A method for selecting and arranging tower groups

Abstract

The present invention discloses a method for selecting and arranging a group of towers, comprising: step 1: establishing a tower crane model parameter library; step 2: analyzing the tower crane construction plan; step 3: determining the plane position boundary range of the tower crane tower body center; step 4: establishing the group tower plane layout constraint conditions and optimization targets; step 5: using an improved genetic algorithm to determine the group tower selection and preliminary plane layout; step 6: using a gradient descent method to optimize the plane layout. The present invention optimizes the selection and layout of tower cranes, achieving the effect of reducing construction costs and improving construction safety.

Description

Technical Field

[0001] The invention relates to the technical field of building construction, and in particular to a method for selecting and arranging a group of towers. Background Art

[0002] Tower cranes (abbreviated as tower cranes) are indispensable lifting equipment in construction, especially in high-rise buildings and large engineering projects. The reasonable selection and layout of tower cranes directly affect the construction efficiency and project cost. Traditional tower crane layout methods often rely on experience and simple geometric layout methods, which have low construction efficiency and high costs, and it is difficult to achieve the optimal layout in a complex construction environment. With the expansion of the scale and increase in complexity of construction projects, how to optimize the selection and layout of tower groups while ensuring construction safety and efficiency, thereby reducing the lifting cost, has become an important research topic in the current field of construction engineering construction technology. Among them, tower groups refer to two or more tower cranes working at the same construction site and having a common coverage area.

[0003] At present, in the early planning stage of construction projects, the commonly used tower crane layout methods mainly include empirical methods and geometric methods. The empirical method relies on the experience and intuition of engineers. Although it is simple and fast, it lacks scientific basis and is prone to multiple rework and modification, resulting in a waste of manpower and time costs. The geometric method determines the position of the tower crane through simple geometric calculations. It is suitable for simple construction sites, but it is not effective in complex environments. At the same time, the geometric method also needs to rely on experience to make a preliminary selection of the tower group, and it is difficult to find the optimal tower group selection and plane layout plan. In addition, the existing methods are generally unable to quickly and accurately evaluate the impact of tower crane selection and layout on lifting costs, making it difficult to achieve true cost optimization. Summary of the invention

[0004] The purpose of the present invention is to provide a method for selecting and arranging a group of towers, so as to solve the problem of optimizing the selection and layout of the group of towers and thus reducing the lifting cost under the premise of ensuring construction safety and efficiency.

[0005] In order to solve the above technical problems, the present invention provides a technical solution: a tower group selection and plane layout method, comprising:

[0006] Step 1: Establish a tower crane model parameter library;

[0007] Step 2: Analyze the tower crane construction plan;

[0008] Step 3: Determine the plane position boundary range of the center of the tower crane;

[0009] Step 4: Establish the constraints and optimization objectives for the tower group layout;

[0010] Step 5: Use the improved genetic algorithm to determine the tower group selection and preliminary layout;

[0011] Step 6: Use gradient descent method to optimize the floor plan.

[0012] Furthermore, in the tower group selection and plane layout method provided by the present invention, in the step 1, the method for establishing the tower crane model parameter library is: collecting the attached tower crane models available for rental, recording the boom length of each tower crane model, and rent , then the tower crane model can be expressed as On this basis, we integrate and establish the tower crane model parameter library .

[0013] Furthermore, in the tower group selection and plane layout method provided by the present invention, in step 2, the method for parsing the tower crane construction plane diagram includes:

[0014] Step 2.1: Analyze the tower crane construction plan to obtain the graphic contour line parameter information of the building red line, construction road, ancillary facilities and the planned building in the tower crane construction plan;

[0015] Step 2.2: Take the lower left corner of the building red line as the coordinate origin (0,0) and establish a plane rectangular coordinate system;

[0016] Step 2.3: In the plane rectangular coordinate system, set a rectangular area. The range of the rectangular area on the horizontal x-axis is the projection range of the building red line on the horizontal x-axis. , the range on the vertical y-axis is the projection range of the building red line on the vertical y-axis , then the rectangular area is represented by ; Mesh the rectangular area, divide the rectangular area into multiple rectangular grids according to a predetermined spacing, mark each vertex of the rectangular grid and number each vertex.

