Digital coal yard stockpile volume high-precision measurement and calculation method based on laser radar and vision fusion

Abstract

The invention discloses a digital coal yard stockpile volume high-precision measurement and calculation method based on laser radar and vision fusion, and relates to the technical field of digital coal yards, and the method comprises the following specific steps: environment parameter monitoring and sensor adjustment: arranging environment sensors at different positions of a coal yard to collect parameters in real time, and transmitting the parameters to a data processing center; by monitoring environmental parameters in real time and automatically adjusting the working mode and parameters of the sensor, the interference of environmental factors on data acquisition is effectively reduced, the accuracy of data is greatly improved, and according to different environmental conditions such as increase of laser radar transmitting power and adjustment of exposure and filter parameters of the visual sensor in case of high dust, the real-time monitoring of the environmental parameters is realized. And meanwhile, an error model is established by utilizing historical data and error compensation is carried out, so that the measurement error is further reduced, the high-precision measurement of the volume of the stockpile in the coal yard is realized, and a reliable basis is provided for precise management of links such as coal production, storage and transportation.

Description

Technical Field

[0001] The present invention relates to the technical field of digital coal yards, and in particular to high-precision measurement of the stockpile volume in a digital coal yard based on the fusion of laser radar and vision. Background Art

[0002] In the production, storage and transportation processes of the coal industry, accurately measuring the volume of coal yard piles is a crucial task. It is of great significance to many links such as inventory management and cost accounting. With the rapid development of science and technology, various measurement technologies have emerged, providing more possibilities for measuring the volume of coal yard piles. Among them, lidar and visual sensor technologies have gradually become research hotspots in this field due to their unique advantages. Lidar can obtain three-dimensional spatial information of the target object by emitting laser beams and measuring the time of reflected light. It has the characteristics of high precision and high resolution. Visual sensors can obtain two-dimensional image information of the target object, which contains rich visual features such as color and texture.

[0003] However, traditional methods for measuring the volume of coal yard stockpiles, such as manual measurement and simple instrument measurement, have many drawbacks. Manual measurement is not only inefficient, but also suffers from significant influences on measurement accuracy due to human factors, making it difficult to guarantee the accuracy of measurement results. Simple instrument measurement also appears to be inadequate in the complex and changing coal yard environment. Coal yard environments are characterized by unstable lighting conditions, high dust concentrations, and large fluctuations in temperature and humidity. These factors can seriously interfere with data acquisition by lidar and vision sensors. For example, high dust concentrations can cause the lidar laser beam to scatter, reducing the accuracy of 3D point cloud data. Changing lighting conditions can cause overexposure or underexposure in images collected by vision sensors, affecting image quality. In addition, existing measurement systems lack the ability to adapt to environmental changes and cannot automatically adjust the sensor's operating mode and parameters according to changes in environmental parameters, resulting in large measurement errors. Furthermore, traditional systems have difficulty tracking the dynamic changes of coal piles in real time and are unable to promptly update the 3D model and volume data of the coal piles, failing to meet the real-time and accuracy requirements of modern coal yard management. Summary of the Invention

[0004] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a high-precision measurement method for the volume of digital coal yard piles based on the fusion of laser radar and vision. It can reasonably arrange a variety of environmental sensors to accurately collect environmental parameters such as lighting conditions, dust concentration, temperature and humidity in the coal yard in real time, and automatically adjust the working mode and parameters of the laser radar and vision sensor according to changes in environmental parameters to reduce the interference of environmental factors on data collection. Then, a specific fusion method is used to process the collected three-dimensional point cloud data and two-dimensional image data to form more comprehensive and accurate coal yard pile fusion data. At the same time, by establishing a relationship model between environmental parameters and measurement errors, the collected data is error compensated to further improve the measurement accuracy. In addition, real-time data is used to dynamically monitor the coal pile, and the three-dimensional model and volume data of the coal pile are updated in time to achieve high-precision, stable, reliable and real-time measurement of the coal yard pile volume.

