CN113762399A - Method for collecting and visually presenting time-space correlation monitoring data of gravity dam - Google Patents
Method for collecting and visually presenting time-space correlation monitoring data of gravity dam Download PDFInfo
- Publication number
- CN113762399A CN113762399A CN202111063482.1A CN202111063482A CN113762399A CN 113762399 A CN113762399 A CN 113762399A CN 202111063482 A CN202111063482 A CN 202111063482A CN 113762399 A CN113762399 A CN 113762399A
- Authority
- CN
- China
- Prior art keywords
- space
- time
- model
- spatial
- correlation
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012544 monitoring process Methods 0.000 title claims abstract description 41
- 230000005484 gravity Effects 0.000 title claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 35
- 238000005070 sampling Methods 0.000 claims abstract description 27
- 238000007405 data analysis Methods 0.000 claims abstract description 10
- 230000000007 visual effect Effects 0.000 claims abstract description 10
- 238000005516 engineering process Methods 0.000 claims description 29
- 238000012800 visualization Methods 0.000 claims description 20
- 241000854291 Dianthus carthusianorum Species 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 9
- 238000001134 F-test Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000009826 distribution Methods 0.000 claims description 6
- 238000003909 pattern recognition Methods 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 5
- 238000012300 Sequence Analysis Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 230000001934 delay Effects 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 2
- 238000011156 evaluation Methods 0.000 abstract description 16
- 238000004458 analytical method Methods 0.000 abstract description 12
- 230000000875 corresponding effect Effects 0.000 description 8
- 238000005065 mining Methods 0.000 description 7
- 238000011160 research Methods 0.000 description 7
- 238000006073 displacement reaction Methods 0.000 description 6
- 230000004927 fusion Effects 0.000 description 5
- 230000002596 correlated effect Effects 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000007476 Maximum Likelihood Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/005—Routing actions in the presence of nodes in sleep or doze mode
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Abstract
The invention relates to a method for acquiring and visually presenting space-time correlation monitoring data of a gravity dam, which is characterized by comprising a space-time correlation-based wireless sensing data acquisition algorithm and a space-time correlation-based data analysis algorithm; the wireless sensing data acquisition algorithm based on the space-time correlation is a time domain sampling frequency adjustment algorithm which is adopted on the basis of considering the space correlation; the data analysis algorithm based on the space-time correlation comprises a space-time autoregressive moving average model and a space-time autoregressive model. The invention covers the monitoring data of multiple parts, multiple measuring points and multiple types of gravity dams, and performs visual comprehensive analysis and rapid evaluation on the working state of the key parts of the gravity dam in real time, so that the dam safety evaluation is shifted to multi-point and multi-project multi-dimensional comprehensive evaluation from the traditional independent analysis and evaluation of single measuring point.
Description
Technical Field
The invention belongs to the technical field of water conservancy and hydropower engineering, and particularly belongs to the technical field of a method for acquiring and visually presenting space-time correlation monitoring data of a gravity dam.
Background
In the traditional analysis process of the measured monitoring data of the gravity dam, a monitoring sequence of a single measuring point is usually modeled, which means that modeling needs to be carried out on all measuring points, and a large amount of model redundancy is caused. The dam monitoring data have time correlation, the dam serves as an integral structure, displacement among monitoring points is correlated, namely space correlation exists, therefore, a visualization technology of a geometric attribute relation of the gravity dam, data time sequence characteristics and dam characteristic parameters based on a space-time correlation technology is constructed, working conditions of all parts of the large dam in a drainage basin are comprehensively analyzed and quickly evaluated in real time, and dam safety evaluation is promoted to be developed from independent analysis evaluation of traditional single-measuring-point to multi-measuring-point and multi-item multi-dimensional comprehensive evaluation.
At present, most of research of the time-space association technology focuses on the aspect of improving the performance of an association rule algorithm, and research on the aspects of accurate expression and application of an association rule result is less, but the problem that the understanding and application of a user on the association rule result are not negligible is solved, and the research is directly related to the value of a result generated in an early mining stage in practical application. Because the visualization technology fully utilizes the image expression capability of the pattern image, the method responds to the sharp perception of human to color and shape, and enables users to observe and analyze more conveniently and effectively. The visualization technology is applied to the association rule mining to display results, and the method is a new field of the mining research of the time-space association technology.
Disclosure of Invention
The invention provides a wireless sensing acquisition and visual presentation method for gravity dam monitoring data, aiming at solving the defects of the problems.
The invention is realized by adopting the following technical scheme.
