A system and method for planning the path of after-compaction rolling mill based on compaction quality evaluation

Abstract

The present invention discloses a re-rolling path planning system based on compaction quality evaluation, including: a central server, a client, a vehicle-mounted controller, a human-machine interface device, a perception module and a wireless transmission module; the perception module includes an RTK-GPS for detecting the position of the roller and an acceleration sensor for detecting the acceleration of the roller wheel; the vehicle-mounted controller receives the detection signal from the perception module, and sends it to the central server and the human-machine interface device after processing; it receives the operation signal from the central server and the human-machine interface device, and outputs the control signal to the roller actuator after processing; the central server has a neural network model built in it, which receives the signal from the vehicle-mounted controller and the client, generates the re-rolling path data after processing and sends it to the vehicle-mounted controller and the human-machine interface device. The present invention also discloses a re-rolling path planning method based on compaction quality evaluation. The present invention uniformly plans the re-rolling path and improves the re-rolling efficiency.

Description

Technical Field

[0001] The invention relates to the field of earth-rock dam rolling construction, and in particular to a system and method for planning a rolling path based on compaction quality evaluation. Background Art

[0002] At present, in the construction process of earth-rock dam, compaction degree is an important criterion for measuring construction quality. The compaction process of earth-rock dam is mainly achieved by manually driving a vibrating roller to roll on the paved dam material with certain construction parameters. In the rolling construction process of earth-rock dam, the rolling parameters (such as: paving thickness, number of rolling passes, driving speed and roller excitation force, etc.) are first determined through on-site rolling experiments, and then the construction management personnel and the roller driver cooperate to carry out construction work on the designated rolling surface. After rolling, the rolling surface needs to be inspected and the unqualified rolling surface needs to be re-rolled. Usually, the unqualified rolling surface is inspected and re-rolled by humans. This inspection and re-rolling method has the following disadvantages: low rolling efficiency, inaccurate judgment of the re-rolling area, irregular driving mode of the roller, and low mechanical utilization rate. Summary of the invention

[0003] In order to solve the technical problems existing in the known technology, the present invention provides a system and method for accurately judging the re-rolling area and improving the rolling efficiency, and a re-rolling path planning system based on compaction quality evaluation is provided.

[0004] The technical solution adopted by the present invention to solve the technical problems existing in the known technology is: a re-rolling path planning system based on compaction quality evaluation, including: a central server for planning the re-rolling path according to the compaction quality information, a client for remotely inputting operation information, a vehicle-mounted controller for controlling the operation of the roller actuator, a human-machine interaction interface device for on-site human-machine interaction with the vehicle-mounted controller, a perception module for detecting the operating status of the roller, and a wireless transmission module for communication between the central server and the vehicle-mounted controller; the perception module includes an RTK-GPS for detecting the position of the roller and a Acceleration sensor; the on-board controller receives the detection signal from the perception module, and sends it to the central server and the human-machine interface device after processing; it receives the operation signal from the central server and the human-machine interface device, and outputs the control signal to the roller actuator after processing; the central server has a neural network model with the roller operation status as input and the compaction degree as output, which receives the signal from the on-board controller and the client, and generates the rolling path data after processing; it sends the rolling path data to the on-board controller, and sends it to the human-machine interface through the on-board controller.

[0005] Furthermore, the vehicle-mounted controller includes an industrial computer, and the industrial computer is electrically connected to the human-machine interaction interface device, the perception module, the wireless transmission module and the actuator of the roller respectively.

[0006] Furthermore, the wireless transmission module includes a data transmission station and an antenna.

[0007] Furthermore, the RTK-GPS detects the center position coordinates of the roller compactor.