[0005] To solve the above technical problems, the present invention provides the following technical solution: high-precision measurement of the volume of stockpiles in a digital coal yard based on the fusion of laser radar and vision, the method comprising the following specific steps: Environmental parameter monitoring and sensor adjustment: Environmental sensors are deployed at different locations in the coal yard to collect parameters in real time and transmit them to the data processing center. Based on the comparison results with preset thresholds and the impact of environmental parameters on sensor performance, the operating mode and parameters of the lidar and vision sensors are automatically adjusted to adapt to environmental changes. Data acquisition and fusion: The LiDAR scans the coal yard stockpile using an adjusted pattern to acquire 3D point cloud data. Simultaneously, the visual sensor captures 2D image data using adjusted parameters. Feature extraction and matching are then performed on the data, fusing the dataset containing 3D spatial and visual information. Error analysis and compensation: Establish a historical data repository to record the environmental parameters and error data of each measurement, analyze the error variation pattern under different environmental parameters, combine the historical error average, the difference between the current and historical average environmental parameters, and the residual error obtained from machine learning model training, comprehensively evaluate and predict the measurement error, and correct the collected data; Dynamic coal pile modeling and updating: Using lidar and vision sensors to continuously collect real-time data on coal piles, we compare and analyze data at different times to determine their shape, volume changes, material flow, and calculate the rate of shape change. Volume measurement: Preprocess the fused coal yard stockpile data and the updated coal pile 3D model, divide the model into simple geometric shapes, calculate the volumes of each, and sum them up to obtain the total volume of the coal pile. Output the results and store relevant data and information.

[0006] Furthermore, in the environmental parameter monitoring and sensor adjustment step, environmental sensors are arranged at different locations in the coal yard to collect parameters in real time and transmit them to the data processing center. The sensors include light sensors, dust concentration sensors, temperature sensors and humidity sensors.

[0007] Furthermore, in the environmental parameter monitoring and sensor adjustment step, the operating modes and parameters of the laser radar and the visual sensor are automatically adjusted, and the adjustment formula is: ,in, is the parameter value of the sensor after adjustment, It is to adjust the parameter value of the front sensor. Is the environmental parameter adjustment coefficient, which is a constant determined by the type of sensor and the degree of influence of environmental parameters on sensor performance. is the change in environmental parameters.

[0008] Furthermore, in the error analysis and compensation step, a historical data repository is established to record and store in detail the environmental parameter data and corresponding measurement error data of each measurement, and the data in the historical data repository is regularly sorted and analyzed to observe the changing pattern of measurement errors under different environmental parameter conditions. According to the relationship between the environmental parameters and the measurement errors obtained by analysis, when the real-time environmental parameter data is obtained during each measurement, the possible errors in the current measurement are predicted, and the data collected by the lidar and vision sensor are corrected accordingly based on the predicted errors.

[0009] Furthermore, in the error analysis and compensation step, the error that may occur in the current measurement is predicted, and the prediction formula is: ,in, is the predicted measurement error, which is used to compensate the error of the collected data. are the influence coefficients of historical error, environmental parameter change error, and residual error, is the average value of historical measurement errors, reflecting the overall level of errors in past measurements. is the difference vector between the current environmental parameters and the historical average environmental parameters, is the historical average environmental parameter vector, It is the residual error obtained through machine learning model training, which is used to capture the error part that cannot be explained by historical errors and changes in environmental parameters.

[0010] Furthermore, in the dynamic coal pile modeling and updating step, laser radar and visual sensors are used to continuously collect real-time data of coal yard piles. By comparing the data collected at different times, it is analyzed whether the shape and volume of the coal pile have changed, and the material flow on the surface of the coal pile is observed. When it is detected that the shape or volume of the coal pile has changed, the three-dimensional model of the coal pile is updated according to the newly collected data, including modifying the coordinates of the points on the surface of the coal pile in the model and adjusting the shape parameters of the model. At the same time, the volume data of the coal pile is recalculated according to the updated three-dimensional model to ensure the real-time and accuracy of the volume data.

