Multi-agent collaborative design and decision optimization method for building full life cycle

Abstract

The invention relates to the technical field of building engineering, and discloses a building full life cycle-oriented multi-agent collaborative design and decision optimization method, which comprises the following steps of S1, constructing a multi-agent system which comprises a design agent, a construction agent, an operation agent and an owner agent; s2, information integration based on the building information model BIM, wherein the information integration based on the building information model BIM comprises geometric information, physical information, construction information and operation information; and S3, a collaborative decision-making mechanism, wherein the collaborative decision-making mechanism comprises collaborative design flow, decision-making optimization, collaboration and optimization of a construction stage and management and optimization of an operation stage. According to the method, the problems of unsmooth information communication, disjunction of all stages and lack of full-life-cycle optimization in an existing building design and decision-making method are solved, efficient collaboration of all participants in the full life cycle of the building is achieved, the scientificity and optimization degree of decision making are improved, and the building quality and operation benefits are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering, and particularly relates to a multi-agent collaborative design and decision-making optimization method for the entire life cycle of a building. Background Art

[0002] With the development of the construction industry, the scale of construction projects is getting larger and the complexity is constantly increasing. Traditional building design and decision-making methods often have problems such as poor information communication, disconnection between each stage, and inability to fully consider the costs and benefits of the entire life cycle. In the entire life cycle of a building, from planning and design, construction to operation and maintenance, there are many involved parties, such as owners, designers, construction units, operation and management teams, etc. The information exchange and collaborative work efficiency among all parties are crucial for the success of the project.

[0003] However, existing methods are difficult to achieve efficient collaboration, resulting in an extended project cycle, increased costs, and difficult-to-guarantee quality. In addition, in the decision-making process, there is a lack of optimization methods that comprehensively consider various factors within the entire life cycle of a building, and it is impossible to provide optimal decision support for the project. For example, in the design stage, only the appearance and function of the building may be emphasized, while the feasibility of construction and the costs of operation and maintenance are ignored. Therefore, we propose a multi-agent collaborative design and decision-making optimization method for the entire life cycle of a building. Summary of the Invention

[0004] The purpose of the present invention is to provide a multi-agent collaborative design and decision-making optimization method for the entire life cycle of a building, so as to solve the problems of poor information communication, disconnection between each stage, and lack of optimization of the entire life cycle in existing building design and decision-making methods, realize the efficient collaboration of all involved parties within the entire life cycle of a building, improve the scientific nature and optimization degree of decision-making, reduce project costs, and improve building quality and operation benefits.

[0005] The above technical purpose of the present invention is achieved through the following technical solutions: A multi-agent collaborative design and decision-making optimization method for the entire life cycle of a building, including the following steps: S1. Construct a multi-agent system, and the multi-agent system includes a design agent, a construction agent, an operation agent, and an owner agent; S2. Information integration based on the Building Information Model (BIM), and the information integration based on the Building Information Model (BIM) includes geometric information, physical information, construction information, and operation information; S3. A collaborative decision-making mechanism, and the collaborative decision-making mechanism includes a collaborative design process, decision-making optimization, collaboration and optimization in the construction stage, and management and optimization in the operation stage.

[0006] Furthermore, in step S1, the designed intelligent agents are responsible for the design work of the building, including schematic design, construction drawing design, etc. They can generate multiple design schemes according to project requirements and relevant specifications, evaluate and optimize the design schemes. The construction intelligent agents formulate construction plans based on the design schemes, simulate the construction process, analyze the risks and problems during construction, and provide corresponding solutions. The operation intelligent agents are responsible for the management during the building operation stage, predict the maintenance requirements of building equipment, optimize energy consumption, and improve operation efficiency. The owner intelligent agents represent the needs and interests of the owners, evaluate and make decisions on the design schemes, construction plans, and operation strategies. Through standardized interfaces and protocols, information interaction is carried out among the intelligent agents to achieve data sharing and collaborative work. For example, the design intelligent agents transfer the design schemes to the construction intelligent agents and the owner intelligent agents. The construction intelligent agents transfer the feedback information during the construction process to the design intelligent agents and the owner intelligent agents. The operation intelligent agents feedback the data and requirements during the operation stage to the design intelligent agents and the construction intelligent agents for optimization in subsequent design and construction.

