Control method for inserting casing pipe into pipe pile

Abstract

The invention relates to the technical field of control, and discloses a control method for inserting a casing into a pipe pile, which comprises the following steps: determining an evaluation coefficient of a target casing according to a target weight and a target length, and determining an initial descending speed of the target casing based on the evaluation coefficient; comparing the current environment data with the historical data, judging whether to correct the initial descent speed according to a comparison result, judging based on the environment similarity to obtain a speed correction coefficient according to the historical data, or determining an environment set according to a clustering algorithm to obtain the speed correction coefficient; the initial descending speed is corrected based on the speed correction coefficient, and the target descending speed of the target casing pipe is determined; performing multi-angle photographing on the target casing pipe and the pipe pile to determine a real-time image, and judging whether the target casing pipe meets the condition of inserting the pipe pile or not according to the real-time image; obtaining a center point descending coordinate of the target casing based on a random forest model; the method has the characteristics of good dynamic adjustment and accurate insertion of the pipe pile.

Description

Technical Field

[0001] The present invention relates to the field of control technology, and in particular to a control method for inserting a casing into a pipe pile. Background Art

[0002] With the continuous development of engineering, the matching operation of casing and pipe piles plays an important role in construction and infrastructure projects. During the operation, the casing will be prefabricated to play a role in shear resistance. The accuracy, alignment effect and insertion speed of the prefabricated casing when inserting it into the pipe pile are important factors to ensure the quality of the building. However, there are certain limitations in the operation of inserting pipe piles.

[0003] In the operation of inserting pipe piles, due to manual operation or lack of fine control mechanism of mechanical system, the casing may deviate from the predetermined insertion position during the insertion of the pipe pile, resulting in inaccurate alignment between the casing and the pipe pile, and the inability to complete the insertion operation, thus affecting the stability and safety of the overall structure. In addition, repeated adjustment of the insertion operation not only increases the construction time, but also affects the construction efficiency. Secondly, during the insertion of the casing, dynamic changes in the environment will affect the speed of the casing insertion. In traditional insertion control, dynamic adjustments cannot be made according to real-time changes in the environment, so that an accurate descent speed cannot be obtained.

[0004] Therefore, how to provide a control method for inserting a casing into a pipe pile is a technical problem that those skilled in the art are in urgent need of solving. Summary of the invention

[0005] In view of this, the present invention proposes a control method for inserting a casing into a pipe pile, aiming to solve the problems of being unable to flexibly adjust the insertion speed according to environmental changes and being unable to accurately align during the insertion process.

[0006] The present invention provides a control method for inserting a casing into a pile, comprising:

[0007] The casing is welded to one end of the end plate, and the other end of the end plate is welded to the lattice angle steel to obtain a steel column, a seamless steel pipe is sleeved on the steel column to obtain a target casing, a target weight and a target length of the target casing are obtained, an evaluation coefficient of the target casing is determined according to the target weight and the target length, and an initial descent speed of the target casing is determined based on the evaluation coefficient;

[0008] Acquire current environment data, compare the current environment data with historical data, determine whether to correct the initial descent speed according to the comparison result, and when it is determined that the initial descent speed is to be corrected, compare the current environment data with the historical data, determine the environment similarity according to the comparison result, determine the speed correction coefficient according to the historical data based on the environment similarity, or determine the environment set according to the clustering algorithm to obtain the speed correction coefficient, correct the initial descent speed based on the speed correction coefficient, and determine the target descent speed of the target casing;

[0009] Taking photos of the target casing and the pipe pile at multiple angles to determine real-time images, the real-time images including the target casing image and the pipe pile image, and judging whether the target casing meets the conditions for inserting the pipe pile according to the real-time images;

[0010] Acquire the mapped insertion alignment point according to the target casing image, acquire the mapped landing alignment point according to the pipe pile image, determine the descent offset value of the target casing based on the insertion alignment point and the landing alignment point, and obtain the descent coordinate of the center point of the target casing based on the random forest model according to the current environment data, the descent offset value and the target length;

[0011] The target casing is inserted into the pipe pile according to the central point descent coordinates and the target descent speed.

[0012] Further, when obtaining the target weight and target length of the target casing and determining the evaluation coefficient of the target casing according to the target weight and the target length, it includes:

[0013] Obtaining the initial weight of the target casing several times, summing the initial weights several times and taking the average value to obtain the target weight, obtaining the initial length of the target casing several times, summing the initial lengths several times and taking the average value to obtain the target length;

[0014] The evaluation coefficient is obtained by the following formula:

[0015] ;

[0016] in, represents the evaluation coefficient, represents the target weight, represents the target length, and represents the weight coefficient, and .

[0017] Further, when determining the initial descending speed of the target casing based on the evaluation coefficient, it includes:

[0018] Presetting a first preset speed, a second preset speed, and a third preset speed;

[0019] when When it is less than or equal to 80, the first preset speed is used as the initial descending speed of the target casing;

[0020] when When the value is greater than 80 and less than or equal to 100, the second preset speed is used as the initial descending speed of the target casing;

[0021] when When it is greater than 100, the third preset speed is used as the initial descending speed of the target casing;

[0022] The first preset speed is greater than the second preset speed, and the second preset speed is greater than the third preset speed.

