Intelligent positioning method for tower crane selection and site visual arrangement

Abstract

The application is a tower crane selection and site visual arrangement intelligent positioning method, which greatly improves the work efficiency of personnel preparing a tower crane scheme by visualizing hoisting requirements and tower crane hoisting capacity, intelligently optimizes the visual arrangement process of the tower crane based on a genetic algorithm improved based on engineering prior knowledge, and finds a scheme with the lowest price while meeting the hoisting weight requirements and space collision-free.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of heavy lifting equipment selection and arrangement, and particularly relates to a tower crane selection and site visual arrangement intelligent positioning method. BACKGROUND

[0002] As a key vertical transport tool, the tower crane plays an irreplaceable role in numerous engineering construction activities, and its importance is self-evident. In the preparation stage of engineering construction, selecting a suitable tower crane and its positioning is a crucial link, which directly affects the smoothness and progress of the entire project. Although the selection and positioning of the tower crane seem to be two separate steps, they are closely related, and the position of the tower crane determines which specific tower crane model needs to be selected.

[0003] In the traditional method, the scheme compiler roughly estimates the tower crane position through the CAD drawing, and relies heavily on engineering experience, which is low in efficiency and difficult to find a low-price and engineering-demand-adaptable selection and arrangement scheme. SUMMARY

[0004] The present application aims to solve the problems of the prior art and provides a tower crane selection and site visual arrangement intelligent positioning method. By visualizing the lifting demand and the tower crane lifting capacity, the working efficiency of the tower crane scheme compiler is greatly improved. Based on the improved genetic algorithm of engineering prior knowledge, the visual arrangement process of the tower crane is intelligently optimized, and the lowest price scheme is found while meeting the lifting capacity demand and the space collision-free.

[0005] To achieve the above-mentioned purpose, the following technical solutions are adopted:

[0006] A tower crane selection and site visual arrangement intelligent positioning method uses a programming language to develop a set of visualization programs, interactively creates a visualization graph by clicking a mouse, intelligently moves the lifting capacity graph of the tower crane, and solves the optimal selection and arrangement scheme.

[0007] The method comprises the following steps:

[0008] S1, a construction site survey is conducted, the site CAD drawing is processed according to the planar arrangement of the above-ground building, underground pipeline and construction temporary facility, the area is divided according to the lifting demand, the lifting capacity demand of each area is calculated, the lifting capacity distribution graph is obtained, the area is divided according to the boundary of the above-ground and underground environment, and the area restriction distribution graph is obtained;

[0009] S2, the pre-processed CAD drawing is converted into image data, the image is subjected to grayscale, Gaussian blur and thresholding processing, the contour detection is performed, and each connected region of the site is segmented; the lifting capacity demand and the area restriction of the segmented region are mapped to the color domain through a normalization mapping function, and the lifting capacity field of the site is formed on the graph , spatial field ;

[0010] S3, investigate the available tower crane models in the construction site, according to the model data of the tower crane product, draw the coverage image according to the length of the tower crane jib, the maximum lifting weight data, and map the lifting capacity field of the tower crane through the normalization mapping function , at the same time, preliminarily design the size, attachment method and attachment structure size of the foundation, and construct the spatial field of the tower crane ;

[0011] S4, according to the principle of specification requirements, spatial field collision-free and tower crane lifting capacity field covering the site weight field, visualize positioning, according to the overall lifting demand select the tower crane with corresponding capacity, move the tower crane in the site weight field , real-time pixel calculation of the image, calculate whether the lifting capacity meets the requirements and the spatial field collides through the judgment function, and visualize the image, gradually calculate the remaining lifting demand and add the tower crane until it fully meets the construction demand;

[0012] S5, use the genetic algorithm improved based on engineering prior knowledge to intelligently optimize the visual arrangement process of the tower crane, find the lowest price solution while meeting the lifting weight demand and the spatial collision-free.

[0013] S1 includes the following steps:

[0014] S11, conduct site survey of the construction site, determine the spatial restriction area of the building height, traffic line in the ground environment, the restriction area of the site material warehouse, material yard, mixing station working face, and the area of underground pipeline restricting the tower crane foundation;

[0015] S12, standardize the CAD drawing, remove the size marks and text, and draw the boundary of the restriction area in step S11 on the CAD drawing to obtain the area restriction distribution map;

[0016] S13, calculate the lifting weight demand of each area according to the construction drawing, redivide the area according to the lifting weight and environmental boundary, draw on the CAD drawing to obtain the lifting weight distribution map for computer vision processing.

