Jmag-based Thermal Modeling Method for Grain Silos
Through the jmag-based granary thermal modeling method, static segmentation and dynamic fine-tuning technology are used to solve the problem of high difficulty in data acquisition and insufficient accuracy in thermal modeling of large granary, and more accurate grain temperature prediction and scientific grain temperature decision-making are achieved.
Patent Information
- Application Number
- CN202510245283.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-04
- Publication Date
- 2025-07-22
- Estimated Expiration
- 2045-03-04
AI Technical Summary
The existing technology lacks thermal modeling methods with underlying principle interpretability in large granaries, resulting in unreliable grain temperature prediction results and high requirements for data acquisition and accuracy.
The granary unit side length is dynamically adjusted by using jmag-based granary thermal modeling method, through static segmentation and dynamic fine-tuning, using jmag modeling software and Python scripts, thermal modeling and simulation are performed in combination with the grain stack historical temperature data, and the granary unit thermal model is optimized.
It improves the accuracy of grain temperature prediction and interpretability of underlying principles, and can more accurately simulate the internal heat conduction process of granaries and provide scientific guidance on grain temperature decisions.
Smart Images
Figure CN119740446B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of grain temperature prediction, and particularly to a method for thermal modeling of a grain bin based on JMAG. Background Art
[0002] At present, during the storage process of grains, they are affected by multiple temperatures. The rise and fall of temperature will cause the migration of moisture inside the grains. Excessive or too low moisture content will affect the quality and storage stability of the grains; too high temperature will make molds and pests reach the suitable breeding temperature. If no intervention is taken at this time and the high temperature is maintained for a long time, the grains in the grain pile will become moldy and deteriorate, resulting in grain loss; in addition, the increase in temperature will also accelerate the respiration of the grains and increase the metabolic rate of the grains, and excessive respiration will lead to the accelerated deterioration and quality decline of the grains.
[0003] The thermal modeling method is based on thermal principles such as heat conduction, convection, and radiation, and uses data such as the geometric structure and heat transfer characteristics of the warehouse for simulation and prediction, such as the finite difference method (FDM) and the finite element method (FEM), etc. It can analyze and model the real thermal system, and conduct a virtual experiment on the basis of the modeling, that is, thermal simulation, by inputting a series of parameters and information data to calculate and predict the heat transformation of the thermal system. And the interior of the grain bin is a thermal system in the state of storing grains. The transfer of temperature inside the grain pile follows the principle of heat conduction. Therefore, the thermal modeling can be used to study the temperature change inside the grain pile.
[0004] In the prior art, when performing thermal modeling on a large grain bin, the length and width of the grain bin can reach more than 20 meters, and there are obvious differences in parameters such as the heat and air flow inside it at different positions. Based on the complex environment inside the above-mentioned grain bin, in a large grain bin, the mainstream grain temperature prediction method usually performs temperature prediction based on a data-driven model (such as LSTM, Transformer, etc.). The whole process includes multiple steps such as data collection, model establishment, training, and prediction.
[0005] However, such methods need to ensure a sufficient amount of temperature sensor data for the pattern learning of the model, which requires high collection difficulty and accuracy of the data, and lacks interpretability at the underlying principle level, making it impossible to model a large grain bin and effectively simulate the real grain pile environment, resulting in unreliable grain temperature prediction results. Summary of the Invention
[0006] Based on this, it is necessary to provide a method for thermal modeling of a grain bin based on JMAG for the above technical problems.
[0007] This specification adopts the following technical solutions:
[0008] This specification provides a method for thermal modeling of a grain bin based on JMAG, including:
[0009] Obtain the basic information parameters of the granary and the grain pile, including: the lengths of the three dimensions of the granary, namely length, width, and height, the historical temperature data of the grain pile, and the coordinates of the temperature measurement points.
[0010] Based on the lengths of the three dimensions of the granary, namely length, width, and height, select the dimension with the smallest length as the reference length, and perform granary segmentation on the other two dimensions from a spatial perspective based on the reference length to obtain multiple segmented granary units.
