Asynchronous sorting device and method for superior and inferior fruits based on vibration signals

Abstract

The invention relates to a good and bad fruit asynchronous sorting device and method based on vibration signals. The device comprises a feeding device and a sorting module, the feeding device is provided with a plurality of parallel feeding channels, and outlets of the channels are correspondingly connected with the sorting module. The sorting module is formed by sequentially arranging an asynchronous feeding device, a vibration signal collecting device and a sorting device. The vibration signal collecting device comprises an obliquely-arranged signal plate, the asynchronous feeding device throws the fruits to the signal plate according to different time sequences, ordered collection of vibration signals is achieved, and signal interference is avoided. And the asynchronous feeding device adopts a multi-hook rotating wheel structure, so that the feeding precision and efficiency are improved. The invention further provides a corresponding sorting method, and the fruit quality is quickly and accurately judged through vibration signal feature extraction and an intelligent analysis algorithm. According to the technology, the problems that a traditional sorting mode is low in efficiency and large in error are solved, the sorting precision and the production efficiency are remarkably improved, and the intelligent sorting requirement of large-scale agricultural production is met.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural product processing, and particularly relates to an asynchronous sorting device and method for good and bad fruits based on vibration signals. Background Art

[0002] The screening of good and bad fruits is a key link in the field of agricultural processing. Its core goal is to efficiently distinguish high-quality fruits from damaged fruits to ensure the smooth progress of subsequent storage, processing, and sales. Currently, a common sorting method is to use the principle of free fall, dropping the fruits from a certain height onto a vibrating plate, and judging the quality of the fruits by collecting the vibration signals on the vibrating plate. However, this method has the following significant problems:

[0003] Low feeding efficiency: The existing system adopts a single-channel conveying method. The fruits need to pass through the dropping port one by one and fall onto the detection plate by gravity. Only after the detection is completed can the next fruit be conveyed. This serial processing method severely limits the feeding efficiency and is difficult to meet the needs of large-scale production.

[0004] Signal interference problem: Due to the direct connection structure design between channels, during the feeding process, the mechanical vibration signals between channels will be coupled and interfered with each other. This interference will reduce the signal-to-noise ratio of the vibration signal acquisition, thereby affecting the detection accuracy.

[0005] Secondary bounce phenomenon: When the fruits fall onto the vibrating plate, they are prone to secondary bounce. The additional vibration signals generated by the secondary collision will interfere with the feature extraction of the initial collision signal. The superposition of the collision energy will change the amplitude-frequency characteristics of the vibration signal, resulting in the distortion of the collected vibration signal, thereby increasing the screening error and affecting the final sorting effect.

[0006] Insufficient signal analysis efficiency and accuracy: The existing vibration signal analysis methods have problems of low efficiency and insufficient accuracy, and it is difficult to quickly and accurately judge the quality of fruits, resulting in low overall screening efficiency and unable to meet the high-efficiency requirements of modern agricultural production.

[0007] In summary, the existing technology has obvious deficiencies in terms of feeding efficiency, signal interference, secondary bounce, and signal analysis accuracy, and cannot screen fruits efficiently and accurately. There is an urgent need for a more efficient and accurate screening device and method for good and bad fruits to solve the above problems. Summary of the Invention

[0008] The purpose of the present invention is to solve at least one of the above technical defects.

[0009] For this reason, one purpose of the present invention is to propose an asynchronous sorting device for good and bad fruits based on vibration signals to solve the problems mentioned in the background art and overcome the deficiencies existing in the prior art.

[0010] To achieve the above object, an asynchronous sorting device for good and bad fruits based on vibration signals according to the present invention includes:

[0011] A feeding device and a sorting module; the feeding device includes a plurality of parallel feeding channels, and a sorting module is correspondingly arranged at the outlet end of each feeding channel; the sorting module includes a plurality of asynchronous feeding devices, a signal acquisition device and a sorting device. The asynchronous feeding device, the vibration signal acquisition device and the sorting device are arranged in sequence along the fruit conveying direction to form a continuous sorting production line; the signal acquisition device includes a signal board, and the signal board is arranged at a predetermined inclination angle relative to the horizontal plane; a plurality of asynchronous feeding devices throw materials onto the signal board according to different time sequences to achieve asynchronous feeding.

[0012] Preferably, it further includes a first motor. The asynchronous feeding device includes a feeding component; the feeding component includes a runner and a plurality of material taking hooks. The runner is drivingly connected to the output end of the first motor, and the material taking hooks are evenly arranged along the circumferential direction of the runner for grasping and throwing fruits onto the signal board; and / or the feeding device further includes a feeding hook, and the feeding hook is arranged at the outlet end of the feeding channel and cooperates with the material taking hook.

[0013] In any of the above solutions preferably, the feeding device includes a frame, feeding rollers, a flow dividing partition component and a second motor. The output end of the second motor is connected to the feeding rollers. A plurality of feeding rollers are arranged in sequence on the frame, and the flow dividing partition component is arranged above the feeding rollers; the feeding channels are jointly defined by the flow dividing partition component and the feeding rollers for diverting and conveying fruits to the corresponding sorting modules.

[0014] In any of the above solutions preferably, the sorting device includes a sorting box and a material selection component. A material selection component is arranged on each of the opposite sides of the sorting box. The material selection component includes a fixed rod, a telescopic rod, a movable rod, a driving arm and a material selection plate. The fixed rod is arranged on the outer side wall of the sorting box. One end of the telescopic rod is connected to the fixed rod, and the other end of the telescopic rod is connected to the movable rod. Both ends of the movable rod are respectively pivotally connected to one end of a driving arm, and both ends of the material selection plate are respectively pivotally connected to the other end of the driving arm to drive the material selection plate to perform a material selection operation through the telescopic movement of the telescopic rod.

[0015] The present invention also discloses a sorting method for an asynchronous sorting device for good and bad fruits based on vibration signals, including the following steps:

[0016] Step S1: Establish a state model of the feeding device and output the transmission speed of the fruits;

[0017] Step S2: Establish a state model of the asynchronous feeding device and control the asynchronous feeding device to throw the fruits onto the signal board according to a preset trajectory;

[0018] Step S3: Establish a signal acquisition device state model, analyze and process the vibration signals collected by the signal board, and send them to the industrial control computer;

[0019] Step S4: Establish a classified signal state model, analyze the vibration signals received by the industrial control computer, and output classified result signals;

[0020] Step S5: Establish a sorting device signal state model, output sorting action signals according to the classified result signals, and control the sorting device to complete fruit sorting.

