CN109299585A - A kind of architectural design method based on artificial intelligence - Google Patents

A kind of architectural design method based on artificial intelligence Download PDF

Info

Publication number
CN109299585A
CN109299585A CN201811554242.XA CN201811554242A CN109299585A CN 109299585 A CN109299585 A CN 109299585A CN 201811554242 A CN201811554242 A CN 201811554242A CN 109299585 A CN109299585 A CN 109299585A
Authority
CN
China
Prior art keywords
building
deep learning
learning model
design method
sample
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201811554242.XA
Other languages
Chinese (zh)
Inventor
刘威鑫
车宁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Hefu Artificial Intelligence Technology Group Co ltd
Original Assignee
Shanghai Hefu Artificial Intelligence Technology Group Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Hefu Artificial Intelligence Technology Group Co ltd filed Critical Shanghai Hefu Artificial Intelligence Technology Group Co ltd
Priority to CN201811554242.XA priority Critical patent/CN109299585A/en
Publication of CN109299585A publication Critical patent/CN109299585A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/10Geometric CAD
    • G06F30/13Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/20Design optimisation, verification or simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Structural Engineering (AREA)
  • Computational Mathematics (AREA)
  • Civil Engineering (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Architecture (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present invention provides a kind of architectural design method based on artificial intelligence, by establishing database purchase building shape product data, carries out ground structure analysis using building of the peripheral hardware to required design, and compare qualifications built in correlation, completes architectural design method.