[0007] Furthermore, in step S2, the Building Information Model (BIM) serves as the basis for information interaction among the intelligent agents to ensure the accuracy and consistency of information. The design intelligent agents carry out design work based on the BIM model. The construction intelligent agents formulate construction plans and conduct construction simulations according to the BIM model. The operation intelligent agents use the BIM model for equipment management and energy optimization. Through the BIM model, each intelligent agent can obtain the required information in real time, avoiding the emergence of information silos. For example, during the construction process, construction workers can view the construction drawings and relevant information in the BIM model in real time through mobile devices to understand the construction progress and quality requirements. During the operation stage, maintenance personnel can quickly locate the equipment position through the BIM model and view the maintenance records and technical parameters of the equipment.

[0008] Furthermore, in step S3, the collaborative design process first depends on the requirements proposed by the owner intelligent agents and the project constraints. Then the design intelligent agents send the preliminary design schemes to the construction intelligent agents and the operation intelligent agents for evaluation. The construction intelligent agents evaluate the design schemes from aspects such as construction feasibility, construction cost, and construction progress, and put forward improvement suggestions. The operation intelligent agents, from the perspective of building operation and maintenance, make decisions and optimize using multi-objective optimization algorithms to optimize multiple objectives throughout the building life cycle. The formula for the multi-objective optimization algorithm is:

[0009] Among them, C_total is the total life-cycle cost, E_impact is the carbon emission impact factor, and S_satisfaction is the user satisfaction. Finally, for the reinforcement learning framework: the Deep Deterministic Policy Gradient (TD3) algorithm is used for policy optimization. The collaboration and optimization in the construction stage include digital twin construction, dynamic adjustment of schedule resources, and safety risk early warning; The digital twin construction: BIM+IoT real-time data mapping; The dynamic adjustment of schedule resources: a resource scheduling algorithm based on reinforcement learning; The safety risk early warning: an LSTM network for predicting the accident probability (accuracy rate ≥ 85%); The management and optimization in the operation stage include equipment failure prediction, maintenance strategy optimization, and dynamic energy consumption management. The equipment failure prediction is based on the time-series data analysis of Transformer. The maintenance strategy optimization considers the preventive maintenance plan of the total life-cycle cost. The dynamic energy consumption management is the closed-loop control of the digital twin and the physical system.

[0010] The beneficial effects of the present invention are as follows: By constructing a multi-agent system, the present invention realizes the efficient collaboration of all parties involved in the whole life cycle of the building. Information can be exchanged between agents in a timely and accurate manner, avoiding the problems of poor information communication and disconnection in each stage, and greatly improving the project promotion efficiency; By adopting a multi-objective optimization algorithm, multiple objectives in the whole life cycle of the building are fully considered, providing a series of optimal decision-making schemes for the owner. The owner can make a choice according to his own needs and preferences, thereby improving the scientificity and rationality of decision-making and realizing the maximization of the overall project benefits; In the design stage, through the participation of construction agents and operation agents, unreasonable points in the design scheme can be discovered in advance, avoiding large-scale modifications and adjustments in the construction and operation stages, thereby reducing the project cost. In the construction stage, through the construction progress simulation and resource allocation optimization, the construction efficiency is improved, resource waste is reduced, and the cost is further reduced. In the operation stage, through equipment maintenance prediction and energy consumption optimization, the operation cost is reduced; Throughout the whole life cycle of the building, each agent evaluates and optimizes the building from different perspectives, ensuring the quality of the building. In the operation stage, through intelligent management and optimization, the operation efficiency of the building is improved, providing a more comfortable and convenient use environment for users. Description of the Drawings

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0012] Figure 1 It is a flowchart of a multi-agent collaborative design and decision optimization method for the whole life cycle of a building according to the present invention.

[0013] Figure 2 It is an architecture diagram of a multi-agent system in a multi-agent collaborative design and decision optimization method for the whole life cycle of a building according to the present invention.

