CN116186634B - Intelligent management system for construction data of building engineering - Google Patents
Intelligent management system for construction data of building engineering Download PDFInfo
- Publication number
- CN116186634B CN116186634B CN202310457096.3A CN202310457096A CN116186634B CN 116186634 B CN116186634 B CN 116186634B CN 202310457096 A CN202310457096 A CN 202310457096A CN 116186634 B CN116186634 B CN 116186634B
- Authority
- CN
- China
- Prior art keywords
- analyzed
- data
- distance
- abnormal
- points
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE
- G01L1/00—Measuring force or stress, in general
- G01L1/10—Measuring force or stress, in general by measuring variations of frequency of stressed vibrating elements, e.g. of stressed strings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention relates to the technical field of electric digital data processing, in particular to an intelligent management system for construction data of building engineering. The system collects temperature, frequency and soil pressure information of the vibrating wire type soil pressure box together. And performing primary data screening on the temperature sequence according to the normal temperature range to obtain a first abnormal moment, and further performing correlation analysis on the temperature information and the frequency information to obtain a scatter diagram to be analyzed. The aggregation distance for clustering is obtained by integrating the areas to be analyzed in the scatter diagram to be analyzed after independent analysis, so that an accurate clustering result is obtained, abnormal data points are screened out according to the clustering result, and further a second abnormal moment is obtained. And taking the soil pressure data corresponding to the first abnormal time and the second abnormal time as the screened data to perform abnormality degree analysis. The invention realizes the secondary screening of the data through the correlation analysis of the temperature and the frequency, and obtains the accurate data to be analyzed abnormally through the screening and the management of the data.
Description
Technical Field
The invention relates to the technical field of electric digital data processing, in particular to an intelligent management system for construction data of building engineering.
Background
Deep foundation pit support is an important link in construction of constructional engineering, and influences the quality, safety and other problems of the constructional engineering. In the process of deep foundation pit supporting construction, the pit body of the deep foundation pit can be subjected to the action of lateral soil pressure from surrounding soil, and the stability and the form of an engineering structure can be absolutely influenced by the action of the lateral soil pressure, so that soil pressure data of the deep foundation pit are required to be monitored in the construction process of the building engineering.
For the soil pressure data acquired in real time, the degree of abnormality of the data in a period of time can be obtained through an abnormality detection algorithm. However, because the construction period of the building engineering is longer and the construction period is more, a large amount of data can appear in the data acquisition process, and more data exist in the data, which belong to normal soil pressure data, and an abnormality detection algorithm is not needed, so that the obtained data are required to be managed to reduce the data amount of abnormality detection, and the data which need to participate in the abnormality detection are screened out. In the prior art, soil pressure data can be screened through a preset data range, so that abnormal data can be obtained. However, the normal data range corresponding to the soil pressure data in different construction scenes is changed, so that the soil pressure data cannot be accurately managed in the prior art, namely, the data to be abnormally analyzed cannot be accurately screened out from the acquired data.
Disclosure of Invention
In order to solve the technical problem of inaccurate data screening of the abnormal analysis of the soil pressure data, the invention aims to provide an intelligent management system for construction data of a building engineering, and the adopted technical scheme is as follows:
the invention provides an intelligent management system for construction data of building engineering, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein when the processor executes the computer program, the following method steps are realized:
obtaining a temperature sequence, a frequency sequence and a soil pressure sequence of a vibrating wire type soil pressure box in a deep foundation pit in the construction process of the building engineering in a preset sampling period according to a preset sampling frequency;
screening out a first abnormal moment in the temperature sequence according to a normal temperature range; constructing a two-dimensional scatter diagram according to the temperature sequence and the frequency sequence, deleting the scatter diagram corresponding to the first abnormal moment, and obtaining the scatter diagram to be analyzed;
uniformly dividing at least two areas to be analyzed in the scatter diagram to be analyzed; obtaining the concentration degree of scattered points in each area to be analyzed; obtaining the dense distance of the area to be analyzed according to the position distribution of the scattered points in the area to be analyzed; acquiring the gathering distance in the scatter diagram to be analyzed according to the gathering degree and the dense distance of all the areas to be analyzed;
clustering scattered points in the scattered points by taking the aggregation distance as a clustering radius in a DBSCAN clustering algorithm, and screening abnormal data points; taking the sampling time of the abnormal data point as a second abnormal time;
and carrying out abnormality degree analysis on elements corresponding to the first abnormality time and the second abnormality time in the soil pressure sequence.
Further, the step of screening the first abnormal time in the temperature sequence according to the normal temperature range includes:
obtaining an environmental temperature sequence in a preset sampling period, and obtaining an environmental temperature range according to the environmental temperature sequence; scaling the environment temperature range according to a preset adjustment coefficient to obtain the normal temperature range; and taking the sampling time corresponding to the element outside the normal temperature range in the temperature sequence as a first abnormal time.
Further, the obtaining the concentration degree of the scattered points in each area to be analyzed includes:
taking the product of standard deviation and polar difference of coordinates in each dimension in the region to be analyzed as the discrete feature of the corresponding dimension; obtaining integral discrete features according to the discrete features in two dimensions; obtaining the concentration degree according to the integral discrete features; the overall discrete features are inversely related to the concentration level.
