CN117173448B - Method and device for intelligently controlling and early warning progress of foundation engineering - Google Patents
Method and device for intelligently controlling and early warning progress of foundation engineering Download PDFInfo
- Publication number
- CN117173448B CN117173448B CN202310879277.5A CN202310879277A CN117173448B CN 117173448 B CN117173448 B CN 117173448B CN 202310879277 A CN202310879277 A CN 202310879277A CN 117173448 B CN117173448 B CN 117173448B
- Authority
- CN
- China
- Prior art keywords
- progress
- picture
- construction
- construction project
- engineering
- 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
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000010276 construction Methods 0.000 claims abstract description 138
- 230000002159 abnormal effect Effects 0.000 claims abstract description 62
- 230000005856 abnormality Effects 0.000 claims abstract description 42
- 238000012544 monitoring process Methods 0.000 claims abstract description 23
- 238000013528 artificial neural network Methods 0.000 claims abstract description 14
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 13
- 238000011176 pooling Methods 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 10
- 238000007689 inspection Methods 0.000 claims description 8
- 239000012212 insulator Substances 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 235000005770 birds nest Nutrition 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 4
- 238000004880 explosion Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 235000005765 wild carrot Nutrition 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 abstract description 7
- 238000007726 management method Methods 0.000 description 16
- 238000013507 mapping Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000000763 evoking effect Effects 0.000 description 2
- 238000013508 migration Methods 0.000 description 2
- 230000005012 migration Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000009440 infrastructure construction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
A method and a device for intelligently controlling and early warning the progress of a foundation engineering, the method comprises the following steps: the method comprises the steps that a full convolution neural network image recognition algorithm based on a region recognizes a construction project progress picture; establishing a formalized model for monitoring the identification result of the progress picture of the foundation engineering, determining a progress abnormality picture, generating a progress abnormality notification card and publishing and early warning; detecting the progress abnormal picture by using a density peak value-fuzzy C-means clustering algorithm, comparing the progress abnormal picture with the construction equipment in the normal construction project progress picture, and determining the construction project progress deviation; and determining an abnormal unit according to the foundation engineering progress deviation, correcting the abnormal unit, and making a new foundation engineering progress strategy. The invention quantifies the internal relation among engineering progress through the deep learning algorithm, digs the internal rule thereof, and provides guarantee for the safe and smooth promotion of the foundation engineering.
Description
Technical Field
The invention relates to the field of construction projects, in particular to an intelligent control and early warning method and device for progress of construction projects.
Background
The territories of China are wide, the terrains are relatively complex, certain difficulty is brought to construction of the trans-regional power grid and the ultra-high voltage transmission line engineering, and the requirements of high efficiency and high speed cannot be met only by means of the existing inspection means and conventional tests. The unmanned aerial vehicle intelligent management system has the advantages that the unmanned aerial vehicle technology is applied, electric power inspection and construction planning tasks can be well completed, and intelligent management is conducted on the processes of infrastructure construction management, inspection and the like through unmanned aerial vehicle inspection application. How to utilize unmanned aerial vehicle autonomous flight, video passback etc. technique, realize the intelligent management and control of capital construction engineering progress and early warning and lean management is the key problem that needs to solve at present.
The invention discloses a detection method of a small sample intelligent substation power equipment component based on deep migration learning from page 44, 1148 of journal 2020 of electric network technology, and provides a single-stage multi-frame detector intelligent substation power equipment image target detection algorithm based on migration learning, which can optimally detect a small sample power equipment data set. The publication of the electric network technology, journal 2021, 45, page 713 discloses a fault positioning method of secondary equipment of an intelligent substation based on deep learning. The application method of deep learning in intelligent defect recognition of transmission line engineering acceptance is disclosed in page 44, page 5 of journal 2020 of Jiangxi electric power, and the method adopts a deep learning algorithm to learn and recognize acquired images, so that the acceptance quality and accuracy are ensured, and the line engineering acceptance is intelligent while the personnel cost is reduced. However, the above prior art is mainly focused on applying deep learning to power grid equipment and fault detection, and there are few studies considering management of progress of a construction project using deep learning.
Disclosure of Invention
The invention aims to overcome the defect and the problem that the progress of the foundation engineering cannot be intelligently controlled and pre-warned in the prior art, and provides a method and a device for intelligently controlling and pre-warning the progress of the foundation engineering, which are beneficial to improving the safety and stability of the foundation engineering and ensuring the correct progress of the foundation engineering.