[0008] The present invention also provides a method for planning a re-rolling path based on compaction quality evaluation, the method comprising: setting a central server for planning a re-rolling path according to compaction quality information, setting a client for remotely inputting operation information, setting a vehicle-mounted controller for controlling the operation of a roller actuator, setting a human-machine interaction interface device for on-site human-machine interaction with the vehicle-mounted controller, setting a perception module for detecting the position coordinates and roller acceleration of the roller, and setting a wireless transmission module for realizing communication between the central server and the vehicle-mounted controller; the perception module adopts RTK-GPS and an acceleration sensor; the RTK-GPS detects the position coordinates of the roller; the acceleration sensor detects the roller acceleration speed; a vehicle-mounted controller, which receives the detection signal from the perception module, and sends it to the central server and the human-machine interface device after processing; it receives the operation signal from the central server and the human-machine interface device, and outputs the control signal to the roller actuator after processing; the central server, which has a neural network model with the roller operation status as input and the compaction degree as output, receives the signal from the vehicle-mounted controller and the client, processes it through the built-in neural network model, obtains the area data with poor compaction quality that needs to be re-rolled, and further generates the re-rolling path data; it sends the re-rolling path data to the vehicle-mounted controller, and sends it to the human-machine interface through the vehicle-mounted controller.

[0009] Furthermore, the method comprises the following steps:

[0010] Step 1: The remote operator inputs operation information to the central server through the client, and the operation information includes roller parameters and bin surface construction parameters;

[0011] Step 2: When the roller is working, the sensing module detects the position coordinates of the roller and the acceleration of the roller wheel in real time, and sends the detection signal to the vehicle controller;

[0012] Step 3, the vehicle-mounted controller processes the received detection signal, generates the position coordinates of the roller and the vibration state data of the roller corresponding to the position coordinates, sends the generated data to the human-machine interface device for display, and sends it to the central server through the wireless transmission module;

[0013] Step 4: After the rolling is completed, the central server processes the data received from the vehicle controller and the client through the built-in neural network model to generate a compaction cloud map and a rolling pass cloud map;

[0014] Step 5: The central server obtains the area with poor compaction quality that needs additional rolling based on the compaction degree cloud map and the rolling pass cloud map, and then generates additional rolling path data based on the position coordinates of each roller and the operation information input by the client. The central server sends the additional rolling path data to the vehicle controller of the corresponding roller through the wireless transmission module;

[0015] Step 6: The vehicle-mounted controller sends the re-rolling path data to the human-machine interface device for display; the on-site operator performs re-rolling operations according to the display.

[0016] Furthermore, in step 1, the roller parameters include: vehicle number, roller width, roller turning radius; the bin surface construction parameters include: bin surface boundary, and rolling times required.

[0017] Furthermore, in step 4, the method for generating the compaction degree cloud map and the rolling pass cloud map comprises the following steps:

[0018] Step A-1, the central server obtains the historical track of the roller operation according to the position coordinates of the roller and the vibration property data of the roller corresponding to the position coordinates, widens the historical track according to the roller width, obtains the rolling number cloud map of the entire warehouse surface, and then obtains the rolling number and the vibration property data of the roller at any point on the warehouse surface;

[0019] Step A-2, during the bunker surface construction process, the quality inspection personnel randomly select a quality inspection point on the bunker surface for rolling operation, inspect the compaction degree of the quality inspection point, obtain the compaction degree data of the quality inspection point, and simultaneously obtain the rolling number of the quality inspection point and the vibration state data of the roller from the position coordinates of the quality inspection point;

[0020] Step A-3, establishing a neural network model; using the rolling number and roller vibration performance data of the quality inspection point as model input, and using the compaction degree of the quality inspection point as model output;

[0021] Step A-4, collecting the rolling times, roller vibration performance data and compaction degree of several quality inspection points, making training sample sets and test sample sets; training and testing the neural network model;

[0022] Step A-5, the trained neural network model is used to predict the compaction degree of any point on the warehouse surface and generate a compaction cloud map.

[0023] Furthermore, in step A-4, new quality inspection point data is regularly collected and added to the training sample set and the test sample set, and the updated training sample set and the test sample set are regularly used to train and predict the neural network model.

[0024] Furthermore, in step 5, the method for planning the path for rolling reinforcement includes the following steps:

[0025] Step B-1, the central server analyzes the compaction degree cloud map and the rolling number cloud map to obtain the area that needs additional rolling and the number of additional rolling passes;

[0026] Step B-2, dividing the rolling bin surface into a number of blocks, taking the center of the block as the node that the roller passes through, and the central server uses the shortest path algorithm based on the position coordinate information of each roller to determine the shortest path for each roller to reach each supplementary rolling area;

[0027] Step B-3: the central server plans the re-rolling path according to the operation information input by the client and based on the shortest path principle.