[0023] Further, when obtaining current environment data, comparing the current environment data with historical data, and judging whether to correct the initial descent speed according to the comparison result, it includes:

[0024] Acquire wind speed data of the current environment according to the airflow sensor, and preprocess the wind speed data, wherein the preprocessing includes data cleaning and data standardization, determine the ambient wind speed value of the current environment based on the result of the preprocessing, and acquire the ambient wind direction of the current environment, wherein the current environment data includes the ambient wind speed value and the ambient wind direction;

[0025] The historical data includes historical qualified wind speed values, historical primary environmental wind speed values, historical primary environmental wind directions and historical speed correction coefficients, and the historical primary environmental wind speed values, the historical primary environmental wind directions and the historical speed correction coefficients correspond to each other, and the environmental wind speed values ​​are compared with the historical qualified wind speed values ​​in the historical data;

[0026] When the ambient wind speed value is greater than the historical qualified wind speed value, it is determined that the initial descent speed is to be corrected;

[0027] When the environmental wind speed value is less than or equal to the historical qualified wind speed value, it is determined that the initial descent speed is not to be corrected, and the initial descent speed is determined as the target descent speed of the target casing.

[0028] Further, when comparing the current environment data with the historical data, determining the environment similarity according to the comparison result, deriving the speed correction coefficient according to the historical data based on the environment similarity, or determining the environment set according to the clustering algorithm to derive the speed correction coefficient, it includes:

[0029] The environmental similarity is determined by comparing the environmental wind speed value, the environmental wind direction and the historical data. The environmental similarity is obtained by the following formula:

[0030] ;

[0031] in, Indicates the environmental similarity, Indicates the ambient wind speed value. Indicates the historical ambient wind speed value. Indicates the ambient wind direction. Indicates the historical environmental wind direction;

[0032] Setting a similarity threshold, and comparing the environment similarity with the similarity threshold, and determining the speed correction coefficient according to the comparison result;

[0033] When there is data in the historical data whose environmental similarity is greater than the similarity threshold, the speed correction coefficient is determined according to the historical speed correction coefficient corresponding to the maximum environmental similarity;

[0034] When there is no data with an environment similarity greater than a similarity threshold in the historical data, the speed correction coefficient is obtained by determining an environment set according to a clustering algorithm.

[0035] Furthermore, when the speed correction coefficient is obtained by determining the environment set according to the clustering algorithm, the initial descent speed is corrected based on the speed correction coefficient to determine the target descent speed of the target casing, the method includes:

[0036] The ambient wind speed value, the ambient wind direction and the historical data are taken as an initial data set, the historical speed correction coefficient corresponding to each data in the initial data set is extracted, and the cluster number k is set to 2, the parameters of the Gaussian distribution are initialized, the probability that each data in the initial data set belongs to each Gaussian distribution is calculated, and the responsibility value is obtained, and the target data set corresponding to the ambient wind speed value and the ambient wind direction is obtained according to the responsibility value, and the target data set is taken as an environment set, and based on the environment set, the mean value of the historical speed correction coefficient in the environment set is determined as the speed correction coefficient;

[0037] The target descending speed of the target casing is a product value of the initial descending speed and the speed correction coefficient.

[0038] Furthermore, the target casing and the pipe pile are photographed at multiple angles to determine a real-time image, wherein the real-time image includes the target casing image and the pipe pile image, including:

[0039] The target casing and the pipe pile are photographed at multiple angles respectively, and the obtained multi-angle images are processed, wherein the image processing includes denoising, geometric correction and contrast adjustment, feature points are extracted from the multi-angle images after image processing according to an image algorithm, matching information between the multi-angle images is determined, splicing processing is performed according to the matching information, an alignment area is determined, and all the multi-angle images are combined according to the alignment area and the shooting angle of each multi-angle image to obtain the target casing image and the pipe pile image.

[0040] Further, when judging whether the target casing meets the condition for inserting the pipe pile according to the real-time image, it includes:

[0041] Based on a plurality of insertion alignment points pre-deployed on the target casing and a plurality of placement alignment points of the pipe pile, all the insertion alignment points are mapped to the target casing image, and all the placement alignment points are mapped to the pipe pile image;

[0042] When the mapped insertion alignment point and the landing alignment point completely overlap, it is determined that the target casing meets the conditions for being inserted into the pile;

[0043] When the mapped insertion alignment point and the landing alignment point do not completely overlap, it is determined that the target casing does not meet the condition for inserting the pipe pile.

[0044] Further, when determining the descent offset value of the target casing based on the insertion alignment point and the landing alignment point, it includes:

[0045] Determine the intervals between all insertion alignment points and placement alignment points, and establish an insertion placement data set based on all the intervals;

[0046] Extract a maximum phase distance and a minimum phase distance from the positioning data set, sum and average the maximum phase distance and the minimum phase distance, and obtain the insertion edge value ;

[0047] When the phase distance is greater than or equal to the insertion edge value When , the corresponding phase distance decreases according to the following formula:

[0048] ;

[0049] in, For the The decrease deviation of the phase distance is For the The phase distance, It represents the sum of all interphase distances in the insertion location dataset;

[0050] When the phase distance is less than the insertion edge value When , the corresponding phase distance decreases according to the following formula:

[0051] ;

[0052] in, The decrease deviation of the phase distance is For the The phase distance, It represents the sum of all interphase distances in the insertion location dataset;

[0053] statistics The number is recorded as the first offset, and the statistics The number is recorded as the second offset, and the falling offset value is determined according to the first offset and the second offset. The falling offset value is obtained by the following formula:

[0054] ;

[0055] in, Indicates the falling offset value, Indicates the first offset, Indicates the second offset.

