BIM (Building Information Modeling) digital collaborative management method for construction progress control

Abstract

The invention discloses a BIM digital collaborative management method for construction progress control, and the method comprises the steps: collecting a plurality of construction-related drawings and construction progress plans, and constructing a BIM model through a BIM software platform; sorting the plurality of construction procedures based on a genetic algorithm and a BIM model to obtain an optimal construction procedure sorting table; construction is conducted according to the optimal construction procedure sorting table, after the current construction procedure is finished, a plurality of sensors are used for collecting construction site data, a three-dimensional model is constructed, and a construction progress model is obtained; performing progress analysis on the construction progress model and the BIM model to obtain a progress deviation value; judging whether the construction progress is delayed or not according to the progress deviation value; if yes, obtaining a progress delay reason by using a Bayesian network model; if not, finishing the construction progress analysis. The invention relates to the technical field of construction progress control, and solves the technical problems of unreasonable procedure arrangement and inaccurate progress delay analysis in an existing method.

Description

Technical Field

[0001] The present invention belongs to the field of construction progress control and relates to digital collaborative management technology, in particular to a BIM digital collaborative management method for construction progress control. Background Art

[0002] In the field of construction engineering, construction progress control is crucial. The construction progress is directly related to the project's delivery time, cost-effectiveness, and the satisfaction of all stakeholders. A reasonable construction schedule can ensure that the project is completed on time and avoid huge fines, contract disputes, and damage to corporate reputation caused by delays. At the same time, through effective progress control, resource allocation can be optimized, resource utilization can be improved, unnecessary waste can be reduced, and project costs can be reduced. Efficient collaborative management can ensure that information is transmitted between all parties in a timely and accurate manner during construction progress control, avoiding duplication of work, misunderstandings, and delays caused by poor information flow. For example, the construction unit can obtain change notices from the design unit in a timely manner and adjust the construction plan; suppliers can supply materials on time according to the construction progress to avoid shutdowns caused by material shortages; and the supervision unit can monitor the construction quality and progress in real time and discover and solve problems in a timely manner.

[0003] However, there are some problems in the existing construction progress control field. In terms of construction process planning, existing technologies often lack systematicity and scientificity. Traditional methods mostly rely on experience or simple templates, and fail to fully consider the logical relationship between processes and the rational allocation of resources. In large-scale construction projects, the relationships between different professional processes are intricate, but these relationships are often ignored, resulting in chaotic processes and unbalanced resource allocation, which in turn affects construction efficiency and progress. At the same time, when analyzing the causes of delays, existing technologies often only attribute them to manpower or material problems, while ignoring factors such as equipment, weather, and technology and their interactions. This analysis method cannot draw on historical experience and it is difficult to judge the specific impact of each factor, resulting in a lack of pertinence and effectiveness in response measures, causing delays to recur. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a BIM digital collaborative management method for construction progress control, which is used to solve the problems of unreasonable process planning and inaccurate analysis of causes of progress delays in existing construction progress control methods.

[0005] To achieve the above object, the present invention provides a BIM digital collaborative management method for construction progress control, comprising:

[0006] S1, collect several construction-related drawings and construction schedules, and use the BIM software platform to build a BIM model to obtain a BIM model;

[0007] S2, sorting a number of construction processes based on the genetic algorithm and the BIM model to obtain an optimal construction process sorting table;

[0008] S3, performing construction according to the optimal construction process ranking table, and after the current construction process is completed, using a number of sensors to collect construction site data and construct a three-dimensional model to obtain a construction progress model;

[0009] S4, performing progress analysis on the construction progress model and the BIM model to obtain a progress deviation value;

[0010] S5, judging whether the construction progress is delayed according to the progress deviation value; if yes, obtaining the cause of the progress delay by using the Bayesian network model; if no, completing the construction progress analysis.

[0011] Based on the above technical process, the present invention realizes a BIM-based digital construction collaborative management method. Among them, by using genetic algorithms combined with BIM models to sort the construction processes, the optimal construction process sorting table can be obtained, which helps to improve construction efficiency; during the construction process, with the help of sensors to collect data, build a progress model, and compare it with the BIM model to obtain the progress deviation value, the actual progress can be grasped in real time and intuitively, and progress differences can be discovered in time; when it is judged that the construction progress is delayed, the Bayesian network model can be used to accurately analyze and obtain the reasons for the progress delay. The Bayesian network model can accurately analyze the reasons for the delay, provide a basis for subsequent adjustments and improvements to the construction arrangements, and ensure that the project is carried out as scheduled.