Description

A kind of architectural design method based on artificial intelligence
Technical field
The present invention relates to field of artificial intelligence, in particular to a kind of architectural design method based on artificial intelligence.
Background technique
With the continuous propulsion of digitlization and IT application process, in existing detail design, designer usually can be applied The design tool software of existing building decoration area is designed, and is accordingly generated and finally presented in the form of electronic data Design result.
Building trade, which works, be unable to do without data information and Heuristics, including drawing, technical documentation, operating instruction and industry Information etc..In current design workflow, architect generally uses traditionally on paper and electronic information document, and partial information is borrowed Internet is helped, but lies substantially in the inefficient state of windows folder management, shows themselves in that data deficiencies genealogical classification, data Amount is few, storage position dispersion, and information island, shortage system integration management join with outside network resource one by one for artificial formation System is weak;Data check mode is single, and search efficiency is low, and data content shows not intuitive.In the information age, architectural design It is the course of work that bulk items information was integrated and created processing by architect, electronic degree is high, and information exchange is frequent, low The design data management mode of effect necessarily affects the promotion of design work efficiency.
Summary of the invention
To realize above-mentioned technical purpose and the technique effect, proposing a kind of architectural design side based on artificial intelligence Method carries out ground structure analysis using building of the peripheral hardware to required design by establishing database purchase building shape product data, and Qualifications built in comparison correlation, complete architectural design method, which is characterized in that specifically comprise the following steps:
Step 1: establishing server, the model data of building is collected by server, and be stored in advance alternative Architectural modulus;
Step 2: acquisition map datum treats construction area and carries out terrain data typing, and carries out the circle on building ground online It is fixed, upload the base CAD file;
Step 3: to the underlying parameter of server typing building, limiting building items basic data;
Step 4: carrying out data comparison in server, pass through the building of internal model data combination step 3 typing Underlying parameter automatically generates building type product model, and generates electronic document output.
Specifically, the architectural modulus includes level-one architectural modulus, second level architectural modulus and three-level architectural modulus.
Specifically, the level-one architectural modulus includes axis net and buildings model.
Specifically, the earphone architectural modulus includes building series, D reconstruction and door and window tool.
Specifically, the three-level architectural modulus includes pillar, beam, floor, wall and stair, project name, door number, class Type and fire-protection rating and lintel, terrace absolute altitude, door pocket pattern, door leaf thickness, door leaf height, door leaf width and remarks.
Specifically, the base CAD file includes the topography and geomorphology in area yet to be built, circle ground range and is building ground scheme again Report.
Specifically, the peripheral hardware includes equipped with remote control equipment and with nobody of infrared sensing and camera shooting photographing device Machine, field surveys equipment and telecommunication equipment.
Specifically, electronic document generated includes design general layout, each layer plane figure of building type, elevation, explanation Book and bill of approximate estimate.
Specifically, the process for automatically generating building pattern type product model is carried out using deep learning model, utilize Each sample that training sample is concentrated, is trained the initial deep learning model constructed in advance, obtains intermediate depth study Model;Wherein, the initial deep learning model are as follows: be loaded with the deep learning model of class object function;
Using the intermediate deep learning model, the feature vector that the training sample concentrates each sample, and root are calculated The feature vector of each sample is concentrated according to the training sample, calculates the initial of the intermediate parameters of central point distance objective function Value;
Central point distance objective function is added in the intermediate deep learning model and loads the intermediate parameters Initial value obtains target deep learning model;
The preset quantity sample of training sample concentration is imported as lot data;
Using current goal deep learning model, the feature vector of each sample in present lot data is calculated, and according to The feature vector of each sample in present lot data, updates the parameter value of the intermediate parameters;
Based on the feature vector of each sample in present lot data, the function of the central point distance objective function is calculated The functional value of value and the class object function, and judge whether the functional value for the central point distance objective function being calculated is received Whether the functional value for the class object function held back to the first predetermined interval, and be calculated converges to the second predetermined interval;
If not, utilizing the back-propagation gradient of central point distance objective function and the reversed biography of the class object function Gradient is broadcast, the parameter of current goal deep learning model is adjusted, and returns and executes the pre- of the importing training sample concentration If the step of quantity sample is as lot data;
If so, terminating the training to current goal deep learning model.
It is different from the prior art, on the basis of existing design method, the artificial intelligence design based on deep learning is added Mode, it is possible to reduce the manpower consumption in design process, designer only need to be audited and be modified on the basis of the drawing produced, Intuitively, the carry out architectural design refined, provides more convenience.
Specific embodiment
Below in conjunction with the embodiment of the present invention, technical scheme in the embodiment of the invention is clearly and completely described, Obviously, the described embodiments are merely a part of the embodiments of the present invention, instead of all the embodiments.Based in the present invention Embodiment, every other embodiment obtained by those of ordinary skill in the art without making creative efforts, all Belong to the scope of protection of the invention.
A kind of architectural design method based on artificial intelligence, by establishing database purchase building shape product data, using outer If the building to required design carries out ground structure analysis, and compares qualifications built in correlation, architectural design method, tool are completed Body includes the following steps:
Step 1: establishing server, the model data of building is collected by server, and be stored in advance alternative Architectural modulus;
Step 2: acquisition map datum treats construction area and carries out terrain data typing, and carries out the circle on building ground online It is fixed, upload the base CAD file;