In order to achieve the above object, the technical solution of the present invention is: a method for intelligently controlling and early warning the progress of a foundation engineering comprises the following steps:
the method comprises the steps that a full convolution neural network image recognition algorithm based on a region recognizes a construction project progress picture;
Establishing a formalized model for monitoring the identification result of the progress picture of the foundation engineering, determining a progress abnormality picture, generating a progress abnormality notification card and publishing and early warning;
detecting the progress abnormal picture by using a density peak value-fuzzy C-means clustering algorithm, comparing the progress abnormal picture with the construction equipment in the normal construction project progress picture, and determining the construction project progress deviation;
And determining an abnormal unit according to the foundation engineering progress deviation, correcting the abnormal unit, and making a new foundation engineering progress strategy.
The method for acquiring the foundation engineering progress picture comprises the following steps:
establishing a scene set covering various foundation projects progress based on pictures collected in unmanned aerial vehicle inspection management and control;
manually labeling the progress targets and the abnormal categories in the picture set one by one according to the different progress targets and the abnormal categories;
Dividing the marked picture set according to the abnormal category proportion to obtain a training set and a testing set.
The abnormal categories comprise bird nest, foreign matters on the wires, insulator self-explosion, insulator pollution, discharge gap damage and wire clamp inclination.
The area-based full convolution neural network image recognition algorithm comprises the following steps:
preprocessing the picture to extract feature vectors;
Sending the preprocessed picture into a pre-trained classification network, and fixing corresponding network parameters; three branches exist on the feature map obtained by the last convolution layer of the pre-trained classification network; the first branch is to perform region selection network operation on the feature map to obtain a corresponding region of interest (ROI); the second branch is to obtain a K x (C+1) dimensional position sensitive score map on the feature map for classification; the third branch is to obtain a 4 XK X K-dimensional position sensitive score map on the feature map for data regression;
And performing a position-sensitive ROI pooling operation on the position-sensitive score mapping in the K X (C+1) dimension and the position-sensitive score mapping in the 4X K dimension respectively to obtain corresponding abnormal category and position information.
The formalized model for monitoring the identification result of the foundation engineering progress picture is as follows:
Wherein FS is a formalized model; TC t is the expected progress task state of the construction project; PC t is the actual progress state of the construction project; The actual state of the construction engineering equipment is established; t is the construction engineering time; k is a certain progress stage of the foundation engineering; The number of the type N of the construction engineering equipment in the period t; N_WIP t is the total amount of the construction engineering equipment; /(I) AndQueuing the sequence for the device waiting to be used and waiting to be finished; /(I)Monitoring equipment for participating in a capital construction project; QP WIP,m is the monitoring result.
And according to the monitored formalized model, the time and place of the progress abnormality picture shooting, the picture specific equipment information and the regional engineering responsible person information are clarified, a progress abnormality notification card is generated and the publication and early warning are carried out, and the progress abnormality notification card comprises progress abnormality description, potential accident types and responsible person information.
The construction project progress deviation J is as follows:
Wherein n is the data quantity of the middle layer after the picture training of the construction engineering equipment; c is the number of picture samples of the construction engineering equipment; w j is the similarity of the picture samples of the construction engineering equipment; mu i,j is a membership matrix of a sample picture of the construction engineering equipment to a clustering center; c i is the sample center; x j is a minimum anchor point frame sample after the picture of the construction engineering equipment is segmented.
An intelligent management and control and early warning device of foundation engineering progress, includes:
the construction project progress picture identification module is used for identifying the construction project progress picture based on a full convolution neural network image identification algorithm of the region;
The progress abnormality picture determining and early warning module is used for establishing a formalized model for monitoring the identification result of the progress picture of the foundation engineering, determining the progress abnormality picture, generating a progress abnormality notification card and publishing and early warning;
The construction project progress deviation determining module is used for detecting the progress abnormal picture by using a density peak value-fuzzy C-means clustering algorithm, comparing the progress abnormal picture with construction equipment in a normal construction project progress picture, and determining the construction project progress deviation;
And the foundation engineering progress correction module is used for determining an abnormal unit according to the foundation engineering progress deviation, correcting the abnormal unit and making a new foundation engineering progress strategy.