[0012] Furthermore, the sequencing of several construction processes based on the genetic algorithm and the BIM model includes:

[0013] S21, integer encoding a number of construction processes to obtain a number of process numbers, and obtaining a basic logical relationship of the construction processes from the BIM model, and obtaining N process sorting schemes according to the basic logical relationship;

[0014] S22, for the process sequencing scheme j, using the BIM model to obtain the estimated duration and basic logical relationship of several construction processes, and obtain the total construction period Tj;

[0015] S23, according to the formula The construction resource balance index B is calculated; where R i represents the total amount of resources required for the i-th construction process, and n represents the number of construction processes;

[0016] S24, constructing a fitness function based on the construction resource balance index B and the total construction period Tj Get several fitness values; among them, F jrepresents the fitness value under process sorting scheme j, ω1 and ω2 represent dynamic weight coefficients, and Δω represents the weight adjustment change in each iteration;

[0017] S25, performing genetic operations according to a number of fitness values ​​and N process sorting schemes to obtain a sorting set, and performing local search optimization on the sorting set based on the BIM model to obtain a current optimal process sorting sequence set;

[0018] S26, starting from the current optimal process sorting solution set, repeating S2-2 to S2-5 until the fitness value no longer increases after a preset number of consecutive iterations, and obtaining the optimal process set;

[0019] S27, comparing the fitness values ​​of several sorting sequences in the optimal process set, marking the sorting sequence with the largest fitness value as the optimal process sorting sequence, and obtaining the optimal construction process sorting table.

[0020] The construction process involves many interrelated factors. It is difficult to accurately determine the optimal construction process sequence by relying solely on experience or simple methods. In order to more comprehensively and accurately find a solution that not only conforms to the logical order of construction but also achieves the optimal balance between construction period and resource utilization, a genetic algorithm combined with construction resources is used to construct a fitness function combined with resource balance, and efficiently screen and optimize among many possible process sequence solutions.

[0021] Furthermore, the genetic operation is performed according to a number of fitness values ​​and N process sorting schemes, including:

[0022] S25-11, obtaining a first sorted set using a roulette wheel selection method according to a number of fitness values;

[0023] S25-12, setting the intersection position k, combining the sorted sequences in the first sorted set in pairs, and in each combination, splitting the sorted sequences at k and then exchanging them, and adjusting the sorted sequences that do not satisfy the basic logical relationship after the exchange, to obtain a second sorted set after single-point crossover;

[0024] S25-13, performing a mutation operation on the second sorted set with a preset mutation probability Pm, and adjusting the sorted sequences that do not satisfy the basic logical relationship after the mutation operation to obtain a sorted set.

[0025] In the problem of construction process sequencing, faced with many possible combinations of solutions, it is necessary to use the selection, crossover and mutation operations in genetic algorithms to continuously explore better sequencing solutions. At the same time, since the construction process has its own inherent logical constraints, it is necessary to adjust the illogical situations to ensure that the generated solution meets the actual construction requirements.

[0026] Furthermore, the performing local search optimization on the sorting set based on the BIM model includes:

[0027] S25-21, for each sequence I in the sorted set, traverse a number of adjacent process pairs (A, B) in sequence I;

[0028] S25-22, obtaining the spatial position coordinates of process A and process B through the BIM model, and judging whether process A and process B are adjacent in space according to the spatial position coordinates; if yes, jump to S25-23; if not, traverse the next adjacent process pair in the sequence and jump to S25-22;

[0029] S25-23, count the number of resources of several types required by process A and process B, and form resource vectors Ra and Rb, and judge whether process A and process B have complementary resources according to Ra and Rb; if yes, jump to S25-24; if no, traverse the next adjacent process pair in the sequence and jump to S25-22;

[0030] S25-24, swapping the positions of process A and process B in sequence I to obtain a new sequence I';

[0031] S25-25, calculate and obtain the fitness values ​​FI and FI' of I and I', and determine whether FI is greater than FI'; if yes, update I with I'; if no, keep I unchanged;

[0032] S25-26, repeat S25-21 to S25-26 until several sequences in the sorted set are traversed to obtain the current optimal process sorted sequence set.

[0033] Based on the BIM model, local search and optimization of the sorting set from the two key perspectives of spatial location and resource complementarity can fully tap the improvement potential of the sorting scheme in the local range, further improve the rationality of process sorting and resource utilization efficiency, make the final optimal process sorting scheme more in line with the actual construction situation, and reduce unnecessary resource allocation and process connection costs.

[0034] Further, judging whether process A and process B are spatially adjacent according to the spatial position coordinates includes:

[0035] A1, according to the spatial position coordinates, use the Euclidean distance formula to calculate and obtain the spatial distance between process A and process B;

[0036] A2, judging whether the spatial distance is greater than a preset spatial distance threshold; if yes, process A and process B are adjacent in space; if no, process A and process B are far away in space.

[0037] Furthermore, judging whether the resources of process A and process B are complementary according to Ra and Rb includes:

[0038] B1, calculate and obtain the difference vector of Ra and Rb according to the vector difference formula;

[0039] B2, determine whether the difference vector is greater than a preset resource complementation threshold vector; if yes, the resources of process A and process B are complementary; if no, the resources of process A and process B are in conflict.

