A sustainable earthwork optimization deployment method based on DC-FACO

Abstract

The application provides a sustainable earthwork optimization and deployment method based on DC-FACO, and relates to the technical field of optimization scheduling. The method comprises the following steps: collecting field construction data of a waterway project, constructing a multi-dimensional data matrix based on the field construction data; constructing a plurality of constraint conditions of earthwork deployment based on earthwork deployment logic and engineering characteristics of the waterway project, and constructing a collaborative optimization model based on the plurality of constraint conditions; and optimizing the collaborative optimization model based on a sustainable earthwork deployment model optimization mode of DC-FACO. The method can construct an economic and ecological benefit collaborative optimization model, meet the core concept of sustainable construction, and adopt a two-stage hybrid optimization framework. Through improvement of an ant colony algorithm, dynamic pheromone updating mechanism and glowworm optimization algorithm are fused to enhance global optimization capability, so that global search capability and stability are collaboratively optimized.

Description

Technical Field

[0001] This application relates to the field of optimization scheduling technology, specifically to a sustainable earthwork optimization allocation method based on DC-FACO. Background Technology

[0002] In the construction of large-scale waterway projects, the optimal allocation of earthwork is a crucial aspect affecting project efficiency, cost control, and ecological sustainability. Research on the optimal allocation of earthwork in large-scale waterway projects is of great significance for improving the economic and environmental benefits of these projects.

[0003] Traditional earthwork allocation thinking typically involves constructing allocation models that include transshipment yards. However, from a natural conditions perspective, some waterways have relatively flat terrain and stable geological conditions, allowing excavated earthwork to be easily backfilled or utilized within a short distance without the need for secondary transportation via transshipment yards. From an economic perspective, constructing transshipment yards involves numerous costs such as site rental, equipment investment, and personnel management. By rationally planning transportation routes and accurately calculating earthwork volumes, excavated earthwork can be directly transported to the filling area, effectively avoiding the high costs associated with transshipment yards and improving economic efficiency.

[0004] Faced with complex and ever-changing large-scale waterway engineering projects, existing research still has shortcomings in terms of model applicability, influencing factors, and the depth of green construction, and further enrichment and optimization of earthwork allocation models are needed. Summary of the Invention

[0005] Based on this, this application provides a sustainable earthwork optimization and allocation method based on DC-FACO. By constructing a synergistic optimization model of economic and ecological benefits, it conforms to the core concept of sustainable construction and adopts a two-stage hybrid optimization framework. By improving the ant colony algorithm, integrating the dynamic pheromone update mechanism and the firefly algorithm to enhance the global optimization capability, the DC-FACO algorithm achieves synergistic optimization of global search capability and stability.

[0006] This application provides a sustainable earthwork optimization allocation method based on DC-FACO, which includes: Collect on-site construction data of waterway engineering, and construct a multi-dimensional data matrix based on the on-site construction data; Based on the earthwork allocation logic and the engineering characteristics of the waterway project, multiple constraints for earthwork allocation are constructed, and a collaborative optimization model is constructed based on these multiple constraints. The optimization method based on the DC-FACO sustainable earthwork allocation model optimizes the collaborative optimization model, including: using a collaborative ant colony algorithm to generate an initial earthwork allocation scheme based on the multidimensional data matrix, optimizing the local search efficiency through a dynamic pheromone update mechanism, introducing a firefly algorithm to perturb and optimize the initial earthwork allocation scheme, and obtaining the global target allocation scheme and the target earthwork allocation path.

[0007] According to one embodiment of this application, the process of collecting on-site construction data of waterway engineering and constructing a multi-dimensional data matrix based on the on-site construction data includes: Collect the geographic coordinates of the filling point, the design capacity, and the transportation distance matrix; and construct the spatial parameter layer of the multidimensional data matrix based on the geographic coordinates of the filling point, the design capacity, and the transportation distance matrix. Collect the pre-excavation volume of earth and rock, the loose fill volume coefficient, and the allocation unit price, and construct the physical and economic parameter layer of the multidimensional data matrix based on the pre-excavation volume of earth and rock, the loose fill volume coefficient, and the allocation unit price; Collect the approval date of the filling point and the thickness of the topsoil backfill, and construct the time environment parameter layer of the multidimensional data matrix based on the approval date of the filling point and the thickness of the topsoil backfill; The pre-excavation volume and design capacity parameters in the on-site construction data are standardized in units, and the spatial parameter layer, the physical and economic parameter layer, and the temporal environmental parameter layer are integrated to generate the multidimensional data matrix compatible with the collaborative optimization model.