[0011] Through the Geometry Editor module of the jmag modeling software, taking the reference length as the standard, set the dynamic fine-tuning amount and fine-tuning range of the side length of the granary unit, call the python loop script to dynamically fine-tune the structure of the granary unit at any corner after segmentation into cubes with multiple side lengths, and obtain the thermal models of multiple dynamically fine-tuned granary units through thermal modeling.
[0012] Input the historical temperature data of the grain pile and the coordinates of the temperature measurement points into the thermal models of multiple dynamically fine-tuned granary units, set the parameters of the thermal models of the granary units, perform simulations on the thermal models of the dynamically fine-tuned granary units one by one, obtain the simulation results of each thermal model of the dynamically fine-tuned granary unit and calculate the mean absolute error between it and the true value, and select the ratio of the reference length with the smallest error as the final reference length of the side length of the granary unit.
[0013] Adjust the structure of the granary unit at the corner after segmentation into a cube with the final reference length as the side length, and use the differences between the lengths of the other two dimensions and the side lengths of the granary units at their corresponding corners to interpolate and adjust the structures of the granary units except those at the corners after segmentation to cubes, obtain the finally segmented granary units and perform thermal modeling on each of them one by one to obtain the final thermal models of each granary unit after segmentation.
[0014] Preferably, the dynamic fine-tuning amount and fine-tuning range of the side length of the granary unit specifically include:
[0015] The dynamic fine-tuning amount is 5% of the reference length;
[0016] The fine-tuning range is from 70% of the reference length to 130% of the reference length.
[0017] Preferably, the parameters of the thermal model of the granary unit specifically include: material parameters, granary parameters, and simulation parameters.
[0018] Preferably, the material parameters include: thermal conductivity and specific heat capacity.
[0019] Preferably, the granary parameters include: constant heat source density, boundary conditions, and initial temperature.
[0020] Preferably, the constant heat source density is used to reflect the ability to generate or consume heat per unit time and unit volume in the granary scenario.
[0021] Preferably, the boundary conditions specifically include:
[0022] Select the granary surface affected by the ambient temperature change and set all the walls during segmentation and simulation; among them, no boundary conditions need to be set at the connection between the two granary units after segmentation; and the boundary conditions of the granary unit thermal model are set by the table of ambient temperature changing with time.
[0023] Preferably, the initial temperature is the temperature of the grain pile corresponding to the moment when the granary unit thermal model starts simulation.
[0024] Preferably, the simulation parameters include: step size and number of steps;
[0025] The number of steps is consistent with the number of temperature time tables set in the boundary conditions.
[0026] The above at least one technical solution adopted in this specification can achieve the following beneficial effects:
[0027] In the granary thermal modeling method based on jmag provided in this specification, first, based on the parameters of the three dimensions of the granary including length, width, and height, the granary is statically segmented from a spatial perspective to obtain multiple segmented granary units; secondly, using a Python loop script, the structure of the granary unit at any corner is dynamically fine-tuned to a cube according to a specified ratio of the reference length and thermal modeling is performed to obtain the thermal models of multiple granary units; the historical temperature data of the grain pile and the coordinates of the temperature measurement points are input into the granary unit thermal model, the granary is simulated by setting the thermal model parameters, and the final reference length is determined according to the simulation results, thus completing the dynamic adjustment of the side length of the granary unit, and adjusting the side lengths of the segmented granary units except the corners to be close to a cube according to the difference between the lengths of the other two dimensions and the side lengths of the corresponding granary units at the corners.
