Risk prediction and adaptive budget allocation system and method based on construction settlement

Abstract

The invention discloses a risk prediction and adaptive budget allocation system and method based on construction settlement, and the method comprises the following steps: S1, collecting historical settlement and initial budget data of a construction project, carrying out the multi-dimensional marking, and generating a structured data set; s2, processing the data by applying multi-level time sequence analysis, and identifying key levels; s3, generating an optimal quantity-price coupling relationship based on a multi-party game model, and performing preliminary tuning; s4, performing hyper-parameter optimization through a dual-domain heterogeneous population tuning mechanism; s5, analyzing the influence of external factors on the cost by using a structural transition model, and carrying out deviation analysis; s6, risk scenes under different construction situations are generated through risk simulation, and risk factors are identified and adjusted; s7, adjusting budget allocation in real time based on an adaptive complex network mechanism; and S8, a review report is generated, and multi-party collaborative review is supported. According to the method, the dynamic adjustment capability of budget allocation is improved through various optimization mechanisms, and audit transparency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of project management and budget analysis, and in particular to a risk prediction and self-adaptive budget allocation system and method based on construction settlement. Background Art

[0002] With the rapid development of engineering management and budget analysis, budget management and preliminary budget review of construction projects have become an indispensable part of construction projects. Traditional budget management methods usually rely on experience and manual operations, involving a large amount of manual calculations and data entry, which is not only inefficient but also prone to errors and deviations. With the expansion of the scale and complexity of construction projects, how to optimize resource allocation and cost control while ensuring project quality has become an important technical challenge in the field of engineering management.

[0003] In the existing technology, the commonly used budget management and review methods are mainly based on static models and manual experience. These traditional methods have obvious limitations and defects when facing complex construction environments, which are specifically reflected in the following aspects:

[0004] 1. Data collection and processing are highly manual: Traditional construction project data collection often relies on manual input, involving a variety of information such as historical settlement data, budget data, and material costs. Due to the lack of automated tool support, manual processing is not only time-consuming and labor-intensive, but also prone to data entry errors, affecting the final budget review results. In addition, existing methods lack the ability to efficiently parse and process unstructured data, resulting in a lot of valuable information not being used in budget analysis in a timely manner.

[0005] 2. Single data analysis method, unable to cope with the complexity of multi-dimensional data: In the existing technology, traditional budget review mostly adopts a single static model, which makes it difficult to effectively analyze and process various multi-dimensional dynamic data in the construction process. This leads to the review results often not accurately reflecting the real-time status of the project, especially when sudden changes occur during the progress of the construction project, the traditional method is difficult to adjust the budget allocation in real time, which leads to waste of funds and unreasonable resource allocation.

[0006] 3. Lack of effective response mechanism for multi-party game: There are multiple stakeholders involved in construction projects, and the distribution of interests among these stakeholders is often complicated. The existing budget management system lacks a game model for different stakeholders, making it difficult to effectively balance the needs of all parties. The traditional static allocation method cannot cope with the complexity of this multi-party game, which easily leads to unfair budget allocation, thus affecting the overall progress and quality of the construction project.

[0007] 4. Hyperparameter tuning lacks flexibility: Existing budget review and adjustment systems usually rely on empirical values ​​or fixed models for parameter settings and lack a dynamic tuning mechanism. This results in insufficient system adaptability in the face of complex and frequently changing construction environments, and the inability to flexibly adjust parameters based on real-time feedback from the project, which in turn affects the accuracy of budget allocation. In addition, since hyperparameters cannot be optimized in real time, the efficiency of project funding is difficult to improve, increasing construction costs.

[0008] 5. Insufficient consideration of external factors: The budget of a construction project is not only affected by internal factors of the project, but also involves external factors such as market price fluctuations, policy changes, natural disasters, etc. In the existing technology, the traditional budget management system lacks sufficient consideration of these external factors and fails to incorporate external variables into the budget model for dynamic adjustment, resulting in the inability of the budget system to respond quickly to sudden external events, which brings greater uncertainty and risk to project management.

[0009] 6. Lack of intelligent risk prediction and simulation mechanism: In the existing budget review system, risk assessment mainly relies on experience judgment or static analysis models, and lacks intelligent risk prediction and simulation tools. Due to the lack of sufficient simulation and prediction of potential risks, traditional systems are difficult to identify potential risks and abnormal factors in construction projects in a timely manner, which leads to problems such as budget overruns and waste of resources in the later stages of the project. In addition, most of the existing risk assessment methods cannot provide multi-scenario simulation and risk response solutions, which affects the flexibility and accuracy of project management.

[0010] 7. Lack of transparency in the audit and review process: In existing technologies, the budget audit and review process of construction projects usually relies on paper documents or simple spreadsheets, with low information transparency and inability to achieve multi-party collaborative review. Since the parties have limited access to budget results, it is easy to cause information asymmetry, which increases the sense of distrust in the audit process. In addition, traditional methods are difficult to effectively record and track key nodes in the budget adjustment process, which brings difficulties to the subsequent audit and correction work.

[0011] Therefore, how to provide a risk prediction and adaptive budget allocation system and method based on construction settlement is an urgent problem that technicians in this field need to solve. Summary of the invention

[0012] One purpose of the present invention is to propose a risk prediction and adaptive budget allocation system and method based on construction settlement. The present invention uses multi-level time series marginal effect analysis, multi-party game equilibrium quantity and price tuning framework, dual-domain heterogeneous population tuning mechanism and cross-deep fake risk simulation and other technologies to describe in detail the whole process of construction project budget review and dynamic adjustment. The method has the advantages of optimizing budget allocation, improving review accuracy, and responding to construction risks in real time, and improves the transparency of the audit process and the efficiency of multi-party collaborative review.

[0013] A risk prediction and adaptive budget allocation system and method based on construction settlement according to an embodiment of the present invention includes the following steps:

[0014] S1. Collect historical settlement data and preliminary budget data of construction projects, perform multi-dimensional annotation, and form a structured data set. Historical settlement data refers to the settlement records of construction costs generated in previous engineering construction projects, including material costs, labor costs, equipment usage costs, etc.