[0028] The advantages and positive effects of the present invention are as follows: by setting up a central server and a remote client to communicate with the on-board controller of the roller on site, the operating status information of the roller can be obtained in real time, including the running track of the roller, the exciting force, the acceleration of the roller wheel, etc., so as to realize the collection of real-time operating data of the roller; the central server predicts the compaction degree of each point on the bin surface by the input data of each on-board controller through the built-in neural network model, thereby analyzing the area that needs to be re-rolled due to poor compaction quality; the central server can use the shortest path algorithm and other intelligent algorithms to uniformly plan the re-rolling path based on the operating information input by the client and the shortest path principle, and send the re-rolling path data to the on-board controller of the corresponding roller, thereby reducing the workload of manual detection of compaction; it also avoids the problem of uneven mechanical utilization and low re-rolling efficiency caused by manual decentralized judgment of the re-rolling area and individual re-rolling operations. The re-rolling path planning system and method based on compaction quality evaluation of the present invention improves the re-rolling efficiency, thereby improving the construction efficiency. In addition, a human-machine interactive interface device is provided to facilitate on-site operators to understand the operating status of the roller in real time and to receive operating information from the central server in a timely manner, making operations more convenient. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a structural block diagram of a rolling path planning system based on compaction quality evaluation of the present invention.

[0030] Figure 2 The present invention is a workflow diagram of a method for planning a re-rolling path based on compaction quality evaluation. DETAILED DESCRIPTION

[0031] In order to further understand the content, features and effects of the present invention, the following embodiments are listed and described in detail with reference to the accompanying drawings:

[0032] See also Figure 1 to Figure 2A system for planning a re-rolling path based on compaction quality evaluation, comprising: a central server for planning a re-rolling path according to compaction quality information, a client for remotely inputting operation information, a vehicle-mounted controller for controlling the operation of a roller actuator, a human-machine interaction interface device for on-site human-machine interaction with the vehicle-mounted controller, a perception module for detecting the operating status of the roller, and a wireless transmission module for communication between the central server and the vehicle-mounted controller; the perception module comprises an RTK-GPS for detecting the position of the roller and an acceleration sensor for detecting the acceleration of the roller wheel; the vehicle-mounted controller receives information from the vehicle-mounted controller; The detection signal of the perception module is processed and sent to the central server and the human-machine interaction interface device; the on-board controller receives the operation signal from the central server and the human-machine interaction interface device, and outputs the control signal to the roller actuator after processing; the central server is equipped with a neural network model whose input is the roller operation status and output is the compaction degree. The central server receives the signal from the on-board controller and the client, and generates the rolling path data after processing; the central server sends the rolling path data to the on-board controller, and sends it to the human-machine interaction interface through the on-board controller.

[0033] The on-board controller can send the processed detection signal and the operation signal from the central server to the human-machine interaction interface device for display. In this way, the human-machine interaction interface device can display the operating status of the roller and the re-rolling path and other instructions sent by the central server in real time. On-site operators can input operation instructions and other operation signals in the human-machine interaction interface device to control the operation of the roller actuator through the on-board controller.

[0034] The vehicle-mounted controller receives the detection signal from the sensing module and the input signal from the human-machine interface device, generates the position coordinates of the roller and the vibration state data of the roller corresponding to the position coordinates after processing, and sends the generated data to the human-machine interface device for display, and sends it to the central server through the wireless transmission module;

[0035] The vehicle-mounted controller, the human-machine interface device, the sensing module, and the wireless transmission module can be arranged on each roller. The central server can be arranged in a remote command center, and the client can be connected to the central server of the remote command center through a network or other communication methods.

[0036] RTK-GPS, which means GPS with carrier phase differential function in Chinese, can select GPS products with carrier phase differential function in existing technologies.

[0037] The central server may be a server in the prior art.

[0038] A client refers to a device that has installed an application corresponding to the central server and can provide local services to users. The client can be a personal computer connected to the central server via a wired network, or a mobile phone or tablet computer connected to the central server via a wireless network.

[0039] The neural network model can adopt a neural network model applicable in the prior art, take the known parameters related to the compaction degree as the input of the model, take the unknown parameters such as the compaction degree as the output of the model, collect corresponding training samples and test samples, train and test the neural network model, and then use the trained neural network model to predict the compaction degree based on the known data.