Further, the obtaining the dense distance of the area to be analyzed according to the position distribution of the scattered points in the area to be analyzed includes:
clustering scattered points in the area to be analyzed according to the coordinate information to obtain at least two first cluster clusters, wherein the first cluster with the largest sample number in the cluster is used as a target cluster;
optionally one dimension as a target dimension; in the target cluster, non-target dimension coordinate differences between scattered points of two identical target dimension coordinates are obtained, and a first reference point pair is formed by two scattered points corresponding to the largest non-target dimension coordinate difference; obtaining the number of scattered points on the connecting line of the two points in the first reference point pair as the reference number; if more than one first reference point pair is provided, taking the average number of scattered points on all the connecting lines as the reference number; taking the ratio of the maximum non-target dimension coordinate difference to the reference number as an initial dense distance in a target dimension;
changing the target dimension to obtain the initial dense distance in each dimension;
the maximum initial dense distance between the two dimensions is taken as the dense distance of the region to be analyzed.
Further, the obtaining the aggregation distance in the scatter plot to be analyzed according to the concentration degree and the dense distance of all the areas to be analyzed includes:
taking the duty ratio of the concentration degree of each region to be analyzed in the concentration degree of all the regions to be analyzed as a weight coefficient; multiplying the weight coefficient by the dense distance of the corresponding area to be analyzed to obtain a weighted dense distance; and accumulating the weighted dense distances of all the areas to be analyzed to obtain the aggregate distance.
Further, the screening out outlier data points includes:
clustering scattered points in the scattered point map by taking the aggregation distance as a clustering radius in a DBSCAN clustering algorithm to obtain a second cluster and discrete points; taking the second cluster with the largest sample number as a cluster to be analyzed; taking the distance between the sample points in the cluster to be analyzed and the cluster center as a judgment distance; obtaining an average judgment distance in the cluster to be analyzed, and taking a sample point corresponding to the judgment distance larger than the average judgment distance as an abnormal sample; and taking the abnormal samples, the samples in the non-to-be-analyzed cluster and the discrete points as the abnormal data points.
Further, the analyzing the abnormality degree according to the element corresponding to the first abnormality time and the second abnormality time in the soil pressure sequence includes:
and performing anomaly degree analysis on elements corresponding to the first anomaly time and the second anomaly time in the soil pressure sequence by using an LOF anomaly detection algorithm.
Further, the constructing a two-dimensional scatter diagram according to the temperature sequence and the frequency sequence includes:
and constructing a two-dimensional coordinate space by taking temperature information as an abscissa and frequency information as an ordinate, and obtaining scattered points in the two-dimensional coordinate space according to the temperature sequence and the frequency sequence to obtain the two-dimensional scattered point diagram.
Further, the uniformly dividing at least two areas to be analyzed in the scatter diagram to be analyzed includes:
obtaining a sensor temperature range of the temperature sequence; and uniformly dividing the temperature range of the sensor into a preset number of sub-ranges, wherein the corresponding region of the sub-ranges in the scatter diagram to be analyzed is the region to be analyzed.
The invention has the following beneficial effects:
according to the invention, the fact that the soil pressure data are directly analyzed is considered to be easy to cause poor final screening effect due to the influence of the data or the influence in the data acquisition process, and the soil pressure data acquired by the soil pressure sensor can be obtained through the temperature and the frequency of the sensor, so that the temperature information, the frequency information and the soil pressure information of the soil pressure sensor are simultaneously acquired, the temperature information and the frequency information are taken as processing information in the subsequent process, and the soil pressure information to be processed abnormally is further obtained. In the processing process, the first abnormal moment is screened out for the first time through the normal temperature range, and then the scattered points in the scattered point diagram are subjected to clustering analysis by utilizing the gathering distance through analyzing the concentration and the scattered point distribution characteristics of the local area, so that the second abnormal moment corresponding to the abnormal data points is obtained. The first abnormal time is the abnormal time which is roughly screened out, the second abnormal time is the abnormal time which further considers the fluctuation and distribution of the data, and the data screening to the to-be-abnormal analysis of the soil pressure data is accurately realized through the screening and the management of the data, so that the high efficiency and the accuracy of the subsequent abnormal analysis process are ensured.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an implementation method of an intelligent management system for construction data of a building engineering according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a scatter diagram to be analyzed according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of specific implementation, structure, characteristics and effects of an intelligent management system for construction data of building engineering according to the invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the intelligent management system for construction data of building engineering provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of an implementation method of an intelligent management system for construction data of a building engineering according to an embodiment of the invention is shown. The embodiment of the invention provides an intelligent management system for construction data of building engineering, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor can realize the following method steps when executing the computer program, and a flow chart corresponding to the method steps is shown in fig. 1, and specifically comprises the following steps:
step S1: and obtaining a temperature sequence, a frequency sequence and a soil pressure sequence of the soil pressure sensor in the deep foundation pit in the construction process of the building engineering according to the preset sampling frequency in a preset sampling period.