[0017] Furthermore, in the tower group selection and plane layout method provided by the present invention, in step 3, the method for determining the plane position boundary range of the tower crane tower center includes:

[0018] Step 3.1: Based on the graphic information of the construction plan obtained by parsing in step 2, establish the graphic contour line function of the building red line, construction road, ancillary facilities and the planned building in the tower crane construction plan f ;

[0019] Step 3.2: Expand the horizontal safety distance along the normal direction of the building red line, construction road, ancillary facilities and the graphic contour line of the proposed building in the tower crane construction plan , get the plane position boundary contour function of the center of the tower crane , thus determining the plane position boundary range of the tower crane tower center within the building red line as ;

[0020] Step 3.3: Filter the vertices of the rectangular grid divided in step 2 and take the plane coordinates within the boundary range Some vertices in the tower form a set of optional points in the center of the tower crane .

[0021] Furthermore, in the tower group selection and plane layout method provided by the present invention, in step 4, the method of establishing the tower group plane layout constraint conditions and optimization objectives includes:

[0022] Step 4.1: Establish the constraints for the tower group layout, including:

[0023] Step 4.1.1: According to formula (1), verify that the horizontal distance between the end of any tower crane boom and the center of the adjacent tower crane tower meets the specification requirements;

[0024] (1);

[0025] In the formula, It's a tower crane and tower crane Horizontal distance from the center of the tower; , and , Tower crane and tower crane The horizontal and vertical coordinate values ​​of the center of the tower; , Tower crane and tower crane The length of the lifting arm; It is the horizontal safety distance between the end of the tower crane boom and the center of the adjacent tower crane tower body required by the specification;

[0026] Step 4.1.2: The building plane of any proposed building is completely covered by the lifting range of the tower crane;

[0027] Step 4.2: Establish optimization objectives for the tower group layout, including:

[0028] Step 4.2.1: Determine the sum of the rental costs of all tower cranes according to formula (2);

[0029] (2);

[0030] In the formula, is the total rental cost of all tower cranes; is the number of tower crane types in the tower group layout plan; The tower crane model is The number of tower cranes; The tower crane model is the rent;

[0031] Step 4.2.2: Determine the cross-operation area ratio of the tower cluster according to formula (3) and formula (4)

[0032] (3);

[0033] (4);

[0034] In the formula, is the total number of tower cranes in the tower group layout plan; is the cross-operation area ratio of the tower groups; It's a tower crane and tower crane Cross-operation area; It's a tower crane operating area;

[0035] Step 4.2.3: Considering the two optimization objectives comprehensively, the objective function of the tower group layout is obtained as formula (5);

[0036] (5);

[0037] In the formula, It is the objective function of the plane layout of the tower groups.

[0038] Furthermore, in the tower group selection and plane layout method provided by the present invention, in step 5, the method of using an improved genetic algorithm to determine the tower group selection and preliminary plane layout includes:

[0039] Step 5.1: Use binary coding to encode the plan layout of a single tower group, where the coding adopts formula (6);

[0040] (6);

[0041] In the formula, It is an individual plan layout plan of a single tower group; , are the binary representations of the serial numbers of the first and nth tower crane models respectively; , The tower crane models are , Binary representation of the number of cranes; , The tower crane models are The first and The binary representation of the number of the point where the center of the tower crane is located; , The tower crane models are The first and The binary representation of the number of the point where the center of the tower crane is located;

[0042] Step 5.2: Initialize the population, and set the initial number of individuals in the population to , use random numbers to generate each individual, and filter out individuals that do not meet the requirements of step 3.3 and step 4.1 until the initial number of individuals is reached; establish a fitness function for evaluating individual fitness according to formula (7);

[0043] (7);

[0044] In the formula, Is an individual The fitness value of Is an individual The objective function value of

[0045] According to the fitness function, calculate the fitness of each individual in the population;

[0046] Step 5.3: Select the top 10% of the best individuals in terms of fitness. Perform local search optimization and make small adjustments to individual The crane model, quantity or location in the filter is used to filter out invalid individuals and individuals that do not meet the requirements of steps 3.3 and 4.1, and generate individuals The set of neighborhood individuals ;

[0047] Step 5.4: For each individual , calculate the set The fitness value of each neighborhood individual in , take the neighborhood individual with the highest fitness value that is higher than the fitness value of the original individual, and replace it with the original individual ;

[0048] Step 5.5: Keep the excellent individuals after replacement and directly enter the next generation;

[0049] Step 5.6: Use roulette wheel selection as the selection operator, so that the probability of each individual entering the next generation is equal to the ratio of its fitness value to the sum of the fitness values ​​of all individuals in the entire population, and select individuals with better fitness to enter the next generation;