[0008] According to one embodiment of this application, the plurality of constraints include multiple constraints among supply and receipt balance constraints, stockpile capacity constraints, land acquisition project approval time constraints, rock excavation priority constraints, topsoil backfilling constraints, zero constraints and non-negative constraints on the amount of excavation at the end of allocation. The supply and receiving balance constraints include:

[0009]

[0010] in, Let i represent the total number of excavation projects, and let i represent the i-th excavation project. For time period t, from the excavation project Transport to the reclamation project The volume of earthwork, For time period t, from the excavation project Transport to the reclamation project The volume of stone, For the filling project Total demand during time period t For the filling project The total amount of storage in time period t+1; The heap capacity constraints include:

[0011] Where k is the total number of construction periods. For the filling project Maximum storage capacity; The time constraints for land acquisition project approvals include:

[0012]

[0013] in, To account for the allocation constraints across n filling projects in each of the four quarters, The matrix element represents the filling project in quarter t. Is it adjustable? The "priority constraint on the nearest stone excavation" indicates that stone excavation should be prioritized for filling projects that are close to each other. The topsoil backfill constraints include:

[0014] Where h represents the average thickness requirement for topsoil backfilling. This is the volume coefficient for loose fill soil. For the filling project The area occupied; The zero constraint on the final excavation volume after allocation includes:

[0015] in, In order to complete the planned excavation volume, In order to complete the planned volume of excavated rock; This represents the total number of filling projects; The nonnegativity constraint indicates that the value of all variables is a nonnegative number.

[0016] According to one embodiment of this application, the objective function of the collaborative optimization model is:

[0017] Where M is the objective function, representing the total allocation cost. For time period t, from the excavation project Transport to the reclamation project of The unit price for allocation, For time period t, from the excavation project Transport to the reclamation project The unit price of the stone mix. Costs for soil and water conservation.

[0018] According to one embodiment of this application, the method of generating an initial earthwork allocation scheme based on the multidimensional data matrix using a collaborative ant colony algorithm, optimizing local search efficiency through a dynamic pheromone update mechanism, and introducing a firefly algorithm to perturb and optimize the initial earthwork allocation scheme to obtain a global target allocation scheme and a target earthwork distribution path includes: The division of labor and collaborative ant colony algorithm is initialized; The algorithm for collaborative ant colony division of labor is configured to generate stone allocation paths for four quarters, update the remaining filling capacity and pheromone concentration, and the algorithm for collaborative ant colony division of labor is configured to generate earthwork allocation paths in combination with topsoil backfill constraints after stone allocation is completed, and update pheromone synchronously. The ant colony re-searches the global path based on the dynamically updated pheromone matrix and calculates the total allocation cost for the current iteration until the maximum number of iterations is reached. The lowest allocation price generated in each iteration is compared, and the minimum of all the lowest allocation prices is taken as the global target allocation price. Based on the global target allocation price, the global target allocation scheme and the target earthwork allocation path are generated in the four quarters.

[0019] According to one embodiment of this application, the step of setting the stone ant colony in the collaborative ant colony algorithm to generate stone allocation paths for four quarters, updating the remaining fill capacity and pheromone concentration, and setting the earthwork ant colony in the collaborative ant colony algorithm to generate earthwork allocation paths in conjunction with topsoil backfill constraints after stone allocation is completed, and synchronously updating pheromones, includes: Based on the exploration strategies and resource allocation for earthwork and rockwork, the ant colonies are divided into earthwork ant colonies and rockwork ant colonies; wherein, the earthwork ant colony is responsible for the allocation of resources for earthwork, and the rockwork ant colony is responsible for the allocation of resources for rockwork. Based on the pheromones left behind by the soil-moving ant colony and the stone-moving ant colony after their respective searches, the migration probability of the total ant colony in the next cycle is updated. ):

[0020]

[0021]

[0022] in, Let be the transition probability matrix of the stone. Let the probability of the stone foundation transfer be denoted as . The pheromone matrix after volatilization. Let be the earthwork transition probability matrix. The probability of earthwork foundation transfer. To volatilize the soil pheromones, The proportion of pheromone evaporation rate, This is the original pheromone matrix of the stone. This is the original pheromone matrix of the earthwork. The contribution of stone pheromone to each path calculated based on fitness. Q represents the contribution of earthwork pheromone to each path calculated based on fitness, where Q is the pheromone intensity. and These are random numbers extracted from a uniform distribution to simulate the fitness of each stone ant and earth ant.

[0023] According to one embodiment of this application, the transfer probability of earthwork and rockwork is:

[0024]

[0025] The total ant colony migration probability is converted to:

[0026]

[0027] in, and Let be the probability of moving stones and the probability of moving earth, respectively, and d(dest) be the distance from the source node src to the destination node dest. and These are the unit prices for stone and earthwork allocation, respectively. As an exponential adjustment factor, and These represent the total probability of stone movement and the total probability of earth movement from the source node src to the target node dest, respectively. and These represent the pheromone concentrations along the rock path and the earth path after volatilization, respectively.

[0028] According to one embodiment of this application, after comparing the lowest allocation price generated in each iteration, taking the minimum of all the lowest allocation prices as the global target allocation price, and generating the global target allocation scheme and target earthwork allocation path for the four quarters based on the global target allocation price, the method further includes: The firefly algorithm is initialized. The allocation path is dynamically adjusted based on the firefly location update mechanism; During the iteration process, candidate solutions that are below the ant colony cost threshold are selected and stored; Extract the lowest allocation price from all the candidate solutions, and output the earthwork allocation path and the stonework allocation path corresponding to the lowest allocation price.