The soil pressure of a deep foundation pit in the construction process of the building engineering is that the pit is lifted to receive lateral soil pressure from surrounding soil. The vibrating wire type soil pressure box is a common tool in the soil pressure detection process of a deep foundation pit, the vibrating wire type soil pressure box uses frequency as an output signal when detecting the soil pressure, and in order to avoid subjective errors generated by manual measurement data, the existing automatic data acquisition is realized by arranging temperature sensors in the vibrating wire type soil pressure box and converting temperature, frequency and soil pressure data through arranging a calculation chip in each sensor. Therefore, in order to avoid the influence of the data itself caused by the direct screening of the soil pressure data, the temperature data and the frequency data can be subjected to the association analysis, and the soil pressure data for anomaly detection can be obtained according to the corresponding relation of the sampling time.
Specifically, in one embodiment of the invention, a construction project including a deep foundation pit is set to 20 periods, data are collected once every one minute in each period, and 1200 times are collected, so that a corresponding temperature sequence, a corresponding frequency sequence and a corresponding soil pressure sequence are obtained. I.e. the sampling frequency is one sub-time, the sampling period is 1200 minutes and the number of sampling periods is 20. It should be noted that, in other embodiments, the specific sampling frequency, sampling period, and the number of sampling periods may be set according to the type of construction and the specific content.
It should be noted that, the temperature information, the frequency information and the soil pressure information acquired directly according to the vibrating wire type soil pressure box have units, so in order to eliminate the influence of dimension, normalization operation needs to be performed on each acquired data, the specific normalization operation is a technical means well known to those skilled in the art, in one embodiment of the present invention, the normalization is performed by using the range normalization, in other embodiments, the normalization operation may also be performed by using other modes, the specific normalization operation is not repeated, and the temperature sequence obtained by normalization is not repeatedFrequency ofSequence and sequence of earth pressureWhereinIs an index of sampling period, i.e. a periodic index of the construction of the building engineering. It should be noted that the data processing methods under different sampling periods are the same, and in the following description, only the data of the first period is used for illustration, i.e. a temperature sequence is obtainedFrequency ofSequence and sequence of earth pressureFor implementation in subsequent steps.
Step S2: screening out a first abnormal moment in the temperature sequence according to the normal temperature range; and constructing a two-dimensional scatter diagram according to the temperature sequence and the frequency sequence, deleting the scatter diagram corresponding to the first abnormal moment, and obtaining the scatter diagram to be analyzed.
The distribution of normal data in the actual data set is dense and concentrated, the distribution of abnormal data is discrete and disordered, and the data volume of the abnormal data is smaller than that of the normal data, so that in the prior art, in order to avoid a great amount of nonsensical calculation of the abnormal detection algorithm in a huge data set, the data to be analyzed can be screened out through a priori data range, and the speed and the efficiency of the abnormal detection algorithm are improved. However, considering that the construction process of the building engineering is affected by complex environment, such as electromagnetic interference of equipment, human operation errors and the like, the data to be analyzed can not be accurately screened out directly according to the prior data range, and further considering that the abnormal characteristics of the temperature data are obvious, in the embodiment of the invention, the first abnormal moment is screened out in the temperature sequence according to the normal temperature range, the screening of the first abnormal moment is realized, and the remaining data are further finely screened in the subsequent steps.
Preferably, in one embodiment of the present invention, the step of screening the first abnormal time in the temperature sequence according to the normal temperature range specifically includes:
in consideration of the fact that the difference between the temperature data in the vibrating wire type soil pressure box and the ambient temperature is small, the temperature sensor can be used for measuring an ambient temperature sequence corresponding to the ambient temperature in the sampling period, and then the ambient temperature range is obtained, namely, the maximum value and the minimum value in the ambient temperature sequence form the ambient temperature range. Based on priori knowledge, the soil temperature range of the deep foundation pit and the environment temperature range have a certain degree of difference, so that the environment temperature range is scaled according to a preset adjustment coefficient, and because the environment temperature is more stable than the soil temperature, the adjusted temperature range can be used as a normal temperature range for initially screening abnormal data, and the sampling time corresponding to elements outside the normal temperature range in the temperature sequence is used as a first abnormal time, namely, the data corresponding to the first abnormal time is regarded as abnormal data.
In one embodiment of the inventionAnd setting the preset adjustment coefficient to 2, and scaling the strategy to enlarge. If the obtained ambient temperature range isWhereinIs the lowest temperature of the environment and,the normal temperature range is the highest temperature of the environment。
And further carrying out association analysis on the temperature information and the frequency information, constructing a two-dimensional scatter diagram according to the temperature sequence and the frequency sequence, wherein a plurality of scattered points are distributed in the two-dimensional scatter diagram, each scattered point corresponds to one coordinate, each scattered point represents the temperature information and the frequency information acquired at the corresponding sampling time, namely, each scattered point simultaneously contains the temperature information and the frequency information, and therefore, the association analysis on the temperature information and the frequency information can be carried out based on the distribution of the scattered points. Because the first abnormal moment is obtained after the preliminary screening, in order to avoid the influence of the data at the first abnormal moment on the subsequent further screening, the scatter points corresponding to the first moment are deleted in the scatter points, and the scatter points to be analyzed are obtained.
Preferably, in one embodiment of the present invention, constructing a two-dimensional scatter plot according to a temperature sequence and a frequency sequence includes:
and constructing a two-dimensional coordinate space by taking the temperature information as an abscissa and the frequency information as an ordinate, and obtaining scattered points in the two-dimensional coordinate space according to the temperature sequence and the frequency sequence to obtain a two-dimensional scattered point diagram.