[0028] Since the lengths of the length and width of large granaries are relatively large, which has a great impact on the simulation results, and the two sides of the granary are more affected by the outside world than the middle, and relatively stronger temperature gradients will be formed. Therefore, the present invention is based on the collected granary dimensions, and uses the idea of static segmentation and dynamic fine-tuning for large granaries for thermal modeling and thermal simulation, so as to more accurately simulate the heat conduction process of the granary, improve the interpretability of the underlying principles, and then effectively simulate the real grain pile environment to improve the accuracy of grain temperature prediction and provide strong scientific guidance for grain temperature decision-making. Description of the Drawings
[0029] The accompanying drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:
[0030] Figure 1 is a schematic flowchart of the grain bin thermal modeling method based on jmag provided in this specification;
[0031] Figure 2 is a front sectional view of the grain bin sensor of the grain bin thermal modeling method based on jmag provided in this specification;
[0032] Figure 3 is a side sectional view of the grain bin sensor of the grain bin thermal modeling method based on jmag provided in this specification;
[0033] Figure 4 is a schematic diagram of the grain bin modeling of the grain bin thermal modeling method based on jmag provided in this specification. Detailed Description of the Embodiments
[0034] To make the objectives, technical solutions, and advantages of this specification clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.
[0035] The following will detail the technical solutions provided in each embodiment of the present application in conjunction with the drawings.
[0036] Figure 1 is a schematic flowchart of a grain bin thermal modeling method based on jmag in this specification, which specifically includes the following steps:
[0037] S101: Obtain the basic information parameters of the grain bin and the grain pile, including: the lengths of the three dimensions of the length, width, and height of the grain bin, the historical temperature data of the grain pile, and the coordinates of the temperature measurement points.
[0038] S102: Based on the lengths of the three dimensions of the grain bin including length, width, and height, select the dimension with the smallest length as the reference length, and perform grain bin segmentation on the other two dimensions from a spatial perspective based on the reference length to obtain multiple segmented grain bin units;
[0039] Among them, when performing grain bin segmentation on the other two dimensions from a spatial perspective based on the reference length, it is not strictly required to segment with a cube with the reference length as the side length, and the number of grain bin segments should be as small as possible.
[0040] S103: Using the Geometry Editor module of the jmag modeling software, with the reference length as the standard, set the dynamic fine-tuning amount and fine-tuning range of the unit side length of the grain bin. Call the python loop script to dynamically fine-tune the unit structure of the grain bin at any corner after segmentation into a cube with multiple side lengths, and obtain multiple thermo-models of the dynamically fine-tuned grain bin units through thermal modeling.
[0041] S104: Input the historical temperature data of the grain pile and the coordinates of the temperature measurement points into multiple thermo-models of the dynamically fine-tuned grain bin units, and set the parameters of the thermo-models of the grain bin units. Simulate each thermo-model of the dynamically fine-tuned grain bin unit one by one, obtain the simulation results of each thermo-model of the dynamically fine-tuned grain bin unit and calculate the mean absolute error between it and the true value. Select the ratio of the reference length with the smallest error as the final reference length of the unit side length of the grain bin.
[0042] S105: Adjust the unit structure of the grain bin at the corner after segmentation into a cube with the final reference length as the side length, and use the difference between the lengths of the other two dimensions and the side length of the grain bin unit at the corresponding corner to interpolate and adjust the unit structure of the grain bin except at the corner after segmentation to a cube, obtain the finally segmented grain bin units and conduct thermal modeling on each of them one by one to obtain the final thermo-models of each grain bin after segmentation. Specifically, it includes:
[0043] The parameters of the thermo-model of the grain bin unit, including: material parameters, grain bin parameters, and simulation parameters;
[0044] The material parameters include: thermal conductivity and specific heat capacity;
[0045] The grain bin parameters include: constant heat source density, boundary conditions, and initial temperature;
[0046] The simulation parameters include: step size and number of steps.
