Fruit transport line dynamic weighing method and system based on multi-angle slope

Abstract

The invention relates to a fruit conveying line dynamic weighing method and system based on a multi-angle slope. The method comprises the following steps: acquiring a normal force data set of a target fruit passing through at least three preset slopes at different angles within a preset time, and performing preprocessing; the average value of the preprocessed normal force data set is calculated, and the corresponding masses of the target fruit passing through the slopes of the different angles are calculated based on the average value, the preset gravitational acceleration and the different angles; and calculating the mass corresponding to the target fruit passing through the slopes of different angles based on a weighted average method, and compensating the mass based on a preset high-speed motion correction factor to obtain a mass value corresponding to the target fruit. According to the method, the normal force is directly measured through the short slopes with different inclination angles, the weighted average algorithm is combined to calculate the mass, the single-point contact error caused by rolling is eliminated, the slopes are designed according to the characteristics that the fruits are round and easy to roll, pressure changes are captured, motion compensation is carried out, and the measurement stability of the fruits with irregular skins is ensured.

Description

Technical Field

[0001] The present application relates to the technical field of fruit sorting and processing, and particularly to a dynamic weighing method and system for a fruit transportation line based on multi-angle slopes. Background Art

[0002] The high-speed fruit sorting production line is a core device in modern fruit and vegetable processing, packaging, and logistics, achieving efficient and precise grading and sorting through automation technology. As a special type of measurement object, fruits have diverse shapes, uneven surfaces, and thin skins that are prone to damage, making traditional weighing methods difficult to apply. Especially on high-speed transportation lines, the movement state, contact instability, and possible rolling of fruits will have a significant impact on the measurement results.

[0003] As shown in Chinese Patent CN109807079B, a fully automatic weighing fruit sorting system is disclosed. The main structure includes a fruit feeding table, an orderly conveying table, a transmission part, a weighing table, a sorting table, a fruit transfer vehicle, and a control console. The fruit feeding table adopts a fully automatic fruit feeding system or a simple fruit dumping device. The fruit conveying port of the fruit feeding table is in contact with the rear end of the orderly conveying table, and fruits enter the orderly conveying table from the fruit feeding table. The front end of the orderly conveying table is fixedly connected to the rear end of the transmission part. The fruit transfer vehicle is installed on the sorting table in a slot type and can operate along a fixed track. The front part of the transmission part is fixedly connected to the rear end of the sorting table, and the lower end of the weighing table is fixedly connected to the upper side of the rear end of the sorting table. The entire set of equipment is controlled by the control console to achieve full automation of fruit sorting. The main structure of this system is reasonable, and the control principle is reliable. It can realize the non-destructive and orderly transfer of fruits in a stacked state, is suitable for sorting and transferring various fruits, and has accurate and high-efficiency fruit sorting.

[0004] However, in the prior art, when weighing target fruits through a weighing table, it is necessary to slow down or stop the high-speed production line, so that the weighing time cannot match the speed of the high-speed production line, resulting in low fruit sorting efficiency. Summary of the Invention

[0005] 1. Problems to be Solved Based on this, it is necessary to provide a dynamic weighing method and system for a fruit transportation line based on multi-angle slopes that can match the speed of the high-speed production line and then improve the fruit sorting efficiency in response to the above technical problems.

[0006] 2. Technical Solutions In a first aspect, the present application provides a dynamic weighing method for a fruit transportation line based on multi-angle slopes. The method includes: Collecting a dataset of normal forces of target fruits passing through at least three different-angle slopes within a preset time and performing preprocessing; Calculate the average value of the preprocessed normal force data set, and calculate the mass of the target fruit corresponding to different angles of the ramp based on the average value of the preprocessed normal force data set, the preset gravitational acceleration, and different angles; Calculate the mass of the target fruit corresponding to different angles of the ramp based on the weighted average method, and obtain the mass value corresponding to the target fruit after compensating the mass based on the preset high-speed motion correction factor.

[0007] In one embodiment, load the preset fruit tray parameter set based on the target fruit type; When the fruit tray is detected, calculate the height of the target fruit on the fruit tray; When the target fruit is inconsistent with the preset height corresponding to the target fruit in the preset fruit tray parameter set, send a fruit tray recycling signal.

[0008] In one embodiment, calculating the mass of the target fruit corresponding to different angles of the ramp based on the average value of the preprocessed normal force data set, the preset gravitational acceleration, and different angles includes: The formula for calculating the mass of the target fruit corresponding to different angles of the ramp is as follows: ; Where, is the mass of the target fruit calculated through the ramp at an angle of , is the normal force detected by the ramp at an angle of for the target fruit, is the gravitational acceleration, is the preset ramp angle; Compare whether the mass of the target fruit corresponding to different angles of the ramp is within the mass range corresponding to the preset target fruit type; If the mass of the target fruit corresponding to different angles of the ramp is outside the mass range corresponding to the preset target fruit type, calculate the difference between the mass of the target fruit corresponding to different angles of the ramp and the mass range corresponding to the preset target fruit type, and set the difference as an outlier. Query the abnormal operation corresponding to the outlier in the preset abnormal operation library, and different ranges of outliers and the abnormal operations corresponding to the outliers are stored in the abnormal operation library.

[0009] In one embodiment, calculating the mass of the target fruit corresponding to different angles of the ramp based on the weighted average method includes: Construct a normal force equation system for different angles; The formula for calculating the mass of the target fruit corresponding to different angles of the ramp based on the weighted average method is as follows: ; Among them, is the mass of the target fruit calculated through the ramp at an angle, is the normal force obtained by detecting the target fruit passing through the ramp at an angle, is the normal force obtained by detecting the target fruit passing through the ramp at an angle, is the normal force obtained by detecting the target fruit passing through the ramp at an angle, is the gravitational acceleration, , and are preset ramp angles; Compensate the mass calculated by the weighted average method based on a preset dynamic compensation coefficient.

