CN117648596B - Digital twin and intelligent sensor fusion method and system for building construction - Google Patents
Digital twin and intelligent sensor fusion method and system for building construction Download PDFInfo
- Publication number
- CN117648596B CN117648596B CN202311596887.0A CN202311596887A CN117648596B CN 117648596 B CN117648596 B CN 117648596B CN 202311596887 A CN202311596887 A CN 202311596887A CN 117648596 B CN117648596 B CN 117648596B
- Authority
- CN
- China
- Prior art keywords
- scaffold
- node
- digital twin
- data
- strain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000009435 building construction Methods 0.000 title claims abstract description 15
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 9
- 230000001133 acceleration Effects 0.000 claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000004422 calculation algorithm Methods 0.000 claims description 20
- 238000013145 classification model Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 15
- 239000000463 material Substances 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 5
- 239000013598 vector Substances 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 238000010276 construction Methods 0.000 description 9
- 238000013527 convolutional neural network Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000012937 correction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 239000004035 construction material Substances 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides a digital twin and intelligent sensor fusion method and system for building construction, and relates to the technical field of data monitoring, wherein the method comprises the steps of determining scaffold information based on a preset coordinate system and a photographed picture of a scaffold; determining the lateral stability of the scaffold, the longitudinal stability of the scaffold and the stability of the support points of the scaffold based on the scaffold information; judging whether the scaffold needs to be comprehensively checked by using a judging model based on the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of supporting points of the scaffold, the stability of the scaffold at a plurality of nodes and the firmness of the scaffold; determining the firmness of the scaffold by using an acceleration digital twin sensor model based on the data of the acceleration digital twin sensor of the scaffold within a preset duration; the method can accurately determine the safety of the scaffold by using the judging model to judge whether the scaffold needs to be comprehensively checked.
Description
Technical Field
The invention relates to the technical field of data monitoring, in particular to a digital twin and intelligent sensor fusion method and system for building construction.
Background
Digital twinning is a multi-physical quantity, multi-scale and multi-probability simulation process, mapping is completed in a virtual space, and accordingly the full life cycle process of corresponding entity equipment is reflected. Digital twinning is a beyond-the-reality concept that can be seen as a digital mapping system of one or more important, mutually dependent equipment systems; with the continuous development of modern engineering construction, the scaffold is used as important equipment in construction, and the safety and stability of the scaffold have a crucial influence on the safety and quality of the whole engineering. However, the conventional scaffold management method often depends on manual inspection and experience judgment, lacks scientificity and accuracy, and cannot discover and process potential safety hazards in time.
How to accurately determine the safety of scaffolds is a current problem to be solved.
Disclosure of Invention
The technical problem to be solved by the invention is how to accurately determine the safety of the scaffold.
In a first aspect, the present invention provides a digital twin and intelligent sensor fusion method for building construction, including: acquiring a shot picture of a scaffold, wherein the scaffold is in data of strain digital twin sensors of a plurality of nodes within a preset time period, and the scaffold is in data of acceleration digital twin sensors within the preset time period; determining scaffold information based on a preset coordinate system and a photographed picture of a scaffold, wherein the scaffold information comprises position coordinates of each main rod of the scaffold, position coordinates of each cross rod of the scaffold, position coordinates of each longitudinal rod of the scaffold, position coordinates of each node, material information of the scaffold, supporting point information, flatness of the scaffold and degree of freshness of scaffold components; determining the lateral stability of the scaffold, the longitudinal stability of the scaffold and the stability of the support points of the scaffold based on the scaffold information; clustering the data sets of the variable digital twin sensor data by using a K-Means algorithm based on the data sets of the strain digital twin sensor data of the plurality of nodes of the scaffold within a preset duration to obtain a plurality of clusters and cluster centers of each cluster, and determining the stability of the scaffold at the plurality of nodes respectively according to the initial node stability of the data of the strain digital twin sensor of each node, the similarity of the data of the strain digital twin sensor of each node and the cluster centers of the cluster to which the data of the strain digital twin sensor of each node belongs;
The initial node stability According to the formulaObtaining;
Wherein, Data x of strain digital twin sensor for node to be detected belongs to state categoryIs the probability of the state categoryComprises a normal state and an abnormal state, wherein the states are classified intoThe method comprises the steps of inputting data x of strain digital twin sensors of nodes into an initial classification model and outputting the data x, wherein the initial classification model is obtained by inputting data of strain digital twin sensors, marked with state types, of all nodes of a scaffold into an initial training network for training;
The similarity I (x) is according to the formula: Obtaining;
Wherein Z is the number of clusters after the data set of the strain digital twin sensor data is clustered, Represents the firstThe strain digital twin sensor data of the cluster center of each cluster, wherein x is the data of the strain digital twin sensor of the node to be detected;
The said difference According to the formulaObtaining;
Wherein, Data of a strain digital twin sensor which is the mth node,The data set of the strain digital twin sensor data is the data of the strain digital twin sensor of the j-th node except the data of the strain digital twin sensor of the m-th node, U is the data set of the strain digital twin sensor data, n is the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data,Standard deviation of the data of the strain digital twin sensor for each of the nodes in the data set of strain digital twin sensor data;
Determining the firmness of the scaffold by using an acceleration digital twin sensor model based on the data of the acceleration digital twin sensor of the scaffold within a preset duration; and judging whether the scaffold needs to be comprehensively checked or not by using a judging model based on the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of supporting points of the scaffold, the stability of the scaffold at a plurality of nodes and the firmness degree of the scaffold.
Further, clustering the data of the strain digital twin sensors based on the plurality of nodes of the scaffold within a preset time period by using a K-Means algorithm, and obtaining a cluster center of each of the plurality of clusters comprises:
Preprocessing data of a plurality of nodes of the scaffold in preset time to obtain an average value of each node, a standard deviation of each node, a maximum value of each node, a minimum value of each node, a peak value of each node and a valley value of each node;
And clustering by using a K-Means algorithm based on the average value of each node, the standard deviation of each node, the maximum value of each node, the minimum value of each node, the peak value of each node and the valley value of each node to obtain a plurality of clusters and a cluster center of each cluster in the plurality of clusters.