[0015] S2. Use multi-level time series marginal effect analysis techniques to clean and integrate the generated data sets and identify the key levels that affect the review results;

[0016] S3. Based on the preprocessed data, the multi-party game equilibrium quantity-price optimization framework is used to build a game model among multiple stakeholders in the construction project, analyze and generate the global optimal quantity-price coupling relationship, and output the initial optimization results;

[0017] S4. Using the dual-domain heterogeneous population tuning mechanism, the generated initial tuning results are optimized for hyperparameters to generate optimization parameters, including quantity-price coupling coefficient, smoothing factor, regression coefficient of high-order features, error term and error smoothing coefficient, etc.;

[0018] S5. Based on the optimized parameters, use the structural change model to analyze the impact of changes in external factors on the construction project cost, identify the deviations caused by external factors through deviation analysis, output the deviation identification results and form an adjustment plan;

[0019] S6. Through cross-deep fake risk simulation, based on the deviation analysis results, generate risk simulation scenarios under different construction situations, compare simulation with actual data, identify potential outliers and risk factors, and feed back to the tuning mechanism for adjustment;

[0020] S7. Apply an adaptive complex network budget adjustment mechanism to adjust budget allocation in real time based on potential outliers and risk factors;

[0021] S8. Integrate analysis results to generate review reports for construction projects, support multi-party collaborative reviews, and optimize the transparency of audits and reviews.

[0022] Optionally, S3 includes the following steps:

[0023] S31. Based on the preprocessed structured data set, extract the key influencing factors related to the construction project, including material cost, labor cost and equipment utilization rate, and perform standardization and denoising on them;

[0024] S32. Obtain the processed data, build a quantity-price coupling analysis model, and set the objective function as:

[0025]

[0026] Among them, Y is the target variable, β0 is the constant term, β i is the regression coefficient of each influencing factor, γ j is the regression coefficient of the high-order feature, β i X i As the influencing factors, is the square term of the independent variable, capturing the nonlinear relationship, δ is the time decay weight, t is the time, τ is the time constant, and ∈ is the error term;

[0027] S33. Based on the objective function, a multi-party game model is constructed, and the quantity-price relationship in the model is adjusted in real time through a dynamic adaptive quantity-price coupling coefficient Q:

[0028]

[0029] Among them, Q t is the dynamic adaptive quantity-price coupling coefficient at time step t, α is the smoothing factor, and (1-α) is used to adjust the impact of the input variable at the current moment on the quantity-price coupling coefficient. represents the kth variable X k The impact on the relationship between quantity and price, represents the influence of nonlinear relationship, X l is the lth input variable, σ is a constant, λ is an adjustment factor, C and L represent cost and load respectively;

[0030] S34. Through the multi-party game equilibrium quantity-price tuning framework, based on the dynamic adjustment results of quantity and price, the global optimal quantity-price coupling relationship in the construction project is generated, and the initial tuning results are output. The results include the optimal values ​​of the quantity-price coupling coefficient and the smoothing factor, which serve as a reference for subsequent hyperparameter optimization and review decisions.

[0031] Optionally, S4 includes the following steps:

[0032] S41. Based on the initial tuning results generated by the game model, a global exploration population and a local fine tuning population are constructed. The global exploration population is used to explore in a wide search space, and the local fine tuning population is used to finely optimize the candidate solutions generated by the global exploration population.

[0033] S42, the global exploration population adopts a multi-scale hierarchical shuttle optimization algorithm, which conducts extensive searches at high-scale levels and performs refined optimization at low-scale levels, dynamically adjusting the search range;

[0034] S43. After individuals in the global exploration population perform shuttle optimization at different scale levels, potential optimal solutions are selected based on the fitness results, and the optimization objective function of multi-scale search is set as:

[0035]

[0036] Among them, Z is the global optimization objective function, β0 is a constant term, and β i is the linear regression coefficient, P i represents the i-th optimization parameter in the global exploration population, γ j is the square term coefficient, Q t is the dynamic adaptive quantity-price coupling coefficient, δ is the adjustment factor that captures the temporal dynamic changes in global exploration, λ is the activation factor, and P k The kth optimization parameter, θ is the threshold, and η is the error term;

[0037] S44, the local fine tuning population further optimizes the candidate solutions of the global exploration population, and sets the local optimization objective function as:

[0038]

[0039] Among them, W is the local optimization objective function, Q t is the dynamic adaptive quantity-price coupling coefficient, μ0 is a constant term, μ k is the local parameter R k The linear regression coefficient of is, γ is the local adjustment factor, is the global optimization objective function Z with respect to the optimization parameter P k The first derivative of Denotes the dynamic adaptive quantity-price coupling coefficient Q t The rate of change over time, δ is the time adjustment factor, t is the time variable, τ1 is the local time constant, and ∈ is the error term;

[0040] S45, the global exploration population and the local fine-tuning population are managed through a step-by-step fitness elimination mechanism to optimize the population. Through the differentiation of fitness standards, the population deeply optimizes the accuracy.

[0041] S46. Based on the optimization results, the global optimal hyperparameter combination is finally generated. Through fine-tuning, the optimization result is finally formed to provide support for subsequent review and construction resource allocation.

[0042] Optionally, S5 includes the following steps:

[0043] S51. Based on the parameters generated by the optimized global exploration and local fine tuning, a structural change model is constructed to capture the dynamic impact of external factors on the construction project cost:

[0044]

[0045] Among them, V t is the impact of external factors on construction cost at time t, θ0 is a constant term, External factors F i The cumulative impact on construction cost is is a nonlinear recursive relationship, sgn(F i is the symbolic function, is the smooth adjustment term for local optimization, μ is the adjustment coefficient, θ i External factors F i The regression coefficient of

[0046] S52. Through the above structural change model, the impact of external factors on construction cost is captured in real time, and by analyzing the changing trend of external factors, the key external factors that cause construction cost fluctuations are identified, and real-time deviation analysis is dynamically generated;

[0047] S53. Based on deviation analysis, identify the deviation of construction project cost caused by external factors. The deviation calculation formula is as follows:

[0048]

[0049] Among them, D t is the deviation value of the external factors on the construction cost at time t, V t External factors F i The partial derivative captures the sensitivity of external factors, For the modulo operation mod(W,F i ) introduces the periodic relationship between local optimization and external factors, is the smooth adjustment of the local optimization result to the overall deviation, and κ is the adjustment coefficient;

[0050] S54, generating an adjustment plan for the construction cost using a void adjustment algorithm according to the result of the deviation analysis;

[0051] S55. Apply the generated adjustment plan to the cost model of the construction project to form a new benchmark cost, and enter it into the cost database. According to changes in external factors, the benchmark cost model is regularly updated.

