Grain flow detection device and detection method thereof

Abstract

The invention discloses a grain flow detection device and a detection method thereof, and belongs to the technical field of signal processing. The grain flow detection device comprises a double-plate differential grain flow sensor composed of two groups of Wheatstone bridges; the first group of Wheatstone bridges is composed of four strain gauges attached to the upper rigid body beam, the second group of Wheatstone bridges is composed of four strain gauges attached to the lower rigid body beam, and the first group of Wheatstone bridges is used for measuring voltage signals generated by deformation of the rigid body beams caused by impact of grains on the upper interception plate and change of the strain gauges caused by mechanical vibration; the second group of Wheatstone bridges is used for measuring a voltage signal generated by the change of the variable piece caused by mechanical vibration; the signal acquisition module is used for receiving a voltage signal of the double-plate differential grain flow sensor and transmitting the voltage signal to the signal processing and analyzing module; and the signal processing and analyzing module is used for converting the received voltage signal into numerical information of the grain flow based on a signal difference algorithm and a grain flow detection program so as to measure the grain flow.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and in particular to a grain flow detection device and a detection method thereof. Background Art

[0002] Grain flow measurement is an important technology in agricultural machinery and equipment. It is widely used in harvesters, seeders and other agricultural machinery to monitor flow parameters in real time during crop harvesting and processing. This technology is particularly important in the implementation of precision agriculture and can provide reliable support for crop yield assessment, operation efficiency optimization and agricultural production data analysis. However, in practical applications, grain flow measurement still faces many technical challenges and shortcomings.

[0003] The existing methods for measuring grain flow rate mainly include volumetric method, mass method and impact force detection method. The volumetric method indirectly calculates the material flow rate by measuring the volume flow rate of grains. However, due to the large variations in the density, water content and flow characteristics of grains in different operating environments, the measurement accuracy of the volumetric method is often difficult to guarantee, especially in high-speed flow conditions, where errors are prone to occur. The mass method directly obtains the mass flow rate of grains through a weighing system. Although the accuracy is high, this method requires a complex weighing device and is easily disturbed by mechanical vibration. At the same time, the dynamic response performance is poor, and it is difficult to adapt to high-frequency and real-time measurement requirements. The impact force detection method uses the relationship between the impact force and flow rate of grains to estimate the grain flow rate by detecting the impact force. The structure of this method is relatively simple and the dynamic response is good, but in actual applications, due to differences in grain shape, speed and impact angle, its detection signal will be subject to large fluctuations. In addition, the impact force detection device is usually directly exposed to the harvester operating environment and is easily affected by mechanical vibration, noise interference and environmental changes (such as humidity, temperature, etc.), resulting in unstable measurement data.

[0004] In response to the above problems, in recent years, grain flow detection technology based on strain gauges has gradually attracted attention. By installing strain gauges on fixed beams or brackets, and utilizing the sensitivity of strain gauges to the force generated by grain impact, high-precision flow detection can be achieved. However, in practical applications, this technology also has the disadvantages of difficulty in distinguishing between grain impact force and strain force signals generated by mechanical vibration, large mechanical vibration interference, and insufficient anti-interference ability of the system. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a grain flow detection device and a detection method thereof.

[0006] The specific scheme of the present invention is as follows: A grain flow detection device comprises:

[0007] A dual-plate differential grain flow sensor consisting of two sets of Wheatstone bridges, a signal acquisition module, and a signal processing and analysis module;

[0008] The first group of Wheatstone bridges is composed of four strain gauges attached to the upper rigid beam, and the second group of Wheatstone bridges is composed of four strain gauges attached to the lower rigid beam. The upper and lower rigid beams are connected through a base as an integrated structure and are installed together at the grain outlet of the harvester through a connecting bracket. The upper rigid beam and the lower rigid beam are respectively provided with an upper interception plate and a lower interception plate. The first group of Wheatstone bridges and the second group of Wheatstone bridges respectively constitute two two-dimensional force sensors. The first group of Wheatstone bridges is used to measure the voltage signal generated by the deformation of the rigid beam caused by the impact of grains on the upper interception plate and the change of the strain gauge caused by mechanical vibration; the second group of Wheatstone bridges is used to measure the voltage signal generated by the change of the strain gauge caused by mechanical vibration;

[0009] The signal acquisition module is a data acquisition card, which is used to receive the voltage signal generated by the first group of Wheatstone bridges and the second group of Wheatstone bridges and transmit it to the signal processing and analysis module;

[0010] The signal processing and analysis module is used to convert the received voltage signal into numerical information of the grain flow based on a signal differential algorithm and a grain flow detection program to measure the grain flow.