[0017] S2 includes the following steps:

[0018] S21, convert the standardized drawing into image data, perform gray processing, Gaussian blur, thresholding processing on the image, and perform contour detection to obtain the contour data of the lifting weight distribution and the contour data of the spatial restriction area. These contour data are saved as a tree structure for subsequent calculation;​

[0019] S22. Using a depth-first search-based contour tree retrieval method, retrieve the contour lines of the lifting weight distribution map. In the standardized drawing image, click on a point in the region. ,calculate The calculation formula for the region where the point is located is as follows:

[0020] ;

[0021] In the formula,

[0022] Indicates the first The number of pixels clicked;

[0023] express The outer boundary contour of the region where the point is located;

[0024] express The coordinates of the pixel;

[0025] Indicates the outline number being retrieved;

[0026] Values ​​range from 1 to , This indicates the number of key pixels on the outline.

[0027] express The coordinates of the point;

[0028] calculate The Euclidean distance from the boundary of each connected region is determined. The outer boundary of the region to which the point belongs;

[0029] S23, Further search results All sub-outlines , ,,,, Further solving yields The formula for the region is as follows:

[0030] ;

[0031] Segment the connected components of the lifting weight distribution map for all different lifting weights;

[0032] S24, will Lifting capacity of the area After normalization, the following results were obtained. Furthermore, an interpolation function is used to map the color to the color space and assign it to... Region, visualize the lifting weight, color interpolation function as follows:

[0033] ;

[0034] In the formula,

[0035] The color after mapping is represented;

[0036] And The color corresponding to the low and high lifting weight is taken in the RGB or HSV space;

[0037] S25, repeat steps S22-S24, convert the weight distribution map into a visual lifting weight field ;

[0038] S26, convert the region limit distribution map into a visual space field using the method of steps S22-S25 , the color interpolation function takes the same color uniformly .

[0039] S3 includes the following steps:

[0040] S31, investigate the available tower crane models in the construction site, and statistically organize the types, models, parameters and rental prices of tower cranes, crawler cranes and automobile cranes;

[0041] S32, for a type of tower crane, draw a coverage image according to the data of the tower crane's lifting arm length and maximum lifting weight, and calculate the color at the center point according to the maximum lifting weight

[0042] ;

[0043] The pixel calculation formula in the tower crane coverage area is as follows:

[0044] ;

[0045] In the formula,

[0046] The distance from the farthest end of the tower crane coverage to the center is represented;

[0047] The Euclidean distance of the point in the region from the center point is represented;

[0048] Form the lifting capacity field of the tower crane ;

[0049] ​S33. Based on the specifications, design the foundation dimensions, attachment method, and attachment structure dimensions of the tower crane. Construct a visual spatial field of the tower crane using the methods in steps S22-S25. The color interpolation function uniformly selects the same color. .

[0050] S4 includes the following steps:

[0051] S41. Based on specifications and on-site construction requirements, visualize the lifting capacity of different models of mobile tower cranes. In the lifting area of ​​the site The system is designed to explore reasonable solutions by performing real-time lifting capacity assessment, spatial collision assessment, and cost calculation. Tower cranes are selected in batches, prioritizing those with larger lifting capacity requirements and gradually conforming to the assessment principles. The overall lifting demand is calculated based on the site's lifting capacity distribution map. :

[0052] ;

[0053] In the formula,

[0054] , , This represents the pixel value of a point in the site lifting weight distribution map;

[0055] Calculate the overall lifting capacity of each type of tower crane. :

[0056] ;

[0057] In the formula,

[0058] , , Indicates the first The pixel values ​​of points in the image representing the lifting capacity of a tower crane;

[0059] S42. Real-time lifting capacity determination, based on the lifting capacity field of the tower crane. Lifting field of the site The coverage capability, the Next, choose the closest one. of A specific tower crane , lifting capacity field Superimposed on the lifting field of the site ,move The center coordinates of the circles are used to determine the lifting weight in real time based on the superimposed state. The determination formula is as follows:

[0060] ;

[0061] In the formula,

[0062] Indicates the first The color value of the pixel after the stacking process;

[0063] This represents the color value of each pixel after the previous overlay.