[0052] Optionally, the S6 includes the following steps:

[0053] S61. Based on the deviation analysis results, we apply cross-deep fake risk simulation and combine it with nonlinear adaptive cross-response mechanism to build multi-dimensional risk scenarios in construction projects, simulate the risk transmission mechanism between external factors and construction cost, and identify potential risks in the construction process:

[0054]

[0055] Among them, R t is the construction cost risk value at time t, θ i External factors F i The weight coefficient, F i (t is the external factor at time t, W is the local optimization parameter, α is the adjustment coefficient, σ t is a random disturbance term, used to capture random risk factors;

[0056] S62. Generate risk propagation paths for different scenarios in construction projects through cross-deepfake risk simulation and nonlinear adaptive cross-response mechanism, dynamically generate outlier detection results, identify high-risk nodes, and generate risk propagation paths and risk assessments for scenarios:

[0057]

[0058] Among them, CrossRisk t is the multidimensional cross-risk response matrix at time t, is the sensitivity of external factors to construction risks, It is an external factor F i The acceleration influence term, is the second-order derivative of the external factor, λ i is the adjustment coefficient, β is the risk response acceleration coefficient, is the volatility adjustment factor;

[0059] S63. Generate risk response strategies for construction projects based on risk simulation results and detected anomalies, combined with actual construction data, and prioritize resource allocation for high-risk nodes;

[0060] S64. Dynamically adjust the resource allocation strategy in the construction project based on the feedback optimization results to match the resource allocation priority with the construction task load and risk level;

[0061] S65. Adjust resource management strategies in construction projects in real time to optimize the allocation of construction resources;

[0062] S66. Enter the optimized resource allocation and tuning results into the cost model of the construction project, generate an adjustment plan for the construction project, and enable the cost model to adapt to real-time changes in construction risks.

[0063] Optionally, the S7 includes the following steps:

[0064] S71. Based on potential outliers and risk factors, a multi-level budget allocation network for construction projects is constructed through an adaptive complex network budget adjustment mechanism, and budget demand changes between nodes are tracked in real time:

[0065]

[0066] in, is the budget allocation value at time t, α i is the budget allocation coefficient, F i (t) is the external factor at time t, L i (t) is the construction load, W i (t) is the workload, β is the adjustment factor, To optimize resource weights;

[0067] S72. Analyze the dependencies and interactions of various risk nodes in the construction project through the adaptive complex network mechanism, and adjust the budget allocation of each node according to the risk factors:

[0068]

[0069] Among them, P t is the budget allocation priority at time t, P0 is the initial budget allocation priority, β i is the node budget priority coefficient, R i (t) is the risk impact value, is an exponential adjustment function used to adjust for changes in budget priorities. Optimize the partial derivatives of the weights for the budget;

[0070] S73. Based on the feedback information of the construction project and in combination with the parallel space-time budget mirror adjustment algorithm, different budget allocation strategies are executed in parallel in multiple space-time mirrors, and the budget execution results under each mirror are compared in real time to select the optimal budget allocation coefficient.

[0071] S74. Based on the optimal budget allocation coefficient, recalculate the budget allocation for each stage of the construction project, and adjust the budget in real time to adapt to new risks and load changes;

[0072] S75. Enter the budget adjustment plan into the system and regularly update the budget allocation plan based on the actual progress and feedback data of the construction project.

[0073] A risk prediction and adaptive budget allocation system based on construction settlement according to an embodiment of the present invention includes:

[0074] Data collection module: collects historical settlement data, estimated data, and budget data of construction projects through a variety of interfaces and sensors, supports multi-dimensional annotation, automatically parses unstructured files and converts them into structured data sets, and extracts project division, list, quota, main material, and equipment information;

[0075] Data processing module: cleans, denoises and normalizes the collected data to generate a unified quantity and price feature representation, and further standardizes the cost structure at different stages;

[0076] Adaptive complex network budget adjustment module: Based on budget allocation, recalculation results of sub-item costs and risk feedback mechanism, it monitors risk nodes and load changes in construction projects in real time and dynamically adjusts budget allocation;

[0077] Parallel space-time budget mirror adjustment module: By executing multiple budget allocation strategies in parallel and comparing the budget execution results in different mirrors, the optimal budget allocation coefficient is selected, the budget allocation model is dynamically optimized, and it is updated in real time based on mirror feedback;

[0078] Cost comparison and analysis module: Based on the calculation results of sub-item costs, it supports cost comparison and analysis of budget, rough estimate and construction settlement, outputs the deviation reasons and correction measures of sub-item costs and engineering quantity indicators, and automatically analyzes them by citing the deviation reason library;

[0079] Optimization feedback module: Based on the real-time feedback information in the construction project, combined with external risk factors and construction load changes, the budget allocation plan and parameters are regularly updated, the budget allocation coefficient is automatically adjusted, and the resource management strategy of the construction project is optimized;

[0080] System management module: Enter the budget adjustment plan into the system, record the actual progress and feedback data of the construction project, and regularly update the budget allocation plan based on the analysis report automatically generated by the system, so as to allocate resources and control costs of construction projects in complex environments.

[0081] The beneficial effects of the present invention are:

[0082] (1) The present invention significantly improves the accuracy and efficiency of construction project budget review by combining multi-level time series marginal effect analysis, a multi-party game equilibrium quantity-price tuning framework, and a dual-domain heterogeneous population tuning mechanism. In particular, the present invention overcomes the shortcomings of traditional methods in terms of budget allocation and real-time hyperparameter optimization for multiple stakeholders, can quickly generate a globally optimal quantity-price coupling relationship, and make budget adjustments in a dynamic environment.