[0021] Preferably, step S1 specifically includes:

[0022] Let the displacement of the fruit be x1(t), the transmission speed of the fruit be x2(t), and the driving force of the second motor be u1(t). Taking the acceleration of the fruit as the description quantity, the acceleration of the fruit is the second derivative of the fruit displacement, and the state equation is:

[0023]

[0024] Among them, F motor is the driving force of the second motor, F friction is the friction force of the fruit transmission, a max is the acceleration of the fruit, m walnut is the mass of the fruit, m walnut ·a max is the inertia of the fruit. Let the constraint condition be:

[0025] F motor ≥μ·N + m walnut ·a max

[0026] μ is the friction coefficient of the feeding roller, N = m walnut ·g, g is the acceleration due to gravity, F friction = μ·N;

[0027] Let the speed of the fruit be V max , and the constraint V max is within the preset range, then:

[0028]

[0029] The output equation of the feeding device state model is:

[0030] y1(t) = C1x1(t) + D1u1(t)

[0031] Among them, the output quantity y1(t) is the transmission speed of the fruit, C1 is the displacement unit matrix, and D1 is the second motor driving force unit matrix.

[0032] In any of the above solutions preferably, step S2 specifically includes:

[0033] Let the angular displacement of the runner be \(x_3(t)\), the angular velocity of the runner be \(x_4(t)\), the driving torque of the first motor be \(u_2(t)\), and the state equation of \(x_4(t)\) be:

[0034]

[0035] The acceleration of \(x_4(t)\) describes the state of the runner changing with time:

[0036]

[0037] where \(J\) is the moment of inertia, and \(T\) friction (t) is the frictional resistance that needs to be overcome when the output end of the first driving motor rotates. The constraint condition of \(J\) is:

[0038] T motor (t) ≤ T max

[0039] T max is the preset torque borne by the runner; \(w\) max is the critical angular velocity preset by the first motor, with the constraint: \(|x_4(t)| ≤ w\) max ;

[0040] Let \(m\) walnut be the mass of the fruit, \(F\) centrifugal be the centrifugal force of the fruit, \(r\) be the radius from the center of the runner to the center of the feeding hook, and the gravity of the fruit be \(m\) walnut *g; The two conditions for the fruit to be thrown out are:

[0041] (1) When \(F\) centrifugal is greater than \(m\) walnut *g, the fruit meets the first condition for being thrown out. The calculation formula for the centrifugal force of the fruit is:

[0042] F centrifugal＝ m walnut *r*x4(t) 2

[0043] Simplified to:

[0044] r*x4(t) 2 >g

[0045] (2) Let \(\theta\) release be the preset angle at which the fruit is thrown out. When \(x_3(t)=\theta\) release , the fruit meets the second condition for being thrown out, and the fruit is thrown out. The linear velocity at which the fruit is thrown out is:

[0046] v t h row (t) = r*x4(t)

[0047] v t h row (t) is the tangential velocity of the fruit being thrown, and the angle at which the fruit is thrown is:

[0048] θ t h row (t) = x3(t)

[0049] θ t h row (t) is the angle at which the fruit is thrown;

[0050] The output equation of the asynchronous feeding device state model is:

[0051] y2(t) = C2x4(t) + D2u2(t)

[0052] y3(t) = C3x3(t)

[0053] Among them, y2(t) is the output parameter of the throwing speed of the fruit, y3(t) is the output parameter of the throwing angle of the fruit, u2(t) is the torque input by the second motor, C2 is the relationship coefficient between the angular velocity and the throwing speed, C3 is the relationship coefficient between the angular displacement and the throwing angle, and D2 is the influence coefficient of u2(t) on the output;

[0054] Preferably, in any of the above solutions, step S3 specifically includes:

[0055] Let the position where the fruit hits the signal plate be x5(t), the impact velocity of the fruit hitting the signal plate be x6(t), and the force of the fruit hitting the signal plate be F1 impact (t), and the frictional force when the fruit collides with the signal plate is F1 friction (t), and the state equation of x5(t) is:

[0056]

[0057] The state equation of the fruit velocity is:

[0058]

[0059] Among them:

[0060] F1 friction (t) = μ1 * N1 * sign(x6(t))

[0061] Among them, m walnut is the mass of the fruit, μ1 is the friction coefficient of the signal plate, N1 is the normal force during the impact of the fruit, and sign(x6(t)) is the sign function of x6(t), indicating that the direction of F1 friction (t) is opposite to the direction of motion of the fruit; for each impact, only a single impact is allowed to fall into the sorting position, x6(tafter ) is the velocity direction after the fruit impact, and the constraint is x6(t after ) < 0, indicating that the velocity direction after the fruit impact is opposite to the velocity direction before the fruit impact;

[0062] Let s(t) be the signal of the actual physical vibration generated by the signal board, and the output equation is:

[0063] s(t) = C4 * F1 impact (t) + D3 * x6(t)

[0064] where C4 is the relationship coefficient between s(t) and F1 impact (t), D3 is the relationship parameter coefficient between s(t) and x6(t). Let the electrical signal y4(t) sent to the industrial control computer after s(t) is f - transformed, and its output equation is:

[0065] y4(t) = f(s(t))

[0066] Preferably, in any of the above - mentioned solutions, step S4 specifically includes:

[0067] Let x7(t) be the fixed - length signal segment obtained after y4(t) undergoes endpoint detection and clipping processing, and x 7k (t) be the k - th mode after x7(t) undergoes VMD decomposition:

[0068] (1) Decompose x7(t) into multiple modes:

[0069]

[0070] k is a positive integer. The k - mode signals undergo the HT transform to obtain the corresponding analytic signals:

[0071]

[0072] Among them, the analytic signal of the mode contains the instantaneous amplitude and instantaneous frequency information of the signal. j1 is the imaginary unit, and H{x 7k (t)} is the Hilbert transform of x 7k (t);

[0073] (2) For each mode x 7k (t), calculate the variance and variance contribution rate between it and the original signal x7(t), and according to the selected modes within the preset range, calculate the state equation of the variance:

[0074] Var k = Var(x 7k (t) - x7(t))

[0075] State equation of the variance contribution rate:

[0076]

[0077] Var k is the variance, and CR k is the variance contribution rate. Reconstruct the signal within a preset modal range and remove the signals that do not meet the preset range. The state equation is:

[0078]

[0079] (3) Convert the reconstructed signal into a linear spectrogram S(t,f). Using x selected (t) as the input, perform the conversion using the short-time Fourier transform STFT(). The output equation is

[0080] S(t,f) = STFT(x selected (t))

[0081] (4) Using S(t,f) as the input, perform classification using the image classification algorithm classifier(). The output equation is:

[0082] y4 = classifier(S(t,f))

[0083] where y4 is the output classification result signal, representing the category of the signal.