[0011] Furthermore, in the dynamic coal pile modeling and updating step, the data collected at different times are compared to analyze whether the shape and volume of the coal pile have changed. The analysis formula is: ,in, It is the rate of change of the shape of the coal pile, which comprehensively reflects the changes in the volume and surface area of ​​the coal pile. is the change in the volume of the coal pile, that is, , is the volume of the coal pile at the previous moment, is the change in the surface area of ​​the coal pile, that is, , is the surface area of ​​the coal pile at the previous moment.

[0012] Furthermore, in the volume calculation step, the model is divided into simple geometric shapes, and the volumes of each are calculated and then summed up to obtain the total volume of the coal pile. The calculation formula is: ,in, is the total volume of the coal pile, is the number of tetrahedrons into which the three-dimensional point cloud data of the coal pile is divided. It is The three-dimensional coordinate vectors of the four vertices of a tetrahedron.

[0013] Compared with existing technologies, this high-precision measurement of the volume of stockpiles in a digital coal yard based on the fusion of LiDAR and vision has the following beneficial effects: 1. The present invention effectively reduces the interference of environmental factors on data collection by real-time monitoring of environmental parameters and automatically adjusts the working mode and parameters of the sensor, greatly improving the accuracy of the data. According to different environmental conditions, such as increasing the laser radar transmission power in high dust conditions, adjusting the exposure and filter parameters of the visual sensor, etc., to ensure the normal operation of the sensor. At the same time, historical data is used to establish an error model and perform error compensation, further reducing the measurement error, thereby achieving high-precision measurement of the volume of coal yard piles, and providing a reliable basis for the precise management of coal production, storage and transportation.

[0014] 2. The present invention can track the shape changes and material flow of coal piles in real time through dynamic coal pile modeling and updating, and timely update the three-dimensional model and volume data of the coal pile, meeting the demand for real-time monitoring of the coal pile volume, so that relevant personnel can grasp the dynamic information of the coal pile at any time, and provide timely and accurate data support for coal inventory management and cost accounting.

[0015] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0017] Figure 1 This is an operation diagram of the high-precision measurement process of the stockpile volume in a digital coal yard based on the fusion of LiDAR and vision. Figure 2 This is a flow chart for high-precision volume calculation of digital coal yard stockpiles based on the fusion of lidar and vision. DETAILED DESCRIPTION

[0018] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0019] Example 1 A large thermal power plant has multiple large coal yards for storing the coal needed for power generation. These coal yards have huge coal reserves and frequent coal loading, unloading, and use operations every day. Therefore, accurately measuring the volume of coal piles is crucial for inventory management and cost accounting. If the volume of coal piles cannot be accurately and timely determined, it may lead to insufficient coal supply, affecting power generation production, or cause inventory backlogs and increased costs.

[0020] High-precision light sensors, dust concentration sensors, temperature sensors, and humidity sensors are installed at different locations in the coal yard, including the top and edge of the coal pile and the open area of ​​the coal yard. These sensors have high sensitivity and stability and can monitor the environmental parameters of the coal yard in real time and accurately, such as Figure 1As shown, for example, light sensors can accurately measure subtle changes in light intensity, and dust concentration sensors can detect tiny fluctuations in dust particle concentration in the air. The sensors use advanced wireless communication technology to quickly transmit the environmental parameter data collected in real time to the data processing center. The data processing center is equipped with high-performance servers and professional data processing software, which can perform real-time analysis and processing of large amounts of environmental parameter data.