[0056] Further, when obtaining the descent coordinates of the center point of the target casing based on the random forest model according to the current environment data, the descent offset value and the target length, it includes:

[0057] Establishing a three-dimensional coordinate system with the center point of the pipe pile;

[0058] Acquire historical offset data from the descending offset value, acquire historical environmental data from the environmental wind speed value and the environmental wind direction, and acquire historical length data from the target length;

[0059] The historical offset data, the historical environment data and the historical length data are used as training sets, and based on the three-dimensional coordinate system, cross-validation is used in combination with grid search to find establishment parameters of a random forest model, a random forest model is established, and the training set is used to train and fit the random forest model;

[0060] The descent coordinates of the center point of the target casing are acquired according to the current environment data, the descent offset value and the target length.

[0061] Compared with the prior art, the beneficial effects of the present invention are as follows: an evaluation coefficient is obtained through the target weight and target length of the target casing, an initial descent speed is determined based on the evaluation coefficient, and a comparison is made based on the current environmental data and the historical data, so as to dynamically correct the initial descent speed, avoid the risk of inaccurate insertion of the pipe pile due to speed mismatch and environmental changes, and improve the reliability and accuracy of the insertion process; a speed correction coefficient is obtained based on historical data based on environmental similarity judgment, or a speed correction coefficient is obtained based on an environmental set determined by a clustering algorithm, so as to ensure that the target casing is always descended at an accurate speed under different environmental conditions, and the unstable factors caused by environmental fluctuations are avoided; by taking photos of the target casing and the pipe pile at multiple angles, obtaining real-time images and analyzing them, it is possible to accurately judge whether the target casing meets the docking requirements for inserting the pipe pile; the descent offset value is calculated based on the insertion alignment point after mapping obtained from the target casing image, and the placement alignment point after mapping obtained from the pipe pile image; and then combined with the current environmental data, a random forest model is used for accurate prediction to ensure that the accurate center point descent coordinates of the target casing are obtained, and repeated adjustments and errors in the insertion operation are reduced, thereby improving the construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0063] Figure 1 A flow chart of a control method for inserting a casing into a pipe pile provided by an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of the structure of a steel column provided in an embodiment of the present invention;

[0065] Figure 3 A schematic diagram of the structure of a seamless steel pipe provided in an embodiment of the present invention;

[0066] Figure 4 A schematic structural diagram of a target sleeve provided in an embodiment of the present invention.

[0067] In the figure: 1. casing; 2. end plate; 3. lattice angle steel; 10. seamless steel pipe. DETAILED DESCRIPTION

[0068] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0069] See also Figure 1-4 As shown, in some embodiments of the present application, a method for controlling a casing to be inserted into a pile includes:

[0070] S100: welding the casing 1 to one end of the end plate 2, and welding the other end of the end plate 2 to the lattice angle steel 3 to obtain a steel column, sleeve the seamless steel pipe 10 on the steel column to obtain a target casing, obtain a target weight and a target length of the target casing, determine an evaluation coefficient of the target casing according to the target weight and the target length, and determine an initial descent speed of the target casing based on the evaluation coefficient;

[0071] S200: Acquire current environment data, compare the current environment data with historical data, and determine whether to correct the initial descent speed according to the comparison result. When it is determined that the initial descent speed is to be corrected, compare the current environment data with the historical data, determine the environment similarity according to the comparison result, determine the speed correction coefficient according to the historical data based on the environment similarity, or determine the environment set according to the clustering algorithm to obtain the speed correction coefficient, correct the initial descent speed based on the speed correction coefficient, and determine the target descent speed of the target casing;

[0072] S300: taking photos of the target casing and the pipe pile at multiple angles to determine a real-time image, the real-time image including the target casing image and the pipe pile image, and judging whether the target casing meets the condition for inserting the pipe pile according to the real-time image;

[0073] S400: acquiring the mapped insertion alignment point according to the target casing image, acquiring the mapped landing alignment point according to the pipe pile image, determining the descent offset value of the target casing based on the insertion alignment point and the landing alignment point, and obtaining the descent coordinates of the center point of the target casing based on the random forest model according to the current environmental data, the descent offset value and the target length;

[0074] S500: inserting the target casing into the pile according to the center point descent coordinates and the target descent speed.