[0084] Preferably, in any of the above solutions, step S5 specifically includes:

[0085] Let the length of the telescopic rod be x8(t), the telescopic speed of the telescopic rod be x9(t), the discrimination instruction of the industrial control computer be u4(t), and the control signal for the industrial control computer to drive the telescopic rod be u5(t);

[0086] The length of the telescopic rod is restricted by the speed of the telescopic rod. The state equation is:

[0087]

[0088] The speed of the telescopic rod is determined by the external driving force and the inertia and friction of the telescopic rod. The state equation is:

[0089]

[0090] where, m rod is the mass of the telescopic rod, F drive (t) is the telescopic force driven by the industrial control computer, and F2 friction (t) is the friction force generated by the telescopic rod during movement;

[0091] F drive (t) = k drive (t) * u5(t)

[0092] where k drive is the conversion coefficient from the control signal to the driving force, and y5(t) is the output signal of the sorting action. The output equation is:

[0093]

[0094] where, x 8t h res h old is the set telescopic length threshold, representing the telescopic length when the sorting action is completed.

[0095] Compared with the prior art, the advantages and beneficial effects of the present invention are as follows:

[0096] 1. The sorting device of the present invention includes a material taking mechanism composed of multiple parallel channels, and a signal plate with a predetermined inclination angle. Moreover, different feeding time sequences are adopted for the multiple parallel channels to achieve orderly acquisition of vibration signals, thereby ensuring the accuracy and reliability of vibration signal acquisition. In addition, multiple feeding hooks are arranged on each rotating wheel, improving the feeding efficiency and ejection accuracy.

[0097] 2. The sorting method of the present invention significantly improves the screening efficiency and accuracy by providing an innovative vibration signal acquisition and analysis process. Among them, signal processing algorithms are adopted, which can quickly and accurately judge the quality of fruits, greatly improving the accuracy and production efficiency of agricultural fruit screening, and meeting the requirements of modern agricultural production for efficient and precise sorting.

[0098] The additional aspects and advantages of the present invention will be partially given in the following description, partially will become obvious from the following description, or will be understood through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] The above and / or additional aspects and advantages of the present invention will become obvious and easy to understand from the description of the embodiments in conjunction with the following drawings, wherein:

[0100] Figure 1 is a schematic structural diagram of a device for asynchronous sorting of good and bad fruits based on vibration signals according to an embodiment of the present invention.

[0101] Figure 2 is a schematic structural diagram of a feeding device in a device for asynchronous sorting of good and bad fruits based on vibration signals according to an embodiment of the present invention.

[0102] Figure 3 is a schematic structural diagram of a feeding component in a device for asynchronous sorting of good and bad fruits based on vibration signals according to an embodiment of the present invention.

[0103] Figure 4It is the front view of the feeding component in an asynchronous sorting device for good and bad fruits based on vibration signals according to an embodiment of the present invention.

[0104] Figure 5 It is the structural schematic diagram of the signal acquisition device in an asynchronous sorting device for good and bad fruits based on vibration signals according to an embodiment of the present invention.

[0105] Figure 6 It is the structural schematic diagram of the sorting device in an asynchronous sorting device for good and bad fruits based on vibration signals according to an embodiment of the present invention.

[0106] Figure 7 It is the structural schematic diagram of the material selection component in an asynchronous sorting device for good and bad fruits based on vibration signals according to an embodiment of the present invention.

[0107] Figure 8a It is the original signal diagram when the fruit is a good fruit.

[0108] Figure 8b It is the original signal diagram when the fruit is a damaged good fruit.

[0109] Figure 9a It is the decomposed modal diagram when the fruit is a good fruit.

[0110] Figure 9b It is the decomposed modal diagram when the fruit is a damaged fruit.

[0111] Figure 10a It is the reconstructed signal diagram when the fruit is a good fruit.

[0112] Figure 10b It is the reconstructed signal diagram when the fruit is a damaged fruit.

[0113] Figure 11a It is the linear spectrum diagram before reconstruction when the fruit is a good fruit.

[0114] Figure 11b It is the linear spectrum diagram before reconstruction when the fruit is a damaged fruit.

[0115] Figure 12a It is the linear spectrum diagram after reconstruction when the fruit is a good fruit.

[0116] Figure 12b It is the linear spectrum diagram after reconstruction when the fruit is a damaged fruit.

[0117] Among them: 1-feeding device; 101-second motor; 102-feeding roller; 103-driving gear; 104-baffle; 105-first diversion baffle; 106-third diversion baffle; 107-baffle fixing frame; 108-feeding hook; 2-frame; 3-asynchronous feeding device; 301-first motor; 302-rotating wheel; 303-feeding hook; 304-rotating shaft; 4-signal acquisition device; 401-vibration signal sensor; 402-signal board; 5-industrial computer; 6-sorting device; 601-arc ceiling; 602-telescopic rod; 603-selection plate; 604-driving arm; 605-movable rod; 606-fixed rod; 607-fixing piece; 608-blanking partition plate; 609-support column; 7-storage box. DETAILED DESCRIPTION

[0118] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0119] like Figures 1 to 7 As shown, an asynchronous sorting device 6 for sorting good and bad fruits based on vibration signals according to an embodiment of the present invention comprises: a feeding device 1 and a sorting module; the feeding device 1 comprises a plurality of parallel feeding channels, and a sorting module is correspondingly configured at the outlet end of each feeding channel; the sorting module comprises a plurality of asynchronous loading devices 3, a signal acquisition device 4 and a sorting device 6, and the asynchronous loading devices 3, the vibration signal acquisition device 4 and the sorting device 6 are arranged in sequence along the fruit conveying direction to constitute a continuous sorting production line; the signal acquisition device 4 comprises a signal board 402, and the signal board 402 is arranged at a predetermined inclination angle relative to the horizontal plane; the plurality of asynchronous loading devices 3 throw materials onto the signal board 402 according to different timings to realize asynchronous loading.

[0120] The asynchronous loading in the present invention is specifically: the asynchronous loading device 3 of each sorting module realizes the loading of single fruit in sequence, and the current fruit triggers the loading of the next fruit after the vibration signal collection is completed, and the loading action of each sorting module remains asynchronous in time frequency, that is, the loading timing of each sorting module is independent of each other and asynchronous to realize the loading of single fruit. The above structure realizes the orderly collection of vibration signals, thereby ensuring the accuracy and reliability of vibration signal collection.