[0083] (2) By introducing the cross-deep fake risk simulation technology, the present invention realizes the simulation and analysis of risk scenarios in different construction scenarios, improves the accuracy of risk identification and the adaptive ability of the system, and can respond to potential anomalies and external risk factors in the construction process in a timely manner. In addition, the adaptive complex network budget adjustment mechanism adjusts the budget allocation in real time by combining risk feedback to ensure the flexibility and accuracy of resource management.

[0084] (3) The present invention provides a comprehensive and efficient construction project budget review and management solution. Through automated data collection, intelligent budget adjustment, risk prediction and feedback optimization, the system can adapt to complex construction environments, dynamically adjust budgets, and optimize resource allocation, thereby improving the cost control and review transparency of construction projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0086] Figure 1 A flowchart of a risk prediction and adaptive budget allocation system and method based on construction settlement proposed by the present invention;

[0087] Figure 2 A schematic diagram of a process flow based on a dual-domain heterogeneous population tuning mechanism proposed by the present invention;

[0088] Figure 3 This is a schematic diagram of risk scenario generation based on a cross-deep fake risk simulation module proposed in the present invention. DETAILED DESCRIPTION

[0089] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0090] refer to Figure 1-3 , a risk prediction and adaptive budget allocation system and method based on construction settlement, comprising the following steps:

[0091] S1. Collect historical settlement data and preliminary budget data of construction projects, perform multi-dimensional annotation, and form a structured data set. Historical settlement data refers to the settlement records of construction costs generated in previous engineering construction projects, including material costs, labor costs, equipment usage costs, etc.

[0092] S2. Use multi-level time series marginal effect analysis techniques to clean and integrate the generated data sets and identify the key levels that affect the review results;

[0093] S3. Based on the preprocessed data, the multi-party game equilibrium quantity-price optimization framework is used to build a game model among multiple stakeholders in the construction project, analyze and generate the global optimal quantity-price coupling relationship, and output the initial optimization results;

[0094] S4. Using the dual-domain heterogeneous population tuning mechanism, the generated initial tuning results are optimized for hyperparameters to generate optimization parameters, including quantity-price coupling coefficient, smoothing factor, regression coefficient of high-order features, error term and error smoothing coefficient, etc.;

[0095] S5. Based on the optimized parameters, use the structural change model to analyze the impact of changes in external factors on the construction project cost, identify the deviations caused by external factors through deviation analysis, output the deviation identification results and form an adjustment plan;

[0096] S6. Through cross-deep fake risk simulation, based on the deviation analysis results, generate risk simulation scenarios under different construction situations, compare simulation with actual data, identify potential outliers and risk factors, and feed back to the tuning mechanism for adjustment;

[0097] S7. Apply an adaptive complex network budget adjustment mechanism to adjust budget allocation in real time based on potential outliers and risk factors;

[0098] S8. Integrate analysis results to generate review reports for construction projects, support multi-party collaborative reviews, and optimize the transparency of audits and reviews.

[0099] In this implementation, S3 includes the following steps:

[0100] S31. Based on the preprocessed structured data set, extract the key influencing factors related to the construction project, including material cost, labor cost and equipment utilization rate, and perform standardization and denoising on them;

[0101] S32. Obtain the processed data, build a quantity-price coupling analysis model, and set the objective function as:

[0102]

[0103] Among them, Y is the target variable, β0 is the constant term, βi is the regression coefficient of each influencing factor, γ j is the regression coefficient of the high-order feature, β i X i As the influencing factors, is the square term of the independent variable, capturing the nonlinear relationship, δ is the time decay weight, t is the time, τ is the time constant, and ∈ is the error term;

[0104] S33. Based on the objective function, a multi-party game model is constructed, and the quantity-price relationship in the model is adjusted in real time through a dynamic adaptive quantity-price coupling coefficient Q:

[0105]

[0106] Among them, Q t is the dynamic adaptive quantity-price coupling coefficient at time step t, α is the smoothing factor, and (1-α) is used to adjust the impact of the input variable at the current moment on the quantity-price coupling coefficient. represents the kth variable X k The impact on the relationship between quantity and price, represents the influence of nonlinear relationship, X l is the lth input variable, σ is a constant, λ is an adjustment factor, C and L represent cost and load respectively;

[0107] S34. Through the multi-party game equilibrium quantity-price tuning framework, based on the dynamic adjustment results of quantity and price, the global optimal quantity-price coupling relationship in the construction project is generated, and the initial tuning results are output. The results include the optimal values ​​of the quantity-price coupling coefficient and the smoothing factor, which serve as a reference for subsequent hyperparameter optimization and review decisions.

[0108] In this implementation manner, S4 includes the following steps:

[0109] S41. Based on the initial tuning results generated by the game model, a global exploration population and a local fine tuning population are constructed. The global exploration population is used to explore in a wide search space, and the local fine tuning population is used to finely optimize the candidate solutions generated by the global exploration population.

[0110] S42, the global exploration population adopts a multi-scale hierarchical shuttle optimization algorithm, which conducts extensive searches at high-scale levels and performs refined optimization at low-scale levels, dynamically adjusting the search range;

[0111] S43. After individuals in the global exploration population perform shuttle optimization at different scale levels, potential optimal solutions are selected based on the fitness results, and the optimization objective function of multi-scale search is set as:

[0112]

[0113] Among them, Z is the global optimization objective function, β0 is a constant term, and β i is the linear regression coefficient, P i represents the i-th optimization parameter in the global exploration population, γ j is the square term coefficient, Q t is the dynamic adaptive quantity-price coupling coefficient, δ is the adjustment factor that captures the temporal dynamic changes in global exploration, λ is the activation factor, and P k The kth optimization parameter, θ is the threshold, and η is the error term;

[0114] S44, the local fine tuning population further optimizes the candidate solutions of the global exploration population, and sets the local optimization objective function as:

[0115]

[0116] Among them, W is the local optimization objective function, Q t is the dynamic adaptive quantity-price coupling coefficient, μ0 is a constant term, μ k is the local parameter R k The linear regression coefficient of is, γ is the local adjustment factor, is the global optimization objective function Z with respect to the optimization parameter P k The first derivative of Denotes the dynamic adaptive quantity-price coupling coefficient Q t The rate of change over time, δ is the time adjustment factor, t is the time variable, τ1 is the local time constant, and ∈ is the error term;

[0117] S45, the global exploration population and the local fine-tuning population are managed through a step-by-step fitness elimination mechanism to optimize the population. Through the differentiation of fitness standards, the population deeply optimizes the accuracy.