[0064] tower crane The color value of each pixel;

[0065] , This indicates the offset distance of the tower crane's center point relative to the center point of the site;

[0066] The calculation results are truncated to RGB values ​​in the range of 0 to 255. The ability of the tower crane to cover the lifting requirements of the area is determined by visual color judgment. RGB(0,0,0) represents black, indicating that it can cover the area.

[0067] Further calculation of the first After the stacking, the remaining lifting requirements ;

[0068] ;

[0069] In the formula,

[0070] , , Indicates the first Pixel values ​​of points in the site lifting weight distribution map after multiple overlays;

[0071] S43. Real-time spatial discrimination is performed, with the discrimination principle being that the tower crane and its foundation do not conflict with existing above-ground or underground structures or structures to be constructed. spatial field Spatial field of the site The data is superimposed, and spatial discrimination is performed at the coordinate position where step S42 is located. The discrimination formula is as follows:

[0072] ;

[0073] Then there is no spatial conflict.

[0074] S44. Following the tower crane model from largest to smallest, then crawler cranes and truck cranes, the... The second choice is closest to of For the corresponding tower crane, repeat steps S41-S43 step by step until the judgment requirements are fully met:

[0075] ;

[0076] The remaining area is satisfied with the lifting demand by mobile lifting equipment.

[0077] S5 comprises the following steps:

[0078] S51, scheme coding is performed on the tower crane and its position [ K , x k , y k ] The initialization setting of the population is combined with prior knowledge, the large-weight tower crane is preferentially placed near the large-weight lifting area, and the engineering prior information that the tower crane is attached to the existing building structure is converted into a probability distribution The scheme population is formed by probabilistic sampling when the population is randomly initialized;

[0079] S52, the fitness is calculated, and the fitness function is as follows:

[0080] ;

[0081] In the formula,

[0082] The total price of the rented tower crane is represented;

[0083] The constraint processing is represented, the constraints are set according to the spatial discrimination and the lifting weight discrimination method, when the discrimination passes 1 is taken, and when it does not pass 0 is taken;

[0084] S53, the scheme with large fitness is selected to perform the crossover and mutation operation, and the probability parameter is set to the product of the fixed probability and the prior probability:

[0085] Position crossover: ;

[0086] Position mutation: ;

[0087] S54, a new population is generated, the fitness is recalculated, and the probability distribution is updated according to the fitness gradient ;

[0088] S55, steps S52-S54 are repeated until the maximum number of evolution times is reached;

[0089] S56, the scheme with the lowest total price is selected as the final tower crane selection and arrangement scheme.

[0090] The beneficial effect of the present application is that the present application effectively visualizes the tower crane selection and arrangement process, visualizes the display by area according to the hoisting weight requirement, greatly improves the efficiency of the scheme compiler, facilitates the construction personnel to intuitively understand the scheme, improves the computability, optimizes the efficiency, and creates economic effects. BRIEF DESCRIPTION OF DRAWINGS

[0091] Fig. 1 is a flow chart of the present application;

[0092] Fig. 2 is a site visualization process chart of the present application;

[0093] Fig. 3 is a genetic algorithm flow chart based on engineering prior knowledge improvement of the present application;

[0094] The present application will be described in detail below with reference to the accompanying drawings. DETAILED DESCRIPTION

[0095] The principles and features of the present application are described below in conjunction with the accompanying drawings, and the examples are only used to explain the present application and are not intended to limit the scope of the present application. In the following paragraphs, the present application is described in more detail with examples. The advantages and features of the present application will be more apparent from the following description. It should be noted that the drawings are very simplified and use non-precise proportions, only to facilitate and clarify the purpose of assisting the description of the embodiments of the present application.

[0096] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used in the specification of the present application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0097] The present application is further described below in conjunction with the accompanying drawings and examples:

[0098] In order to improve the efficiency of tower crane selection and arrangement, reduce the cost of leasing tower cranes, and thus create economic benefits, the present application focuses on the intelligent positioning of tower crane selection and site visualization arrangement.

