CN-121998364-A - Construction job collaborative scheduling method and system based on multi-source data fusion

CN121998364ACN 121998364 ACN121998364 ACN 121998364ACN-121998364-A

Abstract

The invention discloses a construction job collaborative scheduling method and system based on multi-source data fusion, and relates to the technical field of job scheduling. The method comprises the steps of executing multi-source data acquisition, establishing multi-source time sequence state tensors, inputting the state tensors into a BIM model, updating and generating digital twin bodies synchronous with construction states according to BIM model parameters, abstracting an operation unit into a plurality of collaborative agents, utilizing the digital twin bodies to configure twin states and resource constraints of the agents, executing game analysis through a benefit function, establishing local strategies, and aggregating the local strategies and carrying out collaborative scheduling of construction operations. The method solves the technical problems that the construction state is difficult to accurately simulate in real time, the cooperation among the operation units lacks scientific policy guidance, and conflicts and resource waste are easy to occur, and achieves the technical effects of providing a scientific policy for the cooperation of the operation units, reducing conflicts, optimizing resource allocation and improving the construction cooperation efficiency by constructing a digital twin body which synchronously evolves with the construction state.

Inventors

WANG KEXIN
SONG CHENGLIN
CHENG SHIFENG
REN CHANGHE
WU HUITAO
ZHANG ZHIQING
WANG XU
JIANG LIGUO
LIU GUANGUO
WANG SIHAI
ZHANG YU
DONG ZHI
CHEN XIXI
Tian Wenmao
WEI JUN

Assignees

中铁十二局集团有限公司
南京智航科技发展有限公司
中铁建南方建设投资有限公司
苏交科集团股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (10)

