Automatic deviation rectifying system for cantilever beam mechanism of bucket wheel machine

Abstract

The invention provides an automatic deviation rectifying system for a cantilever beam mechanism of a bucket wheel machine, which relates to the technical field of cantilever beams of bucket wheel machines and comprises a cantilever beam, a driving mechanism, a monitoring device and a controller. The cantilever beam is used for driving the bucket wheel to move. The driving mechanism is used for controlling the bucket wheel machine and the cantilever beam to move. The monitoring device is used for monitoring state data and driving data of the bucket wheel machine and target material state data. And the controller is used for predicting the feeding rate at the next moment so as to judge whether the position of the cantilever beam deviates. The control data at the current moment are generated during deviation, and the control data are circularly optimized according to the feedback state data at the next moment, so that the automatic deviation correction of the cantilever beam is realized. According to the method, the multi-view image is identified and processed, whether the cantilever beam has deviation is judged, and the cantilever beam is subjected to cyclic optimization control through the A-C network model when the deviation exists, so that the method has the characteristics of high real-time performance and high precision, the operation efficiency of the bucket wheel machine is effectively improved, and the operation risk of the bucket wheel machine is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of bucket wheel machine cantilever beams, and in particular to an automatic deviation correction system for a bucket wheel machine cantilever beam mechanism. Background Art

[0002] Bucket wheel excavators are a common material handling equipment widely used in ports, mines and other large logistics sites. They are mainly used to load and unload bulk materials such as coal and ore onto transport vehicles. One of the key components of a bucket wheel excavator is its cantilever beam mechanism, which is responsible for the suspension and movement of the bucket wheel, ensuring that the bucket wheel can effectively grab and transport materials during operation. However, since bucket wheel excavators are affected by a variety of factors during operation (such as wind force, changes in the center of gravity of materials, and the operator's control accuracy, etc.), the cantilever beam mechanism is prone to deviation, which not only affects the efficiency of material loading and unloading, but may also damage the equipment and shorten its service life.

[0003] In order to improve the working efficiency and stability of bucket wheel machines, traditional technology usually adopts manual adjustment and correction. This method is not only inefficient, but also easily limited by the operator's experience and judgment ability, which leads to unnecessary losses and delays. At present, although there are control systems for bucket wheel machines on the market, most systems are mainly focused on drive and motion control, and lack the functions of dynamic adjustment and accurate control of cantilever beam mechanisms. This greatly limits the automation level and operating efficiency of bucket wheel machines. Therefore, the demand for automatic correction technology for the cantilever beam mechanism of bucket wheel machines is becoming more and more urgent. Summary of the invention

[0004] The invention provides an automatic deviation correction system for a cantilever beam mechanism of a bucket wheel machine, which is used to solve the defect in the prior art that the cantilever beam mechanism of the bucket wheel machine is difficult to automatically correct its deviation in a timely and effective manner, resulting in low operating efficiency and high risk.

[0005] On the one hand, the present invention provides an automatic deviation correction system for a cantilever beam mechanism of a bucket wheel machine, comprising a cantilever beam, a driving mechanism, a monitoring device and a controller.

[0006] The cantilever beam is used to drive the bucket wheel to move.

[0007] The driving mechanism includes a pitch drive module, a slewing drive module and a travel drive module. The travel drive module is used to drive the bucket wheel machine to move as a whole. The pitch drive module is used to start the contact end of the cantilever beam and the bucket wheel to do pitch movement. The slewing drive module is used to drive the cantilever beam to rotate.

[0008] The monitoring device includes a state monitoring module, a drive monitoring module and a target monitoring module. The state monitoring module is used to monitor the state data of the bucket wheel machine. The drive monitoring module is used to monitor the drive data of the bucket wheel machine. The target monitoring module is used to monitor the state data of the target material.