[0075] Specifically, in S100, a steel column is obtained by welding, and a seamless steel pipe 10 is set on the steel column to obtain a target casing. Since the lattice angle steel 3 is a square structure and is not easy to hoist, a seamless steel pipe 10 is installed on its outside to form a target casing, which is convenient for hoisting during construction. The evaluation coefficient is calculated for the target weight and target length of the target casing. The evaluation coefficient reflects the difficulty of inserting the target casing. The higher the evaluation coefficient, the more difficult it is to insert the pipe pile, so that the initial descent speed when it is descended needs to be in a low speed state. In S200, the change of the current environmental data will have a certain impact on the descent speed of the target casing. By obtaining the current environmental data in real time and comparing it with the historical data, it is determined whether the initial descent speed needs to be corrected. The historical data provides environmental data at different insertion periods. According to the comparison of the two, the environmental similarity can be determined. In addition, the speed correction coefficient is obtained based on historical data according to the environmental similarity, or the speed correction coefficient is obtained by determining the environmental set according to the clustering algorithm, which ensures the adaptability to the similarity of different environments and improves the reliability and accuracy of the correction of the initial descent speed. In S300, the real-time image is determined by taking photos from multiple angles, and the multiple angles include six angles such as up, down, left, right, front and back, so as to obtain a comprehensive and detailed position status of the target casing and the pile. Through the real-time image, it can be evaluated whether the target casing meets the conditions for inserting the pile. In S400, the insertion alignment point after mapping is obtained according to the target casing image, and the landing alignment point after mapping is obtained according to the pile image. The insertion alignment point refers to the position on the target casing that remains stationary after the target casing enters the pile, and the landing alignment point is the position inside the pile corresponding to the insertion alignment point. The descent offset value of the target casing is determined according to the insertion alignment point and the landing alignment point, and the center point descent coordinates of the target casing are obtained based on the random forest model according to the current environmental data, the descent offset value and the target length. The random forest model is a deep learning model that can learn the current environmental data, the descent offset value and the target length to obtain the center point descent coordinates of the target casing. In S500, the target casing is accurately inserted into the pile according to the center point descent coordinates and the target descent speed.

[0076] It can be understood that by calculating the initial descent speed of the target casing and correcting it, the influence of environmental changes on the descent speed can be dynamically responded to, and the alignment accuracy between the target casing and the pile is ensured by the center point descent coordinates, which improves the safety and reliability of the entire operation process, avoids repeated insertion operations, and thus improves efficiency.

[0077] In some embodiments of the present application, when obtaining the target weight and target length of the target casing and determining the evaluation coefficient of the target casing according to the target weight and target length, the process includes:

[0078] The initial weight of the target casing is obtained several times, and the initial weights are summed and averaged to obtain the target weight; the initial length of the target casing is obtained several times, and the initial lengths are summed and averaged to obtain the target length;

[0079] The evaluation coefficient is obtained by the following formula:

[0080] ;

[0081] in, represents the evaluation coefficient, represents the target weight, represents the target length, and represents the weight coefficient, and .

[0082] In some embodiments of the present application, when determining the initial descending speed of the target casing based on the evaluation coefficient, the method includes:

[0083] Presetting a first preset speed, a second preset speed, and a third preset speed;

[0084] when When it is less than or equal to 80, the first preset speed is used as the initial descending speed of the target casing;

[0085] when When the value is greater than 80 and less than or equal to 100, the second preset speed is used as the initial descending speed of the target casing;

[0086] when When it is greater than 100, the third preset speed is used as the initial descending speed of the target casing;

[0087] The first preset speed is greater than the second preset speed, and the second preset speed is greater than the third preset speed.

[0088] Specifically, by obtaining the initial weight of the target casing several times and averaging the target weight, and obtaining the initial length of the target casing several times and averaging the target length, the error or fluctuation of a single measurement is reduced, the deviation caused by individual measurement abnormalities is avoided, and the reliability and accuracy of the target weight and target length are improved. The several times is preferably five times, which can be adjusted according to actual needs. According to different evaluation coefficients, the preset speed is dynamically selected to adapt to different insertion situations. For example: when the target weight is 2000kg and the target length is 15m, and When both are 0.5, the evaluation coefficient is 102. At this time, the third preset speed is used as the initial descent speed of the target casing. When the target weight is 1500kg and the target length is 15m, and When 0.5 is selected for both, the evaluation coefficient is 92. At this time, the second preset speed is used as the initial descent speed of the target casing. By setting multiple preset speeds, dynamic selection is effectively performed for the target weight and target length, thereby avoiding the risks of excessive speed and speed mismatch, improving the efficiency of inserting the target casing into the pile, and enhancing stability and reliability.

[0089] In some embodiments of the present application, when obtaining current environmental data, comparing the current environmental data with historical data, and judging whether to correct the initial descent speed according to the comparison result, the method includes:

[0090] The wind speed data of the current environment is obtained according to the airflow sensor, and the wind speed data is preprocessed, the preprocessing includes data cleaning and data standardization, the ambient wind speed value of the current environment is determined based on the result of the preprocessing, and the ambient wind direction of the current environment is obtained, wherein the current environment data includes the ambient wind speed value and the ambient wind direction;

[0091] The historical data includes historical qualified wind speed values, historical primary environmental wind speed values, historical primary environmental wind directions and historical speed correction coefficients, and the historical primary environmental wind speed values, historical primary environmental wind directions and historical speed correction coefficients correspond to each other, and the environmental wind speed values ​​are compared with the historical qualified wind speed values ​​in the historical data;

[0092] When the ambient wind speed value is greater than the historical qualified wind speed value, it is determined that the initial descent speed should be corrected;

[0093] When the ambient wind speed value is less than or equal to the historical qualified wind speed value, it is determined that the initial descent speed is not to be corrected, and the initial descent speed is determined as the target descent speed of the target casing.

[0094] Specifically, during the insertion of the target casing into the pile, it will be affected by the local wind force, thereby interfering with the initial descent speed during insertion, so the initial descent speed needs to be corrected to reduce the interference of the wind force on the initial descent speed. The wind speed data of the current environment is obtained according to the airflow sensor, and the wind speed data is preprocessed. Data cleaning helps to remove abnormal noise and ensure the accuracy and reliability of the ambient wind speed value. For example, abnormal readings such as extreme values ​​or erroneous data caused by airflow sensor failure are eliminated. Data standardization helps to eliminate the influence of different wind speed measurement units or magnitudes, and facilitates the integration of data sources. Through data cleaning and data standardization, the interference caused by data fluctuations is reduced, thereby accurately determining the ambient wind speed value and ambient wind direction of the current environment.