[0010] In one embodiment, compensating the mass calculated by the weighted average method based on a preset dynamic compensation coefficient includes: Construct a system deviation compensation based on the speed of the target fruit, and the formula is as follows: ; Among them, is the system deviation compensation, is the speed of the target fruit.

[0011] In one embodiment, apply a corresponding compensation coefficient in a preset compensation coefficient library based on the fruit type and weight range. The compensation database stores fruit types, weight ranges, and corresponding compensation coefficients; Dynamically optimize the compensation coefficient based on supervised learning and generate a corresponding table of fruit type, weight range, and compensation coefficient; Update the compensation coefficient based on historical data.

[0012] In a second aspect, the present application also provides a dynamic weighing system for a fruit transportation line based on multi-angle ramps. The system includes: A normal force acquisition module for acquiring a normal force data set of the target fruit passing through at least three different angle ramps within a preset time and performing preprocessing; A mass calculation module for calculating the average value of the preprocessed normal force data set and calculating the mass corresponding to the target fruit passing through different angle ramps respectively based on the average value of the preprocessed normal force data set, a preset gravitational acceleration, and different angles; A mass compensation module for calculating the mass corresponding to the target fruit passing through different angle ramps based on the weighted average method and compensating the mass based on a preset high-speed motion correction factor to obtain the mass value corresponding to the target fruit.

[0013] In a third aspect, the present application also provides a computer system. The computer system includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: Collect a normal force data set of a target fruit passing through at least three different-angled slopes within a preset time and perform preprocessing; Calculate the average value of the preprocessed normal force data set and calculate the mass of the target fruit corresponding to different-angled slopes respectively based on the average value of the preprocessed normal force data set, a preset gravitational acceleration, and different angles; Calculate the mass of the target fruit corresponding to different-angled slopes based on the weighted average method and obtain the mass value corresponding to the target fruit after compensating the mass based on a preset high-speed motion correction factor.

[0014] In a fourth aspect, the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the following steps are implemented: Collect a normal force data set of a target fruit passing through at least three different-angled slopes within a preset time and perform preprocessing; Calculate the average value of the preprocessed normal force data set and calculate the mass of the target fruit corresponding to different-angled slopes respectively based on the average value of the preprocessed normal force data set, a preset gravitational acceleration, and different angles; Calculate the mass of the target fruit corresponding to different-angled slopes based on the weighted average method and obtain the mass value corresponding to the target fruit after compensating the mass based on a preset high-speed motion correction factor.

[0015] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented: Collect a normal force data set of a target fruit passing through at least three different-angled slopes within a preset time and perform preprocessing; Calculate the average value of the preprocessed normal force data set and calculate the mass of the target fruit corresponding to different-angled slopes respectively based on the average value of the preprocessed normal force data set, a preset gravitational acceleration, and different angles; Calculate the mass of the target fruit corresponding to different-angled slopes based on the weighted average method and obtain the mass value corresponding to the target fruit after compensating the mass based on a preset high-speed motion correction factor.

[0016] 3. Beneficial effects This application adopts the above method to directly measure the normal force through short slopes with different inclination angles, and combines a weighted average algorithm to calculate the mass, eliminating the single-point contact error caused by rolling. This solution designs slopes according to the characteristics of round and easily rolling fruits, captures pressure changes and performs motion compensation to ensure the measurement stability of fruits with irregular skins. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 FIG. 6 is a schematic structural diagram of a dynamic weighing method for a fruit transportation line based on multi-angle slopes in an embodiment; Figure 2 FIG. 9 is a flowchart of a dynamic weighing method for a fruit transportation line based on multi-angle slopes in an embodiment; Figure 3 FIG. 12 is a structural block diagram of a dynamic weighing system for a fruit transportation line based on multi-angle slopes in an embodiment; Figure 4 FIG. 15 is an internal structural diagram of a computer system in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to make the objectives, technical solutions and advantages of this application clearer, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0019] Existing fruit weighing technologies face various technical defects and cannot meet the actual requirements of modern high-speed production lines. Although the accuracy of traditional static weighing can reach ±0.1 g, each single measurement takes 2 - 3 seconds and requires manual operation, which is completely unsuitable for the high-speed production line rhythm of processing more than 4 pieces per second; for round fruits, static placement is also prone to instability, affecting the measurement accuracy. Although the response time of conveyor belt dynamic weighing technology is shortened to 0.1 - 0.5 seconds, due to being mainly designed for flat objects, the irregular characteristics of the fruit surface lead to uneven force, and the measurement error can reach ±5 - 50 g. Moreover, the frequent contact between the fruit and the rollers is likely to cause skin abrasion. The response time of hanging weighing technology is relatively long (0.5 - 1 second), far exceeding the 0.1 second required by the production line. In addition, the design of the hanging device destroys the continuity of the production line and may also cause damage to the fruit skin due to improper clamping force. Although the laser volume estimation combined with density measurement technology has a fast scanning speed, due to relying on indirect estimation rather than direct measurement, it is sensitive to the density differences of fruits (such as about 0.85 g / cm³ for apples and about 0.9 g / cm³ for grapes) and the non-uniformity of the internal structure, resulting in an error of 5 - 10%. At the same time, the system complexity is high and the maintenance cost is expensive.