[0121] It also includes an industrial control computer 5, which is used to receive vibration signals and analyze them, and control the sorting device 6 to perform corresponding sorting operations. A storage box 7 is arranged below the sorting device 6 for storing the sorted good and bad fruits. A vibration signal sensor 401 is provided on the signal generating board for collecting the vibration signals on the signal board 402. The sorting device 6 also includes an arc-shaped ceiling 601, and the signal board 402 is arranged inside the arc-shaped ceiling 601. The arc-shaped ceiling 601 is used to form a physical shielding layer to effectively block the interference of external environmental noise on the signal collection process and improve the signal collection accuracy.

[0122] Preferably, the signal board 402 is arranged at an inclination angle greater than zero degrees and less than ninety degrees with respect to the horizontal plane, and its inclination direction extends from the end close to the asynchronous feeding device 3 to the end far from the asynchronous feeding device 3. The preferred inclination angle is sixty degrees, which can effectively suppress the bouncing phenomenon of the fruits. As a preferred embodiment, there are three sorting modules arranged in parallel, and each sorting module works independently without interfering with each other, greatly optimizing the sorting and discrimination effects.

[0123] The fruits are conveyed according to the process and are taken away one by one by the asynchronous feeding device 3 to make a parabolic motion, hitting the signal board 402. The signal board 402 sends the signals to the industrial control computer 5 in real time for real-time analysis and judgment. The industrial control computer 5 drives the sorting device 6 to perform sorting actions to achieve the sorting of good fruits and damaged fruits.

[0124] Specifically, it also includes a first motor 301. The asynchronous feeding device 3 includes a feeding component; the feeding component includes a runner 302 and a plurality of material-taking hooks 303. The runner 302 is drivingly connected to the output end of the first motor 301. The material-taking hooks 303 are uniformly arranged along the circumferential direction of the runner 302 and are used to grab and throw the fruits onto the signal board 402; and / or the feeding device 1 also includes a feeding hook 108, and the feeding hook 108 is arranged at the outlet end of the feeding channel. The feeding hook 108 cooperates with the material-taking hook 303.

[0125] Optionally, the feeding hook 108 is a hook-shaped part with a double-tooth structure, and the material-taking hook 303 is a hook-shaped part with a three-tooth structure. The feeding hook 108 and the material-taking hook 303 are arranged in a staggered manner for cooperation.

[0126] In a preferred embodiment of the present invention, the fruit is transmitted to the end of the feeding device 1 by a feeder device and temporarily stored in a double-toothed feeding hook 108. Since the friction coefficient between the fruits is small, the fruits to be fed will not be stacked, and each double-toothed feeding hook 108 can only accommodate a single fruit. The three-toothed feeding hook 303 of the asynchronous feeding device passes through the double-toothed feeding hook 108 to complete the fruit feeding operation. Among them, the sizes of the feeding hook 108 and the feeding hook 303 can be adjusted or replaced according to needs to adapt to fruits of different specifications. Each rotating wheel 302 is equipped with multiple feeding hooks 108, which effectively improves the feeding efficiency and ejection accuracy, and significantly improves the accuracy and reliability of fruit sorting. Preferably, there are twelve feeding hooks 303 on each rotating wheel 302, which are evenly distributed along the circumferential direction of the rotating wheel 302, and the central angle between each adjacent feeding hook 303 and the line connecting the center of the circle is thirty degrees. The output end of the first motor 301 is connected to the material picking hook 303 through the rotating shaft 304; the rotating shaft 304 passes through the center of the rotating wheel 302, and the rotating shaft 304 is fixedly connected to the rotating wheel 302. When there are multiple rotating wheels 302 arranged in parallel, the rotating shaft 302 is connected to multiple rotating wheels 302, and the material picking hooks on each rotating wheel 302 are staggered, that is, the projections of the material picking hooks on each rotating wheel 302 in the direction parallel to the rotating shaft 304 do not overlap, so that when loading, the loading timings of multiple rotating wheels are ensured to be different, one end of the rotating shaft 304 is connected to the output end of the first motor 304 and one end of the rotating shaft 304, and the other end of the rotating shaft 304 passes through the center of the rotating wheel 302 and is rotatably connected to the frame 2, the first motor drives the rotating shaft to rotate, and the rotating shaft drives the rotating wheel to rotate.

[0127] When multiple parallel asynchronous loading devices 3 are set, for example, three asynchronous loading devices 3 are set, each loading device includes a loading component, and thus is divided into a first loading component, a second loading component and a third loading component. For example, the material picking hooks 303 of each loading component are arranged with a phase difference: a first misalignment angle is formed between the material picking hook 303 of the first loading component and the material picking hook 303 of the second loading component, a second misalignment angle is formed between the material picking hook 303 of the second loading component and the material picking hook 303 of the third loading component, and a third misalignment angle is formed between the material picking hook 303 of the third loading component and the material picking hook 303 of the first loading component; under the drive of the first motor 301, although each loading component rotates synchronously, due to the phase difference arrangement of the material picking hooks 303, a sequential material picking operation sequence is formed: after the first loading component completes the material picking action, the second loading component executes the material picking, and then the third loading component executes the material picking, and the material picking sequence of the three loading components is not synchronous or simultaneous.

[0128] Specifically, the feeding device 1 includes a frame 2, a feeding roller 102, a diverter baffle assembly and a second motor 101. The output end of the second motor 101 is connected to the feeding roller 102. A plurality of feeding rollers 102 are arranged in sequence on the frame 2. The diverter baffle assembly is arranged above the feeding roller 102. The feeding channel is defined by the diverter baffle assembly and the feeding roller 102, and is used to divert the fruit and transport it to the corresponding sorting module.

[0129] Preferably, the first motor 301 and the second motor 101 are stepper motors. There are multiple feed rollers 102, and driving gears 103 are installed on both sides of each feed roller 102. Two adjacent feed rollers 102 are meshed through auxiliary gears. The output end of the second motor 101 is connected to a feed roller, and the first feed roller is driven to rotate, and the first feed roller is connected to the second feed roller through the auxiliary gear. The second feed roller drives the third feed roller to rotate in one direction to feed the material through the auxiliary gear.

[0130] Baffles 104 are also provided at both ends of the feed roller 102 to prevent the walnuts from falling outside during the conveying process. The feed device 1 also includes a frame 2, which is supported at both ends of the feed roller 102. The diverter baffle assembly includes a first diverter baffle 105, a second diverter baffle and at least one third diverter baffle 106, the first diverter baffle 105 is fixed on one side of the frame 2, the second diverter baffle is fixed on the other side of the frame 2, and the third diverter baffle 106 is arranged between the first diverter baffle 105 and the second diverter baffle, and is arranged parallel to the first diverter baffle 105 and the second diverter baffle or at a preset angle to separate the space above the feed roller 102 to form multiple feeding channels.