[0029] According to one embodiment of this application, the initialization settings for the firefly algorithm include:

[0030]

[0031]

[0032] in, and The firefly position matrices are for stonework and earthwork assignments, respectively. The optimal stone allocation scheme obtained through the ant colony algorithm is given by N, where N is the number of fireflies in the colony. Let C represent the optimal earthwork allocation scheme obtained through the ant colony algorithm, where C represents the cost of each firefly. This represents the initial cost of fireflies.

[0033] According to one embodiment of this application, the update mechanism includes:

[0034] in, To characterize the location of fireflies in the current allocation plan, To characterize the location of fireflies in the optimal allocation scheme, As an attracting factor, Let r be the decay rate factor, and r be the Euclidean distance between the two schemes. As a disturbance factor, It is a Gaussian random vector.

[0035] Compared with existing technologies, the beneficial effects of this application are: It introduces the engineering characteristics of waterway engineering and establishes a phased and categorized earthwork allocation system for scenarios without transshipment. First, it designs differentiated transportation strategies based on the differences in physical properties between earthwork and rock; second, it dynamically optimizes the scheduling scheme by combining the approval time of the filling project, which can reduce project downtime and avoid waste of human and material resources caused by time delays. The sustainable earthwork allocation model optimization method (DC-FACO algorithm) proposed in this application, which integrates an improved ant colony algorithm and a firefly algorithm, adopts a two-stage hybrid optimization framework. It improves local search efficiency through an improved ant colony algorithm, enhances global optimization capabilities through a firefly algorithm, and balances the algorithm's convergence speed and accuracy through an adaptive parameter adjustment mechanism, thus better adapting to changes in engineering conditions. Attached Figure Description

[0036] Figure 1 A schematic diagram illustrating the steps of the DC-FACO-based sustainable earthwork optimization allocation method provided in this application embodiment.

[0037] Figure 2 This is a schematic diagram of the process flow in a waterway engineering scenario without transshipment.

[0038] Figure 3 A schematic diagram of the model improvement strategy provided in the embodiments of this application.

[0039] Figure 4 This is a flowchart illustrating the DC-ACO algorithm provided in an embodiment of this application.

[0040] Figure 5 This is a flowchart illustrating the DC-FACO algorithm provided in an embodiment of this application. Detailed Implementation

[0041] The present application will now be described in further detail with reference to experimental examples and specific embodiments. However, this should not be construed as limiting the scope of the subject matter of the present application to the following embodiments; all technologies implemented based on the content of the present application fall within the scope of protection of the present application.

[0042] Unless otherwise specified, the terms "upper," "lower," "left," "right," "center," "inner," "outer," and "side" used in the description of specific embodiments of this application to indicate orientation or positional relationships are based on the orientation or positional relationships shown in the accompanying drawings, or the orientation or positional relationship in which the product / equipment / device is usually placed during use. These terms are merely for the purpose of facilitating the description of the solution in this application or simplifying the description in specific embodiments, so as to enable those skilled in the art to quickly understand the solution, and do not indicate or imply that a particular device / component / element must have a specific orientation, or be constructed and operated in a specific positional relationship. Therefore, they should not be construed as limitations on this application.

[0043] In the description of the embodiments of this application, technical terms such as "first" and "second" only distinguish one entity or operation from another, and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary or secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0044] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0045] Please refer to Figure 1 , Figure 1 A schematic diagram illustrating the steps of the DC-FACO-based sustainable earthwork optimization and allocation method provided in this application embodiment. The DC-FACO-based sustainable earthwork optimization and allocation method may include: S1. Collect on-site construction data for the waterway project and construct a multi-dimensional data matrix based on the on-site construction data. Please refer to [link / reference]. Figure 2 , Figure 2 This is a schematic diagram of the process flow for a waterway engineering project without transshipment. The process flow for a waterway engineering project without transshipment can be summarized as follows: First, earth and rock excavation is carried out, and then the earth and rock are directly transported to the storage site; at the storage site, the site must be cleared first, and soil and water conservation measures must be carried out simultaneously to form a cyclical protection mechanism; the earth and rock transported to the storage site are systematically stockpiled through unloading and layered stockpiling processes.

[0046] S2. Based on the earthwork allocation logic and the engineering characteristics of waterway engineering, construct multiple constraints for earthwork allocation, and construct a collaborative optimization model based on multiple constraints.

[0047] S3. The collaborative optimization model is optimized using the DC-FACO-based sustainable earthwork allocation model optimization method.

[0048] The DC-FACO (Dynamic Cooperative Firefly and AntColony Optimization) provided in this application embodiment is a hybrid optimization framework that integrates dynamic cooperative ant colony optimization and firefly algorithm. It has both the local search capability of division of labor and cooperation and the global optimization capability of firefly, and can be applied to complex constraint scenarios of large-scale earthwork allocation.