[0040] Furthermore, the method of collecting construction site data using a plurality of sensors and constructing a three-dimensional model includes:

[0041] S31, using a number of sensors to collect and obtain three-dimensional point cloud data of the construction site;

[0042] S32, preprocessing the three-dimensional point cloud data of the construction site to obtain preprocessed three-dimensional point cloud data; wherein the preprocessing includes data cleaning, data calibration, and data format conversion;

[0043] S33, constructing the pre-processed three-dimensional point cloud data into a three-dimensional model using point cloud processing software to obtain a construction progress model.

[0044] Furthermore, the progress analysis of the construction progress model and the BIM model includes:

[0045] S4-1, coordinate-aligning the construction progress model with the BIM model, and obtaining characteristic vectors Fpl of several components in the aligned construction progress model, and obtaining characteristic vectors Fbj of several components corresponding to the aligned BIM model according to the construction time; wherein l and j represent components;

[0046] S4-2, calculate the distance dlj between Fpl and Fbj according to the Euclidean distance formula, and calculate the similarity between l and j according to the formula Slj=1 / (1+dlj);

[0047] S4-3, comparing whether the similarity Slj is greater than a preset similarity threshold; if yes, a number of successfully matched component pairs (a, b) are obtained; if no, a set of unconstructed components C is obtained; wherein a represents the components in the aligned construction progress model, and b represents the components in the aligned BIM model;

[0048] S4-4, obtaining a complete component b' corresponding to b from the BIM model, and calculating the ratio of a to b' to obtain a set A of construction schedules of several components a, and calculating the ratio of b to b' to obtain a set B of planned schedules of several components b;

[0049] S4-5, calculate the progress deviation value ΔP according to the formula ΔP=∑Aa-Bb / (Nab+Nc); wherein Nab represents the number of successfully matched component pairs, and Nc represents the number of unconstructed components in set C.

[0050] Through steps such as coordinate alignment, eigenvector calculation, similarity comparison and progress quantification, the deviation between the construction progress and the planned progress can be analyzed scientifically and accurately, avoiding the arbitrariness and uncertainty of subjective judgment, providing precise and objective data support for construction progress control, and facilitating timely and targeted adjustment measures.

[0051] Further, judging whether the construction progress is delayed according to the progress deviation value includes:

[0052] Compare whether the progress deviation value ΔP is greater than or equal to 0; if yes, the construction is normal; if not, the construction is delayed.

[0053] Furthermore, the construction of the Bayesian network model includes:

[0054] S51-1, set several delay indicators: material factor Cm, human factor Cw, equipment factor Ce, weather factor Cn;

[0055] S51-2, set several evaluation indicators: daily material quantity difference Dm(t), equipment quantity difference De(t), worker quantity difference Dw(t), supervisor quantity S(t), material price Pm(t), temperature T(t), humidity H(t), precipitation R(t);

[0056] S51-3, defining the several evaluation indicators and the several delay indicators as a node set V = {Dm(t), De(t), Dw(t), S(t), Pm(t), T(t), H(t), Cm, Cw, Ce, Cn}, defining the causal relationship between the nodes as an edge set E, and obtaining a Bayesian network G = (V, E); wherein the causal relationship between the nodes is determined based on engineering domain knowledge and historical experience, and obtaining several edge connection rules;

[0057] S51-4, according to the evaluation indicators and the delay indicators, collect daily data of N historical construction processes to obtain a data set D = {(D i m(t),D i e(t),D i w(t),S i (t),P i m(t),T i (t),H i (t),C i m,C i w,C i e,C i n)|i=1,2,...,N}; where i represents the construction process;

[0058] S51-5, calculate the conditional probability distribution estimate of several nodes X∈V according to the formula P(X=xj|π(X)=πi)=n(xj,πi) / n(πi), complete the Bayesian network parameter learning, and obtain the Bayesian network model; wherein π(X) represents the parent node set of node X, n(xj,πi) represents the number of samples in the data set D where the node X takes the value xj and the corresponding π(X) takes the value πi, and n(πi) represents the number of samples in the data set D where π(X) takes the value πi.

[0059] The constructed Bayesian network model comprehensively considers various influencing factors and their interrelationships, and learns based on historical data, making its analysis of the causes of construction progress delays more scientific, objective and accurate. In addition, the Bayesian network model can explore the potential probabilistic relationship between various factors and progress delays in complex construction scenarios. It can not only analyze the impact of a single factor, but also consider the interaction between multiple factors, which helps to accurately locate the cause of the delay.

[0060] Furthermore, the use of the Bayesian network model to obtain the cause of the progress delay includes:

[0061] S52-1, obtain several daily evaluation indicators of the current construction process, and obtain the evidence variable node set E = {(D * m(t),D * e(t),D * w(t),S * (t),P * m(t),T * (t),H * (t)}, and set the query variable node set to Q = {Cm, Cw, Ce, Cn}, and the hidden variable set to H = VQE;

[0062] S52-2, using the probability inference formula of the Bayesian network model P(Q|E=e)=∑ h∈H P(Q,h,E=e) / ∑ q∈Q ∑ h∈H P(q,h,E=e) calculates the probability of several delay indicators P(Q|E=e);

[0063] S52-3, among the probabilities of several delay indicators P(Q|E=e), the query variables that are greater than a preset threshold are marked as delay reasons, and several progress delay reasons are obtained.