[0014] Figure 3 It is an architecture diagram of a collaborative decision-making mechanism in a multi-agent collaborative design and decision optimization method for the whole life cycle of a building according to the present invention. Specific embodiments

[0015] The following will clearly and completely describe the technical solutions of the present invention in combination with specific embodiments. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0016] Please refer to Figures 1-3 , the present invention provides a technical solution: a technical solution of a multi-agent collaborative design and decision optimization method for the whole life cycle of a building, including the following steps: S1. Construct a multi-agent system, and the multi-agent system includes a design agent, a construction agent, an operation agent, and an owner agent; S2. Information integration based on the building information model BIM, and the information integration based on the building information model BIM includes geometric information, physical information, construction information, and operation information; S3. Collaborative decision-making mechanism, and the collaborative decision-making mechanism includes a collaborative design process, decision optimization, collaboration and optimization in the construction stage, and management and optimization in the operation stage.

[0017] Furthermore, in step S1, the designed intelligent agents are responsible for the design work of the building, including schematic design, construction drawing design, etc. They can generate multiple design schemes according to project requirements and relevant specifications, evaluate and optimize the design schemes. The construction intelligent agents formulate construction plans based on the design schemes, simulate the construction process, analyze risks and problems during construction, and provide corresponding solutions. The operation intelligent agents are responsible for the management during the building operation stage, predict the maintenance requirements of building equipment, optimize energy consumption, and improve operation efficiency. The owner intelligent agents represent the needs and interests of the owners, evaluate and make decisions on the design schemes, construction plans, and operation strategies. Through information interaction among the intelligent agents via standardized interfaces and protocols, data sharing and collaborative work are achieved. For example, the design intelligent agent transmits the design scheme to the construction intelligent agent and the owner intelligent agent. The construction intelligent agent transmits the feedback information during the construction process to the design intelligent agent and the owner intelligent agent. The operation intelligent agent feeds back the data and requirements during the operation stage to the design intelligent agent and the construction intelligent agent for optimization in subsequent design and construction, realizing the efficient collaboration of all parties involved in the whole life cycle of the building. The intelligent agents can conduct information interaction in a timely and accurate manner, avoiding problems such as poor information communication and disconnection in each stage, and greatly improving the project promotion efficiency.

[0018] Furthermore, in step S2, the Building Information Model (BIM) serves as the basis for information interaction among the intelligent agents to ensure the accuracy and consistency of information. The design intelligent agent conducts design work based on the BIM model. The construction intelligent agent formulates construction plans and conducts construction simulation according to the BIM model. The operation intelligent agent uses the BIM model for equipment management and energy optimization. Through the BIM model, each intelligent agent can obtain the required information in real time, avoiding the emergence of information islands. For example, during the construction process, construction workers can view the construction drawings and relevant information in the BIM model through mobile devices to understand the construction progress and quality requirements. During the operation stage, maintenance personnel can quickly locate the equipment position through the BIM model and view the maintenance records and technical parameters of the equipment.

[0019] Further, in the collaborative design process of step S3, based on the requirements proposed by the owner agent and the project constraints, such as site conditions, budget, functional requirements, etc., a preliminary design plan is generated. Then, the design agent sends the preliminary design plan to the construction agent and the operation agent for evaluation. The construction agent evaluates the design plan from aspects such as construction feasibility, construction cost, and construction schedule, and puts forward improvement suggestions. The operation agent evaluates the design plan from the perspective of building operation and maintenance, such as energy consumption, equipment maintenance convenience, etc., and also puts forward improvement suggestions. The design agent optimizes the design plan according to the feedback from the construction agent and the operation agent. The optimized design plan is sent to the construction agent and the operation agent for evaluation again, and this iterative process continues until all parties reach an agreement. During this process, the owner agent can participate in the evaluation and decision-making at any time to ensure that the design plan meets the owner's expectations. For example, when designing a commercial building, the construction agent may point out that a certain design plan has greater construction difficulty and higher cost, and suggest adjusting the building structure. The operation agent may propose that the air-conditioning system design of this building is not conducive to energy conservation and needs to be optimized. The design agent adjusts the design plan according to these feedbacks and finally forms a design plan that satisfies all parties; Decision optimization uses a multi-objective optimization algorithm to optimize multiple objectives during the entire life cycle of the building, such as cost, quality, construction period, environmental impact, etc. During the decision-making process, each agent provides relevant data and information according to its own objectives and tasks and participates in the multi-objective optimization. For example, the construction agent provides the cost and construction period data of different construction plans, and the operation agent provides the energy consumption and maintenance cost data under different operation strategies, The formula for the multi-objective optimization algorithm is as follows:

