Intelligent layout system and method of heavy construction machinery based on machine learning

Abstract

The present invention discloses a system and method for intelligent layout of heavy construction machinery based on machine learning, which belongs to the technical field of construction machinery. The system includes a first machine learning model, a load position layout module, a second machine learning model and a force verification result marking module. The system can automatically obtain plate structure information, beam structure information, and material property information through the first machine learning model, and obtain N groups of working conditions based on the load layout through the load position layout module, and then automatically verify the force verification results of the beam-slab structure corresponding to the N groups of working conditions through the second machine learning model, and use different colors to distinguish different verification results through the force verification result marking module to obtain a heavy construction machinery layout diagram. Therefore, the system can quickly output a heavy construction machinery layout diagram, and compared with traditional technologies, it has the advantages of fast layout speed, simple operation, no reliance on human experience, and strong adaptability.

Description

Technical Field

[0001] The present invention relates to a heavy construction machinery intelligent arrangement system and method based on machine learning, belonging to the technical field of building machinery construction. Background Art

[0002] During the construction process, heavy construction machinery such as passenger and freight elevators, truck cranes, crawler cranes, etc. are often arranged on the top slabs of various basements. At this time, the beam and slab structure of the basement top slab will bear loads exceeding the design requirements. Therefore, when determining the layout position of heavy construction machinery, it is necessary to verify the force of the beam and slab structure of the basement top slab.

[0003] At present, when arranging heavy construction machinery, finite element software is often used to establish a structural model for verification. For different heavy construction machinery, different finite element models need to be established, and the load position needs to be repeatedly modified. The modeling operation process is cumbersome and time-consuming.

[0004] Therefore, it is necessary to develop a heavy construction machinery intelligent layout system and method based on machine learning. Summary of the invention

[0005] The present invention provides a system and method for intelligent layout of heavy construction machinery based on machine learning, which is used to solve the problem that the traditional method of arranging existing heavy construction machinery on the top plate of a building such as a basement has a cumbersome operation process and a long time consumption.

[0006] In order to solve the above technical problems, the present invention includes the following technical solutions:

[0007] A heavy construction machinery intelligent layout system based on machine learning, comprising:

[0008] The first machine learning model can output the slab structure information, beam structure information, and material property information after inputting the basement top slab floor plan and the general structural design description;

[0009] The load location arrangement module can automatically arrange the load in different areas of the plate according to the load information of heavy construction machinery and the plate structure information. The number of areas is recorded as N, corresponding to N groups of working conditions;

[0010] The second machine learning model can input plate structure information, beam structure information, material property information, load information and N groups of working conditions, and then output the force verification results of the beam-slab structure corresponding to the N groups of working conditions;

[0011] The force verification result marking module can mark the failed areas in the force verification results of the beam-slab structure with a first color, mark the passed areas with a second color, and output the marked heavy construction machinery layout drawing.

[0012] Furthermore, the heavy construction machinery intelligent arrangement system based on machine learning also includes:

[0013] The first information collection and processing module is capable of collecting a first training data set; the first training data set includes a plurality of groups of first training data; each group of first training data includes a basement top plate plan layout and a general description of the structural design, as well as pre-processed plate structure information, beam structure information, and material property information;

[0014] The first model training module trains the first machine learning model according to the training data set preprocessed by the first information acquisition and processing module.

[0015] Furthermore, the heavy construction machinery intelligent arrangement system based on machine learning also includes:

[0016] The second information collection and processing module is capable of collecting a second training data set; the second training data set includes a plurality of groups of second training data; each group of second training data includes various heavy machinery instruction manuals, relevant construction plans, relevant calculation books, and pre-processed plate structure information, beam structure information, material property information, load information, and beam-slab structure verification results;

[0017] The second model training module trains the second machine learning model according to the training data set preprocessed by the second information acquisition and processing module.