[0118] S46. Based on the optimization results, the global optimal hyperparameter combination is finally generated. Through fine-tuning, the optimization result is finally formed to provide support for subsequent review and construction resource allocation.

[0119] In this implementation manner, S5 includes the following steps:

[0120] S51. Based on the parameters generated by the optimized global exploration and local fine tuning, a structural change model is constructed to capture the dynamic impact of external factors on the construction project cost:

[0121]

[0122] Among them, V t is the impact of external factors on construction cost at time t, θ0 is a constant term, External factors F iThe cumulative impact on construction cost is is a nonlinear recursive relationship, sgn(F i is the symbolic function, is the smooth adjustment term for local optimization, μ is the adjustment coefficient, θ i External factors F i The regression coefficient of

[0123] S52. Through the above structural change model, the impact of external factors on construction cost is captured in real time, and by analyzing the changing trend of external factors, the key external factors that cause construction cost fluctuations are identified, and real-time deviation analysis is dynamically generated;

[0124] S53. Based on deviation analysis, identify the deviation of construction project cost caused by external factors. The deviation calculation formula is as follows:

[0125]

[0126] Among them, D t is the deviation value of the external factors on the construction cost at time t, V t External factors F i The partial derivative captures the sensitivity of external factors, For the modulo operation mod(W,F i ) introduces the periodic relationship between local optimization and external factors, is the smooth adjustment of the local optimization result to the overall deviation, and κ is the adjustment coefficient;

[0127] S54, generating an adjustment plan for the construction cost using a void adjustment algorithm according to the result of the deviation analysis;

[0128] S55. Apply the generated adjustment plan to the cost model of the construction project to form a new benchmark cost, and enter it into the cost database. According to changes in external factors, the benchmark cost model is regularly updated.

[0129] In this implementation manner, S6 includes the following steps:

[0130] S61. Based on the deviation analysis results, we apply cross-deep fake risk simulation and combine it with nonlinear adaptive cross-response mechanism to build multi-dimensional risk scenarios in construction projects, simulate the risk transmission mechanism between external factors and construction cost, and identify potential risks in the construction process:

[0131]

[0132] Among them, R t is the construction cost risk value at time t, θ i External factors F iThe weight coefficient, F i (t is the external factor at time t, W is the local optimization parameter, α is the adjustment coefficient, σ t is a random disturbance term, used to capture random risk factors;

[0133] S62. Generate risk propagation paths for different scenarios in construction projects through cross-deepfake risk simulation and nonlinear adaptive cross-response mechanism, dynamically generate outlier detection results, identify high-risk nodes, and generate risk propagation paths and risk assessments for scenarios:

[0134]

[0135] Among them, CrossRisk t is the multidimensional cross-risk response matrix at time t, is the sensitivity of external factors to construction risks, It is an external factor F i The acceleration influence term, is the second-order derivative of the external factor, λ i is the adjustment coefficient, β is the risk response acceleration coefficient, is the volatility adjustment factor;

[0136] S63. Generate risk response strategies for construction projects based on risk simulation results and detected anomalies, combined with actual construction data, and prioritize resource allocation for high-risk nodes;

[0137] S64. Dynamically adjust the resource allocation strategy in the construction project based on the feedback optimization results to match the resource allocation priority with the construction task load and risk level;

[0138] S65. Adjust resource management strategies in construction projects in real time to optimize the allocation of construction resources;

[0139] S66. Enter the optimized resource allocation and tuning results into the cost model of the construction project, generate an adjustment plan for the construction project, and enable the cost model to adapt to real-time changes in construction risks.

[0140] In this implementation manner, the S7 includes the following steps:

[0141] S71. Based on potential outliers and risk factors, a multi-level budget allocation network for construction projects is constructed through an adaptive complex network budget adjustment mechanism, and budget demand changes between nodes are tracked in real time:

[0142]

[0143] in, is the budget allocation value at time t, α iis the budget allocation coefficient, F i (t) is the external factor at time t, L i (t) is the construction load, W i (t) is the workload, β is the adjustment factor, To optimize resource weights;

[0144] S72. Analyze the dependencies and interactions of various risk nodes in the construction project through the adaptive complex network mechanism, and adjust the budget allocation of each node according to the risk factors:

[0145]

[0146] Among them, P t is the budget allocation priority at time t, P0 is the initial budget allocation priority, β i is the node budget priority coefficient, R i (t) is the risk impact value, is an exponential adjustment function used to adjust for changes in budget priorities. Optimize the partial derivatives of the weights for the budget;

[0147] S73. Based on the feedback information of the construction project and in combination with the parallel space-time budget mirror adjustment algorithm, different budget allocation strategies are executed in parallel in multiple space-time mirrors, and the budget execution results under each mirror are compared in real time to select the optimal budget allocation coefficient.

[0148] S74. Based on the optimal budget allocation coefficient, recalculate the budget allocation for each stage of the construction project, and adjust the budget in real time to adapt to new risks and load changes;

[0149] S75. Enter the budget adjustment plan into the system and regularly update the budget allocation plan based on the actual progress and feedback data of the construction project.