[0131] There are two third diverter baffles 106, and the width of the third diverter baffles 106 is greater than that of the first diverter baffle 105. One end of the first diverter baffle 105 away from the outlet of the feeding channel and one end of the second diverter baffle away from the outlet of the feeding channel are both provided with guide arc plates.

[0132] It also includes a partition fixing frame 107 , through which the first flow dividing partition 105 , the second flow dividing partition and the third flow dividing partition 106 are fixedly connected.

[0133] The guide arc plate facilitates the diversion and guidance of the fruit, and cooperates with the diversion baffle assembly to achieve efficient diversion of the fruit.

[0134] Optionally, the sorting device 6 includes a sorting box and a material selection assembly. A material selection assembly is provided on each of the opposite sides of the sorting box. The material selection assembly includes a fixed rod 606, a telescopic rod 602, a movable rod 605, a driving arm 604, and a material selection plate 603. The fixed rod 606 is provided on the outer side wall of the sorting box. One end of the telescopic rod 602 is connected to the fixed rod 606, and the other end of the telescopic rod 602 is connected to the movable rod 605. Both ends of the movable rod 605 are respectively pivotally connected to one end of a driving arm 604. Both ends of the material selection plate 603 are respectively pivotally connected to the other end of the driving arm 604, so as to drive the material selection plate 603 to perform a material selection operation through the telescopic movement of the telescopic rod 602. The sorting box further includes a box body. Support columns 609 are provided at the bottom of the box body. An arc-shaped ceiling 601 is provided on the box body of the sorting box. A signal plate 402 is provided above the material selection plate 603. The bottom end of the signal plate 402 is vertically aligned with the connection end between the two material selection assemblies, that is, the middle end where the two material selection plates 603 face each other. After the fruits fall from the signal plate 402, they fall into the included angle area between the two material selection plates 603.

[0135] It further includes a fixing member 607. The fixing member 607 is provided outside the sorting box. The fixed rod 606 is horizontally arranged. The telescopic rod 602 is perpendicular to the fixed rod 606. The fixing member 607 is used to fix both ends of the fixed rod 606. It further includes a blanking partition plate 608. The blanking partition plate 608 is provided below the included angle area between the two material selection plates 603. The bottom ends of the two material selection plates 603 are respectively provided above the blanking partition plate 608.

[0136] When the telescopic rod extends, it drives the driving arm to act, driving the material selection plate to open; the blanking partition plate divides the sorting box into two independent sorting spaces, and a storage box is respectively provided corresponding to the lower part of each sorting space. A material selection assembly is provided corresponding to each sorting space. The material selection plates of the two material selection assemblies are inclined and form an included angle of α, where α ∈ [0°, 180°]. In a preferred embodiment, α = 120°; after the fruits fall into the included angle area between the two material selection plates, the target material selection plate is opened according to the sorting signal, so that the fruits fall into the corresponding sorting space under the action of gravity; dedicated storage boxes are provided below each sorting space to realize the automatic classification and collection of fruits.

[0137] Working principle of the device of the present invention: Multiple fruits at the front end of the feeding device are conveyed backward by the feeding roller driven by the second motor, and the shunt partition assembly restricts the fruits to form a single-column conveyance and prevents lateral dropping; After the fruits enter the picking hook one by one, they are extracted by the feeding hook of the asynchronous feeding device. When the runner rotates and lifts the fruit-carrying feeding hook to the highest point, the fruits are projected and impact the signal plate on the inner side of the arc-shaped ceiling; The vibration signal sensor converts the impact vibration into an electrical signal and transmits it to the industrial control computer for analysis. At the same time, the fruits fall into the included angle area of the sorting plate; The industrial control computer controls the corresponding sorting plate to open according to the analysis result, so that the fruits fall into the specified sorting space, and finally the classification and collection are completed by the storage box below.

[0138] The present invention also discloses a sorting method applicable to the above-mentioned asynchronous sorting device for good and bad fruits based on vibration signals, including the following steps:

[0139] Step S1: Establish a state model of the feeding device and output the transmission speed of the fruits;

[0140] Step S2: Establish a state model of the asynchronous feeding device and control the asynchronous feeding device to project the fruits to the signal plate according to a preset trajectory;

[0141] Step S3: Establish a state model of the signal acquisition device, analyze and process the vibration signals collected by the signal plate, and send them to the industrial control computer;

[0142] Step S4: Establish a classification signal state model, analyze the vibration signals received by the industrial control computer, and output a classification result signal;

[0143] Step S5: Establish a signal state model of the sorting device, output a sorting action signal according to the classification result signal, and control the sorting device to complete the sorting of the fruits.

[0144] Specifically, step S1 specifically includes:

[0145] Let the displacement of the fruit be x1(t), the transmission speed of the fruit be x2(t), the driving force of the second motor be u1(t), x1(t) and x2(t) be state variables, and u1(t) be an input variable. Taking the acceleration of the fruit as the description quantity, the acceleration of the fruit is the second derivative of the fruit displacement. The state equation of the feeding device is:

[0146]

[0147] Among them, F motor is the driving force of the second motor, F friction is the friction force of the fruit transmission, a max is the acceleration of the fruit, m walnut is the mass of the fruit, m walnut ·a max is the inertia of the fruit, and the constraint condition is the driving force F of the stepping motormotor The frictional force F must be overcome friction and inertia. Let the constraint conditions be as follows:

[0148] F motor ≥μ·N + m walnut ·a max

[0149] where μ is the frictional coefficient of the feeding roller, N = m walnut ·g, g is the acceleration due to gravity, and F friction = μ·N;

[0150] Let the velocity of the fruit be V max , V max is the maximum velocity of the fruit. It must be within a certain range to constrain V max Within the preset range, then:

[0151]

[0152] The output equation of the feeding device state model is:

[0153] y1(t) = C1x1(t) + D1u1(t)

[0154] where the output quantity y1(t) is the conveying velocity of the fruit, C1 is the displacement unit matrix, and D1 is the driving force unit matrix of the second motor. The velocity, as a state variable, is used to describe the state of the feeding device state model and is also one of the output quantities during the feeding stage.