Further, preprocessing the data of the strain digital twin sensors of the plurality of nodes of the scaffold within the preset duration to obtain an average value of each node, a standard deviation of each node, a maximum value of each node, a minimum value of each node, a peak value of each node and a valley value of each node, and further comprising: and filtering the strain digital twin sensor data of each node to remove noise data.
Still further, the method provided by the application further comprises the following steps: acquiring the distance between the data of the strain digital twin sensor of each node in the two clusters and the clustering center of each cluster;
And calculating Euclidean distance between the feature vector of the strain digital twin sensor data of each node and the clustering center of each cluster, and obtaining the similarity between the data of the strain digital twin sensor of each node and the clustering center of each cluster.
According to a second aspect, the present invention provides a digital twin and intelligent sensor fusion system for building construction, comprising: the first acquisition module is used for acquiring the photographed pictures of the scaffold, the data of the strain digital twin sensors of the plurality of nodes of the scaffold in the preset time length, and the data of the acceleration digital twin sensors of the scaffold in the preset time length;
the scaffold information determining module is used for determining scaffold information based on a preset coordinate system and a photographed picture of the scaffold, wherein the scaffold information comprises position coordinates of each main rod of the scaffold, position coordinates of each cross rod of the scaffold, position coordinates of each longitudinal rod of the scaffold, position coordinates of each node, material information of the scaffold, supporting point information, flatness of the scaffold and the degree of freshness of scaffold components;
The reasonable degree determining module is used for determining the lateral stability of the scaffold, the longitudinal stability of the scaffold and the stability of the supporting point of the scaffold based on the scaffold information;
The stability determining module is used for clustering the data sets of the variable digital twin sensor data based on the data sets of the strain digital twin sensor data of the plurality of nodes of the scaffold within the preset duration by using a K-Means algorithm to obtain a plurality of clusters and cluster centers of each cluster, and determining the stability of the scaffold at the plurality of nodes respectively according to the initial node stability of the data of the strain digital twin sensor of each node, the similarity between the data of the strain digital twin sensor of each node and the cluster center of the cluster to which the data of the strain digital twin sensor of each node belongs;
The initial node stability According to the formulaObtaining;
Wherein, Data x of strain digital twin sensor for node to be detected belongs to state categoryIs the probability of the state categoryComprises a normal state and an abnormal state, wherein the states are classified intoThe method comprises the steps of inputting data x of strain digital twin sensors of nodes into an initial classification model and outputting the data x, wherein the initial classification model is obtained by inputting data of strain digital twin sensors, marked with state types, of all nodes of a scaffold into an initial training network for training; wherein, the initial training network can adopt LSTM, CNN, SVM models, etc., which are not limited in this document;
The similarity I (x) is according to the formula: Obtaining;
Wherein Z is the number of clusters after the data set of the strain digital twin sensor data is clustered, Represents the firstThe strain digital twin sensor data of the cluster center of each cluster, wherein x is the data of the strain digital twin sensor of the node to be detected;
The said difference According to the formulaObtaining;
Wherein, Data of a strain digital twin sensor which is the mth node,The data set of the strain digital twin sensor data is the data of the strain digital twin sensor of the j-th node except the data of the strain digital twin sensor of the m-th node, U is the data set of the strain digital twin sensor data, n is the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data,Standard deviation of the data of the strain digital twin sensor for each of the nodes in the data set of strain digital twin sensor data;
The firmness determining module is used for determining the firmness of the scaffold by using the acceleration digital twin sensor model based on the data of the acceleration digital twin sensor of the scaffold within the preset time period;
And the inspection module is used for judging whether the scaffold needs to be comprehensively inspected by using a judgment model based on the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of the support points of the scaffold, the stability of the scaffold at a plurality of nodes and the firmness degree of the scaffold.
Further, the photographed picture of the scaffold is obtained through unmanned aerial vehicle or manual photographing.
Still further, the stability determination module is further configured to:
Preprocessing data of a plurality of nodes of the scaffold in preset time to obtain an average value of each node, a standard deviation of each node, a maximum value of each node, a minimum value of each node, a peak value of each node and a valley value of each node;
And clustering by using a K-Means algorithm based on the average value of each node, the standard deviation of each node, the maximum value of each node, the minimum value of each node, the peak value of each node and the valley value of each node to obtain a plurality of clusters and a cluster center of each cluster in the plurality of clusters.
Still further, the stability determination module is further configured to: and filtering the strain digital twin sensor data of each node to remove noise data.
Further, the stability determining module is specifically configured to calculate euclidean distances between feature vectors of the strain digital twin sensor data of each node and a cluster center of each cluster, so as to obtain similarity between the data of the strain digital twin sensor of each node and the cluster center of each cluster.