[0050] Step 5.7: Use multi-point crossover as the crossover operator, set crossover points at the end of the binary code of the crane model number, the number of cranes corresponding to the crane model, and the number of the center of the tower body of each crane, and then exchange genes between individuals to generate new individuals, and filter out invalid individuals and individuals that do not meet the requirements of steps 3.3 and 4.1;

[0051] Step 5.8: Use basic position mutation as the mutation operator to perform mutation operation on any position of the individual according to a certain mutation probability to generate new individuals, and filter out invalid individuals and individuals that do not meet the requirements of steps 3.3 and 4.1;

[0052] Step 5.9: Aggregate the new individuals generated in steps 5.3 to 5.8 to form a new population;

[0053] Step 5.10: Repeat steps 5.2 to 5.9 until an individual reaches the fitness threshold , and use this individual as the tower group selection and preliminary plan layout plan for the project.

[0054] Furthermore, in the tower group selection and plane layout method provided by the present invention, in step 6, the method of optimizing the plane layout by using the gradient descent method includes:

[0055] Step 6.1: Individual tower group layout plan output from step 5.7 , the tower crane model and the corresponding tower crane quantity have been determined, and the tower center of each tower crane is further iterated and refined. First, the individual is expressed in a new form according to formula (8);

[0056] (8);

[0057] In the formula, Is an individual New representation of , are the numbers of the first and nth tower crane models respectively; , The tower crane models are , The number of tower cranes; , and , The tower crane models in the initial state are The first and The horizontal and vertical coordinates of the center of the tower crane; , and , The tower crane models in the initial state are The first and The horizontal and vertical coordinates of the center of the tower crane;

[0058] Step 6.2: Perform gradient descent iterative update to gradually optimize the layout of each tower crane; modify the original objective function according to the determined tower crane model and the corresponding number of tower cranes, and obtain the new objective function as formula (9);

[0059] (9);

[0060] In the formula, is the new objective function, is the cross-operation area ratio of the tower groups;

[0061] Assume that the current Iterations, for each crane , calculate the gradient of the new objective function with respect to the coordinates , the partial derivative of the objective function at this point is ;

[0062] According to the calculated partial derivatives, the tower center of each tower crane is updated. The specific calculation formula (10) is:

[0063] (10);

[0064] Step 6.3: Calculate the total gradient norm according to formula (11);

[0065] (11);

[0066] In the formula, is the total gradient norm, is the partial derivative of the new objective function at this point;

[0067] Step 6.4: Repeat steps 6.2 to 6.3. After each iteration, determine whether the gradient norm is less than the set threshold. If the conditions are met, the gradient descent process is considered to have converged, and the tower center of each current tower crane is output as the final plane layout of the tower group.

[0068] Furthermore, in the tower group selection and plane layout method provided by the present invention, in step 2.2, other vertices of the building red line are used instead of the lower left corner vertex as the coordinate origin (0,0) to establish a plane rectangular coordinate system.

[0069] Furthermore, in the tower group selection and plane layout method provided by the present invention, in step 2.3, the rectangular area is divided into a plurality of rectangular grids with a predetermined spacing of 5 m.

[0070] Furthermore, in the tower group selection and plane layout method provided by the present invention, in step 6.2, the learning rate , as the step size for updating the horizontal and vertical coordinates of the center of the tower crane in each iteration.

[0071] Compared with the prior art, the tower group selection and plane layout method provided by the present invention has the following beneficial effects:

[0072] 1. By comprehensively considering the building red line, the planned building and other relevant factors, the selection and quantity of tower cranes were optimized, and the plane coordinates of the center of each tower crane were finally determined. The selection and layout of the tower cranes were optimized, achieving the effect of reducing construction costs and improving construction safety.

[0073] 2. By analyzing the tower crane construction plan and setting up a reasonable boundary range for constraint verification, the safety and rationality of the tower crane layout are ensured, saving time for manual repeated verification.

[0074] 3. Establishing the objective function of multi-objective optimization and using genetic algorithm for tower group selection and preliminary plane layout can realize rapid selection, reduce the overall rental cost of tower cranes, and effectively reduce the number of reworks compared with manual selection.

[0075] 4. Optimize the plane layout of tower cranes through the gradient descent method, significantly reduce the cross-operation area between tower cranes, and ensure construction safety.