1. The construction job collaborative scheduling method based on multi-source data fusion is characterized by comprising the following steps of: performing multi-source data acquisition, and establishing multi-source time sequence state tensors by utilizing a space-time correlation mechanism, wherein the multi-source data comprise geological radar and drilling feedback data, construction mechanical equipment operation data, material and personnel positioning data and environment monitoring data; Inputting the multisource time sequence state tensor into a three-dimensional underground space BIM model, and updating according to parameters of the three-dimensional underground space BIM model to generate a digital twin body which synchronously evolves with the applied construction state; After the operation unit is read, the operation unit is abstracted into a plurality of cooperative agents, wherein each cooperative agent has the functions of local perception, task priority evaluation and conflict constraint learning; After the digital twin body is utilized to configure the local twin state and the resource constraint condition of each cooperative agent, game analysis of the cooperative agents is executed through an instant benefit function, and a local execution strategy is established; And carrying out global state aggregation on the local execution strategy, establishing a global correction result, and carrying out cooperative scheduling of construction operation according to the global correction result.
2. The collaborative scheduling method for construction jobs based on multi-source data fusion according to claim 1, wherein performing a game analysis of collaborative agents via instant benefit functions comprises: Extracting a state variable set of a corresponding operation unit from the digital twin body for each cooperative agent, and normalizing the state variable set to construct a local state feature vector; Reading on-site resource use data, establishing resource constraint conditions according to the resource use data, and limiting a policy space of the collaborative intelligent agent according to the resource constraint conditions; Constructing an instant benefit function by utilizing the local state characteristic vector, wherein the instant benefit function comprises a geological disturbance risk item, an equipment load rate item, a job progress deviation item and an energy consumption cost item; And carrying out profit evaluation on the fitting execution policy of each cooperative agent in the policy space by utilizing the instant profit function, sharing local profit and conflict constraint based on a communication topological structure, carrying out game iteration according to a sharing result, and establishing the local execution policy.
3. The collaborative scheduling method for construction jobs based on multi-source data fusion according to claim 2, wherein sharing local benefits and conflict constraints based on a communication topology, performing game iteration according to the sharing result, comprises: constructing a communication topology structure chart according to the spatial position, the operation type, the equipment interference relation and the resource coupling degree of each cooperative agent in the three-dimensional underground space BIM model, wherein the edge weight of the communication topology structure chart dynamically represents the information interaction strength and the resource competition strength between the cooperative agents; After each cooperative agent calculates the profit evaluation, carrying out neighborhood-level weighted broadcast on the local profit vector and the conflict constraint parameter through the communication topological structure; after each cooperative agent receives the neighborhood sharing result, executing the self-adaptive updating of the instant benefit function according to the neighborhood benefit fluctuation rate and the resource competition gradient, generating the strategy searching preference of the local strategy, and executing game iteration according to the self-adaptively updated instant benefit function and the strategy searching preference.
4. The collaborative scheduling method for construction jobs based on multi-source data fusion according to claim 3, wherein performing game iterations according to adaptively updated instant benefit functions and policy search preferences comprises: performing instant profit reevaluation on the candidate strategy set of the corresponding collaborative agent by using the updated instant profit function, and establishing a profit change gradient vector; Generating a strategy migration direction parameter according to the gain change gradient vector and the strategy search preference, sharing the strategy migration direction parameter in a communication topological structure, and executing double-channel game updating, wherein the double channels comprise a synchronous gain migration channel and a conflict constraint feedback channel, the synchronous gain migration channel is used for adjusting the probability distribution of a strategy according to a neighborhood average gain trend, and the conflict constraint feedback channel is used for compensating a resource use plan according to the neighborhood conflict constraint change; And establishing a local execution strategy according to the multi-round double-channel game iteration.
5. The collaborative scheduling method for construction jobs based on multi-source data fusion according to claim 1, wherein performing global state aggregation on the local execution policy to establish a global correction result comprises: acquiring all local execution strategies, and constructing a global state tensor according to the spatial dependency relationship in the digital twin body and task time sequence constraint, wherein the global state tensor characterizes the space-time interference distribution and resource consumption mapping of each cooperative agent in the underground space; Calculating a strategy consistency index and a resource conflict index according to the global state tensor, and marking a local area with the strategy consistency index lower than a first threshold value or the resource conflict index higher than a second threshold value as a global deviation area; and carrying out compensation optimization on the global deviation area under a dynamic constraint alignment mechanism, and establishing a global correction result.
6. The collaborative scheduling method for construction jobs based on multi-source data fusion according to claim 5, wherein the compensation optimization under the dynamic constraint alignment mechanism is performed on the global deviation area, comprising: extracting an associated cooperative agent constraint set in the global deviation area, establishing a constraint coupling matrix according to the cooperative agent constraint set, and calculating constraint differential vectors of all cooperative agents by using the constraint coupling matrix; Performing gradient mapping on the constraint differential vector, executing self-adaptive offset on constraint items of the instant benefit function of the cooperative agent, and establishing a constraint correction vector; And carrying out neighborhood propagation updating of the communication topological structure by using the constraint correction vector, and executing cooperative adjustment to complete compensation optimization.
7. The collaborative scheduling method for construction jobs based on multi-source data fusion according to claim 1, wherein collaborative scheduling for construction jobs is performed according to the global correction result, comprising: Configuring a task verification node by using the global correction result; performing node-by-node collaborative execution state verification based on the task verification node to generate a node verification data set; And performing cooperative scheduling compensation management according to the node verification data set.
8. The collaborative scheduling method for construction operations based on multi-source data fusion according to claim 7, wherein collaborative scheduling compensation management is performed according to the node verification data set, comprising: performing early warning signal matching of task execution deviation according to the node verification data set; And after the early warning strategy is configured by utilizing the early warning signal matching result, early warning report management is executed.
9. The construction job collaborative scheduling method based on multi-source data fusion according to claim 1, wherein the collaborative agents are provided with a rewarding mechanism, nonlinear target weights are cited in the rewarding mechanism, the nonlinear target weights are updated according to external environment evolution characteristics, and game analysis is performed according to the collaborative agents after adjustment of the rewarding mechanism.
10. A construction job collaborative scheduling system based on multi-source data fusion, characterized in that it is used for implementing the construction job collaborative scheduling method based on multi-source data fusion according to any one of claims 1-9, the system comprises: The data acquisition module is used for executing multi-source data acquisition and establishing multi-source time sequence state tensors by utilizing a space-time correlation mechanism, wherein the multi-source data comprise geological radar and drilling feedback data, construction mechanical equipment operation data, material and personnel positioning data and environment monitoring data; The digital twin module inputs the multi-source time sequence state tensor to a three-dimensional underground space BIM model, and generates a digital twin body synchronously evolving with the applied construction state according to the parameter updating of the three-dimensional underground space BIM model; The intelligent agent construction module is used for abstracting the operation unit into a plurality of cooperative intelligent agents after the operation unit is read, wherein each cooperative intelligent agent has the capabilities of local perception, task priority evaluation and conflict constraint learning; The game analysis module is used for executing game analysis of the cooperative agents through the instant benefit function after the digital twin body is utilized to configure the local twin state and the resource constraint condition of each cooperative agent, and establishing a local execution strategy; And the job scheduling module is used for carrying out global state aggregation on the local execution strategy, establishing a global correction result and carrying out cooperative scheduling on construction jobs according to the global correction result.

