An NMR-based graphical detection system and method for multi-field coupling in a granary

Abstract

A graphical detection system and method for multi-field coupling in a granary based on NMR belong to the research field of multi-field coupling effects in a grain bulk ecosystem. The system of the present invention consists of a granary simulation device, a nuclear magnetic resonance imaging analyzer, and a data acquisition and processing system. The present invention first applies nuclear magnetic resonance imaging technology to the detection of multi-field coupling effects in a grain bulk, which can solve problems such as low detection accuracy, susceptibility to interference, difficulty in implementation, and inability to accurately detect the moisture field in a grain bulk in the prior art. Through this system, the simulation of multi-field grain storage scenarios and the precise control of grain storage boundary conditions can be realized. It can achieve rapid, non-destructive, real and accurate detection of multi-physical field parameters such as the temperature, moisture, and relative humidity of a grain bulk, and combine with the grain condition cloud map technology to realize the visualization analysis and prediction of multi-field coupling effects in a grain bulk, providing a new technical means for the research of multi-field coupling effects in a grain bulk ecosystem.

Description

Technical Field

[0001] The present invention relates to the field of research on multi-field coupling effects in a grain pile ecosystem, and particularly to a graphical detection system and method for multi-field coupling in a granary based on NMR. Background Art

[0002] The grain in a granary during storage is a complex multi-field coupling ecosystem. This system consists of biological factors and abiotic factors. The biological factors include grains, microorganisms (such as fungi, bacteria, etc.), stored-grain pests (such as insects, mites, etc.), animals (such as rodents, birds, etc.), etc., which constitute the biological field in the grain pile; the abiotic factors include temperature, moisture, relative humidity, gas composition, micro-airflow, etc., which constitute the physical field in the grain pile. There are numerous complex interactions and connections among these fields, such as the influence of temperature on the respiration of grains, on the growth and reproduction of stored-grain pests, on stored-grain microorganisms, and on the change of stored-grain quality, etc., jointly determining the stability of the grain pile ecosystem.

[0003] Among the numerous factors in the grain pile ecosystem, temperature, moisture, and humidity are key factors affecting the storage stability of the grain pile. The difference in the temperature distribution of the grain pile will cause the migration of grain moisture, resulting in the flow of micro-airflow in the pores of the grain pile, causing the moisture to transfer from the high-temperature area to the low-temperature area; and the humidity to transfer from the low-temperature area to the high-temperature area. Local accumulations of temperature, moisture, and humidity are formed inside the grain pile, inducing grain mildew, heating, and pest infestation, posing a major threat to the safety of stored grains. Therefore, mastering the real-time state changes of the temperature field, moisture field, and humidity field in the grain pile and using the multi-field coupling theory to predict and analyze their state change laws are of great significance for realizing safe grain storage.

[0004] Currently, numerical simulation methods and in-situ experimental methods are mainly used to detect the multi-field coupling effects in the grain pile. The numerical simulation method detects the change laws of the fields in the grain pile by establishing two-dimensional or three-dimensional mathematical models and using simulation analysis software such as CFD and COMSOL for numerical simulation. This type of technology often relies on idealized mathematical models, makes assumptions and simplifications for some grain pile parameters, and requires a lot of time to adjust the parameters, unable to achieve real-time and accurate detection, and cannot be better applied in practice. The in-situ experimental method mostly focuses on the detection of the temperature and humidity fields in the grain pile, detecting the changes in temperature and humidity in the grain pile through temperature and humidity sensors. This type of technology has a long data sampling period, is time-consuming and laborious, and the temperature and humidity sensors are easily affected by external environmental factors and interference signals, making it difficult to ensure the accuracy of the detection results. In addition, the existing grain pile multi-field coupling detection technologies also cannot achieve real and accurate detection of the grain pile moisture field, resulting in the change of the grain pile moisture field under multi-field coupling still being in the "black box" stage. Therefore, there is an urgent need to invent a grain pile multi-field coupling detection system that can precisely detect the three parameters of temperature, humidity, and moisture in the grain pile, providing a new technical means for the research of multi-field coupling effects in the grain pile. Summary of the Invention

[0005] The present invention designs and develops a graphical detection system and method for multi-field coupling in a grain pile based on NMR. A granary simulation device is used to simulate the actual granary environment; a nuclear magnetic resonance imaging analyzer is used to collect data on the moisture change of the grain pile during the grain storage period; multiple temperature sensors with a fixed spacing are used to collect data on the temperature change of the grain pile during the grain storage period. A multi-field coupling mathematical model is used to calculate the humidity change data of the grain pile, realizing the precise detection of the three parameters of temperature, humidity, and moisture in the grain pile, and the cloud maps of the temperature field, moisture field, and humidity field of the grain pile can be reproduced through the supporting cloud map generation software. Through this system, the mutual coupling mechanism between the temperature field, humidity field, and moisture field in the grain pile under various grain storage scenario conditions can be deeply studied, providing a new technical means for the study of the multi-field coupling effect in the grain pile.