[0020] Particularly crucial is that the roller - type dynamic weighing technology, which is closest to the technical path of the present invention, still faces numerous challenges when applied in the fruit field: The contact area between the fruit and the roller is small, and the circular and uneven surface causes significant fluctuations in force transmission, resulting in limited measurement accuracy; The frequent contact between the hard roller and the fruit surface may not only wear the fruit epidermis (especially for perishable fruits such as strawberries and blueberries), but also accelerate the wear of the equipment itself and is not suitable for long - term continuous operation; In addition, the fruit quality varies greatly (ranging from 50g to 300g), and sensors with a higher sampling rate are required to accurately capture force changes. Although the instantaneous force pulse measurement is relatively fast, the impact process may damage the fruit, and it is difficult to form a stable force on the irregular surface. At the same time, complex subsequent conveying designs are required. The vibration analysis method has a long measurement time (more than 1 second), high computational complexity, is not suitable for high - speed production lines, and is sensitive to changes in fruit shape.

[0021] In the prior art, whether it is the roller - type weighing or the weighing platform design, it is difficult to simultaneously solve the problems of force fluctuations caused by fruit rolling and epidermis damage within 0.1 second. This technical difficulty has long restricted the application of high - speed production lines.

[0022] The dynamic weighing method for fruit transportation lines based on multi - angle slopes provided by the embodiments of the present application can be applied to, for example, Figure 1 the structural schematic diagram shown. In this method, a temperature sensor is set to read the ambient temperature. The system is provided with four different sensors (P1, P2, P3, P4), and each sensor has a clear function. P1 (the first photoelectric sensor) is used to detect the leading edge of the fruit tray, trigger the laser scanner (200Hz) and record the starting moment; P2 (the second photoelectric sensor) confirms that the fruit tray enters the measurement area and starts collecting normal force data; P3 (the third photoelectric sensor) may be used to mark the end of the measurement or monitor the position of the fruit tray on the slope; P4 (the fourth photoelectric sensor) confirms that the fruit tray leaves the measurement area and triggers the mass calculation and sorting operation. The running speed of the conveyor belt is set to 1.5m / s. The first photoelectric sensor is used to detect the fruit tray signal. When the first photoelectric sensor detects the leading edge of the fruit tray, it sends a prompt signal to activate the laser scanner. The laser scanner scans the content of the fruit tray at a frequency of 200Hz and records the starting moment at the same time. When the fruit tray passes through the second photoelectric sensor, it enters the measurement stage.

[0023] In one embodiment, as Figure 2 shown, the method includes the following steps: Step 202, collect the normal force data set of the target fruit passing through at least three different - angle slopes within a preset time and perform pre - processing.

[0024] Among them, in this embodiment, the multi-angle ramp system adopts a specific angle combination of 15°, 30°, and 35° respectively; the preprocessing operation includes digital filtering (low-pass filtering with a cut-off frequency of 500 Hz to remove vibration interference at about 15 Hz at a conveyor belt speed of 1.5 m / s) of the force signal at each measurement point, removing outliers, and calculating the average value within the effective time window. After adjusting the force value using the temperature compensation coefficient, the mass is calculated using the weighted average method based on the normal force equations.

[0025] It is worth mentioning that the preprocessing step includes temperature compensation. The measured values of the sensor are affected by changes in the ambient temperature, and the goal is to achieve high-precision mass measurement with an accuracy of ±2 g. Temperature changes may cause the sensor output to drift, introducing measurement errors. Therefore, it is crucial to correct the data through temperature compensation. Regarding the timing of temperature compensation, it is arranged after digital filtering, removing discrete values, and calculating the average value for several reasons: First, digital filtering removes vibration interference (such as 15 Hz) caused by the operation of the conveyor belt (1.5 m / s) through low-pass filtering (cut-off frequency 500 Hz), ensuring signal smoothness; second, removing discrete values can eliminate abnormal points, such as sudden change values caused by sensor noise or fruit jitter, improving data stability; finally, calculating the average value within the effective time window obtains a stable representative value of the normal force, providing a reliable basis for subsequent correction. Temperature compensation is performed after these steps because if compensation is applied before filtering and removing discrete values, noise and outliers may interfere with the correction process, resulting in inaccurate compensation results. On the contrary, performing temperature compensation after obtaining a stable average value can directly correct the clean data, ensuring the effectiveness of the correction. In summary, temperature compensation, as part of the preprocessing, is performed after digital filtering, removing discrete values, and calculating the average value to ensure data stability and compensation accuracy.

[0026] Step 204, calculate the average value of the preprocessed normal force data set and calculate the mass of the target fruit corresponding to different angles of the ramp based on the average value of the preprocessed normal force data set, the preset acceleration due to gravity, and different angles.

[0027] Among them, the formula for calculating the mass of the target fruit corresponding to different angles of the ramp is as follows: ; Among them, is the mass of the target fruit calculated through the ramp at an angle of , is the normal force detected by the target fruit passing through the ramp at an angle of , is the acceleration due to gravity, is the preset ramp angle.

[0028] Step 206: Calculate the mass of the target fruit corresponding to different - angle slopes based on the weighted - average method, and obtain the mass value corresponding to the target fruit after compensating the mass based on a preset high - speed motion correction factor.

[0029] Among them, a normal - force equation system for different angles is constructed; The formula for calculating the mass of the target fruit passing through different - angle slopes based on the weighted - average method is as follows: ; Among them, is the mass of the target fruit calculated by passing through the slope at an angle of , is the normal force detected when the target fruit passes through the slope at an angle of , is the normal force detected when the target fruit passes through the slope at an angle of , is the normal force detected when the target fruit passes through the slope at an angle of , is the acceleration due to gravity, , and are preset slope angles.