[0020] Among them, C_total is the total life cycle cost, E_impact is the carbon emission impact factor, and S_satisfaction is the user satisfaction. Among them, the values of the weight coefficients α, β, and γ need to comprehensively consider the specific situation of the project, such as the budget limit of the project, environmental protection requirements, and the owner's emphasis on user satisfaction, etc. Usually, they can be determined by methods such as expert experience scoring and analytic hierarchy process. The value range is generally between 0 and 1, and α + β + γ = 1; Final reinforcement learning framework: The Deep Deterministic Policy Gradient (TD3) algorithm is used for policy optimization, fully considering multiple objectives within the entire building life cycle, providing a series of optimal decision-making solutions for the owner. The owner can choose according to their own needs and preferences, thereby improving the scientificity and rationality of decision-making and achieving the maximization of the overall project benefits. The multi-objective optimization algorithm generates a series of Pareto optimal solutions based on the data provided by each agent, that is, solutions that satisfy the optimal trade-off between multiple objectives. The owner agent can select the most suitable decision-making solution from the Pareto optimal solutions. For example, if the owner pays more attention to the quality and environmental impact of the building and is willing to increase costs to a certain extent to achieve these two goals, then a solution with better performance in terms of quality and environmental impact and an acceptable increase in cost can be selected from the Pareto optimal solutions; Collaboration and optimization in the construction stage include digital twin construction, dynamic adjustment of schedule resources, and safety risk warning; The digital twin construction: BIM+IoT real-time data mapping; The dynamic adjustment of schedule resources: A resource scheduling algorithm based on reinforcement learning; The safety risk warning: An LSTM network predicts the accident probability (accuracy rate ≥ 85%); The management and optimization in the operation stage include equipment fault prediction, maintenance strategy optimization, and dynamic energy consumption management. The equipment fault prediction is based on the time-series data analysis of the Transformer. In practical applications, the Transformer model has high requirements for data volume and computing resources and may face problems such as insufficient data or low computing efficiency. For insufficient data, data augmentation techniques can be used to expand the data. For the case of low computing efficiency, model compression techniques or distributed computing architectures can be utilized to improve the computing speed. The maintenance strategy optimization considers the preventive maintenance plan with the full life-cycle cost. The dynamic energy consumption management is the closed-loop control between the digital twin and the physical system. Through the participation of construction agents and operation agents, unreasonable points in the design scheme can be discovered in advance, avoiding large-scale modifications and adjustments during the construction and operation stages, thus reducing the project cost. During the construction stage, through construction progress simulation and resource allocation optimization, the construction efficiency is improved, resource waste is reduced, and the cost is further reduced. During the operation stage, through equipment maintenance prediction and energy consumption optimization, the operation cost is reduced. During the construction stage, the construction agent formulates a detailed construction plan according to the optimized design scheme and uses the BIM model for construction progress simulation and resource allocation optimization. The construction agent monitors the progress, quality, safety, etc. during the construction process in real time and feeds the relevant data back to other agents. For example, when the construction progress is delayed, the construction agent analyzes the reasons in time and transmits the information to the design agent and the owner agent. The design agent can make appropriate adjustments to the design scheme according to the situation, and the owner agent can coordinate resources to ensure the smooth progress of the project.

[0021] Meanwhile, the construction agent interacts with the intelligent devices and sensors on-site to achieve the intelligent management of the construction process. For example, the environmental parameters and equipment operation status at the construction site are monitored in real time through sensors, and intelligent devices are used for automated construction to improve the construction efficiency and quality.

[0022] During the building operation stage, the operation agent uses Internet of Things technology and sensors to collect the operation data of building equipment, energy consumption data, personnel activity data, etc. Through the analysis of these data, the operation agent predicts the time of equipment failure, conducts maintenance in advance, optimizes the energy consumption strategy, and improves the operation efficiency of the building.

[0023] The operation agent can also adjust and optimize the use functions of the building according to the feedback from the owner agent and users. For example, according to the feedback from users on the indoor environmental comfort, the operation parameters of the air conditioning system are adjusted; according to the requirements of the owner for cost control, the equipment maintenance plan is optimized.

