Construction project construction scheme process level simulation method

Abstract

The invention provides a construction project construction scheme process-level simulation method, which comprises the following steps: S1, field modeling: constructing a general project construction process model for a construction project; s2, preparing data, and establishing an engineering construction scheme process-level simulation general data structure; s3, preparing an algorithm, and establishing an operation section disturbance event simulation algorithm and a circulation process simulation algorithm; s4, simulation calculation: inputting actual construction process data for simulation, and calculating to obtain full-cycle complete simulation data of an engineering construction stage; and S5, analysis and post-processing: performing post-processing operation of process transverse comparison and disturbance event analysis by using simulation data, evaluating and analyzing problems existing in the construction scheme, and providing an auxiliary decision for adjustment of the construction scheme. According to the method, construction full-period complete simulation data can be obtained through process-level simulation, the construction scheme problem is evaluated and analyzed according to the simulation data, and then auxiliary decision making and dynamic adjustment of the construction scheme are achieved.

Description

Technical Field

[0001] The invention belongs to the field of construction organization information system, and in particular relates to a construction project construction plan process level simulation method suitable for special construction of railway engineering bridges, tunnels and roadbeds. Background Art

[0002] The construction plan of a construction project is a programmatic document for construction organization, and its contents include project overview, construction preparation, construction plan, construction technology, safeguard measures, etc. The preparation of traditional construction plans often draws on similar projects, and the contents of the plans are generally similar. There is a lack of analysis and planning for specific projects, which makes the construction plan lack of pertinence and easily leads to a disconnect between the actual construction on site and the construction plan;

[0003] Secondly, the work efficiency indicators, quality indicators, safety indicators, etc. used in the preparation of traditional construction project construction plans are usually derived from the ideal settings or formula calculations of the specifications. However, in the actual project construction process, different construction units have different management and control levels in terms of quality, safety, progress, cost, etc., and the construction levels of construction teams are different. At the same time, there are other disturbances that affect the construction progress. The actual construction situation on site is usually quite different from the situation set by the construction plan editor, resulting in a mismatch between the construction site and the construction plan. Summary of the invention

[0004] The present invention proposes a process-level simulation method and system for a construction project construction plan, which can obtain complete simulation data of the entire construction cycle through process-level simulation, evaluate and analyze problems in the construction plan in terms of quality, safety, progress, cost, etc. based on the simulation data, and then realize auxiliary decision-making and dynamic adjustment of the construction plan.

[0005] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0006] A construction project construction plan process level simulation method, comprising:

[0007] S1. Domain modeling: building a general engineering construction process model for construction projects;

[0008] S2. Data preparation: Based on the construction process model, a general data structure for process-level simulation of the engineering construction plan is established, and the previous engineering construction data is filled in;

[0009] S3, algorithm preparation, establishing the operation section disturbance event simulation algorithm and the cycle process simulation algorithm, and using historical engineering construction data to train the simulation algorithm;

[0010] S4, simulation calculation, input the actual construction process data for simulation, and calculate and obtain complete simulation data for the entire cycle of the engineering construction stage;

[0011] S5. Analysis and post-processing: Use simulation data to perform horizontal comparison of process steps and post-processing operations such as disturbance event analysis to evaluate and analyze problems in the construction plan and provide auxiliary decision-making for the adjustment of the construction plan.

[0012] Furthermore, the construction of the general engineering construction process model in step S1 includes: dividing the construction process into three layers: operation area, operation section, and cycle process; wherein:

[0013] The work area is the first layer, which is defined as a construction area that is independent of each other in terms of time, space, and resource allocation. Each construction area operates in parallel without interfering with each other;

[0014] The operation section is a secondary layer, belonging to a certain operation area, and is defined as a construction area that is loosely coupled in time and space and independent of each other in resource allocation; each operation section has strong constraints on spatial distance and strong constraints on the order of time;

[0015] The cyclic process is the bottom layer and belongs to a certain operation section. It is defined as a construction operation with strong coupling in time, space and resource allocation. The cyclic processes in the same operation section are constructed in the same space, have a sequence in time, and reuse resources.