The intelligent control and early warning equipment for the progress of the construction project comprises a memory and a processor;
The memory is used for storing computer program codes and transmitting the computer program codes to the processor;
The processor is configured to perform the method as described above according to instructions in the computer program code.
A computer readable storage medium having stored thereon a computer program which when executed by a processor implements a method as described above.
Compared with the prior art, the invention has the beneficial effects that:
In the intelligent management and control and early warning method and device for the progress of the construction project, firstly, a full convolution neural network image recognition algorithm based on a region is used for recognizing a progress picture of the construction project; secondly, a formalized model for monitoring the identification result of the progress picture of the foundation engineering is established, a progress abnormality picture is determined, a progress abnormality notification card is generated, and a publication early warning is carried out; then, detecting the progress abnormal picture by using a density peak value-fuzzy C-means clustering algorithm, comparing the progress abnormal picture with the construction equipment in the normal construction project progress picture, and determining the construction project progress deviation; and finally, determining an abnormal unit according to the construction project progress deviation, correcting the abnormal unit, and making a new construction project progress strategy. The invention quantifies the internal relation among engineering progress through the deep learning algorithm, digs the internal rule thereof, and provides guarantee for the safe and smooth promotion of the foundation engineering.
Drawings
FIG. 1 is a flow chart of a method for intelligent control and early warning of progress of a construction project.
FIG. 2 is a flow chart of a progress anomaly picture intelligent interpretation architecture in an embodiment of the invention.
FIG. 3 is a block diagram of a device for intelligent control and early warning of progress of a construction project.
FIG. 4 is a block diagram of a construction project progress intelligent control and early warning device of the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings and detailed description.
Example 1:
referring to fig. 1, an intelligent management and early warning method for progress of a construction project includes:
s1, identifying a construction project progress picture by a full convolution neural network image identification algorithm based on a region;
The method for acquiring the foundation engineering progress picture comprises the following steps:
Establishing a scene set covering various foundation projects progress based on pictures collected in unmanned aerial vehicle inspection management and control; carrying out statistics and summarization on the abnormal types found in the picture set to obtain common abnormal types; the picture set covers the scene picture of the progress of the construction project;
Manually labeling the progress targets and the abnormal categories in the picture set one by one according to the different progress targets and the abnormal categories; the abnormal categories comprise six categories, namely bird nest, foreign wire bodies, insulator self-explosion, insulator pollution, discharge gap damage and wire clamp inclination;
Dividing the marked picture set according to the abnormal category proportion to obtain a training set and a testing set; the abnormal category proportion is determined according to the proportion of pictures related to different abnormal categories in the actual construction project to the total picture set.
The area-based full convolution neural network image recognition algorithm comprises the following steps:
Preprocessing the picture to extract feature vectors; and (3) carrying out convolution on the whole picture input to obtain feature vectors, obtaining 2000 areas to be detected by using a selective search algorithm, sending the feature images of the areas to be detected to an interesting area pooling layer for unified processing, and finally obtaining the feature images with fixed sizes and inputting the feature images to a subsequent full-connection layer, wherein the process is a feature extraction process.
Sending the preprocessed picture into a pre-trained classification network, and fixing corresponding network parameters; three branches exist on the feature map obtained by the last convolution layer of the pre-trained classification network; the first branch is to perform region selection (Region Proposal Network, RPN) network operation on the feature map to obtain a corresponding region of interest (ROI); the second branch is to obtain a K x (C+1) dimensional position sensitive score map on the feature map for classification; the third branch is to obtain a4 XK X K-dimensional position sensitive score map on the feature map for data regression;
And performing a position-sensitive ROI pooling operation on the position-sensitive score mapping in the K X (C+1) dimension and the position-sensitive score mapping in the 4X K dimension respectively to obtain corresponding abnormal category and position information.
Region selection, namely RPN, is a method for inputting a characteristic diagram after convolution and outputting a candidate block diagram. Each point on the feature map generates an anchor block, each block having corresponding coordinates to the original map, and the anchor blocks are subjected to a classification network convolution process as shown in table 1 below to generate a multi-dimensional vector of the middle layer. Each vector contains the position information of an anchor block, and the ROI is acquired according to the set target sample.
