A material flow detection and quantitative dispensing device and method

Abstract

This invention discloses a material flow detection and quantitative distribution device and method. The device includes a non-contact scanning detection module, a data processing module, a control module, and an actuator. The non-contact scanning detection module is installed above the conveyor belt (1) and is used to perform full-section scanning of the material on the conveyor belt (1) to obtain the spatial distribution data of the material. The data processing module is connected to the non-contact scanning detection module and is used to calculate the real-time flow value of the material based on the spatial distribution data. The control module is connected to the data processing module and is used to generate a speed adjustment command based on the deviation between the real-time flow value and the preset target flow value. This invention solves the technical problems in the prior art, such as easy wear of contact detection elements, low detection accuracy, inability of non-contact single-point ranging scheme to perform full-section real-time modeling of high-speed flowing materials, and inability to dynamically adjust the quantitative distribution of materials according to the real-time flow status of the material.

Description

Technical Field

[0001] This invention belongs to the field of bulk material flow detection and quantitative distribution technology, specifically relating to a material flow detection and quantitative distribution device and method. Background Technology

[0002] Traditional material flow detection methods often employ contact-based approaches such as mechanical weighing, photoelectric sensors, or pressure sensors. These methods suffer from low accuracy, susceptibility to environmental interference, and the inability to monitor the full cross-sectional distribution of materials in real time. Contact-based detection elements experience severe wear due to prolonged friction with the material, resulting in high maintenance costs. Furthermore, their signal stability is poor in high-humidity, dusty industrial environments, making it difficult to guarantee reliable detection during continuous production. While existing non-contact ranging solutions can avoid contact wear, they are primarily suitable for single-point distance detection of static objects and cannot perform real-time full-cross-sectional modeling of high-speed flowing materials, thus failing to meet the demands of industrial production lines for continuous, high-precision material flow detection.

[0003] Industrial production lines commonly use electronic belt scales for real-time weighing. The core of this method is to collect material weight signals through weighing sensors and combine this with a control system to achieve dynamic flow monitoring. The stability of this method is highly dependent on routine maintenance conditions such as conveyor belt cleanliness and bearing lubrication. It is easily affected by environmental factors such as conveyor belt misalignment and vibration, leading to baseline drift. When the material supply is uneven, error accumulation is significant, requiring frequent manual zeroing, making it difficult to meet the requirements of high-precision continuous production.

[0004] The quantitative distribution of materials on each conveyor belt generally relies on manual experience for adjustment or a fixed proportion of mechanical diversion structure. It cannot be dynamically adjusted according to the real-time flow status, which can easily cause material accumulation or uneven distribution, affecting the processing quality of subsequent processes and the overall efficiency of the system.

[0005] In summary, existing technologies have significant shortcomings in terms of material flow detection accuracy, non-contact capability, and dynamic response capability for quantitative distribution. There is an urgent need for a non-contact, high-precision, and dynamically adjustable material flow detection and quantitative distribution technology. Summary of the Invention

[0006] The purpose of this invention is to provide a material flow detection and quantitative distribution device and method to solve the technical problems in the prior art, such as easy wear of contact detection elements, low detection accuracy, inability of non-contact single-point ranging schemes to perform real-time full-section modeling of high-speed flowing materials, and inability to dynamically adjust material quantitative distribution according to the real-time flow status of materials.

[0007] To achieve the above objectives, the present invention provides a material flow detection and quantitative distribution device, comprising a non-contact scanning detection module, a data processing module, a control module, and an actuator. The non-contact scanning detection module is installed above the conveyor belt and is used to perform full-section scanning of the material on the conveyor belt to acquire spatial distribution data of the material. The data processing module is connected to the non-contact scanning detection module and is used to calculate the real-time flow value of the material based on the spatial distribution data. The control module is connected to the data processing module and is used to generate a speed adjustment command based on the deviation between the real-time flow value and a preset target flow value. The actuator is connected to the control module and is used to receive the speed adjustment command and drive the conveyor belt to operate according to the speed adjustment command.