[0095] It is understandable that the historical qualified wind speed value can be set according to the local meteorological wind speed and the required wind speed during the operation, and it can be used as a judgment standard, and its specific value can be adaptively changed in actual application. By introducing the historical qualified wind speed value as a judgment standard, the accuracy of the correction of the initial descent speed is improved. When the ambient wind speed value is greater than the historical qualified wind speed value, it means that it exceeds the requirements of the operation. When the ambient wind speed value is not greater than the historical qualified wind speed value, it means that the current environment meets the operation requirements, and the target casing can be inserted into the pile according to the construction requirements. Determining the historical qualified wind speed value and using it as a judgment standard enhances the adaptability to different environments and improves the reliability of inserting the target casing into the pile.

[0096] In some embodiments of the present application, when comparing the current environment data with the historical data, determining the environment similarity according to the comparison result, deriving the speed correction coefficient according to the historical data based on the environment similarity, or determining the environment set according to the clustering algorithm to derive the speed correction coefficient, it includes:

[0097] The environmental similarity is determined by comparing the ambient wind speed value and ambient wind direction with historical data. The environmental similarity is obtained by the following formula:

[0098] ;

[0099] in, Indicates the environmental similarity, Indicates the ambient wind speed value. Indicates the historical ambient wind speed value. Indicates the ambient wind direction. Indicates the historical environmental wind direction;

[0100] Setting a similarity threshold, and comparing the environment similarity with the similarity threshold, and determining a speed correction coefficient according to the comparison result;

[0101] When there is data in the historical data with an environmental similarity greater than the similarity threshold, the speed correction coefficient is determined according to the historical speed correction coefficient corresponding to the maximum environmental similarity;

[0102] When there is no data with an environment similarity greater than a similarity threshold in the historical data, the environment set is determined according to the clustering algorithm to obtain a speed correction coefficient.

[0103] In some embodiments of the present application, when determining the environment set according to the clustering algorithm to obtain the speed correction coefficient, correcting the initial descent speed based on the speed correction coefficient, and determining the target descent speed of the target casing, the method includes:

[0104] The ambient wind speed value, ambient wind direction and historical data are taken as the initial data set, the historical speed correction coefficient corresponding to each data in the initial data set is extracted, and the cluster number k is set to 2, the parameters of the Gaussian distribution are initialized, and the probability that each data in the initial data set belongs to each Gaussian distribution is calculated to obtain the responsibility value, and the target data set corresponding to the ambient wind speed value and ambient wind direction is obtained according to the responsibility value. The target data set is taken as the environmental set, and the mean value of the historical speed correction coefficient in the environmental set is determined as the speed correction coefficient based on the environmental set;

[0105] The target descent speed of the target casing is the product of the initial descent speed and the speed correction coefficient.

[0106] Specifically, by judging the matching degree between the current environmental data and the historical data through the similarity threshold, the historical speed correction coefficient corresponding to the maximum environmental similarity can be effectively utilized and used as the speed correction coefficient, thereby ensuring the reliability and consistency of the correction process and meeting the conditions for historical correction. The similarity threshold is preferably 0.7. If the matching degree between the current environmental data and the historical data is low, the historical data is analyzed through a clustering algorithm to improve the accuracy of the correction of the initial descent speed. The comprehensive use of massive historical data ensures the accuracy of the speed correction coefficient. Moreover, by continuously accumulating historical data, the corresponding correction scheme can be found when unknown situations occur, thereby improving the efficiency of inserting the target casing into the pile.

[0107] It is understandable that by analyzing historical data through clustering algorithms, the environment set closest to the current environment is found, the accuracy of the speed correction coefficient is guaranteed, and the operation stability of the target casing inserted into the pile is ensured under different environmental wind speed values ​​and environmental wind directions. Whether the historical speed correction coefficient corresponding to the maximum environmental similarity is used as the speed correction coefficient, or the speed correction coefficient obtained by the clustering algorithm, the initial descent speed can be accurately corrected. By multiplying the speed correction coefficient and the initial descent speed, it is ensured that after the initial descent speed is determined, it can be dynamically corrected according to the speed correction coefficient, thereby ensuring the stability and reliability of the target casing inserted into the pile.

[0108] In some embodiments of the present application, the target casing and the pipe pile are photographed at multiple angles to determine a real-time image, and the real-time image includes the target casing image and the pipe pile image, including:

[0109] The target casing and pipe pile are photographed at multiple angles respectively, and the obtained multi-angle images are processed, including denoising, geometric correction and contrast adjustment. Feature points are extracted from the multi-angle images after image processing according to the image algorithm, and matching information between the multi-angle images is determined. Splicing processing is performed based on the matching information to determine the alignment area. According to the alignment area and the shooting angle of each multi-angle image, all the multi-angle images are combined to obtain the target casing image and pipe pile image.