[0016] Furthermore, the establishment of the operation section disturbance event simulation algorithm in step S3 includes:

[0017] S301, create a data set C, C = {c i |i∈[1,a max ]}; where a max Indicates the maximum value of the cumulative number of cycles when a disturbance event occurs; i indicates the cumulative number of cycles; c i is a single data sample, c i =(i, n i , d i , e i ), n i is the cumulative number of disturbances, d i is the duration of disturbance event handling, e i Indicates whether a disturbance event occurs;

[0018] S302, using a logistic regression algorithm, i ) as a data set to input the logistic regression algorithm to build a disturbance event prediction model for the operation section;

[0019] S303: predicting the probability p of a disturbance event occurring at the current cumulative number of cycles x by using the disturbance event prediction model for the operation section 1; Then compare with the model judgment threshold of the current cumulative number of cycles x to determine whether a disturbance event has occurred; the model judgment threshold is: where n x is the cumulative number of disturbances corresponding to the current cumulative number of cycles x, and D is defined as the sum of the cumulative number of disturbances;

[0020] S304, predicting the duration of disturbance event handling d: d = α × d x +β×d avg , α=0.8, β=0.2; where d x is the disturbance treatment duration corresponding to the current cumulative number of cycles x, d avg is the arithmetic mean of all disturbance treatment durations, and α represents d x The weight of the proportion, β represents d avg The proportion weight of .

[0021] Furthermore, the establishment of the cycle process simulation algorithm in step S3 includes:

[0022] S311, create a data set B, B = {b kj |k, j>0}; where b kj is a single data sample, m e represents the e-th construction condition, Indicates the construction condition m e Construction parameters under conditions, s k represents the kth process, r j represents the jth cycle, t kj It represents the j-th cycle time of the k-th process;

[0023] S312, for process s k , get the corresponding data in data set B Based on LSTM algorithm t kj Prediction model: Each process trains an independent prediction model;

[0024] S313. Select a corresponding prediction model according to the process, input the current construction conditions, construction parameters, and number of cycles into the prediction model, and output the time prediction result t of the process under the number of cycles.

[0025] Furthermore, the simulation calculation process in step S4 includes:

[0026] The duration of the key operation section is taken as the total duration of the operation area; let the operation area be u, and the key operation section of the operation area u be v; j represents the number of cycles of the operation section v, the maximum is J, and k represents the number of process items, the maximum is K;

[0027] Then the total duration of the work area u is T u It is expressed as:

[0028]

[0029] Where T uv is the duration of the key operation section v of the operation area u, T uvi is the duration of the jth cycle of the key operation section v of the operation area u, T uvkj is the duration of the i-th process in the j-th cycle of the key operation section v of the operation area u, d uvkj is the duration of disturbance event handling predicted by the operation section disturbance event simulation algorithm, t uvkj The time consumption of the current process is predicted by the cycle process simulation algorithm.

[0030] On the other hand, the present invention also proposes a construction engineering construction plan process level simulation system, comprising:

[0031] Domain modeling module: for construction projects, build a general engineering construction process model;

[0032] Data preparation module: Based on the construction process model, a general data structure for process-level simulation of the engineering construction plan is established, and the previous engineering construction data is filled in;

[0033] Algorithm preparation module: establish the operation section disturbance event simulation algorithm and the cycle process simulation algorithm, and use the historical engineering construction data to train the simulation algorithm;

[0034] Simulation calculation module: input the actual construction process data for simulation, and calculate the complete simulation data of the whole cycle of the engineering construction stage;

[0035] Analysis and post-processing module: Use simulation data to perform post-processing operations such as horizontal comparison of work processes and disturbance event analysis, evaluate and analyze problems in the construction plan, and provide auxiliary decision-making for the adjustment of the construction plan.

[0036] Furthermore, the domain modeling module is divided into work area unit, work section unit, and cycle process unit, including:

[0037] Operation area unit: The operation area is the first layer, which is defined as a construction area that is independent of each other in terms of time, space, and resource allocation. Each construction area operates in parallel without interfering with each other;

[0038] Operation section unit: The operation section is a secondary layer, belonging to a certain operation area, and is defined as a construction area that is loosely coupled in time and space and independent of each other in resource allocation; each operation section has strong constraints on spatial distance and strong constraints on the order of time;

[0039] Cyclic process unit: The cyclic process is the bottom layer and belongs to a certain operation section. It is defined as a construction operation with strong coupling in time, space and resource allocation. The cyclic processes in the same operation section are constructed in the same space, have a sequence in time, and reuse resources.