[0155] Specifically, step S2 specifically includes:

[0156] Let the angular displacement of the rotating wheel be x3(t), the angular velocity of the rotating wheel be x4(t), x3(t) and x4(t) are state variables, the driving torque of the first motor is u2(t), u2(t) is the input variable, and the state equation of x4(t) is:

[0157]

[0158] The acceleration of x4(t) describes the state of the rotating wheel changing with time:

[0159]

[0160] where J is the moment of inertia, T friction (t) is the frictional resistance that needs to be overcome when the output end of the first driving motor rotates. The frictional resistance that needs to be overcome when the output end of the first driving motor rotates is specifically the frictional force on the rotating shaft that needs to be overcome when the first driving motor drives the rotating shaft to rotate, so as to drive the rotating shaft to rotate. The output end of the first motor is connected to the rotating shaft, and the other end of the rotating shaft passes through the rotating wheel and is rotatably connected to the frame. T motor(t) must be less than or equal to the maximum torque T that the runner can withstand motor (t) ≤ T max , the constraint condition of J is:

[0161] T motor (t) ≤ T max

[0162] T max is the preset torque borne by the runner; w max is the critical angular velocity preset by the first motor, and x4(t) must be within a reasonable range, constraint: |x4(t)| ≤ w max ;

[0163] Let m walnut be the mass of the fruit, F centrifugal be the centrifugal force of the fruit, r be the radius from the center of the runner to the center of the feeding hook, the center of the feeding hook coincides with the center of the fruit, and the gravity of the fruit is m walnut *g; g is the acceleration due to gravity, and the two conditions for the fruit to be thrown out are:

[0164] (1) When F centrifugal is greater than m walnut *g, the fruit meets the first condition for being thrown out, and the calculation formula for the centrifugal force of the fruit is:

[0165] F centrifugal＝ m walnut *r*x4(t) 2

[0166] Simplified to:

[0167] r*x4(t) 2 > g

[0168] (2) Let θ release be the preset angle at which the fruit is thrown out. When x3(t) reaches the ideal best preset angle θ release the fruit is released and thrown out, and when x3(t) = θ release the fruit meets the second condition for being thrown out. The throwing speed of the fruit is jointly determined by x4(t) and r. Therefore, the tangential velocity of the fruit thrown out is:

[0169] v t h row (t) = r*x4(t)

[0170] v t h row (t) is the tangential velocity of the fruit thrown out. The angle at which the fruit is thrown out is determined by the angle of the runner. When the angle of the runner reaches a certain predetermined value, the fruit will be thrown out. Therefore, the angle at which the fruit is thrown out is:

[0171] θt h row h(t) = x3(t)

[0172] θ t h row h(t) is the angle at which the fruit is thrown;

[0173] The output equation of the asynchronous feeding device state model is:

[0174] y2(t) = C2x4(t) + D2u2(t)

[0175] y3(t) = C3x3(t)

[0176] Among them, y2(t) is the output parameter of the throwing speed of the fruit, y3(t) is the output parameter of the throwing angle of the fruit, u2(t) is the torque input by the second motor, C2 is the relationship coefficient between the angular velocity and the throwing speed, C3 is the relationship coefficient between the angular displacement and the throwing angle, and D2 is the influence coefficient of u2(t) on the output;

[0177] Specifically, step S3 specifically includes:

[0178] Let the position where the fruit hits the signal board be x5(t), x5(t) is the distance between the position where the fruit hits the signal board and the bottom end of the signal board, the impact speed of the fruit hitting the signal board is x6(t), and the force F1 when the fruit hits the signal board impact F1(t), F1 impact F1(t) is an input variable, jointly determined by the mass m of the fruit walnut and x6(t). The frictional force when the fruit collides with the signal board is F1 friction F1(t), and the state equation of x5(t) is:

[0179]

[0180] x6(t) is determined by the forces acting on the fruit. On the signal board, the main acting forces are F1 impact F1(t) and F1 friction F1(t), and the state equation of the fruit speed is:

[0181]

[0182] Among them:

[0183] F1 friction F1(t) = μ1 * N1 * sign(x6(t))

[0184] Among them, m walnut is the mass of the fruit, μ1 is the friction coefficient of the signal board, N1 is the normal force when the fruit hits the signal board, and sign(x6(t)) is the sign function of x6(t), indicating F1friction (t) is in the opposite direction to the movement direction of the fruit; for each impact, only a single impact is allowed to fall into the sorting position, x6(t after ) is the velocity direction of the fruit after impact. For each impact, only a single impact is allowed to fall into the sorting position, and the constraint is x6(t after ) < 0, indicating that the velocity direction of the fruit after impact is opposite to the velocity direction of the fruit before impact; ensure that there is no secondary impact.

[0185] Let s(t) be the signal of the actual physical vibration generated by the signal board, which can be determined by F1 impact (t) and x6(t). The output equation is:

[0186] s(t) = C4 * F1 impact (t) + D3 * x6(t)

[0187] where C4 and D3 are constant coefficients. C4 is the relationship coefficient between s(t) and F1 impact (t), and D3 is the relationship parameter coefficient between s(t) and x6(t). Let the electrical signal sent to the industrial control computer after s(t) is converted by f be y4(t), and its output equation is:

[0188] y4(t) = f(s(t))

[0189] Specifically, step S4 specifically includes:

[0190] Let x7(t) be the fixed-length signal segment obtained after y4(t) undergoes endpoint detection and clipping processing, and x 7k (t) is the k-th mode after the VMD decomposition of x7(t), and u3(t) is the original signal received by the industrial control computer, which is used to implement the signal reception status monitoring function of the industrial control computer; when the industrial control computer successfully receives the original signal, it is determined to be in a normal working state; when the industrial control computer does not receive the original signal, it is determined to be in an abnormal working state. Figure 8a is the original signal diagram of the fruit being a good fruit, Figure 8b is the original signal diagram of the fruit being a damaged fruit.

[0191] (1) Decompose x7(t) into multiple modes, Figure 9a is the decomposition mode diagram of the fruit being a good fruit, Figure 9b is the decomposition mode diagram of the fruit being a damaged fruit:

[0192]

[0193] k is a positive integer, and the k mode signals undergo the HT transform to obtain the corresponding analytic signals:

[0194]

[0195] Among them, the analytic signal of the mode contains the instantaneous amplitude and instantaneous frequency information of the signal. j1 is the imaginary unit, and H{x 7k (t)} is the Hilbert transform of x 7k (t);

[0196] (2) For each mode x 7k (t), calculate the variance and variance contribution rate between it and the original signal x7(t), and select the modes within the preset range. The modes within the preset range are set according to the debugging results and are the modes that can best represent the characteristics of the current signal. The state equation for variance calculation is:

[0197] Var k = Var(x 7k (t) - x7(t))

[0198] The state equation for variance contribution rate:

[0199]

[0200] Var k is the variance, and CR k is the variance contribution rate. Reconstruct the signal within the preset range of modes, and remove the signals that do not meet the preset range, so as to improve the classification effect of the signal. The state equation is:

[0201]

[0202] Figure 10a is the reconstructed signal diagram of the fruit being a good fruit, Figure 10b is the reconstructed signal diagram of the fruit being a damaged fruit. It can be seen that the discrimination degree of the signal becomes larger after signal reconstruction in the time domain.