[0008] Furthermore, the non-contact scanning detection module includes:

[0009] At least two lidars are symmetrically installed on both sides above the conveyor belt, with their scanning range covering the full width of the conveyor belt, for real-time acquisition of material surface scanning contour data;

[0010] The motion compensation unit is used to acquire the displacement of the conveyor belt in real time and perform coordinate transformation compensation on the scanned contour data.

[0011] The data fusion unit is used to perform spatial registration and fusion on the scanned contour data after coordinate transformation compensation to generate a two-dimensional spatial distribution model of the material layer, and output it to the data processing module.

[0012] Furthermore, the motion compensation unit obtains the displacement of the conveyor belt through a displacement sensor installed in the conveyor belt drive system.

[0013] Furthermore, the data processing module includes:

[0014] The data preprocessing unit uses a statistical outlier removal algorithm to filter noise in the spatial distribution data, separates the material area from the conveyor belt background through dynamic height threshold segmentation, and uses Euclidean clustering algorithm to extract effective material accumulation contour segments.

[0015] The flow calculation unit reconstructs the material layer cross-sectional geometric model based on the effective material accumulation contour segment, and calculates the material cross-sectional area by accumulating the areas of each micro-element of the material layer cross-sectional geometric model. And combined with the linear speed of the conveyor belt (1) obtained by the conveyor belt (1) drive system and scan time interval According to the formula: Calculate the volume of the material ;

[0016] The density correction unit includes a humidity sensor, used to correct the material density value in real time based on the measured moisture content of the material collected by the humidity sensor, using a dynamic density model. The formula for the dynamic density model is:

[0017] ;

[0018] in, This is the corrected material density value. For the material's reference density, This is the humidity correction factor. The measured value of the moisture content of the material;

[0019] The corrected material density value With respect to the volume of the material The fusion calculation outputs a mass flow rate value as the real-time flow rate value to the control module.

[0020] Furthermore, the data processing module also includes a trend prediction unit, which has a built-in traffic trend prediction model for predicting the changing trend of the real-time traffic value based on the historical sequence of the real-time traffic value, and outputting an early warning signal to the control module in advance.

[0021] Furthermore, the control module includes:

[0022] A fuzzy PID controller is used to receive the deviation between the real-time flow value and the preset target flow value, quantify the deviation through a membership function, and generate a speed adjustment amount using a weighted defuzzification rule.

[0023] The feedforward compensation unit has a built-in feedforward compensation prediction model, which is used to train the feedforward compensation prediction model based on the historical sequence of the real-time flow value, output the feedforward compensation coefficient in advance, and superimpose the feedforward compensation coefficient onto the speed adjustment amount to generate the speed adjustment command.

[0024] Furthermore, the actuator is connected to the drive systems of at least two conveyor belts via a high-speed communication bus, and outputs the speed adjustment command independently to each conveyor belt.

[0025] Furthermore, it also includes an auxiliary module, which includes:

[0026] The multi-sensor verification unit includes a weighing platform located at the end of the conveyor belt, which is used to periodically collect the cumulative weight of the material, compare it with the flow integral value calculated based on the real-time flow value, generate an error correction coefficient, and iteratively update the calculation parameters of the real-time flow value in the data processing module using the error correction coefficient.

[0027] The safety protection unit is used to monitor the motor current of the conveyor belt drive system and output protection commands in stages when the limit is exceeded, and to continuously monitor the spatial distribution data and automatically issue a stop command to the conveyor belt drive system when the continuity is interrupted.

[0028] The monitoring and display unit is connected to the data processing module and is used to display the heat map distribution of material flow and the operating status of equipment in real time.

[0029] In another aspect, the present invention provides a method for material flow detection and quantitative distribution, characterized by comprising the following steps:

[0030] Step S1: The non-contact scanning detection module performs a full-section scan of the material on the conveyor belt to obtain the spatial distribution data of the material.

[0031] Step S2: The data processing module preprocesses and calculates the flow rate of the spatial distribution data, and outputs the real-time flow rate value of the material after density correction.

[0032] Step S3: The control module compares the real-time flow value with the preset target flow value, generates a speed adjustment amount, and generates a speed adjustment command by superimposing the feedforward compensation coefficient onto the speed adjustment amount through the feedforward compensation prediction model.