[0018] Furthermore, the plate structure information includes the plate reinforcement information, plate span length L, plate span width B, plate thickness h b ;

[0019] Beam structure information includes beam reinforcement information, beam centerline, beam length L l , beam section width B l , beam section height H l ;

[0020] The material property information includes the elastic modulus E of the component material and the Poisson's ratio ν of the component material.

[0021] Furthermore, the preprocessing process is as follows: labeling the plate structure information, beam structure information, and material property information; extracting the parameter values ​​and units in the labels, and converting the parameter units into preset units; and numbering the beam and plate components in sequence.

[0022] Furthermore, the load information includes the load value F of the heavy construction machinery, the load action length L F and load acting width B F .

[0023] Furthermore, the verification results of the beam-slab structure include:

[0024] If Mu,l <M l or V u,l <V l , the corresponding result is that the beam verification fails;

[0025] If M u,l ≥M l And V u,l ≥V l , the corresponding result is that the beam verification passed;

[0026] If M u,b <M b or V u,b <V b , the corresponding result is that the plate verification fails;

[0027] If M u,b ≥M b And V u,b ≥V b , the corresponding result is that the board verification passed;

[0028] Among them, M u,l 、V u,l are the bending and shear bearing capacities of the beam structure respectively; M l 、V l are the bending moment and shear force of the beam structure respectively; M u,b 、V u,b M are the bending bearing capacity and punching bearing capacity of the plate structure respectively; b 、V b are the bending moment and punching force of the plate structure respectively.

[0029] Accordingly, the present invention also provides a method for intelligent layout of heavy construction machinery based on machine learning, using the intelligent layout system for heavy construction machinery based on machine learning, the layout method comprises the following steps:

[0030] S1. Collect the plan layout and general structural design description of the basement roof where heavy construction machinery is to be deployed;

[0031] S2. Input the plan layout and the general structural design description of the basement top plate into the first machine learning model, and the first machine learning model outputs the plate structure information, the beam structure information, and the material property information;

[0032] S3. Collecting the load information of heavy construction machinery, the load location arrangement module arranges the load in different areas of the floor according to the load information in the heavy load information to form N groups of working conditions;

[0033] S4. Input the plate structure information, beam structure information, material property information, load information and N groups of working conditions into the second machine learning model, and the second machine learning model outputs the verification results of the beam-slab structure corresponding to the N groups of working conditions;

[0034] S5. The beam-slab force verification result marking module marks the area corresponding to the working condition where the verification result of the beam-slab structure is not passed with the first color, marks the area corresponding to the working condition where the verification result is passed with the second color, and outputs a heavy construction machinery layout drawing marked with the first color and / or the second color.

[0035] Due to the adoption of the above technical scheme, the present invention has the following advantages and positive effects compared with the prior art: the intelligent layout system and method of heavy construction machinery based on machine learning provided by the present invention can automatically obtain plate structure information, beam structure information, and material property information through the first machine learning model, and obtain N groups of working conditions based on load layout through the load position layout module, and then automatically verify the force verification results of the beam-slab structure corresponding to the N groups of working conditions through the second machine learning model, and use different colors to distinguish different verification results through the force verification result marking module to obtain the heavy construction machinery layout diagram. Therefore, in the application, only the plan layout diagram of the basement top plate, the general description of the structural design, and the load information of the heavy construction machinery need to be provided. The system can quickly output the heavy construction machinery layout diagram, and the speed is much higher than the traditional method of finite element modeling for calculation. Moreover, the method is based on machine deep learning technology, and has the advantages of simple operation and no reliance on human experience. Moreover, the system can also adapt to basement top plates with different plate spans under any load length and width, and is also suitable for application scenarios of heavy construction machinery such as car cranes, crawler cranes, and passenger and freight elevators, and has the advantage of strong adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a schematic diagram of a heavy construction machinery intelligent arrangement system based on machine learning in one embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of load arrangement in one embodiment of the present invention;

[0038] Figure 3 The present invention is a flowchart of a method for intelligent arrangement of heavy construction machinery based on machine learning in one embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following is a further detailed description of the heavy construction machinery intelligent arrangement system and method based on machine learning provided by the present invention in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer in conjunction with the following description. It should be noted that the accompanying drawings are all in a very simplified form and are not in precise proportions, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. Embodiment 1

[0040] Combination Figure 1 and Figure 2 As shown, this embodiment provides a heavy construction machinery intelligent layout system based on machine learning, including: a first machine learning model, a load position layout module, a second machine learning model and a force verification result marking module.

[0041] After inputting the floor plan and general structural design description of the basement top plate, the first machine learning model can output the plate structure information, beam structure information, and material property information. The floor plan and general structural design description of the basement top plate can be obtained in a specific project. The plate structure information, beam structure information, and material property information are included in the floor plan and general structural design description. The first machine learning model can automatically identify the plate structure information, beam structure information, and material property information through training. For example, the plate structure information includes the plate reinforcement information, plate span length L, plate span width B, and plate thickness h. b ; Beam structure information includes beam reinforcement information, beam centerline, beam length L l , beam section width B l , beam section height H l ; Material property information includes the elastic modulus E of the component material and the Poisson's ratio ν of the component material.

[0042] The load location arrangement module can automatically arrange the load in different areas of the plate according to the load information of the heavy construction machinery and the plate structure information. The number of areas is recorded as N, corresponding to N groups of working conditions. The load information of the heavy construction machinery can be obtained in the specific project. The load information includes the load value F of the heavy construction machinery, the load action length L F and load acting width B F .like Figure 2 The load positions corresponding to the N groups of working conditions and the coordinates of the load center relative to the plate center (x i ,y i ), i=1,2,…,N.

[0043] The second machine learning model can output the force verification results of the beam-slab structure corresponding to the N groups of working conditions after inputting the plate structure information, beam structure information, material property information, load information and N groups of working conditions. The force verification results include pass or fail, specifically: if M u,l <M lor V u,l <V l , the corresponding result is that the beam verification fails; if M u,l ≥M l And V u,l ≥V l , the corresponding result is that the beam verification passes; if M u,b <M b or V u,b <V b , the corresponding result is that the plate verification fails; if M u,b ≥M b And V u,b ≥V b , the corresponding result is that the plate verification passed; among them, M u,l 、V u,l are the bending and shear bearing capacities of the beam structure respectively; M l 、V l are the bending moment and shear force of the beam structure respectively; M u,b 、V u,b M are the bending bearing capacity and punching bearing capacity of the plate structure respectively; b 、V b are the bending moment and punching force of the plate structure respectively. The second machine learning model can derive M based on the plate structure information, beam structure information, and material property information. u,b 、V u,b 、M u,b 、V u,b Combined with the load information and N groups of working conditions, M is obtained. l 、V l 、M b 、V b , and then give the judgment result according to the above judgment method.

[0044] Both the first machine learning model and the second machine learning model are models built based on machine learning technology, embedded with convolutional neural network (CNN), recurrent neural network (RNN), transformer architecture or encoder-decoder architecture, and are trained with corresponding data sets according to the training purpose and have corresponding functions.

[0045] The force calculation result marking module can mark the failed areas in the force calculation results of the beam-slab structure with a first color, mark the passed areas with a second color, and output the marked heavy construction machinery layout diagram. The heavy construction machinery layout diagram can directly show which areas meet the conditions for deploying heavy construction machinery.

[0046] The heavy construction machinery intelligent layout system based on machine learning provided in this embodiment can automatically obtain plate structure information, beam structure information, and material property information through the first machine learning model, and obtain N groups of working conditions based on load layout through the load position layout module, and then automatically verify the force verification results of the beam-slab structure corresponding to the N groups of working conditions through the second machine learning model, and use different colors to distinguish different verification results through the force verification result marking module to obtain the heavy construction machinery layout diagram. Therefore, in the application, only the plan layout diagram of the basement roof, the general description of the structural design, and the load information of the heavy construction machinery need to be provided. The system can quickly output the heavy construction machinery layout diagram, and the speed is much higher than the traditional method of finite element modeling for calculation. Moreover, the system also has the advantage of strong adaptability, and can adapt to basement roofs with different plate spans under any load length and width, and is also suitable for application scenarios of heavy construction machinery such as truck cranes, crawler cranes, and passenger and freight elevators.

[0047] In a specific embodiment, the heavy construction machinery intelligent layout system based on machine learning also includes a first information acquisition and processing module and a first model training module. The first information acquisition and processing module is capable of acquiring a first training data set; the first training data set includes several groups of first training data; each group of first training data includes a basement roof plan layout and a general structural design description, as well as pre-processed plate structure information, beam structure information, and material property information. The first model training module trains the first machine learning model according to the training data set pre-processed by the first information acquisition and processing module. The specific contents of the plate structure information, beam structure information, and material property information can be found in the above description. The preprocessing of information includes labeling the plate structure information, beam structure information, and material property information in the basement roof plan layout and the general structural design description; normalizing the units of the labeled data, such as the plate span length L, plate span width B, and plate thickness h. b , beam length L l , beam section width B l and the beam section height H l The unit is mm, the unit of elastic modulus E of component material is MPa, and the unit of Poisson's ratio ν of component material is dimensionless; beams and plate components are numbered, and the numbering order is from left to right and from top to bottom from the upper left corner of the drawing.

[0048] In a specific embodiment, the heavy construction machinery intelligent layout system based on machine learning also includes a second information collection and processing module, which can collect a second training data set. The second training data set includes several groups of second training data; each group of second training data includes various heavy machinery instruction manuals, relevant construction plans, relevant calculation books, and pre-processed plate structure information, beam structure information, material property information, load information, and beam-slab structure verification results. The second model training module trains the second machine learning model according to the second training data set pre-processed by the second information collection and processing module. As an example, the load information includes the load value F of the heavy construction machinery, the load action length L F and load acting width B F , and also includes the coordinates (x, y) of the load center relative to the center of the plate. The information preprocessing of the second information acquisition and processing module includes: converting the load value F of the heavy construction machinery into kN, the load action length L F , Load width B F The units of x and y of the relative distances between the load center and the slab center are converted into mm. The verification results of the beam-slab structure in the second training data set can be directly obtained from the corresponding finite element calculation results or obtained through other means. Embodiment 2

[0049] This embodiment provides a method for intelligent layout of heavy construction machinery based on machine learning, using the intelligent layout system for heavy construction machinery based on machine learning described in Embodiment 1. The layout method includes the following steps:

[0050] S1. Collect the plan layout and general structural design description of the basement roof where heavy construction machinery is to be deployed;

[0051] S2. Input the plan layout and the general structural design description of the basement top plate into the first machine learning model, and the first machine learning model outputs the plate structure information, the beam structure information, and the material property information;

[0052] S3. Collecting the load information of heavy construction machinery, the load location arrangement module arranges the load in different areas of the floor according to the load information in the heavy load information to form N groups of working conditions;

[0053] S4. Input the plate structure information, beam structure information, material property information, load information and N groups of working conditions into the second machine learning model, and the second machine learning model outputs the verification results of the beam-slab structure corresponding to the N groups of working conditions;

[0054] S5. The beam-slab force verification result marking module marks the area corresponding to the working condition where the verification result of the beam-slab structure is not passed with a first color, and marks the area corresponding to the working condition where the verification result is passed with a second color, and outputs a heavy construction machinery layout diagram marked with the first color and / or the second color. On-site personnel can arrange heavy construction machinery according to the fast heavy construction machinery layout diagram.

[0055] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0056] The above-mentioned embodiments only express several implementation modes of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A heavy construction machinery intelligent layout system based on machine learning, characterized in that: include: The first machine learning model can output the slab structure information, beam structure information, and material property information after inputting the basement top slab floor plan and the general structural design description; The load location arrangement module can automatically arrange the load in different areas of the plate according to the load information of heavy construction machinery and the plate structure information. The number of areas is recorded as N, corresponding to N groups of working conditions; The second machine learning model can input plate structure information, beam structure information, material property information, load information and N groups of working conditions, and then output the force verification results of the beam-slab structure corresponding to the N groups of working conditions; A stress verification result marking module is capable of marking the failed areas in the stress verification result of the beam-slab structure with a first color, marking the passed areas with a second color, and outputting the marked heavy construction machinery layout diagram; The first information collection and processing module is capable of collecting a first training data set; the first training data set includes a plurality of groups of first training data; each group of first training data includes a basement top plate plan layout and a general description of the structural design, as well as pre-processed plate structure information, beam structure information, and material property information; A first model training module, training a first machine learning model according to the training data set preprocessed by the first information acquisition and processing module; The second information collection and processing module is capable of collecting a second training data set; the second training data set includes a plurality of groups of second training data; each group of second training data includes various heavy machinery instruction manuals, relevant construction plans, relevant calculation books, and pre-processed plate structure information, beam structure information, material property information, load information, and beam-slab structure verification results; The second model training module trains the second machine learning model according to the training data set preprocessed by the second information acquisition and processing module.

2. The heavy construction machinery intelligent layout system based on machine learning as claimed in claim 1, characterized in that: The plate structure information includes the plate reinforcement information, plate span length L, plate span width B, plate thickness h b ; Beam structure information includes beam reinforcement information, beam centerline, beam length L l , beam section width B l , beam section height H l ; The material property information includes the elastic modulus E of the component material and the Poisson's ratio ν of the component material.

3. The heavy construction machinery intelligent arrangement system based on machine learning as claimed in claim 2, characterized in that: The preprocessing process is: labeling the plate structure information, beam structure information, and material property information; extracting the parameter values ​​and units in the labels, and converting the parameter units into preset units; and numbering the beam and plate components in sequence.

4. The heavy construction machinery intelligent arrangement system based on machine learning as claimed in claim 1, characterized in that: Load information includes the load value F of heavy construction machinery and the load action length L F and load acting width B F .

5. The heavy construction machinery intelligent arrangement system based on machine learning as claimed in claim 1, characterized in that: The verification results of beam-slab structure include: If M u,l <M l or V u,l <V l , the corresponding result is that the beam verification fails; If M u,l ≥M l And V u,l ≥V l , the corresponding result is that the beam verification passed; If M u,b <M b or V u,b <V b , the corresponding result is that the plate verification fails; If M u,b ≥M b And V u,b ≥V b , the corresponding result is that the board verification passed; Among them, M u,l 、V u,l are the bending and shear bearing capacities of the beam structure respectively; M l 、V l are the bending moment and shear force of the beam structure respectively; M u,b 、V u,b M are the bending bearing capacity and punching bearing capacity of the plate structure respectively; b 、V b are the bending moment and punching force of the plate structure respectively.

6. A method for intelligent arrangement of heavy construction machinery based on machine learning, characterized in that: Using the heavy construction machinery intelligent layout system based on machine learning as claimed in claim 1, the layout method comprises the following steps: S1. Collect the plan layout and general structural design description of the basement roof where heavy construction machinery is to be deployed; S2. Input the plan layout and the general structural design description of the basement top plate into the first machine learning model, and the first machine learning model outputs the plate structure information, the beam structure information, and the material property information; S3. Collecting the load information of heavy construction machinery, the load location arrangement module arranges the load in different areas of the floor according to the load information in the heavy load information to form N groups of working conditions; S4. Input the plate structure information, beam structure information, material property information, load information and N groups of working conditions into the second machine learning model, and the second machine learning model outputs the verification results of the beam-slab structure corresponding to the N groups of working conditions; S5. The beam-slab force verification result marking module marks the area corresponding to the working condition where the verification result of the beam-slab structure is not passed with the first color, marks the area corresponding to the working condition where the verification result is passed with the second color, and outputs a heavy construction machinery layout drawing marked with the first color and / or the second color.

Images