Description

Construction job collaborative scheduling method and system based on multi-source data fusion Technical Field The invention relates to the technical field of job scheduling, in particular to a construction job collaborative scheduling method and system based on multi-source data fusion. Background In the current field of underground space development and large-scale engineering construction, construction operation is rapidly developed towards the directions of large scale, complexity and refinement. The construction site is used as a dynamic complex system, and the interior of the complex system is filled with multi-source, heterogeneous and closely related elements, such as continuously disclosed geological conditions, cooperative operation of a large number of mechanical equipment, precise scheduling of materials and personnel and environmental indexes of time variation. Traditional job scheduling methods rely heavily on pre-established plans and manual experience to make decisions, and are difficult to sense and respond to global states of field transient changes in real time. The hysteresis causes the frequent existence of information barriers among the operation units, which easily causes the problems of resource allocation conflict, unsmooth procedure connection, untimely safety risk early warning and the like, not only restricts the further improvement of the construction efficiency, but also brings great uncertainty and safety concern to project management. Disclosure of Invention The application provides a construction operation collaborative scheduling method and a construction operation collaborative scheduling system based on multi-source data fusion, which solve the technical problems that construction states are difficult to accurately simulate in real time, scientific policy guidance is lacking in cooperation among operation units, and conflicts and resource waste are easy to occur. The application provides a construction job collaborative scheduling method based on multi-source data fusion, which comprises the following steps: The method comprises the steps of acquiring multi-source data, establishing multi-source time sequence state tensors by utilizing a space-time correlation mechanism, inputting the multi-source time sequence state tensors into a three-dimensional underground space BIM model, updating and generating digital twin bodies synchronously evolving with an applied construction state according to parameters of the three-dimensional underground space BIM model, abstracting an operation unit into a plurality of cooperative agents after the operation unit is read, wherein each cooperative agent has local perception, task priority assessment and conflict constraint learning capacity, configuring local twin states and resource constraint conditions of each cooperative agent by utilizing the digital twin bodies, executing game analysis of the cooperative agents through an instant benefit function, establishing a local execution strategy, conducting global state aggregation on the local execution strategy, establishing a global correction result, and conducting cooperative scheduling of construction operation according to the global correction result. In a second aspect of the present application, there is provided a construction job collaborative scheduling system based on multi-source data fusion, the system comprising: The system comprises a data acquisition module, an intelligent agent construction module, a game analysis module and an operation scheduling module, wherein the data acquisition module is used for acquiring multi-source data, the multi-source data comprise geological radar and drilling feedback data, construction machinery operation data, material and personnel positioning data and environment monitoring data, the digital twin module is used for inputting the multi-source time sequence state tensor into a three-dimensional underground space BIM model and generating digital twin bodies synchronously evolving with an applied construction state according to the BIM model parameter update of the three-dimensional underground space, the intelligent agent construction module is used for abstracting an operation unit into a plurality of cooperative intelligent agents after the operation unit is read, each cooperative intelligent agent has the local perception, task priority evaluation and conflict constraint learning capacity, the game analysis module is used for configuring the local twin state and resource constraint condition of each cooperative intelligent agent through instant gain function and then executing game analysis of the cooperative intelligent agent to establish a local execution strategy, and the operation scheduling module is used for conducting global state aggregation on the local execution strategy to establish a global correction result and carrying out construction operation cooperative scheduling according to the global correction result. One or more t