[0030] In the above - mentioned dynamic weighing method for fruit transportation lines based on multi - angle slopes, first, the measurement accuracy is improved. By designing a multi - angle slope force - measuring system, errors of indirect estimation methods such as density assumption and shape modeling are eliminated. The mass of the fruit is calculated by directly measuring the physical quantity related to mass (normal force \(F_N = m\cdot g\cdot\cos\theta\)), ensuring a high accuracy of ±2g. Second, the measurement efficiency is greatly improved. It is ensured that the complete measurement process can be completed within 0.1 second, adapting to the high - speed assembly - line beat requirement of at least 4 pieces per second, meeting the high - efficiency requirements of modern agricultural product processing. Third, considering the special characteristics of fruits such as being round, having a rough surface, and being easily worn, a "slope - flat section" alternating structure and a low - friction contact surface are designed to ensure stable transmission and no damage to the fruits during the measurement process. At the same time, a multi - condition measurement system is used to adapt to fruits of different shapes, improving the versatility and reliability of the overall measurement system. Through the innovative slope - structure design and dynamic compensation algorithm of the present invention, the technical bottleneck of traditional dynamic weighing in fruit measurement is broken through, providing a new technical solution for the fruit sorting assembly line.

[0031] In one embodiment, since there may be empty trays in the high - speed assembly line, before collecting the normal - force data set, empty - tray detection needs to be performed, which specifically includes: Load the preset fruit plate parameter set based on the target fruit type; when a fruit plate is detected, calculate the height of the target fruit on the fruit plate; when the height of the target fruit is inconsistent with the preset height corresponding to the target fruit in the preset fruit plate parameter set, send a fruit plate recycling signal to control the conveyor belt to direct the fruit plate to the recycling channel; when the height of the target fruit is consistent with the preset height corresponding to the target fruit in the preset fruit plate parameter set, collect the normal force data set of the target fruit passing through at least three different angles of slopes within a preset time and perform preprocessing.

[0032] Among them, the parametric definition of the fruit plate structure is as follows: D: Fruit plate diameter (cm); H: Fruit plate height (cm); d: Hollow diameter at the bottom of the fruit plate (cm); t: Thickness of the fruit plate edge (cm); h: Height of the bottom edge of the fruit plate (cm); a: Fruit exposure height (cm); p: Height of the side positioning protrusion (cm); w: Width of the side positioning protrusion (cm). Among them, a (fruit exposure height) is the key parameter and satisfies the following constraints: , where is the slope length, is the maximum slope angle; , where is the fruit diameter.

[0033] The parametric definition of the slope module is as follows: L: Slope length (cm); W: Slope width (cm); Inclination angle of the i-th slope (°); S: Module center spacing (cm); g: Guide groove width (mm); e: Guide groove depth (cm); : Total height of the i-th module (cm); : Bracket height of the i-th module (cm).

[0034] For a fruit with mass m, when passing through a slope with an angle of θ, the force F measured by the sensor satisfies: F = m·g·cosθ. Therefore, by measuring the forces at different angles, a system of equations can be constructed: , , , and the weighted average method is used to solve for the fruit mass m, where the weights are normalized based on the cosθ values: , the module angle selection needs to satisfy that the cosθ values have significant differences; the sampling frequency f needs to satisfy, , where, is the contact time. In addition, the module spacing S > D (ensuring that the fruit plate completely leaves one slope before contacting the next one); W < D (preventing the fruit plate from directly contacting the slope); (ensuring the effective contact area); Module height difference compensation relationship: To ensure the smooth transition of the fruit plate, the total height of each module needs to satisfy, , where, is the total height of the i-th module, L = 3 cm is the slope length, . For example, . In actual design, , the compensation requirement is met by adjusting the bracket height.

[0035] In the overall solution, taking an apple as an example, the design parameters of the fruit tray can be used as a reference as follows: D = 9 cm (fruit tray diameter); H = 4 cm (fruit tray height); d = 6 cm (bottom hollow diameter); t = 0.4 cm (edge thickness); h = 1.2 cm (bottom edge height); a = 1.5 cm (height of fruit exposed); p = 0.5 cm (height of side bulge); w = 0.2 cm (width of side bulge).

[0036] For smaller fruits (such as strawberries), the parameters are adjusted to: D = 5 cm; d = 3 cm; a = 0.8 cm. In the overall solution, the design parameters of the slope module can be used as a reference as follows: L = 3 cm (slope length); W = 8 cm (slope width); = 15°, = 30°, = 35° (slope angle); S = 12 cm (module center spacing); g = 2 mm (width of guide groove); e = 1 cm (depth of guide groove); = 3 cm, = 3.5 cm, = 4 cm (total height of module); = 1.5 cm, = 0.77 cm, = 0.5 cm (height of bracket).

[0037] When the conveyor belt speed v = 1.5 m / s: The single-module contact time = L / v = 0.02 s; The sensor sampling rate of 10 kHz can obtain about 200 data points per pass; The difference in cosθ values: cos15° ≈ 0.966, cos30° ≈ 0.866, cos35° ≈ 0.819; The three-point measurement redundancy design reduces the influence of single-point error. Through parametric design, the device can flexibly adapt to different types and sizes of fruits, achieving high-precision (error < ±2 g), non-destructive, and high-speed (< 0.2 s / fruit) quality measurement, meeting the industrial fruit grading requirements.

[0038] In this embodiment, before measuring the normal force data set, through empty tray detection, the possibility of abnormal measurement results caused by no fruit placed on the fruit tray during the conveyor belt transportation, which in turn affects the measurement accuracy, is reduced.

[0039] In one embodiment, after collecting the normal force data set, it is necessary to compare whether the fruits in the fruit tray are misplaced. The specific detection operations include: Compare whether the mass of the target fruit corresponding to different-angle slopes is within the mass range corresponding to the preset target fruit type; if the mass of the target fruit corresponding to different-angle slopes is within the mass range corresponding to the preset target fruit type, continue to calculate the mass of the target fruit; if the mass of the target fruit corresponding to different-angle slopes is outside the mass range corresponding to the preset target fruit type, calculate the difference between the mass of the target fruit corresponding to different-angle slopes and the mass range corresponding to the preset target fruit type and set the difference as an outlier; query the abnormal operation corresponding to the outlier in the preset abnormal operation library, and different ranges of outliers and the abnormal operations corresponding to the outliers are stored in the abnormal operation library.