[0021] When it is found that the dust concentration in the coal yard increases due to coal loading and unloading operations, the data processing center will use the formula To adjust the sensor parameters, here , corresponding to the four environmental parameters of light, dust concentration, temperature and humidity, for (No. The influence coefficient of various environmental parameters on sensor parameter adjustment) is initially determined by testing the measurement accuracy and signal strength of the lidar in different dust concentration environments, recording the relationship between the change in dust concentration and the change in lidar performance, and assigning an initial influence coefficient to the dust concentration parameter based on these relationships. ,Similarly, similar experimental tests are also conducted on parameters such as light, temperature, and humidity.,During the experiment, the performance indicators of the sensor (such as,measurement error, signal stability, etc.) are used as the basis for measuring the,degree of influence of environmental parameters. The greater the change in the performance,the corresponding The larger the value, the more the system runs and accumulates data. The actual data between the environmental parameter changes and the sensor performance are continuously collected, and these actual data are used to Optimize and adjust by using the regression algorithm in machine learning, taking the change of environmental parameters as input and the change of sensor performance as output, and adjust by training the model The value of enables the model to more accurately predict the sensor performance changes, thus obtaining more reasonable weight, (The initial parameter value of the sensor) is determined by the factory setting or initial calibration of the sensor, (No. The actual measured values ​​of various environmental parameters are collected in real time by the corresponding environmental sensors. (No. The reference values ​​of various environmental parameters are determined according to the sensor technical documentation and the ideal working environment requirements.

[0022] After calculation according to the above formula, the data processing center will send instructions to the lidar to appropriately increase the intensity of its transmitted signal to enhance its ability to penetrate dust. At the same time, it will send instructions to the visual sensor to adjust its exposure time to extend the exposure time and enable special filters to reduce the interference of dust on image acquisition. When the light intensity changes and exceeds the optimal working light range of the visual sensor, the data processing center will adjust the exposure parameters of the visual sensor, such as changing the aperture size and adjusting the sensitivity, to ensure that the collected image is clear and accurate. If the changes in temperature and humidity exceed the set range, the data processing center will fine-tune the internal parameters of the lidar and visual sensor to compensate for the impact of environmental changes on sensor performance. For example, by adjusting the operating voltage, current and other parameters of the sensor's electronic components to adapt to different temperature and humidity conditions.

[0023] In accordance with the adjusted working mode, the lidar performs all-round scanning of multiple coal piles in the coal yard with high frequency and high precision. The emitted laser beam can quickly and accurately measure the distance information of each point on the surface of the coal pile, thereby obtaining accurate three-dimensional point cloud data. These point cloud data depict the shape and spatial structure of the coal pile in detail. The visual sensor simultaneously collects two-dimensional image data of the coal pile. It has the characteristics of high resolution and wide viewing angle, and can capture visual features such as color, texture, and shape of the coal pile. Through advanced image acquisition technology, the collected images are ensured to be of high quality and rich in details, integrating data sets containing three-dimensional spatial and visual information.

[0024] Establish a long-term historical data repository to record in detail the environmental parameters, measurement results and errors with the actual volume of each measurement. Through in-depth analysis of a large amount of historical data, use machine learning algorithms and statistical methods to establish an accurate error prediction model. When the ambient temperature and humidity change significantly, the system will use the formula To predict the measurement error and compensate for it, is the predicted measurement error, which is used to compensate the error of the collected data. is the average value of historical measurement errors, which is calculated from historical measurement data. is the difference vector between the current environmental parameters and the historical average environmental parameters, obtained by statistical analysis of the real-time data collected by environmental sensors and historical environmental data. is the historical average environmental parameter vector, The residual error is obtained by training the machine learning model. The machine learning algorithm is used to train historical data. The difference between the model prediction value and the actual measurement value is the residual error. are the influence coefficients of historical error, environmental parameter change error, and residual error, respectively, and , preliminarily determined When setting, based on experience and simple analysis of historical data, if historical data shows that measurement error is mainly related to historical measurement error, the initial setting If it is found that the change of environmental parameters has a significant impact on the measurement error, then increase it appropriately. For the residual error, since it reflects the error part that is difficult to explain, a smaller value can be set initially. By statistically analyzing the historical data, the contribution rate of different factors to the measurement error is calculated, and the initial weight value is determined based on this. When adjusting, the weight is optimized by cross-validation and other methods. The historical data is divided into a training set and a validation set, and different Take the value combination, calculate the prediction error (such as mean square error, etc.), adjust the value combination according to the prediction error performance on the validation set, and continuously optimize until the prediction error is minimized. The system makes corresponding corrections to the data collected by the lidar and vision sensor based on the prediction results. For example, if it is predicted that the measured distance of the lidar may be deviated, the distance data measured by it will be adjusted. If it is predicted that the image collected by the vision sensor is deformed, the image will be corrected.

