Grain moisture online nondestructive testing method

Abstract

The invention discloses an online nondestructive detection method for grain moisture. The method comprises the following steps: step a), acquiring frequency data generated by a parallel polar plate device; b) acquiring environmental data such as temperature and humidity data and CO2 concentration data acquired by an environmental monitoring sensor in the closed channel; c) acquiring strain data generated by a pressure strain gauge on the inner side of the top of the closed channel; step d), calculating the moisture content of the grain based on the frequency data; a plurality of groups of parallel polar plate devices are arranged, one group of parallel polar plate devices are positioned in the closed channel, and the other parallel polar plate devices are positioned in the through channel which is through up and down; the water content is calculated by taking frequency data generated based on the parallel pole plate device in the closed channel as a basic frequency and based on frequency data generated based on other parallel pole plate devices; and e) correcting the moisture content by using the environmental data and the strain data to obtain real-time moisture content data. The detection method provided by the invention can accurately detect the moisture content of the grains.

Description

Technical Field

[0001] The present invention relates to the technical field of grain moisture content detection, and in particular to an online non-destructive detection method for grain moisture. Background Art

[0002] Grain moisture testing is a critical step in ensuring grain storage safety and quality. However, existing grain moisture testing methods and circuit systems have many shortcomings, making it difficult to meet the requirements of efficient, accurate, and non-destructive testing in modern grain storage and processing.

[0003] Traditional methods for measuring grain moisture are primarily direct and indirect. Direct methods, such as thermal drying and chemical methods, while highly accurate, are time-consuming, cumbersome, and unsuitable for online and on-site testing. For example, thermal drying requires drying grain samples at a specific temperature for extended periods, making it inadequate for rapid testing. Chemical methods, on the other hand, are less commonly used due to the high cost of chemical reagents, potential for corrosion to instruments, and high maintenance costs.

[0004] The indirect method indirectly determines the moisture content of grains by detecting physical quantities related to moisture (such as conductivity, dielectric constant, etc.). It has the advantages of fast speed and easy online detection, but it has the problems of low accuracy and interference from various factors. For example, although the existing capacitive moisture detection device has a fast detection speed, the design of the structure and signal acquisition circuit is easily affected by factors such as grain density, particle size, ambient temperature, vibration, etc., resulting in a decrease in detection accuracy. Therefore, capacitive detection devices are usually used for testing in laboratories. Operators are required to strictly follow the prescribed material sampling and testing procedures. The operation is cumbersome and time-consuming, and has not been effectively used in the online detection of grain moisture.

[0005] In the prior art, the invention patent with publication number CN112198199A proposes a moisture detection device based on IQ modulation technology, which separates the resistance component and the capacitance component through a capacitive sensor, and combines temperature compensation to achieve rapid detection of grain moisture content. However, this method still has significant limitations in detection accuracy and stability under complex storage environments. Specifically, the patent relies on the measurement data of a single capacitive sensor and does not consider the impact of grain porosity on dielectric properties. Since the difference in porosity between grain particles will directly change the equivalent dielectric constant, a single capacitance value measurement is difficult to accurately reflect the true moisture content, especially at different depths of the grain pile, which is prone to systematic deviations. In addition, although the comparative document introduces a temperature compensation mechanism, it does not involve the coupling effects of dynamic factors such as ambient humidity and gas composition (such as CO2 concentration). The accumulation of CO2 generated by grain respiration in the storage environment will change the local dielectric environment, and temperature and humidity fluctuations may cause sensor baseline drift. Simple temperature compensation cannot completely eliminate multi-physical field interference. More critically, the comparative document uses a single-point measurement mode and lacks the integration and correction of spatially distributed data. This can easily lead to distorted measurement results when the detection electrode is in uneven contact with the grain or when there is localized condensation. These shortcomings severely limit the applicability and reliability of this technology in online grain monitoring scenarios. Summary of the Invention

[0006] Purpose of the invention: In order to overcome the deficiencies in the prior art, the present invention provides an online non-destructive detection method for grain moisture that can compensate for the moisture content data detected by the plate device based on environmental data and porosity to obtain more accurate moisture content data.

[0007] Technical solution: To achieve the above-mentioned purpose, the present invention provides an online non-destructive detection method for grain moisture, which is implemented by a microcontroller and includes the following steps a) and d):

[0008] Step a), obtaining frequency data generated by the parallel plate device;

[0009] Step d), calculating the moisture content of the grain based on the frequency data;

[0010] There are multiple groups of parallel plate devices, one of which is located in a closed channel, and the other parallel plate devices are located in a through channel extending vertically through the channel. In step d), the moisture content is calculated based on the frequency data generated by the parallel plate device in the closed channel as a base frequency and based on the frequency data generated by the other parallel plate devices.

[0011] The method further comprises the following steps b) and c) before step d):

[0012] Step b) obtaining environmental data such as temperature and humidity data and CO2 concentration data collected by environmental monitoring sensors in the closed channel;

[0013] Step c) obtaining strain data generated by a pressure strain gauge located on the inner side of the top of the closed channel;

[0014] The method further comprises the following step e) after step d):

[0015] Step e) correcting the moisture content using the environmental data and the strain data to obtain real-time moisture content data.

[0016] Furthermore, before step d), the method further comprises the following steps:

[0017] determining whether there is grain in the through passage based on the frequency data and / or the strain data, and obtaining a determination result;

[0018] When the judgment result is yes, continue to implement the subsequent steps;

[0019] When the judgment result is no, return to step a).

[0020] When determining whether there is food in the through-channel based on the frequency data, the difference between the frequency data generated by the parallel plate device in the through-channel and the reference data can be calculated to see if it is less than a preset threshold. If so, it is determined that there is no food in the through-channel. When determining whether there is food in the through-channel based on the strain data, a pressure value can be calculated based on the strain data. Based on the pressure value, it can be determined whether there is food above the closed channel, and the presence of food in the through-channel can be determined accordingly.

[0021] Furthermore, the strain gauge has a strain bridge (Wheatstone bridge), and the strain bridge is connected to the microcontroller via a strain acquisition circuit;

[0022] The strain acquisition circuit includes a bridge conversion circuit, which includes a first selection switch K1 of a bridge signal processing circuit connected to each other, and also includes a first resistor R1 and a second resistor R2 connected in series, with a line extending between the two resistors and connected to a first selection end of the first selection switch K1, and the second selection end of the first selection switch K1 being connected to the positive output end of the strain bridge; the first resistor R1 and the second resistor R2 are respectively connected to the positive excitation end and the negative excitation end of the strain bridge, and a matching resistor selection circuit is provided between the negative excitation end and the second resistor R2, the matching resistor selection circuit including the second selection switch K2 and multiple optional branches, one of which has no matching resistor, and the other branches have matching resistors of different resistance values;

[0023] The method further comprises the following steps preceding step a):

[0024] Step S101: Controlling the first selection switch K1 to connect to the corresponding selection end based on the type of the strain gauge bridge. Specifically, when the strain gauge bridge is a half bridge or a single bridge, controlling the first selection end of the first selection switch K1 to connect; when the strain gauge bridge is a full bridge, controlling the second selection end of the first selection switch K1 to connect;

[0025] Step S102: Determine the connected branch of the second selection switch K2 based on the connectivity of the first selection switch K1. Specifically, when the first selection end of the first selection switch K1 is turned on, the second selection switch K2 is controlled to turn on the branch with the matching resistor. The resistance value of the matching resistor on the turned-on branch is determined according to the model of the strain bridge. At this time, the resistor divider network composed of the first resistor R1 and the second resistor R2 participates in the single-bridge or half-bridge strain measurement, and together with the strain gauge and the matching resistor, forms a strain test full bridge. When the second selection end of the first selection switch K1 is turned on, the second selection switch K2 is controlled to turn on the branch without the matching resistor.

[0026] Furthermore, the strain acquisition circuit further includes a calibration resistor R3 and a third switch K3 for controlling the on / off of a circuit where the calibration resistor R3 is located;

[0027] The step S102 further includes the following steps:

[0028] Step S103, controlling the third switch K3 to be turned on and executing the calibration procedure;

[0029] Step S104: After the calibration is completed, the third switch K3 is controlled to be disconnected.

[0030] The first selection switch K1 , the second selection switch K2 and the third switch K3 are all electrically controlled switches that can be controlled by a microprocessor to change their states.

[0031] After the third switch K3 is turned on, the calibration resistor R3 can be connected in series or in parallel with the strain bridge. Correspondingly, the series calibration method or the parallel calibration method is used to perform zero point calibration on the strain gauge and calibrate the sensitivity of the strain gauge.

[0032] Furthermore, the number of the through channels is three, and the calculation of the moisture content in step d) specifically includes the following steps:

[0033] Step d1) Calculate the reference dielectric constant based on the frequency f0 generated by the parallel plate device in the through channel:

[0034]

[0035] Among them, ε r0 is the base dielectric constant; is the electrode structure constant; is the no-load calibration capacitance; L is the inductive impedance value, C is the capacitance value; d is the distance between the parallel plates; ε0 is the vacuum dielectric constant; S is the effective area of the electrode part;

[0036] Step d2) Calculate the real-time dielectric constant of the parallel plate device in each closed channel:

[0037]

[0038] Among them, ε ri is the real-time dielectric constant of the parallel plate device in the i-th channel; f i is the frequency generated by the parallel plate arrangement in channel i; is the compensation capacitor of channel i, which is used to eliminate the manufacturing tolerance of the electrode;

[0039] Step d3), construct a multi-channel fusion moisture content model:

[0040]

[0041] Where W is the moisture content; k i is the channel space weight coefficient, β is the temperature compensation factor, T is the measured temperature in the closed channel; T0 is the reference temperature, which can be 25°C.

[0042] Furthermore, the step e) of correcting the moisture content using the environmental data and the strain data to obtain real-time moisture content data includes the following steps:

[0043] e1) Calculate real-time void fraction based on strain data:

[0044] in Where K is the strain coefficient, h is the height of the grain pile, ρ s is the typical density of grain; σ(t) is the output voltage of the strain gauge, reflecting the grain pile pressure;

[0045] e2) Construct a dielectric constant model to quantify the effect of pores:

[0046]

[0047] in, is the pore influence value; ε air is the dielectric constant of air;

[0048] e3) Calibrate the moisture content through the dynamic correction function:

[0049]

[0050] Among them, εcal is the calibrated dielectric constant, λ is the pore coupling coefficient;

[0051] e4) Use Kalman filter to fuse CO2 concentration data and update correction parameters:

[0052]

[0053] Among them, λ t is the respiration correction coefficient; γ is the attenuation factor, which suppresses the dielectric shift caused by microbial metabolism in real time; is the carbon dioxide concentration.

[0054] Beneficial effects: The online non-destructive detection method for grain moisture of the present invention has the following beneficial effects:

[0055] (1) This method for online nondestructive detection of grain moisture collects frequency data in closed channels and through channels through multiple sets of parallel plate devices, calculates moisture content based on the closed channel frequency to ensure measurement accuracy, and corrects moisture content by combining temperature, humidity and CO2 concentration data collected by environmental monitoring sensors and strain data generated by pressure strain gauges. It effectively eliminates the interference of environmental factors and grain porosity on moisture measurement, significantly improves the accuracy and reliability of real-time moisture content data, and realizes efficient and nondestructive online monitoring of grain moisture.

[0056] (2) Based on step S101-step S102, flexible adaptation of different types of strain gauges can be achieved, making the entire detection system highly compatible and convenient for subsequent maintenance. When the strain gauge is damaged, another type of strain gauge can be selected for replacement. The microcontroller can control the two selection switches according to the type of strain gauge to enable different paths to be turned on, thereby achieving compatible adaptation.

[0057] (3) Through the weighted fusion of multi-channel dielectric constant differences and the temperature compensation model, the single-point measurement error and the influence of environmental temperature drift can be eliminated, and more accurate moisture content data can be obtained.

[0058] (4) Through the porosity correction model and Kalman filter dynamic optimization, the interference of grain bulk density changes and respiration on dielectric measurement can be eliminated, the fluctuation amplitude of continuously monitored moisture content can be greatly reduced, and the long-term stability can be greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a three-dimensional structural diagram of the detection device;

[0060] Figure 2 It is a cross-sectional structural diagram of the detection device;

[0061] Figure 3 This is a flow chart of the on-line non-destructive detection method for grain moisture;

[0062] Figure 4 This is the structural diagram of the moisture collection circuit;

[0063] Figure 5 This is the structural diagram of the bridge conversion circuit.

[0064] In the figure: 1-detection plate; 2-mounting base; 2a-enclosed channel; 2b-through channel; 2c-hole; 2d-vent; 21-top cover; 3-temperature and humidity sensor. DETAILED DESCRIPTION

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

[0066] The on-line nondestructive detection method of grain moisture of the present invention is based on Figure 1 and Figure 2 The detection device shown is implemented, and the detection device includes four groups of parallel plate devices, each group of parallel plate devices includes two detection plates 1 installed opposite to each other, and the detection device also includes a mounting base 2 for mounting the parallel plate devices, and the mounting base 2 has a closed channel 2a and three through channels 2b. The closed channel 2a and each through channel 2b have parallel plate devices; the closed channel 2a can communicate gas with its through channel 2b, and can communicate gas with the external space of the mounting base 2, and food cannot enter the closed channel 2a, so that the consistency between the environment inside the closed channel 2a and the external space environment can be ensured, and a temperature and humidity sensor 3 and a CO2 concentration sensor are installed in the closed channel 2a, and a strain gauge is installed on the inner side of the top cover 21 of the closed channel 2a, and the strain gauge is a resistance strain gauge; the through channel 2b is through-through from top to bottom, and has a side wall structure on all sides, and the side wall of the through channel 2b has a hole 2c for food to pass through. The vent holes 2d on the sidewalls of the closed channel 2a are all inclined, with one end of the vent, near the through channel 2b, positioned higher, while the other end, which opens into the through channel 2b or the outside world, is positioned lower. The vent holes are smaller than the grain particle size, preventing grain and impurities from entering the closed channel 2a through the vent holes. This ensures consistent environmental conditions within the closed channel 2a, the outside world, and the through channel 2b. To prevent grain from accumulating at the top of the closed channel 2a and causing inaccurate data, the top of the closed channel 2a is sloped.

[0067] The detection device also includes a microcontroller, which is connected to all the above-mentioned parallel plate devices, and is also connected to the above-mentioned temperature and humidity sensor 3, CO2 concentration sensor and strain gauge.

[0068] like Figure 3 The online non-destructive detection method for grain moisture shown is implemented by a microcontroller and includes the following steps a) to e):

[0069] Step a), obtaining frequency data generated by the parallel plate device;

[0070] Step b) obtaining environmental data such as temperature and humidity data and CO2 concentration data collected by the environmental monitoring sensor in the closed channel 2a;

[0071] Step c) obtaining strain data generated by a pressure strain gauge located on the inner side of the top of the closed channel 2a;

[0072] Step d), calculating the moisture content of the grain based on the frequency data; the moisture content is calculated based on the frequency data generated by the parallel plate device in the closed channel 2a as a base frequency and based on the frequency data generated by other parallel plate devices;

[0073] Step e) correcting the moisture content using the environmental data and the strain data to obtain real-time moisture content data.

[0074] This online nondestructive detection method for grain moisture uses multiple sets of parallel plate devices to collect frequency data in the closed channel 2a and the through channel 2b respectively, and calculates the moisture content based on the frequency of the closed channel 2a to ensure measurement accuracy. At the same time, the moisture content is corrected by combining the temperature, humidity and CO2 concentration data collected by the environmental monitoring sensor and the strain data generated by the pressure strain gauge. This effectively eliminates the interference of environmental factors and grain porosity on moisture measurement, significantly improves the accuracy and reliability of real-time moisture content data, and realizes efficient and nondestructive online monitoring of grain moisture.

[0075] Preferably, before step d), the method further includes the following steps: judging whether there is grain in the through channel 2b based on the frequency data and / or the strain data, and obtaining a judgment result; when the judgment result is yes, continuing to implement subsequent steps; when the judgment result is no, returning to step a).

[0076] When determining whether there is food in through-channel 2b based on the frequency data, the difference between the frequency data generated by the parallel plate assembly in through-channel 2b and the reference data can be calculated to see if it is less than a preset threshold. If so, the absence of food in through-channel 2b is determined. When determining whether there is food in through-channel 2b based on the strain data, a pressure value can be calculated based on the strain data. Based on this pressure value, the presence of food above closed channel 2a can be determined, and the presence of food in through-channel 2b can be determined accordingly. Through these steps, the presence of food in through-channel 2b can be determined in real time, and the decision to proceed with subsequent calculation steps can be made accordingly. If there is no food in through-channel 2b, the calculation can be terminated promptly to avoid wasting computing power and generating erroneous data.

[0077] Preferably, the strain gauge has a strain bridge (Wheatstone bridge), and the strain bridge is connected to the microcontroller via a strain acquisition circuit;

[0078] like Figure 4 As shown, the strain acquisition circuit includes a bridge conversion circuit, a signal processing circuit, a signal conversion circuit, a self-calibration circuit, a self-balancing circuit, a bridge excitation circuit, a microprocessor, and a power module. The signal processing circuit amplifies and filters the analog strain signal to eliminate external interference; the signal conversion circuit converts the analog continuous strain signal into a digital signal for reading by the microcontroller; the bridge excitation circuit provides an excitation voltage for the strain bridge; the self-balancing circuit offsets the initial strain output value; the self-calibration circuit calibrates the system voltage measurement accuracy; and the power module provides power to each component circuit at an appropriate voltage. The microcontroller reads the signal conversion module data and manages the coordinated operation of the other circuits.

[0079] like Figure 5 As shown, the bridge conversion circuit includes a first selection switch K1 of a bridge signal processing circuit connected to each other, and also includes a first resistor R1 and a second resistor R2 connected in series. A line connected to the first selection end of the first selection switch K1 extends between the two resistors, and the second selection end of the first selection switch K1 is connected to the positive output end of the strain bridge; the first resistor R1 and the second resistor R2 are respectively connected to the positive excitation end and the negative excitation end of the strain bridge, and a matching resistor selection circuit is provided between the negative excitation end and the second resistor R2. The matching resistor selection circuit includes the second selection switch K2 and multiple optional branches, one of which has no matching resistor, and the other branches have matching resistors of different resistance values; Figure 5 There are three matching resistors, namely matching resistors R4, R5, and R6.

[0080] The method further comprises the following steps preceding step a):

[0081] Step S101: Controlling the first selection switch K1 to connect to the corresponding selection end based on the type of the strain gauge bridge. Specifically, when the strain gauge bridge is a half bridge or a single bridge, controlling the first selection end of the first selection switch K1 to connect; when the strain gauge bridge is a full bridge, controlling the second selection end of the first selection switch K1 to connect;

[0082] Step S102: Determine the connected branch of the second selection switch K2 based on the connectivity of the first selection switch K1. Specifically, when the first selection end of the first selection switch K1 is turned on, the second selection switch K2 is controlled to turn on the branch with the matching resistor. The resistance value of the matching resistor on the turned-on branch is determined according to the model of the strain bridge. At this time, the resistor divider network composed of the first resistor R1 and the second resistor R2 participates in the single-bridge or half-bridge strain measurement, and together with the strain gauge and the matching resistor, forms a strain test full bridge. When the second selection end of the first selection switch K1 is turned on, the second selection switch K2 is controlled to turn on the branch without the matching resistor.

[0083] Based on step S101-step S102, flexible adaptation of different types of strain gauges can be achieved, making the entire detection system highly compatible and convenient for subsequent maintenance. When the strain gauge is damaged, another type of strain gauge can be selected for replacement. The microprocessor can control the two selection switches to connect different paths according to the type of strain gauge, thereby achieving compatible adaptation.

[0084] Preferably, the strain acquisition circuit further includes a calibration resistor R3 and a third switch K3 for controlling the on / off of a circuit where the calibration resistor R3 is located;

[0085] The step S102 further includes the following steps:

[0086] Step S103, controlling the third switch K3 to be turned on and executing the calibration procedure;

[0087] Step S104: After the calibration is completed, the third switch K3 is controlled to be disconnected.

[0088] The first selection switch K1 , the second selection switch K2 and the third switch K3 are all electrically controlled switches that can be controlled by a microprocessor to change their states.

[0089] After the third switch K3 is turned on, the calibration resistor R3 can be connected in series or in parallel with the strain bridge. Correspondingly, the series calibration method or the parallel calibration method is used to perform zero point calibration on the strain gauge and calibrate the sensitivity of the strain gauge.

[0090] Preferably, the calculation of the moisture content in step d specifically includes the following steps:

[0091] Step d1) Calculate the reference dielectric constant based on the frequency f0 generated by the parallel plate device in the through channel 2b:

[0092]

[0093] Among them, ε r0 is the base dielectric constant; is the electrode structure constant; is the no-load calibration capacitance; L is the inductive impedance value, C is the capacitance value; d is the distance between the parallel plates; ε0 is the vacuum dielectric constant; S is the effective area of the electrode part;

[0094] Step d2) Calculate the real-time dielectric constant of the parallel plate device in each closed channel 2a:

[0095]

[0096] Among them, ε ri is the real-time dielectric constant of the parallel plate device in the i-th channel; f i is the frequency generated by the parallel plate arrangement in channel i; is the compensation capacitor of channel i, which is used to eliminate the manufacturing tolerance of the electrode;

[0097] Step d3), construct a multi-channel fusion moisture content model:

[0098]

[0099] Where W is the moisture content; k i is the channel space weight coefficient, β is the temperature compensation factor, T is the measured temperature in the closed channel 2a; T0 is the reference temperature, which can be 25°C.

[0100] Through the weighted fusion of multi-channel dielectric constant differences and temperature compensation model, single-point measurement errors and the influence of environmental temperature drift can be eliminated to obtain more accurate moisture content data.

[0101] Preferably, the step e) of correcting the moisture content by using the environmental data and the strain data to obtain real-time moisture content data comprises the following steps:

[0102] Step e1), calculate the real-time void fraction based on the strain data:

[0103] in Where K is the strain coefficient, h is the grain pile height (the grain pile height can be automatically collected by installing ultrasonic sensors or other detection sensors in the granary), ρ s is the typical density of grain; σt is the output voltage of the strain gauge, reflecting the grain pile pressure;

[0104] Step e2) constructs a dielectric constant model to quantify the effect of pores:

[0105]

[0106] in, is the pore influence value; ε air is the dielectric constant of air;

[0107] Step e3) calibrate the moisture content using a dynamic correction function:

[0108]

[0109] Among them, ε cal is the calibrated dielectric constant, λ is the pore coupling coefficient;

[0110] Step e4) uses a Kalman filter to fuse CO2 concentration data and update the correction parameters:

[0111]

[0112] Among them, λ t is the respiration correction coefficient; γ is the attenuation factor, which suppresses the dielectric shift caused by microbial metabolism in real time; is the carbon dioxide concentration.

[0113] Through the porosity correction model and Kalman filter dynamic optimization, the interference of grain bulk density changes and respiration on dielectric measurement can be eliminated, the fluctuation amplitude of continuously monitored moisture content can be greatly reduced, and the long-term stability can be greatly improved.

[0114] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for online nondestructive detection of grain moisture, implemented by a microcontroller, comprising the following steps a) and d): Step a), obtaining frequency data generated by the parallel plate device; Step d), calculating the moisture content of the grain based on the frequency data; Its characteristics are: There are multiple groups of parallel plate devices, one of which is located in a closed channel, and the other parallel plate devices are located in a through channel extending vertically through the channel. In step d), the moisture content is calculated based on the frequency data generated by the parallel plate device in the closed channel as a base frequency and based on the frequency data generated by the other parallel plate devices. The method further comprises the following steps b) and c) before step d): Step b) obtaining environmental data such as temperature and humidity data and CO2 concentration data collected by environmental monitoring sensors in the closed channel; Step c) obtaining strain data generated by a pressure strain gauge located on the inner side of the top of the closed channel; The method further comprises the following step e) after step d): Step e) correcting the moisture content using the environmental data and the strain data to obtain real-time moisture content data.