[0047] Among them, the thermal conductivity is a measure of the heat conduction ability of the grain pile. It refers to the amount of heat transferred per unit time through a unit horizontal cross-sectional area when the temperature vertical downward gradient is 1 °C / m. Specifically, it is defined as: Take two parallel planes perpendicular to the heat conduction direction inside the object, with a distance of 1 meter and an area of 1 square meter. If the temperature difference between the two planes is 1 K, then the amount of heat conducted from one plane to the other plane within 1 second is defined as the thermal conductivity of the substance, and its unit is watt-meter -1 ·kelvin -1 (W·m -1 ·K -1 ). The specific heat capacity, also known as the specific heat capacity at constant pressure, is abbreviated as specific heat. It is the heat capacity of a unit mass of a substance, that is, the amount of heat absorbed or released by a unit mass of an object when it changes a unit temperature.
[0048] The constant heat source density refers to the rate of heat release or absorption per unit area or unit volume. In the context of a granary, it describes the ability of the grain pile to stably generate or consume heat per unit time and unit volume.
[0049] The boundary condition is to select the granary surface affected by the ambient temperature change and set it for all walls during segmentation and simulation. Among them, the boundary condition does not need to be set at the connection between the two granary units after segmentation; and the boundary condition of the granary unit thermal model is set by the table of ambient temperature varying with time.
[0050] The initial temperature is set to the temperature of the grain pile at the start time of the simulation.
[0051] Set the simulation step size and the number of steps, and the number of steps is the same as the number of temperature time tables set in the boundary condition.
[0052] In this embodiment, taking a large granary as an example, the thermal modeling and simulation of the granary are completed through the steps of the above method.
[0053] Collect the basic information parameters of the target granary: The target granary is a large flat warehouse, with a length of 29.35 meters, a width of 23.45 meters, a stacking line height of 6 meters. Temperature measurement cables are arranged in the granary, and each temperature measurement cable is equipped with four temperature sensors. The horizontal spacing of the sensors in the length direction of the granary is 4.8 meters, denoted as zones, and a total of 7 zones are deployed; the vertical spacing in the width direction is 4.5 meters, denoted as points, and a total of 6 points are deployed; the vertical spacing between the upper and lower layers is 1.8 meters, denoted as layers, and a total of four layers are deployed. For the specific deployment of sensor points, refer to Figure 2 and Figure 3 ; Collect the historical grain temperature data of the temperature sensors at each point within the collection time at around 9 o'clock every day and store it in units of days.
[0054] Model according to the granary size collected in step S101, refer to Figure 4. Static segmentation: Since the height of the granary (6 meters) is the minimum among the three dimensions, the height is selected as the reference for segmentation. Segmentation in the width direction (point direction): The width of the granary (23.45 meters) is equally divided. After segmentation, the unit of the granary is approximately close to a cube, and the number of granary units for the whole granary segmentation should be as small as possible. Based on 3 equal parts, each part is 7.816 meters. Since the temperature measurement point spacing in the width direction is 4.5 meters, each segmentation block after equal division can cover two columns of temperature measurement points to ensure the uniformity of data collection. Segmentation in the length direction (zone direction): Since the two sides in the length direction of the granary are more affected by the external environment than the middle area, and equal division will cut a row of sensors, making them located on the edge of the segmentation unit block and affecting the simulation results, a non-uniform segmentation strategy is adopted: the side lengths of the two sides are the same as those in the width direction, set to 7.816 meters, to ensure the uniform distribution of temperature measurement points in the segmentation blocks on the left and right sides, and each segmentation block contains two columns of temperature measurement points. Middle part: The remaining 13.718 meters is used as an independent segmentation block, which contains three columns of temperature measurement points to ensure the coherence of data sampling. Finally, the whole granary is segmented into 9 unit blocks of 3 (width direction) × 3 (length direction).