[0025] Using real-time data from lidar and vision sensors, and using advanced data analysis algorithms to continuously monitor the status of the coal pile, the system can quickly detect changes in the volume and shape of the coal pile when new coal is delivered to the coal yard and added to the coal pile. By calculating the shape change rate of the coal pile, the degree and direction of the change can be determined. The formula used to calculate the shape change rate is ,in is the rate of change of the coal pile shape, which comprehensively reflects the changes in the volume and surface area of ​​the coal pile. is the change in the volume of the coal pile, that is , and The volume of the three-dimensional model of the coal pile at the previous and next moments is calculated. is the change in the surface area of ​​the coal pile, that is, , and The surface area of ​​the three-dimensional model of the coal pile at previous and subsequent moments is calculated. Once the change in the coal pile is detected to reach a certain threshold, the system will immediately start the three-dimensional model update program. Based on the newly collected data, the three-dimensional reconstruction technology is used to accurately update the three-dimensional model of the coal pile. The update process includes recalculating the point cloud coordinates of the coal pile surface, adjusting the model's geometric shape and texture information, etc., to ensure that the updated three-dimensional model can accurately reflect the real-time status of the coal pile. At the same time, the volume data of the coal pile is recalculated and inventory information is updated in real time, providing a reliable basis for the power plant's coal procurement and production scheduling.

[0026] The fused data and updated 3D model are pre-processed, and advanced filtering algorithms and data cleaning techniques are used to remove noise and abnormal points. Then the coal pile model is divided into multiple tetrahedrons and the three-dimensional model is divided into four tetrahedrons according to the formula. Calculate the volume of each tetrahedron, where is the total volume of the coal pile, The number of tetrahedrons into which the three-dimensional point cloud data of the coal pile is divided is obtained by processing the three-dimensional point cloud data of the coal pile with the tetrahedron segmentation algorithm. For the The three-dimensional coordinate vectors of the four vertices of a tetrahedron are obtained from the three-dimensional point cloud data collected by the lidar. These point cloud data are pre-processed and used for tetrahedron segmentation. Finally, the volumes of all tetrahedrons are added together to obtain the total volume of the coal pile. The calculation results are displayed in real time through the power plant's monitoring system. Managers can view the volume information of the coal pile anytime and anywhere through computers, mobile phones and other terminal devices. At the same time, the volume data will be stored in the database for subsequent query, analysis and statistical use. The database will also record the time, environmental parameters and other information of each volume measurement, which is convenient for managers to trace and analyze historical data, summarize rules and further optimize coal yard management.

[0027] Example 2 A coal logistics transfer station is responsible for the storage and transfer of large quantities of coal. The daily coal throughput is huge, and coal loading, unloading, and transshipment operations are frequent. Due to the entry and exit of different batches of coal and changes in stacking methods, the shape and volume of the coal pile are constantly changing. Accurately measuring the volume of the coal pile is critical for logistics scheduling (such as the reasonable arrangement of transport vehicles and loading and unloading equipment) and cost settlement (such as the accurate calculation of storage and transportation costs). If the volume measurement is inaccurate, it may lead to waste of logistics resources, increase operating costs, and even cause business disputes.