In another embodiment of the present invention, the frequency information may be used as an abscissa and the temperature information may be used as an ordinate to construct a two-dimensional scattergram.
It should be noted that, when constructing a two-dimensional scatter diagram, there may be two points having the same temperature frequency information at different times, that is, two points that are completely overlapped in two-dimensional coordinates, and in order to reduce the calculation amount in the subsequent analysis process of the scatter diagram to be analyzed, the completely overlapped scatter point is regarded as one scatter point.
Step S3: uniformly dividing at least two areas to be analyzed in the scatter diagram to be analyzed; obtaining the concentration degree of scattered points in each area to be analyzed; obtaining the dense distance of the area to be analyzed according to the position distribution of the scattered points in the area to be analyzed; and obtaining the gathering distance in the scatter diagram to be analyzed according to the gathering degree and the dense distance of all the areas to be analyzed.
Because the temperature data and the frequency data collected at the same moment are abnormal data only if one of the temperature data and the frequency data is abnormal data, the soil pressure data corresponding to the moment is abnormal data, and therefore the scattering points can be further screened through the concentrated and discrete properties of the distribution of the scattering points in the scattering point diagram to be analyzed.
Because the data can present certain distribution characteristics in a certain local range in the construction process of the building engineering, for the accuracy of data analysis, the scatter diagram is uniformly divided into at least two areas to be analyzed, each area to be analyzed is analyzed, the analysis of local scatter distribution is realized, reference data is provided for the whole analysis in the subsequent steps, and the data obtained by the whole analysis is more accurate.
Preferably, one embodiment of the present invention considers that the temperature is continuously changed with time in the same sampling period, that is, the temperature of the soil is changed with the operation of the construction, and the frequency of the vibrating wire type soil pressure box is changed less with time. Therefore, in the same smaller temperature range, the normal soil pressure value obtained by the vibrating wire type soil pressure box is approximate, namely, the scatter diagram to be analyzed can be segmented based on temperature. The sensor temperature range of the temperature sequence is obtained, i.e. the sensor temperature range is formed with the maximum and minimum values of the temperature sequence. Dividing the temperature range of the sensor into a preset number of sub-ranges, wherein the corresponding region of the sub-ranges in the scatter diagram to be analyzed is the region to be analyzed.
Referring to fig. 2, a schematic diagram of a scatter diagram to be analyzed according to an embodiment of the invention is shown. The abscissa of the scatter plot to be analyzed in one embodiment of the invention is temperatureT, the ordinate is the frequency F, the number of the sub-ranges is set to be 5, namely 5 range sections exist in the abscissa, the end point value of each range section is a dividing line, a scattered point area in a scattered point diagram to be analyzed is divided into 5 areas to be analyzed, and the dividing lines in FIG. 2 are respectively:、、、、and. If the scattered points are on the dividing line, the corresponding scattered points are classified into the region on the left side of the dividing line.
Because the temperature data and the frequency data do not change greatly corresponding to normal data in a certain temperature range, namely the scattered point distribution is concentrated, and the temperature difference and the frequency difference corresponding to the scattered point are small. The scattered points of the abnormal data are scattered, and the scattered points of the abnormal data are free from the scattered point gathering position of the normal data. It is therefore necessary to first obtain the concentration degree of the scattered points in each area to be analyzed in order to analyze the distribution of the scattered points.
In one embodiment of the invention, the concentration degree of the area to be analyzed can be obtained directly according to the distance variance between the scattered point coordinates. Preferably, in one embodiment of the present invention, it is considered that the discrete point degree in one area is affected by two variables of x-axis coordinates and y-axis coordinates, and the mean value of the temperature value and the mean value of the frequency value between different areas are not necessarily the same, so that the discrete amplitude of the data points of different areas cannot be compared directly by using the variance, and therefore, the product of the standard deviation and the polar difference of the coordinates in each dimension in the area to be analyzed is taken as the discrete feature of the corresponding dimension. Integral discrete features are obtained from the discrete features in two dimensions. The degree of concentration is obtained from the overall discrete features. The overall discrete features are inversely related to the degree of concentration.
As one example, the calculation formula of the concentration degree includes:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the firstThe concentration of the individual areas to be analyzed,is the standard deviation of the coordinates in the x dimension,is the very bad in the x-dimension,is the standard deviation of the coordinates in the y-dimension,being the very poor in the y-dimension,is a natural constant.
The product of the standard deviation and the polar difference is used as the discrete feature of the corresponding dimension in the calculation formula of the concentration degree, namely, the larger the standard deviation is, the larger the discrete degree of the scattered points in the region is; the larger the pole difference, the larger the discrete magnitude of the scattered points in the area. Therefore, the larger the value of the discrete feature is, the more scattered the data in the corresponding dimension is, and the whole discrete degree is obtained by combining the discrete features in the two dimensions through addition and further mapping through an exponential function based on a natural constant. Because both discrete features are greater than 0, the inverse of the overall degree of discrete is taken as the degree of concentration, i.e., an overall degree of discrete negative correlation mapping is achieved, and the range of mapped values is limited to between 0 and 1.