The position sensitive score mapping is obtained by the following steps: and placing the ROI pooling operation layers between the convolution layers, extracting shared features by using the convolution layers of the common classification network, then extracting feature vectors on a feature map formed by the last layer of network by using the ROI pooling operation, and inputting the feature vectors into the full-connection layer for ROI classification and regression. The ROI is divided into limited sub-regions which map one-to-one with the original, i.e. position sensitive score maps.
S2, a formalized model for monitoring the identification result of the progress picture of the foundation engineering is established, a progress abnormality picture is determined, a progress abnormality notification card is generated, and a publication early warning is carried out;
The formalized model for monitoring the identification result of the foundation engineering progress picture is as follows:
Wherein FS is a formalized model; TC t is the expected progress task state of the construction project; PC t is the actual progress state of the construction project; The actual state of the construction engineering equipment is established; t is the construction engineering time; k is a certain progress stage of the foundation engineering; The number of the type N of the construction engineering equipment in the period t; N_WIP t is the total amount of the construction engineering equipment; /(I) AndQueuing the sequence for the device waiting to be used and waiting to be finished; /(I)Monitoring equipment for participating in a capital construction project; QP WIP,m is the monitoring result.
And according to the monitored formalized model, the time and place of the progress abnormality picture shooting, the picture specific equipment information and the regional engineering responsible person information are clarified, a progress abnormality notification card is generated and the publication and early warning are carried out, and the progress abnormality notification card comprises progress abnormality description, potential accident types and responsible person information. And posting a progress abnormality notification card in the construction area, and simultaneously automatically generating a two-dimensional code of an area progress abnormality list, wherein each project progress manager can comprehensively know and control the project progress of the foundation area by scanning the two-dimensional code.
S3, detecting the progress abnormal picture by using a density peak value-fuzzy C-means clustering algorithm, comparing the progress abnormal picture with the construction equipment in the normal construction project progress picture, and determining the construction project progress deviation;
The construction project progress deviation J is as follows:
Wherein n is the data quantity of the middle layer after the picture training of the construction engineering equipment; c is the number of picture samples of the construction engineering equipment; w j is the similarity of the picture samples of the construction engineering equipment; mu i,j is a membership matrix of a sample picture of the construction engineering equipment to a clustering center; c i is the sample center; x j is a minimum anchor point frame sample after the picture of the construction engineering equipment is segmented.
S4, determining an abnormal unit according to the construction project progress deviation, correcting the abnormal unit, and making a new construction project progress strategy.
The invention identifies the collected pictures of the construction project based on an image identification algorithm of a regional full convolution network (R-FCN), detects the pictures based on the result, and identifies progress abnormal pictures; secondly, based on an R-FCN algorithm, providing a key technical logic of progress monitoring-anomaly discovery-problem tracing for intelligent management and control early warning of the progress of the construction engineering for picture identification results; and finally, providing a control and early warning technical guide for implementing 'clear progress deviation-abnormal unit positioning-evoked factor searching-progress correction strategy' on the progress of the foundation engineering according to key technical logic. Positioning an abnormal unit: and determining an abnormal unit according to the information reflected by the progress abnormal picture in time, and checking related time, area range, responsible person and contact way. Searching for an induction factor: after the related information of abnormal progress is determined, the evoked factors of the weak links of the basic construction project are compared and searched. And (3) correcting a progress strategy: and correcting the progress abnormality related unit in time, changing an error link, making a new progress strategy, and carrying out progress early warning on the progress of the foundation engineering.
As shown in fig. 2, first, collecting data of progress of a construction project, and summarizing samples to obtain categories of progress abnormal pictures; secondly, manually marking targets and anomalies in the picture set according to the progress targets and the progress anomalies; and finally, dividing the marked pictures according to abnormal categories of the progress in the set, wherein four fifths of the pictures are designated as training sets for training the algorithm model, the rest of the pictures are designated as test sets, and the user algorithm model is evaluated.
Table 1 is a categorized network structure. The network structure comprises two layers of convolution layer base layers, 9 layers of convolution layers forming a first convolution group, 9 layers of convolution layers forming a second convolution group, 69 layers of convolution layers forming a third convolution group and a full connection layer. The convolution module 1 performs convolution operation by using 64 convolution check pictures of 7×7; the convolution module 2_x performs a pooling operation using a convolution kernel of 3×3 and performs 3 residual blocks; the convolution module 3_x convolves with 128 3×3 convolution kernels, performing 4 residual blocks; the convolution module 4_x convolves with 256 3 x 3 convolution kernels, performing 6 residual blocks. And finally, carrying out averaging and pooling, and then carrying out full connection layer and softmax activation function classification to obtain 1 multiplied by 1 output.