[0040] Exemplarily, the following are stored in the preset abnormal operation library: 1. When the outlier is less than 500 g, send an abnormal message; 2. When the outlier is greater than 500 g and less than 1000 g, send an abnormal alarm; when the outlier is greater than 1000 g, stop the conveyor belt; if the detected outlier is 600 g, send an abnormal alarm.

[0041] In this embodiment, a preliminary inspection is performed on the mass value of the target fruit on the fruit plate, so that during the mass measurement of the fruit, fruits with large mass differences can be identified, reducing the misplacement of other types of fruits on the fruit plate, resulting in abnormal mass calculation of the target fruit, and then affecting the final mass calculation of the target fruit.

[0042] In one embodiment, when the fruit plate passes by at a high speed (1.5 m / s), due to the inertial effect and short contact time, there will be a certain systematic deviation in the force value of static calibration, and a calibration operation needs to be performed on the systematic deviation, which specifically includes: Construct a system deviation compensation based on the speed of the target fruit, and the formula is as follows: ; where is the system deviation compensation, is the speed of the target fruit.

[0043] Exemplarily, when , through testing, the deviation of about 6% caused by the inertial effect can be corrected to within ±2 g, which is used to correct this systematic deviation.

[0044] First, the system compares the mass values (m_A, m_B, m_C) calculated through three different angle slopes (15°, 30°, 35°). If the standard deviation between these values is less than 2g, the measurement results are considered consistent and reasonable; otherwise, it is regarded as an abnormal measurement and re-detection is triggered. Secondly, the system checks whether the calculated mass conforms to the expected range of the fruit type. For example, the mass of an apple is usually between 100g - 300g. If the result far exceeds this range (such as 500g), it may be a measurement or classification error. In addition, the system verifies the stability of the sensor signal during the measurement process to ensure there are no abnormal fluctuations or drifts, such as signal instability caused by equipment failure or fruit tray deviation. For abnormal results, the system will perform repeated measurements to confirm whether it is a random error or a systematic problem. At the same time, the system will record the abnormal values and corresponding operations (such as sending an alarm or stopping the machine), and analyze its matching degree with the actual abnormal degree for subsequent optimization. In summary, through consistency, physical range, signal stability, and repeatability tests, the rationality of abnormal values and operations is ensured.

[0045] It is worth mentioning that the system will also select different compensation coefficients according to different fruit types and weight ranges, and continuously optimize these parameters through machine learning algorithms. The system then conducts strict data verification, calculates the standard deviation of three independent measurements, and checks whether the force value relationship between measurement points conforms to the theoretical expectation ( ), and determines whether the final mass is within the preset reasonable range. Any abnormality will be recorded and it will be decided whether intervention is needed according to the severity. After the data verification passes, the fruits are classified according to the preset quality grade standards (such as special grade > 250g, first grade 200 - 250g, second grade 150 - 200g, third grade < 150g). The classification results are transmitted to the sorting control system in real time through the industrial Ethernet. When the fourth photoelectric sensor (P4) confirms that the fruit tray has left the measurement area, the sorting mechanism controls the pneumatic diverter according to the quality grade to guide the fruit tray to the corresponding channel. Throughout the process, the fruit never leaves the fruit tray to ensure damage-free processing. The measurement results, timestamps, and classification results of each fruit tray are stored in the database for subsequent statistical analysis and quality traceability.

[0046] It is worth noting that the compensation coefficient is used to correct measurement deviations caused by fruit characteristics or high-speed movement (such as 1.5 m / s) and is closely related to fruit types and weight ranges. Different fruits (such as apples, strawberries, grapes) affect the normal force measurement due to different densities, shapes, and surface characteristics, so different compensation coefficients are required. For example, apples (with a density of about 0.85 g / cm³) may require a smaller compensation coefficient, while strawberries (with a soft surface) may require a larger coefficient to compensate for contact deformation. At the same time, fruits with a larger mass (such as 200 g - 300 g) may require a higher compensation coefficient due to more significant inertial effects, while lighter fruits (such as 50 g - 100 g) require a smaller coefficient. These corresponding relationships are established through experimental calibration: multiple measurements are taken for each fruit type and weight range, the deviation between the measured value and the true value is recorded, the compensation coefficient is fitted, and a corresponding table is generated. For example, the coefficient for the 100 g - 200 g range of apples is 1.05, and for the 200 g - 300 g range is 1.07. In addition, machine learning is introduced to dynamically optimize the compensation coefficient to improve measurement accuracy. The specific method is to use supervised learning, and the training data includes known true mass and measurement data (such as normal force, ambient temperature). The features cover fruit type, weight range, conveyor belt speed (1.5 m / s), temperature, humidity, etc., and the goal is to predict the optimal compensation coefficient K. The model can use linear regression, support vector machine regression, or neural network, and the system will regularly update the model to adapt to environmental changes or equipment aging using new data. For example, the system can adjust the compensation coefficient of apples on a 15° slope from the initial 1.05 to 1.06 based on historical data. In summary, the initial relationship of the compensation coefficient is established through experiments and statistics, and machine learning continuously optimizes it through supervised learning to enhance the system's adaptability.