[0006] The NMR-based graphical detection system for multi-field coupling in a granary of the present invention is characterized in that it is composed of a granary simulation device A, a nuclear magnetic resonance imaging analyzer B, and a data acquisition and processing system C. Among them: in the working state, the granary simulation device A is placed in the large-bore detection coil B2 of the nuclear magnetic resonance imaging analyzer B; the data acquisition and processing system C is built into the control cabinet B3 of the nuclear magnetic resonance imaging analyzer B and is communicatively connected to the granary simulation device A and the large-bore detection coil B2 of the nuclear magnetic resonance imaging analyzer B.

[0007] The granary simulation device A is composed of a grain storage tank A1, a second detection circuit board group A2, a first detection circuit board A3, a top cover assembly A4, and a plug-in 1. Among them:

[0008] The grain storage tank A1 is an open-ended cylindrical shape, and the inner surface of the upper end of its shell 2 is provided with four bosses of a uniformly distributed boss group 3;

[0009] The second detection circuit board group A2 is composed of 24 detection circuit board assemblies with the same structure. Each detection circuit board assembly is composed of a circuit board 4, five temperature sensors of a temperature sensor group 5, and a light-emitting diode 6. The five temperature sensors of the temperature sensor group 5 are uniformly distributed on the upper part of the circuit board 4, and the light-emitting diode 6 is fixedly connected to the lower end near the circuit board 4;

[0010] The first detection circuit board A3 is circular, and it is provided with 24 holes of a positioning hole group 7, which are distributed on three concentric circles with the center of the circle. From the inside to the outside, there are 4, 8, and 12 respectively; four bosses of a boss group 8 are uniformly distributed on the circumference;

[0011] The top cover A4 is umbrella-shaped, with a handle 9 fixedly connected near the front end on the upper surface, a threaded hole 10 on the right side of the upper surface, four chutes of a chute group 11 and four grooves of a groove group 12 on the lower end surface;

[0012] The grain storage tank A1, the second detection circuit board group A2, the first detection circuit board A3, and the top cover assembly A4 are arranged in sequence from bottom to top. The upper ends of the 24 circuit boards of the second detection circuit board group A2 are fixedly connected corresponding to the 24 holes of the first detection circuit board A3. In the working state, the 24 circuit boards of the second detection circuit board group A2 are located inside the grain storage tank A1, and the four bosses of the boss group 8 in the first detection circuit board A3 are in contact with the inner surface of the grain storage tank A1.

[0013] The first detection circuit board A3 is communicatively connected to the plug-in unit 1.

[0014] The plug-in unit 1 is threadedly connected to the threaded hole 10 of the top cover A4.

[0015] The nuclear magnetic resonance imaging analyzer B is composed of a C-shaped large-bore cavity magnet B1, a large-aperture detection coil B2, a control cabinet B3, a display B4, and a coil support B5. Among them: the lower end of the large-aperture detection coil B2 is fixedly connected to the upper end of the coil support B5; the coil support B5 is placed at the central position of the C-shaped large-bore cavity magnet B1; the control cabinet B3 is placed on the right side of the C-shaped large-bore cavity magnet B1 and is communicatively connected to the large-aperture detection coil B2; the display B4 is placed on the upper end of the control cabinet B3 and is communicatively connected to the control cabinet B3.

[0016] The data acquisition and processing system C is composed of an industrial computer C1, nuclear magnetic resonance imaging software C2, temperature data acquisition software C3, and cloud map generation software C4. Among them: the industrial computer C1 is built into the control cabinet B3 and is communicatively connected to the large-aperture detection coil B2 and the grain bin simulation device A; the nuclear magnetic resonance imaging software C2 and the temperature data acquisition software C3 are communicatively connected to the industrial computer C1; the cloud map generation software C4 is communicatively connected to the nuclear magnetic resonance imaging software C2 and the temperature data acquisition software C3; the cloud map generation software C4 includes a data processing module 13, a cloud map generation module 14, and a data storage module 15. The cloud map generation module 14 is communicatively connected to the data processing module 13 and the display B4, and the cloud map generation module 14 and the data storage module 15 are communicatively connected to each other.

[0017] The detection method of the NMR-based multi-field coupling graphical detection system for grain bins includes the following steps:

[0018] 1) Load a certain height of grain, the second detection circuit board group A2, and the first detection circuit board A3 into the grain storage tank A1, install the top cover A4 and make a sealing treatment, and then put it into a simulation environment for storage simulation;

[0019] 2) Calibrate the instrument parameters of the nuclear magnetic resonance imaging analyzer B, such as the center frequency, magnetic field uniformity, soft and hard radio frequency pulse width, etc., with a standard oil sample;

[0020] 3) Take out the grain bin simulation device A from the simulation environment and place it into the large-diameter detection coil B2 of the nuclear magnetic resonance imaging analyzer B. Run the temperature data acquisition software C3 to collect the temperature distribution data of the grain pile inside the grain bin simulation device A at this moment; run the nuclear magnetic resonance imaging software C2, adjust imaging parameters such as the number of layers, layer thickness, and layer spacing, so that the nuclear magnetic resonance imaging position is consistent with the temperature detection position, and collect the moisture distribution data of the grain pile inside the grain bin simulation device A at this moment;