[0049] Optionally, before planning the earthwork allocation path, on-site information collection and preprocessing for large-scale waterway projects can be performed. Data collection is conducted on the target project site, and based on the input requirements of the earthwork allocation mathematical model, feature parameter extraction, attribute classification, and dimensional standardization are performed on the raw data. Step S1 can specifically include: Collect the geographic coordinates of the filling points, design capacity, and transportation distance matrix, and construct a spatial parameter layer of a multi-dimensional data matrix based on the geographic coordinates of the filling points, design capacity, and transportation distance matrix; Collect the pre-excavation volume, loose fill volume coefficient, and allocation unit price of earth and rock, and construct the physical and economic parameter layer of a multi-dimensional data matrix based on the pre-excavation volume, loose fill volume coefficient, and allocation unit price of earth and rock. Collect the approval date and topsoil backfill thickness of the filling points, and construct a time environment parameter layer of a multi-dimensional data matrix based on the approval date and topsoil backfill thickness of the filling points; The pre-excavation volume and design capacity parameters in the on-site construction data are standardized in units, and the spatial parameter layer, physical and economic parameter layer, and temporal environmental parameter layer are integrated to generate a multi-dimensional data matrix compatible with the collaborative optimization model.

[0050] Among them, the raw data can be processed by extracting feature parameters, classifying attributes and standardizing dimensions through the collection of target data at the site of large-scale waterway engineering projects. The dataset sample is divided into the total excavation volume of earthwork and rockwork, the land area occupied by the filling project, the average transportation distance, the design capacity, the filling period planning, and the comprehensive allocation cost (unit price of earthwork and rockwork transportation and unit price of topsoil backfilling).

[0051] In this embodiment, a multi-dimensional data matrix is ​​constructed by hierarchically integrating three types of key parameters (spatial, physical economy, and temporal environment) to provide standardized input for subsequent collaborative optimization models. Its core objective is to structure the complex information of the engineering site, ensuring that the model can simultaneously handle multi-dimensional factors such as spatial layout, resource economy, time constraints, and ecological requirements.

[0052] For example, a spatial parameter layer can be constructed based on geographic coordinates (i.e., the geographical location of the filling point), design capacity (the maximum volume of earth and rock that the filling project can accommodate), and a transportation distance matrix (the transportation distance or cost matrix from the excavation point to the filling point). This spatial parameter layer describes the spatial distribution of filling points and the topology of the transportation network, providing basic data for path selection and capacity matching in optimization algorithms. A physical-economic parameter layer can be constructed based on pre-excavation volume (the total amount of earth and rock to be excavated in each excavation project), loose fill volume coefficient (the volume expansion coefficient of earth and rock during filling), and allocation unit price (the transportation cost per unit of earth and rock). This physical-economic parameter layer can quantify resource scale and economic costs. A temporal environmental parameter layer can be constructed based on the filling point approval date (the quarter or time node in which the filling project is allowed to commence) and topsoil backfill thickness (the minimum thickness of topsoil required to cover the green area). This temporal environmental parameter layer is used to introduce time constraints and ecological requirements.

[0053] In the above implementation process, the pre-excavation volume and design capacity parameters in the on-site construction data are standardized to eliminate dimensional differences and avoid calculation deviations caused by inconsistent units. Spatial, physical-economic, and temporal environmental parameters are integrated into a unified multidimensional data matrix, forming the standard input format for the model. The structure of the multidimensional data matrix perfectly matches the input requirements of the collaborative optimization model, ensuring that the model can directly call parameters for calculation.

[0054] In some embodiments, the multiple constraints include multiple constraints such as supply and receipt balance constraints, stockpile capacity constraints, land acquisition project approval time constraints, rock excavation priority constraints, topsoil backfilling constraints, zero constraints and non-negative constraints on the amount of excavation completed after allocation.

[0055] Supply and receiving balance constraints include:

[0056]

[0057] Where nW represents the total number of excavation projects, and i represents the i-th excavation project. For time period t, from the excavation project Transport to the reclamation project The volume of earthwork, For time period t, from the excavation project Transport to the reclamation project The volume of stone, For the filling project Total demand during time period t For the filling project The total amount of storage in time period t+1; Heap capacity constraints include:

[0058] Where k is the total number of construction periods. For the filling project Maximum storage capacity; The time constraints for land acquisition project approvals include:

[0059]

[0060] in, To account for the allocation constraints across n filling projects in each of the four quarters, The matrix element represents the filling project in quarter t. Is it adjustable? The principle of prioritizing the allocation of rock excavation to nearby filling projects indicates that rock excavation should be prioritized for these projects. Topsoil backfill constraints include:

[0061] Where h represents the average thickness requirement for topsoil backfilling. This is the volume coefficient for loose fill soil. For the filling project The area occupied; The zero constraints for the final excavation volume include:

[0062] in, In order to complete the planned excavation volume, In order to complete the planned volume of excavated rock; This represents the total number of filling projects; The nonnegativity constraint indicates that the value of all variables is nonnegative.