The invention provides a digital twin and intelligent sensor fusion method and system for building construction, wherein the method comprises the steps of obtaining photographed pictures of a scaffold, and obtaining data of strain digital twin sensors of a plurality of nodes of the scaffold within a preset duration, and obtaining data of acceleration digital twin sensors of the scaffold within the preset duration; determining scaffold information based on a preset coordinate system and a photographed picture of a scaffold, wherein the scaffold information comprises position coordinates of each main rod of the scaffold, position coordinates of each cross rod of the scaffold, position coordinates of each longitudinal rod of the scaffold, position coordinates of each node, material information of the scaffold, supporting point information, flatness of the scaffold and degree of freshness of scaffold components; determining the lateral stability of the scaffold, the longitudinal stability of the scaffold and the stability of the support points of the scaffold based on the scaffold information; clustering the data sets of the variable digital twin sensor data by using a K-Means algorithm based on the data sets of the strain digital twin sensor data of the plurality of nodes of the scaffold within a preset duration to obtain a plurality of clusters and cluster centers of each cluster, and determining the stability of the scaffold at the plurality of nodes respectively according to the initial node stability of the data of the strain digital twin sensor of each node, the similarity of the data of the strain digital twin sensor of each node and the cluster centers of the cluster to which the data of the strain digital twin sensor of each node belongs; the initial node stabilityAccording to the formulaObtaining; wherein,Data x of strain digital twin sensor for node to be detected belongs to state categoryIs the probability of the state categoryComprises a normal state and an abnormal state, wherein the states are classified intoThe method comprises the steps of inputting data x of strain digital twin sensors of nodes into an initial classification model and outputting the data x, wherein the initial classification model is obtained by inputting data of strain digital twin sensors, marked with state types, of all nodes of a scaffold into an initial training network for training; the similarity I (x) is according to the formula: obtaining; wherein Z is the number of clusters after the data set of the strain digital twin sensor data is clustered, Represents the firstThe strain digital twin sensor data of the cluster center of each cluster, wherein x is the data of the strain digital twin sensor of the node to be detected; the said differenceAccording to the formulaObtaining; wherein,Data of a strain digital twin sensor which is the mth node,The data set of the strain digital twin sensor data is the data of the strain digital twin sensor of the j-th node except the data of the strain digital twin sensor of the m-th node, U is the data set of the strain digital twin sensor data, n is the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data,Standard deviation of the data of the strain digital twin sensor for each of the nodes in the data set of strain digital twin sensor data; determining the firmness of the scaffold by using an acceleration digital twin sensor model based on the data of the acceleration digital twin sensor of the scaffold within a preset duration; based on the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of supporting points of the scaffold, the stability of the scaffold at a plurality of nodes respectively and the firmness of the scaffold, the method can accurately determine the safety of the scaffold by judging whether the scaffold needs to be comprehensively checked by using a judging model.
Drawings
FIG. 1 is a schematic flow chart of a digital twin and intelligent sensor fusion method for building construction provided by an embodiment of the invention;
fig. 2 is a schematic diagram of a digital twin and intelligent sensor fusion system for building construction according to an embodiment of the present invention.
Detailed Description
In the embodiment of the invention, a digital twin and intelligent sensor fusion method for building construction is provided as shown in fig. 1, and a scaffold monitoring method for the digital twin sensor for building construction comprises the following steps of S1-S7:
Step S1, obtaining photographed pictures of a scaffold, and obtaining data of strain digital twin sensors of a plurality of nodes of the scaffold within a preset time period, and obtaining data of acceleration digital twin sensors of the scaffold within the preset time period.
The shooting picture of the scaffold is obtained through unmanned aerial vehicle or manual shooting. The scaffold is a temporary structure for supporting and fixing construction materials in building construction, and generally consists of a main rod, a cross rod, a longitudinal rod and nodes. In this embodiment, 3D point cloud data of the scaffold is obtained by photographing the scaffold with a binocular camera, a three-dimensional scanner, an RGB-D camera, etc., a standard engineering drawing is generated from the 3D point cloud data with engineering drawing software such as CAD (e.g., a three-dimensional modeling function under a switched working space menu), and then an explanation of scaffold diagrams and related setting parameters (e.g., parameters such as a supporting point position, a supporting manner of a supporting point, a supporting point type, a supporting area, etc.) is generated with automatic calculation software such as a widemontage.
The strain digital twin sensor is used for measuring the strain (namely deformation) of an object and is arranged on a node of the scaffold. The preset time period may be 1 hour, 12 hours, 24 hours, or the like, and is not limited herein.
The acceleration digital twin sensor is used for measuring the acceleration of an object and is arranged at a key position of the scaffold.
The strain digital twin sensor acquires data of an actual strain digital twin sensor and then maps the data in a virtual space to obtain a plurality of strain twin data, so that a set of the data of the strain digital twin sensor is constructed. Furthermore, the subsequent clustering of the data set corresponding to the digital twin sensor data by using the K-Means algorithm can be realized, and the reliability is higher.
The acceleration digital twin sensor acquires data of an actual acceleration digital twin sensor and then maps the data in a virtual space to obtain a plurality of acceleration twin data, so that a set of data of the acceleration digital twin sensor is constructed. Furthermore, the reliability of determining the firmness of the scaffold by using the acceleration digital twin sensor model can be made to be higher based on the data set of the acceleration digital twin sensor of the scaffold within the preset time period.
And S2, determining scaffold information based on a preset coordinate system and a photographed picture of the scaffold, wherein the scaffold information comprises the position coordinate of each main rod of the scaffold, the position coordinate of each cross rod of the scaffold, the position coordinate of each longitudinal rod of the scaffold, the position coordinate of each node, the material information of the scaffold, the supporting point information, the flatness of the scaffold and the degree of freshness of scaffold components.
Specifically, a trained scaffold classification model (such as a Multi-TASK LEARNING Multi-task/Multi-target classification neural network model) is input based on a preset coordinate system and 3D point cloud data of a photographed picture of a scaffold, the neural network is provided with a plurality of output nodes, each node corresponds to one classification task), and the position coordinates of each main rod, the position coordinates of each cross rod, the position coordinates of each longitudinal rod, the position coordinates of each node, material information, supporting point information, flatness of the scaffold and the degree of new and old components of the scaffold are output.
The preset coordinate system is a reference system for describing the position and direction of the object, and is generally composed of three mutually perpendicular xyz axes.
The node is the intersection of the main rod, the cross rod and the longitudinal rod in the scaffold, and the node is used for connecting and fixing different parts.
The material information comprises the material, diameter, wall thickness and the like of the steel pipe and the fastener used by the scaffold.
The supporting point information is the point that the scaffold contacts the ground. The supporting point information comprises supporting point positions, supporting modes, supporting point types, supporting areas and the like.
The support point position refers to a specific position on the ground or structure supporting the scaffold.
The supporting mode of the supporting point comprises an embedded type supporting mode, a desk type supporting mode, a hook type supporting mode and the like. Different supporting modes are suitable for different types of scaffolds and different working environments.
The type of the supporting point may be classified into two types of fixed supporting points and adjustable supporting points. The fixed support points are generally used on the ground or concrete structures, while the adjustable support points can be adjusted in height as required.