[0021] 4) Use the data processing module 13 in the cloud map generation software C4 to process the collected temperature distribution data and moisture distribution data of the grain pile. The processing process is as follows:

[0022] 4.1) Classify and sort the collected temperature distribution data of the grain pile according to the cross-section and longitudinal section of the grain pile, and use the following mathematical interpolation model to calculate the temperature data interpolation matrix of the grain pile cross-section;

[0023]

[0024] Where: x i is a given point, and its function value f(x i ) is known, f is a deterministic continuous function; x ∈ R d ; ||x - x i || is the Euclidean norm; λ i ∈ R; n, b ∈ R d ; a ∈ R; φ is a radial basis function;

[0025] 4.2) Extract the moisture distribution image of the grain pile cross-section from the nuclear magnetic source file of the collected moisture distribution data of the grain pile, and perform binary threshold processing to obtain the image contour of the grain pile cross-section; for each image contour, calculate the number of pixels between the left and right endpoints of each row in this image contour row by row, and record the left and right endpoints of the row where the maximum value is located as x0 and x1; calculate the number of pixels between the upper and lower endpoints of each column column by column, and record the upper and lower endpoints of the column where the maximum value is located as y0 and y1; ignore the distortion of the image in the frequency encoding direction (x direction), and only consider the distortion in the phase encoding direction (y direction), then there is:

[0026] The distortion coefficient of the grain pile cross-section image is:

[0027]

[0028] The distortion coefficient of the grain pile longitudinal section image is:

[0029]

[0030] Where: h is the height of the grain pile, and d is the diameter of the grain pile;

[0031] According to the distortion coefficient corresponding to the grain pile cross-section, the position coordinates of each column of pixels in the contour of the grain pile cross-section image are corrected column by column according to the following formula;

[0032] y i =(y i0 -y center )×S + y center

[0033] Where: where y i is the y coordinate of pixel i after correction, y i0 is the initial y coordinate of pixel i, y center is the y coordinate of the center point of the image contour; for the pixel values of the missing pixels after correction, they are filled in using the mathematical interpolation model in 4.1);

[0034] Convert the moisture content distribution data of the corrected grain pile cross-section into actual moisture content distribution data according to the following formula;

[0035]

[0036] Where: M i is the wet basis moisture content of any point on the contour of the grain pile cross-section image, k is the calibration proportionality coefficient, A all is the sum of the nuclear magnetic signal intensity values of all points on the contour of the grain pile cross-section image, b is the calibration constant, A i is the nuclear magnetic signal intensity value of this point, and n is the total number of points on the contour of the grain pile cross-section image;

[0037] According to the position coordinates of the temperature sensors in the grain pile cross-section, extract the data corresponding to the coordinate points from the actual moisture content distribution data of the grain pile cross-section, and use the mathematical interpolation model in 4.1) to calculate the actual moisture content data interpolation matrix of the grain pile cross-section;

[0038] 4.3) According to the position coordinates of the temperature sensors in the grain pile cross-section, extract the data corresponding to the coordinate points from the temperature data interpolation matrix and the actual moisture content interpolation matrix corresponding to the grain pile cross-section, and calculate the relative humidity data at the specified position in the grain pile cross-section according to the following three-parameter ERH model;

[0039]

[0040] Where: ERH i is the relative humidity of the i-th point in the grain pile cross-section, EMC i is the wet basis moisture content of the grain at this point, T i is the ambient temperature (grain temperature) at this point, and A, B, and C are the equation parameters;

[0041] Substitute the calculated relative humidity data and coordinates of the grain pile cross-section into the mathematical interpolation model in 4.1) to calculate the relative humidity data interpolation matrix of the grain pile cross-section;

[0042] 5) Based on the temperature data interpolation matrix of the grain pile cross-section, the actual moisture content data interpolation matrix of the grain pile cross-section, and the relative humidity data interpolation matrix of the grain pile cross-section, use the cloud map generation module 14 in the cloud map generation software C4 to reproduce the distribution cloud maps of the temperature field, moisture field, and relative humidity field of the grain pile cross-section, and display them on the display B4 of the nuclear magnetic resonance imaging analyzer B;

[0043] 6) During the storage simulation period, repeat steps 2) to 5) at regular time intervals, and use the grain condition cloud map technology to graphically detect the multi-field coupling effect inside the grain pile during the storage period;

[0044] 7) When the storage simulation period ends, store data such as the collection date, collection time, grain pile cross-section identifier, temperature data interpolation matrix, actual moisture content data interpolation matrix, relative humidity data interpolation matrix, temperature field cloud map, moisture field cloud map, and relative humidity field cloud map in the data storage module 15 of the cloud map generation software C4 for subsequent query and use.

[0045] The beneficial effects of the present invention are as follows:

[0046] 1. A multi-field coupling graphical detection system for a granary based on NMR provided by the present invention uses nuclear magnetic resonance imaging technology to detect moisture changes inside the grain pile, uses temperature sensors to detect temperature changes inside the grain pile, calculates humidity changes inside the grain pile through a multi-field coupling model, and uses cloud map technology to monitor grain condition changes in the grain pile, realizing real and accurate detection of the three parameters of temperature, humidity, and moisture in the grain pile, and providing a new technical means for the research of multi-field coupling effects in the grain pile.