[0055] . Dynamic fine-tuning: Use a python script to perform side length fine-tuning and modeling simulation on the unit blocks near the corners of the granary. The height remains the reference unchanged. Take 10% of the reference length of 6 meters, that is, 0.6 meters, as the fine-tuning amount to fine-tune the other two sides. Starting from the initial side length of 7.816 meters after static segmentation, reduce 0.6 meters each time. The fine-tuning range is from 5.4 meters to 7.816 meters, which can ensure that the number of sensors contained in the unit block within this range is the same, and calculate the simulation error each time. Select the side length with the minimum error as the final side length of the unit block. Then, the segmentation blocks in the areas at the intersection points around the granary are consistent with the finally determined unit blocks, and the side lengths of the unit blocks in the remaining areas are equally divided after subtracting the adjusted part from the total length to ensure that the unit blocks in the internal area of the granary are still close to a cube.
[0056] Since there are many divided blocks, taking the three leftmost blocks in the length direction of the granary as an example, simulation operations are carried out. Since grains are poor conductors of heat, the thermal conductivity of wheat is 0.13 - 0.16 W / (m·K), and that of paddy rice is 0.1436 W / (m·K), with a heat capacity of 1550 J / (kg·K); another paper mentions that the thermal conductivity of grains is 0.159 W / (m·K), and the heat capacity is 1871 J / (kg·K); and the constant heat source density of grains is very small and can even be ignored. The parameter settings in our simulation all refer to the above data. The thermal conductivity is set to 0.159 W / (m·K), and the constant heat source density is set to 0.1. The simulation requires setting a thermal transient state study case and setting the simulation step size and time interval. In the simulation conditions, set the initial temperature, i.e., the temperature on the first day of the actual simulation, and a table of the ambient temperature varying with the time step. After setting the above parameters, add them to the simulation model in the form of coordinates according to the positions of the temperature measurement points in the granary. Import the three-dimensional coordinates of the specific temperature measurement points in the Probes of the case to obtain the simulated temperature of that point in the simulation results.
[0057] For the granary in this embodiment, simulations were carried out on some days in February, April, and May in 2024 respectively with dynamic segmentation modeling and without dynamic segmentation modeling. As shown in Table 1, Table 2, and Table 3, they are the average errors between the temperature measurement values of the leftmost two rows of temperature measurement points before and after the granary segmentation in February, April, and May and the true temperature measurement values respectively:
[0058] Table 1 Average errors between the temperature measurement values of the leftmost two rows of temperature measurement points before and after the granary segmentation in February and the true temperature measurement values
[0059]
[0060] Table 2 Average errors between the temperature measurement values of the leftmost two rows of temperature measurement points before and after the granary segmentation in April and the true temperature measurement values
[0061]
[0062] Table 3 Average errors between the temperature measurement values of the leftmost two rows of temperature measurement points before and after the granary segmentation in May and the true temperature measurement values
[0063]
[0064] From the simulation results of the above partial dates in 3 months, it can be seen that the error between the thermal simulation results after segmentation and the measured values is smaller, and it can better reflect the temperature change of the temperature measurement points inside the granary.
[0065] In summary, based on JMAG, the present invention statically divides the granary through benchmark parameters, dynamically fine-tunes the granary units after static division using a Python script program, and the side length of each granary unit can be dynamically adjusted according to the actual situation. Thermal modeling and thermal simulation of the granary are completed. This method can more accurately simulate the heat conduction process of the granary, improve the interpretability of the underlying principles, and effectively simulate the real grain pile environment to improve the accuracy of grain temperature prediction and provide strong scientific guidance for grain temperature decision-making.
[0066] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.