[0063] Alternatively, the objective function of the collaborative optimization model can be:

[0064] Where M is the objective function, representing the total allocation cost. For time period t, from the excavation project Transport to the reclamation project of The unit price for allocation, For time period t, from the excavation project Transport to the reclamation project The unit price of the stone mix. Costs for soil and water conservation.

[0065] Optionally, step S3, which optimizes the co-optimization model based on the DC-FACO sustainable earthwork allocation model, can specifically include a two-stage hybrid optimization process. Please refer to [link / reference]. Figure 3 , Figure 3 This diagram illustrates the model improvement strategy provided in this application's embodiments. By comparing the differences between the traditional and improved earthwork allocation models, the diagram visually presents the optimization direction: the traditional model primarily uses mixed allocation, failing to distinguish between the characteristics of earthwork and rock; the improved model constructs a multi-dimensional collaborative optimization framework by independently allocating earthwork and rock, dynamically allocating based on project characteristics (e.g., prioritizing nearby rockwork), setting allocation priorities, and introducing constraints on land acquisition approval time and ecological requirements for topsoil backfilling. The module and process arrows demonstrate the upgrade path from a single economic objective to multi-constraint collaboration encompassing "economy-ecology-time," improving allocation efficiency and sustainability through hierarchical management, dynamic prioritization, and ecological rules.

[0066] For example, the first stage of the hybrid optimization process can be to generate an initial earthwork allocation scheme based on a multidimensional data matrix using a collaborative ant colony algorithm, and optimize the local search efficiency through a dynamic pheromone update mechanism, specifically including: S31. Initialize the collaborative ant colony algorithm; the initialization settings include setting the parameters of the ant colony algorithm, allocating the endpoint number, the number of quarters, and the total number of ants. Number of rock (soil) ants Number of iterations The importance of pheromones The importance of heuristic functions pheromone volatilization rate Initialize the earthwork allocation data, allocation scheme matrix, ant path pheromone matrix, and earthwork and stone allocation matrix, and set the initial minimum cost to infinity.

[0067] S32. The rock ant colony in the collaborative ant colony algorithm generates rock allocation paths for four quarters, updates the remaining fill capacity and pheromone concentration, and the earthwork ant colony in the collaborative ant colony algorithm generates earthwork allocation paths based on topsoil backfill constraints after rock allocation is completed, and updates pheromones synchronously; Specifically, step S32 includes: Based on their exploration strategies and resource allocation for earthwork and rockwork, the ant colonies are divided into earthwork ant colonies and rockwork ant colonies; among them, the earthwork ant colonies are responsible for the allocation of resources for earthwork, and the rockwork ant colonies are responsible for the allocation of resources for rockwork. Based on the pheromones left by the soil-moving ant colony and the stone-moving ant colony after their respective searches, the migration probability of the total ant colony in the next cycle is updated. ):

[0068]

[0069]

[0070] in, Let be the transition probability matrix of the stone. Let the probability of the stone foundation transfer be denoted as . The pheromone matrix after volatilization. Let be the earthwork transition probability matrix. The probability of earthwork foundation transfer. To volatilize the soil pheromones, The proportion of pheromone evaporation rate, This is the original pheromone matrix of the stone. This is the original pheromone matrix of the earthwork. The contribution of stone pheromone to each path calculated based on fitness. Q represents the contribution of earthwork pheromone to each path calculated based on fitness, where Q is the pheromone intensity. and These are random numbers extracted from a uniform distribution to simulate the fitness of each stone ant and earth ant.

[0071] Stone ant colony Based on the approved time constraints, path selection is performed for each available fill capacity and dynamically adjusted transfer probability, generating four quarterly stone allocation plans. The total remaining stone allocation and each fill capacity are updated until no remaining stone allocation is available. Similarly, the earthmoving ant population... exist After allocation, the same path selection strategy is applied based on the remaining fill volumes and topsoil backfill constraints, updating the total remaining earthwork allocation and each fill volume until no earthwork allocation remains. After pheromone concentration is volatilized, calculations are performed separately. Updated after pheromone contribution.

[0072] Among them, the allocation unit price is affected by the source node ( ) to target node ( The influence of distance is considered in the transfer probability by taking both distance and allocation unit price into account. Based on the principle of separation and independent allocation according to project characteristics, a priority factor is introduced into the stone allocation area, and the weight of stone priority is adjusted accordingly. To achieve the goal of prioritizing the allocation of stone when the transportation distance is shorter, the transfer probabilities of earthwork and stonework are as follows.

[0073]

[0074]

[0075] The overall ant colony migration probability is converted to:

[0076]

[0077] in, and Let be the probability of moving stones and the probability of moving earth, respectively, and d(dest) be the distance from the source node src to the destination node dest. and These are the unit prices for stone and earthwork allocation, respectively. As an exponential adjustment factor, and These represent the total probability of stone movement and the total probability of earth movement from the source node src to the target node dest, respectively. and These represent the pheromone concentrations along the rock path and the earth path after volatilization, respectively.