[0040] The on-board controller can adopt a roller compactor on-board controller applicable in the prior art, and the on-board controller can also include an industrial computer in the prior art, and the industrial computer can be electrically connected to the human-computer interaction interface device, the perception module, the wireless transmission module and the roller compactor's actuator respectively.

[0041] By setting up a central server and a remote client to communicate with the on-board controller of the roller on site, the central server can obtain the operating status information of the roller in real time, including the roller's running trajectory, exciting force, acceleration of the rolling wheel, etc., to realize the collection of real-time operating status data of the roller; the central server receives real-time operating data from the on-board controller of each roller through the built-in neural network model to predict the compaction degree of each point on the bin surface, thereby analyzing the areas with poor compaction quality that need additional rolling; based on the operating information input by the client and the shortest path principle, the additional rolling path is uniformly planned, and the additional rolling path data is sent to the on-board controller of the corresponding roller.

[0042] The vibration characteristics of the roller are the time domain and frequency domain characteristics of the roller vibration acceleration. Through the rolling test, it is found that there is a certain regularity between the change of the roller vibration characteristics and the change of the compaction degree. The roller vibration performance data refers to the vibration acceleration time domain and frequency domain data, which can be calculated from the roller acceleration. A neural network model with the input of the rolling number and the roller vibration performance data and the output of the compaction degree can be set in the central server; the compaction degree of any point on the bin surface can be predicted by the roller vibration performance data of the point through the neural network model.

[0043] The human-machine interface device can adopt a human-machine interface such as a touch screen in the prior art, or edit the human-machine interface on a display through human-machine interaction software, and the operator inputs manual instructions and other information using a keyboard. Setting up a human-machine interaction interface device facilitates on-site operators to understand the roller compactor operation status in real time. On-site operators can also communicate with the central server in real time, timely feedback the on-site operation status and receive information from the central server, so that they can operate more conveniently.

[0044] The wireless transmission module may include a data transmission radio and an antenna. The wireless transmission module may adopt a wireless communication module applicable in the prior art, such as Zigbee, wifi, mobile communication, data transmission radio, etc.

[0045] The RTK-GPS can be located at the top or bottom center of the roller compactor, so as to detect the center coordinates of the roller compactor.

[0046] The acceleration sensor may be any acceleration sensor suitable for use in the prior art.

[0047] The present invention also provides a method for planning a re-rolling path based on compaction quality evaluation, the method comprising: setting a central server for planning a re-rolling path according to compaction quality information, setting a client for remotely inputting operation information, setting a vehicle-mounted controller for controlling the operation of a roller actuator, setting a human-machine interaction interface device for on-site human-machine interaction with the vehicle-mounted controller, setting a perception module for detecting the position coordinates and roller acceleration of the roller, and setting a wireless transmission module for realizing communication between the central server and the vehicle-mounted controller; the perception module adopts RTK-GPS and an acceleration sensor; the RTK-GPS detects the position coordinates of the roller; the acceleration sensor detects the roller acceleration speed; a vehicle-mounted controller, which receives the detection signal from the perception module, and sends it to the central server and the human-machine interface device after processing; it receives the operation signal from the central server and the human-machine interface device, and outputs the control signal to the roller actuator after processing; the central server, which has a neural network model with the roller operation status as input and the compaction degree as output, receives the signal from the vehicle-mounted controller and the client, processes it through the built-in neural network model, obtains the area data with poor compaction quality that needs to be re-rolled, and further generates the re-rolling path data; it sends the re-rolling path data to the vehicle-mounted controller, and sends it to the human-machine interface through the vehicle-mounted controller.

[0048] The central server processes the data received from the vehicle-mounted controller and the client through the built-in neural network model to obtain the data of the area with poor compaction quality that needs re-rolling, and further generates the re-rolling path data, which is sent to the vehicle-mounted controller; the vehicle-mounted controller sends the re-rolling path data to the human-machine interface device for display; the on-site operators input the operation instructions according to the display of the human-machine interface device; the vehicle-mounted controller outputs the control signal according to the operation instructions and the re-rolling path data to control the operation of the roller actuator.

[0049] The on-board controller can generate the position coordinates of the roller and the vibration property data of the roller corresponding to the position coordinates according to the detection signal received from the perception module. A neural network model with the number of rolling passes and the vibration property data of the roller as input and the compaction degree as output can be set in the central server. The compaction degree of any point on the bin surface can be predicted by the vibration property data of the roller at that point through the neural network model.