[0047] 2. The present invention first uses low-field nuclear magnetic resonance imaging technology for detecting the moisture field inside the grain pile, which can avoid problems such as low detection accuracy caused by the susceptibility of existing technology to environmental temperature and interference signal disturbances when using moisture sensors for detection; and difficulties in sampling, time-consuming and laborious, and damage to the storage state of the grain pile when using the sampling method; realizing rapid, non-destructive, and high-precision detection of the moisture in the grain pile.

[0048] 3. The granary simulation device in the present invention is designed according to the structural dimensions of the actual granary, and is smaller in volume compared with the actual granary, facilitating the simulation of various grain storage scenarios and the precise control of grain storage boundary conditions, making the multi-field coupling results of the grain pile obtained through the granary simulation device closer to the actual grain storage situation. Description of the Drawings

[0049] Figure 1 It is a schematic structural diagram of a multi-field coupling graphical detection system for a granary based on NMR;

[0050] Figure 2 It is a structural schematic diagram of a granary simulation device;

[0051] Figure 3 It is a structural schematic diagram of a grain storage tank;

[0052] Figure 4 It is a structural schematic diagram of a circuit board;

[0053] Figure 5 It is a structural schematic diagram of a first detection circuit board;

[0054] Figure 6 It is a structural schematic diagram of a top cover;

[0055] Figure 7 It is a structural schematic diagram of a data acquisition and processing system;

[0056] Figure 8 It is a flowchart of a detection method for a multi-field coupling graphical detection system of a granary based on NMR;

[0057] Figure 9 It is a schematic diagram of a cloud map generation software interface;

[0058] Figure 10 It is a cloud map of the temperature field distribution of the cross-section of a grain pile in the embodiment;

[0059] Figure 11 It is a cloud map of the moisture field distribution of the cross-section of a grain pile in the embodiment;

[0060] Figure 12 It is a cloud map of the humidity field distribution of the cross-section of a grain pile in the embodiment;

[0061] Figure 13 It is a cloud map of the temperature field distribution of the mid-vertical plane of a grain pile in the embodiment;

[0062] Figure 14 It is a cloud map of the moisture field distribution of the mid-vertical plane of a grain pile in the embodiment;

[0063] Figure 15 It is a cloud map of the humidity field distribution of the mid-vertical plane of a grain pile in the embodiment;

[0064] Wherein: A. Grain bin simulation device A1. Grain storage tank A2. Second detection circuit board group A3. First detection circuit board A4. Top cover B. Nuclear magnetic resonance imaging analyzer B1. C-type large-aperture cavity magnet B2. Large-aperture detection coil B3. Control cabinet B4. Display B5. Coil bracket C. Data acquisition and processing system C1. Industrial control computer C2. Nuclear magnetic resonance imaging software C3. Temperature data acquisition software C4. Cloud map generation software 1. Connector 2. Housing 3. Set of bosses 4. Circuit board 5. Temperature sensor 6. Light-emitting diode 7. Positioning hole 8. Set of bosses 9. Handle 10. Threaded hole 11. Set of chutes 12. Set of grooves 13. Data processing module 14. Cloud map generation module 15. Data storage module. Detailed implementation manner

[0065] The present invention will be described below with reference to the accompanying drawings.

[0066] As Figure 1 shown, the NMR-based multi-field coupling graphical detection system for grain bins of the present invention is characterized in that: it is composed of a grain bin simulation device A, a nuclear magnetic resonance imaging analyzer B, and a data acquisition and processing system C, wherein: in the working state, the grain bin simulation device A is placed in the large-aperture detection coil B2 of the nuclear magnetic resonance imaging analyzer B; the data acquisition and processing system C is built in the control cabinet B3 of the nuclear magnetic resonance imaging analyzer B and is communicatively connected to the grain bin simulation device A and the large-aperture detection coil B2 of the nuclear magnetic resonance imaging analyzer B.

[0067] As Figure 2 shown, the grain bin simulation device A is composed of a grain storage tank A1, a second detection circuit board group A2, a first detection circuit board A3, a top cover assembly A4, and a connector 1, wherein:

[0068] As Figure 3 shown, the grain storage tank A1 is an open-ended cylindrical shape, and the inner surface of the upper end of its housing 2 is provided with 4 bosses of the evenly distributed set of bosses 3;

[0069] The second detection circuit board group A2 is composed of 24 detection circuit board assemblies with the same structure. As Figure 4 shown, each detection circuit board assembly is composed of a circuit board 4, 5 temperature sensors of the temperature sensor group 5, and a light-emitting diode 6. The 5 temperature sensors of the temperature sensor group 5 are evenly distributed on the upper part of the circuit board 4, and the light-emitting diode 6 is fixedly connected to the lower end near the circuit board 4;