Table 1 classification network structure
Due to the complexity of the power grid infrastructure engineering, the invention carries out staged treatment on the progress of the infrastructure engineering, namely four stages of early stage, transformer substation engineering, line engineering and communication engineering. Comparing the R-FCN algorithm used by the invention with other two types of algorithms, setting the threshold value of the three algorithms IoU to be 0.5, unifying and normalizing the image resolution of the model training data set to be 800 x 1200, adopting a small batch gradient descent method, dividing the size of each batch into two images, dividing the learning rate into two batches, and training 36000 batches. The results are shown in table 2, and the target accuracy of the early stage of the construction project, the substation project, the line project and the communication project is 84.62%, 94.16%, 70.57% and 84.85% respectively, and the accuracy is higher than that of the other two algorithms.
Table 2R-FCN algorithm versus other algorithm target accuracy
Example 2:
Referring to fig. 3, an intelligent management and control and early warning device for progress of a construction project includes:
the construction project progress picture identification module is used for identifying the construction project progress picture based on a full convolution neural network image identification algorithm of the region;
The method for acquiring the foundation engineering progress picture comprises the following steps:
establishing a scene set covering various foundation projects progress based on pictures collected in unmanned aerial vehicle inspection management and control;
manually labeling the progress targets and the abnormal categories in the picture set one by one according to the different progress targets and the abnormal categories;
Dividing the marked picture set according to the abnormal category proportion to obtain a training set and a testing set.
The abnormal categories comprise bird nest, foreign matters on the wires, insulator self-explosion, insulator pollution, discharge gap damage and wire clamp inclination.
The area-based full convolution neural network image recognition algorithm comprises the following steps:
preprocessing the picture to extract feature vectors;
Sending the preprocessed picture into a pre-trained classification network, and fixing corresponding network parameters; three branches exist on the feature map obtained by the last convolution layer of the pre-trained classification network; the first branch is to perform region selection network operation on the feature map to obtain a corresponding region of interest (ROI); the second branch is to obtain a K x (C+1) dimensional position sensitive score map on the feature map for classification; the third branch is to obtain a 4 XK X K-dimensional position sensitive score map on the feature map for data regression;
And performing a position-sensitive ROI pooling operation on the position-sensitive score mapping in the K X (C+1) dimension and the position-sensitive score mapping in the 4X K dimension respectively to obtain corresponding abnormal category and position information.
The progress abnormality picture determining and early warning module is used for establishing a formalized model for monitoring the identification result of the progress picture of the foundation engineering, determining the progress abnormality picture, generating a progress abnormality notification card and publishing and early warning;
The formalized model for monitoring the identification result of the foundation engineering progress picture is as follows:
Wherein FS is a formalized model; TC t is the expected progress task state of the construction project; PC t is the actual progress state of the construction project; The actual state of the construction engineering equipment is established; t is the construction engineering time; k is a certain progress stage of the foundation engineering; The number of the type N of the construction engineering equipment in the period t; N_WIP t is the total amount of the construction engineering equipment; /(I) AndQueuing the sequence for the device waiting to be used and waiting to be finished; /(I)Monitoring equipment for participating in a capital construction project; QP WIP,m is the monitoring result.
And according to the monitored formalized model, the time and place of the progress abnormality picture shooting, the picture specific equipment information and the regional engineering responsible person information are clarified, a progress abnormality notification card is generated and the publication and early warning are carried out, and the progress abnormality notification card comprises progress abnormality description, potential accident types and responsible person information.
The construction project progress deviation determining module is used for detecting the progress abnormal picture by using a density peak value-fuzzy C-means clustering algorithm, comparing the progress abnormal picture with construction equipment in a normal construction project progress picture, and determining the construction project progress deviation;
The construction project progress deviation J is as follows:
Wherein n is the data quantity of the middle layer after the picture training of the construction engineering equipment; c is the number of picture samples of the construction engineering equipment; w j is the similarity of the picture samples of the construction engineering equipment; mu i,j is a membership matrix of a sample picture of the construction engineering equipment to a clustering center; c i is the sample center; x j is a minimum anchor point frame sample after the picture of the construction engineering equipment is segmented.