[0033] Step S4: The actuator drives the conveyor belt to run according to the speed adjustment command;

[0034] Step S5: The auxiliary module periodically calibrates the calculation parameters of the real-time flow value in the data processing module, monitors the operating status of the conveyor belt drive system and outputs protection commands when abnormal, and displays the heat map distribution of material flow and equipment operating status in real time.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] First, this invention constructs a complete closed-loop system consisting of a non-contact scanning detection module, a data processing module, a control module, and an actuator. It acquires the spatial distribution data of materials through non-contact scanning, calculates the real-time flow rate of materials in real time, and generates speed adjustment commands to drive the conveyor belt. This fundamentally overcomes the defects of traditional contact detection elements that are prone to wear and tear, and also solves the technical problem that existing material distribution relies on manual experience and cannot be dynamically adjusted according to the real-time flow status.

[0037] Secondly, this invention employs at least two lidar units symmetrically installed on both sides above the conveyor belt, with their scanning range covering the entire width of the conveyor belt. Combined with coordinate transformation compensation, it eliminates the distortion of the scanning contour data caused by the movement and vibration of the conveyor belt, effectively solving the technical problem of real-time modeling of the entire cross-section of high-speed flowing materials.

[0038] Third, this invention uses a segmented trapezoidal accumulation method to reconstruct the geometric model of the material layer cross-section and calculate the material cross-sectional area. Furthermore, a dynamic density model is introduced to correct the material density value in real time based on the measured value of the material moisture content, and the mass flow rate value is output as the real-time flow rate value, so that the detection accuracy can still be maintained under the working conditions of dynamic changes in the material state.

[0039] Fourth, the trend prediction unit of this invention outputs an early warning signal to the control module based on the historical sequence of real-time flow values. The feedforward compensation prediction model outputs the feedforward compensation coefficient 5 control cycles in advance and superimposes it on the speed adjustment amount, which effectively suppresses the overshoot caused by system inertia and improves the response speed and stability of the speed adjustment command.

[0040] Fifth, the actuator of this invention is connected to the drive systems of at least two conveyor belts via a high-speed communication bus, and outputs speed adjustment commands to each conveyor belt independently, so as to realize the dynamic quantitative distribution of materials on multi-directional conveyor belts and adapt to multi-station parallel feeding scenarios.

[0041] Sixth, the auxiliary module of this invention integrates multi-sensor periodic calibration, three-level linkage safety protection, and real-time display of material cross-sectional distribution data heat map. The system has self-calibration, self-protection, and visual monitoring capabilities, which greatly reduces the difficulty of operation and maintenance and ensures the long-term reliability of equipment operation. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the system installation structure of the material flow detection and quantitative distribution device in an embodiment of the present invention.

[0043] Figure 2 This is a schematic diagram illustrating the principle of a lidar scanning a full cross-section of materials on a conveyor belt in an embodiment of the present invention.

[0044] Figure 3 This is a schematic diagram of the geometric calculation of the trapezoidal micro-element of the material cross-sectional area in an embodiment of the present invention.

[0045] In the picture:

[0046] 1—Conveyor belt; 2—LiDAR. Detailed Implementation

[0047] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following embodiments use tobacco shreds in a tobacco processing workshop as an example for illustration, but the present invention is also applicable to flow detection and quantitative distribution scenarios for various bulk materials such as granules, powders, and fibers.

[0048] refer to Figure 1 , Figure 2 , Figure 3The material flow detection and quantitative distribution device provided in this embodiment of the invention mainly consists of five parts: a non-contact scanning detection module, a data processing module, a control module, an actuator, and an auxiliary module.

[0049] The non-contact scanning detection module is installed above the conveyor belt 1 to perform full-section scanning of the material on the conveyor belt 1 and obtain the spatial distribution data of the material. The non-contact scanning detection module mainly includes at least two lidars 2, a motion compensation unit and a data fusion unit. At least two lidars 2 are symmetrically installed on both sides above the conveyor belt 1, and their scanning range covers the full width of the conveyor belt 1 to collect the surface scanning contour data of the material in real time.