[0040] Furthermore, the algorithm preparation module is provided with an operation section disturbance event simulation algorithm submodule, including:

[0041] Create a dataset C unit: Create a dataset C, C = {c i |i∈[1,a max ]}; where a max Indicates the maximum value of the cumulative number of cycles when a disturbance event occurs; i indicates the cumulative number of cycles; c i is a single data sample, c i =(i, n i , d i , e i ), n i is the cumulative number of disturbances, d i is the duration of disturbance event handling, e i Indicates whether a disturbance event occurs;

[0042] Construct the disturbance model unit: Use the logistic regression algorithm to transform the perturbation model in the dataset C (i, e i ) as a data set to input the logistic regression algorithm to build a disturbance event prediction model for the operation section;

[0043] Judgment unit: predict the probability p1 of a disturbance event occurring when the current cumulative number of cycles x is reached by using the disturbance event prediction model of the operation section; then compare it with the model judgment threshold of the current cumulative number of cycles x to determine whether a disturbance event occurs; the model judgment threshold is: where n x is the cumulative number of disturbances corresponding to the current cumulative number of cycles x, and D is defined as the sum of the cumulative number of disturbances;

[0044] Prediction duration unit: Prediction of disturbance event handling duration d: d = α × d x +β×d avg , α=0.8, β=0.2; where d x is the disturbance treatment duration corresponding to the current cumulative number of cycles x, d avg is the arithmetic mean of all disturbance treatment durations, and α represents d x The weight of the proportion, β represents d avg The proportion weight of .

[0045] Furthermore, the algorithm preparation module is provided with a cycle process simulation algorithm submodule, including:

[0046] Dataset B unit: Create dataset B, B = {b kj |k, j>0}; where b kj is a single data sample, m e represents the e-th construction condition, Indicates the construction condition m e Construction parameters under conditions, s k represents the kth process, r j represents the jth cycle, t kj It represents the j-th cycle time of the k-th process;

[0047] Process model building unit: for process s k , get the corresponding data in data set B Based on LSTM algorithm t kj Prediction model: Each process trains an independent prediction model;

[0048] Time consumption prediction unit: select the corresponding prediction model according to the process, input the current construction conditions, construction parameters, and number of cycles into the prediction model, and output the time consumption prediction result t of the process under the said number of cycles.

[0049] Furthermore, the simulation calculation module includes:

[0050] The duration of the key operation section is taken as the total duration of the operation area; let the operation area be u, and the key operation section of the operation area u be v; j represents the number of cycles of the operation section v, the maximum is J, and k represents the number of process items, the maximum is K;

[0051] Then the total duration of the work area u is T u It is expressed as:

[0052]

[0053] Where T uv is the duration of the key operation section v of the operation area u, T uvj is the duration of the jth cycle of the key operation section v of the operation area u, T uvkj is the duration of the i-th process in the j-th cycle of the key operation section v of the operation area u, d uvkj is the duration of disturbance event handling predicted by the operation section disturbance event simulation algorithm, t uvkj The time consumption of the current process is predicted by the cycle process simulation algorithm.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. The present invention establishes a general data model and data structure, collects real data from the actual construction process, uses a series of simulation calculation methods to simulate the construction process at the process level, and obtains complete simulation data for the entire cycle of the engineering construction stage. Based on the simulation data, the problems of the construction plan in terms of quality, safety, progress, cost, etc. are evaluated and analyzed, thereby realizing auxiliary decision-making and dynamic adjustment of the construction plan.

[0056] 2. The present invention proposes to apply disturbance factors in the process-level simulation to more realistically restore the time loss caused by the construction team's construction level, work habits, organizational model, etc. at the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a schematic diagram of a method flow of an embodiment of the present invention. DETAILED DESCRIPTION

[0058] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0059] In order to make the purpose and features of the present invention more obvious and understandable, the specific implementation methods of the present invention are further described below in conjunction with the accompanying drawings.

[0060] The construction engineering construction plan process level simulation method of this embodiment is as follows Figure 1 As shown, specifically including:

[0061] S1. Domain Modeling:

[0062] According to the construction scenario of railway engineering, a general data model suitable for railway construction projects such as bridges, tunnels, and roadbeds is constructed. Specifically, the construction process is divided into three layers: work area, work section, and cycle process. Among them:

[0063] (1) The work area is the first layer and is defined as a construction area that is independent of each other in terms of time, space, and resource allocation. It is characterized by parallel operation in each construction area without interference with each other. It is generally clearly defined in the construction plan. The work area is represented by the symbol A, and multiple work areas are represented by numerical subscripts (e.g. A 1 , A 2 ).

[0064] (2) The operation section is a secondary layer, belonging to a certain operation area, and is defined as a construction area that is loosely coupled in time and space and independent in resource allocation. Its characteristics are that each operation section can operate in parallel, but there are strong constraints on spatial distance and sequence in time, but it does not require temporal continuity. The operation section is represented by the symbol S, and multiple sections are represented in the order of operation time using numerical subscripts (such as: S 1 , S2 ).

[0065] (3) The cyclic process is the bottom layer and belongs to a certain operation section. It is defined as a construction operation with strong coupling in time, space, and resource allocation. It is characterized by the unified operation area of ​​the cyclic process. All processes in the same operation section are constructed in the same space, have a sequence in time sequence, and reuse resources. The cyclic process is represented by the symbol R. The cyclic processes of each operation section are arranged in sequence in time sequence, using numerical subscripts (such as: S 1 R 1 , S 1 R B ).

[0066] S2. Data preparation:

[0067] Based on the domain modeling level division, a general data structure for process-level simulation of engineering construction schemes is established. According to the data structure, actual engineering construction data is collected and sorted for data filling, which is used as the input of subsequent simulation algorithms. Research is conducted on previous construction cases. The data specifically includes:

[0068] Process inspection data (D1), including "operation area, starting mileage, ending mileage, operation section, operation process, inspection time, approval time, whether the first inspection is qualified" data fields. The data comes from the process inspection information system;

[0069] Safety control data (D2), including data fields such as “operating area, starting mileage, ending mileage, hidden danger level, hidden danger description, troubleshooting time, rectification time, and closing time”. The data comes from the safety hidden danger troubleshooting system;

[0070] Concrete supply data (D3), including "operation area, starting mileage, ending mileage, operation section, operation process, design volume, production volume" data fields. The data comes from the mixing plant system;

[0071] Cycle process data (D4), including "operation area, start mileage, end mileage, operation section, operation process, start time, end time, process duration, overlap duration" data fields. The data comes from the cycle process construction ledger;

[0072] Disturbance event data (D5) includes the data fields of “operation area, start mileage, end mileage, operation section, operation process, disturbance event type, treatment start time, treatment end time, and treatment duration”. A disturbance event is a critical event that occurs during a cycle process and causes an increase or decrease in construction time.

[0073] S3. Algorithm preparation:

[0074] S3.1 Operation section disturbance event simulation algorithm:

[0075] Through the study of previous construction cases, it is found that under the same construction conditions, the same construction team, the disturbance events that occur within a certain construction mileage range in the operating section are regularly distributed. Therefore, the present invention establishes a function of the cumulative number of cycles (the number of cycles between the occurrence of two disturbance events in the operating section) and the disturbance event (probability of occurrence, duration of treatment) to predict the probability of the disturbance event occurring in the current operating section cycle, as well as the duration required for treatment after the occurrence (applicable to all major disturbance events in railway engineering construction). The specific method is as follows:

[0076] (1) Dataset preparation:

[0077] The cumulative number of disturbances indicates the total number of disturbances that occur under the cumulative number of cycles. Suppose the cumulative number of disturbances when the cumulative number of cycles is i is n. i , the duration of handling disturbance events is d i The cumulative number of cycles, the cumulative number of disturbances and the duration of treatment (in minutes) when a disturbance event occurs constitute data b i :b i =(i, n i , d i );

[0078] Then the set B formed by the cumulative number of cycles, the cumulative number of disturbances, and the treatment duration is:

[0079] B = {b i |b i =(i, n i , d i ); i∈[1, a max ]}; a max Indicates the maximum value of the accumulated number of cycles when a disturbance event occurs.

[0080] When no disturbance event occurs at location i, let n i Equal to 0, d i is equal to 0. Let:

[0081]

[0082] c i is a single data sample consisting of the cumulative number of cycles, the cumulative number of disturbances, the duration of treatment, and whether a disturbance event occurred, which can be simplified as: c i =(i, n i , d i , e i ), e i Indicates whether a disturbance event occurs;

[0083] Then c i The set C ofi |i∈[1,a max ]}.

[0084] (2) Building a prediction model:

[0085] The logistic regression algorithm is used to build a disturbance event prediction model for the operation section. The logistic regression algorithm is a commonly used algorithm for binary classification (true and false), and its principle is not described here. i ) is used as a data set to input the logistic regression algorithm to generate a prediction model. A disturbance event is a general concept that defines a behavior and represents the full set of all disturbance factors.

[0086] (3) Predict whether a disturbance event occurs:

[0087] The disturbance event data set of the operation section is divided into training set, validation set and test set at a ratio of 3:1:1. The training set is input into the model for a certain round of training, and the validation set is used for validation, and the model with the smallest loss value under the validation set is selected as the final training result.

[0088] The cumulative number of cycles x to be tested in the test set is input into the prediction model to obtain the probability p of a disturbance event occurring at the cumulative number of cycles. 1 ;

[0089] Calculate the sum of the cumulative perturbations in each cycle in the training set. Define a set of cycles as starting from 1 cumulative cycle and ending with the cumulative number of cycles when a perturbation event occurs. Therefore, this value is equivalent to the number of cycle groups, denoted as D, and its calculation formula is as follows:

[0090]

[0091] The cumulative number of perturbations in the training set corresponding to the cumulative number of cycles is x is n x , calculate its ratio to the number of cycle groups D as the model judgment threshold under the current cumulative number of cycles.

[0092]

[0093] If p 1 ≥ratio, it is True, indicating that a disturbance event has occurred; otherwise, it is False, indicating that no disturbance event has occurred.

[0094] In order to make the model better adaptable to the current engineering project, the current judgment threshold is dynamically updated after each test. Let the cumulative number of cycles be x. If a disturbance event is determined to occur, the number of cycle groups D is increased by 1, n x Add 1, recalculate the model judgment threshold ratio under the current cumulative number of cycles. If it is determined that no disturbance event has occurred, the number of cycle groups D is increased by 1, n xThe value remains unchanged, and the model judgment threshold ratio under the current cumulative number of cycles is recalculated.

[0095] (4) Prediction of the duration of disturbance event handling:

[0096] According to the current cumulative number of cycles, take the disturbance treatment time d x , calculate the arithmetic mean of all disturbance treatment durations d avg , and finally calculate the disturbance event handling time d:

[0097] d=α×d x +β×d avg , α = 0.8, β = 0.2;

[0098] Among them, α and β represent the weights of the disturbance handling time of the current cumulative number of cycles and the arithmetic mean of the handling time of all disturbance events.

[0099] S3.2 Cycle process simulation algorithm:

[0100] By tracking a large number of construction projects, it is found that under similar working conditions, with the same construction parameters such as personnel, machinery, materials, and construction methods, the operation duration (time consumption) of each process within a certain construction range shows regularity. The present invention proposes a cyclic process simulation algorithm based on LSTM (Long sort-term memory). The algorithm obtains the time consumption of the process operation by simulating the construction process, making the construction process simulation more realistic and providing important data reference for the adjustment of the construction plan. It is an important part of the dynamic adjustment method of the construction plan. The calculation method is as follows:

[0101] (1) Unified definition and representation of data

[0102] Let M be the set of all construction conditions, m e Represents the e-th construction condition, e=1,2,...,E, and the set M is expressed as follows:

[0103] M={m 1 , m 2 , ..., m E};

[0104] Represent construction parameters such as personnel, machinery, materials, and construction methods as a whole Let P be the set of all construction parameters, Indicates the construction condition m e The construction parameters under these conditions, set P, are expressed as follows:

[0105]

[0106] Assume that in the same operation section, S is the set of cycle processes, sk represents the kth process. The set S is represented as follows:

[0107] S={s 1 ,s 2 , ..., s k , ...}

[0108] (2) Dataset creation

[0109] In the construction condition, m τ The construction parameters are When the kth process is represented by s k , k = 1, 2, ..., K. The jth cycle is represented by r j , j = 1, 2, ..., J, then the j-th cycle time of the k-th process is expressed as t kj , let b kj is the set of the above data, then b kj For a single data sample, it is represented as follows:

[0110]

[0111] Therefore, the set B consisting of data samples of construction conditions, construction parameters, construction processes, number of cycles, and process duration is expressed as:

[0112] B = {b kj |k,j>0}

[0113] Considering the time consumption of different processes, different processes are formed into different data sets, process s k The corresponding data samples constitute data set B k .

[0114] (3) Building a prediction model

[0115] For process s i , take data set B k In Let the LSTM algorithm build and t kj Each process needs to train an independent prediction model. The LSTM algorithm is a common neural network algorithm, and its principle will not be described here.

[0116] (4) Predicting process time

[0117] The data set is divided into training set, validation set and test set in a ratio of 3:1:1. The training set is input into the model for a certain round of training, and the validation set is used for validation. The model with the smallest loss value under the validation set is selected as the final training result.

[0118] Select the corresponding prediction model according to the process, and select process s k , assuming the current construction condition is m e , construction parameters are The number of cycles r j Input the prediction model and output the jth cycle r j Middle process k The time-consuming prediction result t.

[0119] S4. Simulation calculation:

[0120] The construction process-level simulation simulates the construction process of the work area, work section, and cycle process.

[0121] The calculation process is as follows:

[0122] The duration of the key operation section is taken as the total duration of the operation area; let the operation area be u, and the key operation section of the operation area u be v; j represents the number of cycles of the operation section v, the maximum is J, and k represents the number of process items, the maximum is K;

[0123] Then the total duration of the work area u is T u It is expressed as:

[0124]

[0125] Where T uv is the duration of the key operation section v of the operation area u, T uvi is the duration of the jth cycle of the key operation section v of the operation area u, T uvkj is the duration of the i-th process in the j-th cycle of the key operation section v of the operation area u, d uvkj is the duration of disturbance event handling predicted by the operation section disturbance event simulation algorithm, t uvkj The time consumption of the current process is predicted by the cycle process simulation algorithm.

[0126] The principle of this step is: through S1 domain modeling, it is known that the operation sections have a loose coupling relationship in time and space, and there are safety step requirements for different operation sections in space. Therefore, each operation section can operate in parallel (organize the construction of different operation sections at the same time), and there is no requirement for continuity in time, but there is a sequence in time sequence. Through the study of a large number of construction projects, it is found that there are key operation sections in different engineering operation areas, which are characterized by having safety step requirements with other operation sections (such as the face operation section in tunnel construction), and the time consumption plays a decisive role in the total duration of the entire operation area. Based on this, the duration of the key operation section is determined as the total duration of the operation area.

[0127] S5. Analysis and post-processing:

[0128] The obtained complete simulation data of the whole cycle of the engineering construction stage is used to perform post-processing operations such as horizontal comparison of processes and disturbance event analysis. This is to evaluate the task allocation and resource scheduling under different processes, the collaborative efficiency between processes and the reasons for progress delays; analyze the quality risks under different processes; evaluate the consumption and cost of human, machine and material resources under different processes, and analyze the economic benefits of each process. In response to potential problems and risks found in the simulation, decision-making opinions and plans are proposed to adjust the construction sequence, optimize resource allocation, and improve the efficiency and quality of the project.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A construction project construction plan process level simulation method, characterized in that: include: S1. Domain modeling: building a general engineering construction process model for construction projects; S2. Data preparation: Based on the construction process model, a general data structure for process-level simulation of the engineering construction plan is established, and the previous engineering construction data is filled in; S3, algorithm preparation, establishing the operation section disturbance event simulation algorithm and the cycle process simulation algorithm, and using historical engineering construction data to train the simulation algorithm; S4, simulation calculation, input the actual construction process data for simulation, and calculate and obtain complete simulation data for the entire cycle of the engineering construction stage; S5. Analysis and post-processing: Use simulation data to perform horizontal comparison of process steps and post-processing operations such as disturbance event analysis to evaluate and analyze problems in the construction plan and provide auxiliary decision-making for the adjustment of the construction plan.

2. The construction engineering construction plan process level simulation method according to claim 1 is characterized in that: The construction of the general engineering construction process model in step S1 includes: dividing the construction process into three layers: operation area, operation section, and cycle process; wherein: The work area is the first layer, which is defined as a construction area that is independent of each other in terms of time, space, and resource allocation. Each construction area operates in parallel without interfering with each other; The operation section is a secondary layer, belonging to a certain operation area, and is defined as a construction area that is loosely coupled in time and space and independent of each other in resource allocation; each operation section has strong constraints on spatial distance and strong constraints on the order of time; The cyclic process is the bottom layer and belongs to a certain operation section. It is defined as a construction operation with strong coupling in time, space and resource allocation. The cyclic processes in the same operation section are constructed in the same space, have a sequence in time, and reuse resources.

3. The construction engineering construction plan process level simulation method according to claim 1 is characterized in that: The establishment of the operation section disturbance event simulation algorithm in step S3 includes: S301, create a data set C, C = {c i |i∈[1,a max ]}; where a max Indicates the maximum value of the cumulative number of cycles when a disturbance event occurs; i indicates the cumulative number of cycles; c i is a single data sample, c i =(i,n i ,d i ,e i ), n i is the cumulative number of disturbances, d i is the duration of disturbance event handling, e i Indicates whether a disturbance event occurs; S302, using a logistic regression algorithm, i ) as a data set to input the logistic regression algorithm to build a disturbance event prediction model for the operation section; S303, predicting the probability p1 of a disturbance event occurring at the current cumulative number of cycles x through the operation section disturbance event prediction model; then comparing it with the model judgment threshold of the current cumulative number of cycles x to determine whether a disturbance event occurs; the model judgment threshold is: where n x is the cumulative number of disturbances corresponding to the current cumulative number of cycles x, and D is defined as the sum of the cumulative number of disturbances; S304, predicting the duration of disturbance event handling d: d = α × d x +β×d av$ ; where d x is the disturbance treatment duration corresponding to the current cumulative number of cycles x, d avg is the arithmetic mean of all disturbance treatment durations, and α represents d x The weight of the proportion, β represents d avg The proportion weight of .

4. The construction engineering construction plan process level simulation method according to claim 1 is characterized in that: The establishment of the cycle process simulation algorithm in step S3 includes: S311, create a data set B, B = {b k( |k,j>0}; where b k( is a single data sample, m0 represents the e-th construction condition, represents the construction parameters under the construction condition m0, s k represents the kth process, r ( represents the jth cycle, t k( It represents the j-th cycle time of the k-th process; S312, for process s k , get the corresponding data in data set B Based on LSTM algorithm t k( Prediction model: Each process trains an independent prediction model; S313. Select a corresponding prediction model according to the process, input the current construction conditions, construction parameters, and number of cycles into the prediction model, and output the time prediction result t of the process under the number of cycles.

5. The construction engineering construction plan process level simulation method according to claim 1 is characterized in that: The simulation calculation process in step S4 includes: The duration of the key operation section is taken as the total duration of the operation area; let the operation area be u, and the key operation section of the operation area u be v; j represents the number of cycles of the operation section v, the maximum is J, and k represents the number of process items, the maximum is K; The total duration T5 of the work area u is expressed as: Where T 56 is the duration of the key operation section v of the operation area u, T 568 is the duration of the jth cycle of the key operation section v of the operation area u, T ；vk( is the duration of the i-th process in the j-th cycle of the key operation section v of the operation area u, d ；vk( is the duration of disturbance event handling predicted by the operation section disturbance event simulation algorithm, t ；vk( The time consumption of the current process is predicted by the cycle process simulation algorithm.

6. A construction project construction plan process level simulation system, characterized in that: include: Domain modeling module: for construction projects, build a general engineering construction process model; Data preparation module: Based on the construction process model, a general data structure for process-level simulation of the engineering construction plan is established, and the previous engineering construction data is filled in; Algorithm preparation module: establish the operation section disturbance event simulation algorithm and the cycle process simulation algorithm, and use the historical engineering construction data to train the simulation algorithm; Simulation calculation module: input the actual construction process data for simulation, and calculate the complete simulation data of the whole cycle of the engineering construction stage; Analysis and post-processing module: Use simulation data to perform post-processing operations such as horizontal comparison of work processes and disturbance event analysis, evaluate and analyze problems in the construction plan, and provide auxiliary decision-making for the adjustment of the construction plan.