[0150] A risk prediction and adaptive budget allocation system based on construction settlement according to an embodiment of the present invention includes:

[0151] Data collection module: collects historical settlement data, estimated data, and budget data of construction projects through a variety of interfaces and sensors, supports multi-dimensional annotation, automatically parses unstructured files and converts them into structured data sets, and extracts project division, list, quota, main material, and equipment information;

[0152] Data processing module: cleans, denoises and normalizes the collected data to generate a unified quantity and price feature representation, and further standardizes the cost structure at different stages;

[0153] Adaptive complex network budget adjustment module: Based on budget allocation, recalculation results of sub-item costs and risk feedback mechanism, it monitors risk nodes and load changes in construction projects in real time and dynamically adjusts budget allocation;

[0154] Parallel space-time budget mirror adjustment module: By executing multiple budget allocation strategies in parallel and comparing the budget execution results in different mirrors, the optimal budget allocation coefficient is selected, the budget allocation model is dynamically optimized, and it is updated in real time based on mirror feedback;

[0155] Cost comparison and analysis module: Based on the calculation results of sub-item costs, it supports cost comparison and analysis of budget, rough estimate and construction settlement, outputs the deviation reasons and correction measures of sub-item costs and engineering quantity indicators, and automatically analyzes them by citing the deviation reason library;

[0156] Optimization feedback module: Based on the real-time feedback information in the construction project, combined with external risk factors and construction load changes, the budget allocation plan and parameters are regularly updated, the budget allocation coefficient is automatically adjusted, and the resource management strategy of the construction project is optimized;

[0157] System management module: Enter the budget adjustment plan into the system, record the actual progress and feedback data of the construction project, and regularly update the budget allocation plan based on the analysis report automatically generated by the system, so as to allocate resources and control costs of construction projects in complex environments.

[0158] Embodiment 1:

[0159] In order to verify the practical feasibility of the present invention, the system was applied to an infrastructure construction project in a large city, aiming to optimize the project's construction budget management and preliminary budget review process. The project includes multiple sub-projects, such as bridge and road construction, involving multiple stakeholders. The total project budget is RMB 5 billion and the construction period is expected to be 3 years. However, in the actual construction process, traditional budget management methods have failed to effectively deal with complex interest distribution, market fluctuations and construction risk issues, resulting in frequent project delays, cost overruns and waste of resources. For this reason, the project party decided to introduce the risk prediction and adaptive budget allocation system based on construction settlement of the present invention to improve the efficiency of construction management and the rationality of budget allocation.

[0160] In this embodiment, the historical settlement data and preliminary budget data of the project are first collected to generate a multi-dimensionally labeled structured data set. The system extracts detailed construction data from each sub-project through the data collection module, including information such as construction period, budget, material cost, and staffing. Subsequently, the data processing module uses multi-level time series marginal effect analysis technology to clean and normalize the data to identify the key influencing levels in the budget review, such as the impact of steel price fluctuations on the budget. Through this step, the data processing time is reduced by about 30%, and the data accuracy is significantly improved, ensuring the reliability of subsequent analysis and review.

[0161] After data preprocessing, the system used a multi-party game equilibrium quantity-price tuning framework to analyze the needs of stakeholders in the construction project. Through analysis, the system identified the key needs of material suppliers and construction parties, such as material suppliers wanting to increase prices, and construction parties asking for an extension of the construction period. In the end, the system reached a budget allocation plan acceptable to both parties without affecting progress and costs. The tuning process reduced conflicts of interest among all parties by about 25%, significantly reduced budget allocation disputes, and increased review efficiency by about 40%.

[0162] Specifically, the system uses the following methods in multiple steps: A structural change model is constructed based on optimized parameters to capture the dynamic impact of external factors on construction costs. The model captures the impact of changes in steel market prices on project budgets in real time. As steel prices rose by 20% in the middle of the project, the system identified the deviations caused by price fluctuations and calculated the sensitivity of each external factor, so as to give priority to these key external factors in the budget. Based on the results of the deviation analysis, the system uses the void adjustment algorithm to generate a budget adjustment plan for steel price fluctuations. The system promptly recommends that the construction party use alternative materials, which alleviates the cost pressure caused by rising steel prices and ultimately reduces the additional cost of the project by about 5%. The system enters the adjusted baseline cost into the cost database and updates the model regularly to ensure that the budget can adapt to external market changes.

[0163] In the middle of the project, due to fluctuations in the raw material market and the increase in project complexity, the system activated the cross-deep fake risk simulation function to generate risk simulation scenarios under different construction scenarios. By comparing the simulation with the actual data, the system identified the potential risks brought about by steel price fluctuations and construction delays, and took the following measures: In the process of constructing multidimensional risk scenarios, the system generated a multidimensional risk scenario of rising steel prices based on deviation analysis, and used a nonlinear adaptive cross-response mechanism to simulate the risk transmission between external factors and construction costs. This mechanism identifies potential risk nodes for project construction delays and forms a multidimensional cross-risk response matrix. Based on the risk simulation results and combined with the actual construction data, the system generated a response strategy for high-risk nodes and prioritized the allocation of resources to key project sections. At the same time, the system dynamically adjusted the resource priority in the budget allocation to ensure flexible matching of risk node needs during project progress. In the end, project resource allocation was more reasonable, avoiding resource waste in the later stage.

[0164] During the project advancement, the system continuously applies the adaptive complex network budget adjustment mechanism, combined with the spatiotemporal budget mirror adjustment algorithm, to dynamically optimize the budget. Based on the real-time feedback of the construction project, the system executes different budget allocation strategies in multiple spatiotemporal mirrors, compares the budget execution results under each mirror, and finally selects the optimal budget allocation coefficient. For example, the system finds that the bridge construction is smooth and cost-saving, and timely allocates the surplus budget to the road construction project to optimize resource allocation. Based on the optimal budget allocation coefficient, the system recalculates the budget of each stage of the project in real time and dynamically adjusts it to adapt to changes in risks and loads. This adjustment mechanism improves the efficiency of project budget utilization by about 12% and ensures more flexible construction fund allocation. The system enters the adjustment plan into the budget management system and regularly updates the budget allocation plan based on actual progress and feedback data. When the construction progress is smooth, the system further optimizes the budget allocation to ensure that the funds can support the smooth advancement of each sub-project.

[0165] During the project cycle, the system generates multiple construction project review reports, and through multi-party collaborative audit and review, the transparency of the audit is improved. The system automatically records the key nodes of each budget adjustment and generates a detailed audit report for the project party and stakeholders to review. This method has increased audit efficiency by about 20%, greatly improved information transparency, and ensured the standardization and transparency of project management.