The support area refers to the surface area of the support point in contact with the ground or structure. The larger the supporting area is, the stronger the bearing capacity of the scaffold is, and the stability is also better.
The flatness of the scaffold is the degree to which the scaffold surface is flat.
The degree of freshness of the scaffold members has a great influence on the performance of the scaffold members, and the scaffold with shorter service time has stronger stability than the scaffold with longer service time.
In some embodiments, the scaffold information can be obtained by processing a preset coordinate system and a photographed picture of the scaffold through a scaffold processing model. The scaffold treatment model is a convolutional neural network model. The convolutional neural network (Convolutional Neural Network, CNN) is a deep learning algorithm. Consists of an input layer, a convolution layer, a pooling layer and a full connection layer. The scaffold treatment model can be trained by a gradient descent method to obtain a trained scaffold treatment model. The scaffold processing model can identify the position coordinates of the main rod, the cross rod, the longitudinal rod and the node of the scaffold, and the attributes of material information, supporting point information, flatness, degree of old and new and the like.
And acquiring detailed information of the scaffold through the scaffold processing model so as to analyze the state and safety of the scaffold later.
Step one, constructing a digital twin model of the scaffold, constructing a model of a basic unit of the scaffold by using a pointer to the scaffold object, and constructing the digital twin model from two aspects of multi-field model construction and 'geometric-scaffold-behavior-rule' multi-dimensional model construction. The 'geometry-scaffold-behavior-rule' model can be used for describing the geometrical characteristics, scaffold characteristics, behavior coupling relation, evolution rules and the like of scaffold objects; the multi-domain model is used for comprehensively describing the thermal and mechanical characteristics of the scaffold object by respectively constructing the domain models related to the scaffold object. Through multi-dimensional model construction and multi-domain model construction, accurate construction of a digital twin model is realized;
Step two, assembling a scaffold digital twin model, namely firstly, constructing a hierarchical relation of the model and defining an assembling sequence of the model so as to avoid the situation of difficult assembly; secondly, proper space constraint conditions are required to be added in the assembly process, the models of different levels are required to pay attention to and have certain difference with the added space constraint relationship, for example, constraint relationships such as angle constraint, contact constraint, offset constraint and the like between the models and the added scaffold are required to be constructed, the model assembly process from equipment to production line to workshop is realized, and the model assembly is realized based on the constructed constraint relationship and the model assembly sequence;
Thirdly, fusing the digital twin model of the scaffold, wherein coupling relations exist among different systems, so that the construction of the digital twin model of the scaffold is realized, and the external wall, the electricity, the liquid and other multi-field models are fused;
and step four, verifying the digital twin model of the scaffold, wherein the method specifically comprises the steps of verifying characteristics of the scaffold and information such as correction objects, correction parameters, correction results and the like, and is beneficial to the application of the digital twin model of the scaffold to construction of different building groups. Further, the photographed picture of the scaffold is obtained through unmanned aerial vehicle or manual photographing.
And step S3, determining the lateral stability of the scaffold, the longitudinal stability of the scaffold and the stability of the supporting points of the scaffold based on the scaffold information.
The lateral stability of the scaffold refers to the stability of the scaffold in the horizontal direction, reflecting the ability of the scaffold to resist horizontal loads.
The longitudinal stability of the scaffold is the stability of the scaffold in the vertical direction, reflecting the capacity of the scaffold to resist vertical loads.
The stability of the support points of the scaffold indicates the firmness and stability of the support points of the scaffold, and the stability of the whole scaffold is directly affected.
In some embodiments, the scaffold information may be processed by a stability processing model to obtain the lateral stability of the scaffold, the longitudinal stability of the scaffold, and the support point stability of the scaffold. The stability processing model is a deep neural network model. Deep neural networks include convolutional neural networks (Convolutional Neural Network, CNN), recurrent neural networks (Recurrent Neural Network, RNN), and transducers, among others. The stability processing model is based on the scaffold information obtained in the step S2, and the transverse stability, the longitudinal stability and the supporting point stability of the scaffold can be obtained after comprehensive evaluation by analyzing the position coordinates and the connection relation of the main rod, the cross rod, the longitudinal rod and the nodes, the number and the position of the supporting points and other information.
And S4, clustering the data sets of the digital twin sensor data by using a K-Means algorithm based on the data sets of the strain digital twin sensor data of the plurality of nodes of the scaffold within the preset duration to obtain a plurality of clusters and a cluster center of each cluster in the plurality of clusters.
The K-Means algorithm is a cluster analysis algorithm that iteratively divides data into K clusters such that the sum of the distances of each data point from the cluster center of the cluster to which it belongs is minimized.
In some embodiments, clustering using a K-Means algorithm based on data of strain digital twin sensors of a plurality of nodes of the scaffold over a preset period of time, the obtaining a cluster center of the plurality of clusters and each of the plurality of clusters comprises:
Preprocessing data of a plurality of nodes of the scaffold in preset time to obtain an average value of each node, a standard deviation of each node, a maximum value of each node, a minimum value of each node, a peak value of each node and a valley value of each node;
In some embodiments, the strain digital twin sensor data for each node may be filtered to remove noise data and determine an average value for each node, a standard deviation for each node, a maximum value for each node, a minimum value for each node, a peak value for each node, and a valley value for each node.
And clustering by using a K-Means algorithm based on the average value of each node, the standard deviation of each node, the maximum value of each node, the minimum value of each node, the peak value of each node and the valley value of each node to obtain a plurality of clusters and a cluster center of each cluster in the plurality of clusters.
And S5, carrying out weighted summation on the data of the strain digital twin sensors of each node and the differences of the data of the strain digital twin sensors of each node and the data of each strain digital twin sensor in the cluster according to the initial node stability of the data of the strain digital twin sensors of each node and the similarity of the data of the strain digital twin sensors of each node and the cluster center of the cluster to which the data of the strain digital twin sensors of each node belong, and determining the stability of the scaffold in a plurality of nodes respectively.