[0050] Specifically, in this embodiment, two industrial-grade LiDARs 2 are symmetrically mounted on supports on both sides above a bidirectional conveyor belt 1, at a height of 0.75 meters above the conveyor belt 1. After the material falls into the center of the conveyor belt 1 from above, the material is directionally transported in direction 1 (left) or direction 2 (right) by controlling the running direction of the conveyor belt 1. The combined scanning range of the two LiDARs 2 covers the entire width of the conveyor belt 1 (in this embodiment, the full width of the conveyor belt 1 is 1.2m). At the same time, the height information of the material at each measurement point is obtained by measuring the distance on the material surface at each measurement point, thereby realizing a full cross-sectional scan of the material cross-section in both width and height directions on the conveyor belt 1, providing complete cross-sectional data for the subsequent generation of a two-dimensional spatial distribution model of the material layer.

[0051] The motion compensation unit is used to acquire the displacement of conveyor belt 1 in real time and perform coordinate transformation compensation on the scanned contour data.

[0052] Specifically, in this embodiment, the conveyor belt 1 experiences continuous displacement during operation (each scanning time interval) The continuous forward movement of the inner conveyor belt 1, coupled with mechanical vibration, causes displacement distortion in the scan profile data acquired by the lidar 2, affecting the accuracy of cross-sectional area calculation. The non-contact scanning detection module is equipped with a motion compensation unit. This unit uses displacement sensors installed in the conveyor belt 1 drive system to acquire the actual displacement of the conveyor belt 1 in real time during each scanning cycle. It then performs coordinate transformation compensation on the scan profile data, mapping the coordinates of each scanning point to the same reference position. This eliminates the distortion in the scan profile data caused by the combined movement and vibration of the conveyor belt 1, improving the spatial accuracy of the scan profile data.

[0053] The data fusion unit is used to spatially register and fuse the scanned contour data after coordinate transformation compensation to generate a two-dimensional spatial distribution model of the material layer, and output it to the data processing module.

[0054] Specifically, in this embodiment, the data fusion unit performs spatial registration and fusion on the scan contour data of the two industrial-grade LiDARs 2 after coordinate transformation compensation, and generates a two-dimensional spatial distribution model of the material layer with a resolution of not less than 2mm×2mm. The two-dimensional spatial distribution model of the material layer is then output to the data processing module as the basic data for the data processing module to calculate the real-time flow value of the material.

[0055] The data processing module, connected to the non-contact scanning detection module, is used to calculate the real-time flow rate of materials based on spatial distribution data. The data processing module mainly includes a data preprocessing unit, a flow rate calculation unit, a density correction unit, and a trend prediction unit.

[0056] The data preprocessing unit uses a statistical outlier removal algorithm to filter noise in the spatially distributed data, separates the material area from the background of conveyor belt 1 through dynamic height threshold segmentation, and uses Euclidean clustering algorithm to extract effective material accumulation contour segments.

[0057] Specifically, in this embodiment, the data preprocessing unit performs the following processing on the spatially distributed data received by the data processing module in sequence:

[0058] ① Noise Filtering: The Statistical Outlier Removal (SOR) algorithm is used to filter the scan profile data. The SOR algorithm calculates the average distance between each data point in the scan profile data and its m neighboring data points. Data points whose average distance exceeds a certain multiple of the global mean standard deviation are identified as outliers and removed. In this embodiment, the number of neighboring points m = 50 and the standard deviation multiple is 1.5. Scan profile data points generated by environmental dust are sparse and scattered, with large average neighbor distances, and can be identified as outliers and removed, while scan profile data points generated by material accumulation are dense and continuous, with small average neighbor distances, and can be completely retained.

[0059] ② Background Separation: Based on the ranging results of two industrial-grade LiDARs 2, the height value of each scanning point relative to the reference surface of the conveyor belt 1 is calculated, and dynamic height threshold segmentation is used to separate the material area from the background of the conveyor belt 1. The height of the reference surface of the conveyor belt 1 is dynamically corrected to compensate for the height fluctuation of the conveyor belt 1 during operation. In this embodiment, data points with a height value lower than 10mm correspond to the surface of the conveyor belt 1 and the area where a thin layer of dust adheres, and are classified as background and removed; data points with a height value higher than 100mm are regarded as abnormal points such as structural component obstruction and are filtered out; data points with a height value in the range of 10mm to 100mm are effective material scanning contour data points and are retained.