[0011] Specifically, in some possible implementations, the resistance of the first group of Wheatstone bridges and the second group of Wheatstone bridges are both 350Ω, which is used to improve the sensitivity of the strain gauge to stress changes. The 350Ω strain gauge has a small temperature drift and low power consumption. When performing impulse method grain flow detection, it can reduce power consumption while ensuring measurement accuracy, reduce the impact of temperature drift, and improve anti-interference ability and dynamic response performance.

[0012] Specifically, in some possible implementations, the signal acquisition module is combined with a data acquisition card to support multi-channel signal synchronous acquisition, and realizes data interaction with the signal processing and analysis module through DLL function calls.

[0013] Specifically, in some possible implementations, the signal difference algorithm adopts a least mean square LMS algorithm based on adaptive filtering, which can dynamically adjust the filter weights to optimize signal quality and suppress noise interference. The algorithm includes the following steps:

[0014] The input signal is fitted and calculated through the adaptive filter matrix to generate an error signal;

[0015] The error signal is used to adjust the filter weights so that the mean square error between the signal and the expected signal is minimized;

[0016] The filter weight update rule is:

[0017]

[0018] Among them, d i (m) is the input signal, u i (m) is the filter input, ω i (m) is the filter weight, u is the step size factor, N s is the number of input channels.

[0019] Specifically, optionally, in some possible implementations, the upper rigid beam and the lower rigid beam are made of a material with a high elastic modulus to ensure the structural stability and measurement accuracy of the system.

[0020] Specifically, in some possible implementations, the signal processing and analysis module obtains the linear coefficient a and the offset b through a calibration experiment, and converts the voltage signal U into the grain flow M, and the relationship is:

[0021] M=aU+b

[0022] Among them, a and b are calibration coefficients obtained through experimental calibration.

[0023] Specifically optionally, in some possible embodiments, it is characterized in that the sampling frequency of the sensor system is 5000 Hz to capture detailed information of dynamic changes in grain flow.

[0024] Specifically, optionally, in some possible implementations, the signal processing and analysis module is implemented based on LabVIEW software, performs spectrum analysis, filtering and numerical conversion on the collected voltage signal, and provides display and storage functions for real-time grain flow information.

[0025] In addition, the present invention also provides a grain flow detection method, characterized in that the detection method is applicable to the grain flow detection device described in any one of the above items, comprising:

[0026] The voltage signal generated by the double-plate differential grain flow sensor composed of two groups of Wheatstone bridges, wherein the voltage signal generated by the first group of Wheatstone bridges is caused by the deformation of the rigid beam caused by the impact of grain on the upper interception plate and the change of the strain gauge caused by mechanical vibration, and the voltage signal generated by the second group of Wheatstone bridges is the voltage signal caused by the change of the strain gauge caused by mechanical vibration;

[0027] The signal acquisition module receives the voltage signals generated by the first group of Wheatstone bridges and the second group of Wheatstone bridges and transmits them to the signal processing and analysis module;

[0028] The signal processing and analysis module in the host computer converts the received voltage signal into numerical information of the grain flow based on the signal difference algorithm and the grain flow detection program to measure the grain flow.

[0029] In the technical solution of the present invention, the upper interception plate and the lower interception plate are respectively installed on the upper rigid beam and the lower rigid beam, wherein the upper interception plate is used to transmit the force generated by the impact of grains and the strain force caused by the vibration of the machine to the strain gauge on the upper rigid beam, while the lower interception plate focuses on detecting the strain force generated by the vibration of the machine. Since the upper and lower rigid beams are installed together at the grain outlet of the harvester as an integrated structure, the machine vibration noise signal detected by the two rigid beams is a common mode signal. By applying a signal differential algorithm, the common mode noise can be eliminated, thereby improving the signal-to-noise ratio and enhancing the measurement accuracy and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present invention.

[0031] Figure 1 This is a schematic diagram of the pasting position of the strain gauge bridge;

[0032] Figure 2 It is a strain gauge bridge circuit;

[0033] Figure 3 is a block diagram of a dual-plate differential grain flow sensor;

[0034] Figure 4 is a schematic diagram of the installation of a dual-plate differential grain flow sensor;

[0035] Figure 5 It is the structural diagram of the adaptive wavelet transform filtering algorithm model.