In other embodiments of the present invention, other basic operations such as multiplication may be used to obtain the integral discrete features, or a decreasing function mapping method may be used to map the integral discrete features in a negative correlation manner to obtain the concentration degree, which is not limited herein.
Further, according to the position distribution of the scattered points in the area to be analyzed, the dense distance of the area to be analyzed is obtained, the dense distance represents the distribution range of the scattered points which are intensively distributed in the area to be analyzed, and the distribution characteristics of the scattered points corresponding to the area to be analyzed can be commonly represented with the concentration degree by obtaining the dense distance.
Preferably, the obtaining the dense distance of the area to be analyzed according to the position distribution of the scattered points in the area to be analyzed according to one embodiment of the present invention includes:
clustering the scattered points in the area to be analyzed according to the coordinate information to obtain at least two first cluster clusters, wherein the first cluster with the largest sample number in the cluster is used as a target cluster, namely the target cluster is the scattered point cluster intensively distributed in the area to be analyzed. One dimension is optionally used as the target dimension. And in the target cluster, obtaining non-target dimension coordinate differences between scattered points of two identical target dimension coordinates, and forming a first reference point pair by two scattered points corresponding to the largest non-target dimension coordinate differences. And obtaining the number of scattered points on the connecting line of the two points in the first reference point pair as the reference number. If there is more than one first reference point pair, the average number of scattered points on all the lines is used as the reference number. The ratio of the maximum non-target dimension coordinate difference to the reference number is taken as the initial dense distance in the target dimension, because the maximum non-target dimension coordinate difference represents the maximum distance of the corresponding target cluster in the non-target dimension, the larger the maximum distance is, the larger the initial dense distance is, because the data of two different moments but the same temperature frequency data can be regarded as a scattered point, and the analysis is performed in the target cluster with more samples, so that the smaller the reference number is, the more aggregated the data distribution between the first reference points is, namely the smaller the reference number is, the larger the initial dense distance is. The target dimensions are changed to obtain an initial dense distance in each dimension. The maximum initial dense distance between the two dimensions is taken as the dense distance of the region to be analyzed. The formula for the dense distance is:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the firstThe dense distance of the individual areas to be analyzed,for the maximum non-target dimension coordinate difference with the y-dimension as the target dimension,for the reference number in the y dimension as the target dimension,for the maximum non-target dimension coordinate difference with the x dimension as the target dimension,for the reference number in the y dimension as the target dimension,the function is chosen for the maximum value.
In one embodiment of the invention, the clustering method of the first cluster uses a K-Means clustering algorithm, where K is set to 3. The K-Means clustering algorithm is a technical Means well known to those skilled in the art, and will not be described in detail herein.
The concentration degree and the dense distance jointly represent the distribution characteristics of the scatter points in the to-be-analyzed area, namely the local analysis of the to-be-analyzed scatter point diagram is realized, the local analysis results are further integrated to obtain the whole analysis result, namely the concentration distance in the to-be-analyzed scatter point diagram is obtained according to the concentration degree and the dense distance of all the to-be-analyzed areas.
Preferably, the method for acquiring the specific aggregation distance in one embodiment of the present invention includes:
the duty ratio of the concentration degree of each area to be analyzed in the concentration degree of all the areas to be analyzed is taken as a weight coefficient. I.e. all weight coefficients in the scatter diagram to be analyzed are added to be 1. Multiplying the weight coefficient by the dense distance of the corresponding area to be analyzed to obtain the weighted dense distance. And accumulating the weighted dense distances of all the areas to be analyzed to obtain an aggregate distance. The formula of the aggregate distance is expressed as:
wherein, the liquid crystal display device comprises a liquid crystal display device,in order to gather the distance of the distance,for the number of areas to be analyzed,is the firstThe dense distance of the individual areas to be analyzed,is the firstThe concentration of the individual areas to be analyzed,is the firstConcentration of the individual areas to be analyzed.
And (3) taking the duty ratio of the concentration degree as a weight to adjust the dense distance of the corresponding region to be analyzed, namely, the larger the concentration degree is, the larger the reference of the corresponding dense distance as the concentrated distance is, and obtaining the concentrated distance representing the whole of the scatter diagram to be analyzed through weighted summation.
Step S4: clustering scattered points in the scattered points by taking the aggregation distance as a clustering radius in a DBSCAN clustering algorithm, and screening out abnormal data points; the sampling time of the abnormal data point is taken as a second abnormal time.
Because the aggregation distance represents the aggregation characteristic of the whole scattered points in the scattered points to be analyzed, namely, more scattered points are aggregated in the range of the aggregation distance, the aggregation distance is used as the clustering radius in the DBSCAN clustering algorithm, a classification result with accurate classification can be obtained, and abnormal data points can be screened out according to the clustering result, and the method specifically comprises the following steps:
and clustering the scattered points in the scattered points by taking the aggregation distance as a clustering radius in a DBSCAN clustering algorithm to obtain a second clustering cluster and the scattered points. Because the scattered points corresponding to the normal data are larger in number of samples, the second cluster with the largest number of samples is used as the cluster to be analyzed, namely, other clusters which are not to be analyzed are clusters with the scattered points corresponding to the abnormal data. It should be noted that, because the abnormal data is a small amount of data, the non-to-be-analyzed cluster is a cluster composed of a small amount of scattered points.