Step 3: to the underlying parameter of server typing building, limiting building items basic data;
Step 4: carrying out data comparison in server, pass through the building of internal model data combination step 3 typing Underlying parameter automatically generates building type product model, and generates electronic document output.
Wherein, the architectural modulus includes level-one architectural modulus, second level architectural modulus and three-level architectural modulus.
The level-one architectural modulus includes axis net and buildings model.
The earphone architectural modulus includes building series, D reconstruction and door and window tool.
The three-level architectural modulus includes pillar, beam, floor, wall and stair, project name, door number, type and prevents Fiery grade and lintel, terrace absolute altitude, door pocket pattern, door leaf thickness, door leaf height, door leaf width and remarks.
The base CAD file includes the topography and geomorphology in area yet to be built, circle ground range and is building ground scheme report again.
The peripheral hardware includes equipped with remote control equipment and with infrared sensing and the unmanned plane, the scene survey that image photographing device Equipment and telecommunication equipment are drawn, unmanned plane can set cruise route, and remote controlled drone is cruised, while pass through carrying Remote sensing equipment on unmanned plane shoots regional neighboring area yet to be built in each monitoring position.
Electronic document generated includes design general layout, each layer plane figure of building type, elevation, specification and general Calculate book.
Further, the process for automatically generating building pattern type product model is carried out using deep learning model, benefit The each sample concentrated with training sample, is trained the initial deep learning model constructed in advance, obtains intermediate depth Practise model;Wherein, the initial deep learning model are as follows: be loaded with the deep learning model of class object function;
Using the intermediate deep learning model, the feature vector that the training sample concentrates each sample, and root are calculated The feature vector of each sample is concentrated according to the training sample, calculates the initial of the intermediate parameters of central point distance objective function Value;
Central point distance objective function is added in the intermediate deep learning model and loads the intermediate parameters Initial value obtains target deep learning model;
The preset quantity sample of training sample concentration is imported as lot data;
Using current goal deep learning model, the feature vector of each sample in present lot data is calculated, and according to The feature vector of each sample in present lot data, updates the parameter value of the intermediate parameters;
Based on the feature vector of each sample in present lot data, the function of the central point distance objective function is calculated The functional value of value and the class object function, and judge whether the functional value for the central point distance objective function being calculated is received Whether the functional value for the class object function held back to the first predetermined interval, and be calculated converges to the second predetermined interval;
If not, utilizing the back-propagation gradient of central point distance objective function and the reversed biography of the class object function Gradient is broadcast, the parameter of current goal deep learning model is adjusted, and returns and executes the pre- of the importing training sample concentration If the step of quantity sample is as lot data;
If so, terminating the training to current goal deep learning model.
Server is used to store the architectural modulus for user setting;Wherein server is stored with building for user setting The parameters such as frame, architectural design details are built, service implement body can be using any proper data storage type or storage rack The database of structure.
Preferably, can scheme constructs hardware system according to the present invention, including one or more processors and deposit Reservoir.Wherein, processor and memory can be connected by bus or other modes,
Memory as a kind of non-volatile computer readable storage medium storing program for executing, can be used for storing non-volatile software program, Non-volatile computer executable program and module, such as the architectural design based on Building Information Model in the embodiment of the present invention Corresponding program instruction/the module of method.Processor by run non-volatile software program stored in memory, instruction with And module, thereby executing the various function application and data processing of server, i.e., realization above method embodiment based on building Build the architectural design method of information model.
Memory may include storing program area and storage data area, wherein storing program area can storage program area, extremely Application program required for a few function;Storage data area, which can be stored, uses created number according to architectural design system According to etc..In addition, memory may include high-speed random access memory, it can also include nonvolatile memory, for example, at least One disk memory, flush memory device or other non-volatile solid state memory parts.In some embodiments, memory can Choosing includes the memory remotely located relative to processor, these remote memories can be by network connection to based on building letter Cease the architectural design system of model.The example of above-mentioned network includes but is not limited to internet, intranet, local area network, movement Communication network and combinations thereof.
One or more of module storages in the memory, are executed when by one or more of processors When, execute above-mentioned architectural design method.
By adopting the above technical scheme, by unmanned plane place surrounding enviroment can be carried out with the acquisition of image, and be based on The analysis that image includes color is handled, with the dominant hue of environment and each position institute of building where determination area yet to be built Corresponding details tone (i.e. selection tone);Based on the analysis of above-mentioned tone data, so that the tone of building can be with periphery Primary climate be consistent, and may make place relative to component environment details carry out tone fitting so that place and peripheral ring The total tune degree in border is significantly improved.At the same time, the above-mentioned large size regional architecture yet to be built based on mobile platform analysis Design method, also can personnel for surrounding enviroment and vehicle flow while the tone to place surrounding enviroment is monitored It is monitored, so that designer would know that the magnitude of traffic flow numerical value of regional periphery each region yet to be built, and is based on above-mentioned number Value carries out the roading of public place, so that the quick circulation of the achievable personnel in area yet to be built planned, so that its Service efficiency is obviously improved
The above description is only an embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair Equivalent structure or equivalent flow shift made by bright description, or direct or brief introduction are used in other correlative technology fields, It is included within the scope of the present invention.