[0028] Environmental sensors are rationally and evenly arranged in different areas of the transfer yard, including coal-intensive areas, loading and unloading areas, and surrounding open areas. These sensors include high-precision light sensors, dust concentration sensors that can detect the concentration of fine dust particles, temperature sensors that can accurately measure temperature changes, and humidity sensors that are sensitive to humidity changes. All types of sensors have strong anti-interference capabilities and fast response speeds. The sensors transmit the real-time collected environmental parameter data to the data processing center at an extremely fast speed through a low-power, high-bandwidth wireless communication network. Figure 2As shown in the figure, the data processing center is equipped with a powerful data analysis server and an intelligent data processing software system, which can receive, store and deeply analyze massive amounts of environmental parameter data in real time. When encountering severe weather (such as cloudy days, haze, etc.) that causes lighting conditions to deteriorate, the data processing center responds quickly and automatically adjusts the exposure parameters of the visual sensor based on the pre-established correspondence model between environmental parameters and sensor performance. It not only increases the sensitivity, but also intelligently adjusts the aperture size and shutter speed to ensure that clear and complete coal pile image data can be collected. At the same time, according to the real-time changes in dust concentration, the transmission power of the lidar is accurately adjusted. When the dust concentration is high, the transmission power is greatly increased to enhance the penetration ability of the laser, ensure the accuracy and integrity of the three-dimensional point cloud data, and reduce the interference of dust on data collection.

[0029] The lidar uses advanced scanning technology and, in accordance with the adjusted working mode, rapidly scans the coal piles in the transfer yard with high resolution and wide coverage. The emitted laser beam can accurately measure the three-dimensional spatial coordinates of each point on the surface of the coal pile and obtain high-precision three-dimensional point cloud data. These point cloud data present the three-dimensional shape and spatial distribution of the coal pile in detail. The visual sensor is equipped with a high-performance image sensor and high-quality optical lens, and synchronously collects two-dimensional image data of the coal pile. It can capture the rich color information, unique texture features and overall shape contours of the coal pile. Through the optimized image acquisition algorithm, the collected images are ensured to have high clarity, high contrast and accurate color reproduction, integrating data sets containing three-dimensional space and visual information.

[0030] A historical data repository is established to comprehensively record the environmental parameters of each measurement (including detailed information such as lighting, dust, temperature, and humidity), measurement results (such as the measured coal pile volume and shape), and the error compared to the actual volume (obtained by comparison with high-precision measurement equipment or methods). Data mining technology and machine learning algorithms are used to conduct in-depth analysis and mining of massive amounts of historical data. When a change in environmental parameters is detected, the system automatically inputs the current environmental parameters into an error prediction model trained on a large amount of data. The model can predict possible measurement errors based on the complex relationship between environmental parameters and measurement errors in the historical data. For example, when temperature and humidity change significantly at the same time, the model predicts that the measurement accuracy of the lidar may be affected because changes in temperature and humidity may cause changes in the propagation medium of the laser, which in turn affects the propagation speed and reflection characteristics of the laser. Based on the predicted results, the system will compensate the collected data accordingly. By processing the lidar measurement data with temperature and humidity compensation algorithms, the deviation of the measured distance is corrected. The visual sensor image data is filtered, denoised, and geometrically corrected to eliminate image deformation and noise interference caused by environmental changes, thereby improving measurement accuracy and reducing errors.

[0031] Utilizing real-time data from lidar and vision sensors, a dynamic monitoring algorithm is adopted to continuously monitor the status of the coal pile. By conducting detailed comparative analysis of the data collected at different times, the system can keenly detect changes in the volume and shape of the coal pile. When coal is transported out of the coal pile, the system can not only detect the reduction in volume, but also accurately determine the specific changes in the shape of the coal pile, such as the location and degree of depression on the surface of the coal pile.