[0070] As Figure 5 shown, the first detection circuit board A3 is circular, and it is provided with 24 holes of the positioning hole group 7, and they are distributed on three concentric circles with the center of the circle. From the inside to the outside, there are 4, 8, and 12 respectively; 4 bosses of the set of bosses 8 are evenly distributed on the circumference;

[0071] As shown Figure 6 in FIG. 1, the top cover A4 is umbrella-shaped, fixedly connected with a handle 9 near the front end on the upper surface, provided with a threaded hole 10 on the right side of the upper surface, and provided with 4 chutes of a chute group 11 and 4 grooves of a groove group 12 on the lower end surface;

[0072] The grain storage tank A1, the second detection circuit board group A2, the first detection circuit board A3 and the top cover assembly A4 are arranged in sequence from bottom to top. The upper ends of the 24 circuit boards of the second detection circuit board group A2 are fixedly connected corresponding to the lower sides of 24 holes of the first detection circuit board A3. In the working state, the 24 circuit boards of the second detection circuit board group A2 are located in the grain storage tank A1, and the 4 bosses of the boss group 8 in the first detection circuit board A3 are in contact with the inner surface of the grain storage tank A1;

[0073] The first detection circuit board A3 is communicatively connected with the connector 1;

[0074] The connector 1 is threadedly connected with the threaded hole 10 of the top cover A4.

[0075] As shown Figure 7 in FIG. 2, the described data acquisition and processing system C is composed of an industrial control computer C1, a nuclear magnetic resonance imaging software C2, a temperature data acquisition software C3 and a cloud map generation software C4. The interface of the cloud map generation software C4 is as shown Figure 9 in FIG. 3, wherein: the industrial control computer C1 is built in the control cabinet B3 and communicatively connected with the large-aperture detection coil B2 and the grain bin simulation device A; the nuclear magnetic resonance imaging software C2 and the temperature data acquisition software C3 are communicatively connected with the industrial control computer C1; the cloud map generation software C4 is communicatively connected with the nuclear magnetic resonance imaging software C2 and the temperature data acquisition software C3; the cloud map generation software C4 includes a data processing module 13, a cloud map generation module 14 and a data storage module 15. The cloud map generation module 14 is communicatively connected with the data processing module 13 and the display B4, and the cloud map generation module 14 and the data storage module 15 are communicatively connected with each other.

[0076] The detection method of the described NMR-based multi-field coupling graphical detection system for grain bins, the process of which is as shown Figure 8 in FIG. 4, includes the following steps:

[0077] 1) Load a certain height of grain, the second detection circuit board group A2 and the first detection circuit board A3 into the grain storage tank A1, install the top cover A4 and perform a sealing treatment, and then put it into a simulation environment for storage simulation;

[0078] 2) Calibrate the instrument parameters such as the center frequency, magnetic field uniformity, and soft and hard radio frequency pulse widths of the nuclear magnetic resonance imaging analyzer B with a standard oil sample;

[0079] 3) Take out the grain bin simulation device A from the simulation environment and place it into the large-diameter detection coil B2 of the nuclear magnetic resonance imaging analyzer B. Run the temperature data acquisition software C3 to collect the temperature distribution data of the grain pile in the grain bin simulation device A at this moment; run the nuclear magnetic resonance imaging software C2, adjust the imaging parameters such as the number of layers, layer thickness, and layer spacing, so that the nuclear magnetic resonance imaging position is consistent with the temperature detection position, and collect the moisture distribution data of the grain pile in the grain bin simulation device A at this moment;

[0080] 4) Use the data processing module 13 in the cloud map generation software C4 to process the collected grain pile temperature distribution data and grain pile moisture distribution data. The processing process is as follows:

[0081] 4.1) Classify and sort the collected grain pile temperature distribution data according to the cross-section and longitudinal section of the grain pile, and use the following mathematical interpolation model to calculate the temperature data interpolation matrix of the grain pile cross-section;

[0082]

[0083] Where: x i is the given point, and its function value f(x i ) is known, f is a deterministic continuous function; x ∈ R d ; ||x - x i || is the Euclidean norm; λ i ∈ R; n, b ∈ R d ; a ∈ R; φ is the radial basis function;

[0084] 4.2) Extract the moisture distribution image of the grain pile cross-section from the nuclear magnetic source file of the collected grain pile moisture distribution data, and perform binary threshold processing to obtain the image contour of the grain pile cross-section; for each image contour, calculate the number of pixels between the left and right endpoints of each row in this image contour row by row, and record the left and right endpoints of the row where the maximum value is located as x0 and x1; calculate the number of pixels between the upper and lower endpoints of each column column by column, and record the upper and lower endpoints of the column where the maximum value is located as y0 and y1; ignoring the distortion of the image in the frequency encoding direction (x direction), only considering the distortion in the phase encoding direction (y direction), then there is:

[0085] The distortion coefficient of the grain pile cross-section image is:

[0086]

[0087] The distortion coefficient of the grain pile longitudinal section image is:

[0088]

[0089] Where: h is the height of the grain pile, and d is the diameter of the grain pile;