[0060] ③ Material Region Clustering: The material scan contour data after background separation is processed using the Euclidean clustering algorithm. The Euclidean clustering algorithm groups adjacent points with a spatial distance less than a set threshold into the same cluster: the material accumulation area scan contour data is dense and continuous, with the distance between adjacent points usually not exceeding 5mm, forming large continuous scan contour data groups; while isolated noise points remaining after the first two steps are removed because they lack surrounding neighboring points and cannot meet the clustering conditions. In this embodiment, the distance threshold is set to 5mm to extract effective material accumulation contour segments for subsequent material cross-sectional area analysis. Calculation;

[0061] The flow calculation unit reconstructs the geometric model of the material layer cross-section based on the effective material accumulation profile segment, and calculates the material cross-sectional area by accumulating the areas of each micro-element of the material layer cross-sectional geometric model. And combined with the linear speed of conveyor belt 1 obtained by the drive system of conveyor belt 1 and scan time interval According to the formula: Calculate the volume of the material ;

[0062] Specifically, in this embodiment, the cross-sectional area The calculation uses a step-by-step trapezoidal summation method, and the specific calculation formula is as follows:

[0063]

[0064]

[0065]

[0066]

[0067]

[0068] in, The vertical height of the internal pole of lidar 2 from the bottom surface of the conveyor (a fixed value, determined by the installation position). This is the current radar ranging value. The current radar scanning angle, This will be the next radar ranging value. For the next radar scan angle, Let the cross-sectional area be a single trapezoidal element. Take the absolute value to ensure that the area of ​​the infinitesimal element is always positive. This represents the final accumulated cross-sectional area of ​​the material.

[0069] Obtain the cross-sectional area of ​​the material Then, combined with the linear speed of conveyor belt 1 obtained by the drive system of conveyor belt 1. and scan time interval The material volume is calculated using the following formula. :

[0070]

[0071] The density correction unit includes a humidity sensor, which is used to correct the material density value in real time based on the measured moisture content of the material collected by the humidity sensor and a dynamic density model.

[0072] Specifically, in this embodiment, since the material density varies with the moisture content, fixing the material density parameter would introduce systematic errors. Therefore, the density correction unit uses a dynamic density model to correct the material density value in real time. The formula for the dynamic density model is:

[0073] ;

[0074] in, This is the corrected material density value. For the material's reference density, This is the humidity correction factor. This represents the measured moisture content of the material.

[0075] The corrected material density value With material volume The fusion calculation outputs the mass flow rate value as a real-time flow rate value to the control module. The mass flow rate value formula is: ;

[0076] The trend prediction unit has a built-in traffic trend prediction model, which is used to predict the changing trend of real-time traffic values ​​based on the historical sequence of real-time traffic values, and output early warning signals to the control module in advance.

[0077] Specifically, in this embodiment, the traffic trend prediction model is implemented using an LSTM (Long Short-Term Memory) neural network. It takes a time series of real-time traffic values ​​from 10 consecutive scan cycles as input, and the network's hidden layer contains 32 neurons. This traffic trend prediction model learns the temporal patterns in historical real-time traffic value sequences to predict the trend of real-time traffic value changes in the next 3-5 seconds, and outputs an early warning signal to the control module in advance, enabling the control module to trigger feedforward compensation ahead of time.

[0078] The control module, connected to the data processing module, is used to generate speed adjustment commands based on the deviation between the real-time flow value and the preset target flow value. The control module mainly includes a fuzzy PID controller and a feedforward compensation unit.

[0079] The fuzzy PID controller receives the deviation between the real-time flow rate and the preset target flow rate, quantifies the deviation using a membership function, and generates the speed adjustment amount using weighted defuzzification rules. The fuzzy PID controller calculates the deviation in real time. ;in The real-time traffic value output by the data processing module. The preset target flow rate value.

[0080] The deviation interval [-5%, +5%] was divided into 7 fuzzy subsets. The membership function of each fuzzy subset adopted a triangular function, and overlapping intervals were set between adjacent subsets to ensure a smooth transition in the mapping. For the deviation... After performing fuzzy inference, a weighted inference method is used to defuzzify the data and output a precise speed adjustment amount. ,

[0081] The feedforward compensation unit has a built-in feedforward compensation prediction model, which is used to train the feedforward compensation prediction model based on the historical sequence of real-time flow values, output the feedforward compensation coefficients in advance, and superimpose the feedforward compensation coefficients onto the speed adjustment amount to generate speed adjustment commands.

[0082] Specifically, in this embodiment, the feedforward compensation prediction model uses historical sequences of real-time flow values ​​as training data to predict future trends in real-time flow values ​​and establishes a mapping relationship between this trend and the time lag between the issuance of the speed regulation command and the material flow response. It outputs a feedforward compensation coefficient five control cycles in advance and superimposes this coefficient onto the speed regulation amount. The system generates a final speed adjustment command, effectively suppressing overshoot caused by system inertia. The control module transmits the speed adjustment command to the actuator via a high-speed communication bus with a control cycle of no more than 100ms.

[0083] The actuator, connected to the control module, is used to receive speed adjustment commands and drive the conveyor belt 1 to run according to the speed adjustment commands.

[0084] Specifically, in this embodiment, a bidirectional conveyor belt 1 is used as an example for explanation. In other embodiments, the actuator can also be connected to the drive systems of at least two independent conveyor belts 1 simultaneously via a high-speed communication bus, and output speed adjustment commands independently to each conveyor belt 1 to adapt to multi-station parallel feeding application scenarios.

[0085] In this embodiment, conveyor belt 1 is driven by a closed-loop vector control variable frequency speed-regulating motor (rated power 7.5kW). A 17-bit absolute encoder is integrated at the motor shaft end to provide real-time speed feedback, forming a closed-loop speed control circuit. After receiving the speed adjustment command, the actuator plans the speed transition trajectory according to an S-shaped acceleration curve. The speed response time is no greater than 50ms, and the steady-state speed fluctuation rate is no greater than 0.5%. The speed adjustment range of conveyor belt 1 is 0.1–2.5m / s. The S-shaped curve planning effectively avoids material accumulation or discontinuity caused by rapid acceleration or deceleration.

[0086] The auxiliary module includes a multi-sensor verification unit, a safety protection unit, and a monitoring and display unit.

[0087] The multi-sensor calibration unit includes a weighing platform located at the end of conveyor belt 1, used to periodically collect the cumulative weight of the material. The error correction coefficient is compared with the traffic integral value calculated based on the real-time traffic value to generate an error correction coefficient, and the error correction coefficient is used to iteratively update the calculation parameters of the real-time traffic value in the data processing module.

[0088] Specifically, in this embodiment, a high-precision weighing platform is set at the end of the conveyor belt 1 in the multi-sensor verification unit, and the cumulative weight of the material in each batch is collected after each batch is completed. The total weight of the materials With traffic integral value Compare and generate error correction coefficients. The error correction coefficient is then used to iteratively update the calculation parameters of the real-time flow value in the data processing module (i.e., the material baseline density in the dynamic density model). The updated formula is as follows This enables periodic self-calibration of traffic data.

[0089] The safety protection unit is used to monitor the motor current of the conveyor belt 1 drive system and output protection commands in stages when the limit is exceeded, as well as to continuously monitor the spatial distribution data and automatically issue a stop command to the conveyor belt 1 drive system when the continuity is interrupted.

[0090] Specifically, in this embodiment, the safety protection unit implements a three-level linkage response mechanism by monitoring the motor current of the conveyor belt 1 drive system:

[0091] Level 1: When the motor current exceeds the rated value by 10%, an audible and visual alarm is triggered and the abnormal waveform is recorded;

[0092] Second stage: When the motor current exceeds the rated value by 20%, a linear speed reduction command is sent to the drive system of conveyor belt 1;

[0093] Level 3: When the motor current exceeds the rated value by 30% and lasts for more than 1 second, an emergency braking command is issued, and the braking time is no more than 100ms.