A method for collecting and visually presenting space-time correlation monitoring data of a gravity dam comprises a space-time correlation-based wireless sensing data collection algorithm and a space-time correlation-based data analysis algorithm;
the wireless sensing data acquisition algorithm based on the space-time correlation is a time domain sampling frequency adjustment algorithm which is adopted on the basis of considering the space correlation;
the data analysis algorithm based on the space-time correlation comprises a space-time autoregressive moving average model and a space-time autoregressive model.
The wireless sensing data acquisition algorithm based on the time-space association is that initial aggregation nodes are searched in a network based on a minimum spanning tree, then clustering is carried out by utilizing a K mean value, and the whole network is divided into a plurality of clusters; selecting a cluster head for the nodes in the cluster according to the weight; sampling by nodes in the cluster in turn, and enabling the nodes which are not sampled to enter a sleep state; the cluster head nodes in each cluster determine the sampling frequency of all member nodes in the cluster according to the history model of the sensing data and the change condition of the current sampling data; and finally, taking the sink node as a root node, forming a minimum spanning tree by the cluster head nodes, and sending information to the sink node in a multi-hop manner.
The invention relates to a wireless sensing data acquisition algorithm based on space-time correlation, which comprises the following steps:
step 1) initializing a known network topology structure;
step 2) selecting an initial convergence vertex in a network area based on a minimum spanning tree algorithm;
step 3) clustering nodes in the network by using a K-means algorithm;
step 4), selecting the node with the most residual energy as a cluster head node in each K-means cluster, and establishing a minimum spanning tree of all cluster head nodes by taking the sink node as a root node;
step 5), selecting m nodes to work in each round according to the residual energy, continuously adjusting sampling intervals according to a time sequence analysis method in the sampling process, keeping event monitoring, increasing the value of m when an event is monitored, and keeping the value of m when the event is stable;
predicting the value of Rt by adopting a first-order AR prediction model according to the requirements of the accuracy and the speed of the stable data prediction and considering the limitation of calculation and storage;
in the formula, constant
Representing the use of Yule-Walker estimation, i.e.
Finally, a prediction formula based on a minimum mean square error form is given as
The space-time autoregressive moving average model provided by the invention is characterized in that the overall monitoring physical quantity of a plurality of space points is analyzed and forecasted by considering the associated information among observation points, and for a stable space-time sequence, the form of the space-time autoregressive moving average model is as follows:
in the formula, alphatIs random error and follows normal distribution; p and q are respectively time autoregressive and moving average order;
mk、nkthe spatial autoregressive order and the moving average order are respectively the time delay k; w (h) is a weight matrix of h-order spatial delays, which represents spatial h-order adjacency, and w (0) ═ E.
The space-time autoregressive model of the invention is that the space order is not changed along with the time order p,
the spatial weight matrix reflects the influence between the adjacent points in the space, and for the area with regular distribution, the spatial weight matrix is generally determined according to the spatial proximity, namely, the element corresponding to the point adjacent to the target point is 1, and the other corresponding elements are zero; the influence of the spatial weight matrix is inconsistent due to different numbers of adjacent points of each spatial point, so as to ensure that the total influence of the adjacent points of each spatial point is consistent; and (3) carrying out standardization processing on the spatial weight array:
in the formula, ωijIs the element of the ith row and the jth column of the normalized pre-space weight array; omegaiIs the sum of the elements of the ith row of the spatial weight matrix; wijThe normalized spatial weight array elements.
The space-time autoregressive model comprises the following three steps: step 1) pattern recognition; step 2), parameter estimation; and 3) carrying out model inspection.
The method comprises the following steps that 1) pattern recognition is carried out, wherein the type of a model is determined through the truncation of an autocorrelation coefficient and a partial autocorrelation coefficient; the common method for determining the order of the model is a residual sum of squares criterion, a white noise test criterion, an Akaike information criterion and an F-criterion; among the widely used Akaike information criteria are
AIC＝lnσ2+p/N (7)
BIC＝lnσ2+plnN/N (8)
In the formula, σ2Is the variance of the residual; p is the order of the model; and N is the data length.
The invention selects BIC to determine the order of the model.
The parameter estimation in the step 2) is that a least square estimation is adopted by a space-time autoregressive model.
The step 3) model of the invention is verified as,
checking whether the established model is suitable or not, and checking parameters; by constructing the global statistics of the model:
wherein N is the sample length; p is the number of parameters; generally, the significance level alpha of F-test is given to be 0.05, and the model is subjected to overall significance test; the F-test can only show that the model is significant in general, but cannot guarantee that each parameter of the model is significant, so that the significance test needs to be carried out on each parameter; by constructing statistics:
in the formula, qiAs a parameter
Co-factor of (c);
is the variance; the significance of each parameter is tested through T-test, so that a final space-time autoregressive model is determined; and predicting the space-time sequence according to a model established by BIM, and then processing the obtained data by combining a visualization technology to obtain corresponding pictures and tables, so that the data is perceived more intuitively.
Compared with the prior art, the invention has the beneficial effects that:
(1) the existing association rule visualization technology mainly comprises the following steps: and the visualization technologies comprise an association rule table, a two-dimensional matrix, a directed graph, a parallel coordinate, a three-dimensional coordinate, map display and the like. However, each technology has corresponding advantages and disadvantages, and based on the analysis of the rule visualization technology, most of the technologies cannot be popularized to the visualization expression of the spatio-temporal association rule due to the defects of the technologies, but the association rule table visualization technology and the map visualization technology can well express the spatio-temporal association rule.
A spatio-temporal association rule contains at least 6 parameters: the rule is premised on that the result of the rule, and the spatial correlation among the item sets in the rule, namely the spatial predicate, the time constraint, the support degree of the rule and the confidence degree of the rule, so that a visual graph of the mining result of one space-time correlation rule at least comprises the six parameters. In the method, the space-time association rule is expressed by combining the table visualization technology and the map visualization technology.
(2) At present, most of the research on association rules focuses on the aspect of improving the performance of an association rule algorithm, and the research on the aspects of accurate expression and application of association rule results is less, but the data analysis and presentation visualization technology is applied to association rule mining and used for displaying the results, so that the method is a new field of association rule mining research. Because the visualization technology fully utilizes the image expression capability of the pattern image, the method responds to the sharp perception of human to color and shape, and enables users to observe and analyze more conveniently and effectively. Therefore, the understanding and the application of the association rule result by the user are enhanced, and the value of the result generated in the early mining stage in the actual application is better.
The invention is further explained below with reference to the drawings and the detailed description.
Drawings
Fig. 1 is a diagram of an intelligent integration system for monitoring dam safety.
Detailed Description
The invention discloses a visualization technology of a geometric attribute relation, a data time sequence characteristic and a dam characteristic parameter of a gravity dam based on a space-time correlation technology. The technology constructs a space-time data analysis system, the system covers multiple parts, multiple measuring points and multiple types of monitoring data of the gravity dam, visual comprehensive analysis and rapid evaluation are carried out on the working performance of the key parts of the gravity dam in real time, and the dam safety evaluation is changed from the traditional independent analysis evaluation of single measuring point to the multi-dimensional comprehensive evaluation of multiple measuring points and multiple projects.
The gravity dam space-time correlation monitoring data analysis system as shown in fig. 1 takes forward thinking, reverse thinking, system thinking and feedback thinking as core thinking modes; according to the characteristics of the dam safety monitoring field, intelligent fusion is divided into four levels of data level fusion, analysis level fusion, diagnosis level fusion and evaluation level fusion, so that the safety of the gravity dam is judged and predicted. And the realization of the steps requires modeling and data collection processing. In the traditional analysis process of the actually measured monitoring data of the gravity dam, a monitoring sequence of a single measuring point is modeled generally, and under the real condition, the monitoring data of the gravity dam not only has time correlation, but also the gravity dam is taken as an integral structure, and the displacement among the monitoring points is correlated, namely, space correlation exists, so that the geometric attribute relation, the data time sequence characteristic and the dam characteristic parameter of the gravity dam are visualized by adopting a space-time correlation technology, and the working state of each part of the drainage basin dam is comprehensively analyzed and rapidly evaluated in real time.
1) Basic technology and principle of wireless sensing data acquisition algorithm based on space-time correlation
In this technique, wireless sensor nodes are grouped into disjoint sets, with the sampling and transmission of each group being managed by a designated clusterhead. Most of clustering protocols of the wireless sensor network are top-down methods, the overall situation of the wireless sensor network is firstly obtained, and the upper layer of clustering is established by selecting certain nodes as cluster heads. The remaining nodes of each group then go to the cluster member of the designated cluster.
The specific process of the algorithm is as follows: searching an initial aggregation node in the network based on the minimum spanning tree, then clustering by using a K mean value, and dividing the whole network into a plurality of clusters; selecting a cluster head for the nodes in the cluster according to the weight; sampling by nodes in the cluster in turn, and enabling the nodes which are not sampled to enter a sleep state; and the cluster head nodes in each cluster determine the sampling frequency (namely the next sampling moment) of all member nodes in the cluster according to the historical model of the sensing data and the change condition of the current sampling data. And finally, taking the sink node as a root node, forming a minimum spanning tree by the cluster head nodes, and sending information to the sink node in a multi-hop manner. The algorithm is a time domain sampling frequency adjustment algorithm which is adopted on the basis of considering spatial correlation, and is a space-time correlated sampling control algorithm.
The algorithm process is as follows:
(1) the network topology structure is known, and initialization is carried out;
(2) selecting an initial convergence vertex in a network area based on a minimum spanning tree algorithm;
(3) clustering nodes in the network by using a K mean algorithm;
(4) selecting the node with the most residual energy as a cluster head node in each K-means cluster, and establishing a minimum spanning tree of all cluster head nodes by taking the sink node as a root node;
(5) and selecting m nodes to work in each round according to the residual energy, continuously adjusting the sampling interval according to a time sequence analysis method in the sampling process, keeping monitoring the event, increasing the value of m when the event is monitored, and keeping the value of m when the event is stable.
And predicting the value of Rt by adopting a first-order AR prediction model according to the requirements of the accuracy and the speed of the smooth data prediction and considering the limitation of calculation and storage.
In the formula, constant
Representing the use of Yule-Walker estimation, i.e.
Finally, a prediction formula based on a minimum mean square error form is given as
2) Basic technology and principle of data analysis and presentation technology based on spatio-temporal correlation
The spatio-temporal autoregressive model is a modeling method for analyzing a sequence according to the time correlation and the space correlation of a spatio-temporal sequence, and is used for analyzing the correlation relationship in a random variable.
And the method considers the relevant information among the observation points, analyzes and forecasts the integral monitoring physical quantity of the space multiple points, and possibly realizes a larger breakthrough of deformation analysis. For stationary space-time sequences, the general form of its spatio-temporal autoregressive moving average model is:
in the formula, alphatIs random error and follows normal distribution; p and q are respectively time autoregressive and moving average order; m isk、nkThe spatial autoregressive order and the moving average order are respectively the time delay k; w (h) is a weight matrix of h-order spatial delays, which represents spatial h-order adjacency, and w (0) ═ E. The weight matrix is typically given when building the model.
Here, a space-time autoregressive model with a spatial order not changing with a temporal order p is mainly discussed:
the spatial weight matrix reflects the influence between spatially adjacent points, and for a region with a relatively regular distribution, the spatial weight matrix is generally determined according to spatial proximity, that is, an element corresponding to a point adjacent to a target point is 1, and other corresponding elements are zero. The influence of the spatial weight matrix is inconsistent because the number of the adjacent points of each spatial point is different, so as to ensure that the total influence of the spatially adjacent points of each point is consistent. And (3) carrying out standardization processing on the spatial weight array:
in the formula, ωijIs the element of the ith row and the jth column of the normalized pre-space weight array; omegaiIs the sum of the elements of row i of the spatial weight matrix; wijThe normalized spatial weight array elements.
The main process of constructing the space-time sequence modeling by using the BIM is as follows:
(1) pattern recognition
The type of the model is determined by the truncation of the autocorrelation coefficients and the partial autocorrelation coefficients. Common methods for determining the model order are residual sum of squares criteria, white noise test criteria, Akaike information criteria, and F-criteria, among others. The residual sum of squares criterion is determined by whether the residual sum of squares of the high-order model is obviously lower than that of the low-order model, and has no definite measurement standard, and the determined model order is generally higher; according to the white noise test criterion, due to the limitation of a sample, the statistic of a residual error has certain uncertainty; when the two comparison models are not applicable, the F-criterion cannot give a judgment; the Akaike information criterion considers the increase of the model order, which may make the calculation more complicated and the calculation error larger. Among the widely used Akaike information criteria are
AIC＝lnσ2+p/N (7)
BIC＝lnσ2+plnN/N (8)
In the formula, σ2Is the variance of the residual; p is the order of the model; and N is the data length. Theory has demonstrated that the order determined by AIC is not a consistent estimate of its truth; the BIC considers the order of the model more and is consistency estimation of a true value of the order of the model, and the BIC is selected to determine the order of the model;
(2) parameter estimation
The least square method and the maximum likelihood estimation method are commonly used in the parameter estimation of the space-time autoregressive model. When the maximum likelihood estimation calculates the 0-order time delay corresponding coefficient, the calculation of the second derivative of the requirement likelihood function is complex. Least squares estimation is employed herein.
(3) Model inspection
The established model is checked for suitability and the parameters are checked. By constructing the global statistics of the model:
wherein N is the sample length; p is the number of parameters. The model is generally subjected to an overall significance test given an F-test significance level a of 0.05. The F-test only indicates that the model is significant overall, but does not guarantee that every parameter of the model is significant, and therefore, significance testing is required for every parameter. By constructing statistics:
in the formula, qiAs a parameter
Co-factor of (c);
is the variance. The significance of each parameter is tested through T-test, so that a final space-time autoregressive model is determined; and predicting the space-time sequence according to the model established by the BIM. And then the obtained data is combined with a visualization technology to process to obtain corresponding pictures and tables, so that the data is more intuitively perceived by readers.
The practical application is as follows:
the technology is explained by taking the measured data of the dam crest tension line of a certain concrete gravity dam as an example. The maximum dam height of the dam is 85.8m, the maximum bottom width is 68m, and the total storage capacity is 42.9 hundred million m when the normal water storage level is 108m3. The dam is provided with automatic monitoring systems such as a tension line, an inverted plumb bob, a static level, seepage monitoring, buoyancy monitoring, water level measurement and the like. The displacement monitoring data of 17 measuring points on the dam crest tension line during the period from 1 st to 28 th 8 th 28 th 2008 are collected, the data sampling rate is 1d (actually, the average value of 1d displacement), and in order to compare the time sequence of each measuring point, the data of each point are added with the same constant one by one.
The building of the dam space-time autoregressive model is used for testing the space prediction capability of the model, only 16 measuring point data in a dam crest tension line are selected to participate in modeling during modeling, and the No. 8 point does not participate in modeling and is used for testing the space prediction accuracy of the model.
(1) Obtaining spatiotemporal correlated wireless sensory data
Clustering nodes of the whole network based on an improved K-means algorithm by utilizing the data space correlation of the wireless sensor network to obtain each clustered sub-network, and selecting a cluster head from each sub-cluster; and energy is saved by utilizing a node alternate working mechanism in the cluster.
And secondly, by utilizing the time correlation, the cluster head node realizes time domain AR model prediction data of member nodes in the cluster according to a sampling frequency prediction model self-adaptive control algorithm. And carrying out self-adaptive optimization adjustment on the sampling frequency. If the perceived scene data has large fluctuation, increasing the sampling frequency so as to ensure the quality of the sampled data; if the dynamic data is small, the sampling of redundant data is reduced.
(2) Data processing
And eliminating the data with gross errors by using a 3-time medium error method, and then interpolating the missing data by using a Lagrange interpolation method. The analysis of the stabilized displacement data sequence shows that all the measuring point displacement sequences have obvious annual cycle terms, and the annual cycle terms are fitted by adopting a formula (11) for the stabilized data sequence and then deducted from the data sequence.
y＝α0+α1sin(t)+α2cos(t) (11)
(3) Determining spatial weight matrices
The spatial weight matrix is established by directly utilizing the adjacency in space. Due to the existence of the edge points, the spatial weight matrix W is normalized.
(4) Model identification
Selecting a data sequence which is stabilized in the period from 1 st of 2008 9 to 1 st of 2011 and 9, and calculating a space-time autocorrelation coefficient (table 1) and a partial correlation coefficient (table 2) of 3-order spatial delay and 6-order time delay.
TABLE 1 autocorrelation coefficients of spatio-temporal sequences
TABLE 2 partial correlation coefficients of space-time sequences
As can be seen from tables 1 and 2, the partial autocorrelation coefficients have truncation, and are suitable for building a spatio-temporal autocorrelation model, and then the BIC values shown in table 3 are calculated.
TABLE 3 BIC calculation results
As can be seen from Table 3, the value of BIC tends to stabilize after the spatial delay is 2 and the time delay is 2. therefore, the STAR (2,1) model can be established as shown in equation (12).
(5) Parameter estimation and model verification
The least squares estimation was performed on the model (9) using the smoothed data sequence, and after the model passed the F-test, the parameters of each model were tested and the results are shown in table 4.
TABLE 4F-test of estimated parameters
It can be seen from table 4 that the P values for all parameters are 0, significantly less than the given significance level of 0.05, and all parameters pass the test. The final spatio-temporal autoregressive model is
Zt＝0.209W(1)Zt+0.738W(2)Zt+0.612Zt-1-0.067W(1)Zt-1-0.456W(1)Zt-1+0.329Zt-2-0.122W(1)Zt-2-0.244W(2)Zt-2 (13)
4 indexes of correlation coefficient (fitting correlation coefficient) of the fitting data sequence and the observed data sequence, RMS value (fitting RMS) of the fitting residual data sequence, correlation coefficient (predicted correlation coefficient) of the predicted data sequence and the observed data sequence, RMS value (predicted RMS) of the predicted residual data sequence and the like are selected to perform comparative analysis on the space-time autoregressive model and the single-point autoregressive model, and the results are shown in Table 5.
TABLE 5 evaluation of accuracy
| Correlation of fitting values | Fitting RMS | Predicting the number of correlations | Predicting RMS | |
| STAR | 0.997 | 0.030 | 0.965 | 0.071 |
| AR | 0.987 | 0.066 | 0.915 | 0.131 |
| Exl_8(STAR) | 0.856 | 0.256 |
As seen from table 5: the space-time model is slightly better than the single-point autoregressive model in fitting and prediction accuracy. The space-time model covers multiple parts, multiple measuring points and multiple types of monitoring data of the gravity dam, deformation of the positions of the monitoring points on the dam body is predicted, the instant space autoregressive model has not only time prediction capability but also certain space prediction capability, and dam safety evaluation is changed from traditional single-measuring-point independent analysis evaluation to multi-measuring-point and multi-project multi-dimensional comprehensive evaluation.
The foregoing is only a part of the specific embodiments of the present invention and specific details or common general knowledge in the schemes have not been described herein in more detail. It should be noted that the above-mentioned embodiments do not limit the present invention in any way, and all technical solutions obtained by means of equivalent substitution or equivalent transformation for those skilled in the art are within the protection scope of the present invention. The scope of the claims of the present application shall be defined by the claims and the description of the embodiments and the like used for explaining the claims.
Claims (10)
1. A method for collecting and visually presenting space-time correlation monitoring data of a gravity dam is characterized by comprising a space-time correlation-based wireless sensing data collection algorithm and a space-time correlation-based data analysis algorithm;
the wireless sensing data acquisition algorithm based on the space-time correlation is a time domain sampling frequency adjustment algorithm which is adopted on the basis of considering the space correlation;
the data analysis algorithm based on the space-time correlation comprises a space-time autoregressive moving average model and a space-time autoregressive model.
2. The gravity dam space-time association monitoring data acquisition and visualization presentation method according to claim 1, wherein the space-time association-based wireless sensing data acquisition algorithm is that an initial aggregation node is searched in a network based on a minimum spanning tree, and then clustering is performed by using a K mean value to divide the whole network into a plurality of clusters; selecting a cluster head for the nodes in the cluster according to the weight; sampling by nodes in the cluster in turn, and enabling the nodes which are not sampled to enter a sleep state; the cluster head nodes in each cluster determine the sampling frequency of all member nodes in the cluster according to the history model of the sensing data and the change condition of the current sampling data; and finally, taking the sink node as a root node, forming a minimum spanning tree by the cluster head nodes, and sending information to the sink node in a multi-hop manner.
3. The gravity dam space-time correlation monitoring data acquisition and visual presentation method according to claim 2, wherein the space-time correlation based wireless sensing data acquisition algorithm comprises the following steps:
step 1) initializing a known network topology structure;
step 2) selecting an initial convergence vertex in a network area based on a minimum spanning tree algorithm;
step 3) clustering nodes in the network by using a K-means algorithm;
step 4), selecting the node with the most residual energy as a cluster head node in each K-means cluster, and establishing a minimum spanning tree of all cluster head nodes by taking the sink node as a root node;
step 5), selecting m nodes to work in each round according to the residual energy, continuously adjusting sampling intervals according to a time sequence analysis method in the sampling process, keeping event monitoring, increasing the value of m when an event is monitored, and keeping the value of m when the event is stable;
predicting the value of Rt by adopting a first-order AR prediction model according to the requirements of the accuracy and the speed of the stable data prediction and considering the limitation of calculation and storage;
in the formula, constant
Representing the use of Yule-Walker estimation, i.e.
Finally, a prediction formula based on a minimum mean square error form is given as
4. The gravity dam space-time correlation monitoring data acquisition and visualization presentation method according to claim 1, wherein the space-time autoregressive moving average model is used for analyzing and forecasting the integral monitoring physical quantity of a plurality of spatial points by considering the correlation information between observation points, and for a stable space-time sequence, the form of the space-time autoregressive moving average model is as follows:
in the formula, alphatIs random error and follows normal distribution; p and q are respectively time autoregressive and moving average order; m isk、nkThe spatial autoregressive order and the moving average order are respectively the time delay k; w (h) is a weight matrix of h-order spatial delays, which represents spatial h-order adjacency, and w (0) ═ E.
5. The gravity dam space-time correlation monitoring data acquisition and visualization presentation method according to claim 1, wherein the space-time autoregressive model is that the spatial order is invariant with the temporal order p,
the spatial weight matrix reflects the influence between the adjacent points in the space, and for the area with regular distribution, the spatial weight matrix is generally determined according to the spatial proximity, namely, the element corresponding to the point adjacent to the target point is 1, and the other corresponding elements are zero; the influence of the spatial weight matrix is inconsistent due to different numbers of adjacent points of each spatial point, so as to ensure that the total influence of the adjacent points of each spatial point is consistent; and (3) carrying out standardization processing on the spatial weight array:
in the formula, ωijIs the element of the ith row and the jth column of the normalized pre-space weight array; omegaiIs the sum of the elements of the ith row of the spatial weight matrix; wijThe normalized spatial weight array elements.
6. The gravity dam space-time correlation monitoring data acquisition and visual presentation method according to claim 1, wherein the space-time autoregressive model comprises the following three steps: step 1) pattern recognition; step 2), parameter estimation; and 3) carrying out model inspection.
7. The gravity dam space-time correlation monitoring data acquisition and visual presentation method according to claim 6, wherein the step 1) pattern recognition is to determine the type of the model through the truncation of the autocorrelation coefficients and the partial autocorrelation coefficients; the common method for determining the order of the model is a residual sum of squares criterion, a white noise test criterion, an Akaike information criterion and an F-criterion; among the widely used Akaike information criteria are
AIC＝lnσ2+p/N (7)
BIC＝lnσ2+plnN/N (8)
In the formula, σ2Is the variance of the residual; p is the order of the model; and N is the data length.
8. The gravity dam spatiotemporal correlation monitoring data acquisition and visual presentation method according to claim 7, wherein BIC is selected to determine the order of the model.
9. The gravity dam space-time correlation monitoring data acquisition and visual presentation method according to claim 6, wherein the parameter estimation of the step 2) is that a space-time autoregressive model adopts least square estimation.
10. The gravity dam space-time correlation monitoring data acquisition and visual presentation method according to claim 6, wherein the model of the step 3) is tested as,
checking whether the established model is suitable or not, and checking parameters; by constructing the global statistics of the model:
wherein N is the sample length; p is the number of parameters; generally, the significance level alpha of F-test is given to be 0.05, and the model is subjected to overall significance test; the F-test can only show that the model is significant in general, but cannot guarantee that each parameter of the model is significant, so that the significance test needs to be carried out on each parameter; by constructing statistics:
in the formula, qiAs a parameter
Co-factor of (c);
is the variance; by T-testChecking the significance of each parameter to determine a final spatio-temporal autoregressive model; and predicting the space-time sequence according to a model established by BIM, and then processing the obtained data by combining a visualization technology to obtain corresponding pictures and tables, so that the data is perceived more intuitively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111063482.1A CN113762399A (en) | 2021-09-10 | 2021-09-10 | Method for collecting and visually presenting time-space correlation monitoring data of gravity dam |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111063482.1A CN113762399A (en) | 2021-09-10 | 2021-09-10 | Method for collecting and visually presenting time-space correlation monitoring data of gravity dam |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113762399A true CN113762399A (en) | 2021-12-07 |
Family
ID=78794883
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111063482.1A Pending CN113762399A (en) | 2021-09-10 | 2021-09-10 | Method for collecting and visually presenting time-space correlation monitoring data of gravity dam |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113762399A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116029166A (en) * | 2023-03-30 | 2023-04-28 | 浙江省水利河口研究院(浙江省海洋规划设计研究院) | Method and device for determining number of times of reservoir dam re-accumulation based on pattern recognition and electronic equipment |
| CN116842339A (en) * | 2023-06-20 | 2023-10-03 | 北京讯腾智慧科技股份有限公司 | Dam data analysis method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101511099A (en) * | 2009-04-01 | 2009-08-19 | 南京邮电大学 | Collection method for wireless sensor network data based on time series prediction model |
| CN112989563A (en) * | 2021-02-02 | 2021-06-18 | 中国水利水电科学研究院 | Dam safety monitoring data analysis method |
| CN113065702A (en) * | 2021-03-31 | 2021-07-02 | 中国地质大学（武汉） | Landslide displacement multi-linear prediction method based on ST-SEEP segmentation method and space-time ARMA model |
| CN113065000A (en) * | 2021-03-29 | 2021-07-02 | 泰瑞数创科技(北京)有限公司 | Multisource heterogeneous data fusion method based on geographic entity |
-
2021
- 2021-09-10 CN CN202111063482.1A patent/CN113762399A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101511099A (en) * | 2009-04-01 | 2009-08-19 | 南京邮电大学 | Collection method for wireless sensor network data based on time series prediction model |
| CN112989563A (en) * | 2021-02-02 | 2021-06-18 | 中国水利水电科学研究院 | Dam safety monitoring data analysis method |
| CN113065000A (en) * | 2021-03-29 | 2021-07-02 | 泰瑞数创科技(北京)有限公司 | Multisource heterogeneous data fusion method based on geographic entity |
| CN113065702A (en) * | 2021-03-31 | 2021-07-02 | 中国地质大学（武汉） | Landslide displacement multi-linear prediction method based on ST-SEEP segmentation method and space-time ARMA model |
Non-Patent Citations (4)
| Title |
|---|
| 张美燕，蔡文郁，周丽萍: "基于聚类模型预测的无线传感网自适应采样技术研究", 《电子与信息学报》, vol. 37, no. 1, pages 2 * |
| 李广春，戴吾蛟，杨国祥，刘斌: "时空自回归模型在大坝变形分析中的应用", 《武汉大学学报·信息科学版》, vol. 40, no. 7, pages 1 - 2 * |
| 杨阳，王佳林，许后磊: "大坝监测信息的多维时空可视化", 《河海大学学报(自然科学版)》, vol. 38, no. 4, pages 1 - 2 * |
| 杨阳;许后磊;方春晖;徐波;: "基于监测信息的坝体三维可视化及系统集成", 水电能源科学, no. 01 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116029166A (en) * | 2023-03-30 | 2023-04-28 | 浙江省水利河口研究院(浙江省海洋规划设计研究院) | Method and device for determining number of times of reservoir dam re-accumulation based on pattern recognition and electronic equipment |
| CN116029166B (en) * | 2023-03-30 | 2023-06-20 | 浙江省水利河口研究院(浙江省海洋规划设计研究院) | Method and device for determining number of times of reservoir dam re-accumulation based on pattern recognition and electronic equipment |
| CN116842339A (en) * | 2023-06-20 | 2023-10-03 | 北京讯腾智慧科技股份有限公司 | Dam data analysis method and device |
| CN116842339B (en) * | 2023-06-20 | 2024-03-12 | 北京讯腾智慧科技股份有限公司 | Dam data analysis method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6114692B2 (en) | System and method for building motion sensor data | |
| CN103842923B (en) | For forecast and the Sequence kernel regression modeling system of prediction | |
| Di Prinzio et al. | Data-driven catchment classification: application to the pub problem | |
| CN113762399A (en) | Method for collecting and visually presenting time-space correlation monitoring data of gravity dam | |
| CN107610021A (en) | The comprehensive analysis method of environmental variance spatial and temporal distributions | |
| CN108268597A (en) | A kind of moving-target activity probability map construction and behavior intension recognizing method | |
| CN111639243B (en) | Space-time data progressive multi-dimensional mode extraction and anomaly detection visual analysis method | |
| CN109543356A (en) | Consider the ocean interior temperature-salinity structure remote sensing inversion method of Space atmosphere | |
| CN114446033A (en) | Dynamic alarm threshold value determination method | |
| JP2013175108A (en) | Clustering device and clustering program | |
| CN110851931A (en) | Optimal arrangement method for flow monitoring points of urban water supply pipe network | |
| CN113065702B (en) | Landslide displacement multi-linear prediction method based on ST-SEEP segmentation method and space-time ARMA model | |
| CN113901384A (en) | Ground PM2.5 concentration modeling method considering global spatial autocorrelation and local heterogeneity | |
| CN103914558A (en) | Method for mining space-time aggregation patterns of meteorological elements on basis of space-time statistics | |
| CN109063885A (en) | A kind of substation's exception metric data prediction technique | |
| CN115775634A (en) | Multivariable kernel density estimation-based entomogenous disease prediction method and system | |
| CN102156641A (en) | Prediction method and system for confidence interval of software cost | |
| Fischer | Spatial analysis in geography | |
| Switzer | Statistical considerations in network design | |
| CN113128875A (en) | Root cause positioning method and device for index abnormity of multi-dimensional data set | |
| Sun et al. | Flow measurement-based self-adaptive line segment clustering model for leakage detection in water distribution networks | |
| CN116304950A (en) | Multi-source heterogeneous data fusion method and device for power distribution network and storage medium | |
| Haining | Data problems in spatial econometric modeling | |
| CN115712847A (en) | High arch dam modal parameter automatic identification method based on strong earthquake monitoring data | |
| CN106792968A (en) | The secondary cluster algorithm of the wireless sensor network based on bilayer model system |
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 |