Claims (9)

1. a kind of architectural design method based on artificial intelligence utilizes peripheral hardware by establishing database purchase building shape product data Ground structure analysis is carried out to the building of required design, and compares qualifications built in correlation, completes architectural design method, it is special Sign is, specifically comprises the following steps:
Step 1: establishing server, the model data of building is collected by server, and alternative building is stored in advance Parameter;
Step 2: acquisition map datum treats construction area and carries out terrain data typing, and carries out the delineation on building ground online, on Pass the base CAD file;
Step 3: to the underlying parameter of server typing building, limiting building items basic data;
Step 4: carrying out data comparison in server, pass through the basis of the building of internal model data combination step 3 typing Parameter automatically generates building type product model, and generates electronic document output.
2. a kind of architectural design method based on artificial intelligence according to claim 1, it is characterised in that: the building ginseng Number includes level-one architectural modulus, second level architectural modulus and three-level architectural modulus.
3. a kind of architectural design method based on artificial intelligence according to claim 2, it is characterised in that: the level-one is built Building parameter includes axis net and buildings model.
4. a kind of architectural design method based on artificial intelligence according to claim 2, it is characterised in that: the earphone is built Building parameter includes building series, D reconstruction and door and window tool.
5. a kind of architectural design method based on artificial intelligence according to claim 2, it is characterised in that: the three-level is built Build parameter include pillar, beam, floor, wall and stair, project name, door number, type and fire-protection rating and lintel, Level ground absolute altitude, door pocket pattern, door leaf thickness, door leaf height, door leaf width and remarks.
6. a kind of architectural design method based on artificial intelligence according to claim 1, it is characterised in that: the CAD base Ground file includes the topography and geomorphology in area yet to be built, circle ground range and is building ground scheme report again.
7. a kind of architectural design method based on artificial intelligence according to claim 1, it is characterised in that: the peripheral hardware packet Include unmanned plane, field surveys equipment and the telecommunication equipped with remote control equipment and with infrared sensing and camera shooting photographing device Equipment.
8. a kind of architectural design method based on artificial intelligence according to claim 1, it is characterised in that: electricity generated Subfile includes design general layout, each layer plane figure of building type, elevation, specification and bill of approximate estimate.
9. it is according to claim 1 it is a kind of based on manually can only architectural design method, it is characterised in that: described is automatic The process for generating building pattern type product model is carried out using deep learning model, and each sample concentrated using training sample is right The initial deep learning model constructed in advance is trained, and obtains intermediate deep learning model;Wherein, initial deep learning model Are as follows: it is loaded with the deep learning model of class object function;
Using the intermediate deep learning model, the feature vector that the training sample concentrates each sample is calculated, and according to institute The feature vector that training sample concentrates each sample is stated, the initial value of the intermediate parameters of central point distance objective function is calculated;
Central point distance objective function is added in intermediate deep learning model and is loaded the initial value of the intermediate parameters, is obtained To target deep learning model;
The preset quantity sample of training sample concentration is imported as lot data;
Using current goal deep learning model, the feature vector of each sample in present lot data is calculated, and according to current The feature vector of each sample in lot data, updates the parameter value of intermediate parameters;
Based on the feature vector of each sample in present lot data, calculate the central point distance objective function functional value and The functional value of class object function, and judge whether the functional value for the central point distance objective function being calculated converges to first Predetermined interval, and whether the functional value for the class object function being calculated converges to the second predetermined interval;
If not, utilizing the back-propagation gradient of central point distance objective function and the backpropagation ladder of the class object function Degree, adjusts the parameter of current goal deep learning model, and returns and execute the preset quantity for importing training sample concentration The step of sample is as lot data;
If so, terminating the training to current goal deep learning model.
CN201811554242.XA 2018-12-18 2018-12-18 A kind of architectural design method based on artificial intelligence Pending CN109299585A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811554242.XA CN109299585A (en) 2018-12-18 2018-12-18 A kind of architectural design method based on artificial intelligence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811554242.XA CN109299585A (en) 2018-12-18 2018-12-18 A kind of architectural design method based on artificial intelligence

Publications (1)

Publication Number Publication Date
CN109299585A true CN109299585A (en) 2019-02-01

Family

ID=65142979

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811554242.XA Pending CN109299585A (en) 2018-12-18 2018-12-18 A kind of architectural design method based on artificial intelligence

Country Status (1)

Country Link
CN (1) CN109299585A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110020798A (en) * 2019-03-28 2019-07-16 上海荷福人工智能科技(集团)有限公司 A kind of approaches to IM of tracing to the source of the project progress based on artificial intelligence
CN110096835A (en) * 2019-05-14 2019-08-06 上海荷福人工智能科技(集团)有限公司 A kind of arrangement and method for construction optimum management method based on artificial intelligence engineering design
CN114641753A (en) * 2019-11-18 2022-06-17 欧特克公司 Composite data generation and Building Information Model (BIM) element extraction from floor plan drawings using machine learning
WO2024079761A1 (en) * 2022-10-13 2024-04-18 Visava Labs Private Limited A system and method for generating architectural designs and specifications for buildings

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103150751A (en) * 2013-01-10 2013-06-12 江苏易图地理信息工程有限公司 Three-dimensional modeling method for achieving building inside and outside integration in digital map
WO2014079477A1 (en) * 2012-11-20 2014-05-30 Siemens Aktiengesellschaft Method for the automatic creation of two- or three-dimensional building models
CN105574302A (en) * 2016-02-29 2016-05-11 江阴市建筑设计研究院有限公司 Large-scale public place building design method based on mobile platform analysis
CN107590355A (en) * 2017-10-27 2018-01-16 巧夺天宫(深圳)科技有限公司 Architectural design method, design system and electronic equipment based on BIM
CN108734193A (en) * 2018-03-27 2018-11-02 合肥麟图信息科技有限公司 A kind of training method and device of deep learning model