[0090] According to the distortion coefficient corresponding to the grain pile cross-section, the position coordinates of each column of pixels in the contour of the grain pile cross-section image are corrected column by column according to the following formula;

[0091] y i =(y i0 y center )×S + y center

[0092] Where: where y i is the corrected y coordinate of pixel i, y i0 is the initial y coordinate of pixel i, y center is the y coordinate of the center point of the image contour; for the pixel values of the missing pixels after correction, they are completed using the mathematical interpolation model in 4.1);

[0093] Convert the moisture distribution data of the corrected grain pile cross-section into actual moisture content distribution data according to the following formula;

[0094]

[0095] Where: M i is the wet basis moisture content of any point on the contour of the grain pile cross-section image, k is the calibration proportionality coefficient, A all is the sum of the nuclear magnetic signal intensity values of all points on the contour of the grain pile cross-section image, b is the calibration constant, A i is the nuclear magnetic signal intensity value of this point, and n is the total number of points on the contour of the grain pile cross-section image;

[0096] According to the position coordinates of the temperature sensors in the grain pile cross-section, extract the data corresponding to the coordinate points from the actual moisture content distribution data of the grain pile cross-section, and calculate the actual moisture content data interpolation matrix of the grain pile cross-section using the mathematical interpolation model in 4.1);

[0097] 4.3) According to the position coordinates of the temperature sensors in the grain pile cross-section, extract the data corresponding to the coordinate points from the temperature data interpolation matrix and the actual moisture content interpolation matrix corresponding to the grain pile cross-section, and calculate the relative humidity data at the specified position in the grain pile cross-section according to the following three-parameter ERH model;

[0098]

[0099] Where: ERH i is the relative humidity of the i-th point in the grain pile cross-section, EMC i is the wet basis moisture content of the grain at this point, T i is the ambient temperature (grain temperature) at this point, and A, B, and C are the equation parameters;

[0100] Substitute the calculated relative humidity data and coordinates of the grain pile cross-section into the mathematical interpolation model in 4.1) to calculate the relative humidity data interpolation matrix of the grain pile cross-section;

[0101] 5) Based on the temperature data interpolation matrix of the grain pile cross-section, the actual moisture content data interpolation matrix of the grain pile cross-section, and the relative humidity data interpolation matrix of the grain pile cross-section, use the cloud map generation module 14 in the cloud map generation software C4 to reproduce the distribution cloud maps of the temperature field, moisture field, and relative humidity field of the grain pile cross-section, and display them on the display B4 of the nuclear magnetic resonance imaging analyzer B;

[0102] 6) During the storage simulation period, repeat steps 2) to 5) at certain time intervals, and use the grain condition cloud map technology to graphically detect the multi-field coupling effect inside the grain pile during the storage period;

[0103] 7) At the end of the storage simulation period, store data such as the collection date, collection time, grain pile cross-section identifier, temperature data interpolation matrix, actual moisture content data interpolation matrix, relative humidity data interpolation matrix, temperature field cloud map, moisture field cloud map, and relative humidity field cloud map in the data storage module 15 of the cloud map generation software C4 for subsequent query and use.

[0104] Example

[0105] In this example, two types of corn with different moisture contents are loaded inside the grain bin simulation device A. The height of the grain pile is 120 mm and the diameter is 140 mm. The corn with a moisture content of 25.21% (w.b.) is located at the center of the grain bin simulation device A, forming a cylinder with a diameter of 40 mm and a height of 120 mm. The moisture content of the remaining corn is 18.78% (w.b.). Place the grain bin simulation device A in an incubator at a temperature of 30 °C to simulate the multi-field coupling effect caused by the relatively high local moisture content in the grain pile during the actual grain storage process. Take out the grain bin simulation device A from the incubator every 1 day and place it inside the large-bore detection coil B2 of the nuclear magnetic resonance imaging analyzer B to collect the temperature distribution data and moisture distribution data inside the grain pile. The experiment was carried out for 15 days. The temperature sensor model used is DS18B20Z+, and the layout spacing on the circuit board 4 is 30 mm; the nuclear magnetic resonance imaging sequence used is the SE sequence, and its main parameters are as follows: the simulation gain RG is 30 dB, the preamplifier gear PRG is 2, the digital gain DRG is 5 dB, the waiting time TR is 1500 ms, the echo time TE is 8 ms, the number of accumulations Averages is 16, the number of frequency encoding steps is 256, the number of phase encoding steps is 192, and the slice thickness is 20 mm. Figures 10 to 12 They are the distribution cloud maps of the temperature field, moisture field, and humidity field of the grain pile cross-section at a position 49 mm from the bottom of the grain pile.

[0106] In another embodiment, only corn with one moisture content is loaded inside the granary simulation device A. The height of the grain pile is 120 mm, the diameter is 140 mm, the moisture content of the corn is 16.64% (w.b.), and the remaining conditions remain unchanged. Figures 13 to 15 They are the temperature field distribution nephogram, moisture field distribution nephogram, and humidity field distribution nephogram of the vertical plane of the grain pile in this embodiment.

[0107] Although the embodiments of the present invention have been disclosed above, they are not limited to only the applications listed in the specification and the embodiments. It can be fully applied to various fields suitable for the present invention. For those familiar with the field, additional modifications can be easily achieved. Therefore, without departing from the general concept defined by the claims and the equivalent scope, the present invention is not limited to specific details and the illustrated and described examples here.

Claims

1. An NMR-based graphical detection system for multi-field coupling in a granary, characterized in that: It consists of a grain bin simulation device (A), a nuclear magnetic resonance imaging analyzer (B), and a data acquisition and processing system (C). Among them: The grain bin simulation device (A) consists of a grain storage tank (A1), a second detection circuit board group (A2), a first detection circuit board (A3), a top cover assembly (A4), and a connector (1). Among them: The grain storage tank (A1) is an open-ended cylindrical shape, and the inner surface of the upper end of its shell (2) is provided with four bosses of a uniformly distributed boss group (3); The second detection circuit board group (A2) consists of 24 detection circuit board assemblies with the same structure. Each detection circuit board assembly consists of a circuit board (4), five temperature sensors of a temperature sensor group (5), and a light-emitting diode (6). The five temperature sensors of the temperature sensor group (5) are uniformly distributed on the upper part of the circuit board (4), and the light-emitting diode (6) is fixedly connected to the lower end near the circuit board (4); The first detection circuit board (A3) is circular, and it is provided with 24 holes of a positioning hole group (7), and they are distributed on three circumferences concentric with the center of the circle. From the inside to the outside, there are 4, 8, and 12 respectively; There are four bosses of a boss group (8) uniformly distributed on the circumference; The top cover (A4) is umbrella-shaped, with a handle (9) fixedly connected near the front end on the upper surface, a threaded hole (10) on the right side of the upper surface, four chutes of a chute group (11) and four grooves of a groove group (12) on the lower end surface; The grain storage tank (A1), the second detection circuit board group (A2), the first detection circuit board (A3), and the top cover assembly (A4) are arranged in sequence from bottom to top. The upper ends of the 24 circuit boards of the second detection circuit board group (A2) are fixedly connected corresponding to the lower sides of the 24 holes of the first detection circuit board (A3); In the working state, the 24 circuit boards of the second detection circuit board group (A2) are located inside the grain storage tank (A1), and the four bosses of the boss group (8) in the first detection circuit board (A3) are in contact with the inner surface of the grain storage tank (A1); The first detection circuit board (A3) is communicatively connected to the connector (1); The connector (1) is threadedly connected to the threaded hole (10) of the top cover (A4); The nuclear magnetic resonance imaging analyzer (B) consists of a C-shaped large-bore cavity magnet (B1), a large-bore detection coil (B2), a control cabinet (B3), a display (B4), and a coil support (B5); The data acquisition and processing system (C) consists of an industrial control computer (C1), nuclear magnetic resonance imaging software (C2), temperature data acquisition software (C3), and cloud map generation software (C4); The grain bin simulation device (A) is placed in the large-bore detection coil (B2) of the nuclear magnetic resonance imaging analyzer (B); The industrial control computer (C1) of the data acquisition and processing system (C) is built into the control cabinet (B3) of the nuclear magnetic resonance imaging analyzer (B), and is communicatively connected to the grain bin simulation device (A) and the large-bore detection coil (B2) of the nuclear magnetic resonance imaging analyzer (B).

2. The NMR-based multi-field coupling graphical detection system for grain bins according to claim 1, wherein: The lower end of the large-diameter detection coil (B2) of the nuclear magnetic resonance imaging analyzer (B) is fixedly connected to the upper end of the coil bracket (B5); the coil bracket (B5) is placed at the central position of the C-shaped large-hole cavity magnet (B1); the control cabinet (B3) is placed on the right side of the C-shaped large-hole cavity magnet (B1) and is communicatively connected to the large-diameter detection coil (B2); the display (B4) is placed on the upper end of the control cabinet (B3) and is communicatively connected to the control cabinet (B3).

3. The NMR-based multi-field coupling graphical detection system for grain bins according to claim 1, wherein: The nuclear magnetic resonance imaging software (C2), temperature data acquisition software (C3) of the data acquisition and processing system (C) are communicatively connected to the industrial control computer (C1); the cloud map generation software (C4) is communicatively connected to the nuclear magnetic resonance imaging software (C2), temperature data acquisition software (C3); the cloud map generation software (C4) includes a data processing module (13), a cloud map generation module (14) and a data storage module (15), the cloud map generation module (14) is communicatively connected to the data processing module (13) and the display (B4), and the cloud map generation module (14) and the data storage module (15) are communicatively connected to each other.