[0166] Table 1: Construction project implementation data table

[0167] index Specific data Improved data processing efficiency 30％ Fewer disputes over budget allocations 25％ Shortened decision-making time 40％ Improved budget allocation accuracy 10％ Reduced material procurement costs 20 million yuan Risk simulation to avoid additional material expenditure 5％ Reduced money waste 12％

[0168] Through the application of the method of the present invention, the efficiency of budget management and review of construction projects has been significantly improved. Specific data include: data processing efficiency in the early stage of the project has been improved by 30%; through the game equilibrium tuning framework, budget allocation disputes have been reduced by 25%, and decision-making time has been shortened by 40%; after the introduction of the dual-domain heterogeneous population tuning mechanism, the budget allocation accuracy has been improved by 10%, and the material procurement cost has been reduced by 20 million yuan; the application of risk simulation has enabled the project to avoid 5% of additional material expenditures; the budget adjustment and resource allocation in the later stage of the project have been more reasonable, and the waste of funds has been reduced by 12%. The entire project received comprehensive budget management support through this system, and was finally completed smoothly on time and on budget.

[0169] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A risk prediction and adaptive budget allocation method based on construction settlement, characterized in that: The steps include: S1. Collect historical settlement data and preliminary budget data of construction projects, perform multi-dimensional annotation, and form a structured data set; S2. Use multi-level time series marginal effect analysis techniques to clean and integrate the generated data sets and identify the key levels that affect the review results; S3. Based on the preprocessed data, the multi-party game equilibrium quantity-price optimization framework is used to build a game model among multiple stakeholders in the construction project, analyze and generate the global optimal quantity-price coupling relationship, and output the initial optimization results; S4, using the dual-domain heterogeneous population tuning mechanism, performs hyperparameter optimization on the generated initial tuning results to generate optimized parameters; S5. Based on the optimized parameters, use the structural change model to analyze the impact of changes in external factors on the construction project cost, identify the deviations caused by external factors through deviation analysis, output the deviation identification results and form an adjustment plan; S6. Through cross-deep fake risk simulation, based on the deviation analysis results, generate risk simulation scenarios under different construction situations, compare simulation with actual data, identify potential outliers and risk factors, and feed back to the tuning mechanism for adjustment; S7. Apply an adaptive complex network budget adjustment mechanism to adjust budget allocation in real time based on potential outliers and risk factors; S8. Integrate analysis results to generate review reports for construction projects, support multi-party collaborative reviews, and optimize the transparency of audits and reviews.

2. The risk prediction and adaptive budget allocation method based on construction settlement according to claim 1 is characterized in that: The S3 specifically includes: S31. Based on the preprocessed structured data set, extract the key influencing factors related to the construction project, including material cost, labor cost and equipment utilization rate, and perform standardization and denoising on them; S32. Obtain the processed data, build a quantity-price coupling analysis model, and set the objective function as: Among them, Y is the target variable, β0 is the constant term, β i is the regression coefficient of each influencing factor, γ j is the regression coefficient of the high-order feature, β i X i As the influencing factors, is the square term of the independent variable, capturing the nonlinear relationship, δ is the time decay weight, t is the time, τ is the time constant, and ∈ is the error term; S33. Based on the objective function, a multi-party game model is constructed, and the quantity-price relationship in the model is adjusted in real time through a dynamic adaptive quantity-price coupling coefficient Q: Among them, Q t is the dynamic adaptive quantity-price coupling coefficient at time step t, α is the smoothing factor, and (1-α) is used to adjust the impact of the input variable at the current moment on the quantity-price coupling coefficient. represents the kth variable X k The impact on the relationship between quantity and price, represents the influence of nonlinear relationship, X l is the lth input variable, σ is a constant, λ is an adjustment factor, C and L represent cost and load respectively; S34. Through the multi-party game equilibrium quantity-price tuning framework, based on the dynamic adjustment results of quantity and price, optimize the resource allocation in the construction project, generate the global optimal quantity-price coupling relationship, output the initial tuning results, and support subsequent hyperparameter optimization and review decisions.

3. The risk prediction and adaptive budget allocation method based on construction settlement according to claim 1 is characterized in that: The S4 specifically includes: S41. Based on the initial tuning results generated by the game model, a global exploration population and a local fine tuning population are constructed. The global exploration population is used to explore in a wide search space, and the local fine tuning population is used to finely optimize the candidate solutions generated by the global exploration population. S42, the global exploration population adopts a multi-scale hierarchical shuttle optimization algorithm, which conducts extensive searches at high-scale levels and performs refined optimization at low-scale levels, dynamically adjusting the search range; S43. After individuals in the global exploration population perform shuttle optimization at different scale levels, potential optimal solutions are selected based on the fitness results, and the optimization objective function of multi-scale search is set as: Among them, Z is the global optimization objective function, β0 is a constant term, and β i is the linear regression coefficient, P i represents the i-th optimization parameter in the global exploration population, γ j is the square term coefficient, Q t is the dynamic adaptive quantity-price coupling coefficient, δ is the adjustment factor that captures the temporal dynamic changes in global exploration, λ is the activation factor, and P k The kth optimization parameter, θ is the threshold, and η is the error term; S44, the local fine tuning population further optimizes the candidate solutions of the global exploration population, and sets the local optimization objective function as: Among them, W is the local optimization objective function, Q t is the dynamic adaptive quantity-price coupling coefficient, μ0 is a constant term, μ k is the local parameter R k The linear regression coefficient of is, γ is the local adjustment factor, is the global optimization objective function Z with respect to the optimization parameter P k The first derivative of Denotes the dynamic adaptive quantity-price coupling coefficient Q t The rate of change over time, δ is the time adjustment factor, t is the time variable, τ1 is the local time constant, and ∈ is the error term; S45, the global exploration population and the local fine-tuning population are managed through a step-by-step fitness elimination mechanism to optimize the population. Through the differentiation of fitness standards, the population deeply optimizes the accuracy. S46. Based on the optimization results, the global optimal hyperparameter combination is finally generated. Through fine-tuning, the optimization result is finally formed to provide support for subsequent review and construction resource allocation.