Wherein, initial node stabilityAccording to the formulaThe product is obtained by, among other things,Data x of strain digital twin sensor for node to be detected belongs to state categoryProbability of (2), state classIncluding normal and abnormal states, state categoriesThe data x of the strain digital twin sensor of the node is input into an initial classification model and then output. The initial classification model is obtained by inputting data of strain digital twin sensors marked with state types of all nodes of the scaffold into an initial training network for training. It can be appreciated that the reliability of the initial node stability obtained in the above manner is high.
Similarity degreeObtaining, wherein Z is the number of clusters after the data set of the strain digital twin sensor data is clustered,Represents the firstAnd the data of the strain digital twin sensors of the cluster centers of the individual clusters are obtained, and x is the data of the strain digital twin sensors of the nodes to be detected. It can be appreciated that the reliability of the similarity obtained by the above-described manner is high.
In addition, the similarity of the data of the strain digital twin sensors of the nodes in the same layer (i.e. the same height) of the scaffold from top to bottom is high, that is, the data of the strain digital twin sensors on the same layer with the same height are clustered into a cluster. And if the similarity between the data of the strain digital twin sensor of the node to be detected and the corresponding clustering center is low, the data of the strain digital twin sensor of the node to be detected is abnormal. Further, the stability of the scaffold may be characterized by the similarity of the data of the strain digital twin sensors of the nodes in the layers.
Differentiation ofAccording to the formulaThe product is obtained by, among other things,Data of a strain digital twin sensor which is the mth node,The data set of the strain digital twin sensor data is the data of the strain digital twin sensor of the j-th node except the data of the strain digital twin sensor of the m-th node, U is the data set of the strain digital twin sensor data, n is the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data,Is the standard deviation of the data of the strain digital twin sensor for each node in the data set of the strain digital twin sensor data. It can be appreciated that the variability obtained by the above manner is highly reliable.
In addition, the data of the strain digital twin sensors of the nodes in each layer of the scaffold from top to bottom are small in variability (i.e., high in similarity). If the data of the strain digital twin sensor of the node to be detected has large difference with the data of the strain digital twin sensors of other nodes to be detected in the same cluster, the data of the strain digital twin sensor of the node to be detected is abnormal. Further, the stability of the scaffold may be characterized by the variability of the data of the strain digital twin sensors of the nodes in each layer.
In summary, it can be understood that, according to the initial node stability of the data of the strain digital twin sensor of each node and the similarity between the data of the strain digital twin sensor of each node and the clustering center of the cluster to which the strain digital twin sensor belongs, the differences between the data of the strain digital twin sensor of each node and the data of each strain digital twin sensor in the cluster to which the strain digital twin sensor belongs are weighted and summed, so that the accuracy of determining the stability of the scaffold at a plurality of nodes is high.
And S6, determining the firmness of the scaffold by using an acceleration digital twin sensor model based on the data of the acceleration digital twin sensor of the scaffold in a preset time period.
The acceleration digital twin sensor model is a transducer model. The transducer model can process sequence data. The input of the transducer model is the data of the acceleration digital twin sensor of the scaffold in a preset time period, and the output of the transducer model is the firmness of the scaffold. The transducer model consists of an encoder (Encoder) and a Decoder (Decoder), each part being stacked of multiple layers. The encoder is responsible for encoding the input sequence information, and the decoder is responsible for generating a target sequence according to the output of the encoder. The acceleration digital twin sensor model can obtain acceleration and vibration conditions of the scaffold in different directions according to data acquired by the acceleration digital twin sensor, so that the firmness of the scaffold is evaluated.
And S7, judging whether the scaffold needs to be comprehensively checked by using a judging model based on the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of supporting points of the scaffold, the distance between the data of the strain digital twin sensor of each node in the two clusters and the clustering center of each cluster and the firmness degree of the scaffold.
The judgment model is a deep neural network model. The judgment model can comprehensively consider the transverse stability of the scaffold, the longitudinal stability of the scaffold, the stability of the support points of the scaffold, the stability of the scaffold at a plurality of nodes respectively and the firmness degree of the scaffold, and then judge whether the scaffold needs to be comprehensively checked. The input of the judgment model is the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of the support points of the scaffold, the stability of the scaffold at a plurality of nodes respectively and the firmness degree of the scaffold, and the output of the judgment model is the condition that the comprehensive inspection is needed or not needed.
Based on the same inventive concept, fig. 2 is a schematic diagram of a digital twin and intelligent sensor fusion system for building construction according to an embodiment of the present invention, where the system includes:
A first obtaining module 21, configured to obtain a photographed picture of the scaffold, data of strain digital twin sensors of a plurality of nodes of the scaffold within a preset duration, and data of acceleration digital twin sensors of the scaffold within the preset duration;
A scaffold information determining module 22, configured to determine scaffold information based on a preset coordinate system and a photographed picture of a scaffold, where the scaffold information includes a position coordinate of each main rod of the scaffold, a position coordinate of each crossbar of the scaffold, a position coordinate of each longitudinal rod of the scaffold, a position coordinate of each node, material information of the scaffold, supporting point information, flatness of the scaffold, and a degree of freshness of scaffold members;
A reasonable degree determining module 23, configured to determine lateral stability of the scaffold, longitudinal stability of the scaffold, and stability of a supporting point of the scaffold based on scaffold information;
the stability determining module 24 is configured to perform clustering on the data set of the variable digital twin sensor data by using a K-Means algorithm based on the data set of the strain digital twin sensor data of the plurality of nodes of the scaffold within the preset duration, obtain a plurality of clusters and cluster centers of each of the plurality of clusters, and determine stability of the scaffold at the plurality of nodes respectively according to initial node stability of the data of the strain digital twin sensor of each node, similarity of the strain digital twin sensor data of each node and the cluster center of the cluster to which the strain digital twin sensor data of each node belongs; wherein, initial node stability According to the formulaThe product is obtained by, among other things,Data x of strain digital twin sensor for node to be detected belongs to state categoryProbability of (2), state classIncluding normal and abnormal states, state categoriesThe method comprises the steps that data x of strain digital twin sensors of nodes are input into an initial classification model and then output, wherein the initial classification model is obtained by inputting data of strain digital twin sensors, marked with state types, of all nodes of a scaffold into an initial training network for training; similarity degreeObtaining, wherein Z is the number of clusters after the data set of the strain digital twin sensor data is clustered,Represents the firstThe strain digital twin sensor data of the cluster center of each cluster, wherein x is the data of the strain digital twin sensor of the node to be detected; differentiation ofAccording to the formulaThe product is obtained by, among other things,Data of a strain digital twin sensor which is the mth node,The data set of the strain digital twin sensor data is the data of the strain digital twin sensor of the j-th node except the data of the strain digital twin sensor of the m-th node, U is the data set of the strain digital twin sensor data, n is the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data,Standard deviation of the data of the strain digital twin sensor for each node in the data set of the strain digital twin sensor data;
a firmness determining module 25, configured to determine a firmness of the scaffold using an acceleration digital twin sensor model based on data of the acceleration digital twin sensor of the scaffold within a preset time period;
and the checking module 26 is used for judging whether the scaffold needs to be comprehensively checked by using a judging model based on the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of the supporting points of the scaffold, the stability of the scaffold at a plurality of nodes and the firmness degree of the scaffold.