[0078] S33. The ant colony re-searches the global path based on the dynamically updated pheromone matrix, calculates the total allocation cost for the current iteration, and continues until the maximum number of iterations is reached. During this process, the earthwork allocation data and allocation scheme matrix are initialized, and the ant colony uses the updated pheromone matrix... and The same strategy is used for path selection until the total amount of stone and earthwork allocated is exhausted, then the total allocation cost is calculated. This process is iterated until the maximum number of iterations is reached. If the result is positive, proceed to step S34; otherwise, return to step S32.

[0079] S34. Compare the lowest allocation price generated in each iteration, take the minimum of all lowest allocation prices as the global target allocation price, and generate the global target allocation plan and target earthwork allocation path in the four quarters based on the global target allocation price.

[0080] In step S3, the collaborative optimization model is optimized based on the DC-FACO sustainable earthwork allocation model. The second stage of the optimization process may involve introducing the firefly algorithm to perturb and optimize the initial earthwork allocation scheme, thereby obtaining the global target allocation scheme and the target earthwork distribution path. Specifically, this includes: S35. Initialize the firefly algorithm; including setting the number of fireflies. Number of iterations Disturbance factor Attraction factor decay rate factor The optimized solution is used as the initial solution for the firefly algorithm. .

[0081]

[0082]

[0083]

[0084] in, and The firefly position matrices are for stonework and earthwork assignments, respectively. The optimal stone allocation scheme obtained through the ant colony algorithm is given by N, where N is the number of fireflies in the colony. Let C represent the optimal earthwork allocation scheme obtained through the ant colony algorithm, where C represents the cost of each firefly. This represents the initial cost of fireflies.

[0085] S36. Dynamically adjust the allocation path according to the firefly location update mechanism; the update mechanism includes:

[0086] in, To characterize the location of fireflies in the current allocation plan, To characterize the location of fireflies in the optimal allocation scheme, As an attracting factor, Let r be the decay rate factor, and r be the Euclidean distance between the two schemes. As a disturbance factor, Let the vector be a Gaussian random vector. Compare the cost of each pair of fireflies; if the fireflies... The cost is higher than ,but Will to move.

[0087] S37. During the iteration process, select and store candidate solutions that are lower than the ant colony cost threshold; where the result of each run (total allocation price) is stored, and the globally optimal allocation price recorded by the ant colony is set as the threshold. If the current cost is less than Then store it in a table. Iteration, if the maximum number of iterations is reached... If the result is positive, proceed to step 38; otherwise, return to step 36.

[0088] S38. Extract the lowest allocation price from all candidate solutions and output the earthwork allocation path and stonework allocation path corresponding to the lowest allocation price. Among them, from... Extract all valid results (less than) Output the minimum total allocation price and its corresponding stone and earthwork allocation scheme.

[0089] The following description is a comparative experiment conducted in this embodiment on the traditional ant colony algorithm (ACO algorithm), the division of labor cooperative ant colony algorithm (DC-ACO), and the DC-FACO algorithm provided in this application embodiment.

[0090] ACO algorithm: All ants share pheromones indiscriminately.

[0091] The DC-ACO algorithm employs a two-stage hybrid optimization framework. First, based on different exploration strategies and resource allocation for earthwork, the ant colony is divided into different roles for specialized searches, followed by collaborative work. Pheromones are dynamically updated, and the allocation distance and unit price are closely linked to improve local search efficiency. Please refer to [link to relevant documentation]. Figure 4 , Figure 4This is a flowchart illustrating the DC-ACO algorithm. The process includes: parameter initialization: setting global parameters such as ant colony size, pheromone evaporation coefficient, and maximum iteration count, and loading a multi-dimensional data matrix (e.g., fill volume, allocation unit price, quarterly constraints, etc.). The division of labor includes independent optimization of earthwork and rockwork: based on the approval time of the filling project, it determines whether allocation is permitted in the current quarter. If not, the quarter is skipped; if permitted, the process proceeds to the division of labor optimization. The ant path selection is as follows: Rockwork ants: select a path based on fill volume, remaining rockwork allocation, and rockwork transfer probability (combined with transport distance and unit price), prioritizing allocation to nearby rockwork. If not permitted, the quarter is skipped; if permitted, the process proceeds to the division of labor optimization. Earthwork ants: select a path based on remaining fill volume, remaining earthwork allocation, and earthwork transfer probability, while satisfying the topsoil backfill constraint. After each ant completes its path selection, the fill volume and remaining earthwork volume are updated. The collaborative region includes pheromone fusion and global optimization: After the division of labor phase, the pheromone contribution values ​​of the rock-moving ants and the soil-moving ants are calculated, and the rock pheromone matrix and the soil pheromone matrix are updated. Based on the fused pheromone matrix, the entire ant colony generates a globally optimal allocation scheme, taking into account economic efficiency (allocation unit price) and ecological constraints (topsoil backfill thickness). The DC-FACO algorithm combines the improved DC-ACO algorithm with the firefly algorithm to enhance global optimization capabilities. Please see [link / reference]. Figure 5 , Figure 5 This is a flowchart illustrating the DC-FACO algorithm provided in an embodiment of this application.

[0092] The performance of the three algorithms was evaluated, mainly by considering the optimal objective function, standard deviation, and coefficient of variation (CV). The comprehensive evaluation score was calculated based on the objective function value and stability ranking (3 points for the first place). The experimental results are shown in Table 1.

[0093] Table 1

[0094] As shown in Table 1, the DC-FACO algorithm not only outperforms the ACO and DC-ACO algorithms in terms of objective function value, but also optimizes the problems of premature convergence in traditional ACO and insufficient stability in DC-ACO. It demonstrates its significant advantages in solution quality and stability, and has the highest comprehensive evaluation score, providing a better intelligent decision-making technology for the optimization and allocation of earthwork.

[0095] In summary, the solutions provided in this application can incorporate the characteristics of waterway engineering to establish a phased and categorized earthwork allocation system in a no-transfer scenario. First, differentiated transportation strategies are designed based on the physical characteristics of earthwork and rockwork. Second, the scheduling scheme is dynamically optimized in conjunction with the approval time of the filling project to reduce project downtime and avoid waste of manpower, materials, and other resources due to delays. Finally, the constraint of topsoil backfilling in the greening area during the later stages of construction is introduced to construct a synergistic optimization model for economic and ecological benefits, which aligns with the core concept of sustainable construction. The sustainable earthwork allocation model optimization method (DC-FACO algorithm) proposed in this application, which integrates the improved ant colony algorithm and the firefly algorithm, adopts a two-stage hybrid optimization framework. The improved ant colony algorithm enhances local search efficiency, while the firefly algorithm strengthens global optimization capabilities. An adaptive parameter adjustment mechanism balances the algorithm's convergence speed and accuracy, better adapting to changes in engineering conditions.

[0096] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A sustainable earthwork optimization allocation method based on DC-FACO, characterized in that, include: Collect on-site construction data of waterway engineering, and construct a multi-dimensional data matrix based on the on-site construction data; Based on the earthwork allocation logic and the engineering characteristics of the waterway project, multiple constraints for earthwork allocation are constructed, and a collaborative optimization model is constructed based on these multiple constraints. The collaborative optimization model is optimized based on the DC-FACO algorithm, including using a collaborative ant colony algorithm to generate an initial earthwork allocation scheme based on the multidimensional data matrix, optimizing local search efficiency through a dynamic pheromone update mechanism, and introducing a firefly algorithm to perturb and optimize the initial earthwork allocation scheme in order to obtain a global target allocation scheme and a target earthwork allocation path. Among them, the multiple constraints include multiple constraints such as supply and receipt balance constraints, stockpile capacity constraints, land acquisition project approval time constraints, rock excavation priority constraints, topsoil backfilling constraints, zero constraints and non-negative constraints on the amount of excavation after allocation. The supply and receiving balance constraints include: in, Let i represent the total number of excavation projects, and let i represent the i-th excavation project. For time period t, from the excavation project Transport to the reclamation project The volume of earthwork, For time period t, from the excavation project Transport to the reclamation project The volume of stone, For the filling project Total demand during time period t For the filling project The total amount of storage in time period t+1; The heap capacity constraints include: Where k is the total number of construction periods. For the filling project Maximum storage capacity; The time constraints for land acquisition project approvals include: in, To account for the allocation constraints across n filling projects in each of the four quarters, The matrix element represents the filling project in quarter t. Is it adjustable? The "priority constraint on the nearest stone excavation" indicates that stone excavation should be prioritized for filling projects that are close to each other. The topsoil backfill constraints include: Where h represents the average thickness requirement for topsoil backfilling. This is the volume coefficient for loose fill. For the filling project The area occupied; The zero constraint on the final excavation volume after allocation includes: in, In order to complete the planned excavation volume, In order to complete the planned volume of excavated rock; This represents the total number of filling projects; The nonnegativity constraint indicates that the value of all variables is a nonnegative number; The objective function of the collaborative optimization model is: Where M is the objective function, representing the total allocation cost. For time period t, from the excavation project Transport to the reclamation project of The unit price for allocation, For time period t, from the excavation project Transport to the reclamation project The unit price of the stone mix. Costs for soil and water conservation.

2. The method according to claim 1, characterized in that, The collected on-site construction data of the waterway project is used to construct a multi-dimensional data matrix, including: Collect the geographic coordinates of the filling point, the design capacity, and the transportation distance matrix; and construct the spatial parameter layer of the multidimensional data matrix based on the geographic coordinates of the filling point, the design capacity, and the transportation distance matrix. Collect the pre-excavation volume of earth and rock, the loose fill volume coefficient, and the allocation unit price, and construct the physical and economic parameter layer of the multidimensional data matrix based on the pre-excavation volume of earth and rock, the loose fill volume coefficient, and the allocation unit price; Collect the approval date of the filling point and the thickness of the topsoil backfill, and construct the time environment parameter layer of the multidimensional data matrix based on the approval date of the filling point and the thickness of the topsoil backfill; The pre-excavation volume and design capacity parameters in the on-site construction data are standardized in units, and the spatial parameter layer, the physical and economic parameter layer, and the temporal environmental parameter layer are integrated to generate the multidimensional data matrix compatible with the collaborative optimization model.