[0047] During the system maintenance and optimization phase, the system is designed with a perfect self-calibration mechanism that automatically calibrates the zero point once an hour to eliminate environmental drift and automatically adjusts the dynamic compensation coefficient after processing 1000 fruits. The system also continuously monitors the operating status, including the consistency of the fruit tray passing time and the stability of the sensor signals. The HMI display screen real-time shows the sensor status, current measured value, and throughput, and the mass distribution histogram is dynamically updated for easy monitoring by the operator. All measurement data and abnormal events are completely recorded to support batch traceability and problem backtracking analysis. At the same time, the system provides an API interface for integration with the enterprise ERP system and packaging and labeling systems. Through this process design, the device can achieve non-destructive and high-precision mass measurement of fruits at a speed of 1.5 m / s, with a total measurement time of about 0.18 seconds, meeting the requirements of high-frequency assembly line operations, while maintaining the measurement accuracy within the range of ±2 g, significantly improving the efficiency and accuracy of fruit grading.

[0048] There are indeed dynamic measurement solutions in the prior art, but these solutions usually rely on high-performance and high-cost weighing sensor systems (for example, high-precision strain gauge sensors with a cost of over 10,000 yuan per channel are required for roller-type dynamic weighing, and the cost of laser volume estimation systems is even higher), and these systems still face huge challenges when applied to fruits. The core innovation of the present invention is to creatively solve the problem through structural design and physical principles, achieving high-precision measurement with ordinary sensors, and significantly reducing the technical threshold and implementation cost. The following are the three core points to be protected in the present invention, with particular emphasis on the innovative design for fruit characteristics: 1) Multi-angle ramp force measurement system for the irregular and easy-to-roll characteristics of the fruit surface The biggest characteristic of fruits different from industrial standard items is that their surfaces are highly irregular and easy to roll, which leads to large signal fluctuations and poor stability in measurement by traditional dynamic weighing systems. The multi-angle ramp system of the present invention (a specific angle combination of 15°, 30°, and 35°) specifically addresses this characteristic and creatively solves the rolling control problem of round fruits in high-speed motion through the "ramp - flat section" alternating structure. Compared with the weighing platform design that relies on the deceleration measurement of the fruit conveyor, the present invention can achieve continuous dynamic measurement without deceleration, ensuring that stable measurement data can be obtained even when the fruit surface is rough and irregular, and solving the core problem of "unstable contact" during dynamic measurement of fruits in the industry for a long time. Especially for round fruits such as apples and oranges, traditional flat weighing or roller weighing may cause a measurement error of ±20g due to rolling, while the present invention reduces the force fluctuation to ±2% through the ramp angle design, and controls the error within ±2g.

[0049] 2) Parameterized fruit tray and low-friction contact system The fruit tray structure of the present invention is parametrically designed, and the core parameters include: fruit tray diameter D, bottom hollow diameter d, fruit exposure height a, etc. Among them, the fruit exposure height a satisfies two constraint conditions: a ≤ L·tanθ_max and a ≤ 0.3D_f to ensure stable measurement of the fruit. The low-friction PTFE coating (μ < 0.1) combined with the hollow fruit tray design not only reduces the damage to the fruit skin but also ensures effective contact with the ramp, which is especially suitable for easily damaged fruits such as strawberries and blueberries. Compared with CN109807079B, which uses a hard fruit conveyor to directly contact the fruit and the damage rate may reach 5 - 10%, the present invention reduces the friction damage to less than 1% through the PTFE coating. The protected point lies in this parametric design method combined with fruit characteristics and the selection of low-friction materials, enabling the system to adapt to various fruits from small grapes to large pineapples, significantly improving the non-destructiveness and versatility.

[0050] 3) High-frequency sensor and dynamic compensation algorithm The present invention collects the peak normal force in real time through a high-frequency force sensor (sampling rate of 10 kHz), and calculates the mass by using a weighted average algorithm combined with a dynamic compensation coefficient. The key points of protection lie in the sensor integration and the data processing system: the sensor and the ramp are seamlessly connected through a shock pad to reduce vibration interference; the PLC controller and the synchronous acquisition card ensure the precise alignment of the data of the three groups of sensors, supporting the completion of the calculation within 0.1 second. Compared with the simple weight processing without dynamic compensation mentioned in CN109807079B, the present invention corrects about 6% of the inertial deviation under high-speed motion (1.5 m / s) through Kv, reducing the error from ±5 g to ±2 g. This design avoids the complex indirect estimation methods relying on shape and density in the prior art, and realizes the rapid, accurate and non-destructive measurement of fruits through direct measurement and physical law calculation, which is particularly suitable for various fruits with irregular surfaces and different shapes.

[0051] It is worth mentioning that for objects of different sizes and shapes, a replaceable fruit tray module and an adjustable ramp angle can be designed (such as designing square or oval brackets or stepwise adjustment of the ramp angle from 0° to 45°), and at the same time, the low-friction coating (μ < 0.1) is replaced with a ramp surface covered with a flexible material (such as a silicone pad) to increase buffering and reduce the contact pressure, so as to further improve the adaptability to different fruits (such as strawberries and watermelons) and the protection effect on extremely vulnerable fruits (such as blueberries) on the basis of directly measuring the normal force (FN = m·g·cosθ), while maintaining the high precision of ±2 g and the non-destructive characteristics. Through the software adaptive calculation formula, it is extended to high-speed assembly line scenarios such as express package sorting and pharmaceutical packaging inspection, adapting to diverse surfaces and maintaining an accuracy of ±2 g.

[0052] The present invention can be migrated to the rapid weighing of precision electronic components or the non-destructive sorting of biological samples in the medical field. The low-friction PTFE coating is used to avoid surface scratches, and with a response speed of 0.1 second, it meets the industrial and medical requirements of high cleanliness and high precision (±2 g).

[0053] This weighing module can be compatible and parallel with other modules. For example, combined with a volume measurement module, it can measure the density of fruits, or combined with other quality measurement modules.