[0024] In summary, the multi-agent collaborative design and decision-making optimization method for the whole life cycle of buildings according to the present invention can effectively improve the collaborative efficiency, optimize the decision-making quality, reduce the project cost, improve the building quality and operation benefits in practical projects, and has good application prospects.

[0025] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-agent collaborative design and decision-making optimization method for the whole life cycle of buildings, characterized in that It includes the following steps: S1. Construct a multi-agent system, which includes a design agent, a construction agent, an operation agent, and an owner agent; S2. Information integration based on Building Information Modeling (BIM), and the information integration based on BIM includes geometric information, physical information, construction information, and operation information; S3. A collaborative decision-making mechanism, which includes a collaborative design process, decision optimization, collaboration and optimization in the construction stage, and management and optimization in the operation stage.

2. The multi-agent collaborative design and decision-making optimization method for the whole life cycle of a building according to claim 1, characterized in that: In step S1, the design agent is responsible for the design work of the building, including schematic design, construction drawing design, etc., and can generate multiple design schemes according to project requirements and relevant specifications, and evaluate and optimize the design schemes.

3. A multi-agent collaborative design and decision-making optimization method for the whole life cycle of buildings according to claim 1, characterized in that: In step S1, the construction agent formulates a construction plan according to the design scheme, simulates the construction process, analyzes the risks and problems in construction, and provides corresponding solutions.

4. A multi-agent collaborative design and decision-making optimization method for the whole life cycle of buildings according to claim 1, characterized in that: In step S1, the operation agent is responsible for the management in the building operation stage, predicts the maintenance needs of building equipment, optimizes energy consumption, and improves operation efficiency.

5. A multi-agent collaborative design and decision optimization method for the whole life cycle of buildings according to claim 1, characterized in that: In step S1, the owner agent represents the needs and interests of the owner, and evaluates and makes decisions on the design scheme, construction plan, and operation strategy.

6. The multi-agent collaborative design and decision-making optimization method for the whole life cycle of a building according to claim 1, characterized in that: In step S2, the Building Information Modeling (BIM) serves as the basis for information interaction among agents, ensuring the accuracy and consistency of information. The design agent conducts design work based on the BIM model, the construction agent formulates a construction plan and conducts construction simulation according to the BIM model, and the operation agent uses the BIM model for equipment management and energy optimization.

7. A multi-agent collaborative design and decision optimization method for the whole life cycle of buildings according to claim 1, characterized in that: In step S3, in the collaborative design process, first, according to the requirements proposed by the owner agent and the project constraints, then the design agent sends the preliminary design scheme to the construction agent and the operation agent for evaluation. The construction agent evaluates the design scheme from aspects such as construction feasibility, construction cost, and construction schedule, and puts forward improvement suggestions. The operation agent views from the perspective of building operation and maintenance.

8. A multi-agent collaborative design and decision optimization method for the whole life cycle of buildings according to claim 1, characterized in that: In step S3, decision optimization uses a multi-objective optimization algorithm to optimize multiple objectives during the whole life cycle of the building. The formula of the multi-objective optimization algorithm is: F = α·C_total + β·E_impact + γ·S_satisfaction where C_total is the total life cycle cost, E_impact is the carbon emission impact factor, and S_satisfaction is the user satisfaction; Finally, a reinforcement learning framework: The Deep Deterministic Policy Gradient (TD3) algorithm is used for policy optimization.

9. A multi-agent collaborative design and decision-making optimization method for the whole life cycle of buildings according to claim 1, characterized in that: In step S3, the collaboration and optimization in the construction stage include digital twin construction, dynamic adjustment of progress resources, and safety risk warning; The digital twin construction: BIM + IoT real-time data mapping; The dynamic adjustment of progress resources: A resource scheduling algorithm based on reinforcement learning; The safety risk warning: The LSTM network predicts the accident probability (accuracy rate ≥ 85%).

10. A multi-agent collaborative design and decision optimization method for the whole life cycle of buildings according to claim 1, characterized in that: The management and optimization in the operation stage in step S3 include equipment fault prediction, maintenance strategy optimization, and energy consumption dynamic management. The equipment fault prediction is based on the time series data analysis of the Transformer. The maintenance strategy optimization considers the preventive maintenance plan with the full life cycle cost. The energy consumption dynamic management is the closed-loop control between the digital twin and the physical system.

Images