Claims
1. A method for thermal modeling of a grain bin based on Jmag, characterized in that, It includes the following steps: Obtain the basic information parameters of the granary and the grain pile, including: the lengths of the three dimensions of the granary, namely length, width and height, the historical temperature data of the grain pile, and the coordinates of the temperature measuring points; Based on the lengths of the three dimensions of the granary, namely length, width and height, select the dimension with the smallest length as the reference length, and perform granary segmentation on the other two dimensions from a spatial perspective based on the reference length to obtain multiple segmented granary units; when performing granary segmentation on the other two dimensions from a spatial perspective based on the reference length, it is not strictly required to segment with a cube with the reference length as the side length, and the number of granary segments should be as small as possible; Among them, the granary segmentation on the other two dimensions from a spatial perspective based on the reference length specifically includes: the height of the granary is the minimum value among the lengths of the three dimensions, and the height is selected as the reference for segmentation; The segmentation in the width direction includes: equally dividing the width of the granary. Based on 3 equal parts, each part is 7.816 meters; the segmentation in the length direction includes: the lengths of the two side edges are the same as those in the width direction, which is 7.816 meters; the remaining 13.718 meters is used as an independent segmentation block; Through the Geometry Editor module of the jmag modeling software, taking the reference length as the standard, set the dynamic fine-tuning amount and fine-tuning range of the side length of the granary unit, call the python loop script to dynamically fine-tune the structure of the granary unit at any corner after segmentation into cubes with multiple side lengths, and obtain the thermal models of multiple dynamically fine-tuned granary units through thermal modeling; Among them, select the granary surface affected by the ambient temperature change and set all the walls during segmentation and simulation; among them, the boundary conditions do not need to be set at the connection between the two segmented granary units; and the boundary conditions of the granary unit thermal model are set by a table of ambient temperature changing with time; Among them, the setting of the dynamic fine-tuning amount and fine-tuning range of the side length of the granary unit, and calling the python loop script to dynamically fine-tune the structure of the granary unit at any corner after segmentation into cubes with multiple side lengths specifically includes: Call the python script to perform side length fine-tuning modeling and simulation on the granary unit at the granary corner. Use 10% of the reference length of 6 meters, that is, 0.6 meters, as the fine-tuning amount to fine-tune the other two sides. Take the initial side length of 7.816 meters of the segmented granary unit and reduce it by 0.6 meters each time. The fine-tuning interval is from 5.4 meters to 7.816 meters; and calculate the simulation error each time, and select the side length with the minimum error as the final side length of the granary unit; adjust the granary units at the intersection areas around the granary to the final side length of the finally determined granary unit, and the side lengths of the unit blocks in the remaining areas are equally divided after subtracting the adjusted part from the total length; Input the historical temperature data of the grain pile and the coordinates of the temperature measuring points into the thermal models of multiple dynamically fine-tuned granary units, set the parameters of the granary unit thermal models, perform simulations on the thermal models of the dynamically fine-tuned granary units one by one, obtain the simulation results of each thermal model of the dynamically fine-tuned granary unit and calculate the average absolute error between it and the true value, and select the ratio of the reference length with the smallest error as the final reference length of the side length of the granary unit; Adjust the structure of the granary unit at the corner after segmentation into a cube with the final reference length as the side length, and use the difference between the lengths of the other two dimensions and the side length of the granary unit at its corresponding corner to interpolate and adjust the structure of the granary units except those at the corners after segmentation to a cube, obtain the finally segmented granary units, and perform thermal modeling on each of them one by one to obtain the final thermal model of each granary unit after segmentation.
2. The jmag-based thermal modeling method for grain silos according to claim 1, characterized in that The parameters of the thermal model of the granary unit specifically include: material parameters, granary parameters, and simulation parameters.
3. The jmag-based thermal modeling method for grain bins according to claim 2, wherein, The material parameters include: thermal conductivity and specific heat capacity.
4. The jmag-based thermal modeling method for grain silos according to claim 2, characterized in that The granary parameters include: constant heat source density, boundary conditions, and initial temperature.
5. The jmag-based thermal modeling method for grain silos according to claim 4, characterized in that, The constant heat source density is used to reflect the ability to generate or consume heat per unit time and unit volume in the granary scenario.
6. The jmag-based thermal modeling method for grain silos according to claim 4, characterized in that The initial temperature is the temperature of the grain pile corresponding to the moment when the thermal model of the granary unit starts simulation.
7. The jmag-based thermal modeling method for grain bins according to claim 2, wherein The simulation parameters include: step size and number of steps; The number of steps is consistent with the number of temperature time tables set in the boundary conditions.