Taking the distance between the sample points in the cluster to be analyzed and the cluster center as a judgment distance; and obtaining the average judgment distance in the cluster to be analyzed, and taking the sample point corresponding to the judgment distance larger than the average judgment distance as an abnormal sample. Because the cluster to be analyzed is a cluster with a large number of normal data scattered points aggregated, the abnormal samples in the cluster are regarded as samples which deviate from the fluctuation of the normal data range, and further abnormality detection and analysis are needed to be carried out to judge whether the abnormal samples really occur. Thus, abnormal samples, samples in non-to-be-analyzed clusters and discrete points are taken as abnormal data points.
In one embodiment of the invention, the minimum number of neighbors in the DBSCAN clustering algorithm is 2.
Taking the sampling time of the abnormal data point as a second abnormal time, namely, the screened second abnormal time is the result after the data is finely screened.
Step S5: and carrying out abnormality degree analysis on elements corresponding to the first abnormality time and the second abnormality time in the soil pressure sequence.
The first abnormal time and the second abnormal time are obtained through the correlation analysis of the temperature data and the frequency data, and the soil pressure data corresponding to the first abnormal time and the second abnormal time are abnormal soil pressure or suspected abnormal soil pressure, so that the elements corresponding to the first abnormal time and the second abnormal time can be subjected to the abnormality analysis, and the real abnormal state can be judged according to the abnormality analysis result. Through screening and managing the data, the data quantity in the abnormality detection process is effectively reduced, and meanwhile, the referential property of the data to be analyzed is improved.
Preferably, an embodiment of the present invention uses an LOF anomaly detection algorithm to analyze the anomaly degree of the element corresponding to the first anomaly time and the second anomaly time in the soil pressure sequence. I.e. a larger LOF abnormality factor obtained indicates a greater degree of abnormality. It should be noted that, the LOF anomaly detection algorithm is a technical means well known to those skilled in the art, and will not be described herein. In other embodiments, the anomaly degree analysis may also be performed by other anomaly detection algorithms such as COF, isolated forests, and the like.
In summary, the embodiment of the invention collects temperature, frequency and soil pressure information of the vibrating wire type soil pressure box together. And performing primary data screening on the temperature sequence according to the normal temperature range to obtain a first abnormal moment, and further performing correlation analysis on the temperature information and the frequency information to obtain a scatter diagram to be analyzed. The aggregation distance for clustering is obtained by integrating the areas to be analyzed in the scatter diagram to be analyzed after independent analysis, so that an accurate clustering result is obtained, abnormal data points are screened out according to the clustering result, and further a second abnormal moment is obtained. And taking the soil pressure data corresponding to the first abnormal time and the second abnormal time as the screened data to perform abnormality degree analysis. According to the embodiment of the invention, the secondary screening of the data is realized through the correlation analysis of the temperature and the frequency, and the accurate data to be analyzed abnormally is obtained through the screening and the management of the data.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (7)
1. An intelligent management system for construction data of building engineering, comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the following method steps are realized when the processor executes the computer program:
obtaining a temperature sequence, a frequency sequence and a soil pressure sequence of a vibrating wire type soil pressure box in a deep foundation pit in the construction process of the building engineering in a preset sampling period according to a preset sampling frequency;
screening out a first abnormal moment in the temperature sequence according to a normal temperature range; constructing a two-dimensional scatter diagram according to the temperature sequence and the frequency sequence, deleting the scatter diagram corresponding to the first abnormal moment, and obtaining the scatter diagram to be analyzed;
uniformly dividing at least two areas to be analyzed in the scatter diagram to be analyzed; obtaining the concentration degree of scattered points in each area to be analyzed; obtaining the dense distance of the area to be analyzed according to the position distribution of the scattered points in the area to be analyzed; acquiring the gathering distance in the scatter diagram to be analyzed according to the gathering degree and the dense distance of all the areas to be analyzed;
clustering scattered points in the scattered points by taking the aggregation distance as a clustering radius in a DBSCAN clustering algorithm, and screening abnormal data points; taking the sampling time of the abnormal data point as a second abnormal time;
performing anomaly degree analysis on elements corresponding to the first anomaly time and the second anomaly time in the soil pressure sequence;
the obtaining the concentration degree of the scattered points in each area to be analyzed comprises the following steps:
taking the product of standard deviation and polar difference of coordinates in each dimension in the region to be analyzed as the discrete feature of the corresponding dimension; obtaining integral discrete features according to the discrete features in two dimensions; obtaining the concentration degree according to the integral discrete features; the integral discrete features and the concentration degree form a negative correlation relation;
the obtaining the dense distance of the area to be analyzed according to the position distribution of the scattered points in the area to be analyzed comprises the following steps:
clustering scattered points in the area to be analyzed according to the coordinate information to obtain at least two first cluster clusters, wherein the first cluster with the largest sample number in the cluster is used as a target cluster;
optionally one dimension as a target dimension; in the target cluster, non-target dimension coordinate differences between scattered points of two identical target dimension coordinates are obtained, and a first reference point pair is formed by two scattered points corresponding to the largest non-target dimension coordinate difference; obtaining the number of scattered points on the connecting line of the two points in the first reference point pair as the reference number; if more than one first reference point pair is provided, taking the average number of scattered points on all the connecting lines as the reference number; taking the ratio of the maximum non-target dimension coordinate difference to the reference number as an initial dense distance in a target dimension;
changing the target dimension to obtain the initial dense distance in each dimension;
the maximum initial dense distance between the two dimensions is taken as the dense distance of the region to be analyzed.
2. The intelligent management system for construction data of building engineering according to claim 1, wherein the step of screening out the first abnormal time in the temperature sequence according to the normal temperature range comprises the steps of:
obtaining an environmental temperature sequence in a preset sampling period, and obtaining an environmental temperature range according to the environmental temperature sequence; scaling the environment temperature range according to a preset adjustment coefficient to obtain the normal temperature range; and taking the sampling time corresponding to the element outside the normal temperature range in the temperature sequence as a first abnormal time.
3. The intelligent management system for construction data of construction engineering according to claim 1, wherein the obtaining the concentration distance in the scatter diagram to be analyzed according to the concentration degree and the dense distance of all the areas to be analyzed comprises:
taking the duty ratio of the concentration degree of each region to be analyzed in the concentration degree of all the regions to be analyzed as a weight coefficient; multiplying the weight coefficient by the dense distance of the corresponding area to be analyzed to obtain a weighted dense distance; and accumulating the weighted dense distances of all the areas to be analyzed to obtain the aggregate distance.
4. The intelligent management system for construction data of building engineering according to claim 1, wherein the screening out abnormal data points comprises:
clustering scattered points in the scattered point map by taking the aggregation distance as a clustering radius in a DBSCAN clustering algorithm to obtain a second cluster and discrete points; taking the second cluster with the largest sample number as a cluster to be analyzed; taking the distance between the sample points in the cluster to be analyzed and the cluster center as a judgment distance; obtaining an average judgment distance in the cluster to be analyzed, and taking a sample point corresponding to the judgment distance larger than the average judgment distance as an abnormal sample; and taking the abnormal samples, the samples in the non-to-be-analyzed cluster and the discrete points as the abnormal data points.
5. The intelligent management system for construction data of construction engineering according to claim 1, wherein the performing the abnormality degree analysis on the elements corresponding to the first abnormality time and the second abnormality time in the sequence of earth pressures includes:
and performing anomaly degree analysis on elements corresponding to the first anomaly time and the second anomaly time in the soil pressure sequence by using an LOF anomaly detection algorithm.
6. The intelligent management system for construction data of building engineering according to claim 1, wherein the constructing a two-dimensional scatter diagram according to the temperature sequence and the frequency sequence comprises:
and constructing a two-dimensional coordinate space by taking temperature information as an abscissa and frequency information as an ordinate, and obtaining scattered points in the two-dimensional coordinate space according to the temperature sequence and the frequency sequence to obtain the two-dimensional scattered point diagram.
7. The intelligent management system for construction data of building engineering according to claim 1, wherein the uniformly dividing at least two areas to be analyzed in the scatter diagram to be analyzed comprises:
obtaining a sensor temperature range of the temperature sequence; and uniformly dividing the temperature range of the sensor into a preset number of sub-ranges, wherein the corresponding region of the sub-ranges in the scatter diagram to be analyzed is the region to be analyzed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310457096.3A CN116186634B (en) | 2023-04-26 | 2023-04-26 | Intelligent management system for construction data of building engineering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310457096.3A CN116186634B (en) | 2023-04-26 | 2023-04-26 | Intelligent management system for construction data of building engineering |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116186634A CN116186634A (en) | 2023-05-30 |
CN116186634B true CN116186634B (en) | 2023-07-07 |
Family
ID=86449276
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310457096.3A Active CN116186634B (en) | 2023-04-26 | 2023-04-26 | Intelligent management system for construction data of building engineering |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116186634B (en) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116502107B (en) * | 2023-06-21 | 2023-09-01 | 杭州端点网络科技有限公司 | Data intelligent operation platform supply chain data processing system based on artificial intelligence |
CN116738353B (en) * | 2023-08-15 | 2023-10-13 | 安拓思纳米技术(苏州)有限公司 | Pharmaceutical workshop air filter element performance detection method based on data analysis |
CN116975771B (en) * | 2023-09-25 | 2023-12-08 | 苏州保邦电气有限公司 | Automatic abnormality identification method and system for motor production |
CN116992322B (en) * | 2023-09-25 | 2024-01-16 | 广东申创光电科技有限公司 | Smart city data center management system |
CN116993229B (en) * | 2023-09-25 | 2023-12-19 | 山东高速工程检测有限公司 | Digital management method for construction quality of cross-sea bridge pile foundation |
CN117236084B (en) * | 2023-11-16 | 2024-02-06 | 青岛永强木工机械有限公司 | Intelligent management method and system for woodworking machining production |
CN117421618B (en) * | 2023-11-24 | 2024-04-05 | 上海东方低碳科技产业股份有限公司 | Building energy consumption monitoring method and system |
CN117313222B (en) * | 2023-11-29 | 2024-02-02 | 青岛东捷建设集团有限公司 | Building construction data processing method based on BIM technology |
CN117522350A (en) * | 2024-01-04 | 2024-02-06 | 深圳市毅霖建设集团有限公司 | Intelligent management method and system for green architectural design and decoration engineering |
CN117540238B (en) * | 2024-01-05 | 2024-03-22 | 长春同泰企业管理服务有限责任公司 | Data security management method for industrial digital information acquisition device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018141663A (en) * | 2017-02-27 | 2018-09-13 | 日本電信電話株式会社 | Condition measurement device, condition measurement system, and condition measurement method |
CN111614690A (en) * | 2020-05-28 | 2020-09-01 | 上海观安信息技术股份有限公司 | Abnormal behavior detection method and device |
CN114004512A (en) * | 2021-11-04 | 2022-02-01 | 西安热工研究院有限公司 | Multi-unit operation state outlier analysis method and system based on density clustering |
CN115950557A (en) * | 2023-03-08 | 2023-04-11 | 深圳市特安电子有限公司 | Intelligent temperature compensation method based on pressure transmitter |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6816811B2 (en) * | 2001-06-21 | 2004-11-09 | Johnson Controls Technology Company | Method of intelligent data analysis to detect abnormal use of utilities in buildings |
CN109977996B (en) * | 2019-02-12 | 2021-04-02 | 华北水利水电大学 | Hydraulic structure running state monitoring system based on time series clustering fusion |
CN114116689A (en) * | 2021-10-25 | 2022-03-01 | 浙江瑞邦科特检测有限公司 | Big data cleaning method based on building structure safety monitoring |
KR20220122923A (en) * | 2021-11-29 | 2022-09-05 | 주식회사 알엠에이 | System for monitoring and acquring data in smart factory |
CN115575007A (en) * | 2022-09-15 | 2023-01-06 | 大连理工大学 | Soil pressure and temperature monitoring and early warning method for soil covering tank based on digital twinning technology |
-
2023
- 2023-04-26 CN CN202310457096.3A patent/CN116186634B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018141663A (en) * | 2017-02-27 | 2018-09-13 | 日本電信電話株式会社 | Condition measurement device, condition measurement system, and condition measurement method |
CN111614690A (en) * | 2020-05-28 | 2020-09-01 | 上海观安信息技术股份有限公司 | Abnormal behavior detection method and device |
CN114004512A (en) * | 2021-11-04 | 2022-02-01 | 西安热工研究院有限公司 | Multi-unit operation state outlier analysis method and system based on density clustering |
CN115950557A (en) * | 2023-03-08 | 2023-04-11 | 深圳市特安电子有限公司 | Intelligent temperature compensation method based on pressure transmitter |
Non-Patent Citations (3)
Title |
---|
Multiplexing of LPFG resonant wavelengths for abnormal reaction detection in large-scale plants by distributed high temperature monitoring;Yutaka Katsuyama 等;IEEE;第1-4页 * |
基于K-均值聚类的工业异常数据检测;张仁斌;许辅昊;刘飞;李思娴;;计算机应用研究(07);第2180-2184页 * |
聚类分析在桥梁监测异常数据处理中的应用;李西芝;胡靖;;黑龙江交通科技(12);第88-90、92页 * |
Also Published As
Publication number | Publication date |
---|---|
CN116186634A (en) | 2023-05-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116186634B (en) | Intelligent management system for construction data of building engineering | |
CN110287932B (en) | Road blocking information extraction method based on deep learning image semantic segmentation | |
CN104702680B (en) | A kind of monitoring resource method and system | |
CN108667684B (en) | Data flow anomaly detection method based on local vector dot product density | |
CN110134907B (en) | Rainfall missing data filling method and system and electronic equipment | |
CN110059713A (en) | Precipitation type identification method based on precipitation particle multi-feature parameters | |
Sun et al. | Remote estimation of grafted apple tree trunk diameter in modern orchard with RGB and point cloud based on SOLOv2 | |
CN113139880A (en) | Wind turbine generator actual power curve fitting method, device, equipment and storage medium | |
CN111967717A (en) | Data quality evaluation method based on information entropy | |
CN116416884A (en) | Testing device and testing method for display module | |
CN116739619A (en) | Energy power carbon emission monitoring analysis modeling method and device | |
CN116738261A (en) | Numerical characteristic discretization attribution analysis method and device based on clustering and binning | |
CN102901697B (en) | Porosity detection method for soil | |
CN114217025B (en) | Analysis method for evaluating influence of meteorological data on air quality concentration prediction | |
CN113890833B (en) | Network coverage prediction method, device, equipment and storage medium | |
CN111612734B (en) | Background clutter characterization method based on image structure complexity | |
Li et al. | Automatic identification of modal parameters for high arch dams based on SSI incorporating SSA and K-means algorithm | |
CN113139337A (en) | Partitioned interpolation processing method and device for lake terrain simulation | |
KR20220123845A (en) | Meathod and device for measuring similarity between time series data | |
CN117493921B (en) | Artificial intelligence energy-saving management method and system based on big data | |
CN109931987A (en) | A kind of intelligent vegetable planting machine environment based on cloud precisely monitors system and method | |
CN113098963B (en) | Processing and analyzing system for cloud computing of Internet of things | |
CN115902814B (en) | Method and device for evaluating performance of target recognition model based on information space measurement | |
CN117540325A (en) | Business database anomaly detection method and system based on data variation capture | |
Aris et al. | Application of Mahalanobis-Taguchi system in Rainfall Distribution |
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 |