And the foundation engineering progress correction module is used for determining an abnormal unit according to the foundation engineering progress deviation, correcting the abnormal unit and making a new foundation engineering progress strategy.
Example 3:
Referring to fig. 4, a device for intelligently controlling and early warning progress of a construction project comprises a memory and a processor;
The memory is used for storing computer program codes and transmitting the computer program codes to the processor;
the processor is used for executing an intelligent management, control and early warning method for the progress of the construction project according to instructions in the computer program codes.
A computer readable storage medium, on which a computer program is stored, which when executed by a processor implements a method for intelligent management and early warning of progress of a construction project.
In general, the computer instructions to implement the methods of the present invention may be carried in any combination of one or more computer-readable storage media. The non-transitory computer-readable storage medium may include any computer-readable medium, except the signal itself in temporary propagation.
The computer readable storage medium can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAn), a read-only memory (ROn), an erasable programmable read-only memory (EKROn or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROn), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
Computer program code for carrying out operations of the present invention may be written in one or more programming languages, including an object oriented programming language such as Java, snalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages, and in particular, kython, and TensorFlow, kyTorch-based platform frameworks suitable for neural network computing may be used. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any number of types of networks, including a Local Area Network (LAN) or a Wide Area Network (WAN), or be connected to an external computer (for example, through the Internet using an Internet service provider).
The above devices and non-transitory computer readable storage medium may refer to a specific description of a method for intelligently controlling and early warning progress of a construction project and beneficial effects, and will not be described herein.
While embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives, and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.
Claims (6)
1. The intelligent control and early warning method for the progress of the construction project is characterized by comprising the following steps of:
the method comprises the steps that a full convolution neural network image recognition algorithm based on a region recognizes a construction project progress picture;
the area-based full convolution neural network image recognition algorithm comprises the following steps:
preprocessing the picture to extract feature vectors;
Sending the preprocessed picture into a pre-trained classification network, and fixing corresponding network parameters; three branches exist on the feature map obtained by the last convolution layer of the pre-trained classification network; the first branch is to perform region selection network operation on the feature map to obtain a corresponding region of interest (ROI); the second branch is to obtain a K x (C+1) dimensional position sensitive score map on the feature map for classification; the third branch is to obtain a 4 XK X K-dimensional position sensitive score map on the feature map for data regression;
Performing a position-sensitive ROI pooling operation on the position-sensitive score map in the kxkx (c+1) dimension and the position-sensitive score map in the 4 xkx K dimension, respectively, to obtain corresponding anomaly category and position information;
Establishing a formalized model for monitoring the identification result of the progress picture of the foundation engineering, determining a progress abnormality picture, generating a progress abnormality notification card and publishing and early warning;
The formalized model for monitoring the identification result of the foundation engineering progress picture is as follows:
Wherein FS is a formalized model; TC t is the expected progress task state of the construction project; PC t is the actual progress state of the construction project; the actual state of the construction engineering equipment is established; t is the construction engineering time; k is a certain progress stage of the foundation engineering; /(I) The number of the type N of the construction engineering equipment in the period t; N_WIP t is the total amount of the construction engineering equipment; /(I)AndQueuing the sequence for the device waiting to be used and waiting to be finished; /(I)Monitoring equipment for participating in a capital construction project; QP WIP,m is the monitoring result;
according to the monitored formalized model, the time and place of the progress abnormality picture shooting, the picture specific equipment information and the regional engineering responsible person information are clarified, a progress abnormality notification card is generated and is subjected to publication early warning, and the progress abnormality notification card comprises progress abnormality description, potential accident types and responsible person information;
detecting the progress abnormal picture by using a density peak value-fuzzy C-means clustering algorithm, comparing the progress abnormal picture with the construction equipment in the normal construction project progress picture, and determining the construction project progress deviation;
The construction project progress deviation J is as follows:
wherein n is the data quantity of the middle layer after the picture training of the construction engineering equipment; c is the number of picture samples of the construction engineering equipment; w j is the similarity of the picture samples of the construction engineering equipment; mu i,j is a membership matrix of a sample picture of the construction engineering equipment to a clustering center; c i is the sample center; x j is a minimum anchor point frame sample after the picture of the construction engineering equipment is segmented;
And determining an abnormal unit according to the foundation engineering progress deviation, correcting the abnormal unit, and making a new foundation engineering progress strategy.