[0099] With the development of high-tech such as robots, artificial intelligence, intelligent construction and wisdom civil engineering are new productive forces to improve the development momentum of civil engineering. Intelligent algorithms based on machine learning and computer vision have been widely used in civil engineering construction, structure monitoring and other fields, creating great value. However, in the field of tower crane selection and arrangement, there is a lack of corresponding intelligent methods to optimize the scheme, so the tower crane selection and site visual arrangement intelligent positioning method based on the improved genetic algorithm of engineering prior knowledge emerges as the times require, which improves the work efficiency of the personnel who prepares the tower crane scheme by visualizing the lifting demand and tower crane lifting capacity, and quickly finds a reasonable and economically beneficial scheme.

[0100] A tower crane selection and site visual arrangement intelligent positioning method uses a programming language to develop a set of visualization programs, interactively creates a visualization graph by clicking the mouse, and intelligently moves the lifting capacity graph of the tower crane to solve the optimal selection and arrangement scheme.

[0101] As shown in Figs. 1-3 , it includes the following steps:

[0102] S1, carry out construction site survey, process site CAD drawings according to above-ground buildings, underground pipelines, construction temporary facilities plan, etc., and divide the area according to the lifting demand, calculate the lifting weight demand of each area, get the lifting weight distribution map, divide the site area according to the above-ground and underground environment boundary, get the area restriction distribution map;

[0103] Specifically:

[0104] S11, carry out construction site survey, determine the spatial restriction area of building height, traffic line, etc. in the above-ground environment, the restriction area of site working face such as site material warehouse, material yard, mixing station, etc., and the area that restricts the tower crane foundation such as underground pipeline, etc.;

[0105] S12, standardize the CAD drawing, remove the size mark and text, and draw the boundary of the restriction area in step S11 on the CAD drawing to get the area restriction distribution map;

[0106] S13, calculate the lifting weight demand of each area according to the construction drawing, redivide the area according to the lifting weight and environment boundary, draw on the CAD drawing to get the lifting weight distribution map for computer vision processing.

[0107] S2, convert the preprocessed CAD drawing into image data, perform grayscale, Gaussian blur and thresholding processing on the image, perform contour detection, and segment each connected region of the site; map the lifting weight demand and area restriction of the segmented area to the color domain through a normalization mapping function to form a lifting weight field and a space field on the drawing. ; ​

[0108] Specifically,

[0109] S21, the standardized drawing is converted into image data, the image is subjected to gray processing, Gaussian blur, threshold processing, and contour detection, to obtain contour data of the lifting weight distribution and contour data of the spatial limitation area, which are saved as tree structure for subsequent calculation;

[0110] S22, a contour tree retrieval method based on depth-first search (DFS) is adopted to retrieve the contour line of the lifting weight distribution map, and one point in the area is clicked in the standardized drawing image , the area where the point is located is calculated, and the calculation formula is as follows:

[0111] ;

[0112] In the formula,

[0113] represents the pixel point clicked for the n-th time;

[0114] represents the outer boundary contour of the area where the point is located; represents the coordinates of the pixel point;

[0115] represents the contour line number retrieved;

[0116] represents the contour line number retrieved;

[0117] represents the contour line number retrieved; , represents the number of key pixel points on the contour line;

[0118] represents the coordinates of the point;

[0119] represents the Euclidean distance from each connected area boundary, to determine the outer boundary of the area to which the point belongs; S23, all sub-contours of are further retrieved

[0120] , , , , , and the area where is located is further solved, and the formula is as follows:

[0121] ;

[0122] ​​​​This allows for the segmentation of all connected components with different lifting capacities in the lifting capacity distribution map;

[0123] S24, will Lifting capacity of the area After normalization, the result is obtained Further, an interpolation function is used to map the color to the color space and assign it to... The region visualizes the lifting weight, and the color interpolation function is as follows:

[0124] ;

[0125] In the formula,

[0126] Indicates the mapped color;

[0127] and The color corresponding to the low and high bits of the lifting weight is taken in RGB or HSV space.

[0128] S25. Repeat steps S22-S24 to transform the weight distribution map into a visualized lifting field. ;

[0129] S26. Using the methods in steps S22-S25, transform the regional restricted distribution map into a visualized spatial field. The color interpolation function uniformly selects the same color. .

[0130] S3. Investigate the available tower crane models in the construction area. Based on the tower crane product model data, draw a coverage map according to the tower crane's boom length, maximum lifting capacity, and other data. Map the lifting capacity field of the tower crane using a normalized mapping function. Simultaneously, preliminary design of the foundation dimensions, attachment method, and attachment structure dimensions is carried out to construct the spatial field of the tower crane. ;

[0131] Specifically:

[0132] S31. Investigate the available tower crane models in the construction area and compile statistics on the types, models, parameters, and rental prices of tower cranes, crawler cranes, and truck cranes.