Further, the photographed picture of the scaffold is obtained through unmanned aerial vehicle or manual photographing.
Still further, the stability determination module is further configured to:
Preprocessing data of a plurality of nodes of the scaffold in preset time to obtain an average value of each node, a standard deviation of each node, a maximum value of each node, a minimum value of each node, a peak value of each node and a valley value of each node;
And clustering by using a K-Means algorithm based on the average value of each node, the standard deviation of each node, the maximum value of each node, the minimum value of each node, the peak value of each node and the valley value of each node to obtain a plurality of clusters and a cluster center of each cluster in the plurality of clusters.
Still further, the stability determination module is further configured to: and filtering the strain digital twin sensor data of each node to remove noise data.
Further, the stability determining module is specifically configured to calculate euclidean distances between feature vectors of the strain digital twin sensor data of each node and a cluster center of each cluster, so as to obtain similarity between the data of the strain digital twin sensor of each node and the cluster center of each cluster.
Claims (10)
1. The digital twin and intelligent sensor fusion method for building construction is characterized by comprising the following steps:
Acquiring a shot picture of a scaffold, wherein the scaffold is in data of strain digital twin sensors of a plurality of nodes within a preset time period, and the scaffold is in data of acceleration digital twin sensors within the preset time period;
Determining scaffold information based on a preset coordinate system and a shot picture of the scaffold, wherein the scaffold information comprises position coordinates of each main rod of the scaffold, position coordinates of each cross rod of the scaffold, position coordinates of each longitudinal rod of the scaffold, position coordinates of each node, material information of the scaffold, supporting point information, flatness of the scaffold and degree of freshness of scaffold components;
Determining lateral stability of the scaffold, longitudinal stability of the scaffold and supporting point stability of the scaffold based on the scaffold information;
Clustering the data set of the strain digital twin sensor data by using a K-Means algorithm based on the data set of the strain digital twin sensor data of the scaffold in a preset time period to obtain a plurality of clusters and cluster centers of each cluster, and determining the stability of the scaffold in a plurality of nodes according to the initial node stability of the strain digital twin sensor data of each node, the similarity of the strain digital twin sensor data of each node and the cluster centers of the cluster to which the strain digital twin sensor data of each node belongs;
The initial node stability According to the formula/>Obtaining;
Wherein, Data x of strain digital twin sensor for node to be detected belongs to state category/>Is said state class/>Including normal and abnormal states, the state class/>The method comprises the steps of inputting data x of strain digital twin sensors of nodes into an initial classification model and outputting the data x, wherein the initial classification model is obtained by inputting data of strain digital twin sensors, marked with state types, of all nodes of a scaffold into an initial training network for training;
The similarity I (x) is according to the formula: Obtaining;
Wherein Z is the number of clusters after the data set of the strain digital twin sensor data is clustered, Represents the firstThe strain digital twin sensor data of the cluster center of each cluster, wherein x is the data of the strain digital twin sensor of the node to be detected;
The said difference According to the formula/>Obtaining;
Wherein, Data of strain digital twin sensor for mth node,/>The data set of the strain digital twin sensor data is the data of the strain digital twin sensor of the j-th node except the data of the strain digital twin sensor of the m-th node, U is the data set of the strain digital twin sensor data, n is the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data, and the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data is/>Standard deviation of the data of the strain digital twin sensor for each of the nodes in the data set of strain digital twin sensor data;
Determining the firmness of the scaffold by using an acceleration digital twin sensor model based on the data of the acceleration digital twin sensor of the scaffold in a preset duration;
And judging whether the scaffold needs to be comprehensively checked or not by using a judging model based on the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of the supporting points of the scaffold, the stability of the scaffold at a plurality of nodes and the firmness degree of the scaffold.
2. The method of claim 1, wherein the photographed picture of the scaffold is taken by an unmanned aerial vehicle or by hand.
3. The method of claim 1, wherein clustering using a K-Means algorithm based on data of the strain digital twin sensors of the plurality of nodes of the scaffold over a preset period of time, the obtaining a plurality of clusters and a cluster center of each of the plurality of clusters comprises:
Preprocessing data of the strain digital twin sensors of the plurality of nodes of the scaffold within a preset duration to obtain an average value of each node, a standard deviation of each node, a maximum value of each node, a minimum value of each node, a peak value of each node and a valley value of each node;
And clustering by using a K-Means algorithm based on the average value of each node, the standard deviation of each node, the maximum value of each node, the minimum value of each node, the peak value of each node and the valley value of each node to obtain a plurality of clusters and a cluster center of each cluster in the plurality of clusters.