4. A detection method for the multi-field coupling graphical detection system of a granary based on NMR according to claim 1, characterized in that It includes the following steps: 1) Load a certain height of grain, the second detection circuit board group (A2) and the first detection circuit board (A3) into the grain storage tank (A1), install the top cover (A4) and perform a sealing treatment, and then place it in a simulated environment for storage simulation; 2) Calibrate the instrument parameters such as the center frequency, magnetic field uniformity, soft and hard radio frequency pulse widths of the nuclear magnetic resonance imaging analyzer (B) with a standard oil sample; 3) Take out the grain bin simulation device (A) from the simulated environment and place it into the large-diameter detection coil (B2) of the nuclear magnetic resonance imaging analyzer (B). Run the temperature data acquisition software (C3) to collect the temperature distribution data of the grain pile in the grain bin simulation device (A) at this moment; run the nuclear magnetic resonance imaging software (C2), adjust the imaging parameters such as the number of layers, layer thickness, and layer spacing, so that the nuclear magnetic resonance imaging position is consistent with the temperature detection position, and collect the moisture distribution data of the grain pile in the grain bin simulation device (A) at this moment; 4) Use the data processing module (13) in the cloud map generation software (C4) to process the collected grain pile temperature distribution data and grain pile moisture distribution data. The processing process is as follows: 4.1) Classify and sort the collected grain pile temperature distribution data according to the cross-section and longitudinal section of the grain pile, and use the following mathematical interpolation model to calculate the temperature data interpolation matrix of the grain pile cross-section; Where: x i is a given point, and its function value f(x i ) is known, and f is a deterministic continuous function; x ∈ R d ; ||x - x i || is the Euclidean norm; λ i ∈ R; n, b ∈ R d ; a ∈ R; φ is a radial basis function; 4.2) Extract the moisture distribution image of the grain pile cross-section from the nuclear magnetic source file of the collected grain pile moisture distribution data, and perform binary threshold processing to obtain the image contour of the grain pile cross-section; for each image contour, calculate the number of pixels between the left and right endpoints of each row in this image contour row by row, and record the left and right endpoints of the row where the maximum value is located as x0 and x1; calculate the number of pixels between the upper and lower endpoints of each column column by column, and record the upper and lower endpoints of the column where the maximum value is located as y0 and y1; ignoring the distortion of the image in the frequency encoding direction - the x direction, only considering the distortion in the phase encoding direction - the y direction, then there is: The distortion coefficient of the grain pile cross-sectional image is: The distortion coefficient of the longitudinal section image of the grain pile is as follows: Where: h is the height of the grain pile, and d is the diameter of the grain pile; According to the distortion coefficient corresponding to the grain pile cross-section, the position coordinates of each column of pixels in the contour of the grain pile cross-section image are corrected column by column according to the following formula; y i = (y i0 - y center ) × S + y center Where: where y i is the corrected y - coordinate of pixel i, and y i0 is the initial y - coordinate of pixel i, and y center is the y - coordinate of the center point of the image contour; for the pixel values of the missing pixels after correction, they are complemented using the mathematical interpolation model in 4.1). The corrected moisture distribution data of the grain pile cross-section is converted into actual moisture content distribution data according to the following formula; Where: M i is the wet basis moisture content of any point on the contour of the grain pile cross-sectional image, k is the calibration proportionality coefficient, A all is the sum of the nuclear magnetic signal intensity values of all points on the contour of the grain pile cross-sectional image, b is the calibration constant, A i is the nuclear magnetic signal intensity value of this point, and n is the total number of points on the contour of the grain pile cross-sectional image; According to the position coordinates of the temperature sensors in the grain pile cross-section, the data corresponding to the coordinate points are extracted from the actual moisture content distribution data of the grain pile cross-section, and the actual moisture content data interpolation matrix of the grain pile cross-section is calculated using the mathematical interpolation model in 4.1); 4.3) According to the position coordinates of the temperature sensors in the grain pile cross-section, the data corresponding to the coordinate points are extracted from the temperature data interpolation matrix and the actual moisture content interpolation matrix corresponding to the grain pile cross-section, and the relative humidity data at the specified position in the grain pile cross-section is calculated according to the following three-parameter ERH model; Where: ERH i is the relative humidity at the i-th point in the cross-section of the grain heap, EMC i is the moisture content on a wet basis of the grain at this point, T i is the grain temperature of the environment where this point is located, and A, B, and C are equation parameters; The calculated relative humidity data and coordinates of the grain pile cross-section are substituted into the mathematical interpolation model in 4.1) to calculate the relative humidity data interpolation matrix of the grain pile cross-section; 5) Based on the temperature data interpolation matrix of the grain pile cross-section, the actual moisture content data interpolation matrix of the grain pile cross-section, and the relative humidity data interpolation matrix of the grain pile cross-section, the distribution cloud maps of the temperature field, moisture field, and relative humidity field of the grain pile cross-section are reproduced through the cloud map generation module (14) in the cloud map generation software (C4) and displayed on the display (B4) of the nuclear magnetic resonance imaging analyzer (B); 6) During the storage simulation period, repeat steps 2) to 5) at a certain time interval, and use the grain condition cloud map technology to graphically detect the multi-field coupling effect inside the grain pile during the storage period; 7) At the end of the storage simulation period, data such as the collection date, collection time, grain pile cross-section identification, temperature data interpolation matrix, actual moisture content data interpolation matrix, relative humidity data interpolation matrix, temperature field cloud map, moisture field cloud map, and relative humidity field cloud map are stored in the data storage module (15) of the cloud map generation software (C4) for subsequent query and use.

Images