[0076] 5. It combines multi-objective optimization and dynamic adjustment mechanism, aiming to comprehensively consider the selection, plane layout and lifting cost of tower cranes by establishing mathematical models and optimization algorithms, and further refine the plane position of the tower group on the basis of rapid preliminary determination, so as to finally achieve global optimization of the tower group layout.

[0077] 6. The present invention can reduce construction costs, improve construction efficiency, ensure construction safety, and can minimize the number of tower cranes used and the lifting cost while meeting construction requirements, thereby providing an efficient and economical solution for building tower crane construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 A flow chart of a method for selecting and arranging tower groups according to an embodiment of the present invention;

[0079] Figure 2 A schematic diagram of rectangular mesh division and vertex marking according to an embodiment of the present invention;

[0080] Figure 3 This is a schematic diagram of the plane position boundary range of the center of the tower crane body according to an embodiment of the present invention;

[0081] Figure 4 A schematic diagram of a set of optional points at the center of a tower crane body according to an embodiment of the present invention;

[0082] Figure 5 It is a schematic diagram of the plan layout of the tower groups finally adopted in the embodiment of the present invention. DETAILED DESCRIPTION

[0083] The present invention is described in detail below in conjunction with the accompanying drawings: The advantages and features of the present invention will become more apparent from the following description. It should be noted that the accompanying drawings are in very simplified form and in non-precise proportions, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0084] Please refer to Figure 1 The embodiment of the present invention provides a method for selecting and arranging a group of towers, including:

[0085] Step 1: Establish a tower crane model parameter library.

[0086] Collect the models of attached tower cranes available for rental from different manufacturers on the market (such as Yongmao Construction Machinery, Xugong Construction Machinery, Zhejiang Construction Machinery, etc.), and record the boom length L of each tower crane model i and rent R i , then the tower crane model can be expressed as T i =(L i ,R i ), on this basis, the tower crane model parameter library T=[T1,T2,……T n ], part of the tower crane model parameter library is shown in Table 1. A total of about 1000 different tower crane models were collected, and only 5 are listed in Table 1.

[0087]

[0088] Step 2: Analyze the tower crane construction plan.

[0089] Step 2.1: Analyze the tower crane construction plan and obtain the graphic contour parameter information of the building red line, construction road, ancillary facilities and the planned building in the tower crane construction plan. Figure 2 Building 1, Building 2 and Building 3 are planned buildings.

[0090] Step 2.2: Establish a plane rectangular coordinate system with the lower left corner of the building red line as the coordinate origin (0,0).

[0091] Step 2.3: In the plane rectangular coordinate system, set a rectangular area. The range of the rectangular area on the horizontal x-axis is the projection range of the building red line on the horizontal x-axis [0,300], and the range on the vertical y-axis is the projection range of the building red line on the vertical y-axis [0,150]. The rectangular area can be expressed as A=[0,300]×[0,150]. Grid the rectangular area, including but not limited to dividing the rectangular area into multiple rectangular grids with a spacing of 5m, and marking each vertex of the rectangular grid, such as Figure 2 As shown, each vertex can be numbered from top to bottom and from left to right; each vertex of the rectangular network can also be numbered in a clockwise order; each vertex of the rectangular network can also be numbered in a counterclockwise order.

[0092] Step 3: Determine the plane position boundary range of the center of the tower crane.

[0093] Step 3.1: Based on the graphic information of the construction plan obtained by parsing in step 2, establish the graphic contour line function f(x,y) of the building red line, construction road, ancillary facilities and planned building in the tower crane construction plan.

[0094] Step 3.2: Expand the horizontal safety distance along the normal direction of the building red line, construction road, ancillary facilities and the graphic contour line of the proposed building in the tower crane construction plan According to the specification requirements, take , and obtain the plane position boundary contour function f'(x,y) of the center of the tower crane body. Thus, the plane position boundary range of the center of the tower crane body within the building red line is determined to be B, such as Figure 3 The shaded area in the figure. According to the specification requirements, the horizontal distance between the center of the tower crane and the building red line, construction road, ancillary facilities and the planned building must be greater than the safety distance.

[0095] Step 3.3: Filter the vertices of the rectangular grid divided in step 2, and select some vertices whose plane coordinates are within the boundary range B to form a set of optional points at the center of the tower crane { },like Figure 4 shown.

[0096] Step 4: Establish the constraints and optimization objectives for the tower group layout.