[0094] Meanwhile, the safety protection unit continuously monitors the spatial distribution data. When the continuity of the spatial distribution data is interrupted for more than 30 seconds, it automatically sends a stop command to the conveyor belt 1 drive system to achieve material interruption protection.

[0095] The monitoring and display unit, connected to the data processing module, is used to display the heat map distribution of material flow and the operating status of equipment in real time.

[0096] Specifically, in this embodiment, the monitoring and display unit visualizes the material cross-sectional distribution data output by the data processing module in the form of a heat map in real time, showing the uniformity of material distribution in the width direction of conveyor belt 1, and simultaneously displays the operating status, alarm information and historical trends of each device, so that operators can keep abreast of the system operation in real time.

[0097] This invention also provides a method for material flow detection and quantitative distribution based on the above-mentioned device, comprising the following steps:

[0098] Step S1: The non-contact scanning detection module performs a full-section scan of the material on conveyor belt 1, and after motion compensation processing and data fusion, obtains the spatial distribution data of the material;

[0099] Step S2: The data processing module sequentially performs noise filtering, background separation, and material region clustering on the spatially distributed data, reconstructs the geometric model of the material layer cross-section, and calculates the material cross-sectional area by accumulating the areas of each micro-element of the geometric model of the material layer cross-section. And according to the formula Calculate material volume ;

[0100] Then through dynamic density model ;

[0101] Perform density correction and output the mass flow rate value as the real-time flow rate value of the material;

[0102] Step S3: The control module compares the real-time flow value with the preset target flow value, generates a speed adjustment amount through a fuzzy PID controller, and generates a speed adjustment command by superimposing the feedforward compensation coefficient onto the speed adjustment amount through a feedforward compensation prediction model.

[0103] Step S4: The actuator drives the conveyor belt 1 to run according to the speed adjustment command;

[0104] Step S5: The auxiliary module periodically calibrates the calculation parameters of the real-time flow value in the data processing module, monitors the operating status of the conveyor belt 1 drive system and outputs protection commands when abnormal, and displays the heat map distribution of material flow and equipment operating status in real time.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the embodiments of the present invention have been described in detail, those skilled in the art should understand that modifications or equivalent substitutions to the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of protection of the claims of the present invention.

Claims

1. A material flow detection and quantitative distribution device, characterized in that, include: The non-contact scanning detection module is installed above the conveyor belt (1) and is used to scan the material on the conveyor belt (1) across the entire cross section to obtain the spatial distribution data of the material. A data processing module, connected to the non-contact scanning and detection module, is used to calculate the real-time flow value of the material based on the spatial distribution data; The control module, connected to the data processing module, is used to generate a speed adjustment command based on the deviation between the real-time flow value and the preset target flow value. An actuator, connected to the control module, is used to receive the speed adjustment command and drive the conveyor belt (1) to run according to the speed adjustment command.

2. The material flow detection and quantitative distribution device according to claim 1, characterized in that, The non-contact scanning detection module includes: At least two lidars (2) are symmetrically installed on both sides above the conveyor belt (1), and their scanning range covers the full width of the conveyor belt (1) for real-time acquisition of material surface scanning contour data; The motion compensation unit is used to acquire the displacement of the conveyor belt (1) in real time and perform coordinate transformation compensation on the scanned contour data; The data fusion unit is used to perform spatial registration and fusion on the scanned contour data after coordinate transformation compensation to generate a two-dimensional spatial distribution model of the material layer, and output it to the data processing module.

3. The material flow detection and quantitative distribution device according to claim 2, characterized in that, The motion compensation unit obtains the displacement of the conveyor belt (1) through a displacement sensor installed in the drive system of the conveyor belt (1).

4. The material flow detection and quantitative distribution device according to claim 1, characterized in that, The data processing module includes: The data preprocessing unit uses the statistical outlier removal algorithm to filter the spatial distribution data for noise, separates the material area from the background of the conveyor belt (1) by dynamic height threshold segmentation, and uses the Euclidean clustering algorithm to extract the effective material accumulation contour segment. The flow calculation unit reconstructs the material layer cross-sectional geometric model based on the effective material accumulation contour segment, and calculates the material cross-sectional area by accumulating the areas of each micro-element of the material layer cross-sectional geometric model. And combined with the linear speed of the conveyor belt (1) obtained by the conveyor belt (1) drive system and scan time interval According to the formula: Calculate the volume of the material ; The density correction unit includes a humidity sensor, used to correct the material density value in real time based on the measured moisture content of the material collected by the humidity sensor, using a dynamic density model. The formula for the dynamic density model is: ； in, This is the corrected material density value. For the material's reference density, This is the humidity correction factor. The measured value of the moisture content of the material; The corrected material density value With respect to the volume of the material The fusion calculation outputs a mass flow rate value as the real-time flow rate value to the control module.

5. The material flow detection and quantitative distribution device according to claim 4, characterized in that, The data processing module also includes a trend prediction unit, which has a built-in traffic trend prediction model for predicting the changing trend of the real-time traffic value based on the historical sequence of the real-time traffic value, and outputting an early warning signal to the control module in advance.

6. The material flow detection and quantitative distribution device according to claim 1, characterized in that, The control module includes: A fuzzy PID controller is used to receive the deviation between the real-time flow value and the preset target flow value, quantify the deviation through a membership function, and generate a speed adjustment amount using a weighted defuzzification rule. The feedforward compensation unit has a built-in feedforward compensation prediction model, which is used to train the feedforward compensation prediction model based on the historical sequence of the real-time flow value, output the feedforward compensation coefficient in advance, and superimpose the feedforward compensation coefficient onto the speed adjustment amount to generate the speed adjustment command.

7. The material flow detection and quantitative distribution device according to claim 1, characterized in that, The actuator is connected to the drive system of at least two conveyor belts (1) via a high-speed communication bus, and outputs the speed adjustment command to each conveyor belt (1) independently.

8. The material flow detection and quantitative distribution device according to claim 1, characterized in that, It also includes an auxiliary module, which includes: The multi-sensor verification unit includes a weighing platform set at the end of the conveyor belt (1), which is used to periodically collect the cumulative weight W of the material, compare it with the flow integral value calculated based on the real-time flow value, generate an error correction coefficient, and iteratively update the calculation parameters of the real-time flow value in the data processing module using the error correction coefficient. The safety protection unit is used to monitor the motor current of the conveyor belt (1) drive system and output protection commands in stages when the limit is exceeded, and to continuously monitor the spatial distribution data and automatically issue a stop command to the conveyor belt (1) drive system when the continuity is interrupted. The monitoring and display unit is connected to the data processing module and is used to display the heat map distribution of material flow and the operating status of equipment in real time.

9. A method for detecting and quantitatively distributing material flow, characterized in that, Includes the following steps: Step S1: The non-contact scanning detection module performs a full-section scan of the material on the conveyor belt (1), and after motion compensation processing and data fusion, obtains the spatial distribution data of the material; Step S2: The data processing module sequentially performs noise filtering, background separation, and material region clustering on the spatially distributed data to obtain effective material accumulation contour segments; The material layer cross-sectional geometric model is reconstructed for the effective material accumulation contour segment, and the material cross-sectional area is calculated by accumulating the areas of each micro-element of the material layer cross-sectional geometric model. And according to the formula Calculate material volume Then, based on the measured moisture content of the material collected by the humidity sensor... Through dynamic density model Perform density correction and output the mass flow rate value as the real-time flow rate value of the material; Step S3: The control module compares the real-time flow value with the preset target flow value, generates a speed adjustment amount through a fuzzy PID controller, and generates a speed adjustment command by superimposing the feedforward compensation coefficient onto the speed adjustment amount through a feedforward compensation prediction model. Step S4: The actuator drives the conveyor belt (1) to run according to the speed adjustment command; Step S5: The auxiliary module periodically calibrates the calculation parameters of the real-time flow value in the data processing module, monitors the operating status of the conveyor belt (1) drive system and outputs protection commands when abnormal, and displays the material cross-section distribution data in the form of a heat map in real time and synchronously displays the equipment operating status.

Images