An automatic generation method of high-rise building core tube based on differential evolution algorithm

Through differential evolution algorithm, the mutual conversion relationship and advantage and disadvantage evaluation function of the core cylinder and vector of high-rise buildings is constructed, which solves the problem of core cylinder design dependence in the existing technology, and realizes efficient generation of core cylinder layout that meets fire protection specifications, and improves the design Efficiency and safety.

CN117892386BActive Publication Date: 2024-11-08CHONGQING UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311711316.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-13
Publication Date
2024-11-08
Estimated Expiration
2043-12-13

AI Technical Summary

Technical Problem

The existing technology lacks efficient modeling methods and generation methods to design the core cylinder of high-rise buildings. Especially when considering the stability and fire prevention needs of buildings under external forces such as earthquakes and wind loads, the design depends on the designer's Experience, lack of automated and optimized design processes.

Method used

Using a differential evolution algorithm method, by constructing the mutual conversion relationship and advantages and disadvantage evaluation function of the core cylinder and vector, the core cylinder vector is randomly generated, and iterative update is used to select the variant cross-select iteratively until the iteration stop condition is met, and the minimum area that meets the fire protection specification is generated. Core cylinder layout.

Benefits of technology

Generate a core cylinder layout that complies with fire protection specifications in a short time, improving design efficiency and quality, and ensuring the safety and use needs of the building.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117892386B_ABST
    Figure CN117892386B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, comprising the following steps: 1) constructing a mutual conversion relationship between a core tube of a high-rise building and a vector, and a core tube quality evaluation function; 2) using the core tube quality evaluation function to evaluate the core tube vector to obtain a quality evaluation result; if there is a quality evaluation result that satisfies the iteration stop condition, the core tube vector corresponding to the quality evaluation result is output as the optimal individual; 3) using the mutual conversion relationship between the core tube of a high-rise building and the vector, the optimal individual is converted into a core tube layout of a high-rise building. The present invention proposes a modeling method for mutual conversion between a core tube and a vector, and further uses a differential evolution algorithm to explore a reasonable and compliant core tube arrangement method with the smallest area, so that a core tube that meets fire protection specifications can be generated in a short time and deployed in actual production.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of information technology, and in particular to an automatic generation method of a high-rise building core tube based on a differential evolution algorithm. Background Art

[0002] The core tube is an important structural component in a building. Its design needs to consider the stability of the building under external forces such as earthquakes and wind loads to ensure the safety of the entire building structure. It includes important equipment and passages such as elevators, stairs, pipes, and ventilation systems. The core tube design must take these functional requirements into consideration to ensure that it is effective and meets the use requirements and fire prevention and disaster prevention requirements of the building. In reality, the core tube design often relies on the designer's experience and there is no effective modeling and generation method. Therefore, an efficient high-rise building core tube modeling and automatic generation method can provide great help. Summary of the invention

[0003] The purpose of the present invention is to provide a method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, comprising the following steps:

[0004] 1) Construct the mutual conversion relationship between the core tube of high-rise buildings and vectors, as well as the core tube quality evaluation function;

[0005] 2) randomly generating a plurality of distributed core barrel vectors to represent the initial population, and using the core barrel quality evaluation function to evaluate these randomly generated core barrel vectors to obtain a quality evaluation result;

[0006] If there is a good or bad evaluation result that meets the iteration stop condition, the core cylinder vector corresponding to the good or bad evaluation result is output as the optimal individual, and jump to step 4), otherwise, go to step 3);

[0007] 3) Iteratively update the individuals in the population by using the mutation crossover selection method, and calculate the quality evaluation results of the individuals after iterative update, until there is a quality evaluation result that meets the iteration stop condition, and output the core cylinder vector corresponding to the quality evaluation result as the optimal individual;

[0008] 4) Using the mutual conversion relationship between the core tube of the high-rise building and the vector, the optimal individual is converted into the core tube layout of the high-rise building.

[0009] Furthermore, after the optimal individual is converted into the core tube layout of the high-rise building in step 4), the core tube layout of the high-rise building is also detected to identify and record the wall type of each room and the type of each antechamber, and doors and windows are opened according to the wall type, and air shafts are set in the antechamber.

[0010] Furthermore, the core tube quality evaluation function is used to determine the core tube layout with the smallest building area while complying with fire protection regulations.

[0011] Furthermore, the mutual conversion relationship between the core tube of the high-rise building and the vector includes a vector expression for describing the core tube of the high-rise building, and a method for generating the layout of the core tube of the high-rise building based on the vector expression.

[0012] Further, the vector expression is an N-dimensional vector;

[0013] Among them, the N1-dimensional vector is used to describe the scissor ladder information of the core tube of a high-rise building;

[0014] The N2-dimensional vector is used to describe the elevator information of the core tube of a high-rise building;

[0015] The N3-dimensional vector is used to describe the fire elevator information of the core tube of a high-rise building;

[0016] The N4-dimensional vector is used to describe the strong and weak common electric well information of the core tube of a high-rise building;

[0017] The N5-dimensional vector is used to describe the water well information of the core tube of a high-rise building; N = N1 + N2 + N3 + N4 + N5.

[0018] Further, the scissors ladder information includes the horizontal coordinate and the vertical coordinate of the lower left corner of the scissors ladder, the length and width of the front room, the rotation direction of the scissors ladder, the direction of the front room on the scissors ladder, and the direction of the front room on the scissors ladder;

[0019] The elevator information includes the horizontal coordinate, vertical coordinate, rotation direction of the lower left corner of the elevator, and the length and width of the elevator lobby;

[0020] The fire elevator information includes the horizontal coordinate, vertical coordinate, rotation direction of the lower left corner of the fire elevator, and the length and width of the fire elevator vestibule;

[0021] The strong and weak common electric well information includes the horizontal coordinate, vertical coordinate, and rotation direction of the lower left corner of the strong and weak common electric well;

[0022] The water well information includes the horizontal coordinate, vertical coordinate, and rotation direction of the lower left corner of the water well. Further, the core tube layout generation method of the high-rise building based on the vector expression includes:

[0023] a) Generate equipment and corresponding antechamber based on vector expressions; the equipment includes scissor ladders, elevators, fire elevators, strong and weak common power wells, and water wells;

[0024] b) Determine whether the antechambers of these devices overlap. If so, further determine whether they need to be shared. If so, integrate the overlapping antechambers into a shared antechamber and record the device type corresponding to the shared antechamber;

[0025] c) Generate corridors to maintain connectivity inside the core tube, traverse the antechambers again, record the number of disconnected antechambers, calculate the number of corridors required and generate corridors.

[0026] Furthermore, the core tube quality evaluation function is as follows:

[0027] f(x)=P overlap (x)+P room (x)+P connect (x)+S traffic-core (x)+α*S min_rectangle (x) (1)

[0028] Where P overlap (x) = 100*S overlap represents the core tube equipment overlap penalty function; S overlap is the sum of the overlapping areas between the devices; P room (x) = 100*n is the penalty function when each antechamber of the core tube does not meet the fire protection regulations; n is the number of antechambers that do not meet the fire protection regulations; P connect (x) = 100*θ is the penalty function when the core tube is not connected; θ = 0 when the evacuation exit is normal and the corridor is not blocked by other equipment, otherwise θ = 1; S traffic-core (x) is the area of ​​the current core tube; S min_rectangle (x) is the area of ​​the smallest rectangle surrounding the current core tube; α is the parameter that controls the degree of relaxation; f(x) is the evaluation result.

[0029] Furthermore, the value of the quality evaluation result f(x) is negatively correlated with the quality of the core tube.

[0030] Furthermore, the steps of iteratively updating individuals in the population by using the mutation crossover selection method include:

[0031] 1) Perform mutation operations on individuals in the population and obtain:

[0032] v i (g) = x r1 +F*(x r2 (g)-x r3 (g)) (2)

[0033] In the formula, x r1 , x r2 , x r3 are three individuals in the current population that are not related to individual x i The same and different individuals; F is the variation factor; v i (g) is the individual x i (g) the corresponding variant individuals;

[0034] 2) Perform crossover operation on individuals to obtain:

[0035]

[0036] In the formula, x i,j (g) is x i (g) the value at the jth position; u i,j (g) is x after the crossover i (g) is the value at the jth position; CR is the crossover probability factor; j rand is a random component, which is used to ensure that at least one dimension of the experimental individual after crossover is provided by the variant individual; v i,j (g) is a variant individual;

[0037] 3) Perform selection operations on individuals and obtain:

[0038]

[0039] In the formula, x i (g+1) is the individual after iterative update. i (g) is the individual after crossover.

[0040] Furthermore, the iteration stopping condition includes that the number of iterations reaches a preset number, or the relative error of the solution between two iterations is less than a preset error threshold.

[0041] The technical effect of the present invention is unquestionable. The present invention proposes a modeling method for the mutual transformation between core tubes and vectors, and further uses a differential evolution algorithm to explore a reasonable and compliant core tube arrangement with the smallest area. It can generate a core tube that meets fire protection regulations in a short time and deploy it in actual production. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 The present invention discloses a flow chart of a method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm;

[0043] Figure 2 The modeling process of converting the core cylinder into a vector is used for the present invention;

[0044] Figure 3 The modeling process of using vectors to transform into core barrels for the present invention;

[0045] Figure 4 A flow chart of using a differential evolution algorithm for the present invention;

[0046] Figure 5 This is a result diagram automatically generated according to the present invention. DETAILED DESCRIPTION

[0047] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.

[0048] Embodiment 1:

[0049] See also Figures 1 to 5 , a method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, comprising the following steps:

[0050] 1) Construct the mutual conversion relationship between the core tube of high-rise buildings and vectors, as well as the core tube quality evaluation function;

[0051] 2) randomly generating a plurality of distributed core barrel vectors to represent the initial population, and using the core barrel quality evaluation function to evaluate these randomly generated core barrel vectors to obtain a quality evaluation result;

[0052] If there is a good or bad evaluation result that meets the iteration stop condition, the core cylinder vector corresponding to the good or bad evaluation result is output as the optimal individual, and jump to step 4), otherwise, go to step 3);

[0053] 3) Iteratively update the individuals in the population by using the mutation crossover selection method, and calculate the quality evaluation results of the individuals after iterative update, until there is a quality evaluation result that meets the iteration stop condition, and output the core cylinder vector corresponding to the quality evaluation result as the optimal individual;

[0054] 4) Using the mutual conversion relationship between the core tube of the high-rise building and the vector, the optimal individual is converted into the core tube layout of the high-rise building.

[0055] Step 4) After converting the optimal individual into the core tube layout of the high-rise building, the core tube layout of the high-rise building is also tested to identify and record the wall type of each room and the type of each antechamber, open doors and windows according to the wall type, and set a wind shaft in the antechamber.

[0056] The core tube quality evaluation function is used to determine the core tube layout with the smallest building area while complying with fire protection regulations.

[0057] The mutual conversion relationship between the core tube of the high-rise building and the vector includes a vector expression for describing the core tube of the high-rise building, and a method for generating the layout of the core tube of the high-rise building based on the vector expression.

[0058] The vector expression is an N-dimensional vector;

[0059] Among them, the N1-dimensional vector is used to describe the scissor ladder information of the core tube of a high-rise building;

[0060] The N2-dimensional vector is used to describe the elevator information of the core tube of a high-rise building;

[0061] The N3-dimensional vector is used to describe the fire elevator information of the core tube of a high-rise building;

[0062] The N4-dimensional vector is used to describe the strong and weak common electric well information of the core tube of a high-rise building;

[0063] The N5-dimensional vector is used to describe the water well information of the core tube of a high-rise building; N = N1 + N2 + N3 + N4 + N5.

[0064] The scissors ladder information includes the horizontal coordinate and the vertical coordinate of the lower left corner of the scissors ladder, the length and width of the front room, the rotation direction of the scissors ladder, the direction of the front room on the scissors ladder, and the direction of the front room on the scissors ladder;

[0065] The elevator information includes the horizontal coordinate, vertical coordinate, rotation direction of the lower left corner of the elevator, and the length and width of the elevator lobby;

[0066] The fire elevator information includes the horizontal coordinate, vertical coordinate, rotation direction of the lower left corner of the fire elevator, and the length and width of the fire elevator vestibule;

[0067] The strong and weak common electric well information includes the horizontal coordinate, vertical coordinate, and rotation direction of the lower left corner of the strong and weak common electric well;

[0068] The water well information includes the horizontal coordinate, the vertical coordinate, and the rotation direction of the lower left corner of the water well.

[0069] The methods for generating the core tube layout of high-rise buildings based on vector expressions include:

[0070] a) Generate equipment and corresponding antechamber based on vector expressions; the equipment includes scissor ladders, elevators, fire elevators, strong and weak common power wells, and water wells;

[0071] b) Determine whether the antechambers of these devices overlap. If so, further determine whether they need to be shared. If so, integrate the overlapping antechambers into a shared antechamber and record the device type corresponding to the shared antechamber;

[0072] c) Generate corridors to maintain connectivity inside the core tube, traverse the antechambers again, record the number of disconnected antechambers, calculate the number of corridors required and generate corridors.

[0073] The core tube quality evaluation function is as follows:

[0074] f(x)=P overlap (x)+P room (x)+P connect (x)+S traffic-core (x)+α*S min_rectangle (x) (1)

[0075] Where P overlap (x) = 100*S overlap represents the core tube equipment overlap penalty function; S overlap is the sum of the overlapping areas between the devices; P room (x) = 100*n is the penalty function when each antechamber of the core tube does not meet the fire protection regulations; n is the number of antechambers that do not meet the fire protection regulations; P connect (x) = 100*θ is the penalty function when the core tube is not connected; θ = 0 when the evacuation exit is normal and the corridor is not blocked by other equipment, otherwise θ = 1; S traffic-core (x) is the area of ​​the current core tube; S min_rectangle (x) is the area of ​​the smallest rectangle surrounding the current core tube; α is the parameter that controls the degree of relaxation; f(x) is the evaluation result.

[0076] The value of the quality evaluation result f(x) is negatively correlated with the quality of the core tube.

[0077] The steps of iteratively updating individuals in the population by mutation crossover selection include:

[0078] 1) Perform mutation operations on individuals in the population and obtain:

[0079] v i (g) = x r1 +F*(x r2 (g)-x r3 (g)) (2)

[0080] In the formula, x r1 , x r2 , x r3 are three individuals in the current population that are not related to individual x i The same and different individuals; F is the variation factor; v i (g) is the individual x i (g) the corresponding variant individuals;

[0081] 2) Perform crossover operation on individuals to obtain:

[0082]

[0083] In the formula, x i,j (g) is x i (g) the value at the jth position; u i,j (g) is x after the crossover i (g) is the value at the jth position; CR is the crossover probability factor; j rand is a random component, which is used to ensure that at least one dimension of the experimental individual after crossover is provided by the variant individual; vi,j (g) is a variant individual;

[0084] 3) Perform selection operations on individuals and obtain:

[0085]

[0086] In the formula, x i (g+1) is the individual after iterative update. i (g) is the individual after crossover.

[0087] The iteration stopping condition includes that the number of iterations reaches a preset number, or the relative error of the solution quality between two iterations is less than a preset error threshold.

[0088] Embodiment 2:

[0089] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm comprises the following steps:

[0090] 1) Construct the mutual conversion relationship between the core tube of high-rise buildings and vectors, as well as the core tube quality evaluation function;

[0091] 2) randomly generating a plurality of distributed core barrel vectors to represent the initial population, and using the core barrel quality evaluation function to evaluate these randomly generated core barrel vectors to obtain a quality evaluation result;

[0092] If there is a good or bad evaluation result that meets the iteration stop condition, the core cylinder vector corresponding to the good or bad evaluation result is output as the optimal individual, and jump to step 4), otherwise, go to step 3);

[0093] 3) Iteratively update the individuals in the population by using the mutation crossover selection method, and calculate the quality evaluation results of the individuals after iterative update, until there is a quality evaluation result that meets the iteration stop condition, and output the core cylinder vector corresponding to the quality evaluation result as the optimal individual;

[0094] 4) Using the mutual conversion relationship between the core tube of the high-rise building and the vector, the optimal individual is converted into the core tube layout of the high-rise building.

[0095] Embodiment 3:

[0096] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as that of Example 2. Furthermore, in step 4), after converting the optimal individual into a core tube layout of a high-rise building, the core tube layout of the high-rise building is also detected to identify and record the wall type of each room and the type of each antechamber, and doors and windows are opened according to the wall type, and a wind shaft is set in the antechamber.

[0097] Embodiment 4:

[0098] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of Examples 2-3. Furthermore, the core tube quality evaluation function is used to determine the core tube layout with the smallest building area while complying with fire protection regulations.

[0099] Embodiment 5:

[0100] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of Examples 2-4. Furthermore, the mutual conversion relationship between the core tube of the high-rise building and the vector includes a vector expression for describing the core tube of the high-rise building, and a method for generating a layout of the core tube of the high-rise building based on the vector expression.

[0101] Embodiment 6:

[0102] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of Embodiments 2-5, and further, the vector expression is an N-dimensional vector;

[0103] Among them, the N1-dimensional vector is used to describe the scissor ladder information of the core tube of a high-rise building;

[0104] The N2-dimensional vector is used to describe the elevator information of the core tube of a high-rise building;

[0105] The N3-dimensional vector is used to describe the fire elevator information of the core tube of a high-rise building;

[0106] The N4-dimensional vector is used to describe the strong and weak common electric well information of the core tube of a high-rise building;

[0107] The N5-dimensional vector is used to describe the water well information of the core tube of a high-rise building; N = N1 + N2 + N3 + N4 + N5.

[0108] Embodiment 7:

[0109] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of embodiments 2-6, and further, the scissor ladder information includes the horizontal coordinate and the vertical coordinate of the lower left corner of the scissor ladder, the length and width of the front room, the rotation direction of the scissor ladder, the direction of the upper front room of the scissor ladder, and the direction of the lower front room of the scissor ladder;

[0110] The elevator information includes the horizontal coordinate, vertical coordinate, rotation direction of the lower left corner of the elevator, and the length and width of the elevator lobby;

[0111] The fire elevator information includes the horizontal coordinate, vertical coordinate, rotation direction of the lower left corner of the fire elevator, and the length and width of the fire elevator vestibule;

[0112] The strong and weak common electric well information includes the horizontal coordinate, vertical coordinate, and rotation direction of the lower left corner of the strong and weak common electric well;

[0113] The water well information includes the horizontal coordinate, the vertical coordinate, and the rotation direction of the lower left corner of the water well.

[0114] Embodiment 8:

[0115] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of Embodiments 2-7, and further, a method for generating a core tube layout of a high-rise building based on a vector expression includes:

[0116] a) Generate equipment and corresponding antechamber based on vector expressions; the equipment includes scissor ladders, elevators, fire elevators, strong and weak common power wells, and water wells;

[0117] b) Determine whether the antechambers of these devices overlap. If so, further determine whether they need to be shared. If so, integrate the overlapping antechambers into a shared antechamber and record the device type corresponding to the shared antechamber;

[0118] c) Generate corridors to maintain connectivity inside the core tube, traverse the antechambers again, record the number of disconnected antechambers, calculate the number of corridors required and generate corridors.

[0119] Embodiment 9:

[0120] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of Embodiments 2-8, and further, the core tube quality evaluation function is as follows:

[0121] f(x)=P overlap (x)+P room (x)+P connect (x)+S traffic-core (x)+α*S min_rectangle (x) (1)

[0122] Where P overlap (x) = 100*S overlap represents the core tube equipment overlap penalty function; S overlap is the sum of the overlapping areas between the devices; P room (x) = 100*n is the penalty function when each antechamber of the core tube does not meet the fire protection regulations; n is the number of antechambers that do not meet the fire protection regulations; P connect (x) = 100*θ is the penalty function when the core tube is not connected; θ = 0 when the evacuation exit is normal and the corridor is not blocked by other equipment, otherwise θ = 1; S traffic-core (x) is the area of ​​the current core tube; S min_rectangle (x) is the area of ​​the smallest rectangle surrounding the current core tube; α is the parameter that controls the degree of relaxation; f(x) is the evaluation result.

[0123] Embodiment 10:

[0124] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of embodiments 2-9, and further, the value of the quality evaluation result f(x) is negatively correlated with the quality of the core tube.

[0125] Embodiment 11:

[0126] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of Embodiments 2-10, and further, the step of iteratively updating individuals in a population by using a mutation crossover selection method comprises:

[0127] 1) Perform mutation operations on individuals in the population and obtain:

[0128] v i (g) = x r1 +F*(x r2 (g)-x r3 (g)) (2)

[0129] In the formula, x r1 , x r2 , x r3 are three individuals in the current population that are not related to individual x i The same and different individuals; F is the variation factor; v i (g) is the individual x i (g) the corresponding variant individuals;

[0130] 2) Perform crossover operation on individuals to obtain:

[0131]

[0132] In the formula, x i,j (g) is x i (g) the value at the jth position; u i,j (g) is x after the crossover i (g) is the value at the jth position; CR is the crossover probability factor; j rand It is a random component, which is used to ensure that at least one dimension of the experimental individuals after crossover is provided by the variant individuals;

[0133] 3) Perform selection operations on individuals and obtain:

[0134]

[0135] In the formula, x i (g+1) is the individual after iterative update.

[0136] Embodiment 12:

[0137] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of Examples 2-11. Furthermore, the iteration stopping condition includes the number of iterations reaching a preset number, or the relative error of the solution quality between two iterations being less than a preset error threshold.

[0138] Embodiment 13:

[0139] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of embodiments 2-12, and further, the preset number of times is 1000 times.

[0140] Embodiment 14:

[0141] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the technical content of which is the same as any one of embodiments 2-13, and further, the preset error threshold is 0.1%.

[0142] Embodiment 15:

[0143] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the steps comprising:

[0144] S101. Conduct mathematical modeling of the core tube, design a method for converting the core tube to a vector, and design a core tube evaluation function to indicate the quality of the core tube.

[0145] S102. Randomly generate multiple core tube vectors as an initial population, and calculate an evaluation value for each vector in the population.

[0146] S103, continuously iteratively updating individual solutions in the population by means of mutation crossover selection until the iteration end condition is reached, and converting the searched optimal vector into a core tube layout.

[0147] S104. An efficient boundary detection method is used to detect the optimal core tube generated by S103, and a wind shaft is set in the front room, and doors and windows are set in each room.

[0148] Step S101 includes:

[0149] The core tube is modeled and position codes are generated. This method mainly records the lower left corner coordinates, directions, and length and width of the elevator, fire elevator, scissor ladder, electric shaft, water well, and each antechamber of the core tube, and generates vectors to represent the core tube.

[0150] The core tube evaluation function is designed to indicate the quality of the core tube. The evaluation function is the area of ​​the core tube when it is reasonable and compliant. The evaluation value is calculated for each core tube solution in the population in S101. The lower the evaluation function value, the lower the construction cost of the core tube and the better the quality of the core tube.

[0151] Step S102 includes:

[0152] According to the modeling method in S101, 1000 core tube vectors are randomly generated, each vector represents a different distribution form inside the core tube, and each core tube vector is evaluated and recorded.

[0153] Step S103 includes:

[0154] The vector population is iterated, and the vector is mutated, crossed, and selected during each iteration. Each iteration can continuously improve the quality of the population while maintaining the diversity of the population. The algorithm ends when the number of algorithm iterations reaches the predetermined maximum number or the optimal solution of the population reaches the predetermined error accuracy.

[0155] Step S104 includes:

[0156] According to the modeling method in S101, the optimal vector found after iteration is converted into the core tube layout. An efficient method for detecting irregular graphic boundaries is used to distinguish the types of each antechamber and set up air shafts; the walls of each room are detected and doors and windows are opened according to the wall type.

[0157] Embodiment 16:

[0158] A method for automatically generating a core tube of a high-rise building based on a differential evolution algorithm, the steps comprising:

[0159] S101. Design the core tube modeling method and evaluation method. The main steps are as follows:

[0160] a) Use a 21-dimensional vector to describe the core tube: Figure 2 As shown in the figure, the first 7 bits represent the information of the scissor ladder, which are: the horizontal coordinate, vertical coordinate, length and width of the front room, rotation direction of the scissor ladder, direction of the upper front room, and direction of the lower front room of the scissor ladder; bits 8-11 are the horizontal coordinate, vertical coordinate, rotation direction, length and width of the elevator lobby at the lower left corner; bits 12-15 are the horizontal coordinate, vertical coordinate, rotation direction, length and width of the fire elevator at the lower left corner; bits 16-18 are the horizontal coordinate, vertical coordinate, and rotation direction of the lower left corner of the strong and weak common power shaft; bits 19-21 are the horizontal coordinate, vertical coordinate, and rotation direction of the lower left corner of the water well. Through this method, the internal layout of the core tube can be accurately described by vectors.

[0161] a) The design vector is converted into the core tube layout process, such as Figure 3 The following steps are shown:

[0162] Generate equipment and antechambers. Generate elevators, fire elevators, scissor ladders, pipe shafts and their respective antechambers according to the coordinates and directions in the vector.

[0163] Generate a shared antechamber. Use breadth-first traversal to determine the overlapping relationship between antechambers, determine whether they need to be shared, integrate overlapping antechambers into a large shared antechamber, and record the type of the shared antechamber.

[0164] Generate corridors to keep the core tube connected. Traverse the antechambers again, record the number of disconnected antechambers, calculate the number of corridors required and generate corridors.

[0165] c) An evaluation function for the core tube vector is proposed to ensure that the core tube can minimize the building area while meeting the requirements of fire prevention and disaster prevention regulations. The evaluation function is:

[0166] f(x)=P overlap (x)+P room (x)+P connect (x)+S traffic-core (x)+α*S min_rectangle (x)

[0167] P overlap (x) = 100*S overlap , represents the core tube equipment overlap penalty function, which is used to punish the overlapping of elevators, fire elevators, scissor ladders, and pipe shafts. S overlap It is the sum of the overlapping areas between the devices.

[0168] P room (x) = 100*n, which is the penalty function when the antechambers of the core tube do not meet the fire protection standards. Penalty is imposed when it is detected that the area and shortest side of the antechamber or shared antechamber do not meet the fire protection standards. n is the number of antechambers that do not meet the fire protection standards.

[0169] P connect (x) = 100*θ, which is the penalty function when the core tube is not connected. When the evacuation exit is normal and the corridor is not blocked by other equipment, θ = 0, otherwise θ = 1.

[0170] S traffic-core (x) is the area of ​​the current core tube. S min_rectangle (x) is the area of ​​the minimum rectangle surrounding the current core tube, α is the parameter that controls the degree of relaxation, generally, α∈[0.1a0.2].

[0171] S102. According to the modeling method of S101, 1000 distributed core barrel vectors are randomly generated in a rectangular space with (0, 0) and (20, 20) at the lower left corner and the upper right corner to represent the initial population, and each core barrel vector in the population is evaluated using the evaluation method of S101.

[0172] S103, using the idea of ​​differential evolution to perform 1000 iterations, such as Figure 4 As shown, each iteration mainly consists of the following steps:

[0173] a) Mutation. Use different vectors to interfere with the target vector and perform differential operations to ensure the diversity of the population. The mutation formula is:

[0174] v i (g) = x r1 +F*(x r2 (g)-x r3 (g))

[0175] x r1 , x r2 , x r3 The three species of the current population that are not related to x i The same and different individuals, F is the variation factor, v i (g) is the individual x i (g) The corresponding mutant individual. When the mutant individual does not meet the boundary requirements, a random mutation vector is regenerated.

[0176] b) Crossover. The crossover between the target vector and the mutation vector is used to increase the parameter diversity of the individual vector. The crossover formula is:

[0177]

[0178] x i,j (g) is x i (g) The value at the jth position, u i,j (g) is x after the crossover i (g) At the jth position, CR = 0.3 is the crossover probability factor, j rand is a random component, ensuring that at least one dimension of the experimental individuals after crossover is provided by the mutant individuals.

[0179] c) Selection. A greedy strategy is adopted to retain the better individuals among the current target individuals and the mutation crossover individuals as the next generation. The selection formula is:

[0180]

[0181] Through the greedy strategy, the vector that makes the evaluation function f(x) smaller is always retained. Through multiple iterations of crossover mutation selection, the quality of the vector is continuously improved while ensuring the diversity of the population, preventing it from falling into the local optimal solution. Until the number of iterations is reached or the optimal solution of the population reaches the predetermined error accuracy, the currently searched optimal solution is output.

[0182] S104: For the output optimal solution vector, first convert it into a core tube layout according to the steps in S101, scan the entire core tube using scanning lines parallel to the x-axis and y-axis respectively, and record the walls of each room that belong to the overall boundary. Identify the type of each room, set up a wind shaft on the boundary wall of the front room, and set a lighting window on the boundary wall of the scissor ladder.

[0183] The above experimental results prove that by using mathematical methods to model the core tube layout, proposing an evaluation function for the core tube layout, and further using the differential evolution algorithm, while ensuring the search accuracy, a core tube layout that meets fire protection regulations can be generated in a short time and can be deployed in actual production.

Claims

1. A method for automatically generating the core tube of a high-rise building based on a differential evolution algorithm, characterized in that: The following steps are involved: 1) Construct the mutual conversion relationship between the core tube of high-rise buildings and vectors, as well as the core tube quality evaluation function; 2) randomly generating a plurality of distributed core barrel vectors to represent the initial population, and using the core barrel quality evaluation function to evaluate these randomly generated core barrel vectors to obtain a quality evaluation result; If there is a good or bad evaluation result that meets the iteration stop condition, the core cylinder vector corresponding to the good or bad evaluation result is output as the optimal individual, and jump to step 4), otherwise, go to step 3); 3) Iteratively update the individuals in the population by using the mutation crossover selection method, and calculate the quality evaluation results of the individuals after iterative update, until there is a quality evaluation result that meets the iteration stop condition, and output the core cylinder vector corresponding to the quality evaluation result as the optimal individual; 4) Using the mutual conversion relationship between the core tube of a high-rise building and the vector, the optimal individual is converted into the core tube layout of the high-rise building; The mutual conversion relationship between the core tube of the high-rise building and the vector includes a vector expression for describing the core tube of the high-rise building, and a method for generating the layout of the core tube of the high-rise building based on the vector expression; The vector expression is an N-dimensional vector; Among them, the N1-dimensional vector is used to describe the scissor ladder information of the core tube of a high-rise building; The N2-dimensional vector is used to describe the elevator information of the core tube of a high-rise building; The N3-dimensional vector is used to describe the fire elevator information of the core tube of a high-rise building; The N4-dimensional vector is used to describe the strong and weak common electric well information of the core tube of a high-rise building; The N5-dimensional vector is used to describe the water well information of the core tube of a high-rise building; N = N1 + N2 + N3 + N4 + N5; The scissors ladder information includes the horizontal coordinate and the vertical coordinate of the lower left corner of the scissors ladder, the length and width of the front room, the rotation direction of the scissors ladder, the direction of the front room on the scissors ladder, and the direction of the front room on the scissors ladder; The elevator information includes the horizontal coordinate, vertical coordinate, rotation direction of the lower left corner of the elevator, and the length and width of the elevator lobby; The fire elevator information includes the horizontal coordinate, vertical coordinate, rotation direction of the lower left corner of the fire elevator, and the length and width of the fire elevator vestibule; The strong and weak common electric well information includes the horizontal coordinate, vertical coordinate, and rotation direction of the lower left corner of the strong and weak common electric well; The water well information includes the horizontal coordinate, vertical coordinate, and rotation direction of the lower left corner of the water well; The methods for generating the core tube layout of high-rise buildings based on vector expressions include: a) Generate equipment and corresponding antechamber based on vector expressions; the equipment includes scissor ladders, elevators, fire elevators, strong and weak common power wells, and water wells; b) Determine whether the antechambers of these devices overlap. If so, further determine whether they need to be shared. If so, integrate the overlapping antechambers into a shared antechamber and record the device type corresponding to the shared antechamber; c) Generate corridors to maintain connectivity inside the core tube, traverse the antechambers again, record the number of disconnected antechambers, calculate the number of corridors required and generate corridors.

2. According to the method for automatically generating the core tube of a high-rise building based on the differential evolution algorithm according to claim 1, it is characterized in that: Step 4) After converting the optimal individual into the core tube layout of the high-rise building, the core tube layout of the high-rise building is also tested to identify and record the wall type of each room and the type of each antechamber, open doors and windows according to the wall type, and set a wind shaft in the antechamber.

3. According to the method for automatically generating the core tube of a high-rise building based on a differential evolution algorithm according to claim 1, it is characterized in that: The core tube quality evaluation function is used to determine the core tube layout with the smallest building area while complying with fire protection regulations.

4. The method for automatically generating a high-rise building core tube based on a differential evolution algorithm according to claim 1 is characterized in that: The core tube quality evaluation function is as follows: f(x)=P overlap (x)+P room (x)+P connect (x)+S traffic-core (x)+α*S min_rectangle (x) (1) Where P overlap (x) = 100*S overlap represents the core tube equipment overlap penalty function; S overlap is the sum of the overlapping areas between the devices; P room (x) = 100*n is the penalty function when each antechamber of the core tube does not meet the fire protection regulations; n is the number of antechambers that do not meet the fire protection regulations; P connect (x) = 100*θ is the penalty function when the core tube is not connected; θ = 0 when the evacuation exit is normal and the corridor is not blocked by other equipment, otherwise θ = 1; S traffic-core (x) is the area of ​​the current core tube; S min_rectangle (x) is the area of ​​the smallest rectangle surrounding the current core tube; α is the parameter that controls the degree of relaxation; f(x) is the evaluation result.

5. According to the method of automatic generation of high-rise building core tube based on differential evolution algorithm according to claim 4, it is characterized in that: The value of the quality evaluation result f(x) is negatively correlated with the quality of the core tube.

6. The method for automatically generating a high-rise building core tube based on a differential evolution algorithm according to claim 1 is characterized in that: The steps of iteratively updating individuals in the population by mutation crossover selection include: 1) Perform mutation operations on individuals in the population and obtain: v i (g)=x r1 (g)+F*(x r2 (g)-x r3 (g)) (2) In the formula, x r1 ,x r2 ,x r3 are three individuals in the current population that are not related to individual x i (g) The same and different individuals; F is the variation factor; v i (g) is the individual x i (g) the corresponding variant individuals; 2) Perform crossover operation on individuals to obtain: In the formula, x i,j (g) is x i (g) the value at the jth position; u i,j (g) is x after the crossover i (g) is the value at the jth position; CR is the crossover probability factor; j rand is a random component, which is used to ensure that at least one dimension of the experimental individual after crossover is provided by the variant individual; v i,j (g) is a variant individual; 3) Perform selection operation on individuals and obtain: In the formula, x i (g+1) is the individual after iterative update; u i (g) is the individual after crossover.

7. The method for automatically generating a high-rise building core tube based on a differential evolution algorithm according to claim 1 is characterized in that: The iteration stopping condition includes that the number of iterations reaches a preset number, or the relative error of the solution between two iterations is less than a preset error threshold.