[0203] Figure 11a is the linear spectrum diagram before reconstruction of the fruit being a good fruit, Figure 11b is the linear spectrum diagram before reconstruction of the fruit being a damaged fruit.

[0204] (3) Convert the reconstructed signal into a linear spectrogram S(t, f). Using x selected (t) as the input, perform the conversion using the short-time Fourier transform STFT(), and the output equation is

[0205] S(t, f) = STFT(x selected (t))

[0206] Figure 12a is the linear spectrum diagram after reconstruction of the fruit being a good fruit, Figure 12b is the linear spectrum diagram after reconstruction of the fruit being a damaged fruit, and the discrimination degree becomes significantly larger.

[0207] (4) Using S(t,f) as the input, classify it using the image classification algorithm classifier(), and the output equation is:

[0208] y4 = classifier(S(t,f))

[0209] Where y4 is the output classification result signal, representing the category of the signal. y4 can output 0 or 1. For example, 0 represents a damaged fruit, and 1 represents a complete fruit.

[0210] Specifically, step S5 specifically includes:

[0211] Let the length of the telescopic rod be x8(t), the telescopic speed of the telescopic rod be x9(t), x8(t) and x9(t) are state variables, the discrimination instruction of the industrial control computer is u4(t), and the control signal for the industrial control computer to drive the telescopic rod is u5(t); u4(t) and u5(t) are input variables.

[0212] The length of the telescopic rod is restricted by the speed of the telescopic rod, and the state equation is:

[0213]

[0214] The speed of the telescopic rod is determined by the external driving force, as well as the inertia and friction of the telescopic rod. The external driving force is converted from the control signal u5(t), and the state equation is:

[0215]

[0216] Where, m rod is the mass of the telescopic rod, F drive (t) is the telescopic force driven by the industrial control computer, F2 friction (t) is the friction force generated by the telescopic rod during movement and is proportional to x9(t);

[0217] F drive (t) = k drive (t) * u5(t)

[0218] Where k drive is the conversion coefficient from the control signal to the driving force, and y5(t) is the output signal of the sorting action. The output equation is:

[0219]

[0220] Where, x 8t h res h old is the set telescopic length threshold, representing the telescopic length when the sorting action is completed.

[0221] In the above invention, t is a time parameter. When the telescopic rod reaches a predetermined telescopic length threshold, a sorting action is performed. The output signal y5(t) is 1, indicating that the sorting action is completed; otherwise, the output is 0, indicating that the sorting action is not completed. The output equation describes whether the telescopic rod has completed the sorting action. Through x8(t), it is judged whether the action has completed the sorting action. The sorting method of the present invention significantly improves the screening efficiency and accuracy by providing an innovative vibration signal acquisition and analysis process. Among them, signal processing algorithms are used to quickly and accurately judge the quality of fruits, greatly improving the accuracy and production efficiency of agricultural fruit screening, and meeting the requirements of modern agricultural production for efficient and precise sorting.

[0222] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0223] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention without departing from the principles and purposes of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An asynchronous sorting device for good and bad fruits based on vibration signals, comprising: Feeding device and sorting module; characterized in that, The feeding device comprises a plurality of parallel feeding channels, and the outlet end of each feeding channel is correspondingly provided with one of the sorting modules; The sorting module comprises a plurality of asynchronous feeding devices, a signal collecting device and a sorting device, wherein the asynchronous feeding device, the vibration signal collecting device and the sorting device are arranged in sequence along the fruit conveying direction to form a continuous sorting production line; The signal collection device comprises a signal plate, and the signal plate is arranged at a predetermined inclination angle relative to a horizontal plane; The plurality of asynchronous feeding devices throw materials onto the signal board in different time sequences to achieve asynchronous feeding.

2. The device for asynchronously sorting good and bad fruits based on vibration signals as claimed in claim 1, characterized in that: It also includes a first motor, and the asynchronous feeding device includes a feeding assembly; the feeding assembly includes a rotating wheel and a plurality of feeding hooks, the rotating wheel is drivingly connected to the output end of the first motor, and the feeding hooks are evenly arranged along the circumferential direction of the rotating wheel, and are used to grab and throw fruits onto the signal plate; and / or the feeding device also includes a feeding hook, the feeding hook is arranged at the outlet end of the feeding channel, and the feeding hook cooperates with the feeding hook.

3. The device for asynchronously sorting good and bad fruits based on vibration signals as claimed in claim 1, characterized in that: The feeding device includes a frame, a feeding roller, a diverter baffle assembly and a second motor, the output end of the second motor is connected to the feeding roller, a plurality of feeding rollers are arranged in sequence on the frame, and the diverter baffle assembly is arranged above the feeding roller; the feeding channel is defined by the diverter baffle assembly and the feeding roller, and is used to divert the fruit and transport it to the corresponding sorting module.

4. The device for asynchronously sorting good and bad fruits based on vibration signals as claimed in claim 1, characterized in that: The sorting device includes a sorting box and a material selection component. One material selection component is respectively arranged on the opposite sides of the sorting box. The material selection component includes a fixed rod, a telescopic rod, a movable rod, a driving arm and a material selection plate. The fixed rod is arranged on the outer side wall of the sorting box, one end of the telescopic rod is connected to the fixed rod, and the other end of the telescopic rod is connected to the movable rod. The two ends of the movable rod are respectively pivoted to one end of one of the driving arms, and the two ends of the material selection plate are respectively pivoted to the other end of the driving arm, so that the material selection plate can be driven to perform material selection operations through the telescopic movement of the telescopic rod.

5. A method for sorting good and bad fruits by an asynchronous sorting device based on vibration signals as described in any one of claims 1 to 4, characterized in that: The steps include: Step S1: Establish a feeding device state model and output the transmission speed of the fruit; Step S2: establishing a state model of an asynchronous feeding device, and controlling the asynchronous feeding device to throw the fruit to the signal board according to a preset trajectory; Step S3: Establishing a state model of the signal acquisition device, analyzing and processing the vibration signal collected by the signal board, and sending it to the industrial computer; Step S4: establishing a classification signal state model, analyzing the vibration signal received by the industrial computer, and outputting a classification result signal; Step S5: Establish a signal state model of the sorting device, output a sorting action signal according to the classification result signal, and control the sorting device to complete fruit sorting.