[0110] Specifically, denoising in image processing is achieved by Gaussian filtering or median filtering, and geometric correction is used to remove geometric distortion caused by factors such as lens distortion or photographing errors, thereby ensuring that the geometric shape of the multi-angle image conforms to the current actual state of the target casing and pipe pile, thereby improving the accuracy of the multi-angle image. Through contrast adjustment, the difference between the bright and dark parts of the multi-angle image is improved, and the details of the target casing and pipe pile in the multi-angle image are improved, laying the foundation for subsequent stitching processing. Through the SIFT (Scale Invariant Feature Transform) and RANSAC algorithms in the image algorithm, feature points can be extracted from the multi-angle image after image processing, thereby determining the matching information between the multi-angle images, such as the relative position of the target casing and pipe pile. According to the matching information, multi-band fusion is used for stitching processing to determine the alignment area, and then all multi-angle images are combined to obtain the target casing image and pipe pile image.

[0111] It can be understood that by taking multi-angle photos of the target casing and pipe piles and performing a series of image processing on the acquired multi-angle images, a clear and accurate image basis is provided for the subsequent feature point extraction. Multi-band fusion is used for splicing processing based on the matching information, which improves the availability and integrity of the target casing and pipe pile images and provides reliable image data for subsequent analysis.

[0112] In some embodiments of the present application, when judging whether the target casing meets the condition for inserting the pipe pile according to the real-time image, it includes:

[0113] Based on a plurality of insertion alignment points pre-deployed on the target casing and a plurality of placement alignment points of the pipe pile, all the insertion alignment points are mapped to the target casing image, and all the placement alignment points are mapped to the pipe pile image;

[0114] When the mapped insertion alignment point and the landing alignment point completely overlap, it is determined that the target casing meets the conditions for inserting the pile;

[0115] When the mapped insertion alignment point and the landing alignment point do not completely overlap, it is determined that the target casing does not meet the conditions for inserting the pipe pile.

[0116] Specifically, based on multiple insertion alignment points pre-deployed on the target casing and multiple landing alignment points of the pipe pile, the number of insertion alignment points is preferably fifteen, and the number of landing alignment points is preferably fifteen. The specific number of deployments can be adjusted according to actual needs, and whether the target casing meets the conditions for inserting the pipe pile is judged based on the overlap, thereby improving the reliability and accuracy of the target casing inserting the pipe pile. When the conditions for inserting the pipe pile are met, it indicates that the target casing at this time can just complete the insertion of the pipe pile, and there is no scratch on the wall of the pipe pile. The insertion operation can be completed by lowering the target casing according to the target descent speed. When the conditions for inserting the pipe pile are not met, it is necessary to determine the coordinates of the target casing based on the deviation that occurs during the descent process, so as to ensure the accurate insertion of the pipe pile.

[0117] In some embodiments of the present application, when determining the descent offset value of the target casing based on the insertion alignment point and the landing alignment point, the method includes:

[0118] Determine the intervals between all insertion alignment points and placement alignment points, and establish an insertion placement data set based on all the intervals;

[0119] Extract a maximum phase distance and a minimum phase distance from the placement data set, sum the maximum phase distance and the minimum phase distance and take the average to obtain the insertion edge value ;

[0120] When the phase distance is greater than or equal to the insertion edge value When , the corresponding phase distance decreases according to the following formula:

[0121] ;

[0122] in, For the The decrease deviation of the phase distance is For the The phase distance, It represents the sum of all interphase distances in the insertion location dataset;

[0123] When the phase distance is less than the insertion edge value When , the corresponding phase distance decreases according to the following formula:

[0124] ;

[0125] in, The decrease deviation of the phase distance is For the The phase distance, It represents the sum of all interphase distances in the insertion location dataset;

[0126] statistics The number is recorded as the first offset, and the statistics The number is recorded as the second offset, and the drop offset value is determined according to the first offset and the second offset. The drop offset value is obtained by the following formula:

[0127] ;

[0128] in, Indicates the falling offset value, Indicates the first offset, Indicates the second offset.

[0129] In some embodiments of the present application, when obtaining the descent coordinates of the center point of the target casing based on the random forest model according to the current environment data, the descent offset value and the target length, it includes:

[0130] Establish a three-dimensional coordinate system based on the center point of the pipe pile;

[0131] Obtain historical offset data in the descent offset value, obtain historical environmental data in the ambient wind speed value and ambient wind direction, and obtain historical length data in the target length;

[0132] The historical offset data, historical environmental data and historical length data are used as training sets, and based on the three-dimensional coordinate system, cross-validation and grid search are used to find the establishment parameters of the random forest model, and the random forest model is established. The training set is used to train and fit the random forest model.

[0133] According to the current environment data, the descent offset value and the target length, the descent coordinates of the center point of the target casing are obtained.

[0134] Specifically, the descent offset value of the target casing is determined according to the first offset and the second offset, and the deviation of the target casing from the pipe pile during the descent process can be obtained. The descent offset value lays a data foundation for the subsequent accurate descent coordinates of the center point. The historical offset data is obtained from the descent offset value, the historical environmental data is obtained from the ambient wind speed value and the ambient wind direction, and the historical length data is obtained from the target length. These data record the actual situation of the target casing inserted into the pipe pile at different times. The historical offset data, historical environmental data, and historical length data are used as training sets to ensure the generalization ability of the random forest model. Cross-validation divides the data into several parts and trains the model multiple times to verify its stability and performance. Grid search exhaustively searches the establishment parameters of the random forest model in the parameter space, such as: tree branches, number of samples of leaf nodes, etc.