[0054] Experiments were conducted on the present invention. Fruit tray structure parameters: Fruit tray diameter (D): 7 cm; Fruit tray height (H): 3 cm; Bottom hollow diameter (d): 5 cm; Fruit tray edge thickness (t): 0.5 cm; Bottom edge height (h): 0.8 cm; Fruit exposed height (a): 1.5 cm (meeting the constraint conditions of a ≤ L·tanθmax and a ≤ 0.3Df); Side positioning protrusion height (p): 0.6 cm; Side positioning protrusion width (w): 0.8 cm. Ramp module parameters: Ramp length (L): 3.3 cm × 3 segments, total length 10 cm; Ramp width (W): 10 cm; Inclination angles (θ) of the three ramps: 15°, 30°, and 35° respectively; Module center spacing (S): 4 cm; Guide groove width (g): 500 g; Guide groove depth (e): 0.5 cm. Measurement system parameters: Force sensor sampling rate: 10 kHz; Sensor sensitivity: 0.005 N; Conveyor belt speed: 1 m / s; Synchronous acquisition card time error: < 1 ms; PTFE coating friction coefficient: μ < 0.1.

[0055] Three typical fruits were selected for testing, with each fruit tested 30 times. The results are as follows: 1) Apple (diameter about 7 cm, weight about 150 g): Static weighing average: 150.2 g; Measurement average of the present invention: 149.8 g; Measurement error range: ±1.8 g; Single measurement time: 0.096 s; Epidermal damage rate: No obvious damage was observed; 2) Strawberry (weight about 25 g): Static weighing average: 25.1 g; Measurement average of the present invention: 24.9 g; Measurement error range: ±1.2 g; Single measurement time: 0.093 s; Epidermal damage rate: No damage was observed; 3) Kiwifruit (irregular shape, weight about 100 g): Static weighing average: 100.3 g; Measurement average of the present invention: 99.7 g; Measurement error range: ±2.0 g; Single measurement time: 0.102 s; Epidermal damage rate: No obvious damage was observed; In the high-speed continuous measurement experiment, the system can stably process 4 fruits per second at a conveyor belt speed of 1 m / s, and the measurement accuracy of the three groups of sensor data processed by the weighted average algorithm remains within the range of ±2 g.

[0056] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless specifically stated herein, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0057] Based on the same inventive concept, an embodiment of the present application further provides a dynamic weighing system for a fruit transportation line based on multi-angle slopes for implementing the above-mentioned dynamic weighing method for a fruit transportation line based on multi-angle slopes. The solution provided by this system for solving problems is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the dynamic weighing system for a fruit transportation line based on multi-angle slopes provided below can refer to the limitations for the dynamic weighing method for a fruit transportation line based on multi-angle slopes in the foregoing text, and will not be repeated here.

[0058] In one embodiment, as Figure 3 shown, a dynamic weighing system for a fruit transportation line based on multi-angle slopes is provided, including: an empty tray detection module, a normal force acquisition module, a mass calculation module, and a mass compensation module, where: The normal force acquisition module is configured to acquire a normal force data set of a target fruit passing through at least three different-angle slopes preset within a preset time and perform preprocessing; The mass calculation module is configured to calculate the average value of the preprocessed normal force data set and calculate the mass corresponding to the target fruit passing through different-angle slopes respectively based on the average value of the preprocessed normal force data set, a preset gravitational acceleration, and different angles; The mass compensation module is configured to calculate the mass corresponding to the target fruit passing through different-angle slopes based on the weighted average method and obtain the mass value corresponding to the target fruit after compensating the mass based on a preset high-speed motion correction factor.

[0059] In one embodiment, the empty tray detection module is further configured to: load a preset fruit tray parameter set based on the target fruit type; calculate the height of the target fruit on the fruit tray when detecting the fruit tray; and send a fruit tray recycling signal when the height of the target fruit is inconsistent with the preset height corresponding to the target fruit in the preset fruit tray parameter set.

[0060] In one embodiment, the mass calculation module is further configured to: calculate the mass of the target fruit passing through slopes at different angles according to the following formula: ; where is the mass of the target fruit calculated through the slope at angle, is the normal force detected when the target fruit passes through the slope at angle, is the acceleration due to gravity, is the preset slope angle; compare whether the mass of the target fruit passing through slopes at different angles is within the mass range corresponding to the preset target fruit type; if the mass of the target fruit passing through slopes at different angles is outside the mass range corresponding to the preset target fruit type, calculate the difference between the mass of the target fruit passing through slopes at different angles and the mass range corresponding to the preset target fruit type and set the difference as an outlier; query the abnormal operation corresponding to the outlier in the preset abnormal operation library, and different ranges of outliers and the abnormal operations corresponding to the outliers are stored in the abnormal operation library.

[0061] In one embodiment, the mass compensation module is further configured to: construct a system of equations for normal forces at different angles; calculate the mass of the target fruit passing through slopes at different angles based on the weighted average method according to the following formula: ; where is the mass of the target fruit calculated through the slope at angle, is the normal force detected when the target fruit passes through the slope at angle, is the normal force detected when the target fruit passes through the slope at angle, is the normal force detected when the target fruit passes through the slope at angle, is the acceleration due to gravity, , and are the preset slope angles; compensate the mass calculated by the weighted average method based on the preset dynamic compensation coefficient.

[0062] In one embodiment, the mass compensation module is further configured to: construct a system deviation compensation based on the speed of the target fruit according to the following formula: ; where is the system deviation compensation, is the speed of the target fruit.

[0063] In one embodiment, the quality compensation module is further configured to: apply a corresponding compensation coefficient from a preset compensation coefficient library based on the fruit type and weight range, where the compensation database stores fruit types, weight ranges, and corresponding compensation coefficients; dynamically optimize the compensation coefficient based on supervised learning and generate a correspondence table of fruit types, weight ranges, and compensation coefficients; and update the compensation coefficient based on historical data.

[0064] Each module in the above-mentioned dynamic weighing system for fruit transportation line based on multi-angle slopes can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above modules can be embedded in the processor in the computer system in hardware form or be independent of it, or can be stored in the memory in the computer system in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0065] In one embodiment, a computer system is provided. The computer system can be a server, and its internal structure diagram can be as Figure 4 shown. The computer system includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer system is used to provide computing and control capabilities. The memory of the computer system includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer system is used to store data. The network interface of the computer system is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a dynamic weighing method for a fruit transportation line based on multi-angle slopes.

[0066] In one embodiment, a computer system is provided. The computer system can be a terminal, and its internal structure diagram can be as Figure 4 shown. The computer system includes a processor, a memory, a communication interface, a display screen, and an input system connected through a system bus. Among them, the processor of the computer system is used to provide computing and control capabilities. The memory of the computer system includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer system is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a dynamic weighing method for a fruit transportation line based on multi-angle slopes.

[0067] Those skilled in the art can understand that Figure 4The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer system to which the solution of this application is applied. The specific computer system may include more or fewer components than those shown in the figure, or combine some components, or have a different component arrangement.

[0068] In one embodiment, a computer system is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0069] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0070] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0071] It should be noted that the user information (including but not limited to user system information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0072] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAM), magnetoresistive random access memories (MRAM), ferroelectric random access memories (FRAM), phase change memories (PCM), graphene memories, etc. Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0073] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0074] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A dynamic weighing method for a fruit transport line based on a multi-angle slope, characterized in that: The method comprises: Collecting and preprocessing a normal force data set of a target fruit passing through at least three preset slopes of different angles within a preset time; Calculate the average value of the preprocessed normal force data set and calculate the mass of the target fruit passing through the slopes at different angles based on the average value of the preprocessed normal force data set, the preset gravity acceleration and different angles; The mass of the target fruit passing through slopes of different angles is calculated based on the weighted average method, and the mass value corresponding to the target fruit is obtained after the mass is compensated based on a preset high-speed motion correction factor.

2. The dynamic weighing method for a fruit transport line based on a multi-angle slope according to claim 1, characterized in that: The method further comprises: Load a preset set of fruit plate parameters based on the target fruit type; When a fruit tray is detected, the height of the target fruit on the fruit tray is calculated; When the target fruit is inconsistent with the preset height corresponding to the target fruit in the preset fruit plate parameter set, a fruit plate recovery signal is sent.

3. The dynamic weighing method for a fruit transport line based on a multi-angle slope according to claim 2, characterized in that: The method of calculating the mass of the target fruit passing through slopes of different angles based on the average value of the preprocessed normal force data set, the preset gravity acceleration and different angles includes: The formula for calculating the mass of the target fruit passing through slopes of different angles is as follows: ； in, To pass The slope of the angle is calculated to determine the mass of the target fruit. For the target fruit by The normal force obtained by the slope detection of the angle, is the acceleration due to gravity, is the preset slope angle; Compare whether the mass of the target fruit passing through the slopes at different angles is within the preset mass range corresponding to the target fruit type; If the mass of the target fruit passing through the slopes at different angles is outside the preset mass range corresponding to the target fruit type, the difference between the mass of the target fruit passing through the slopes at different angles and the preset mass range corresponding to the target fruit type is calculated and the difference is set as an abnormal value; An abnormal operation corresponding to the abnormal value is searched in a preset abnormal operation library, wherein the abnormal operation library stores abnormal values ​​of different ranges and abnormal operations corresponding to the abnormal values.

4. The dynamic weighing method for a fruit transport line based on a multi-angle slope according to claim 1 is characterized in that: The method of calculating the mass of the target fruit corresponding to the slopes at different angles based on the weighted average method includes: Construct the normal force equations for different angles; The formula for calculating the mass of the target fruit passing through slopes of different angles based on the weighted average method is as follows: ； in, To pass The mass of the target fruit is calculated by the slope of the angle, For the target fruit by The normal force obtained by the slope detection of the angle, For the target fruit by The normal force obtained by the slope detection of the angle, For the target fruit by The normal force obtained by the slope detection of the angle, is the acceleration due to gravity, , and is the preset slope angle; The quality calculated by the weighted average method is compensated based on the preset dynamic compensation coefficient.

5. The dynamic weighing method for a fruit transport line based on a multi-angle slope according to claim 4 is characterized in that: The compensating the quality calculated by the weighted average method based on the preset dynamic compensation coefficient includes: The system deviation compensation is constructed based on the speed of the target fruit, and the formula is as follows: ； in, To compensate for system deviation, is the speed of the target fruit.

6. The dynamic weighing method for a fruit transport line based on a multi-angle slope according to claim 5 is characterized in that: The method further comprises: Applying corresponding compensation coefficients in a preset compensation coefficient library based on the fruit type and weight range, wherein the compensation coefficient library stores fruit types, weight ranges and corresponding compensation coefficients; Dynamically optimize the compensation coefficient based on supervised learning and generate a corresponding table of fruit type, weight range and compensation coefficient; The compensation coefficient is updated based on historical data.

7. A dynamic weighing system for a fruit transport line based on a multi-angle slope, characterized in that: The system comprises: A normal force acquisition module is used to collect and pre-process the normal force data set of the target fruit passing through at least three preset slopes of different angles within a preset time; A mass calculation module, used to calculate the average value of the preprocessed normal force data set and calculate the mass of the target fruit passing through slopes of different angles based on the average value of the preprocessed normal force data set, a preset gravity acceleration and different angles; The mass compensation module is used to calculate the mass of the target fruit passing through slopes of different angles based on the weighted average method and to obtain the mass value corresponding to the target fruit after compensating the mass based on a preset high-speed motion correction factor.

8. A computer system comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images