[0133] S32. For a certain type of tower crane, draw a coverage area image based on the tower crane's boom length, maximum lifting capacity, and other data. Then, based on the maximum lifting capacity... Calculate the color at the center point:

[0134] ;

[0135] The formula for calculating pixels within the tower crane's coverage area is as follows:

[0136] ;

[0137] In the formula,

[0138] This indicates the distance from the farthest point of the tower crane's coverage area to the center.

[0139] This represents the Euclidean distance from a point within the region to the center point;

[0140] This creates a lifting capacity field for the tower crane. ;

[0141] S33. Based on the specifications, design the foundation dimensions, attachment method, and attachment structure dimensions of the tower crane. Use the methods in steps S22-S25 to construct a visual spatial field of the tower crane. The color interpolation function uniformly selects the same color. .

[0142] S4. Based on the specifications, the principle of no collision in the spatial field, and the principle that the tower crane's lifting capacity field covers the site's lifting capacity field, visual positioning is performed according to the overall lifting requirements. Select a tower crane with appropriate capacity for the site's lifting capacity. Lifting capacity field of mobile tower crane The system performs real-time pixel-level calculations on the image, uses a discriminant function to determine whether the lifting capacity meets requirements and whether there is a collision in the spatial field, and displays the results visually on the image, progressively calculating the remaining lifting requirements. And add tower cranes until they fully meet the construction requirements;

[0143] Specifically:

[0144] S41. Based on specifications and on-site construction requirements, visualize the lifting capacity of different models of mobile tower cranes. In the lifting area of ​​the site The system is designed to explore reasonable solutions by performing real-time lifting capacity assessment, spatial collision assessment, and cost calculation. Tower cranes are selected in batches, prioritizing those with larger lifting capacity requirements and gradually conforming to the assessment principles. The overall lifting demand is calculated based on the site's lifting capacity distribution map. :

[0145] ;

[0146] In the formula,

[0147] , , This represents the pixel value of a point in the site lifting weight distribution map;

[0148] Calculate the overall lifting capacity of each type of tower crane. :

[0149] ;

[0150] In the formula,

[0151] , , Indicates the first The pixel values ​​of points in the image representing the lifting capacity of a tower crane;

[0152] S42. Real-time lifting capacity determination, based on the lifting capacity field of the tower crane. Lifting field of the site The coverage capability, the Next, choose the closest one. of A corresponding tower crane , lifting capacity field Superimposed on the lifting field of the site ,move The center coordinates of the circles are used to determine the lifting weight in real time based on the superimposed state. The determination formula is as follows:

[0153] ;

[0154] In the formula,

[0155] Indicates the first The color value of the pixel after the stacking process;

[0156] This represents the color value of each pixel after the previous overlay.

[0157] tower crane The color value of each pixel;

[0158] , This indicates the offset distance of the tower crane's center point relative to the center point of the site;

[0159] The calculation results are truncated to RGB values ​​in the range of 0 to 255. The ability of the tower crane to cover the lifting requirements of the area is determined by visual color judgment. RGB(0,0,0) represents black, indicating that it can cover the area.

[0160] Further calculation of the first After the stacking, the remaining lifting requirements ;

[0161] ;

[0162] In the formula,

[0163] , , Indicates the first Pixel values ​​of points in the site lifting weight distribution map after multiple overlays;

[0164] S43. Real-time spatial discrimination is performed, with the discrimination principle being that the tower crane and its foundation do not conflict with existing or planned structures above or below ground. spatial field Spatial field of the site The data is superimposed, and spatial discrimination is performed at the coordinate position where step S42 is located. The discrimination formula is as follows:

[0165] ;

[0166] Then there is no spatial conflict.

[0167] S44. Following the tower crane model from largest to smallest, then crawler cranes and truck cranes, the... The second choice is closest to of For the corresponding tower crane, repeat steps S41-S43 step by step until the judgment requirements are fully met:

[0168] ;

[0169] The remaining areas will have their lifting needs met by mobile lifting equipment.

[0170] S5. Employs a genetic algorithm improved based on prior engineering knowledge to intelligently optimize the visualized layout process of tower cranes, finding the lowest-priced solution while meeting lifting capacity requirements and avoiding space collisions.