4. The method of claim 3, wherein preprocessing the data of the strain digital twin sensors of the plurality of nodes of the scaffold within a preset time length to obtain an average value of each node, a standard deviation of each node, a maximum value of each node, a minimum value of each node, a peak value of each node, and a valley value of each node, further comprises: and filtering the strain digital twin sensor data of each node to remove noise data.
5. The method of claim 1, wherein the method further comprises:
acquiring the distance between the data of the strain digital twin sensor of each node in the two clusters and the clustering center of each cluster;
And calculating Euclidean distance between the feature vector of the strain digital twin sensor data of each node and the clustering center of each cluster, and obtaining the similarity between the data of the strain digital twin sensor of each node and the clustering center of each cluster.
6. A digital twin and smart sensor fusion system for building construction, said system being based on the method of any one of claims 1-5, comprising:
The first acquisition module is used for acquiring the photographed pictures of the scaffold, the data of the strain digital twin sensors of the plurality of nodes of the scaffold in the preset time length, and the data of the acceleration digital twin sensors of the scaffold in the preset time length;
The scaffold information determining module is used for determining scaffold information based on a preset coordinate system and a shot picture of the scaffold, wherein the scaffold information comprises position coordinates of each main rod of the scaffold, position coordinates of each cross rod of the scaffold, position coordinates of each longitudinal rod of the scaffold, position coordinates of each node, material information of the scaffold, supporting point information, flatness of the scaffold and degree of freshness of scaffold components;
The reasonable degree determining module is used for determining the lateral stability of the scaffold, the longitudinal stability of the scaffold and the stability of the supporting point of the scaffold based on the scaffold information;
The stability determining module is used for clustering the data set of the strain digital twin sensor data of the plurality of nodes of the scaffold within a preset duration by using a K-Means algorithm to obtain a plurality of clusters and cluster centers of each cluster, and determining the stability of the scaffold at the plurality of nodes respectively according to the initial node stability of the strain digital twin sensor data of each node, the similarity of the strain digital twin sensor data of each node and the cluster center of the cluster to which the strain digital twin sensor data of each node belongs;
The initial node stability According to the formula/>Obtaining;
Wherein, Data x of strain digital twin sensor for node to be detected belongs to state category/>Is said state class/>Including normal and abnormal states, the state class/>The method comprises the steps of inputting data x of strain digital twin sensors of nodes into an initial classification model and outputting the data x, wherein the initial classification model is obtained by inputting data of strain digital twin sensors, marked with state types, of all nodes of a scaffold into an initial training network for training;
The similarity I (x) is according to the formula: Obtaining;
Wherein Z is the number of clusters after the data set of the strain digital twin sensor data is clustered, Represents the firstThe strain digital twin sensor data of the cluster center of each cluster, wherein x is the data of the strain digital twin sensor of the node to be detected;
The said difference According to the formula/>Obtaining;
Wherein, Data of strain digital twin sensor for mth node,/>The data set of the strain digital twin sensor data is the data of the strain digital twin sensor of the j-th node except the data of the strain digital twin sensor of the m-th node, U is the data set of the strain digital twin sensor data, n is the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data, and the number of the data of the strain digital twin sensor of each node in the data set of the strain digital twin sensor data is/>Standard deviation of the data of the strain digital twin sensor for each of the nodes in the data set of strain digital twin sensor data;
The firmness determining module is used for determining the firmness of the scaffold by using an acceleration digital twin sensor model based on the data of the acceleration digital twin sensor of the scaffold in a preset time period;
And the inspection module is used for judging whether the scaffold needs to be comprehensively inspected by using a judgment model based on the lateral stability of the scaffold, the longitudinal stability of the scaffold, the stability of the support points of the scaffold, the stability of the scaffold at a plurality of nodes respectively and the firmness degree of the scaffold.
7. The system of claim 6, wherein the photographed picture of the scaffold is taken by an unmanned aerial vehicle or by a human.
8. The system of claim 6, wherein the stability determination module is further to:
Preprocessing data of the strain digital twin sensors of the plurality of nodes of the scaffold within a preset duration to obtain an average value of each node, a standard deviation of each node, a maximum value of each node, a minimum value of each node, a peak value of each node and a valley value of each node;
And clustering by using a K-Means algorithm based on the average value of each node, the standard deviation of each node, the maximum value of each node, the minimum value of each node, the peak value of each node and the valley value of each node to obtain a plurality of clusters and a cluster center of each cluster in the plurality of clusters.
9. The system of claim 8, wherein the stability determination module is further to: and filtering the strain digital twin sensor data of each node to remove noise data.