[0097] Step 4.1: Establish the constraints for the tower group layout, including:

[0098] Step 4.1.1: According to formula (1), the horizontal distance between the end of any tower crane boom and the center of the adjacent tower crane body must meet the requirements of the specification. The specific calculation formula is:

[0099] (1);

[0100] In the formula, d(T i ,T j ) is tower crane T i And tower crane T j Horizontal distance from the center of the tower; x i , y i and; x j , y j They are tower crane T i And tower crane T j The horizontal and vertical coordinates of the center of the tower; L i , L j They are tower crane T i And tower crane T j The length of the crane boom; d2 is the horizontal safety distance between the end of the tower crane boom and the center of the adjacent tower crane tower body required by the specification. According to the requirements of the specification, d2=2m.

[0101] Taking the 1# and 2# tower cranes in the final tower group layout plan as an example, the specific calculation process is as follows:

[0102]

[0103]

[0104]

[0105] L1+d2=37.14+2=39.14m;

[0106] L2+d2=51.62+2=53.62m;

[0107] From this we can see that d(1# tower crane, 2# tower crane)>L1+d2 and d(1# tower crane, 2# tower crane)>L2+d2, and the verification passed.

[0108] Step 4.1.2: For any proposed building, its building plane shall be completely covered by the lifting range of the tower crane to avoid secondary handling.

[0109] Step 4.2: Establish optimization objectives for the tower group layout, including:

[0110] Step 4.2.1: Determine the sum of the rental costs of all tower cranes according to formula (2); the sum of the rental costs of all tower cranes should be as low as possible to reduce the economic cost of construction;

[0111] (2);

[0112] Where R is the total rental cost of all tower cranes; n is the number of tower crane models in the tower group layout plan; q i The tower crane model is T iThe number of tower cranes; R i The tower crane model is T i of rent.

[0113] Step 4.2.2: Determine the cross-operation area ratio of the tower groups according to formula (3) and formula (4). The cross-operation area ratio of the tower groups should be as small as possible to avoid too dense arrangement of tower cranes. The specific calculation formula is:

[0114] (3);

[0115] (4);

[0116] Where m is the total number of tower cranes in the tower group layout plan; r A is the cross-operation area ratio of the tower group, that is, the ratio of the sum of the cross-operation areas between all tower cranes to the sum of the operating areas of all tower cranes; A(T i ,T j ) is tower crane T i And tower crane T j The cross-operation area; A i It is a tower crane T i working area.

[0117] Step 4.2.3: Considering the two optimization objectives comprehensively, the objective function of the tower group layout is obtained as formula (5);

[0118] (5);

[0119] Where OBJ is the objective function of the tower group layout.

[0120] Step 5: Use the improved genetic algorithm to determine the tower group selection and preliminary layout.

[0121] Step 5.1: Use binary coding to encode a single tower group plan layout scheme. Take the final tower group plan layout scheme as an example, where the coding uses formula (6);

[0122] (6);

[0123]

[0124] Where I is the individual tower group layout plan; T1, T n are the binary representations of the serial numbers of the first and nth crane models respectively; q 1、 q n The tower crane models are T1 and T n Binary representation of the number of cranes; P 11 , are the binary representations of the numbers of the points where the centers of the tower bodies of the first and q1 tower cranes of model T1 are located; P n1 , The tower crane models are T n The first and q n The binary representation of the number of the point where the center of the tower crane is located.

[0125] Step 5.2: Initialize the population. When initializing the population, the initial number of individuals Q(I) of the population can be preset based on engineering experience. In this embodiment, Q(I)=50 is taken. Each individual is generated using a random number, and individuals that do not meet the requirements of step 3.3 and step 4.1 are filtered until the initial number of individuals is reached. A fitness function for evaluating individual fitness is established according to formula (7);

[0126] (7);

[0127] In the formula, Is individual I I The fitness value of The larger the value, the more individual I I The higher the fitness, the higher the OBJ(I I ) is individual I I The objective function value of .

[0128] Taking the final tower group layout plan as an example, the specific calculation process is as follows:

[0129] ;

[0130] ;

[0131] ;

[0132] .

[0133] It can be seen that the fitness of the individuals in the final tower group layout plan is 0.164. Similarly, the fitness of each individual in the population is calculated according to the fitness function.

[0134] Step 5.3: Select the top 10% of the best individuals in terms of fitness. I Perform local search optimization and slightly adjust individual I I The crane model, quantity or location in the figure is filtered out, invalid individuals and individuals that do not meet the requirements of steps 3.3 and 4.1 are generated, and individual I is generated. I The set of neighborhood individuals .

[0135] Step 5.4: For each individual I I, calculate the set I I(n) The fitness value of each neighborhood individual in , take the neighborhood individual with the highest fitness value and higher than the fitness value of the original individual, and replace it with the original individual I I .

[0136] Step 5.5: Keep the excellent individuals after replacement and directly enter the next generation, where the next generation refers to the next iteration.

[0137] Step 5.6: Use roulette wheel selection as the selection operator to make the probability of each individual entering the next generation equal to the ratio of its fitness value to the sum of the fitness values ​​of all individuals in the entire population, and select individuals with better fitness to enter the next generation.

[0138] Step 5.7: Use multi-point crossover as the crossover operator, set crossover points at the end of the binary codes of the crane model number, the number of cranes corresponding to the crane model, and the number of the center of the tower body of each crane, and then exchange genes between individuals to generate new individuals, and filter out invalid individuals and individuals that do not meet steps 3.3 and 4.1.

[0139] Step 5.8: Use basic position mutation as the mutation operator, perform mutation operation on any position on the individual according to a certain mutation probability, generate new individuals, and filter out invalid individuals and individuals that do not meet steps 3.3 and 4.1.

[0140] Step 5.9: Aggregate the new individuals generated in steps 5.3 to 5.8 to form a new population.

[0141] Step 5.10: Repeat steps 5.2 to 5.9 until an individual reaches the fitness threshold ,Pick , and the individual is used as the tower group selection and preliminary layout plan for the project. The number of tower groups and the selection of each tower crane are shown in Table 2. The tower crane models of 1# tower crane and 3# tower crane are the same.

[0142]

[0143] Step 6: Use gradient descent method to optimize the floor plan.

[0144] Step 6.1: For the tower group layout plan individual I output in step 5.7, the tower crane model and the corresponding tower crane quantity have been determined. It is necessary to further iterate and refine the tower center of each tower crane. First, the individual is expressed in a new form according to formula (8). Taking the tower group layout plan finally adopted as an example, it is specifically expressed as follows:

[0145] ;

[0146] In the formula, is a new representation of individual I; , are the numbers of the first and nth tower crane models respectively; , The tower crane models are , The number of tower cranes; , and , The tower crane models in the initial state are The first and The horizontal and vertical coordinates of the center of the tower crane; , and , The tower crane models in the initial state are The first and The horizontal and vertical coordinates of the center of the tower crane.

[0147] Step 6.2: Select the learning rate α=0.1m as the step size for updating the horizontal and vertical coordinates of the tower crane center in each iteration to perform gradient descent iteration update to gradually optimize the layout of each tower crane. Since the tower crane model and the corresponding number of tower cranes have been determined, the original objective function is modified to obtain the new objective function as formula (9);

[0148] (9);

[0149] Where OBJ' is the new objective function, r A It is the cross-operation area ratio of the tower groups.

[0150] Taking the final tower group layout plan as an example, the specific calculation process is as follows:

[0151] ;

[0152] ;

[0153] It can be seen that the cross-operation area ratio of the tower group in the final tower group plan layout is 0.037, which is significantly better than the 0.052 of the final output individual in step 5.

[0154] Taking the final tower group layout plan as an example, the embodiment of the present invention is the k=2136th iteration. For the 1# tower crane, the 2# tower crane and the 3# tower crane, the gradient of the new objective function with respect to the coordinates is calculated. , that is, the partial derivative of the new objective function at this point is .

[0155] According to the calculated partial derivatives, the tower center of each tower crane is updated. The specific calculation is as follows:

[0156] (10);

[0157] The final tower group layout plan (including the tower center of each tower crane) is as follows: Figure 5 shown.

[0158] Step 6.3: Calculate the total gradient norm according to formula (11). Taking the final tower group layout scheme as an example, the specific calculation formula is as follows:

[0159] (11);

[0160] In the formula, is the total gradient norm, is the partial derivative of the new objective function at this point.

[0161] Step 6.4: Repeat steps 6.2 to 6.3. After each iteration, determine whether the gradient norm is less than the set threshold. If the conditions are met, the gradient descent process is considered to have converged, and the tower center of each current tower crane is output as the final plane layout of the tower group.

[0162] The present invention is not limited to the above-mentioned specific implementation modes. Obviously, the above-mentioned embodiments are only some embodiments of the embodiments of the present invention, but not all embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention. Those skilled in the art can make other levels of modifications and changes to the present invention. In this way, if these modifications and changes of the present invention fall within the scope of the claims of the present invention, the present invention is also intended to include these changes and changes.

Claims

1. A method for selecting and arranging a group of towers, characterized in that: include: Step 1: Establish a tower crane model parameter library; Step 2: parsing the tower crane construction plan; in step 2, the method for parsing the tower crane construction plan includes: Step 2.1: Analyze the tower crane construction plan to obtain the graphic contour line parameter information of the building red line, construction road, ancillary facilities and the planned building in the tower crane construction plan; Step 2.2: Take the lower left corner of the building red line as the coordinate origin (0,0) and establish a plane rectangular coordinate system; Step 2.3: In the plane rectangular coordinate system, set a rectangular area. The range of the rectangular area on the horizontal x-axis is the projection range of the building red line on the horizontal x-axis. , the range on the vertical y-axis is the projection range of the building red line on the vertical y-axis , then the rectangular area is represented by ; Meshing the rectangular area, dividing the rectangular area into a plurality of rectangular grids at a predetermined interval, marking each vertex of the rectangular grid and numbering each vertex; Step 3: Determine the plane position boundary range of the center of the tower crane body; in step 3, the method for determining the plane position boundary range of the center of the tower crane body includes: Step 3.1: Based on the graphic information of the construction plan obtained in step 2, establish the graphic contour line function of the building red line, construction road, ancillary facilities and proposed building in the tower crane construction plan f ; Step 3.2: Expand the horizontal safety distance along the normal direction of the building red line, construction road, ancillary facilities and the graphic contour line of the proposed building in the tower crane construction plan , get the plane position boundary contour function of the center of the tower crane , thus determining the plane position boundary range of the tower crane tower center within the building red line as ; Step 3.3: Filter the vertices of the rectangular grid divided in step 2 and take the plane coordinates within the boundary range Some vertices in the tower form a set of optional points in the center of the tower crane ; Step 4: Establishing constraints and optimization targets for the plane layout of the tower group; in step 4, the method for establishing constraints and optimization targets for the plane layout of the tower group includes: Step 4.1: Establish the constraints for the tower group layout, including: Step 4.1.1: According to formula (1), verify that the horizontal distance between the end of any tower crane boom and the center of the adjacent tower crane tower meets the specification requirements; （1）； In the formula, It's a tower crane and tower crane Horizontal distance from the center of the tower; , and , Tower crane and tower crane The horizontal and vertical coordinate values ​​of the center of the tower; , Tower crane and tower crane The length of the lifting arm; It is the horizontal safety distance between the end of the tower crane boom and the center of the adjacent tower crane tower body required by the specification; Step 4.1.2: The building plane of any proposed building is completely covered by the lifting range of the tower crane; Step 4.2: Establish optimization objectives for the tower group layout, including: Step 4.2.1: Determine the sum of the rental costs of all tower cranes according to formula (2); （2）； In the formula, is the total rental cost of all tower cranes; is the number of tower crane types in the tower group layout plan; The tower crane model is The number of tower cranes; The tower crane model is the rent; Step 4.2.2: Determine the cross-operation area ratio of the tower cluster according to formula (3) and formula (4) （3）； （4）； In the formula, is the total number of tower cranes in the tower group layout plan; is the cross-operation area ratio of the tower groups; It's a tower crane and tower crane Cross-operation area; It's a tower crane operating area; Step 4.2.3: Considering the two optimization objectives comprehensively, the objective function of the tower group layout is obtained as formula (5); （5）； In the formula, is the objective function of the tower group layout; Step 5: Using an improved genetic algorithm to determine the tower group selection and preliminary plane layout; In step 5, the method of using an improved genetic algorithm to determine the tower group selection and preliminary plane layout includes: Step 5.1: Use binary coding to encode the plan layout of a single tower group, where the coding adopts formula (6); （6）； In the formula, It is an individual plan layout plan of a single tower group; , are the binary representations of the serial numbers of the first and nth tower crane models respectively; , The tower crane models are , Binary representation of the number of cranes; , The tower crane models are The first and The binary representation of the number of the point where the center of the tower crane is located; , The tower crane models are The first and The binary representation of the number of the point where the center of the tower crane is located; Step 5.2: Initialize the population, and set the initial number of individuals in the population to , use random numbers to generate each individual, and filter out individuals that do not meet the requirements of step 3.3 and step 4.1 until the initial number of individuals is reached; establish a fitness function for evaluating individual fitness according to formula (7); （7）； In the formula, Is an individual The fitness value of Is an individual The objective function value of According to the fitness function, calculate the fitness of each individual in the population; Step 5.3: Select the top 10% of the best individuals in terms of fitness. Perform local search optimization and make small adjustments to individual The crane model, quantity or location in the filter is used to filter out invalid individuals and individuals that do not meet the requirements of steps 3.3 and 4.1, and generate individuals The neighborhood individuals set ; Step 5.4: For each individual , calculate the set The fitness value of each neighborhood individual in , take the neighborhood individual with the highest fitness value that is higher than the fitness value of the original individual, and replace it with the original individual ; Step 5.5: Keep the excellent individuals after replacement and directly enter the next generation; Step 5.6: Use roulette wheel selection as the selection operator, so that the probability of each individual entering the next generation is equal to the ratio of its fitness value to the sum of the fitness values ​​of all individuals in the entire population, and select individuals with better fitness to enter the next generation; Step 5.7: Use multi-point crossover as the crossover operator, set crossover points at the end of the binary code of the crane model number, the number of cranes corresponding to the crane model, and the number of the center of the tower body of each crane, and then exchange genes between individuals to generate new individuals, and filter out invalid individuals and individuals that do not meet the requirements of steps 3.3 and 4.1; Step 5.8: Use basic position mutation as the mutation operator to perform mutation operation on any position of the individual according to a certain mutation probability to generate new individuals, and filter out invalid individuals and individuals that do not meet the requirements of steps 3.3 and 4.1; Step 5.9: Aggregate the new individuals generated in steps 5.3 to 5.8 to form a new population; Step 5.10: Repeat steps 5.2 to 5.9 until an individual reaches the fitness threshold , using this individual as the project's tower group selection and preliminary plan layout plan; Step 6: Optimizing the plane layout by using the gradient descent method; In step 6, the method of optimizing the plane layout by using the gradient descent method includes: Step 6.1: Individual tower group layout plan output from step 5.7 , the tower crane model and the corresponding tower crane quantity have been determined, and the tower center of each tower crane is further iterated and refined. First, the individual is expressed in a new form according to formula (8); ； In the formula, Is an individual New representation of , are the numbers of the first and nth tower crane models respectively; , The tower crane models are , The number of tower cranes; , and , The tower crane models in the initial state are The first and The horizontal and vertical coordinates of the center of the tower crane; , and , The tower crane models in the initial state are The first and The horizontal and vertical coordinates of the center of the tower crane; Step 6.2: Perform gradient descent iterative update to gradually optimize the layout of each tower crane; modify the original objective function according to the determined tower crane model and the corresponding number of tower cranes, and obtain the new objective function as formula (9); （9）； In the formula, is the new objective function, is the cross-operation area ratio of the tower groups; Assume that the current Iterations, for each crane , calculate the gradient of the new objective function with respect to the coordinates , the partial derivative of the new objective function at this point is ; According to the calculated partial derivatives, the tower center of each tower crane is updated. The specific calculation formula (10) is: （10）； Step 6.3: Calculate the total gradient norm according to formula (11); （11）； In the formula, is the total gradient norm, is the partial derivative of the new objective function at this point; Step 6.4: Repeat steps 6.2 to 6.

3. After each iteration, determine whether the gradient norm is less than the set threshold. If the conditions are met, the gradient descent process is considered to have converged, and the tower center of each current tower crane is output as the final plane layout of the tower group.

2. The tower group selection and plane layout method according to claim 1 is characterized in that: In step 1, the method for establishing a tower crane model parameter library is to collect the attached tower crane models available for rental, record the boom length of each tower crane model, and and rent , then the tower crane model can be expressed as On this basis, we integrate and establish the tower crane model parameter library .

3. The tower group selection and plane layout method according to claim 1 is characterized in that: In step 2.2, the lower left vertex is replaced by other vertices of the building red line as the coordinate origin (0,0) to establish a plane rectangular coordinate system.

4. The tower group selection and plane layout method according to claim 1 is characterized in that: In step 2.3, the rectangular area is divided into a plurality of rectangular grids with a predetermined spacing of 5 m.

5. The tower group selection and plane layout method according to claim 1 is characterized in that: In step 6.2, the learning rate is , as the step size for updating the horizontal and vertical coordinates of the center of the tower crane in each iteration.

Images