[0064] By combining the constructed model with the current actual situation, the key causes of delays are screened out through probability comparison to meet the actual needs of accurately analyzing the causes of delays in complex construction environments.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] Genetic algorithms combined with BIM models are used to sort construction processes, and the optimal solution is determined by comprehensively considering factors such as logical relationships, construction period and resource balance. The BIM model is also used to optimize the sorting set from the perspective of space and resource complementarity, tap local potential, improve sorting rationality and efficiency, and make the solution more suitable for actual construction.

[0067] With the help of sensors to collect construction site data, a 3D construction progress model is constructed after pre-processing such as data cleaning, calibration, and format conversion. It is then compared with the BIM model. Through steps such as coordinate alignment, feature vector calculation, similarity comparison, and progress quantification, the deviation between the construction progress and the planned progress can be accurately analyzed. The actual progress can be intuitively grasped in real time, and progress differences can be discovered in a timely manner.

[0068] A Bayesian network model was constructed to integrate various influencing factors and learn based on historical data, to explore the probabilistic relationship between various factors and delays, and to accurately locate the problem by considering the interaction of multiple factors. Combined with the actual situation of the current construction stage, the key causes of delays were screened out to provide a reliable basis for subsequent construction arrangement adjustments and ensure the smooth progress of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0070] Figure 1 A flow chart of the BIM digital collaborative management method for construction progress control provided by the present invention;

[0071] Figure 2 A schematic diagram of the flow chart of the local search optimization method provided by the present invention;

[0072] Figure 3 A schematic diagram of the process flow of the progress analysis provided by the present invention. DETAILED DESCRIPTION

[0073] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] See also Figure 1-Figure 3The embodiment of the present invention provides a BIM digital collaborative management method for construction progress control, including:

[0075] S1, collect several construction-related drawings and construction schedules, and use the BIM software platform to build a BIM model to obtain a BIM model;

[0076] S2, sorting several construction processes based on genetic algorithm and BIM model to obtain the optimal construction process sorting table;

[0077] S3, performing construction according to the optimal construction process ranking table, and after the current construction process is completed, using a number of sensors to collect construction site data and build a three-dimensional model to obtain a construction progress model;

[0078] S4, performing progress analysis on the construction progress model and the BIM model to obtain a progress deviation value;

[0079] S5, judging whether the construction progress is delayed according to the progress deviation value; if yes, using the Bayesian network model to obtain the cause of the progress delay; if no, completing the construction progress analysis.

[0080] In the field of modern construction, the smooth progress of construction projects involves the efficient and coordinated operation of a large amount of human, material and financial resources. Accurate construction sequencing can enable resources to be invested in the corresponding processes at the right time, reducing idleness and waste.

[0081] Therefore, in this embodiment, in order to achieve refined management of the construction process, in steps S1-S2, it is necessary to collect a number of construction-related drawings and construction schedules, and use the BIM software platform to build a BIM model to obtain a BIM model; then, based on the genetic algorithm and the BIM model, a number of construction processes are sorted to obtain an optimal construction process sorting table;

[0082] Specifically, firstly, several construction processes are integer-coded to obtain several process numbers, and the basic logical relationship of the construction processes is obtained from the BIM model, and N process sorting schemes are obtained based on the basic logical relationship;

[0083] Then, for each process sequencing scheme j, the BIM model is used to obtain the estimated duration and logical relationship of the construction process, so as to calculate the earliest start time, earliest completion time, latest start time and latest completion time under the sequencing scheme, and further determine the critical path, thereby obtaining the corresponding total construction period Tj; since the order of processes in different sequencing schemes is different, and the connection and waiting time between processes are also different, the total construction period T will change with the change of the sequencing scheme.

[0084] Then, according to the formula The construction resource balance index B is calculated and then further calculated according to the formula The fitness value of each sorting scheme is calculated; where R i represents the total amount of resources required for the i-th construction process, n represents the number of construction processes, and F j represents the fitness value under process sorting scheme j, ω1 and ω2 represent dynamic weight coefficients, and Δω represents the weight adjustment change in each iteration; the fitness value here reflects the pros and cons of each sorting scheme in terms of comprehensive consideration of construction period and resource balance;

[0085] Next, genetic operations are performed based on several fitness values ​​and N process sorting schemes to obtain the sorted set:

[0086] S25-11, using the roulette wheel selection method to obtain the first sorted set:

[0087] The fitness value of each process sorting scheme is calculated based on the constructed fitness function. Then, the formula Pj=Fj / ∑ N j＝1 Fj is used to determine the probability of each individual being selected;

[0088] Use a random number generator to generate a random number between 0 and 1; starting from the first sorting scheme, accumulate the probability of being selected in order, and when the accumulated sum is greater than or equal to N, the corresponding individual is selected to enter the first sorting set; repeat this process until N sorting schemes are selected to obtain the first sorting set; since the selection probability of each scheme is different, there will be repeated schemes in the N sorting schemes, thereby filtering out the sorting schemes with lower fitness;