[0032] By accurately calculating the shape change rate of the coal pile and setting a reasonable threshold to judge the magnitude of the change, once it is detected that the change of the coal pile exceeds the threshold, the system will immediately start the 3D model update program. Based on the newly collected high-precision data, the advanced 3D reconstruction algorithm is used to quickly and accurately update the 3D model of the coal pile. The update process includes recalculating the point cloud coordinates of the coal pile surface, accurately adjusting the geometric shape of the model to make it completely match the shape of the actual coal pile, and updating the texture information of the model so that it can truly reflect the appearance characteristics of the coal pile. The updated 3D model can reflect the current status of the coal pile in real time and accurately, providing accurate basic data for logistics scheduling. The system will also synchronously recalculate the volume data of the coal pile to ensure the real-time and accuracy of the volume data, providing a reliable basis for cost settlement.

[0033] The fused coal pile data and updated 3D model are preprocessed using a variety of advanced data processing algorithms, such as median filtering to remove noise points and statistical analysis to identify and eliminate outliers. The coal pile model is then segmented into multiple regular tetrahedrons. High-precision geometric calculation methods are used to accurately calculate the volume of each tetrahedron based on its vertex coordinates. Finally, the volumes of all tetrahedrons are added together to obtain the total volume of the coal pile. The calculation results are fed back to relevant personnel in real time through the visual interface of the logistics management system. Relevant personnel can view the coal pile volume information anytime and anywhere through terminal devices such as computers, tablets, or smartphones. At the same time, the volume data is securely and reliably stored in a database. The database also records in detail the time of each volume measurement, environmental parameters, and relevant data during the measurement process, facilitating subsequent query, analysis, and statistics. By analyzing historical volume data, logistics transfer stations can summarize the patterns of coal storage and transportation, optimize logistics scheduling plans, improve operational efficiency, and reduce costs. At the same time, accurate volume data provides an accurate basis for cost settlement and avoids business disputes caused by inaccurate volume measurement.

[0034] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. High-precision measurement of the volume of stockpiled materials in a digital coal yard based on the fusion of laser radar and vision, characterized by: The method comprises the following specific steps: Environmental parameter monitoring and sensor adjustment: Environmental sensors are deployed at different locations in the coal yard to collect parameters in real time and transmit them to the data processing center. Based on the comparison results with preset thresholds and the impact of environmental parameters on sensor performance, the operating mode and parameters of the lidar and vision sensors are automatically adjusted to adapt to environmental changes. Data acquisition and fusion: The LiDAR scans the coal yard using an adjusted pattern to obtain 3D point cloud data. Simultaneously, the visual sensor captures 2D image data using adjusted parameters. Feature extraction and matching are performed on the data, and the resulting data set is fused to include both 3D spatial and visual information. Error analysis and compensation: Establish a historical data repository to record the environmental parameters and error data of each measurement, analyze the error variation pattern under different environmental parameters, combine the historical error average, the difference between the current and historical average environmental parameters, and the residual error obtained from machine learning model training, comprehensively evaluate and predict the measurement error, and correct the collected data; Dynamic coal pile modeling and updating: Using lidar and vision sensors to continuously collect real-time data on coal piles, we compare and analyze data at different times to determine their shape, volume changes, material flow, and calculate the rate of shape change. Volume measurement: Preprocess the fused coal yard stockpile data and the updated coal pile 3D model, divide the model into simple geometric shapes, calculate the volumes of each, and sum them up to obtain the total volume of the coal pile. Output the results and store relevant data and information.

2. The high-precision measurement of the volume of stockpiled materials in a digital coal yard based on the fusion of laser radar and vision according to claim 1 is characterized in that: In the environmental parameter monitoring and sensor adjustment step, environmental sensors are arranged at different locations in the coal yard to collect parameters in real time and transmit them to the data processing center. The sensors include light sensors, dust concentration sensors, temperature sensors and humidity sensors.