[0171] Specifically:

[0172] S51. Code the scheme for tower cranes and their locations. [ K , x k , y k ] By combining prior knowledge to initialize the population, prior engineering information such as prioritizing the placement of heavy-duty tower cranes near heavy-duty lifting areas and attaching tower cranes to existing building structures is transformed into a probability distribution. When randomly initializing the population, a scheme population is formed by sampling according to probability.

[0173] S52. Calculate the fitness function as follows:

[0174] ;

[0175] In the formula,

[0176] represents the total price of the rented tower crane;

[0177] represents a constraint processing, constraints are set according to the space discrimination, hoisting weight discrimination method, when discrimination passes take 1, when it does not pass take 0;

[0178] S53, select the scheme with large fitness for cross and mutation operation, the probability parameter is set to the product of the fixed probability and the prior probability:

[0179] Position crossover: ;

[0180] Position mutation: ;

[0181] S54, generate a new population, and recalculate the fitness, and update the probability distribution according to the fitness gradient ;

[0182] S55, repeat steps S52-S54 until the maximum number of evolution times ;

[0183] S56, select the scheme with the lowest total price as the final tower crane selection and arrangement scheme. Specific embodiment 1:

[0185] combined with Fig. 2 , Fig. 3 , a tower crane selection and site visualization arrangement intelligent positioning method in the application is used to develop a scheme for the site tower crane selection and arrangement of a super high-rise building project.

[0186] Construction site field survey is carried out, and the hoisting weight tons of each region in the whole construction process are calculated. The central region of the super high-rise core tube is provided with steel structures such as rigid beams and steel staircases, and the hoisting weight is between 0.1-4.5t. The hoisting weight distribution and regional restriction are drawn on the CAD drawing respectively.

[0187] The hoisting weight distribution map and the regional restriction distribution map are converted into image data in png format, a program is written in Python language to process the image, contour detection is carried out, the contour of the hoisting weight distribution map is obtained, each region is segmented by clicking the mouse, and the color interpolation function is assigned, the assignment is mapped from the hoisting weight space to the RGB space, take (0, 0, 255), take (255, 0, 0).

[0188] The local tower crane leasing price is shown in Table 1, the leasing car hoist price is 500-2000 yuan / day, the preliminary design basis size, the attachment mode, the drawing tower crane foundation occupies the area, covers the visual diagram of the hoisting capacity of the field.

[0189]

[0190] The tower crane center point is moved on the site map, and whether the space conflicts or the hoisting capacity can be covered is judged in real time, the value of RGB is calculated by using (0~255) truncation, and RGB (0, 0, 0) black means that it can be covered, according to engineering experience, the large tower crane SYT80A (T6010-6) is placed first, and gradually covers the whole map.

[0191] According to engineering experience, the tower crane is arranged near the core tube with a high probability, and can cover the material yard area, the prior information probability distribution is initialized , the probability sampling and coding are carried out, 20 sets of scheme coding are set, the population is initialized, the fixed probability of crossover and mutation is set to 0.4, 0.1, the maximum evolution number is set to 5000 times, after the calculation is finished, the scheme with the lowest price is selected as the selection and arrangement scheme.

[0192] The present application effectively visualizes the tower crane selection and arrangement process, and visualizes the hoisting capacity demand by region, greatly improves the efficiency of the scheme compiler, facilitates the construction personnel to intuitively understand the scheme, and creates economic effect.

[0193] The present application effectively converts the tower crane selection and arrangement scheme preparation problem into a mathematical process, improves the calculability and optimization efficiency.

[0194] On the basis of the genetic algorithm, the present application sets engineering prior knowledge and integrates engineering experience, effectively constrains the optimization direction of the scheme, and improves the optimization efficiency.

[0195] The present application is not only suitable for tower crane selection and arrangement optimization, but also suitable for other construction equipment selection and arrangement problems, has wide application range and is easy to popularize.

[0196] The present application has been described above in conjunction with the drawings, and obviously the specific implementation of the present application is not limited by the above mode, as long as various improvements are made by using the method concept and technical scheme of the present application, or the present application is directly applied to other occasions without improvement, all of which are within the protection scope of the present application.

