demand model, drying parameters for each region are accurately predicted; combined with intelligent optimization algorithms and real-time feedback control, dynamic adjustment and accurate matching of the drying process are achieved. This method significantly improves the efficiency and uniformity of straw drying, avoids problems of insufficient or excessive drying caused by density and permeability differences, effectively reduces energy consumption and safety risks during the drying process, and ensures the consistency of drying quality and the reliability of equipment operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 FIG. is a schematic flow chart of the method for monitoring the straw drying state based on multi-sensors according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0038] Embodiment 1: Figure 1 The method for monitoring the straw drying state based on multi-sensors according to the present invention is given, including:

[0039] S1, collecting data from density sensors and airflow sensors, generating a three-dimensional density distribution model of straw accumulation through multi-point interpolation and data fusion, and locating high-density regions and low-density regions.

[0040] S2, injecting a quantitative gas and monitoring the airflow response data, monitoring the airflow response data, obtaining a permeability distribution map through dynamic modeling, and identifying airflow blockage or inefficient regions.

[0041] S3, combining the density distribution, permeability characteristics, and initial moisture content data, establishing a regional drying demand model and generating the target drying curve for each region.

[0042] S4, adjusting the corresponding drying parameters based on the target drying curve to achieve regional adaptive control.

[0043] During the straw drying process, the distribution of the straw bulk density directly determines the drying efficiency and uniformity. Due to the internal non-uniformity of the straw pile during the stacking process, traditional methods cannot obtain the internal density information. Existing solutions usually place sensors inside the straw pile, which not only affects the stacking structure but also has problems such as limited detection range and difficult maintenance. In this step, a matrix density sensor and an air flow sensor arranged on the inner wall of the drying equipment are used. Through multi-point measurement, data fusion, and inverse modeling, a three-dimensional density distribution model of the straw stack is constructed to locate the high-density, void, and abnormal areas inside the stack.

[0044] Step S1 includes the following content:

[0045] S1.1, Uniformly arrange a matrix density sensor and an air flow sensor on the inner wall of the drying equipment to avoid interfering with the straw stacking structure through the inner wall layout of the equipment. The density sensor uses electromagnetic field induction to measure the local mass distribution of the straw area near the inner wall of the equipment and generates multi-point density data on a two-dimensional plane. The air flow sensor records the pressure difference data of each channel by sensing the pressure change of the ventilation path, which is used to reflect the internal permeability characteristics of the stack. The collected data includes the spatial position, measurement value, and environmental conditions of each sensor, which are used as the input for subsequent modeling.

[0046] S1.2, Using the multi-point measurement data of the density sensor, an initial density distribution model of the straw stack surface is established through an interpolation algorithm. To improve the modeling accuracy, the inverse distance weighted interpolation method is used to assign weights to the density values of each measurement point according to its distance to the target point, generating a density distribution on a two-dimensional plane. The interpolation process constructs the mass characteristics of the local area through the spatial position relationship of the sensors, which better reflects the surface changes of the stacking structure.

[0047] For example, using the three-dimensional inverse distance weighted interpolation method (IDW), the formula is as follows: ;

[0048] Where:

[0049] is the density value of point .

[0050] is the density measurement value of sensor .

[0051] is the Euclidean distance between point and sensor .

[0052] is the distance attenuation exponent, which controls the attenuation speed of the interpolation weight, and usually takes a value range of 1.5 to 2.5.

[0053] The number of sampling points participating in the interpolation calculation.

[0054] The interpolation of density data is restricted by the sensor distribution density and the device structure. The interpolation model may have relatively large errors in areas far from the sensors. Accuracy needs to be improved through subsequent calibration steps.

[0055] S1.3. Use the pressure data of the air flow sensor to calibrate the initial density distribution model. The pressure difference reflects the resistance on the ventilation path, and the non-linear relationship between the resistance and the local bulk density is quantified through modeling. The density distribution calibration logic uses the pressure difference data to enhance the density weight in areas with difficult ventilation and weaken the interference in areas with low resistance through exponential transformation. This process introduces the dynamic information of air flow into the density distribution, making up for the limitations of single density sensor measurement.

[0056] Assume that the pressure difference of the ventilation path has a non-linear relationship with the regional density, and boundary restrictions will be imposed on abnormal points (such as local abnormal high pressure differences) to avoid over-adjustment.

[0057] The density value is calibrated using the following formula: ;

[0058] Where:

[0059] is the calibrated density.

[0060] is the pressure difference at point

[0061] is the maximum pressure difference within the area, used for normalization.

[0062] This process improves the accuracy of the model by increasing the weight of high-density areas and reducing the interference of low-density areas.

[0063] S1.4. Based on the density interpolation model and the air flow calibration data, use the constrained optimization algorithm to generate a three-dimensional density distribution model. The optimization objective function balances the differences between the initial density model and the air flow calibration data. The constraint conditions include density range, density gradient continuity, and physical consistency requirements. Through iterative solution, the density distribution values of each spatial point are finally obtained, and a three-dimensional structure model of the straw stack is constructed. The optimized model can accurately reflect high-density stacking areas, potential ventilation bottlenecks, and low-density voids.

[0064] Optimization objective function: Minimize the difference between the initial density model and the calibrated model. The formula is as follows: ;

[0065] Constraint conditions:

[0066] Density range ​Reasonable.

[0067] Density gradient continuity .

[0068] S1.5. Based on the optimized three-dimensional density model, analyze the density gradient and local anomalies, and mark the high-density areas and low-density areas where uneven drying or air flow blockage may exist. These areas will serve as the key basis for subsequent zoning optimization of the drying strategy. The model output includes a three-dimensional density distribution map, marked abnormal areas, and their physical characteristic data to ensure the accuracy of the subsequent drying control process.

[0069] The process of marking high-density and low-density areas is first based on the local density values of the three-dimensional density distribution model and the density gradient changes of its adjacent points to determine the density characteristic range within the area. The specific operation is as follows: According to the optimized three-dimensional density model, extract the distribution range of the global density values, and calculate the upper and lower quantiles of its statistical distribution (such as the 25th percentile and the 75th percentile) as the initial reference thresholds for low density and high density; then at each density sampling point, use the density change rate gradient of its adjacent points for local correction. The correction logic for the high-density threshold area is to detect whether the density change gradually stabilizes within a certain range (i.e., the gradient value is less than the set gradient upper limit), and if satisfied, mark it as a stable high-density area; for the low-density threshold area, judge whether the density value is less than the initial low-density threshold and the local gradient change is significant (i.e., there is a large-scale downward trend in the density distribution), and mark it as an abnormal low-density area when the condition is met. Finally, generate a regional high-low density distribution map through these high-low density marked points, and adjust the boundary of the transition area in combination with the global characteristics of the model to form an accurate high-low density annotation result.

[0070] Through the matrix density sensors and air flow sensors arranged on the inner wall of the drying equipment, combined with multi-point interpolation and inverse modeling techniques, step S1 accurately constructs a three-dimensional density distribution model of the straw stack. The model effectively characterizes the high-density areas, low-density voids, and density gradient changes inside the straw stack, laying a reliable data foundation for subsequent permeability analysis and zoning optimization of the drying process. The modeling method of the density distribution integrates the dynamic correction of density and air flow data, improving the accuracy and applicability of the model, and ensuring a more comprehensive and accurate description of the characteristics of the stacking structure.

[0071] During the straw drying process, the air permeability of the airflow directly affects the drying efficiency and uniformity. Due to the uneven density distribution inside the straw pile, the ventilation path will show a significant non-linear distribution, resulting in problems such as airflow blockage or inefficient ventilation in some areas. However, relying solely on static monitoring of airflow velocity and pressure difference data cannot comprehensively characterize the complex permeability distribution. To solve this problem, in this step, a quantitative gas is injected into the drying equipment, the response data of the airflow sensor is dynamically monitored, and combined with dynamic modeling and path reconstruction, a high-resolution distribution map reflecting the internal permeability is generated to accurately locate the ventilation bottleneck area.

[0072] Step S2 includes the following:

[0073] S2.1, Inject a quantitative gas into the air inlet of the drying equipment. By precisely controlling the injection flow rate and initial pressure, ensure that the airflow flowing into the equipment is uniform. The matrix airflow sensors arranged on the inner wall of the equipment record the velocity and pressure of the airflow simultaneously. The airflow sensors output the corresponding airflow velocity values and pressure values by sensing the airflow responses at different points. Each sensor records the time, position coordinates, and response data, and this information forms the basis for analyzing the permeability. The data collection is completed in multiple rounds, and each round lasts for a period of time to ensure a complete capture of the dynamic diffusion process of the airflow inside the straw pile.

[0074] S2.2, Based on the velocity and pressure difference recorded by the airflow sensors, preliminarily estimate the permeability distribution of the straw pile. The air permeability of the airflow is closely related to the pressure drop and flow velocity on the path. By analyzing the flow velocity and pressure difference data at the sensor position points, calculate the local permeability of each sensor point. The path length, as the geometric constraint of the area through which the airflow passes, combined with the relationship between the flow rate and pressure loss, obtains the preliminary permeability distribution.

[0075] Use the velocity and pressure recorded by the airflow sensors to model the diffusion path of the airflow in the straw pile. Assume that the permeability through which the airflow passes has a non-linear relationship with the pressure drop on the path, and estimate the initial permeability distribution through the following formula: ;

[0076] Where:

[0077] is the permeability of the path (unit: ) ).

[0078] is the airflow injection volume.

[0079] is the path length, calculated by the geometric distance of the sensor positions in three-dimensional space.

[0080] is the pressure difference of the path .

[0081] S2.3. Combine the airflow response data collected at multiple moments to construct a dynamic model of airflow diffusion. The diffusion process of the airflow is significantly affected by permeability and density distribution. A dynamic diffusion model is used to describe the change of airflow concentration in time and space. Through the concentration change trend of the airflow sensor, the initial permeability distribution is corrected to make it better reflect the actual situation. The dynamic model iteratively simulates the airflow diffusion path and gradually adjusts the permeability distribution so that the final result can fit the characteristics of the dynamic concentration distribution.

[0082] Based on the airflow response data at each moment within the sampling period, a dynamic diffusion model is constructed to analyze the regional characteristics of the airflow. It is assumed that the airflow diffusion satisfies the following non-linear diffusion equation: ;

[0083] where:

[0084] is the position at time of the airflow concentration.

[0085] is the diffusion coefficient, which is related to density and permeability.

[0086] is the dissipation coefficient, which reflects the weakening effect of airflow resistance on diffusion.

[0087] Discretely solve this equation by the finite difference method, simulate the airflow diffusion process, and fit the model with the measured data of the sensor to adjust the initial permeability distribution and generate a corrected permeability distribution. . The dissipation coefficient is further corrected through the correlation function with the pressure gradient to optimize the regional distribution of the airflow.

[0088] S2.4. The process of marking high-permeability and low-permeability regions is completed based on the corrected permeability distribution data by analyzing the permeability gradient and regional characteristics. First, calculate the permeability gradient of each spatial point, that is, the change rate of the permeability values of adjacent points, and screen out the gradient mutation points as potential boundary regions; conduct statistical analysis on the permeability values of the regions with gradients lower than the standard, and take their average value as the global reference permeability threshold; when marking high-permeability regions, screen out the regions with continuous gradients and permeability values higher than the corresponding threshold to ensure unobstructed and uniform ventilation paths; when marking low-permeability regions, screen out the regions with permeability values lower than the corresponding threshold and significant gradient changes, and further analyze the airflow retention characteristics in combination with the dynamic diffusion model to confirm whether there is a potential airflow blockage phenomenon. Finally, generate a regional distribution map, where the boundaries between high-permeability regions and low-permeability regions are composed of gradient change points, and the stability and accuracy of the results are verified through multiple rounds of data.

[0089] By injecting a fixed amount of gas into the drying equipment and monitoring the response data of the airflow sensor, step S2 generates a permeability distribution map inside the straw stack using dynamic diffusion modeling technology. This distribution map clearly identifies the airflow blockage areas and high-permeability areas in the straw stack, and improves the resolution and accuracy of the analysis results by combining the correction method of airflow dynamic characteristics, providing a key support basis for the regional optimization and risk control of subsequent drying strategies. The permeability distribution analysis further combines the dynamic diffusion law to ensure the accurate positioning of bottleneck areas and potential inefficient areas.

[0090] In the first two steps, the internal structural characteristics of the straw pile have been comprehensively grasped through the density distribution model (step S1) and the permeability distribution map (step S2). To achieve an efficient and uniform drying process, these structural characteristics must be combined with the initial moisture content data to accurately predict the drying requirements of each area. This step aims to generate a target drying curve for each area through multi-dimensional data fusion and complex model establishment, ensuring that the drying process can be adaptively adjusted to meet the specific requirements of different areas.

[0091] Step S3 includes the following:

[0092] S3.1, Integrate the three-dimensional density distribution model and the permeability distribution map obtained in steps S1 and S2, and combine with the initial moisture content data of the straw stack. The initial moisture content is measured in each area by a portable moisture sensor, and the initial moisture content of each area is recorded. . In addition, collect environmental parameter data, such as the outside temperature and humidity for dynamic adjustment in the model.

[0093] S3.2, Based on the density distribution , permeability distribution and initial moisture content , construct a regional drying demand prediction function , the purpose of which is to predict the specific drying requirements of each spatial point during the drying process. This function takes into account the moisture migration dynamics of the straw, the influence of permeability on airflow, and the dynamic changes of environmental parameters.

[0094] Drying demand prediction function formula: ;

[0095] Where:

[0096] is the drying demand index of the spatial point .

[0097] is the drying demand sensitivity parameter, reflecting the non-linear relationship between moisture and density and permeability.

[0098] is the environmental parameter influence coefficient, which adjusts the influence degrees of temperature and humidity on the drying demand.

[0099] is the base of the natural logarithm.

[0100] This function reflects the complex influence of density and permeability on moisture migration through non-linear combination, and dynamically responds to the changes of environmental parameters to ensure that the drying demand model can adapt to real-time conditions.

[0101] S3.3. Use machine learning algorithms (such as support vector machine regression or neural network) to optimize the drying demand prediction function. Through historical drying data and real-time monitoring data, train the model to improve the prediction accuracy. The optimization process includes parameter adjustment, model validation and cross-validation to ensure that the model has good generalization ability under different straw pile structures and environmental conditions.

[0102] In the model optimization stage, by introducing data-driven methods, the accuracy and robustness of drying demand prediction are improved to ensure that the generated target drying curve can truly reflect the specific demands of each region.

[0103] S3.4. Based on the optimized regional drying demand model, generate the target drying curve for each region , where represents time. The target drying curve includes the dynamic change trends of drying temperature, air flow velocity and drying time to ensure that the demands in each region during the drying process are accurately met. Through spatial mapping technology, the target drying curve is assigned to specific regional control units to achieve regional drying parameter setting.

[0104] By comprehensively considering the density distribution, permeability characteristics and initial moisture content data, step S3 constructs a refined regional drying demand model and generates the target drying curve for each region. This model uses a non-linear drying demand prediction function and combines machine learning optimization technology to ensure high-precision prediction and dynamic adaptation of drying demand. The generated target drying curve provides specific drying parameter guidance for each region, realizes the precise matching and optimization of the drying process, significantly improves the drying efficiency and uniformity, and ensures that the final product meets high-quality requirements.

[0105] In the previous steps, the internal structural characteristics of straw stacking have been comprehensively grasped through the density distribution model (step S1) and the permeability distribution map (step S2), and the target drying curves for each region have been generated through the regional drying demand model (step S3). These target curves detail the temperature, air velocity, and time requirements for each region during the drying process. To achieve efficient and uniform drying, these target drying curves must be translated into specific drying parameter adjustment operations to ensure that the drying conditions in each region can dynamically respond to its specific needs. This step aims to precisely adjust the operating parameters of the drying equipment based on the target drying curves through an intelligent control system to achieve regional adaptive drying control.

[0106] Step S4 includes the following:

[0107] S4.1, Analyze the target drying curves for each region generated in step S3 to extract the specific values of the temperature and air velocity required for each region at different time points. Through spatial mapping technology, map each region in three-dimensional space to the specific control units of the drying equipment to ensure that each control unit can independently receive and execute the corresponding drying parameters.

[0108] S4.2, Based on the target drying curves, construct an intelligent control algorithm and adopt a multivariable nonlinear optimization method to ensure that the dynamic adjustment of drying parameters can accurately match the needs of each region. The specific steps include:

[0109] Establishment of a multivariable optimization model: Establish a multivariable optimization model for temperature and air velocity, considering the spatio-temporal variation characteristics of the target curves for each region. The model form is as follows: ;

[0110] Where:

[0111] and are the temperature and air velocity of the current region at time respectively.

[0112] and are the temperature and air velocity specified in the target drying curve.

[0113] is the total drying time.

[0114] Setting of constraint conditions: Based on the physical limitations and safety parameters of the equipment, set the upper and lower limits of temperature and air velocity:

[0115] ;

[0116] ;

[0117] and represent the lowest and highest drying temperatures allowed at the spatial point .

[0118] represent the lowest and highest air flow velocities allowed at the spatial point .

[0119] Optimization algorithm application: Global optimization techniques such as genetic algorithms or particle swarm optimization are used to solve the model, obtaining the optimal temperature and air flow velocity settings for each region at different time points. The obtained optimal temperature and air flow velocity settings are distributed to the execution units of the corresponding regions in real time through the control module, dynamically adjusting the operating parameters of the drying equipment to match the target requirements of each region.

[0120] In step S4, by intelligently analyzing the target drying curve, the temperature and air flow velocity requirements of each region are accurately mapped to the control unit of the drying equipment. A multivariable nonlinear optimization algorithm is used to dynamically adjust the drying parameters to ensure that the drying conditions of each region highly match their specific requirements. At the same time, the system uses high-precision sensors to monitor the actual temperature and air flow velocity in real time, compares and analyzes them with the target curve, and uses an adaptive control algorithm to immediately correct the deviation to ensure the continuous optimization and stable operation of the drying process.

[0121] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulations to get a formula that is closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0122] Only some exemplary embodiments of the present invention have been described above by way of illustration. Undoubtedly, for those of ordinary skill in the art, without departing from the spirit and scope of the present invention, the described embodiments can be modified in various different ways. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0123] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0124] As described above, the above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A straw drying state monitoring method based on multiple sensors, characterized in that: Includes steps: S1, collects data from density sensors and airflow sensors, generates a three-dimensional density distribution model of straw accumulation through multi-point interpolation and data fusion, and locates high-density areas and low-density areas; S2, injecting a certain amount of gas and monitoring the airflow response data, monitoring the airflow response data, and obtaining a permeability distribution map through dynamic modeling; S3, combining the three-dimensional density distribution model, permeability distribution map and initial moisture content data, constructing the regional drying demand prediction function and generating the target drying curve for each region; S4, adjusting the corresponding drying parameters based on the target drying curve to achieve regional adaptive control; Step S3 includes the following contents: S3.1, integrating the obtained three-dimensional density distribution model and permeability distribution map, combined with the initial moisture content data of the straw pile, and in addition, collecting environmental parameter data; S3.2, based on three-dimensional density distribution , Permeability distribution and initial moisture content , construct regional drying demand prediction function , predict the specific drying requirements of each spatial point during the drying process, and the drying requirement prediction function formula is: ;in: For space point Drying demand index; It is the drying demand sensitivity parameter, reflecting the nonlinear relationship between moisture, density and permeability; is the environmental parameter influence coefficient, which adjusts the influence of temperature and humidity on drying requirements; is the base of natural logarithms; and are the outside temperature and humidity respectively; S3.3, optimizing the drying demand prediction function using a machine learning algorithm; S3.4, based on the optimized drying demand prediction function, generate the target drying curve for each area ,in Indicates time.

2. The method for monitoring the drying state of straw based on multiple sensors according to claim 1, characterized in that: Step S1 includes the following contents: S1.1, evenly arrange matrix density sensors and airflow sensors on the inner wall of the drying equipment; S1.2, using the multi-point measurement data of the density sensor, an initial density distribution model of the straw pile surface is established through an interpolation algorithm; using the inverse distance weighted interpolation method, the density value of each measurement point is weighted according to its distance to the target point to generate a density distribution in a two-dimensional plane; S1.3, using the pressure data of the airflow sensor, the initial density distribution model is corrected; the pressure difference reflects the resistance on the ventilation path, and the nonlinear relationship between the resistance and the local bulk density is quantified through modeling; the density distribution correction logic uses the pressure difference data to enhance the density weight of the ventilation difficult area through exponential transformation, while weakening the interference of the low resistance area; S1.4, based on the density interpolation model and the airflow correction data, a three-dimensional density distribution model is generated using a constrained optimization algorithm; the objective function is optimized to balance the difference between the initial density model and the airflow correction data; through iterative solution, the density distribution value of each spatial point is finally obtained to construct a three-dimensional density distribution model of the straw pile.

3. The method for monitoring the drying state of straw based on multiple sensors according to claim 2, characterized in that: Step S1 also includes the following contents: S1.5, the process of marking high-density areas and low-density areas first analyzes the local density value of the three-dimensional density distribution model and the density gradient changes of its adjacent points to determine the density feature range in the area. The specific operation is: according to the optimized three-dimensional density distribution model, the distribution range of the global density value is extracted, and the upper and lower quantiles of its statistical distribution are calculated as the initial reference thresholds for low density and high density; then at each density sampling point, the density change rate gradient of its adjacent points is used for local correction. The correction logic of the high-density threshold area is to detect whether the density change gradually tends to be stable within a certain range. If it meets the requirements, it is marked as a stable high-density area; for the low-density threshold area, it is judged whether the density value is less than the initial low-density threshold and the local gradient changes significantly. If the conditions are met, it is marked as an abnormal low-density area; finally, a regional high and low density distribution map is generated through high and low density marking points.

4. The method for monitoring the drying state of straw based on multiple sensors according to claim 3, characterized in that: Step S2 includes the following contents: S2.1, inject a fixed amount of gas into the air inlet of the drying equipment, and ensure the uniformity of the airflow into the equipment by controlling the injection flow rate and initial pressure; the matrix airflow sensor arranged on the inner wall of the equipment records the airflow speed and pressure at the same time; the airflow sensor senses the airflow response at different points and outputs the corresponding airflow speed value and pressure value; S2.2, based on the velocity and pressure difference recorded by the airflow sensor, preliminarily estimate the permeability distribution of the straw pile; the permeability of the airflow is closely related to the pressure drop and flow velocity along the path. By analyzing the flow velocity and pressure difference data at the sensor location point, the local permeability of each sensor point is calculated; S2.3, combining the airflow response data collected at multiple times, constructing a dynamic model of airflow diffusion; the airflow diffusion process is significantly affected by the permeability and density distribution, and a dynamic diffusion model is used to describe the changes in airflow concentration in time and space; the airflow diffusion process is simulated through discrete solution using the finite difference method, and the model is fitted in combination with the actual sensor data, the initial permeability distribution is adjusted, and a corrected permeability distribution is generated.

5. The method for monitoring the drying state of straw based on multiple sensors according to claim 4, characterized in that: Step S2 includes the following contents: S2.4, the process of marking high-permeability and low-permeability areas is based on the corrected permeability distribution data and is completed by analyzing the permeability gradient and regional characteristics; First, the permeability gradient of each spatial point is calculated, that is, the rate of change of the permeability values ​​of adjacent points, and the gradient mutation points are screened out as potential boundary areas; the permeability values ​​of areas with gradients below the standard are statistically analyzed, and the average value is taken as the global benchmark permeability threshold; when marking high-permeability areas, areas with continuous gradients and permeability values ​​higher than the corresponding threshold are screened to ensure that the ventilation path is unobstructed and evenly distributed; when marking low-permeability areas, areas with permeability values ​​lower than the corresponding threshold and drastic gradient changes are screened, and the airflow retention characteristics are analyzed in combination with the dynamic diffusion model to confirm whether there is potential airflow blockage; finally, a permeability distribution map is generated.

6. The method for monitoring the drying state of straw based on multiple sensors according to claim 5, characterized in that: Step S4 includes the following contents: S4.1, analyze the target drying curve of each area and extract the temperature required for each area at different time points and air velocity Through spatial mapping technology, each area in the three-dimensional space corresponds to the specific control unit of the drying equipment, ensuring that each control unit can independently receive and execute the corresponding drying parameters; S4.2, based on the target drying curve, build an intelligent control algorithm and use a multivariable nonlinear optimization method to ensure that the dynamic adjustment of drying parameters can accurately match the needs of each area. The specific steps include: Establishment of multivariable optimization model: Establish a multivariable optimization model of temperature and airflow velocity. The model form is as follows: ;in: and The current area at time temperature and air velocity; and is the temperature and airflow velocity specified in the target drying curve; is the total drying time; Constraint setting: Set upper and lower limits for temperature and airflow speed based on the physical limitations and safety parameters of the equipment: ; ;in, and Indicates a point in space The minimum and maximum drying temperatures allowed; and Indicates a point in space The minimum and maximum air velocity allowed at the location; Application of optimization algorithm: Genetic algorithm or particle swarm optimization global optimization technology is used to solve the model and obtain the optimal temperature and airflow speed settings for each area at different time points. The obtained optimal temperature and airflow speed settings are distributed to the execution units of the corresponding areas in real time through the control module, and the operating parameters of the drying equipment are dynamically adjusted to match the target requirements of each area.

Images