7. The construction engineering construction plan process level simulation system according to claim 6 is characterized in that: The domain modeling module is divided into work area unit, work section unit, and cycle process unit, including: Operation area unit: The operation area is the first layer, which is defined as a construction area that is independent of each other in terms of time, space, and resource allocation. Each construction area operates in parallel without interfering with each other; Operation section unit: The operation section is a secondary layer, belonging to a certain operation area, and is defined as a construction area that is loosely coupled in time and space and independent of each other in resource allocation; each operation section has strong constraints on spatial distance and strong constraints on the order of time; Cyclic process unit: The cyclic process is the bottom layer and belongs to a certain operation section. It is defined as a construction operation with strong coupling in time, space and resource allocation. The cyclic processes in the same operation section are constructed in the same space, have a sequence in time, and reuse resources.

8. The construction engineering construction plan process level simulation system according to claim 6 is characterized in that: The algorithm preparation module has an operation section disturbance event simulation algorithm submodule, including: Create a dataset C unit: Create a dataset C, C = {c i |i∈[1,a max ]}; where a max Indicates the maximum value of the cumulative number of cycles when a disturbance event occurs; i indicates the cumulative number of cycles; c i is a single data sample, c i =(i,n i ,d i ,e i ), n i is the cumulative number of disturbances, d i is the duration of disturbance event handling, e i Indicates whether a disturbance event occurs; Construct the disturbance model unit: Use the logistic regression algorithm to transform the perturbation model in the dataset C (i, e i ) as a data set to input the logistic regression algorithm to build a disturbance event prediction model for the operation section; Judgment unit: predict the probability p1 of a disturbance event occurring when the current cumulative number of cycles x is reached by using the disturbance event prediction model of the operation section; then compare it with the model judgment threshold of the current cumulative number of cycles x to determine whether a disturbance event occurs; the model judgment threshold is: where n x is the cumulative number of disturbances corresponding to the current cumulative number of cycles x, and D is defined as the sum of the cumulative number of disturbances; Prediction duration unit: Prediction of disturbance event handling duration d: d = α × d x +β×d av$ ; where d x is the disturbance treatment duration corresponding to the current cumulative number of cycles x, d avg is the arithmetic mean of all disturbance treatment durations, and α represents d x The weight of the proportion, β represents d avg The proportion weight of .

9. The construction engineering construction plan process level simulation system according to claim 6, characterized in that: The algorithm preparation module has a cycle process simulation algorithm submodule, including: Dataset B unit: Create dataset B, B = {b k( |k,j>0}; where b k( is a single data sample, m0 represents the e-th construction condition, represents the construction parameters under the construction condition m0, s k represents the kth process, r ( represents the jth cycle, t k( It represents the j-th cycle time of the k-th process; Process model building unit: for process s k , get the corresponding data in data set B Based on LSTM algorithm t k( Prediction model: Each process trains an independent prediction model; Time consumption prediction unit: select the corresponding prediction model according to the process, input the current construction conditions, construction parameters, and number of cycles into the prediction model, and output the time consumption prediction result t of the process under the said number of cycles.

10. The construction engineering construction plan process level simulation system according to claim 6, characterized in that: The simulation calculation modules include: The duration of the key operation section is taken as the total duration of the operation area; let the operation area be u, and the key operation section of the operation area u be v; j represents the number of cycles of the operation section v, the maximum is J, and k represents the number of process items, the maximum is K; The total duration T5 of the work area u is expressed as: Where T 56 is the duration of the key operation section v of the operation area u, T 568 is the duration of the jth cycle of the key operation section v of the operation area u, T ；vk( is the duration of the i-th process in the j-th cycle of the key operation section v of the operation area u, d ；vk( is the duration of disturbance event handling predicted by the operation section disturbance event simulation algorithm, t ；vk( The time consumption of the current process is predicted by the cycle process simulation algorithm.

Images