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014079477A1 (en) * 2012-11-20 2014-05-30 Siemens Aktiengesellschaft Method for the automatic creation of two- or three-dimensional building models
CN103150751A (en) * 2013-01-10 2013-06-12 江苏易图地理信息工程有限公司 Three-dimensional modeling method for achieving building inside and outside integration in digital map
CN105574302A (en) * 2016-02-29 2016-05-11 江阴市建筑设计研究院有限公司 Large-scale public place building design method based on mobile platform analysis
CN107590355A (en) * 2017-10-27 2018-01-16 巧夺天宫(深圳)科技有限公司 Architectural design method, design system and electronic equipment based on BIM
CN108734193A (en) * 2018-03-27 2018-11-02 合肥麟图信息科技有限公司 A kind of training method and device of deep learning model

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
孙澄等: "面向人工智能的建筑计算性设计研究", 《建筑学报》 *
孙澄等: "面向人工智能的建筑计算性设计研究", 《建筑学报》, no. 09, 20 September 2018 (2018-09-20) *
石春民: "建筑设计的计算机及信息化应用趋势与思考", 《江苏建筑》 *
石春民: "建筑设计的计算机及信息化应用趋势与思考", 《江苏建筑》, 23 November 2018 (2018-11-23), pages 21 - 22 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110020798A (en) * 2019-03-28 2019-07-16 上海荷福人工智能科技(集团)有限公司 A kind of approaches to IM of tracing to the source of the project progress based on artificial intelligence
CN110096835A (en) * 2019-05-14 2019-08-06 上海荷福人工智能科技(集团)有限公司 A kind of arrangement and method for construction optimum management method based on artificial intelligence engineering design
CN114641753A (en) * 2019-11-18 2022-06-17 欧特克公司 Composite data generation and Building Information Model (BIM) element extraction from floor plan drawings using machine learning
WO2024079761A1 (en) * 2022-10-13 2024-04-18 Visava Labs Private Limited A system and method for generating architectural designs and specifications for buildings

Similar Documents

Publication Publication Date Title
CN109299585A (en) A kind of architectural design method based on artificial intelligence
CN109657347A (en) A kind of architectural working drawing design method based on artificial intelligence
Zhang et al. An improved ant colony algorithm for path planning in one scenic area with many spots
Zang et al. A review of nature-inspired algorithms
Varshney et al. Targeting villages for rural development using satellite image analysis
CN109478045A (en) Goal systems is controlled using prediction
CN106648852A (en) Multi-satellite mission scheduling method and device based on double ant colonies
CN110428111A (en) Multi-Tasking method for planning track when UAV/UGV collaboration is long
CN110096835A (en) A kind of arrangement and method for construction optimum management method based on artificial intelligence engineering design
CN112883476B (en) Layout method and device of building space and electronic equipment
CN109766590A (en) One kind method for reporting based on artificial intelligence check of drawings and out
KR102204337B1 (en) Method for providing real-time construction estimator service using intuitive step-by-step choice
CN103699106A (en) Multi-unmanned aerial vehicle cooperative task planning simulation system based on VR-Forces simulation platform
Dyer et al. The biosphere-reserve concept: needs for a network design
Porst et al. How scale matters in translocality: uses and potentials of scale in translocal research
CN106156934A (en) Distributed well engineering and planning
CN106656308A (en) Method and device for planning tasks in space information network
CN109885082A (en) The method that a kind of lower unmanned aerial vehicle flight path of task based access control driving is planned
CA2942351C (en) Planning drilling operations using models and rig market databases
CN109543634A (en) Data processing method, device, electronic equipment and storage medium in position fixing process
Bournaris et al. A DSS for planning the agricultural production
Wildenberg et al. Use and evaluation of FCM as a tool for long term socio ecological research
Li et al. Research on task allocation in multiple logistics robots based on an improved ant colony algorithm
CA3150928A1 (en) Offset well analysis using well trajectory similarity
Jianfeng et al. Construction of intelligent building design system based on BIM and AI

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20190201