6. The method for sorting good and bad fruits by using an asynchronous sorting device based on vibration signals as claimed in claim 5, characterized in that: The step S1 specifically includes: Assume that the displacement of the fruit is x1(t), the transmission speed of the fruit is x2(t), the driving force of the second motor is u1(t), and the acceleration of the fruit is used as the descriptor. The acceleration of the fruit is the second-order derivative of the displacement of the fruit. The state equation is: Among them, F motor is the second motor driving force, F friction is the friction force of the fruit transmission, a max is the acceleration of the fruit, m walnut is the mass of the fruit, m walnut ·a max is the inertia of the fruit, and the constraints are: F motor ≥μ·N+m walnut ·a max μ is the friction coefficient of the feed roller, N = m walnut g, g is the acceleration due to gravity, F friction =μ·N; Let the speed of the fruit be V max , constrain V max Within the preset range, then: The output equation of the feeding device state model is: y1(t)=C1x1(t)+D1u1(t) The output y1(t) is the conveying speed of the fruit, C1 is the displacement unit matrix, and D1 is the second motor driving force unit matrix.

7. The method for sorting good and bad fruits by using an asynchronous sorting device based on vibration signals as claimed in claim 6, characterized in that: The step S2 specifically includes: Assume that the angular displacement of the wheel is x3(t), the angular velocity of the wheel is x4(t), the driving torque of the first motor is u2(t), and the state equation of x4(t) is: The acceleration x4(t) describes the state of the wheel over time: Where J is the moment of inertia, T friction (t) is the friction resistance that needs to be overcome when the output end of the first drive motor rotates, and the constraint condition of J is: T motor (t)≤T max T max is the preset torque that the wheel is subjected to; w max The preset critical angular velocity of the first motor, constrained by: |x4(t)|≤w max ; Assume m walnut is the mass of the fruit, F centrifugal is the centrifugal force of the fruit, r is the radius from the center of the wheel to the center of the feeding hook, and the fruit gravity is m walnut *g; Two conditions for the fruit to be thrown: (1) When F centrifugal Greater than m walnut *g, the fruit meets the first condition to be thrown out, and the formula for calculating the centrifugal force of the fruit is: F centrifugal＝ m walnut *r*x4(t) 2 Simplified to: r*x4(t) 2 >g (2) Let θ release is the preset angle at which the fruit is thrown, x3(t)=θ release When , the fruit meets the second condition of being thrown, the fruit is thrown, and the linear velocity of the fruit is: v t h row (t)=r*x4(t) v t h row (t) is the linear velocity of the fruit, and the angle of the fruit is: θ t h row (t)=x3(t) θ t h row (t) is the angle at which the fruit is thrown; The output equation of the asynchronous feeding device state model is: y2(t)=C2x4(t)+D2u2(t) y3(t)=C3x3(t) Among them, y2(t) is the output parameter of the fruit throwing speed, y3(t) is the output parameter of the fruit throwing angle, u2(t) is the torque input by the second motor, C2 is the relationship coefficient between angular velocity and throwing speed, C3 is the relationship coefficient between angular displacement and throwing angle, and D2 is the influence coefficient of u2(t) on the output.

8. The method for sorting good and bad fruits by using an asynchronous sorting device based on vibration signals as claimed in claim 7, characterized in that: The step S3 specifically includes: Assume that the position where the fruit hits the signal plate is x5(t), the impact speed of the fruit hitting the signal plate is x6(t), and the force of the fruit hitting the signal plate is F1 impact (t), the friction force when the fruit collides with the signal plate is F1 friction (t), x5(t) state equation is: The fruit velocity state equation is: in: F1 friction (t)=μ1*N1*sign(x6(t)) Among them, m walnut is the fruit mass, μ1 is the friction coefficient of the signal plate, N1 is the normal force when the fruit hits, sign(x6(t)) is the sign function of x6(t), indicating F1 friction (t) is opposite to the movement direction of the fruit; for each collision, only one collision is allowed to fall into the sorting position, x6(t after ) is the velocity direction of the fruit after impact, and the constraint is x6(t after )<0, indicating that the direction of the fruit's velocity after impact is opposite to that before impact; Assume s(t) is the actual physical vibration signal generated by the signal board, and the output equation is: s(t)=C4*F1 impact (t)+D3*x6(t) Where C4 is s(t) and F1 impact (t), D3 is the relationship coefficient between s(t) and x6(t), and the electrical signal sent to the industrial computer after s(t) is converted by f is y4(t), and its output equation is: y4(t)=f(s(t)) 9. The method for sorting good and bad fruits by using an asynchronous sorting device based on vibration signals as claimed in claim 8, characterized in that: The step S4 specifically includes: Let x7(t) be the fixed-length signal segment obtained after endpoint detection and clipping of y4(t), x 7k (t) is the kth mode of x7(t) after VMD decomposition: (1) Decompose x7(t) into multiple modes: k is a positive integer, and k modal signals are transformed by HT to obtain the corresponding analytical signals: Among them, the analytical signal of the mode Contains the instantaneous amplitude and instantaneous frequency information of the signal, j1 is an imaginary unit, H{x 7k (t)} is x 7k Hilbert transform of (t); (2) For each mode x 7k (t), calculate the variance and variance contribution rate between the original signal x7(t), and calculate the state equation of the variance according to the mode selected in the preset range: Yes k =There is(x 7k (t)-x7(t)) Variance contribution state equation: Var k is the variance, CR k is the variance contribution rate, reconstructs the signal within the preset modal range, removes the signal that does not meet the preset range, and the state equation is: (3) Convert the reconstructed signal into a linear spectrum S(t,f), x selected (t) is used as input and converted using short-time Fourier transform STFT(). The output equation is S(t,f)=STFT(x selected (t)) (4) S(t,f) is used as input and classified using the image classification algorithm classifier(). The output equation is: y4=classifier(S(t,f)) Where y4 is the output classification result signal, indicating the category of the signal.

10. The method for sorting good and bad fruits by using an asynchronous sorting device based on vibration signals as claimed in claim 9, characterized in that: The step S5 specifically includes: Assume that the length of the telescopic rod is x8(t), the telescopic rod extension speed is x9(t), the identification instruction of the industrial computer is u4(t), and the control signal of the industrial computer driving the telescopic rod is u5(t); The length of the telescopic rod is constrained by the velocity of the telescopic rod, and the state equation is: The speed of the telescopic rod is determined by the external driving force and the inertia and friction of the telescopic rod, and the state equation is: Among them, m rod is the mass of the telescopic rod, F drive (t) is the extension force driven by the industrial computer, F2 friction (t) is the friction force generated by the telescopic rod during movement; F drive (t)=k drive (t)*u5(t) where k drive is the conversion coefficient from control signal to driving force, y5(t) is the output signal of the sorting action, and the output equation is: Among them, x 8t h res h old It is the set telescopic length threshold, which indicates the telescopic length when the sorting action is completed.

Images