3. The high-precision measurement of the volume of stockpiled materials in a digital coal yard based on the fusion of laser radar and vision according to claim 1 is characterized in that: In the environmental parameter monitoring and sensor adjustment step, the operating modes and parameters of the laser radar and visual sensor are automatically adjusted. The adjustment formula is: ,in, is the parameter value of the sensor after adjustment, It is to adjust the parameter value of the front sensor. Is the environmental parameter adjustment coefficient, which is a constant determined by the type of sensor and the degree of influence of environmental parameters on sensor performance. is the change in environmental parameters.

4. The high-precision measurement of the volume of stockpiled materials in a digital coal yard based on the fusion of laser radar and vision according to claim 1 is characterized in that: In the error analysis and compensation step, a historical data repository is established to record and store the environmental parameter data and corresponding measurement error data of each measurement in detail, and the data in the historical data repository is regularly sorted and analyzed to observe the changing pattern of the measurement error under different environmental parameter conditions. According to the relationship between the environmental parameters and the measurement error obtained by analysis, when the real-time environmental parameter data is obtained during each measurement, the possible error of the current measurement is predicted, and the data collected by the lidar and visual sensor are corrected accordingly based on the predicted error.

5. The high-precision measurement of the volume of stockpiled materials in a digital coal yard based on the fusion of laser radar and vision according to claim 1 is characterized in that: In the error analysis and compensation step, the error that may occur in the current measurement is predicted, and the prediction formula is: ,in, is the predicted measurement error, which is used to compensate the error of the collected data. are the influence coefficients of historical error, environmental parameter change error, and residual error, is the average value of historical measurement errors, reflecting the overall level of errors in past measurements. is the difference vector between the current environmental parameters and the historical average environmental parameters, is the historical average environmental parameter vector, It is the residual error obtained through machine learning model training, which is used to capture the error part that cannot be explained by historical errors and changes in environmental parameters.

6. The high-precision measurement of the volume of stockpiled materials in a digital coal yard based on the fusion of laser radar and vision according to claim 1 is characterized in that: In the dynamic coal pile modeling and updating step, laser radar and visual sensors are used to continuously collect real-time data of coal yard piles. By comparing the data collected at different times, it is analyzed whether the shape and volume of the coal pile have changed, and the material flow on the surface of the coal pile is observed. When a change in the shape or volume of the coal pile is detected, the three-dimensional model of the coal pile is updated according to the newly collected data, including modifying the coordinates of the points on the surface of the coal pile in the model and adjusting the shape parameters of the model. At the same time, the volume data of the coal pile is recalculated according to the updated three-dimensional model to ensure the real-time and accuracy of the volume data.

7. The high-precision measurement of the volume of stockpiled materials in a digital coal yard based on the fusion of laser radar and vision according to claim 1 is characterized in that: In the dynamic coal pile modeling and updating step, the data collected at different times are compared to analyze whether the shape and volume of the coal pile have changed. The analysis formula is: ,in, It is the rate of change of the shape of the coal pile, which comprehensively reflects the changes in the volume and surface area of ​​the coal pile. is the change in the volume of the coal pile, that is, , is the volume of the coal pile at the previous moment, is the change in the surface area of ​​the coal pile, that is, , is the surface area of ​​the coal pile at the previous moment.

8. The high-precision measurement of the volume of stockpiled materials in a digital coal yard based on the fusion of laser radar and vision according to claim 1 is characterized in that: In the volume calculation step, the model is divided into simple geometric shapes, and the volumes of each are calculated and then summed up to obtain the total volume of the coal pile. The calculation formula is: ,in, is the total volume of the coal pile, is the number of tetrahedrons into which the three-dimensional point cloud data of the coal pile is divided. It is The three-dimensional coordinate vectors of the four vertices of a tetrahedron.

Images