[0089] S25-12, setting the intersection position k, combining the sorted sequences in the first sorted set in pairs, and in each combination, splitting the sorted sequences at k and then exchanging them, and adjusting the sorted sequences that do not satisfy the basic logical relationship after the exchange, to obtain a second sorted set after single-point crossover;

[0090] S25-13, performing a mutation operation on the second sorted set with a preset mutation probability Pm, and adjusting the sorted sequences that do not satisfy the basic logical relationship after the mutation operation, to obtain a sorted set;

[0091] Then, based on the BIM model, the sorting set is locally searched and optimized to obtain the current optimal process sorting sequence set;

[0092] Next, starting from the current optimal process sorting solution set, the above steps are repeated until the fitness value no longer increases after a preset number of consecutive iterations, and the optimal process set is obtained;

[0093] Finally, the fitness values ​​of several sorting sequences in the optimal process set are compared, and the sorting sequence with the largest fitness value is marked as the optimal process sorting sequence to obtain the optimal construction process sorting table.

[0094] In the construction process sequencing, in addition to macro factors such as the overall construction period and resource balance, micro factors such as the spatial layout between processes and resource complementarity also have an important impact on construction efficiency and resource utilization. In order to achieve a better state in the details of the sequencing plan, it is necessary to use the space and resource information provided by the BIM model to perform such local search optimization, so as to discover and utilize potential optimization points between processes and improve the overall quality of the plan;

[0095] Therefore, in this embodiment, local search optimization is performed on the sorting set based on the BIM model, including the following steps:

[0096] S25-21, for each sequence I in the sorted set, traverse a number of adjacent process pairs (A, B) in sequence I;

[0097] S25-22, obtaining the spatial position coordinates of process A and process B through the BIM model, and calculating and obtaining the spatial distance between process A and process B using the Euclidean distance formula;

[0098] Determine whether the spatial distance is greater than a preset spatial distance threshold; if yes, process A and process B are spatially adjacent, and jump to S25-23; if no, process A and process B are spatially distant, continue to traverse the next adjacent process pair in the sequence, and jump to S25-22;

[0099] S25-23, counting the number of resources of several types required by process A and process B, and forming resource vectors Ra and Rb, and calculating and obtaining the difference vector of Ra and Rb according to the vector difference formula;

[0100] Then, it is determined whether the difference vector is greater than the preset resource complementation threshold vector; if so, the resources of process A and process B are complementary, and the process jumps to S25-24; if not, the resources of process A and process B conflict, and the process continues to traverse the next adjacent process pair in the sequence and jumps to S25-22;

[0101] S25-24, swapping the positions of process A and process B in sequence I to obtain a new sequence I';

[0102] S25-25, calculate and obtain the fitness values ​​FI and FI' of I and I', and determine whether FI is greater than FI'; if yes, update I with I'; if no, keep I unchanged;

[0103] S25-26, repeat S25-21 to S25-26 until several sequences in the sorted set are traversed to obtain the current optimal process sorted sequence set.

[0104] The rational allocation and utilization of resources is one of the key factors to be considered in the sequencing of construction processes. Judging whether resources are complementary between processes is crucial to optimizing the sequence of processes and improving resource utilization efficiency. By combining the vector difference formula with the preset threshold vector, the relatively abstract concept of resource complementarity can be transformed into a quantifiable and comparable judgment standard, thus providing a scientific basis for the optimization of construction process sequencing based on resource factors.

[0105] After obtaining the optimal construction process ranking table, in step S3, construction is carried out according to the ranking, and after the current construction process is completed, the three-dimensional point cloud data of the construction site is collected by sensors, and then preprocessed, such as data cleaning, data calibration, data format conversion, etc., and then the preprocessed three-dimensional point cloud data is constructed into a three-dimensional model using point cloud processing software to obtain a construction progress model.

[0106] S4, then the construction progress model and the BIM model are analyzed to obtain a progress deviation value;

[0107] Specifically, first, align the coordinates of the construction progress model and the BIM model to ensure that the two can accurately correspond in spatial position. Then, for the aligned construction progress model, obtain the characteristic vectors Fpl of several components in it in sequence according to certain rules and order (here l is used to identify different components). At the same time, according to the corresponding construction time, obtain the characteristic vectors Fbj (j is also used to distinguish different components) of the same corresponding components from the aligned BIM model; the characteristic vector can reflect the relevant characteristics of the component, which is helpful for comparison in similarity and other aspects;

[0108] Next, the Euclidean distance formula is used to calculate the distance dlj between each group of corresponding Fpl and Fbj. On this basis, the similarity between l and j is further calculated according to the formula Slj=1 / (1+dlj). The difference in features between components is converted into an intuitive similarity value, which is convenient for subsequent matching judgment. The higher the similarity, the closer the corresponding components are in all aspects.