4. The risk prediction and adaptive budget allocation method based on construction settlement according to claim 1 is characterized in that: The S5 specifically includes: S51. Based on the parameters generated by the optimized global exploration and local fine tuning, a structural change model is constructed to capture the dynamic impact of external factors on the construction project cost: Among them, V t is the impact of external factors on construction cost at time t, θ0 is a constant term, External factors F i Cumulative impact on construction cost, is a nonlinear recursive relationship, sgn(F i is the symbolic function, is the smooth adjustment term for local optimization, μ is the adjustment coefficient, θ i External factors F i The regression coefficient of S52. Through the above structural change model, the impact of external factors on construction cost is captured in real time, and by analyzing the changing trend of external factors, the key external factors that cause construction cost fluctuations are identified, and real-time deviation analysis is dynamically generated; S53. Based on deviation analysis, identify the deviation of construction project cost caused by external factors. The deviation calculation formula is as follows: Among them, D t is the deviation value of the external factors on the construction cost at time t, V t External factors F i The partial derivative captures the sensitivity of external factors, For the modulo operation mod(W,F i ) introduces the periodic relationship between local optimization and external factors, is the smooth adjustment of the local optimization result to the overall deviation, and κ is the adjustment coefficient; S54, generating an adjustment plan for the construction cost using a void adjustment algorithm according to the result of the deviation analysis; S55. Apply the generated adjustment plan to the cost model of the construction project to form a new benchmark cost, and enter it into the cost database. According to changes in external factors, the benchmark cost model is regularly updated.

5. The risk prediction and adaptive budget allocation method based on construction settlement according to claim 1 is characterized in that: The S6 specifically includes: S61. Based on the deviation analysis results, we apply cross-deep fake risk simulation and combine it with nonlinear adaptive cross-response mechanism to build multi-dimensional risk scenarios in construction projects, simulate the risk transmission mechanism between external factors and construction cost, and identify potential risks in the construction process: Among them, R t is the construction cost risk value at time t, θ i External factors F i The weight coefficient, F i (t is the external factor at time t, W is the local optimization parameter, α is the adjustment coefficient, σ t is a random disturbance term, used to capture random risk factors; S62. Generate risk propagation paths for different scenarios in construction projects through cross-deepfake risk simulation and nonlinear adaptive cross-response mechanism, dynamically generate outlier detection results, identify high-risk nodes, and generate risk propagation paths and risk assessments for scenarios: Among them, CrossRisk t is the multidimensional cross-risk response matrix at time t, is the sensitivity of external factors to construction risks, It is an external factor F i The acceleration influence term, is the second-order derivative of the external factor, λ i is the adjustment coefficient, β is the risk response acceleration coefficient, is the volatility adjustment factor; S63. Generate risk response strategies for construction projects based on risk simulation results and detected anomalies, combined with actual construction data, and prioritize resource allocation for high-risk nodes; S64. Dynamically adjust the resource allocation strategy in the construction project based on the feedback optimization results to match the resource allocation priority with the construction task load and risk level; S65. Adjust resource management strategies in construction projects in real time to optimize the allocation of construction resources; S66. Enter the optimized resource allocation and tuning results into the cost model of the construction project, generate an adjustment plan for the construction project, and enable the cost model to adapt to real-time changes in construction risks.

6. The risk prediction and adaptive budget allocation method based on construction settlement according to claim 1 is characterized in that: The S7 specifically includes: S71. Based on potential outliers and risk factors, a multi-level budget allocation network for construction projects is constructed through an adaptive complex network budget adjustment mechanism, and budget demand changes between nodes are tracked in real time: in, is the budget allocation value at time t, α i is the budget allocation coefficient, F i (t) is the external factor at time t, L i (t) is the construction load, W i (t) is the workload, β is the adjustment factor, To optimize resource weights; S72. Analyze the dependencies and interactions of various risk nodes in the construction project through the adaptive complex network mechanism, and adjust the budget allocation of each node according to the risk factors: Among them, P t is the budget allocation priority at time t, P0 is the initial budget allocation priority, β i is the node budget priority coefficient, R i (t) is the risk impact value, is an exponential adjustment function used to adjust for changes in budget priorities. Optimize the partial derivatives of the weights for the budget; S73. Based on the feedback information of the construction project and in combination with the parallel space-time budget mirror adjustment algorithm, different budget allocation strategies are executed in parallel in multiple space-time mirrors, and the budget execution results under each mirror are compared in real time to select the optimal budget allocation coefficient. S74. Based on the optimal budget allocation coefficient, recalculate the budget allocation for each stage of the construction project, and adjust the budget in real time to adapt to new risks and load changes; S75. Enter the budget adjustment plan into the system and regularly update the budget allocation plan based on the actual progress and feedback data of the construction project.

7. A risk prediction and adaptive budget allocation system based on construction settlement, characterized in that: include: Data collection module: collects historical settlement data, estimated data, and budget data of construction projects through a variety of interfaces and sensors, supports multi-dimensional annotation, automatically parses unstructured files and converts them into structured data sets, and extracts project division, list, quota, main material, and equipment information; Data processing module: cleans, denoises and normalizes the collected data to generate a unified quantity and price feature representation, and further standardizes the cost structure at different stages; Adaptive complex network budget adjustment module: Based on budget allocation, recalculation results of sub-item costs and risk feedback mechanism, it monitors risk nodes and load changes in construction projects in real time and dynamically adjusts budget allocation; Parallel space-time budget mirror adjustment module: By executing multiple budget allocation strategies in parallel and comparing the budget execution results in different mirrors, the optimal budget allocation coefficient is selected, the budget allocation model is dynamically optimized, and it is updated in real time based on mirror feedback; Cost comparison and analysis module: Based on the calculation results of sub-item costs, it supports cost comparison and analysis of budget, rough estimate and construction settlement, outputs the deviation reasons and correction measures of sub-item costs and engineering quantity indicators, and automatically analyzes them by citing the deviation reason library; Optimization feedback module: Based on the real-time feedback information in the construction project, combined with external risk factors and construction load changes, the budget allocation plan and parameters are regularly updated, the budget allocation coefficient is automatically adjusted, and the resource management strategy of the construction project is optimized; System management module: Enter the budget adjustment plan into the system, record the actual progress and feedback data of the construction project, and regularly update the budget allocation plan based on the analysis report automatically generated by the system, so as to allocate resources and control costs of construction projects in complex environments.

Images