[0135] It can be understood that using the training set for training and fitting the random forest model reduces the risk of overfitting and improves the reliability of the random forest model. According to the trained random forest model, the center point descent coordinates of the target casing can be obtained through the random forest model according to the current environmental data, descent offset value and target length. The center point descent coordinates of the target casing reflect the relative position coordinates when aligned with the center point of the pile. According to the center point descent coordinates and the obtained target descent speed, the target casing can be accurately inserted into the pile, thereby improving the reliability and efficiency of the target casing inserted into the pile.

[0136] In summary, the beneficial effects of the present invention are as follows: an evaluation coefficient is obtained through the target weight and target length of the target casing, an initial descent speed is determined based on the evaluation coefficient, and a comparison is made based on the current environmental data and the historical data, so as to dynamically correct the initial descent speed, thereby avoiding the risk of inaccurate insertion of the pipe pile due to speed mismatch and environmental changes, and improving the reliability and accuracy of the insertion process. A speed correction coefficient is obtained based on historical data based on environmental similarity judgment, or a speed correction coefficient is obtained based on an environmental set determined by a clustering algorithm, so as to ensure that the target casing is always descended at an accurate speed under different environmental conditions, and the unstable factors caused by environmental fluctuations are avoided. By taking photos of the target casing and the pipe pile from multiple angles, obtaining real-time images and analyzing them, it is possible to accurately determine whether the target casing meets the docking requirements for inserting the pipe pile, and calculate the descent offset value based on the insertion alignment point after mapping obtained from the target casing image and the placement alignment point after mapping obtained from the pipe pile image, and then combined with the current environmental data, a random forest model is used for accurate prediction to ensure that the accurate center point descent coordinates of the target casing are obtained, and repeated adjustments and errors in the insertion operation are reduced, thereby improving the construction efficiency.

[0137] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0138] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0139] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0140] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0141] 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 above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A control method for inserting a casing into a pile, characterized in that: include: The casing is welded to one end of the end plate, and the other end of the end plate is welded to the lattice angle steel to obtain a steel column, a seamless steel pipe is sleeved on the steel column to obtain a target casing, a target weight and a target length of the target casing are obtained, an evaluation coefficient of the target casing is determined according to the target weight and the target length, and an initial descent speed of the target casing is determined based on the evaluation coefficient; Acquire current environment data, compare the current environment data with historical data, determine whether to correct the initial descent speed according to the comparison result, and when it is determined that the initial descent speed is to be corrected, compare the current environment data with the historical data, determine the environment similarity according to the comparison result, determine the speed correction coefficient according to the historical data based on the environment similarity, or determine the environment set according to the clustering algorithm to obtain the speed correction coefficient, correct the initial descent speed based on the speed correction coefficient, and determine the target descent speed of the target casing; Taking photos of the target casing and the pipe pile at multiple angles to determine real-time images, the real-time images including the target casing image and the pipe pile image, and judging whether the target casing meets the conditions for inserting the pipe pile according to the real-time images; Acquire the mapped insertion alignment point according to the target casing image, acquire the mapped landing alignment point according to the pipe pile image, determine the descent offset value of the target casing based on the insertion alignment point and the landing alignment point, and obtain the descent coordinate of the center point of the target casing based on the random forest model according to the current environment data, the descent offset value and the target length; The target casing is inserted into the pipe pile according to the central point descent coordinates and the target descent speed.

2. The control method for inserting a casing into a pile according to claim 1, characterized in that: When the target weight and the target length of the target casing are obtained and the evaluation coefficient of the target casing is determined according to the target weight and the target length, the method includes: Obtaining the initial weight of the target casing several times, summing the initial weights several times and taking the average value to obtain the target weight, obtaining the initial length of the target casing several times, summing the initial lengths several times and taking the average value to obtain the target length; The evaluation coefficient is obtained by the following formula: ； in, represents the evaluation coefficient, represents the target weight, represents the target length, and represents the weight coefficient, and .

3. The control method for inserting a casing into a pile according to claim 2, characterized in that: When determining the initial descending speed of the target casing based on the evaluation coefficient, it includes: Presetting a first preset speed, a second preset speed, and a third preset speed; when When it is less than or equal to 80, the first preset speed is used as the initial descending speed of the target casing; when When the value is greater than 80 and less than or equal to 100, the second preset speed is used as the initial descending speed of the target casing; when When it is greater than 100, the third preset speed is used as the initial descending speed of the target casing; The first preset speed is greater than the second preset speed, and the second preset speed is greater than the third preset speed.

4. The control method for inserting a casing into a pile according to claim 3, characterized in that: When obtaining current environment data, comparing the current environment data with historical data, and judging whether to correct the initial descent speed according to the comparison result, the method includes: Acquire wind speed data of the current environment according to the airflow sensor, and preprocess the wind speed data, wherein the preprocessing includes data cleaning and data standardization, determine the ambient wind speed value of the current environment based on the result of the preprocessing, and acquire the ambient wind direction of the current environment, wherein the current environment data includes the ambient wind speed value and the ambient wind direction; The historical data includes historical qualified wind speed values, historical primary environmental wind speed values, historical primary environmental wind directions and historical speed correction coefficients, and the historical primary environmental wind speed values, the historical primary environmental wind directions and the historical speed correction coefficients correspond to each other, and the environmental wind speed values ​​are compared with the historical qualified wind speed values ​​in the historical data; When the ambient wind speed value is greater than the historical qualified wind speed value, it is determined that the initial descent speed is to be corrected; When the environmental wind speed value is less than or equal to the historical qualified wind speed value, it is determined that the initial descent speed is not to be corrected, and the initial descent speed is determined as the target descent speed of the target casing.