[0109] For each group of similarities Slj calculated, compare it with the preset similarity threshold. If the similarity Slj is greater than the preset similarity threshold, then it can be determined that the corresponding components l and j are successfully matched, and these successfully matched components are marked as a (the component from the aligned construction progress model) and b (the component from the aligned BIM model), and several successfully matched component pairs (a, b) are summarized;

[0110] On the contrary, if the similarity Slj is less than or equal to the preset similarity threshold, it is determined that these components are not matched successfully, and these unmatched components are summarized to obtain the unconstructed component set C. These components may not have been started for construction or have not been reflected in the matching relationship between the construction schedule model and the BIM model as planned due to some reasons;

[0111] Then, find and obtain the complete component b' corresponding to b from the BIM model, where b' is the complete component corresponding to b. Then, by calculating the ratio of a to b', measure the completion of each component a in terms of actual construction progress relative to the complete plan, and obtain a set A of construction progress of several components a;

[0112] Similarly, the ratio of b to b' is calculated to reflect the planned progress of component b, and a set B of planned progress of several components b is obtained;

[0113] Finally, the progress deviation value ΔP is calculated according to the formula ΔP = ∑(Aa-Bb) / (Nab+Nc). In this formula, Nab represents the number of successfully matched component pairs, and Nc represents the number of unconstructed components in set C, accurately obtaining the overall progress deviation value ΔP.

[0114] In step S5, it is compared whether the progress deviation value ΔP is greater than or equal to 0; if yes, it indicates that the construction is normal; if no, it indicates that the construction is delayed, and then the Bayesian network model is used to analyze the cause of the progress delay;

[0115] Specifically, we first build a Bayesian network model, including:

[0116] S51-1, based on the actual project situation and past experience, set a number of delay indicators that may affect the construction progress, including material factors Cm, manpower factors Cw, equipment factors Ce, weather factors Cn, etc.;

[0117] S51-2, set a number of evaluation indicators, including: daily material quantity difference Dm(t), equipment quantity difference De(t), worker quantity difference Dw(t), supervisor quantity S(t), material price Pm(t), temperature T(t), humidity H(t), precipitation R(t);

[0118] S51-3, define several evaluation indicators and several delay indicators as a node set V = {Dm(t), De(t), Dw(t), S(t), Pm(t), T(t), H(t), Cm, Cw, Ce, Cn}, define the causal relationship between nodes as an edge set E, and obtain the Bayesian network G = (V, E); wherein the causal relationship between nodes is determined based on engineering domain knowledge and historical experience, and the connection rules of several edges are obtained, which clarifies the logical relationship of mutual influence and correlation between various factors, so that the Bayesian network can accurately simulate the complex action mechanism between factors related to the construction progress;

[0119] S51-4, based on a number of evaluation indicators and a number of delay indicators, collect daily data of N historical construction processes to obtain a data set D = {(D i m(t),D i e(t),D i w(t),S i (t),P i m(t),T i (t),H i (t),C i m,C i w,C i e,C i n)|i=1,2,...,N}; where i represents the construction process;

[0120] S51-5, calculate the conditional probability distribution estimate of several nodes X∈V according to the formula P(X=xj|π(X)=πi)=n(xj,πi) / n(πi), complete the parameter learning of the Bayesian network, and obtain the Bayesian network model; wherein π(X) represents the parent node set of node X, n(xj,πi) represents the number of samples in the data set D where the node X is xj and the corresponding π(X) is πi, and n(πi) represents the number of samples in the data set D where π(X) is πi; assign a reasonable probability distribution to each node in the Bayesian network based on historical data, so that it can perform effective reasoning analysis based on new data input, and provide a strong model support for finding the cause of the progress delay;

[0121] Then, the Bayesian network model is used to obtain the causes of schedule delays, including:

[0122] S52-1, obtain several daily evaluation indicators of the current construction process, and obtain the evidence variable node set E = {(D * m(t),D * e(t),D * w(t),S * (t),P * m(t),T * (t),H* (t)}, and set the query variable node set to Q = {Cm, Cw, Ce, Cn}, and the hidden variable set to H = VQE;

[0123] S52-2, using the probability inference formula of the Bayesian network model P(Q|E=e)=∑ h∈H P(Q,h,E=e) / ∑ q∈Q ∑ h∈H P(q,h,E=e) is used to calculate the probability P(Q|E=e) of several delay indicators. Based on the probability theory of Bayesian networks, by comprehensively considering all possible values ​​of hidden variables, the probability of querying variables under the condition of given evidence variables is calculated, thereby obtaining the possibility of delays caused by problems in various delay indicators under the current actual construction situation, and converting abstract causal relationships into quantifiable probability values, which is convenient for further analysis and judgment.

[0124] S52-3, mark the query variables that are greater than the preset threshold value among the probabilities of several delay indicators P(Q|E=e) as delay reasons, and obtain several reasons for progress delay, providing accurate and valuable basis for subsequent targeted improvement measures and adjustment of construction plans.