Claims

1. A method for intelligent positioning of tower crane selection and site visualization layout, characterized in that, Includes the following steps: S1. Conduct on-site survey of the construction site, process the site CAD drawings based on the layout of above-ground buildings, underground pipelines, and temporary construction facilities, divide the area according to lifting requirements, calculate the lifting capacity requirements of each area, obtain the lifting capacity distribution map, and divide the site area according to the above-ground and underground environmental boundaries to obtain the area restriction distribution map. S2. Convert the preprocessed CAD drawing into image data, perform grayscale conversion, Gaussian blurring, and thresholding on the image, perform contour detection, and segment the connected regions of the site; map the lifting capacity requirements and regional constraints of the segmented regions to the color space using a normalization mapping function to form the lifting capacity field of the site on the image. spatial field ; Includes the following steps: S21. Convert the standardized drawings into image data, perform grayscale processing, Gaussian blurring, and thresholding on the images, and perform contour detection to obtain contour data of the lifting weight distribution and contour data of the spatially restricted area. Save these contour data as a tree structure for subsequent calculations. S22. Using a depth-first search-based contour tree retrieval method, retrieve the contour lines of the lifting weight distribution map. In the standardized drawing image, click on a point in the region. ,calculate The calculation formula for the region where the point is located is as follows: ； In the formula, Indicates the first The number of pixels clicked; express The outer boundary contour of the region where the point is located; express The coordinates of the pixel; Indicates the outline number being retrieved; Values ​​range from 1 to , This indicates the number of key pixels on the outline. express The coordinates of the point; calculate The Euclidean distance from the boundary of each connected region is determined. The outer boundary of the region to which the point belongs; S23, Further search results All sub-outlines , ,,,, Further solving yields The formula for the region is as follows: ； Segment the connected components of the lifting weight distribution map for all different lifting weights; S24, will Lifting capacity of the area After normalization, the following results were obtained. Further, an interpolation function is used to map the color to the color space and assign it to... The region visualizes the lifting weight, and the color interpolation function is as follows: ； In the formula, Indicates the mapped color; and The color corresponding to the low and high bits of the lifting weight is taken in RGB or HSV space. S25. Repeat steps S22-S24 to transform the weight distribution map into a visualized lifting field. ; S26. Using the methods in steps S22-S25, transform the regional restricted distribution map into a visualized spatial field. The color interpolation function uniformly selects the same color. ; S3. Investigate the available tower crane models in the construction area. Based on the tower crane product model data, draw a coverage map according to the tower crane's boom length and maximum lifting capacity data. Map the lifting capacity field of the tower crane using a normalized mapping function. Simultaneously, the preliminary design of the foundation dimensions, attachment method, and attachment structure dimensions was carried out, and the spatial field of the tower crane was constructed. ; S4. Based on the specifications, the principle of non-collision in the spatial field, and the principle that the tower crane's lifting capacity field covers the site's lifting capacity field, visual positioning is performed according to the overall lifting requirements. Select a tower crane with appropriate capacity for the site's lifting capacity. Lifting capacity field of mobile tower crane The system performs real-time pixel-level calculations on the image, uses a discriminant function to determine whether the lifting capacity meets requirements and whether there is a collision in the spatial field, and displays the results visually on the image, progressively calculating the remaining lifting requirements. And add tower cranes until they fully meet the construction requirements; S5. Employing a genetic algorithm improved based on prior engineering knowledge, the system intelligently optimizes the visualized layout process of tower cranes, finding the lowest-priced solution while meeting lifting capacity requirements and avoiding spatial collisions.

2. The intelligent positioning method for tower crane selection and site visualization layout according to claim 1, characterized in that, S1 includes the following steps: S11. Conduct on-site surveys of the construction site to determine the spatial restrictions on building height and traffic lines in the above-ground environment, the restricted areas of on-site material warehouses, material storage yards, and mixing plant work surfaces, as well as the restricted areas for tower crane foundations due to underground pipelines; S12. Standardize the CAD drawing, remove dimension marks and text, and draw the boundary of the restricted area in step S11 on the CAD drawing to obtain the area restriction distribution map; S13. Calculate the lifting capacity requirements of each area based on the construction drawings, redivide the areas according to the lifting capacity and environmental boundaries, and draw them on the CAD drawing to obtain the lifting capacity distribution map for computer vision processing.

3. The intelligent positioning method for tower crane selection and site visualization layout according to claim 2, characterized in that, S3 includes the following steps: S31. Investigate the available tower crane models in the construction area and compile statistics on the types, models, parameters, and rental prices of tower cranes, crawler cranes, and truck cranes. S32. For a certain type of tower crane, draw a coverage area image based on the crane's boom length and maximum lifting capacity data, and then, based on the maximum lifting capacity... Calculate the color at the center point: ； The formula for calculating pixels within the tower crane's coverage area is as follows: ； In the formula, This indicates the distance from the farthest point of the tower crane's coverage area to the center. This represents the Euclidean distance between a point within the region and the center point. Forming the lifting capacity field of the tower crane ; S33. Based on the specifications, design the foundation dimensions, attachment method, and attachment structure dimensions of the tower crane. Construct a visual spatial field of the tower crane using the methods in steps S22-S25. The color interpolation function uniformly selects the same color. .

4. The intelligent positioning method for tower crane selection and site visualization layout according to claim 3, characterized in that, S4 includes the following steps: S41. Based on specifications and on-site construction requirements, visualize the lifting capacity of different models of mobile tower cranes. In the lifting area of ​​the site The system is designed to explore reasonable solutions by performing real-time lifting capacity assessment, spatial collision assessment, and cost calculation. Tower cranes are selected in batches, prioritizing those with larger lifting capacity requirements and gradually conforming to the assessment principles. The overall lifting demand is calculated based on the site's lifting capacity distribution map. : ； In the formula, , , This represents the pixel value of a point in the site lifting weight distribution map; Calculate the overall lifting capacity of each type of tower crane. : ； In the formula, , , Indicates the first The pixel values ​​of points in the image representing the lifting capacity of a tower crane; S42. Real-time lifting capacity determination, based on the lifting capacity field of the tower crane. Lifting field of the site The coverage capability, the Next, choose the closest one. of A corresponding tower crane , lifting capacity field Superimposed on the lifting field of the site ,move The center coordinates of the circles are used to determine the lifting weight in real time based on the superimposed state. The determination formula is as follows: ； In the formula, Indicates the first The color value of the pixel after the stacking process; This represents the color value of each pixel after the previous overlay. tower crane The color value of each pixel; , This indicates the offset distance of the tower crane's center point relative to the center point of the site; The calculation results are truncated to RGB values ​​in the range of 0 to 255. The ability of the tower crane to cover the lifting requirements of the area is determined by visual color judgment. RGB(0,0,0) represents black, indicating that it can cover the area. Further calculation of the first After the stacking, the remaining lifting requirements ; ； In the formula, , , Indicates the first Pixel values ​​of points in the site lifting weight distribution map after multiple overlays; S43. Real-time spatial discrimination is performed, with the discrimination principle being that the tower crane and its foundation do not conflict with existing above-ground or underground structures or structures to be constructed. spatial field Spatial field of the site The data is superimposed, and spatial discrimination is performed at the coordinate position where step S42 is located. The discrimination formula is as follows: ； Then the spaces do not conflict; S44. Following the tower crane model from largest to smallest, then crawler cranes and truck cranes, the... The second choice is closest to of For the corresponding tower crane, repeat steps S41-S43 step by step until the judgment requirements are fully met: ； The remaining areas will have their lifting needs met by mobile lifting equipment.

5. The intelligent positioning method for tower crane selection and site visualization layout according to claim 4, characterized in that, S5 includes the following steps: S51. Code the tower crane and its location. By combining prior knowledge to initialize the population, prior engineering information such as prioritizing the placement of heavy-duty tower cranes near heavy-duty lifting areas and attaching tower cranes to existing building structures is transformed into a probability distribution. When randomly initializing the population, a scheme population is formed by sampling according to probability. S52. Calculate the fitness function as follows: ； In the formula, This indicates the total price of renting the tower crane; This indicates constraint processing, setting constraints according to spatial discrimination and lifting capacity discrimination methods. When the discrimination passes... Take 1, if it fails Set to 0; S53. Select a scheme with high fitness for crossover and mutation operations, setting the probability parameter to the product of a fixed probability and a prior probability: Positional overlap: ; Positional variation: ; S54. Generate a new population, recalculate the fitness, and update the probability distribution based on the fitness gradient. ; S55. Repeat steps S52-S54 until the maximum number of evolutions is reached. End of time; S56. Select the option with the lowest total price as the final tower crane selection and layout scheme.

Images