[0036] Figure markings: 1-upper rigid beam; 2-lower rigid beam; 3-first group of Wheatstone bridges composed of strain gauges; 4-second group of Wheatstone bridges composed of strain gauges; 5-connecting bracket; 6-base; 7-upper intercepting plate; 8-lower intercepting plate; 9-grain outlet of harvester; 10-double-plate differential grain flow sensor installed at the grain outlet of harvester. DETAILED DESCRIPTION

[0037] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0038] The present invention relates to a grain flow detection device and a detection method thereof, which are particularly suitable for detecting grain flow in a complex agricultural machinery operation environment. By optimizing the structural design and signal processing algorithm, the present invention has improved the anti-interference ability, signal separation accuracy and dynamic response performance of the measurement system, and can solve the problems of large measurement error, weak anti-vibration ability, low signal-to-noise ratio and the like existing in the prior art.

[0039] like Figure 1 , Figure 2 and Figure 3 As shown, the specific technical solutions of the present invention are as follows:

[0040] A grain flow detection device based on a signal differential algorithm comprises: a double-plate differential grain flow sensor 10 composed of two groups of Wheatstone bridges, a signal acquisition module and a signal processing and analysis module;

[0041] The first group of Wheatstone bridges 3 is composed of four strain gauges attached to the upper rigid beam 1, and the second group of Wheatstone bridges 4 is composed of four strain gauges attached to the lower rigid beam 2. The upper and lower rigid beams are connected through a base 6 as an integrated structure and are installed together at the grain outlet 9 of the harvester through a connecting bracket 5. The upper rigid beam 1 and the lower rigid beam 2 are respectively provided with an upper interception plate 7 and a lower interception plate 8. The first group of Wheatstone bridges 3 and the second group of Wheatstone bridges 4 respectively constitute two two-dimensional force sensors. The first group of Wheatstone bridges 1 is used to measure the voltage signal generated by the deformation of the rigid beam caused by the impact of grains on the upper interception plate and the change of the strain gauge caused by mechanical vibration; the second group of Wheatstone bridges 2 is used to measure the voltage signal generated by the change of the strain gauge caused by mechanical vibration;

[0042] A signal acquisition module, used for receiving the voltage signals generated by the first group of Wheatstone bridges 3 and the second group of Wheatstone bridges 4 and transmitting the voltage signals to the signal processing and analysis module;

[0043] The signal processing and analysis module is used for converting the received voltage signal into numerical information of grain flow based on a signal differential algorithm to measure the grain flow.

[0044] It is easy to understand that the scraper elevator of the harvester is used to lift the grain to a set height and then throw it out through the grain outlet 9 of the harvester. The double-plate differential grain flow sensor 10 is fixed on the grain throwing path, and the upper interception plate 7 is fixed on the upper rigid beam 1. Similarly, the lower interception plate 8 is fixed on the lower rigid beam 2. The upper rigid beam 1 and the lower rigid beam 2 are respectively affixed with four strain gauges, which respectively form a first group of Wheatstone bridges 3 and a second group of Wheatstone bridges 4, forming two two-dimensional force sensors. The first group of Wheatstone bridges 1 is used to measure the voltage signal generated by the deformation of the rigid beam caused by the impact of grain on the upper interception plate and the change of the strain gauge caused by mechanical vibration; the second group of Wheatstone bridges 2 is used to measure the voltage signal generated by the change of the strain gauge caused by mechanical vibration; the measured voltage signal is transmitted to the computer by the data acquisition module via the data acquisition card, and finally the grain flow data information is formed on the computer through the signal processing and analysis module written based on LabVIEW based on the signal processing algorithm (such as signal differential algorithm) and the grain flow detection program.

[0045] like Figure 1 The figure shows the schematic diagram of the pasting of strain gauges on the rigid beam. Figure 2 It is a Wheatstone bridge circuit composed of strain gauges. The design uses a metal foil resistance strain gauge with a resistance of 350Ω. The following is an analysis of the output of the corresponding bridge circuit under the impact force P.