[0125] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0126] Working principle of the present invention:

[0127] First, drawings and plans are collected to build a BIM model. Then, the construction processes are sorted based on the genetic algorithm and the BIM model. Construction is carried out according to the optimal sorting, and after the current process is completed, the three-dimensional point cloud data of the construction site is collected by sensors to build a construction progress model. Then, the construction progress model is aligned with the BIM model coordinates, and the component feature vectors are obtained to calculate the similarity. The similarity is compared to determine the matching component pairs and the set of unconstructed components to calculate the progress deviation value. Finally, whether the construction is delayed is determined based on the progress deviation, and the specific reasons for the delay are analyzed using the Bayesian network model.

[0128] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A BIM digital collaborative management method for construction progress control, characterized in that: include: S1, collect several construction-related drawings and construction schedules, and use the BIM software platform to build a BIM model to obtain a BIM model; S2, sorting several construction processes based on genetic algorithm and BIM model to obtain the optimal construction process sorting table; S3, performing construction according to the optimal construction process ranking table, and after the current construction process is completed, using a number of sensors to collect construction site data and construct a three-dimensional model to obtain a construction progress model; S4, performing progress analysis on the construction progress model and the BIM model to obtain a progress deviation value; S5, judging whether the construction progress is delayed according to the progress deviation value; if yes, obtaining the cause of the progress delay by using the Bayesian network model; if no, completing the construction progress analysis.

2. The BIM digital collaborative management method for construction progress control according to claim 1 is characterized in that: The sequencing of several construction processes based on the genetic algorithm and the BIM model includes: S21, integer encoding a number of construction processes to obtain a number of process numbers, and obtaining a basic logical relationship of the construction processes from the BIM model, and obtaining N process sorting schemes according to the basic logical relationship; S22, for the process sequencing scheme j, the estimated duration and basic logical relationship of several construction processes are obtained using the BIM model to obtain the total construction period Tj; S23, according to the formula The construction resource balance index B is calculated; where R i represents the total amount of resources required for the i-th construction process, and n represents the number of construction processes; S24, constructing a fitness function based on the construction resource balance index B and the total construction period Tj Get several fitness values; among them, F j represents the fitness value under process sorting scheme j, ω1 and ω2 represent dynamic weight coefficients, and Δω represents the change in weight adjustment in each iteration; S25, performing genetic operations according to a number of fitness values ​​and N process sorting schemes to obtain a sorting set, and performing local search optimization on the sorting set based on the BIM model to obtain a current optimal process sorting sequence set; S26, starting from the current optimal process sorting solution set, repeating S2-2 to S2-5 until the fitness value no longer increases after a preset number of consecutive iterations, and obtaining the optimal process set; S27, comparing the fitness values ​​of several sorting sequences in the optimal process set, marking the sorting sequence with the largest fitness value as the optimal process sorting sequence, and obtaining the optimal construction process sorting table.

3. The BIM digital collaborative management method for construction progress control according to claim 2 is characterized in that: The genetic operation is performed according to a number of fitness values ​​and N process sorting schemes, including: S25-11, obtaining a first sorted set using a roulette wheel selection method according to a number of fitness values; S25-12, setting the intersection position k, combining the sorted sequences in the first sorted set in pairs, and in each combination, splitting the sorted sequences at k and then exchanging them, and adjusting the sorted sequences that do not satisfy the basic logical relationship after the exchange, to obtain a second sorted set after single-point crossover; S25-13, performing a mutation operation on the second sorted set with a preset mutation probability Pm, and adjusting the sorted sequences that do not satisfy the basic logical relationship after the mutation operation to obtain a sorted set.

4. The BIM digital collaborative management method for construction progress control according to claim 2 is characterized in that: The local search optimization of the sorting set based on the BIM model includes: S25-21, for each sequence I in the sorted set, traverse a number of adjacent process pairs (A, B) in sequence I; S25-22, obtaining the spatial position coordinates of process A and process B through the BIM model, and judging whether process A and process B are adjacent in space according to the spatial position coordinates; if yes, jump to S25-23; if not, traverse the next adjacent process pair in the sequence and jump to S25-22; S25-23, count the number of resources of several types required by process A and process B, and form resource vectors Ra and Rb, and judge whether process A and process B have complementary resources according to Ra and Rb; if yes, jump to S25-24; if no, traverse the next adjacent process pair in the sequence and jump to S25-22; S25-24, swapping the positions of process A and process B in sequence I to obtain a new sequence I'; S25-25, calculate and obtain the fitness values ​​FI and FI' of I and I', and determine whether FI is greater than FI'; if yes, update I with I'; if no, keep I unchanged; S25-26, repeat S25-21 to S25-26 until several sequences in the sorted set are traversed to obtain the current optimal process sorted sequence set.