[0050] The method may include the following steps:

[0051] Step 1: The remote operator inputs operation information to the central server through the client, and the operation information includes roller parameters and bin surface construction parameters;

[0052] Step 2: When the roller is working, the sensing module detects the position coordinates of the roller and the acceleration of the roller wheel in real time, and sends the detection signal to the vehicle controller;

[0053] Step 3, the vehicle-mounted controller processes the received detection signal, generates the position coordinates of the roller and the vibration state data of the roller corresponding to the position coordinates, sends the generated data to the human-machine interface device for display, and sends it to the central server through the wireless transmission module;

[0054] Step 4: After the rolling is completed, the central server processes the data received from the vehicle controller and the client through the built-in neural network model to generate a compaction cloud map and a rolling pass cloud map;

[0055] Step 5: The central server obtains the area with poor compaction quality that needs additional rolling based on the compaction degree cloud map and the rolling pass cloud map, and then generates additional rolling path data based on the position coordinates of each roller and the operation information input by the client. The central server sends the additional rolling path data to the vehicle controller of the corresponding roller through the wireless transmission module;

[0056] Step 6: The vehicle-mounted controller sends the re-rolling path data to the human-machine interface device for display; the on-site operator performs re-rolling operations according to the display.

[0057] Furthermore, in step 1, the roller parameters may include: vehicle number, roller width, roller turning radius; the bin surface construction parameters may include: bin surface boundary, and rolling number requirements.

[0058] Furthermore, in step 4, the method for generating the compaction degree cloud map and the rolling pass cloud map may include the following steps:

[0059] Step A-1, the central server obtains the historical track of the roller operation according to the position coordinates of the roller and the vibration property data of the roller corresponding to the position coordinates, widens the historical track according to the roller width, obtains the rolling number cloud map of the entire warehouse surface, and then obtains the rolling number and the vibration property data of the roller at any point on the warehouse surface;

[0060] Step A-2, during the bunker surface construction process, the quality inspection personnel randomly select a quality inspection point on the bunker surface for rolling operation, inspect the compaction degree of the quality inspection point, obtain the compaction degree data of the quality inspection point, and simultaneously obtain the rolling number of the quality inspection point and the vibration state data of the roller from the position coordinates of the quality inspection point;

[0061] Step A-3, establishing a neural network model; using the rolling number and roller vibration performance data of the quality inspection point as model input, and using the compaction degree of the quality inspection point as model output;

[0062] Step A-4, collecting the rolling times, roller vibration performance data and compaction degree of several quality inspection points, making training sample sets and test sample sets; training and testing the neural network model;

[0063] Step A-5, the trained neural network model is used to predict the compaction degree of any point on the warehouse surface and generate a compaction cloud map.

[0064] Furthermore, in step A-4, new quality inspection point data can be regularly collected and added to the training sample set and the test sample set, and the updated training sample set and the test sample set can be regularly used to train and predict the neural network model.

[0065] Furthermore, in step 5, the method for planning the path for additional rolling may include the following steps:

[0066] Step B-1, the central server analyzes the compaction degree cloud map and the rolling number cloud map to obtain the area that needs additional rolling and the number of additional rolling passes;

[0067] Step B-2, divide the rolling bin surface into several blocks, take the center of the block as the node that the roller passes through, and the central server uses the shortest path algorithm based on the position coordinate information of each roller to determine the shortest path for each roller to reach each re-rolling area; the rolling bin surface can be divided into several blocks according to a certain granularity, the side length of the block can be equal to the width of the roller, and the passable node connection line can be used as the optional path of the roller. The shortest path algorithm in the prior art can be selected to calculate the shortest driving path for each roller to travel to a certain re-rolling area. The passable paths and the inaccessible paths can be marked separately, and the movement amount from the initial position of the roller to the target position is calculated separately. Starting from the initial position, the next movable position with the smallest movement amount around is continuously selected until it moves to the end point.

[0068] Step B-3, the central server plans the re-rolling path based on the operation information input by the client and the shortest path principle. The central server corresponds to a re-rolling area, selects the roller with the smallest shortest path value to the re-rolling area, plans the re-rolling path of the roller according to the corresponding operation information, and sends the re-rolling path information to the on-board controller of the roller.

[0069] The working process and working principle of the present invention are further described below by a preferred embodiment of the present invention:

[0070] See also Figure 1 A re-rolling path planning system based on compaction quality evaluation, the system includes: a vehicle-mounted controller located on the roller, a perception module, a wireless transmission module, and a remote central server and client. The vehicle-mounted controller uses a high-performance industrial computer; the perception module includes perception devices including: RTK-GPS, acceleration sensor; the wireless transmission module includes a data transmission radio and antenna, a neural network model is set in the central server, and a client application corresponding to the central server is installed in the client. The vehicle-mounted controller can communicate with the perception module and the wireless transmission module in real time through the serial port.

[0071] The roller position and roller acceleration are sensed by the sensing module installed on the roller. The vehicle-mounted controller transmits data to and from the central server through the wireless transmission module installed on the roller. The vehicle-mounted controller installed on the roller processes the signal detected by the sensing module, and sends the processed data to the central server through the wireless transmission module. The central server stores and analyzes the data, and generates a compaction cloud map, a rolling pass cloud map, and a re-rolling path. Parameters are input and compaction quality is checked through the client. A human-machine interface device connected to the vehicle-mounted controller is set up, and the on-site operator can view the re-rolling path and real-time rolling operation status through the human-machine interface device to perform re-rolling operations.

[0072] like Figure 2 As shown, a method for planning a path for supplementary rolling based on compaction quality evaluation is provided. The method utilizes the above-mentioned system for planning a path for supplementary rolling based on compaction quality evaluation. The method specifically comprises the following steps:

[0073] Step (1): the on-site operator starts the roller compactor, and the remote operator starts the central server and the client; the server application is installed on the central server, and the client application is installed on the client.

[0074] Step (2), the vehicle-mounted controller, wireless transmission module and perception module are powered on and started; the power lines of the vehicle-mounted controller, wireless transmission module and perception module are connected to the onboard power supply of the roller, and when the roller is powered on, the vehicle-mounted controller and perception module are automatically powered on and started.

[0075] Step (3), input roller parameters and bin surface construction parameters on the client; roller parameters refer to roller width, antenna height, roller number, roller turning radius; bin surface construction information refers to bin surface boundary, roller vibration frequency, and vehicle speed.

[0076] Step (4), when the roller is in operation, the sensing module senses the roller position and roller acceleration in real time and sends them to the controller; the RTK-GPS of the sensing module senses the roller position coordinates in real time, and the acceleration sensor of the sensing module senses the roller acceleration. The sensing module's detection information is transmitted to the vehicle controller via the serial port.

[0077] (5) The on-board controller processes the signals detected and collected by the sensing module in real time, packages them and sends them to the wireless transmission module.

[0078] (6) The wireless transmission module sends the data packet to the central server.

[0079] (7) The central server parses the data packet and judges the compaction quality through the built-in neural network model, generates a corresponding compaction quality cloud map, plans the re-rolling path based on the compaction quality cloud map, and sends the re-rolling path data to the on-board controller and client of the roller. The compaction quality cloud map includes the compaction degree cloud map and the rolling pass cloud map.

[0080] The method for generating the compaction quality cloud map is as follows: by connecting the collected coordinates of the center position of the roller, the historical trajectory of the roller operation can be obtained. By widening the trajectory according to the width of the roller, the rolling number cloud map of the entire bin surface can be obtained, and then the rolling number of any point on the bin surface can be calculated. During the bin surface construction process, the quality inspectors will test and measure the compaction degree P of the point at the randomly sampled quality inspection point on the bin surface of the rolling operation by the pit inspection method. By using the position coordinates of the quality inspection point as a label, the rolling number of the quality inspection point, the roller vibration state data and the compaction degree data can be obtained. The rolling number and the roller vibration state data are used as the input data of the neural network model, and the compaction degree data is used as the output data of the neural network model. The training samples and test samples are made from the rolling number of the quality inspection points, the roller vibration state data and the compaction degree data. The training samples and the test samples are used to train and test the neural network model to obtain a trained neural network model.

[0081] The trained neural network model can predict the compaction degree of any point on the warehouse surface. The training samples and test samples of the model are not static. As the dam construction progresses, more quality inspection point data are regularly added to the training sample set and the test sample set. The neural network model regularly updates the training sample set and the test sample set to improve the compaction degree prediction accuracy. In this way, the compaction value of any point on the warehouse surface can be accurately obtained, so that a real-time compaction degree cloud map of the construction warehouse surface can be generated.

[0082] (8) The central server determines whether the compaction quality meets the requirements based on the obtained data analysis.

[0083] By comparing the compaction quality evaluation results with the required standards, determine whether the compaction quality of any point on the bin surface meets the requirements. If not, determine the missing number of rolling passes and status and mark the corresponding area as a supplementary rolling area.

[0084] (9) The central server plans the re-rolling path and sends the re-rolling path data to the wireless transmission module.

[0085] The method for planning the path for the additional rolling can be as follows: simplify the rolling bin surface into blocks according to a certain granularity, and record the shortest path blocks that have been considered and the blocks that are no longer considered, respectively calculate the movement amount from the initial position of the roller to the moving position and the estimated value from the moving position to the end point, and start from the initial position, continuously select the next movable position with the minimum movement amount and the estimated value from the moving position to the end point until it moves to the end point. The lines connecting the selected blocks are the path for the additional rolling.

[0086] (10) The wireless transmission module sends the received compaction quality data and re-rolling path data to the vehicle controller.

[0087] (11) The on-board controller analyzes the data and displays the compaction quality cloud map and re-rolling path through the human-computer interaction interface.

[0088] (12) On-site workers carry out supplementary rolling operations based on the compaction quality cloud map and supplementary rolling path. The supplementary rolling operation is completed when the compaction degree of the bin surface meets the requirements.

[0089] (13) The crushing is completed.

[0090] The embodiments described above are only used to illustrate the technical ideas and features of the present invention, and their purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. The patent scope of the present invention cannot be limited only by this embodiment, that is, any equivalent changes or modifications made to the spirit disclosed by the present invention still fall within the patent scope of the present invention.

Claims

1. A rolling path planning system based on compaction quality evaluation, characterized in that: include: A central server for planning a re-rolling path according to compaction quality information, a client for remotely inputting operation information, a vehicle-mounted controller for controlling the operation of the roller actuator, a human-machine interface device for on-site human-machine interaction with the vehicle-mounted controller, a perception module for detecting the operation status of the roller, and a wireless transmission module for communication between the central server and the vehicle-mounted controller; the perception module includes an RTK-GPS for detecting the position of the roller and an acceleration sensor for detecting the acceleration of the roller wheel; the vehicle-mounted controller receives the detection signal from the perception module, and sends it to the central server and the human-machine interface device after processing; It receives operating signals from the central server and the human-machine interaction interface device, and outputs control signals to the roller actuator after processing; the central server is equipped with a neural network model whose input is the roller operating status and output is the compaction degree. It receives signals from the vehicle-mounted controller and the client, and generates supplementary rolling path data after processing; it sends the supplementary rolling path data to the vehicle-mounted controller, and sends it to the human-machine interaction interface through the vehicle-mounted controller.

2. The re-rolling path planning system based on compaction quality evaluation according to claim 1 is characterized in that: The vehicle-mounted controller includes an industrial computer, and the industrial computer is electrically connected to the human-machine interaction interface device, the sensing module, the wireless transmission module and the actuator of the roller respectively.

3. The re-rolling path planning system based on compaction quality evaluation according to claim 1 is characterized in that: The wireless transmission module includes a data transmission station and an antenna.

4. The re-rolling path planning system based on compaction quality evaluation according to claim 1 is characterized in that: The RTK-GPS detects the center position coordinates of the roller.

5. A method for planning a rolling path based on compaction quality evaluation, characterized in that: The method comprises the following steps: setting a central server for planning a re-rolling path according to compaction quality information, setting a client for remotely inputting operation information, setting a vehicle-mounted controller for controlling the operation of a roller actuator, setting a human-machine interaction interface device for on-site human-machine interaction with the vehicle-mounted controller, setting a perception module for detecting the position coordinates and roller acceleration of the roller, and setting a wireless transmission module for realizing communication between the central server and the vehicle-mounted controller; the perception module adopts RTK-GPS and an acceleration sensor; the RTK-GPS detects the position coordinates of the roller; and the acceleration sensor detects the roller acceleration; The on-board controller receives the detection signal from the perception module, and sends it to the central server and the human-machine interface device after processing; it receives the operation signal from the central server and the human-machine interface device, and outputs the control signal to the roller actuator after processing; the central server has a built-in neural network model with the roller operation status as input and the compaction degree as output, which receives the signal from the on-board controller and the client, processes it through the built-in neural network model, obtains the area data with poor compaction quality that needs to be re-rolled, and further generates the re-rolling path data; it sends the re-rolling path data to the on-board controller, and sends it to the human-machine interface through the on-board controller.

6. The method for planning a re-rolling path based on compaction quality evaluation according to claim 5, characterized in that: The method comprises the following steps: Step 1: The remote operator inputs operation information to the central server through the client, and the operation information includes roller parameters and bin surface construction parameters; Step 2: When the roller is working, the sensing module detects the position coordinates of the roller and the acceleration of the roller wheel in real time, and sends the detection signal to the vehicle controller; Step 3, the vehicle-mounted controller processes the received detection signal, generates the position coordinates of the roller and the vibration state data of the roller corresponding to the position coordinates, sends the generated data to the human-machine interface device for display, and sends it to the central server through the wireless transmission module; Step 4: After the rolling is completed, the central server processes the data received from the vehicle controller and the client through the built-in neural network model to generate a compaction cloud map and a rolling pass cloud map; Step 5: The central server obtains the area with poor compaction quality that needs additional rolling based on the compaction degree cloud map and the rolling pass cloud map, and then generates additional rolling path data based on the position coordinates of each roller and the operation information input by the client. The central server sends the additional rolling path data to the vehicle controller of the corresponding roller through the wireless transmission module; Step 6: The vehicle-mounted controller sends the re-rolling path data to the human-machine interface device for display; the on-site operator performs re-rolling operations according to the display.

7. The method for planning a re-rolling path based on compaction quality evaluation according to claim 6, characterized in that: In step 1, the roller parameters include: vehicle number, roller width, roller turning radius; the bin surface construction parameters include: bin surface boundary, rolling times required.

8. The method for planning a re-rolling path based on compaction quality evaluation according to claim 6, characterized in that: In step 4, the method for generating the compaction degree cloud map and the rolling pass cloud map includes the following steps: Step A-1, the central server obtains the historical track of the roller operation according to the position coordinates of the roller and the vibration property data of the roller corresponding to the position coordinates, widens the historical track according to the roller width, obtains the rolling number cloud map of the entire warehouse surface, and then obtains the rolling number and the vibration property data of the roller at any point on the warehouse surface; Step A-2, during the bunker surface construction process, the quality inspection personnel randomly select a quality inspection point on the bunker surface for rolling operation, inspect the compaction degree of the quality inspection point, obtain the compaction degree data of the quality inspection point, and simultaneously obtain the rolling number of the quality inspection point and the vibration state data of the roller from the position coordinates of the quality inspection point; Step A-3, establishing a neural network model; using the rolling number and roller vibration performance data of the quality inspection point as model input, and using the compaction degree of the quality inspection point as model output; Step A-4, collecting the rolling times, roller vibration performance data and compaction degree of several quality inspection points, making training sample sets and test sample sets; training and testing the neural network model; Step A-5, the trained neural network model is used to predict the compaction degree of any point on the warehouse surface and generate a compaction cloud map.

9. The method for planning a re-rolling path based on compaction quality evaluation according to claim 8, characterized in that: In step A-4, new quality inspection point data is collected regularly and added to the training sample set and the test sample set, and the updated training sample set and the test sample set are used regularly to train and predict the neural network model.

10. The method for planning a re-rolling path based on compaction quality evaluation according to claim 6, characterized in that: In step 5, the method for planning the path for rolling reinforcement includes the following steps: Step B-1, the central server analyzes the compaction degree cloud map and the rolling number cloud map to obtain the area that needs additional rolling and the number of additional rolling passes; Step B-2, dividing the rolling bin surface into a number of blocks, taking the center of the block as the node that the roller passes through, and the central server uses the shortest path algorithm based on the position coordinate information of each roller to determine the shortest path for each roller to reach each supplementary rolling area; Step B-3: the central server plans the re-rolling path according to the operation information input by the client and based on the shortest path principle.

Images