[0046] When the combine harvester (i.e., harvester) is working in the field, the grain particle flow is driven by the scraper elevator and impacts the upper interception plate 7 through the grain outlet 9 of the harvester. The upper interception plate 7 is fixed together with the free end of the upper rigid beam 1, so the force acting on the interception plate will also act on the right end of the upper rigid beam 1 at the same time. The point of action of the grain flow impact force on the upper interception plate will change with the changes of factors such as the sensor installation position, the instantaneous mass flow rate of grain, and the working speed of the elevator. Therefore, the torque M equivalent to the free end of the parallel beam is a variable, and the output of the sensor should not be affected by the torque M.

[0047] The circuit is designed as a full-bridge mode, and the voltage sensitivity of the full-bridge is 4 times higher than that of a single bridge. In addition, the full-bridge can eliminate nonlinear errors like the differential bridge and has a continuous drift compensation effect.

[0048] Figure 2 U in is the bridge voltage, and Ui is the bridge output signal. The strains at the four patches are:

[0049]

[0050] Where E is the elastic modulus of the beam material, W is the bending section modulus of the parallel beam at each strain gauge, M1, M2, M3, and M4 are the moments acting at the four patches of the rigid beam caused by the end moment M, where M1 = -M4 and M2 = -M3.

[0051] Under the action of external force, the strain gauge deforms along with the beam. For the metal resistance strain gauge, its resistance value will change accordingly, and its resistance change value is: ΔR=R(1+2v)ε=RKε, where ν is the Poisson's ratio of the strain gauge material, R is the resistance value of the strain gauge when it is not deformed, K is a constant, generally called the strain coefficient or sensitivity of the resistance strain gauge; ε is the strain of the resistance strain gauge.

[0052] Therefore, the output of the rigid beam bridge is:

[0053]

[0054] Because the strain is very small, ignoring high-order small quantities:

[0055]

[0056] The changes to the patch should be:

[0057]

[0058] Therefore, it can be clearly seen from the above formula that the output of the bridge is only related to the impact force P and the structural parameters of the sensor, but has nothing to do with the torque M, that is, it has nothing to do with the point of application of the impact force.

[0059] When a combine harvester is traveling in the field, it will inevitably produce violent mechanical vibrations. These vibration signals are intertwined in the process of collecting sensor signals, resulting in the weak voltage signals output by the sensor being often masked by strong vibration interference, which brings great challenges to the subsequent signal processing work. To this end, the present invention adopts a double-plate structure impulse grain flow sensor design. The sensor is equipped with two detection boards: a front detection board composed of a first group of Wheatstone bridges, an upper rigid beam and an upper interception plate, which is dedicated to the dual detection of grain impact signals and machine vibration signals, and a rear detection board composed of a second group of Wheatstone bridges, a lower rigid beam and a lower interception plate, which is dedicated to the capture of machine vibration signals. Among them, the machine vibration signal picked up by the front detection board is recorded as F1, and that obtained by the rear detection board is F2. Since the two detection boards are firmly mounted on the harvester and connected to each other, the machine vibration signals detected by them show a high degree of common mode characteristics. Based on this, the present invention uses a signal differential algorithm to achieve effective elimination of machine vibration signals, thereby improving the signal-to-noise ratio of the signal:

[0060] F1=F2

[0061] F1-F2=0

[0062] However, the working environment of the sensor is extremely harsh. It not only faces the intrusion of machine vibration, but also the complex dynamics of the harvester when traveling in the field has a significant impact on the work of the sensor. Compared with the strong machine vibration and the noise such as the tilt and shake of the harvester body, the grain flow signal is particularly weak. Therefore, on the basis of adopting a double-plate differential structure and a differential signal processing algorithm to weaken the machine vibration signal, the present invention also urgently needs a method that can effectively collect weak low-frequency signals under a strong noise background. To this end, the wavelet filter construction method based on the adaptive filtering algorithm is improved, and its core lies in constructing a wavelet transform subband adaptive filter suitable for weak signal extraction in a strong noise environment. By deeply integrating wavelet transform technology with adaptive filtering technology and applying it to the detection of weak signals, the signal processing capability is further improved. In particular, a method for detecting electrocardiogram signals based on adaptive wavelet transform is adopted. This method uses the subband coding theory of wavelet transform to improve the detection accuracy of weak signals through adaptive matching and post-recovery fitting of multiple subband weights. This signal processing algorithm has demonstrated certain performance in capturing weak signal characteristics and improving the signal-to-noise ratio, and is an effective method to address the problem of weak signal detection in complex noise environments.

[0063] Using wavelet technology for signal denoising can retain the peaks and mutation features in useful signals. In contrast, when Fourier analysis is used for filtering, although the signal components are mainly concentrated in the low-frequency band, and the noise is scattered in the high-frequency band, it can be filtered by a low-pass filter, but this method is difficult to effectively distinguish the high-frequency components in the useful signal from the high-frequency interference caused by the noise. Specifically, if the bandwidth of the low-pass filter is set too wide, a large amount of noise will still remain in the filtered signal; conversely, if the filter bandwidth is too narrow, some valuable high-frequency signals may be misjudged as noise and eliminated. Therefore, in dealing with the denoising problem of non-stationary signals, wavelet analysis shows certain advantages over Fourier analysis.

[0064] Wavelet transform is a multi-scale mathematical approximation of the signal f(t), which is equivalent to filtering the signal f(t) using a multi-sampling rate filter group. It consists of a low-pass filter H(ω) and a band-pass and high-pass filter G(ω) associated with the scale function φ(t) and the wavelet function ψ(t). In the case of wavelet decomposition, these filters do not have overlapping passband parts, but together cover the entire frequency band of the input signal, so that each sub-band filter has a shorter impulse response than the full-band filter. In the process of adaptive updating of the weights of each sub-band, different step sizes can be used, which will result in a faster convergence speed than the full-band adaptive filter. Sub-band adaptive weak signal extraction algorithm based on wavelet transform. In this algorithm, the error signal is estimated in each sub-band, and then the objective function taking local errors into account is minimized through adaptive approximation. The model of the algorithm is as follows Figure 5 shown.

[0065] It can be seen that both the input signal x(k) and the desired signal d(k) are decomposed into multiple sub-bands by the scaling function. Then, each sub-band signal of the input signal is processed by the weights (wavelet coefficients) through the adaptive filter matrix to obtain an output signal that is compared with the corresponding frequency band of the desired signal in the sub-band. In this algorithm, our goal is to minimize the sub-band error energy. In this model, the objective function can be a linear combination of the squares of the local error amplitudes, that is,

[0066]

[0067] Because the adaptive filter matrix is ​​a diagonal matrix and the subband signal is complex, the update equation of the subband adaptive filter matrix of this algorithm is:

[0068]

[0069] Where N s is the length of the adaptive filter in the ith subband; γ is a small constant selected to prevent the update factor from being too large; the signal e i (m) is the i-th subband error signal; u i (m) is the input signal of the i-th adaptive filter;

[0070]

[0071] Where α is a small factor (0≤α≤0.1). Based on the knowledge of the normalized LMS algorithm, it can be inferred that the typical range of the convergence factor is 0<μ<1.

[0072] The present invention uses LabVIEW software to collect and process the output analog signal through a data acquisition card. As a professional software in the field of automated testing and measurement, LabVIEW integrates more than 600 analysis functions for various data analysis applications such as signal generation, frequency analysis, probability, statistics, mathematical operations, curve fitting, interpolation, digital signal processing, etc. In addition, LabVIEW software also provides additional tool software professionally applied to certain signal processing applications, such as sound and vibration, machine vision, RF / communication measurement, transient / short-duration signal analysis, etc.

[0073] The seamless combination of LabVIEW software and data acquisition card adopts the Dll function calling method, and selects the Coif3 wavelet function in LabVIEW software to extract the grain mass flow signal submerged in strong noise. The sampling frequency is 5000Hz, and there are many irregular complex vibration interference signals in the collected signal. At the same time, the grain flow detection and signal processing algorithm program is written based on LabVIEW software.

[0074] In the grain flow detection device, the relationship between the output yield signal and the physical quantity it represents is reflected by the sensor calibration equation. The sensor needs to use these calibration equations to convert the sensor output signal into actual physical quantities. The calibration process of the grain flow sensor can be divided into two steps: one is the linear calibration of the elastic element force and voltage, and the other is the yield calibration of the grain sensor.

[0075] The output voltage of the elastic element composed of a rigid beam and a strain gauge is linearly related to the force it receives. Therefore, the voltage signal output by the grain flow sensor is also linearly related to the actual output physical quantity (kg / s). According to the average value U1 of the sensor's output voltage and the sensor's output value when there is no wheat impact,

[0076] The base voltage U2 can be used to calculate the average voltage U (U = U1-U2) generated by the wheat impact on the sensor. Therefore, the grain mass is:

[0077] M=aU+b

[0078] Where M is the mass of grain, U is the average voltage value generated by the impact of grain on the sensor, and a and b can be obtained through linear calibration experiments.

[0079] In addition, the present invention also provides a grain flow detection method applicable to the grain flow detection device, specifically comprising: a voltage signal generated by a double-plate differential grain flow sensor composed of two groups of Wheatstone bridges, wherein the voltage signal generated by the first group of Wheatstone bridges is generated by the deformation of the rigid beam caused by the impact of grain on the upper interception plate and the change of the strain gauge caused by mechanical vibration, and the voltage signal generated by the second group of Wheatstone bridges is the voltage signal generated by the change of the strain gauge caused by mechanical vibration;

[0080] The signal acquisition module receives the voltage signals generated by the first group of Wheatstone bridges and the second group of Wheatstone bridges and transmits them to the signal processing and analysis module;

[0081] The signal processing and analysis module in the host computer converts the received voltage signal into numerical information of the grain flow based on the signal difference algorithm and the grain flow detection program to measure the grain flow.

[0082] In summary, the present invention solves the following key problems through structural design and signal processing strategy:

[0083] 1. Suppression of mechanical vibration interference: By utilizing the integrated design of the upper and lower rigid beam structures and the signal differential algorithm, the common-mode vibration noise signal detected by the two beams is eliminated, thereby improving the system signal-to-noise ratio.

[0084] 2. Signal separation and precision improvement: The upper interceptor not only detects the stress signal generated by the grain impact, but also detects the stress signal caused by vibration; the lower interceptor focuses on detecting the common mode signal generated by vibration. By performing differential calculation on the two signals, the grain impact force signal is extracted, effectively improving the measurement accuracy.

[0085] 3. Adaptive signal optimization: The system introduces adaptive filtering based on the least mean square (LMS) method, which can dynamically adjust the filtering weights according to the noise characteristics, further suppress the interference of environmental noise on the signal, and realize real-time signal optimization.

[0086] 4. Efficient signal acquisition and processing: The signal processing and analysis module written in LabVIEW is combined with the data acquisition module to realize multi-channel synchronous signal acquisition, filtering, spectrum analysis and data storage, which can obtain the dynamic change information of grain flow in real time and provide accurate measurement results.

[0087] The core of the present invention is:

[0088] First, by installing an upper interception plate and a lower interception plate on the upper and lower rigid beams respectively, the impact force of grains and the vibration of the machine can be effectively distinguished. The upper interception plate transmits the impact force and vibration strain force of grains to the strain gauge on the upper rigid beam, while the lower interception plate focuses on detecting the strain force signal caused by mechanical vibration. Since the upper and lower rigid beams are an integrated structure and are installed together at the grain outlet of the harvester, the mechanical vibration signal detected by them is a common mode signal. The signal differential algorithm can effectively eliminate the interference of common mode signals, thereby improving the signal-to-noise ratio.

[0089] Secondly, the strain signals detected by the two beams are processed by a signal differential algorithm to eliminate the common-mode noise signal caused by mechanical vibration and retain only the grain impact force signal, thereby significantly improving the signal quality.

[0090] Thirdly, the adaptive filtering module based on the LMS algorithm is used to further optimize the differential signal, suppress non-common mode noise and improve signal stability by dynamically adjusting weights.

[0091] Finally, the data processing module was developed based on the LabVIEW software platform, which can complete signal acquisition, filtering, spectrum analysis, data storage and real-time display, and also provide the calibration function of the linear model M=aU+b, which is used to convert the voltage signal into numerical information of grain flow.

[0092] Compared with the prior art, the present invention has the following significant advantages:

[0093] 1. Strong anti-interference ability: Through the combined application of signal differential algorithm and adaptive filtering module, the interference of mechanical vibration and environmental noise on the measurement signal is suppressed, and the signal-to-noise ratio is improved.

[0094] 2. High measurement accuracy: Based on the integrated design of rigid beam and interception plate and signal differential algorithm, the grain impact signal and mechanical vibration signal are separated, which improves the accuracy and reliability of the measurement system.

[0095] 3. Strong real-time performance: The efficient data acquisition and processing module can obtain the dynamic change information of grain flow in real time, meeting the needs of modern agricultural machinery for real-time monitoring.

[0096] 4. Wide adaptability: This system is not only suitable for grain flow monitoring of harvesters, but can also be applied to particle flow detection scenarios involved in other agricultural equipment.

[0097] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A grain flow detection device, characterized in that: The detection device is designed based on a signal differential algorithm and includes: A dual-plate differential grain flow sensor consisting of two sets of Wheatstone bridges, a signal acquisition module, and a signal processing and analysis module; The first group of Wheatstone bridges is composed of four strain gauges attached to the upper rigid beam, and the second group of Wheatstone bridges is composed of four strain gauges attached to the lower rigid beam. The upper and lower rigid beams are connected through a base as an integrated structure and are installed together at the grain outlet of the harvester through a connecting bracket. The upper rigid beam and the lower rigid beam are respectively provided with an upper interception plate and a lower interception plate. The first group of Wheatstone bridges and the second group of Wheatstone bridges respectively constitute two two-dimensional force sensors. The first group of Wheatstone bridges is used to measure the voltage signal generated by the deformation of the rigid beam caused by the impact of grains on the upper interception plate and the change of the strain gauge caused by mechanical vibration; the second group of Wheatstone bridges is used to measure the voltage signal generated by the change of the strain gauge caused by mechanical vibration; A signal acquisition module, used for receiving the voltage signals generated by the first group of Wheatstone bridges and the second group of Wheatstone bridges and transmitting the voltage signals to the signal processing and analysis module; The signal processing and analysis module is located in the host computer and is used to convert the received voltage signal into numerical information of the grain flow based on the signal difference algorithm and the grain flow detection program to measure the grain flow.

2. The grain flow detection device according to claim 1, characterized in that: The resistance values ​​of the first group of Wheatstone bridges and the second group of Wheatstone bridges are both 350Ω, which improves the sensitivity of the strain gauge to stress changes.

3. The grain flow detection device according to claim 1, characterized in that: The signal acquisition module is combined with the data acquisition card to support multi-channel signal synchronous acquisition, and realizes data interaction with the signal processing and analysis module through DLL function calls.

4. The grain flow detection device according to claim 1, characterized in that: The signal difference algorithm adopts the least mean square LMS algorithm based on adaptive filtering, which can dynamically adjust the filter weights to optimize signal quality and suppress noise interference. The algorithm includes the following steps: The input signal is fitted and calculated through the adaptive filter matrix to generate an error signal; The error signal is used to adjust the filter weights so that the mean square error between the signal and the expected signal is minimized; The filter weight update rule is: Among them, d i (m) is the input signal, u i (m) is the filter input, ω i (m) is the filter weight, u is the step factor, N s is the number of input channels.

5. The grain flow detection device according to claim 1, characterized in that: The upper rigid beam and the lower rigid beam are made of a material with a high elastic modulus to ensure the structural stability and measurement accuracy of the system.

6. The grain flow detection device according to claim 1, characterized in that: The signal processing and analysis module obtains the linear coefficient a and the offset b through calibration experiments, and converts the voltage signal U into the grain flow M, and the relationship is: M=aU+b Among them, a and b are calibration coefficients obtained through experimental calibration.

7. The grain flow detection device according to claim 1, characterized in that: The sensor system samples at 5000 Hz to capture detailed information on the dynamics of grain flow.

8. The grain flow detection device according to claim 1 or 3, characterized in that: The signal processing and analysis module is implemented based on LabVIEW software, which performs spectrum analysis, filtering and numerical conversion on the collected voltage signal, and provides display and storage functions for real-time grain flow information.

9. A method for detecting grain flow, characterized in that: The detection method is applicable to the grain flow detection device described in any one of claims 1 to 8, comprising: The voltage signal generated by the double-plate differential grain flow sensor composed of two groups of Wheatstone bridges, wherein the voltage signal generated by the first group of Wheatstone bridges is caused by the deformation of the rigid beam caused by the impact of grain on the upper interception plate and the change of the strain gauge caused by mechanical vibration, and the voltage signal generated by the second group of Wheatstone bridges is the voltage signal caused by the change of the strain gauge caused by mechanical vibration; The signal acquisition module receives the voltage signals generated by the first group of Wheatstone bridges and the second group of Wheatstone bridges and transmits them to the signal processing and analysis module; The signal processing and analysis module in the host computer converts the received voltage signal into numerical information of the grain flow based on the signal difference algorithm and the grain flow detection program to measure the grain flow.

Images