2. The method for online nondestructive detection of grain moisture according to claim 1, characterized in that: Before step d), the method further comprises the following steps: Based on the frequency data and / or the strain data, it is determined whether there is food in the through passage to obtain a determination result; when the determination result is yes, the subsequent steps are continued; when the determination result is no, the process returns to step a).

3. The method for online nondestructive detection of grain moisture according to claim 1, characterized in that: The strain gauge has a strain bridge, and the strain bridge is connected to the microcontroller via a strain acquisition circuit; The strain acquisition circuit includes a bridge conversion circuit, which includes a first selection switch K1 of a bridge signal processing circuit connected to each other, and also includes a first resistor R1 and a second resistor R2 connected in series, with a line extending between the two resistors and connected to a first selection end of the first selection switch K1, and the second selection end of the first selection switch K1 being connected to the positive output end of the strain bridge; the first resistor R1 and the second resistor R2 are respectively connected to the positive excitation end and the negative excitation end of the strain bridge, and a matching resistor selection circuit is provided between the negative excitation end and the second resistor R2, the matching resistor selection circuit including the second selection switch K2 and multiple optional branches, one of which has no matching resistor, and the other branches have matching resistors of different resistance values; The method further comprises the following steps preceding step a): Step S101: Controlling the first selection switch K1 to connect to the corresponding selection end based on the type of the strain gauge bridge. Specifically, when the strain gauge bridge is a half bridge or a single bridge, controlling the first selection end of the first selection switch K1 to connect; when the strain gauge bridge is a full bridge, controlling the second selection end of the first selection switch K1 to connect; Step S102: Determine the connected branch of the second selection switch K2 based on the connectivity of the first selection switch K1. Specifically, when the first selection terminal of the first selection switch K1 is turned on, the second selection switch K2 is controlled to connect the branch with the matching resistor; when the second selection terminal of the first selection switch K1 is turned on, the second selection switch K2 is controlled to connect the branch without the matching resistor.

4. The method for online nondestructive detection of grain moisture according to claim 3, characterized in that: The strain acquisition circuit further includes a calibration resistor R3 and a third switch K3 for controlling the on / off of a circuit where the calibration resistor R3 is located; The step S102 further includes the following steps: Step S103, controlling the third switch K3 to be turned on and executing the calibration procedure; Step S104: After the calibration is completed, the third switch K3 is controlled to be disconnected.

5. The method for online nondestructive detection of grain moisture according to claim 1, characterized in that: The number of the through channels is three, and the calculation of the moisture content in step d) specifically includes the following steps: Step d1) Calculate the reference dielectric constant based on the frequency f0 generated by the parallel plate device in the through channel: Among them, ε r0 is the base dielectric constant; is the electrode structure constant; is the no-load calibration capacitance; L is the inductive impedance value, C is the capacitance value; d is the distance between the parallel plates; ε0 is the vacuum dielectric constant; S is the effective area of the electrode part; Step d2) Calculate the real-time dielectric constant of the parallel plate device in each closed channel: Among them, ε ri is the real-time dielectric constant of the parallel plate device in the i-th channel; f i is the frequency generated by the parallel plate arrangement in channel i; is the compensation capacitance of the i-th channel; Step d3), construct a multi-channel fusion moisture content model: Where W is the moisture content; k i is the channel space weight coefficient, β is the temperature compensation factor, T is the measured temperature in the closed channel; T0 is the reference temperature.

6. The method for online nondestructive detection of grain moisture according to claim 5, characterized in that: The step e) of correcting the moisture content using the environmental data and the strain data to obtain real-time moisture content data includes the following steps: e1) Calculate real-time void fraction based on strain data: in Where K is the strain coefficient, h is the height of the grain pile, ρ s is the typical density of grain; σ(t) is the output voltage of the strain gauge; e2) Construct a dielectric constant model to quantify the effect of pores: in, is the pore influence value; ε air is the dielectric constant of air; e3) Calibrate the moisture content through the dynamic correction function: Among them, ε cal is the calibrated dielectric constant, λ is the pore coupling coefficient; e4) Use Kalman filter to fuse CO2 concentration data and update correction parameters: Among them, λ t is the respiration correction coefficient; γ is the attenuation factor, which suppresses the dielectric shift caused by microbial metabolism in real time; is the carbon dioxide concentration.

Images