3. The method according to claim 1, characterized in that, The method employs a collaborative ant colony algorithm to generate an initial earthwork allocation scheme based on the multidimensional data matrix. A dynamic pheromone update mechanism optimizes local search efficiency, and a firefly algorithm is introduced to perturb and optimize the initial earthwork allocation scheme, thereby obtaining the global target allocation scheme and the target earthwork distribution path. This includes: The division of labor and collaborative ant colony algorithm is initialized; The algorithm for collaborative ant colony division of labor is configured to generate stone allocation paths for four quarters, update the remaining filling capacity and pheromone concentration, and the algorithm for collaborative ant colony division of labor is configured to generate earthwork allocation paths in combination with topsoil backfill constraints after stone allocation is completed, and update pheromone synchronously. The ant colony re-searches the global path based on the dynamically updated pheromone matrix and calculates the total allocation cost for the current iteration until the maximum number of iterations is reached. The lowest allocation price generated in each iteration is compared, and the minimum of all the lowest allocation prices is taken as the global target allocation price. Based on the global target allocation price, the global target allocation scheme and the target earthwork allocation path are generated in the four quarters.

4. The method according to claim 3, characterized in that, The algorithm for collaborative ant colony work involves setting up a stone ant colony to generate stone allocation paths for four quarters, updating the remaining fill capacity and pheromone concentration, and setting up an earthwork ant colony in the algorithm to generate earthwork allocation paths based on topsoil backfill constraints after stone allocation is completed, and synchronously updating pheromones, including: Based on the exploration strategies and resource allocation for earthwork and rockwork, the ant colonies are divided into earthwork ant colonies and rockwork ant colonies; wherein, the earthwork ant colony is responsible for the allocation of resources for earthwork, and the rockwork ant colony is responsible for the allocation of resources for rockwork. Based on the pheromones left behind by the soil-moving ant colony and the stone-moving ant colony after their respective searches, the migration probability of the total ant colony in the next cycle is updated. ): in, Let be the transition probability matrix of the stone. Let the probability of the stone foundation transfer be denoted as . The pheromone matrix after volatilization. Let be the earthwork transition probability matrix. The probability of earthwork foundation transfer. To volatilize the soil pheromones, The proportion of pheromone evaporation rate, This is the original pheromone matrix of the stone. This is the original pheromone matrix of the earthwork. The contribution of stone pheromone to each path calculated based on fitness. Q represents the contribution of earthwork pheromone to each path calculated based on fitness, where Q is the pheromone intensity. and These are random numbers extracted from a uniform distribution, representing the fitness of each stone ant and earth ant, respectively.

5. The method according to claim 4, characterized in that, The transfer probabilities for earthwork and stonework are: The total ant colony migration probability is converted to: in, and Let be the probability of moving stones and the probability of moving earth, respectively, and d(dest) be the distance from the source node src to the destination node dest. and These are the unit prices for stone and earthwork allocation, respectively. As an exponential adjustment factor, and These represent the total probability of stone movement and the total probability of earth movement from the source node src to the destination node dest, respectively. and These represent the pheromone concentrations along the rock path and the earth path after volatilization, respectively.

6. The method according to claim 3, characterized in that, After comparing the lowest allocation price generated in each iteration, taking the minimum of all the lowest allocation prices as the global target allocation price, and generating the global target allocation scheme and target earthwork allocation path for the four quarters based on the global target allocation price, the method further includes: The firefly algorithm is initialized. The allocation path is dynamically adjusted based on the firefly location update mechanism; During the iteration process, candidate solutions that are below the ant colony cost threshold are selected and stored; Extract the lowest allocation price from all the candidate solutions, and output the earthwork allocation path and the stonework allocation path corresponding to the lowest allocation price.

7. The method according to claim 6, characterized in that, The initialization settings for the firefly algorithm include: in, and The firefly position matrices are for stonework and earthwork assignments, respectively. The optimal stone allocation scheme obtained through the ant colony algorithm is given by N, where N is the number of fireflies in the colony. Let C represent the optimal earthwork allocation scheme obtained through the ant colony algorithm, where C represents the cost of each firefly. This represents the initial cost of fireflies.

8. The method according to claim 6, characterized in that, The update mechanism includes: in, To characterize the location of fireflies in the current allocation plan, To characterize the location of fireflies in the optimal allocation scheme, As an attracting factor, Let r be the decay rate factor, and r be the Euclidean distance between the two schemes. As a disturbance factor, It is a Gaussian random vector.

Images