2. The method for intelligently controlling and early warning the progress of the construction project according to claim 1, wherein the method for acquiring the picture of the progress of the construction project is as follows:
establishing a scene set covering various foundation projects progress based on pictures collected in unmanned aerial vehicle inspection management and control;
manually labeling the progress targets and the abnormal categories in the picture set one by one according to the different progress targets and the abnormal categories;
Dividing the marked picture set according to the abnormal category proportion to obtain a training set and a testing set.
3. The intelligent control and early warning method for the progress of the construction project according to claim 2, wherein the abnormal categories comprise six categories, namely bird nest, foreign matters on wires, self-explosion of insulators, pollution of insulators, damage of discharge gaps and inclination of wire clamps.
4. The utility model provides a construction engineering progress intelligent management and control and early warning device which characterized in that includes:
the construction project progress picture identification module is used for identifying the construction project progress picture based on a full convolution neural network image identification algorithm of the region;
the area-based full convolution neural network image recognition algorithm comprises the following steps:
preprocessing the picture to extract feature vectors;
Sending the preprocessed picture into a pre-trained classification network, and fixing corresponding network parameters; three branches exist on the feature map obtained by the last convolution layer of the pre-trained classification network; the first branch is to perform region selection network operation on the feature map to obtain a corresponding region of interest (ROI); the second branch is to obtain a K x (C+1) dimensional position sensitive score map on the feature map for classification; the third branch is to obtain a 4 XK X K-dimensional position sensitive score map on the feature map for data regression;
Performing a position-sensitive ROI pooling operation on the position-sensitive score map in the kxkx (c+1) dimension and the position-sensitive score map in the 4 xkx K dimension, respectively, to obtain corresponding anomaly category and position information;
The progress abnormality picture determining and early warning module is used for establishing a formalized model for monitoring the identification result of the progress picture of the foundation engineering, determining the progress abnormality picture, generating a progress abnormality notification card and publishing and early warning;
The formalized model for monitoring the identification result of the foundation engineering progress picture is as follows:
Wherein FS is a formalized model; TC t is the expected progress task state of the construction project; PC t is the actual progress state of the construction project; the actual state of the construction engineering equipment is established; t is the construction engineering time; k is a certain progress stage of the foundation engineering; /(I) The number of the type N of the construction engineering equipment in the period t; N_WIP t is the total amount of the construction engineering equipment; /(I)AndQueuing the sequence for the device waiting to be used and waiting to be finished; /(I)Monitoring equipment for participating in a capital construction project; QP WIP,m is the monitoring result;
according to the monitored formalized model, the time and place of the progress abnormality picture shooting, the picture specific equipment information and the regional engineering responsible person information are clarified, a progress abnormality notification card is generated and is subjected to publication early warning, and the progress abnormality notification card comprises progress abnormality description, potential accident types and responsible person information;
The construction project progress deviation determining module is used for detecting the progress abnormal picture by using a density peak value-fuzzy C-means clustering algorithm, comparing the progress abnormal picture with construction equipment in a normal construction project progress picture, and determining the construction project progress deviation;
The construction project progress deviation J is as follows:
wherein n is the data quantity of the middle layer after the picture training of the construction engineering equipment; c is the number of picture samples of the construction engineering equipment; w j is the similarity of the picture samples of the construction engineering equipment; mu i,j is a membership matrix of a sample picture of the construction engineering equipment to a clustering center; c i is the sample center; x j is a minimum anchor point frame sample after the picture of the construction engineering equipment is segmented;
And the foundation engineering progress correction module is used for determining an abnormal unit according to the foundation engineering progress deviation, correcting the abnormal unit and making a new foundation engineering progress strategy.
5. A construction progress intelligent control and early warning device is characterized in that,
Comprising a memory and a processor;
The memory is used for storing computer program codes and transmitting the computer program codes to the processor;
the processor being configured to perform the method of any one of claims 1 to 3 according to instructions in the computer program code.