10. The system of claim 6, wherein the stability determining module is specifically configured to calculate euclidean distances between feature vectors of the strain digital twin sensor data of each node and a cluster center of each cluster, so as to obtain similarity between the data of the strain digital twin sensor of each node and the cluster center of each cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311596887.0A CN117648596B (en) | 2023-11-28 | 2023-11-28 | Digital twin and intelligent sensor fusion method and system for building construction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311596887.0A CN117648596B (en) | 2023-11-28 | 2023-11-28 | Digital twin and intelligent sensor fusion method and system for building construction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117648596A CN117648596A (en) | 2024-03-05 |
| CN117648596B true CN117648596B (en) | 2024-04-30 |
Family
ID=90048872
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311596887.0A Active CN117648596B (en) | 2023-11-28 | 2023-11-28 | Digital twin and intelligent sensor fusion method and system for building construction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117648596B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112647590A (en) * | 2020-12-01 | 2021-04-13 | 河北建工集团有限责任公司 | Construction method for prefabricated assembled reinforced concrete beam column steel node prefabricated combination column |
| CN114332741A (en) * | 2022-03-08 | 2022-04-12 | 盈嘉互联（北京）科技有限公司 | Video detection method and system for building digital twins |
| CN115424155A (en) * | 2022-11-04 | 2022-12-02 | 浙江大华技术股份有限公司 | Illegal construction detection method, illegal construction detection device and computer storage medium |
| CN115857447A (en) * | 2022-11-28 | 2023-03-28 | 安徽宝信信息科技有限公司 | Complex industrial system operation monitoring method and system based on digital twins |
| CN116214542A (en) * | 2023-03-23 | 2023-06-06 | 广东省特种设备检测研究院东莞检测院 | Digital twin system of spherical tank inner wall climbing robot |
| CN116449771A (en) * | 2023-05-10 | 2023-07-18 | 中国标准化研究院 | Digital twin modeling method of numerical control machine tool |
| CN116498061A (en) * | 2023-04-13 | 2023-07-28 | 中建二局第二建筑工程有限公司 | High formwork construction method for arc-shaped structure |
| WO2023159810A1 (en) * | 2022-02-25 | 2023-08-31 | 山东大学 | Digital twinning-based construction monitoring method and system for concrete-filled steel tube arch bridges |
| KR20230129137A (en) * | 2022-02-28 | 2023-09-06 | 주식회사 한국산업기술원 | IoT, IoT, Meta works, construction, management system |
| CN117036999A (en) * | 2023-08-15 | 2023-11-10 | 中国南方电网有限责任公司超高压输电公司广州局 | Digital twinning-based power transformation equipment modeling method |
| CN117094660A (en) * | 2023-07-14 | 2023-11-21 | 中建三局上海有限公司 | Construction monitoring method and system based on digital twin technology |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114746815A (en) * | 2019-09-24 | 2022-07-12 | 西门子股份公司 | Method for creating a digital twin of a facility or device |
-
2023
- 2023-11-28 CN CN202311596887.0A patent/CN117648596B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112647590A (en) * | 2020-12-01 | 2021-04-13 | 河北建工集团有限责任公司 | Construction method for prefabricated assembled reinforced concrete beam column steel node prefabricated combination column |
| WO2023159810A1 (en) * | 2022-02-25 | 2023-08-31 | 山东大学 | Digital twinning-based construction monitoring method and system for concrete-filled steel tube arch bridges |
| KR20230129137A (en) * | 2022-02-28 | 2023-09-06 | 주식회사 한국산업기술원 | IoT, IoT, Meta works, construction, management system |
| CN114332741A (en) * | 2022-03-08 | 2022-04-12 | 盈嘉互联（北京）科技有限公司 | Video detection method and system for building digital twins |
| CN115424155A (en) * | 2022-11-04 | 2022-12-02 | 浙江大华技术股份有限公司 | Illegal construction detection method, illegal construction detection device and computer storage medium |
| CN115857447A (en) * | 2022-11-28 | 2023-03-28 | 安徽宝信信息科技有限公司 | Complex industrial system operation monitoring method and system based on digital twins |
| CN116214542A (en) * | 2023-03-23 | 2023-06-06 | 广东省特种设备检测研究院东莞检测院 | Digital twin system of spherical tank inner wall climbing robot |
| CN116498061A (en) * | 2023-04-13 | 2023-07-28 | 中建二局第二建筑工程有限公司 | High formwork construction method for arc-shaped structure |
| CN116449771A (en) * | 2023-05-10 | 2023-07-18 | 中国标准化研究院 | Digital twin modeling method of numerical control machine tool |
| CN117094660A (en) * | 2023-07-14 | 2023-11-21 | 中建三局上海有限公司 | Construction monitoring method and system based on digital twin technology |
| CN117036999A (en) * | 2023-08-15 | 2023-11-10 | 中国南方电网有限责任公司超高压输电公司广州局 | Digital twinning-based power transformation equipment modeling method |
Non-Patent Citations (1)
| Title |
|---|
| 智慧建造在建筑工程安全施工管理中的应用;何小勇 等;《智能建筑与智慧城市》;20210930(第9期);86-87 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117648596A (en) | 2024-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102788572B (en) | Method, device and system for measuring attitude of lifting hook of engineering machinery | |
| CN113916130B (en) | Building position measuring method based on least square method | |
| CN117252863B (en) | Quick detection and analysis method for geographic information abnormal data | |
| CN112967312B (en) | Real-time robust displacement monitoring method and system for field rigid body target | |
| CN111709988A (en) | Method and device for determining characteristic information of object, electronic equipment and storage medium | |
| CN113408034A (en) | Engineering construction intelligent cloud management platform based on BIM building three-dimensional model technology | |
| CN108830317B (en) | Rapid and fine evaluation method for joint attitude of surface mine slope rock mass based on digital photogrammetry | |
| Barrile et al. | Structural modeling of a historic castle using close range photogrammetry | |
| CN116518864A (en) | Engineering structure full-field deformation detection method based on three-dimensional point cloud comparison analysis | |
| CN117076928A (en) | Bridge health state monitoring method, device and system and electronic equipment | |
| CN117648596B (en) | Digital twin and intelligent sensor fusion method and system for building construction | |
| CN115311574B (en) | Building monitoring method, equipment and medium | |
| TWI636276B (en) | Method of determining earthquake with artificial intelligence and earthquake detecting system | |
| CN113822892B (en) | Evaluation method, device and equipment of simulated radar and computer storage medium | |
| CN115713548A (en) | Automatic registration method for multi-stage live-action three-dimensional model | |
| CN114299421A (en) | Method for detecting bolt looseness, convolutional neural network, training method and system | |
| Ostrowski et al. | A framework for computer vision-based health monitoring of a truss structure subjected to unknown excitations | |
| CN107358655B (en) | Identification method of hemispherical surface and conical surface models based on discrete stationary wavelet transform | |
| CN117058152B (en) | Intelligent detection method for construction quality of fabricated building based on point cloud data | |
| CN110909067A (en) | Visual analysis system and method for ocean multidimensional data | |
| CN116681698B (en) | Spring automatic assembly quality detection method and system | |
| CN117952565B (en) | BIM model-based intelligent management method and system for fabricated building | |
| CN116433755B (en) | Structure dense displacement recognition method and system based on deformable three-dimensional model and optical flow representation learning | |
| CN117114782B (en) | Construction engineering cost analysis method | |
| Chen et al. | Characterizing point cloud data density for spatial change-based maintenance planning of civil infrastructure systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |