A method for detecting foreign matter in a grain bin based on elastic resonance waves

Abstract

This invention discloses a method for detecting foreign objects in grain silos based on elastic resonance waves, belonging to the field of grain condition monitoring technology. A low-frequency three-component vibration pickup is selected to collect background vibration noise. Utilizing the resonance characteristics of elastic waves, foreign objects inside the bulk grain in the grain silo are detected. The steps are: S1, selecting a low-frequency three-component vibration pickup; S2, deploying the pickup on the grain surface at designated detection points in the grain silo, adjusting the pickup's orientation and level, setting acquisition parameters, and recording the background vibration signal; S3, removing abnormal vibration noise and calculating the power spectrum of the three vibration components segment by segment; S4, calculating the horizontal and vertical spectral ratio curves; S5, based on the horizontal and vertical spectral ratio curves, performing foreign object identification and calculating burial depth parameters. This method achieves the identification of foreign objects in the grain silo, with a simple data post-processing workflow, minimal impact from human intervention, and stable and reliable processing results.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of grain storage detection, in particular to a grain storage foreign matter detection method based on elastic resonance waves. BACKGROUND

[0002] Grain storage foreign matter detection is an important work for accurate supervision and auditing of reserve grain, and grain storage foreign matter refers to objects different from normal grain, including cavities, clumps and other impurities. The existence of grain storage foreign matter affects the accurate calculation of the quantity of grain in the warehouse, and therefore it is very important to develop convenient, accurate and intelligent grain storage foreign matter detection technology.

[0003] At present, the methods for non-destructive detection of grain storage bulk grain mainly include impact echo detection method, gravity measurement method, excitation scanning detection method, nuclear detection method, pressure sensor network method, electromagnetic wave (radar) detection method, nuclear magnetic resonance detection method, etc. The impact echo measurement method can detect the grain surface height in the warehouse on the outer wall of the warehouse. The three-dimensional excitation scanning detection technology can obtain the accurate grain pile surface fluctuation form and surface area, laying a foundation for calculating the entire volume of the grain pile. Terahertz waves are electromagnetic radiation waves between infrared and microwaves, and terahertz spectrum detection and imaging technology has certain application value in grain quality identification and classification, grain freshness, grain fungus contamination and grain pest detection. The principle of gravity measurement is to infer the detection target according to the gravity difference caused by the density difference, but the resolution is obviously affected by the volume effect, and there are still certain difficulties in the application of grain storage foreign matter detection. The pressure sensor detection technology is to analyze the mechanical properties of the grain storage, combine the material properties of different grains and the structure of different grain stores, establish a pressure sensor network of the grain store, and realize the measurement method of the quantity of grain in the grain store. The electromagnetic wave (radar) detection technology is to emit electromagnetic pulses, and after the propagation in the grain in the grain store, according to the information such as time, amplitude and waveform of the received electromagnetic wave signals, imaging and pattern recognition technologies are used to achieve the purpose of non-destructive detection of the internal structure of the grain pile. The nuclear magnetic detection technology is to use the different hydrogen content and nuclear magnetic effect strength of different media to judge the physical property difference of the medium and realize the detection of abnormality in the grain store.

[0004] At present, there are Chinese invention patents with publication numbers CN103063136B, CN103307976B, CN104296847B, CN104330137B, CN104331591B, CN105403294B, CN105424148B, CN105928474B, CN105931238B, CN107843321A, CN110823334B, CN109444875B, CN111721448B, etc. in the prior art.

[0005] Existing papers include:

[0006] Sun Qinqing, et al. Research on Evidence Theory in the Detection of Grain Storage Holes in Grain Depot. Computer and Digital Engineering, 2009, Vol. 37 (No. 2), pp. 4-6.

[0007] Qin Yao, et al. Detection of Grain Storage Information in Grain Depot by Radar Tomography. Journal of Wave Science, 2010, Vol. 25 (No. 1), pp. 66-72.

[0008] Zhang Dexian, et al. Online Detection Method of Grain Quantity in Grain Depot Based on Pressure Sensor. China Oils and Fats, 2014, Vol. 29 (No. 4), pp. 98-100.

[0009] Zhang Xinxin, et al. Research on Electromagnetic Wave Detection Technology of Grain Quantity. Computer Knowledge and Technology, 2014, Vol. 10 (No. 10), pp. 2452-2456.

[0010] Zhu Yuhua, et al. Research Progress of Grain Quantity Detection Technology. Journal of Henan University of Technology (Natural Science Edition), 2016, Vol. 37 (No. 6), pp. 116-122.

[0011] The above patents or documents mainly introduce the types of grain depot bulk grain nondestructive detection methods, wherein the methods capable of being used for foreign matter detection in grain pile are electromagnetic wave method, nuclear magnetic resonance method and gravity measurement method. Since metal has great influence on the electromagnetic wave method and the nuclear magnetic resonance method, the application scene and detection precision are greatly limited. For the gravity measurement method, the detection resolution is low, and it is difficult to identify small-scale foreign matters in the warehouse. Moreover, the data collection work of the above methods has the problems of low flexibility, complex process and tedious data post-processing, and the application in the grain depot nondestructive detection is limited to a certain extent. SUMMARY

[0012] The present application aims at the above problems, and provides a grain depot foreign matter detection method based on elastic resonance wave. A low-frequency three-component vibration pickup is selected to collect background vibration noise, and the detection of foreign matters in the grain depot bulk grain is realized by using the resonance characteristics of elastic wave, thereby providing a new method for grain depot grain quantity supervision and auditing.

[0013] The technical scheme of the present application is as follows:

[0014] S1, a low-frequency three-component vibration pickup is selected;

[0015] S2, the vibration pickup is arranged on the grain surface of the designated detection point in the grain depot, the orientation and level of the vibration pickup are adjusted, the collection parameters are set, and the background vibration signal is recorded;

[0016] S3, remove abnormal vibration noise, segment calculation of three component vibration signal power spectrum;

[0017] S4, calculate the horizontal and vertical spectrum ratio curve;

[0018] S5, based on the horizontal and vertical spectrum ratio curve, implement the grain warehouse foreign matter identification and buried depth parameter calculation.

[0019] Further, the low-frequency three-component vibration pickup 9 selected in step S1 includes a shell 12, a three-component vibration sensor 13 fixedly arranged in the shell 12, a data storage module 17, a wireless transmission module 11, a rechargeable battery 16, a level 14 and a compass 15 fixedly arranged on the shell, a magnetic attraction switch 10, and a foot cone 18 threadedly connected to the bottom of the shell 12;

[0020] The directions of the three components in the three-component vibration sensor 13 are orthogonal, and the three-component vibration sensor 13 is connected to the wireless transmission module 11 through the data storage module 17, and sends the collected data to the outside through the wireless transmission module 11; the rechargeable battery 16 is connected to the three-component vibration sensor 13, the data storage module 17, and the wireless transmission module 11 through the magnetic attraction switch 10; the three-component vibration sensor 13, the data storage module 17, and the wireless transmission module 11 are powered by the rechargeable battery 16, and the magnetic attraction switch 10 controls the on-off of the power, and the frequency response range of the three-component vibration sensor 13 can reach as low as 0.2 Hz;

[0021] The current orientation and level of the low-frequency three-component vibration pickup 9 are displayed through the level 14 and the compass 15.

[0022] The metal cone is used as the vibration pickup foot cone, inserted into the bulk grain, and the vibration pickup and the bulk grain are fully coupled, and the length of the foot cone is in the range of 0.2m-0.5m.

[0023] Further, when adjusting the orientation and level of the low-frequency three-component vibration pickup 9 in step S2, by observation, the two horizontal components in the three-component vibration sensor 9 are respectively directed to the north and the east, and the vertical component is the plumb direction.

[0024] Further, after inserting the foot cone 18 of the low-frequency three-component vibration pickup 9 into the grain pile and ensuring good coupling, the sampling frequency, the recording length, and the measurement point number are set, the magnetic attraction switch is used to turn on the vibration pickup, and the background noise vibration signal is started to be recorded. Further, the sampling frequency is selected in the range of 128Hz-512Hz.

[0025] Further, the background vibration signal is a weak earth tremor signal caused by natural or non-subjective human factors, the abnormal vibration noise is a short-time interference signal with an amplitude much larger than the background vibration signal, and the STA / LTA algorithm is used to remove the abnormal vibration noise in step S3.

[0026] Furthermore, the identification of foreign objects in grain warehouses and the calculation of burial depth parameters are information on the existence and burial depth of non-grain occupants inside the bulk grain in grain warehouses.

[0027] Observe whether the horizontal and vertical spectral ratio curves calculated in step S4 have more than one peak to determine whether there are non-grain occupants inside the bulk grain in the grain warehouse. If foreign objects are present, calculate their burial depth information according to the following formula: Where f1 is the frequency corresponding to the foreign object indicator peak, V S The average shear wave velocity of the bulk grain;

[0028] The average shear wave velocity of the above-mentioned bulk grain is calculated using the following formula: V S =4f0h, where f0 is the frequency corresponding to the peak of the bottom indicator wave, and h is the thickness of the grain pile at the detection point.

[0029] The physical basis of this invention is the resonance phenomenon of elastic waves in grain storage piles, with the recorded data being the vibration signal of the medium. The vibration characteristics of the medium are closely related to its elastic parameters, such as shear modulus, Young's modulus, Lamé coefficient, bulk modulus, Poisson's ratio, and density, but independent of its electrical parameters (dielectric coefficient, resistivity, etc.) and magnetic parameters (permeability, etc.). Compared to conventional media, metallic materials have extremely low resistivity and dielectric constant, thus significantly affecting electromagnetic wave propagation and NMR effects. Consequently, conventional non-destructive testing methods based on electromagnetic waves and NMR principles are greatly affected by metallic materials. In contrast, elastic wave-based detection methods are unaffected by metallic materials, demonstrating a significant advantage in grain storage inspection applications where large amounts of metallic materials are present.

[0030] The advantages of the foreign object detection method for grain silos based on elastic resonance waves described in this invention are:

[0031] I. The low-frequency three-component vibration pickup is selected, which can obtain accurate low-frequency background vibration noise signals, thus having the characteristics of large detection depth, unaffected by the metal body of the grain silo, and suitable for the detection of foreign objects in bulk grain in various types of grain silos.

[0032] Second, the selected vibration pickup has independent data acquisition and storage functions. Multiple vibration pickups can be deployed on the surface of the grain pile to realize the synchronous acquisition of background vibration noise at multiple points and to perform three-dimensional detection of foreign objects in the grain pile.

[0033] Third, based on the resonance characteristics of elastic waves, the identification of foreign objects in grain silos is achieved by calculating the horizontal / vertical spectral ratio curve. The data post-processing process is simple, less affected by human intervention, and the processing results are stable and reliable.

[0034] Four, the method belongs to passive source detection method, without excitation source, detection work is simple, low cost, high precision, can provide accurate foreign matter occurrence information for grain storage quantity supervision and audit. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 is the grain foreign matter detection method flowchart provided by the method of the application based on elastic resonance wave;

[0036] Figure 2 is the data acquisition schematic diagram of the grain foreign matter detection method provided by the method of the application based on elastic resonance wave;

[0037] Figure 3 is the model diagram of the low-frequency three-component vibration pickup combination device for wireless real-time transmission;

[0038] Figure 4 is the foreign matter-free model of example one and the corresponding horizontal / vertical spectrum ratio curve characteristic map thereof;

[0039] Figure 5 is the foreign matter-containing model of example two and the corresponding horizontal / vertical spectrum ratio curve characteristic map thereof;

[0040] In the figure: 1-warehouse roof; 2-warehouse wall; 3-grain surface; 4-loose grain; 5-warehouse bottom; 6-foreign matter; 7-ground background noise elastic wave; 8-resonance wave; 9-three-component vibration pickup; 10-magnetic attraction switch; 11-wireless transmission module; 12-outer shell; 13-three-component vibration sensor; 14-level meter; 15-compass; 16-battery; 17-storage module; 18-foot cone. DETAILED DESCRIPTION

[0041] To clearly illustrate the technical features of the patent, the patent will be described in detail below through specific embodiments and in combination with the accompanying drawings.

[0042] Example one:

[0043] In example one, the detection object is a foreign matter-free grain warehouse model as shown in Figure 4 , and the detection point is 6m away from the warehouse bottom. The body wave resonance theory is used for simulation, and the grain foreign matter detection method flowchart provided by the method of the application based on elastic resonance wave is used as shown in Figure 1 , which specifically includes:

[0044] S1, select a low-frequency three-component vibration pickup 9 with wireless transmission as the signal acquisition equipment. The selected low-frequency three-component vibration pickup is a combination device with built-in direction orthogonal three-component vibration sensor 13, data storage module 17, data wireless transmission module 11, level meter 14, compass 17, rechargeable battery 16 and magnetic attraction switch 10, and the frequency response range of the sensor can reach at least 0.2Hz.

[0045] S2, Press Figure 2 The diagram shown is a data acquisition schematic of the grain silo foreign object detection method based on elastic resonance waves provided by the present invention. Three-component vibration pickups 9 are deployed on the grain surface inside the silo. The structure of the three-component vibration pickups 9 is as follows: Figure 3 As shown in the diagram, 1 represents the top of the storage silo, 2 the storage wall, 3 the grain surface, 4 the loose grain, 5 the storage floor, 6 the foreign object, 7 the ground background noise elastic wave, and 8 the resonant wave. First, the foot cone 18 is threaded onto the bottom of the vibration pickup 9. Using a compass 17 and a level 14, the orientation and level are adjusted so that the two horizontal components of the three-component vibration sensor 9 point to due north and due east, respectively, and the vertical component is aligned with the plumb line. Then, the foot cone 18 is inserted into the grain pile to fix the vibration pickup in the grain pile, ensuring good coupling between the two. Finally, the sampling frequency is set to 256Hz, the recording length to 30 minutes, and the measurement point number is set. The vibration pickup is then turned on using a magnetic switch to begin recording the background noise vibration signal.

[0046] S3. For the acquired background noise vibration signal, set the segment duration to 20s, and use the STA / LTA algorithm to identify strong interference noise signals, ensuring that the signal segments containing them are not included in subsequent processing. For the stable background noise vibration signal without interference, calculate the power spectrum of the three component signals segment by segment.

[0047] S4. For the calculated power spectrum curve, a smoothing filter is applied to suppress high-frequency jitter. Then, the power spectrum curves for all time periods of the same component are superimposed to obtain the spectral ratio curves for the horizontal component (X and Y components) and the vertical component (Z component), respectively. The horizontal / vertical spectral ratio (H / V) curve is then calculated using the following formula:

[0048]

[0049] In equation (1), P E (f), P N (f) and P Z (f) shows the power spectra of the east-west (X), north-south (Y), and vertical (Z) components, respectively. It can be seen that the spectral ratio curve is a function of frequency variation.

[0050] S5. The calculated horizontal / vertical spectral ratio curve is as follows: Figure 4As shown in (b), the horizontal axis represents frequency, and the vertical axis represents the ratio of the power spectrum of the horizontal component signal to the power spectrum of the vertical component. A distinct peak can be clearly seen at approximately 5.5 Hz. This peak is the bottom indicator peak, amplified by the background vibration noise incident on the bottom of the silo, resulting from the body wave resonance between the grain surface and the silo bottom. The frequency corresponding to the peak is the inherent resonant frequency related to the shear wave velocity of the bulk grain and the thickness of the grain pile. Based on this peak frequency and the actual thickness of the bulk grain pile, the average shear wave velocity V of the bulk grain can be calculated. S The formula is as follows:

[0051] V S =4f0h (2)

[0052] In equation (1), f0 is the frequency corresponding to the peak of the bottom indicator wave, and h is the thickness of the grain pile at the detection point. In this example, the average shear wave velocity of the bulk grain is 132 m / s, which is consistent with the model design parameters.

[0053] Since the spectral ratio curve has only one obvious peak indicating the bottom of the silo, and there are no other adjacent obvious peaks, it is determined that the grain silo model does not contain foreign objects.

[0054] Example 2:

[0055] In Example 2, the detection target is as follows: Figure 5 (a) shows a grain silo model containing foreign objects, with the distance from the grain surface to the silo bottom at the detection point being 6m. The simulation is performed using the body wave resonance theory. Figure 1 The flowchart of the foreign object detection method for grain silos based on elastic resonance waves provided by the present invention specifically includes:

[0056] S1. Select a low-frequency three-component pickup with wireless transmission as the signal acquisition equipment (the specific requirements are the same as in Example 1, and will not be repeated).

[0057] S2, Press Figure 2 The diagram shows a data acquisition schematic of the grain silo foreign object detection method based on elastic resonance waves provided by the present invention. Three-component vibration pickups are deployed on the grain surface inside the grain silo (the specific requirements are the same as in Example 1, and will not be repeated).

[0058] S3. Calculate the power spectrum of the three component signals of the acquired background noise vibration signal (the specific requirements are the same as in Example 1, and will not be repeated here).

[0059] S4. Calculate the horizontal / vertical spectral ratio (H / V) curve according to formula (1) in Example 1.

[0060] S5. The calculated horizontal / vertical spectral ratio curve is as follows: Figure 5(b) shown, the abscissa is frequency, and the ordinate is the ratio of the horizontal component signal power spectrum to the vertical component power spectrum. It can be clearly seen that at a frequency of about 5.5 Hz, there is an obvious peak, which is the bottom of the bin indicating the peak, and at 10 Hz, there is a peak corresponding to the presence of a foreign object. When there is a foreign object in the bulk grain pile, the spectral ratio curve will have multiple peaks. In this example, since the spectral ratio curve has two obvious peaks, corresponding to the peak indicating the bottom of the bin and the peak indicating the foreign object, it can be determined that the grain bin model contains a foreign object.

[0061] According to formula (2) in Example 1, the average transverse wave speed of the bulk grain can be calculated to be 132 m / s. Here, the abnormal depth z can be calculated according to the following formula:

[0062]

[0063] In formula (3), f1 is the frequency corresponding to the foreign object indicating peak, V S is the average transverse wave speed of the bulk grain. Here, the foreign object depth can be further calculated to be 3.3 m, which is consistent with the model design parameters.

[0064] As can be seen from Example 1 and Example 2, using the method of the present application, the background vibration noise can be used without the need for an excitation source, and the data collection and analysis are convenient, and the existence of a foreign object in the bulk grain pile and the depth information thereof can be obtained, greatly improving the efficiency and accuracy of warehouse grain supervision and auditing.

[0065] As described above, when the horizontal and vertical spectral ratio curves are single-peak, it indicates that there is no foreign object; when the horizontal and vertical spectral ratio curves are double-peak, it indicates that there is a foreign object, and the peak corresponding to the lower frequency is the bottom of the bin indicating peak. The reason is that compared with the depth of the top interface of the foreign object, the depth of the bottom of the bin is larger, so the resonance frequency caused by the two is compared, and the resonance frequency corresponding to the bottom of the bin is lower. Accordingly, the peaks indicating the bottom of the bin and the foreign object can be distinguished.

[0066] Whether there is a foreign object or not, the bottom of the bin will cause resonance, so the indicating peak exists (as shown in Example 1 and Example 2), and the resonance frequency f0 corresponding to the bottom of the bin indicating peak is stable; accurate reading of the resonance frequencies (f0 and f1) corresponding to the bottom of the bin and the foreign object, using f0 and the thickness h of the grain pile (the thickness can be observed on site), the average transverse wave speed of the grain pile can be calculated according to the formula V S = 4f0h. S The foreign object depth can be calculated according to the formula

[0067] ​The present application has many specific implementation approaches, and the above description is only the preferred embodiment of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, several improvements can be made, and these improvements should also be considered as the protection scope of the present application.

Claims

1. A method for detecting foreign matter in a grain storage based on elastic resonance waves, characterized by, Adopting body wave resonance theory to simulate, detecting according to the following steps: S1, selecting low-frequency three-component vibration pickup; S2, arranging the vibration pickup on the grain surface of the designated detection point in the grain depot, adjusting the orientation and level of the vibration pickup, setting the collection parameters, and recording the background vibration signal; S3, removing abnormal vibration noise, and calculating the power spectrum of the three-component vibration signal in sections; The background vibration signal is a weak vibration signal of the earth caused by natural or non-subjective human factors, the abnormal vibration noise is a short-time interference signal with an amplitude much larger than the background vibration signal, and the STA / LTA algorithm is used to remove the abnormal vibration noise in step S3; S4, calculating the horizontal and vertical spectral ratio curve; S5, based on the horizontal and vertical spectral ratio curve, implementing the grain depot foreign matter identification and buried depth parameter calculation; The grain depot foreign matter identification and buried depth parameter calculation are whether the non-grain occupying body exists in the internal grain of the grain depot and the buried depth information. The observation step S4 calculates whether there is a peak greater than 1 in the horizontal and vertical spectral ratio curve, and judges whether there is a non-grain occupying body in the bulk grain in the grain depot. If there is a foreign matter, the buried depth information is calculated according to the following formula: Wherein, f1 is the resonance frequency of the foreign matter indicating peak, V S is the average transverse wave velocity of the bulk grain; The average transverse wave velocity of the bulk grain is calculated according to the following formula: V S = 4f0h, wherein f0 is the peak resonance frequency of the bottom of the bin, and h is the thickness of the grain pile at the detection point.

Images