6. A computer readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, implements the method according to any of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310879277.5A CN117173448B (en) | 2023-07-18 | 2023-07-18 | Method and device for intelligently controlling and early warning progress of foundation engineering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310879277.5A CN117173448B (en) | 2023-07-18 | 2023-07-18 | Method and device for intelligently controlling and early warning progress of foundation engineering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117173448A CN117173448A (en) | 2023-12-05 |
| CN117173448B true CN117173448B (en) | 2024-05-24 |
Family
ID=88943812
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310879277.5A Active CN117173448B (en) | 2023-07-18 | 2023-07-18 | Method and device for intelligently controlling and early warning progress of foundation engineering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117173448B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020164270A1 (en) * | 2019-02-15 | 2020-08-20 | 平安科技（深圳）有限公司 | Deep-learning-based pedestrian detection method, system and apparatus, and storage medium |
| CN113807229A (en) * | 2021-09-13 | 2021-12-17 | 深圳市巨龙创视科技有限公司 | Non-contact attendance checking device, method, equipment and storage medium for intelligent classroom |
| CN115953666A (en) * | 2023-03-15 | 2023-04-11 | 国网湖北省电力有限公司经济技术研究院 | Transformer substation field progress identification method based on improved Mask-RCNN |
| CN116307566A (en) * | 2023-03-12 | 2023-06-23 | 武汉大学 | Dynamic design system for large-scale building construction project construction organization scheme |
-
2023
- 2023-07-18 CN CN202310879277.5A patent/CN117173448B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020164270A1 (en) * | 2019-02-15 | 2020-08-20 | 平安科技（深圳）有限公司 | Deep-learning-based pedestrian detection method, system and apparatus, and storage medium |
| CN113807229A (en) * | 2021-09-13 | 2021-12-17 | 深圳市巨龙创视科技有限公司 | Non-contact attendance checking device, method, equipment and storage medium for intelligent classroom |
| CN116307566A (en) * | 2023-03-12 | 2023-06-23 | 武汉大学 | Dynamic design system for large-scale building construction project construction organization scheme |
| CN115953666A (en) * | 2023-03-15 | 2023-04-11 | 国网湖北省电力有限公司经济技术研究院 | Transformer substation field progress identification method based on improved Mask-RCNN |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117173448A (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108022235B (en) | Method for identifying defects of key components of high-voltage transmission iron tower | |
| CN109858367B (en) | Visual automatic detection method and system for worker through supporting unsafe behaviors | |
| CN110580475A (en) | line diagnosis method based on unmanned aerial vehicle inspection, electronic device and storage medium | |
| CN109872317A (en) | A kind of defect identification method based on power equipments defect identification learning model | |
| CN109580004A (en) | A kind of temperature checking method and device | |
| CN110728300A (en) | Method and system for identifying fault type based on turnout action current curve | |
| CN112419268A (en) | Method, device, equipment and medium for detecting image defects of power transmission line | |
| CN110992307A (en) | Insulator positioning and identifying method and device based on YOLO | |
| CN116503318A (en) | Aerial insulator multi-defect detection method, system and equipment integrating CAT-BiFPN and attention mechanism | |
| CN116681962A (en) | Power equipment thermal image detection method and system based on improved YOLOv5 | |
| Hao et al. | Detection of bird nests on power line patrol using single shot detector | |
| CN116309303A (en) | Electrical equipment defect detection method and device based on infrared image and related equipment | |
| CN115984158A (en) | Defect analysis method and device, electronic equipment and computer readable storage medium | |
| CN117114420B (en) | Image recognition-based industrial and trade safety accident risk management and control system and method | |
| CN114227717A (en) | Intelligent inspection method, device, equipment and storage medium based on inspection robot | |
| CN112419243B (en) | Power distribution room equipment fault identification method based on infrared image analysis | |
| Özer et al. | An approach based on deep learning methods to detect the condition of solar panels in solar power plants | |
| CN111191027B (en) | Generalized zero sample identification method based on Gaussian mixture distribution (VAE) | |
| CN117173448B (en) | Method and device for intelligently controlling and early warning progress of foundation engineering | |
| CN117252840A (en) | Photovoltaic array defect elimination evaluation method and device and computer equipment | |
| CN117152528A (en) | Insulator state recognition method, insulator state recognition device, insulator state recognition apparatus, insulator state recognition program, and insulator state recognition program | |
| Wang et al. | High-Voltage Transmission Line Foreign Object and Power Component Defect Detection Based on Improved YOLOv5 | |
| CN115984759A (en) | Substation switch state identification method and device, computer equipment and storage medium | |
| CN115937492A (en) | Transformer equipment infrared image identification method based on feature identification | |
| CN115908295A (en) | Power grid insulator defect detection method and system based on deep learning |
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 |