[0009] The controller includes a feed rate prediction module, a deviation judgment module, and an automatic deviation correction module. The feed rate prediction module is used to predict the feed rate at the next moment based on the bucket wheel machine parameter information, the current bucket wheel machine state data, the drive data, and the target material state data. The deviation judgment module is used to determine whether the error between the current feed rate and the current predicted feed rate exceeds the preset threshold range, if so, there is a deviation in the cantilever beam position. The automatic deviation correction module is used to generate the current moment control data based on the current state data and the current target material position and the current target material shape, and cyclically optimize the control data based on the feedback of the next moment state data to realize automatic deviation correction of the cantilever beam.

[0010] The automatic deviation correction system of the cantilever beam mechanism of the bucket wheel machine monitors the bucket wheel machine and the target material in real time to determine whether the cantilever beam has deviations, and performs adaptive deviation correction when the cantilever beam has deviations. In the judgment process, the target material is photographed at multiple angles to obtain images of the target material from different perspectives, and then the shape and size of the target material are obtained by identifying and processing the multi-perspective images, so as to calculate the ideal feed rate according to the state data and drive data of the bucket wheel machine, and determine whether the cantilever beam has deviations by comparing the ideal feed rate with the actual feed rate. When there is a deviation, the drive data for deviation correction is calculated through the Actor network, and the position of the cantilever beam after deviation correction is evaluated by the Critic network, so as to realize the cyclic optimization control of the cantilever beam, which has the characteristics of high real-time performance and high precision, effectively improves the operation efficiency of the bucket wheel machine, and reduces the operation risk of the bucket wheel machine.

[0011] According to the automatic deviation correction system of the cantilever beam mechanism of a bucket wheel machine provided by the present invention, the state data includes the rotation center position of the cantilever beam, the cantilever beam angle, the bucket wheel machine moving speed, the bucket wheel speed, and the feed rate. The driving data includes the pitch control data, the rotation control data, the travel control data, and the bucket wheel speed control data.

[0012] According to an automatic deviation correction system for a cantilever beam mechanism of a bucket wheel excavator provided by the present invention, a target monitoring module includes an image acquisition unit, a preprocessing unit, a target recognition unit and a three-dimensional modeling unit. The image acquisition unit is used to acquire an image of a target material and use a radar detector to acquire the distance between the bucket wheel and the target material. The preprocessing unit is used to preprocess the image of the target material to obtain a preprocessed image. The target recognition unit is used to perform edge detection and contour extraction on the preprocessed image to obtain a contour image of the target material. The three-dimensional modeling unit is used to construct a three-dimensional shape of the target material based on contour images at multiple angles and calculate the three-dimensional size of the target material.

[0013] According to an automatic deviation correction system for a cantilever beam mechanism of a bucket wheel machine provided by the present invention, a three-dimensional modeling unit uses a stereoscopic vision method to construct a three-dimensional object model, and the specific steps include:

[0014] Calibrate the camera to obtain the camera's intrinsic and extrinsic parameters.

[0015] Feature points are extracted from two contour images extracted from different angles.

[0016] Calculate the disparity between two silhouette images.

[0017] Depth is calculated based on disparity and the camera's baseline.

[0018] Use the camera parameters to convert the depth into 3D coordinates in the camera coordinate system.

[0019] Different feature points and three-dimensional coordinates are used to generate point cloud data, and the point cloud data is used for three-dimensional reconstruction to obtain the three-dimensional shape and three-dimensional size of the target material.

[0020] According to an automatic deviation correction system for a bucket wheel machine cantilever beam mechanism provided by the present invention, the calculation formula of the parallax is expressed as:

[0021] d = x L -x R

[0022] Where d is the parallax, x is L 、x R are the horizontal coordinates of the left and right contour images respectively. The calculation formula of depth is expressed as:

[0023]

[0024] Where Z is the depth, f is the focal length of the camera, and B is the baseline of the camera.

[0025] The three-dimensional horizontal and vertical coordinates are calculated according to the depth. The formula is:

[0026]

[0027] In the formula, c x 、c y are the principal point coordinates of the camera, y L are the ordinates of the left contour image respectively.

[0028] The final three-dimensional coordinates are expressed as:

[0029]

[0030] Where P is the three-dimensional coordinate of the feature point.

[0031] According to an automatic deviation correction system for a cantilever beam mechanism of a bucket wheel excavator provided by the present invention, the prediction process of the feed rate prediction module includes:

[0032] A three-dimensional coordinate system is established with the rotation center of the cantilever beam as the origin, the travel direction as the X-axis, and the direction perpendicular to the ground as the Z-axis, and the bucket wheel excavator and the target material are mapped into the three-dimensional coordinate system.

[0033] The ideal state data of the bucket wheel at the next moment is calculated based on the current driving data and the current state data. The formula is expressed as:

[0034] C t+1 =f(Q t , C t )

[0035] In the formula, C t+1 is the bucket wheel machine state data at the next moment, f() is the bucket wheel machine motion equation, C t is the current state data, Q t It is the current control data.

[0036] According to the next state data, the distance between the bucket wheel and the target material surface and the linear speed of the bucket wheel are calculated, and then the feed rate at the next moment is calculated. The formula is expressed as:

[0037] v m =Wsv l

[0038] Where W is the bucket wheel width, s is the distance between the bucket wheel edge and the target material surface, and v l is the linear speed of the bucket wheel.

[0039] According to an automatic deviation correction system for a cantilever beam mechanism of a bucket wheel machine provided by the present invention, the automatic deviation correction module uses an AC network model to adjust the driving data of the bucket wheel machine to realize automatic deviation correction of the cantilever beam. The construction process of the AC network model includes:

[0040] Construct an Actor network. The formula for generating control actions by the Actor network is expressed as:

[0041] Q t =π(C t ,θ actor )

[0042] Where π represents the neural network that approximates the Actor policy network, and θactor is the parameter of the Actor network.

[0043] Construct a Critic network. The formula of the value function of the Critic network is expressed as:

[0044]

[0045] Where τ is the parameter of the Critic network.

[0046] According to an automatic deviation correction system for a cantilever beam mechanism of a bucket wheel machine provided by the present invention, the automatic deviation correction module uses the TD error to update the value function of the Critic network, and the formula is expressed as follows:

[0047]

[0048] Where α is the learning rate of the Critic network.

[0049] Update the parameters of the Actor network according to the evaluation of the Critic network. The formula is expressed as:

[0050]

[0051] Where β is the learning rate of the Actor network.

[0052] A controller for an automatic deviation correction system of a bucket wheel excavator cantilever beam mechanism provided according to the present invention also includes a deviation analysis module, which is used to analyze the cause of the deviation of the cantilever beam when there is a deviation in the cantilever beam position, and upload the deviation data and the deviation cause to a log.

[0053] The automatic deviation correction system of the cantilever beam mechanism of the bucket wheel provided by the present invention monitors the bucket wheel and the target material in real time, determines whether the cantilever beam has deviation, and performs adaptive deviation correction when the cantilever beam has deviation. In the judgment process, by taking photos of the target material at multiple angles, images of the target material at different perspectives are obtained, and then the shape and size of the target material are obtained by identifying and processing the multi-perspective images, so as to calculate the ideal feed rate according to the state data and drive data of the bucket wheel, and by comparing the ideal feed rate with the actual feed rate, determine whether the cantilever beam has deviation, and when there is deviation, calculate the drive data for deviation correction through the Actor network, and use the Critic network to evaluate the position of the cantilever beam after deviation correction, so as to realize the cyclic optimization control of the cantilever beam, which has the characteristics of high real-time performance and high precision, effectively improves the operation efficiency of the bucket wheel and reduces the operation risk of the bucket wheel. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the present invention or 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 It is a schematic diagram of the architecture of the automatic deviation correction system of the cantilever beam mechanism of a bucket wheel excavator provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Combine the following Figure 1 The invention discloses an automatic deviation-correcting system for a cantilever beam mechanism of a bucket wheel machine.

[0058] Figure 1 It is a schematic diagram of the architecture of the automatic deviation correction system of the cantilever beam mechanism of a bucket wheel excavator provided in an embodiment of the present invention.

[0059] like Figure 1 As shown, the automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism provided by the embodiment of the present invention includes a cantilever beam, a driving mechanism, a monitoring device and a controller.

[0060] The cantilever beam is used to drive the bucket wheel to move. The cantilever beam body is usually made of high-strength steel to withstand large dynamic loads. The cross-sectional design of the beam may be a hollow rectangle or I-shaped to provide good strength and rigidity. The front end of the cantilever beam is connected to the bucket wheel, which is equipped with multiple excavator buckets for excavating materials. The bucket wheel on the cantilever beam is responsible for direct excavation. By rotating the bucket wheel, the excavator bucket can grab and crush the material on the ground. After the excavator bucket grabs the material from the ground, the cantilever beam lifts it and transports it to the conveyor belt system behind it, which then transports the material to the stockpile area or transportation system.

[0061] The driving mechanism includes a pitch drive module, a slewing drive module, and a travel drive module. The travel drive module is used to drive the bucket wheel machine to move as a whole. The pitch drive module is used to start the contact end of the cantilever beam and the bucket wheel to do pitch movement. The slewing drive module is used to drive the cantilever beam to rotate.

[0062] The monitoring device includes a state monitoring module, a drive monitoring module and a target monitoring module.

[0063] The condition monitoring module is used to monitor the status data of the bucket wheel machine, including the rotation center position p of the cantilever beam. c 、Cantilever beam angle θ c , bucket wheel excavator moving speed v c 、Bucket wheel speedω c , feed rate v m .

[0064] The drive monitoring module is used to monitor bucket wheel machine drive data, including pitch control data. q , rotation control data q , walking control data w qand bucket wheel speed control data b In this embodiment, the driving data is recorded as Q(u q , r q , w q , r b ), the state data is recorded as C(p c ,θ c , v c ,ω c , v m ).

[0065] The target monitoring module is used to monitor the target material status data. In this embodiment, by performing video monitoring on the target material, a real-time video of the target material is obtained, and a frame is extracted from the real-time video to obtain a material image, and the shape and position of the material image are recognized to obtain the actual position and shape of the target material.

[0066] The target monitoring module includes an image acquisition unit, a preprocessing unit, a target recognition unit and a three-dimensional modeling unit.

[0067] The image acquisition unit is used to acquire the image of the target material, and uses a radar detector to acquire the distance between the bucket wheel and the target material. In actual detection, multiple cameras are installed in different directions of the material, or multiple cameras are installed at different positions on the bucket wheel and the cantilever beam to take multi-directional photos of the target material and acquire multi-view images of the target material.

[0068] The preprocessing unit is used to preprocess the target material image, including grayscale conversion and image denoising, to obtain a preprocessed image.

[0069] Among them, the formula for grayscale conversion is expressed as:

[0070] I gray (x, y)=0.299R(x, y)+0.587G(x, y)+0.114B(x, y)

[0071] In the formula, I gray is the grayscale value, (x, y) is the pixel coordinate, R, G, B are the red, green, and blue color channel brightness of the target material image respectively.

[0072] This embodiment uses Gaussian filtering to remove noise, and the formula is expressed as:

[0073]

[0074] In the formula, I G is the image after Gaussian filtering, I(x+i, y+i) is the pixel value of the original image I at position (x+i, y+i), and k is the filter radius.

[0075] The target recognition unit is used to perform edge detection and contour extraction on the preprocessed image to obtain the recognition shape and recognition size of the target material.

[0076] Edge detection: Use the Canny algorithm to perform edge detection and extract the single-view contour of the target material.

[0077] For the target image I G , use the Sobel operator to calculate the horizontal and vertical gradients, then the horizontal gradient is expressed as:

[0078]

[0079] The vertical gradient is expressed as:

[0080]

[0081] In the formula

[0082] Correspondingly, the gradient amplitude is expressed as:

[0083]

[0084] Directions:

[0085]

[0086] Then, the edge is refined by non-maximum suppression. For each pixel:

[0087] ①. Determine the main direction of the gradient (horizontal, vertical or diagonal) according to the gradient direction θ.

[0088] ② Check the gradient values ​​of the two neighbors before and after the pixel in the gradient direction. If the gradient value of the current pixel is a local maximum, keep the point, otherwise suppress it to zero.

[0089] Finally, the edges are tracked using hysteresis thresholding:

[0090] Set the high threshold T H and low threshold T L , all gradient values ​​above T H The pixels below T are marked as strong edges. L The pixels are marked as non-edge.

[0091] For the gradient value at T H and T L Pixels between are marked as edges if they are adjacent to any strong edge pixels; otherwise, they are suppressed as non-edges.

[0092] The contours of the target material are extracted from the multi-angle images acquired at the same time to obtain the multi-angle contour images of the target material.

[0093] The three-dimensional modeling unit is used to construct the three-dimensional shape of the target material according to the contour images of the target material at multiple angles and calculate the three-dimensional size of the target material.

[0094] In this embodiment, a three-dimensional object model is constructed based on the contour images of the target material from different viewing angles using a stereoscopic vision method. The specific steps include:

[0095] Camera calibration: Before stereo vision reconstruction, the camera needs to be calibrated to obtain the camera's intrinsic and extrinsic parameters. The intrinsic parameters include the camera's focal length, principal point coordinates, and distortion coefficients. The extrinsic parameters describe the camera's posture and translation vector relative to the world coordinate system.

[0096] Feature point extraction and matching: Extract feature points from the left and right views, for example, using SIFT, SURF or ORB feature extraction algorithms. Feature point matching can be performed by calculating the similarity of descriptors.

[0097] Calculate parallax: Parallax is the difference in horizontal coordinates of the same object point in the left and right images. Set the coordinates of the feature points in the left and right views to be (x L ,y L )、(x R ,y R ), then the parallax is expressed as:

[0098] d = x L -x R

[0099] Where d is the parallax.

[0100] Calculate depth information: Based on the disparity d and the camera baseline B (the distance between the two camera centers), the depth can be calculated. The formula is:

[0101]

[0102] Where Z is the depth and f is the focal length of the camera.

[0103] Calculate 3D coordinates: Once the depth information is obtained, it can be converted into 3D coordinates in the camera coordinate system using the camera parameters, including:

[0104]

[0105] In the formula, c x 、c y are the principal point coordinates of the camera respectively.

[0106] So the final three-dimensional coordinates can be expressed as:

[0107]

[0108] Point cloud generation: Repeat the above processing results, use different matching points and camera parameters to generate point cloud data, and finally use the point cloud for 3D reconstruction to obtain the 3D shape and 3D size of the target material.

[0109] The controller includes a feed rate prediction module, a deviation judgment module, a deviation analysis module, and an automatic deviation correction module.

[0110] The feed rate prediction module is used to predict the feed rate at the next moment based on bucket wheel machine parameter information, current bucket wheel machine state data, drive data and target material state data.

[0111] Specifically, first, a three-dimensional coordinate system is established with the rotation center of the cantilever beam as the origin, the travel direction as the X-axis, and the direction perpendicular to the ground as the Z-axis, and the bucket wheel excavator and the target material are mapped into the three-dimensional coordinate system.

[0112] Secondly, the feed rate is calculated based on the bucket wheel speed, bucket wheel machine speed, current drive data, and the distance between the bucket wheel and the target material. The specific steps include:

[0113] The bucket wheel grabs material in each rotation. The grab area can be approximated as the width of the bucket wheel (usually the effective height of the bucket width W), multiplied by the linear speed of the bucket wheel. l It can be calculated by the bucket wheel speed (angular velocity) ω c And bucket wheel radius R, the formula is expressed as:

[0114] v l =Rω c

[0115] The feed rate can be expressed as the grab volume of the bucket wheel, that is:

[0116] v m =Wsv l

[0117] Where W is the bucket wheel width and s is the distance between the bucket wheel edge and the target material surface.

[0118] The ideal state data of the bucket wheel at the next moment is calculated based on the current driving data and the current state data. The formula is expressed as:

[0119] C t+1 =f(Q t , C t )

[0120] In the formula, C t+1 is the bucket wheel machine state data at the next moment, f() is the bucket wheel machine motion equation, C t is the current state data, Q t It is the current control data.

[0121] The distance between the bucket wheel and the surface of the target material and the linear speed of the bucket wheel are calculated according to the next state data, and then the feed rate at the next moment is calculated.

[0122] The deviation judgment module is used to judge whether the error between the current feed rate and the current predicted feed rate exceeds the preset threshold range. If so, there is a deviation in the cantilever beam position, and relevant data is sent to the deviation analysis module and the automatic correction module respectively. Otherwise, the cantilever beam position is normal. For example, in this embodiment, the error is set to 10%, that is, when the error between the current feed rate and the predicted feed rate exceeds 10%, it is determined that there is a deviation in the cantilever beam position. In addition, it is also possible to judge whether there is a deviation in the cantilever beam position based on the error between the predicted bucket wheel machine state data at the next moment and the actual state data of the bucket wheel machine at the next moment, or to judge whether there is a deviation in the cantilever beam by combining the cantilever beam position and the feed rate. The judgment standard is set to: E=a1E v +a2E p There is a deviation when ≥E0, where E is the comprehensive error value, a1 is the feed rate error weight, and E v is the feed rate error value, a2 is the cantilever beam position error weight, E p is the cantilever beam position error value, and E0 is the maximum error threshold.

[0123] The deviation analysis module is used to analyze the cause of the cantilever beam deviation when there is a deviation in the cantilever beam position, and upload the deviation data and the deviation cause to the log. The causes of the cantilever beam position deviation mainly include mechanical quality problems, manual operation problems, external force problems, environmental interference problems and system program problems. By checking various problems one by one, find the actual cause of the cantilever beam position deviation and make timely adjustments to ensure the safe, stable and efficient operation of the bucket wheel excavator.

[0124] The automatic deviation correction module is used to generate the control data at the current moment according to the current state data, the current target material position, and the current target material shape, and to optimize the control data according to the feedback state data at the next moment to realize the automatic deviation correction of the cantilever beam. The specific steps include:

[0125] Calculate the control data at the current moment, the formula is expressed as:

[0126]

[0127] Where Ct is the ideal state data of the bucket wheel excavator at the next moment.

[0128] Whether the deviation correction is completed is determined based on the actual status data and actual feed rate at the next moment.

[0129] In this embodiment, the AC (Actor-Critic) network model is used to adjust the driving data of the bucket wheel machine to achieve automatic deviation correction of the cantilever beam. Specifically, the state variable is set to S t (C t , M t ), the action variable is Q t . Set the reward function of the AC network model to:

[0130] R t =-(ω1|δC t , δM t |+ω2|δv l |)

[0131] Where ω1 and ω2 are weights, which are used to adjust the influence of the cantilever state, target position state and bucket wheel speed.

[0132] The AC network model includes the Actor network and the Critic network. In the Actor network, the input is the state variable S t , action variable Q t , the output is the action variable Q t+1 .

[0133] The formula for generating control actions by the Actor network is expressed as:

[0134] Q t =π(C t ,θ actor )

[0135] Where π represents the neural network that approximates the Actor policy network, and θactor is the parameter of the Actor network.

[0136] In the Critic network, the input is the state variable S t :The output is the value V(C t , Q t ), the formula of the value function is expressed as:

[0137]

[0138] Where τ is the parameter of the Critic network.

[0139] The reward is calculated based on the next state data of the feedback, and then the TD error is used to update the value function of the Critic network. The calculation formula of the TD error is expressed as:

[0140] Δ t =Δ t +γV(C t+1 ,μ(C t+1 ,θ actor ))-V(Ct , Q t )

[0141] In the formula, γ is the discount factor.

[0142] Use TD error to update the parameters of the Critic network. The formula is:

[0143]

[0144] Where α is the learning rate of the Critic network.

[0145] Then, the parameters of the Actor network are updated according to the evaluation of the Critic network. The formula is expressed as:

[0146]

[0147] Where β is the learning rate of the Actor network.

[0148] The AC network model is trained using bucket wheel excavator cantilever beam operation data, and the trained Actor-Critic model is tested in a simulation environment to observe its performance under different conditions, and the reward function and model parameters are adjusted in a timely manner according to the test results.

[0149] The trained model is deployed to the actual bucket wheel excavator control system to realize adaptive deviation correction control of the cantilever beam of the bucket wheel excavator.

[0150] The automatic deviation correction system of the cantilever beam mechanism of the bucket wheel machine of this embodiment monitors the bucket wheel machine and the target material in real time to determine whether the cantilever beam has deviations, and performs adaptive deviation correction when the cantilever beam has deviations. In the judgment process, by taking photos of the target material at multiple angles, images of the target material at different perspectives are obtained, and then the shape and size of the target material are obtained by identifying and processing the multi-perspective images, so as to calculate the ideal feed rate according to the state data and drive data of the bucket wheel machine, and by comparing the ideal feed rate with the actual feed rate, it is determined whether the cantilever beam has deviations, and when there are deviations, the drive data for deviation correction is calculated through the Actor network, and the position of the cantilever beam after deviation correction is evaluated by the Critic network, so as to realize the cyclic optimization control of the cantilever beam, which has the characteristics of high real-time performance and high precision, effectively improves the operation efficiency of the bucket wheel machine, and reduces the operation risk of the bucket wheel machine.

[0151] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0152] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An automatic deviation correction system for a cantilever beam mechanism of a bucket wheel excavator, characterized in that: include: A cantilever beam is used to drive the bucket wheel to move; The driving mechanism includes a pitch driving module, a slewing driving module and a walking driving module. The walking driving module is used to drive the bucket wheel machine to move as a whole; the pitch driving module is used to start the contact end of the cantilever beam and the bucket wheel to do pitch movement; the slewing driving module is used to drive the cantilever beam to rotate; The monitoring device comprises a state monitoring module, a drive monitoring module and a target monitoring module; the state monitoring module is used to monitor the state data of the bucket wheel machine; the drive monitoring module is used to monitor the drive data of the bucket wheel machine; the target monitoring module is used to monitor the state data of the target material; The controller includes a feed rate prediction module, a deviation judgment module, and an automatic deviation correction module; the feed rate prediction module is used to predict the feed rate at the next moment according to bucket wheel machine parameter information, current bucket wheel machine state data, drive data, and target material state data; The deviation judgment module is used to judge whether the error between the current feed rate and the current predicted feed rate exceeds a preset threshold range, and if so, there is a deviation in the cantilever beam position; The automatic deviation correction module is used to generate current control data according to current state data, current target material position, and current target material shape, and cyclically optimize control data according to feedback of next-moment state data to achieve automatic deviation correction of the cantilever beam.

2. The automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism according to claim 1 is characterized in that: The state data includes the rotation center position of the cantilever beam, the cantilever beam angle, the bucket wheel machine speed, the bucket wheel speed, and the feed rate; the drive data includes pitch control data, rotation control data, travel control data, and bucket wheel speed control data.

3. The automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism according to claim 1 is characterized in that: The target monitoring module includes an image acquisition unit, a preprocessing unit, a target recognition unit and a three-dimensional modeling unit; the image acquisition unit is used to acquire the image of the target material and use a radar detector to acquire the distance between the bucket wheel and the target material; the preprocessing unit is used to preprocess the image of the target material to obtain a preprocessed image; the target recognition unit is used to perform edge detection and contour extraction on the preprocessed image to obtain a contour image of the target material; the three-dimensional modeling unit is used to construct the three-dimensional shape of the target material according to the contour images at multiple angles and calculate the three-dimensional size of the target material.

4. The automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism according to claim 3 is characterized in that: The three-dimensional modeling unit uses a stereoscopic vision method to construct a three-dimensional object model, and the specific steps include: Calibrate the camera to obtain the camera's intrinsic and extrinsic parameters; Extracting feature points from the two contour images extracted from different angles; Calculating the disparity between the two contour images; Calculating depth based on the disparity and a camera baseline; Convert the depth into three-dimensional coordinates in a camera coordinate system using camera parameters; Point cloud data is generated using the different feature points and the three-dimensional coordinates, and the point cloud data is used to perform three-dimensional reconstruction to obtain the three-dimensional shape and three-dimensional size of the target material.

5. The automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism according to claim 4 is characterized in that: The calculation formula of the parallax is expressed as: d=x L -x R Where d is the parallax, x is L 、x R are the horizontal coordinates of the left and right contour images respectively.

6. The automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism according to claim 4 is characterized in that: The calculation formula of the depth is expressed as: Where Z is the depth, f is the focal length of the camera, and B is the baseline of the camera; The three-dimensional abscissa and ordinate are calculated according to the depth, and the formula is expressed as: In the formula, c x 、c y are the principal point coordinates of the camera, y L are the ordinates of the left contour image respectively; The final three-dimensional coordinates are expressed as: Where P is the three-dimensional coordinate of the feature point.

7. The automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism according to claim 1 is characterized in that: The prediction process of the feed rate prediction module includes: A three-dimensional coordinate system is established with the rotation center of the cantilever beam as the origin, the travel direction as the X-axis, and the direction perpendicular to the ground as the Z-axis, and the bucket wheel excavator and the target material are mapped to the three-dimensional coordinate system; The ideal state data of the bucket wheel at the next moment is calculated based on the current driving data and the current state data. The formula is expressed as: C t+1 =f(Q t ,C t ) In the formula, C t+1 is the bucket wheel machine state data at the next moment, f() is the bucket wheel machine motion equation, C t is the current state data, Q t is the current control data; According to the next state data, the distance between the bucket wheel and the target material surface and the linear speed of the bucket wheel are calculated, and then the feed rate at the next moment is calculated. The formula is expressed as: in m =Wsv l Where W is the bucket wheel width, s is the distance between the bucket wheel edge and the target material surface, and v l is the linear speed of the bucket wheel.

8. The automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism according to claim 1 is characterized in that: The automatic deviation correction module uses an AC network model to adjust the driving data of the bucket wheel machine to achieve automatic deviation correction of the cantilever beam. The construction process of the AC network model includes: Construct an Actor network. The formula for generating control actions by the Actor network is expressed as: Q t =π(C t ,the actor ) Where π represents the neural network that approximates the Actor policy network, and θactor is the parameter of the Actor network; Construct a Critic network, the value function of which is expressed as follows: Where τ is the parameter of the Critic network.

9. The automatic deviation correction system for the cantilever beam mechanism of a bucket wheel excavator according to claim 8, characterized in that: The automatic correction module uses the TD error to update the value function of the Critic network, which is expressed as: Where α is the learning rate of the Critic network; The parameters of the Actor network are updated according to the evaluation of the Critic network. The formula is expressed as: Where β is the learning rate of the Actor network.

10. The automatic deviation correction system of the bucket wheel excavator cantilever beam mechanism according to claim 1, characterized in that: The controller further comprises a deviation analysis module, which is used for analyzing the cause of the deviation of the cantilever beam when there is a deviation in the position of the cantilever beam, and uploading the deviation data and the cause of the deviation to a log.

Images