5. The control method for inserting a casing into a pile according to claim 4, characterized in that: When comparing the current environment data with the historical data, determining the environment similarity according to the comparison result, deriving a speed correction coefficient according to the historical data based on the environment similarity, or determining an environment set according to a clustering algorithm to derive the speed correction coefficient, it includes: The environmental similarity is determined by comparing the environmental wind speed value, the environmental wind direction and the historical data. The environmental similarity is obtained by the following formula: ； in, Indicates the environmental similarity, Indicates the ambient wind speed value. Indicates the historical ambient wind speed value. Indicates the ambient wind direction. Indicates the historical environmental wind direction; Setting a similarity threshold, and comparing the environment similarity with the similarity threshold, and determining the speed correction coefficient according to the comparison result; When there is data in the historical data whose environmental similarity is greater than the similarity threshold, the speed correction coefficient is determined according to the historical speed correction coefficient corresponding to the maximum environmental similarity; When there is no data with an environment similarity greater than a similarity threshold in the historical data, the speed correction coefficient is obtained by determining an environment set according to a clustering algorithm.

6. The control method for inserting a casing into a pile according to claim 5, characterized in that: When the speed correction coefficient is obtained by determining the environment set according to the clustering algorithm, the initial descent speed is corrected based on the speed correction coefficient, and the target descent speed of the target casing is determined, the method includes: The ambient wind speed value, the ambient wind direction and the historical data are taken as an initial data set, the historical speed correction coefficient corresponding to each data in the initial data set is extracted, and the cluster number k is set to 2, the parameters of the Gaussian distribution are initialized, the probability that each data in the initial data set belongs to each Gaussian distribution is calculated, and the responsibility value is obtained, and the target data set corresponding to the ambient wind speed value and the ambient wind direction is obtained according to the responsibility value, and the target data set is taken as an environment set, and based on the environment set, the mean value of the historical speed correction coefficient in the environment set is determined as the speed correction coefficient; The target descending speed of the target casing is a product value of the initial descending speed and the speed correction coefficient.

7. The control method for inserting a casing into a pile according to claim 6, characterized in that: The target casing and the pipe pile are photographed at multiple angles to determine a real-time image, wherein the real-time image includes a target casing image and a pipe pile image, including: The target casing and the pipe pile are photographed at multiple angles respectively, and the obtained multi-angle images are processed, wherein the image processing includes denoising, geometric correction and contrast adjustment, feature points are extracted from the multi-angle images after image processing according to an image algorithm, matching information between the multi-angle images is determined, splicing processing is performed according to the matching information, an alignment area is determined, and all the multi-angle images are combined according to the alignment area and the shooting angle of each multi-angle image to obtain the target casing image and the pipe pile image.

8. The control method for inserting a casing into a pile according to claim 7, characterized in that: When judging whether the target casing meets the condition for inserting the pipe pile according to the real-time image, the method includes: Based on a plurality of insertion alignment points pre-deployed on the target casing and a plurality of placement alignment points of the pipe pile, all the insertion alignment points are mapped to the target casing image, and all the placement alignment points are mapped to the pipe pile image; When the mapped insertion alignment point and the landing alignment point completely overlap, it is determined that the target casing meets the conditions for being inserted into the pile; When the mapped insertion alignment point and the landing alignment point do not completely overlap, it is determined that the target casing does not meet the condition for inserting the pipe pile.

9. The control method for inserting a casing into a pile according to claim 8, characterized in that: When determining the descent offset value of the target casing based on the insertion alignment point and the landing alignment point, the method includes: Determine the intervals between all insertion alignment points and placement alignment points, and establish an insertion placement data set based on all the intervals; Extract a maximum phase distance and a minimum phase distance from the positioning data set, sum and average the maximum phase distance and the minimum phase distance, and obtain the insertion edge value ; When the phase distance is greater than or equal to the insertion edge value When , the corresponding phase distance decreases according to the following formula: ； in, For the The decrease deviation of the phase distance is For the The phase distance, It represents the sum of all interphase distances in the insertion location dataset; When the phase distance is less than the insertion edge value When , the corresponding phase distance decreases according to the following formula: ； in, The decrease deviation of the phase distance is For the The phase distance, It represents the sum of all interphase distances in the insertion location dataset; statistics The number is recorded as the first offset, and the statistics The number is recorded as the second offset, and the falling offset value is determined according to the first offset and the second offset. The falling offset value is obtained by the following formula: ； in, Indicates the falling offset value, Indicates the first offset, Indicates the second offset.

10. The control method for inserting a casing into a pile according to claim 9, characterized in that: When obtaining the descent coordinates of the center point of the target casing based on the random forest model according to the current environment data, the descent offset value and the target length, the method includes: Establishing a three-dimensional coordinate system with the center point of the pipe pile; Acquire historical offset data from the descending offset value, acquire historical environmental data from the environmental wind speed value and the environmental wind direction, and acquire historical length data from the target length; The historical offset data, the historical environment data and the historical length data are used as training sets, and based on the three-dimensional coordinate system, cross-validation is used in combination with grid search to find establishment parameters of a random forest model, a random forest model is established, and the training set is used to train and fit the random forest model; The descent coordinates of the center point of the target casing are acquired according to the current environment data, the descent offset value and the target length.

Images