5. The BIM digital collaborative management method for construction progress control according to claim 4 is characterized in that: The step of judging whether process A and process B are spatially adjacent to each other according to the spatial position coordinates includes: A1, according to the spatial position coordinates, use the Euclidean distance formula to calculate and obtain the spatial distance between process A and process B; A2, judging whether the spatial distance is greater than a preset spatial distance threshold; if yes, process A and process B are adjacent in space; if no, process A and process B are far away in space.

6. The BIM digital collaborative management method for construction progress control according to claim 4 is characterized in that: The determining whether process A and process B have complementary resources according to Ra and Rb includes: B1, calculate and obtain the difference vector of Ra and Rb according to the vector difference formula; B2, determine whether the difference vector is greater than a preset resource complementation threshold vector; if yes, the resources of process A and process B are complementary; if no, the resources of process A and process B are in conflict.

7. The BIM digital collaborative management method for construction progress control according to claim 1 is characterized in that: The method of collecting construction site data using a number of sensors and constructing a three-dimensional model includes: S31, using a number of sensors to collect and obtain three-dimensional point cloud data of the construction site; S32, preprocessing the three-dimensional point cloud data of the construction site to obtain preprocessed three-dimensional point cloud data; wherein the preprocessing includes data cleaning, data calibration, and data format conversion; S33, constructing the pre-processed three-dimensional point cloud data into a three-dimensional model using point cloud processing software to obtain a construction progress model.

8. The BIM digital collaborative management method for construction progress control according to claim 1 is characterized in that: The progress analysis of the construction progress model and the BIM model includes: S4-1, coordinate-aligning the construction progress model with the BIM model, and obtaining characteristic vectors Fpl of several components in the aligned construction progress model, and obtaining characteristic vectors Fbj of several components corresponding to the aligned BIM model according to the construction time; wherein l and j represent components; S4-2, calculate the distance dlj between Fpl and Fbj according to the Euclidean distance formula, and calculate the similarity between l and j according to the formula Slj=1 / (1+dlj); S4-3, comparing whether the similarity Slj is greater than a preset similarity threshold; if yes, a number of successfully matched component pairs (a, b) are obtained; if no, a set of unconstructed components C is obtained; wherein a represents the components in the aligned construction progress model, and b represents the components in the aligned BIM model; S4-4, obtaining a complete component b' corresponding to b from the BIM model, and calculating the ratio of a to b' to obtain a set A of construction schedules of several components a, and calculating the ratio of b to b' to obtain a set B of planned schedules of several components b; S4-5, calculate the progress deviation value ΔP according to the formula ΔP=∑Aa-Bb / (Nab+Nc); wherein Nab represents the number of successfully matched component pairs, and Nc represents the number of unconstructed components in set C.

9. The BIM digital collaborative management method for construction progress control according to claim 1 is characterized in that: The construction of the Bayesian network model includes: S51-1, set several delay indicators: material factor Cm, human factor Cw, equipment factor Ce, weather factor Cn; S51-2, set several evaluation indicators: material quantity difference Dm(t), equipment quantity difference De(t), worker quantity difference Dw(t), supervisor quantity S(t), material price Pm(t), temperature T(t), humidity H(t), precipitation R(t); where t represents time; S51-3, defining the several evaluation indicators and the several delay indicators as a node set V = {Dm(t), De(t), Dw(t), S(t), Pm(t), T(t), H(t), Cm, Cw, Ce, Cn}, defining the causal relationship between the nodes as an edge set E, and obtaining a Bayesian network G = (V, E); S51-4, collecting daily data of N historical construction processes according to the evaluation indicators and the delay indicators, and obtaining a data set D = {(D i m(t),D i e(t),D i w(t),S i (t),P i m(t),T i (t),H i (t),C i m,C i w,C i e,C i n)|i=1,2,...,N}; where i represents the construction process; S51-5, calculate the conditional probability distribution estimate of several nodes X∈V according to the formula P(X=xj|π(X)=πi)=n(xj,πi) / n(πi), complete the Bayesian network parameter learning, and obtain the Bayesian network model; wherein π(X) represents the parent node set of node X, n(xj,πi) represents the number of samples in the data set D where the node X takes the value xj and the corresponding π(X) takes the value πi, and n(πi) represents the number of samples in the data set D where π(X) takes the value πi.

10. The BIM digital collaborative management method for construction progress control according to claim 1, characterized in that: The reasons for the progress delay obtained by using the Bayesian network model include: S52-1, obtain several daily evaluation indicators of the current construction process, and obtain the evidence variable node set E = {(D * m(t),D * e(t),D * w(t),S * (t),P * m(t),T * (t),H * (t)}, and set the query variable node set to Q = {Cm, Cw, Ce, Cn}, and the hidden variable set to H = VQE; S52-2, using the probability inference formula of the Bayesian network model P(Q|E=e)=∑ h∈H P(Q,h,E=e) / ∑ q∈Q ∑ h∈H P(q,h,E=e) calculates the probability of several delay indicators P(Q|E=e); S52-3, among the probabilities of several delay indicators P(Q|E=e), the query variables that are greater than a preset threshold are marked as delay reasons, and several progress delay reasons are obtained.

Images