CN117807469A - Underwater sensor data acquisition method, medium and system - Google Patents
Underwater sensor data acquisition method, medium and system Download PDFInfo
- Publication number
- CN117807469A CN117807469A CN202410223731.6A CN202410223731A CN117807469A CN 117807469 A CN117807469 A CN 117807469A CN 202410223731 A CN202410223731 A CN 202410223731A CN 117807469 A CN117807469 A CN 117807469A
- Authority
- CN
- China
- Prior art keywords
- data
- sensor
- underwater
- sensors
- matrix
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 239000011159 matrix material Substances 0.000 claims abstract description 81
- 238000012423 maintenance Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 8
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 43
- 238000003064 k means clustering Methods 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 150000003839 salts Chemical class 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 description 8
- 238000009826 distribution Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 238000009360 aquaculture Methods 0.000 description 1
- 244000144974 aquaculture Species 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention provides a method, a medium and a system for acquiring data of an underwater sensor, which belong to the technical field of underwater sensors and comprise the following steps: acquiring coordinates of a plurality of underwater sensors and data acquired in real time, and preprocessing the coordinates and the data to obtain preprocessed data; processing the preprocessed data by adopting a random matrix to obtain a high-order matrix; extracting features of the high-order matrix by adopting a sliding window to obtain a feature matrix; clustering the feature matrix of each underwater sensor to obtain a plurality of clustering centers, wherein the number of the clustering centers is smaller than that of the underwater sensors; calculating a sensor to be reduced, and transmitting a newly added underwater sensor coordinate set to an operation and maintenance person; data was collected again. The method solves the technical problems that the acquisition range is small or the acquisition precision is insufficient due to too little sensor deployment because the sensor deployment relies on manual experience in the current underwater sensor data acquisition process.
Description
Technical Field
The invention belongs to the technical field of underwater sensors, and particularly relates to an underwater sensor data acquisition method, medium and system.
Background
An underwater sensor is a device specifically designed for underwater monitoring and plays an important role in many fields. For example, they may be used in the fields of marine research, environmental monitoring, aquaculture, underwater archaeology, oil exploitation, etc. The main functions of the underwater sensor include measuring water temperature, water pressure, water flow speed, water flow direction, chemical components of water (such as dissolved oxygen, pH value, salinity and the like), the number and distribution of underwater organisms (such as fish, plankton and the like), underwater topography, the condition of underwater structures and the like. At present, the arrangement of the underwater sensor network mainly relies on expert experience to carry out manual arrangement, and the manual experience method has the following problems that monitoring blind areas are easily caused by the manual arrangement, some important areas cannot be well covered, and the monitoring data are inaccurate. The resource waste phenomenon exists, too many redundant sensors are distributed in the areas with low importance, and the system cost is increased. That is, in the process of acquiring underwater sensor data in the prior art, manual experience is generally relied on for deploying the sensor underwater, so that the technical problems that the sensor deployment is denser, the acquisition range is small, or the acquisition precision is insufficient are caused.
Disclosure of Invention
In view of the above, the invention provides a method, medium and system for acquiring underwater sensor data, which can solve the technical problems that the existing sensors are densely deployed to cause a small acquisition range or too few to cause insufficient acquisition precision due to the fact that the underwater deployment sensors generally depend on manual experience in the current underwater sensor data acquisition process.
The invention is realized in the following way:
the first aspect of the invention provides an underwater sensor data acquisition method, which comprises the following steps:
s10, acquiring coordinates of a plurality of underwater sensors and data acquired in real time, wherein the coordinates, the water temperature and the salinity are included;
s20, preprocessing the acquired data to obtain preprocessed data;
s30, processing the preprocessed data by adopting a random matrix to obtain a high-order matrix;
s40, carrying out feature extraction on the high-order matrix by adopting a sliding window to obtain a feature matrix;
s50, clustering the feature matrix of each underwater sensor to obtain a plurality of clustering centers, wherein the number of the clustering centers is smaller than that of the underwater sensors;
s60, comparing the clustering degree of each feature matrix with the clustering center corresponding to the feature matrix, and if the clustering degree is larger than the preset threshold upper limit, marking the underwater sensor corresponding to the feature matrix as a sensor to be reduced; if the clustering degree is smaller than the preset threshold lower limit, searching M optimal coordinates by adopting an interpolation method according to the coordinates of a plurality of underwater sensors corresponding to the clustering center, wherein M is larger than or equal to 1, and the M is used as a coordinate set of the newly added underwater sensor;
s70, sending the sensor to be reduced and the newly added underwater sensor coordinate set to operation and maintenance personnel;
and S80, the personnel to be operated and maintained adjusts the deployment of the underwater sensors according to the sensors to be reduced and the newly added underwater sensor coordinate set, and then re-acquires the data acquired by the plurality of underwater sensors in real time and sends the data to the personnel to be operated and maintained.
On the basis of the technical scheme, the underwater sensor data acquisition method can be further improved as follows:
the method for acquiring coordinates of a plurality of underwater sensors and data acquired in real time comprises the following steps of: according to the deployment position of the underwater sensor, the space coordinate of each sensor is recorded in advance; each underwater sensor collects water quality data of the current position at a certain time interval, including water temperature and salinity; and matching the space coordinates of each sensor with water quality data of corresponding time points to form a data set.
The step of preprocessing the collected data to obtain preprocessed data specifically comprises the following steps: detecting whether the data of each sensor is in a normal range, and removing errors and outlier data; for data which are not read by the individual sensors due to equipment faults, carrying out spatial interpolation calculation by the readings of the adjacent sensors, and filling the missing data; carrying out time series smoothing filtering on the data of each sensor to remove measurement noise; the water temperature and salinity data are normalized and mapped to between 0 and 1.
The step of processing the preprocessed data by adopting a random matrix to obtain a high-order matrix specifically comprises the following steps: constructing an m×n random matrix, wherein m is the time sequence length and n is the sensor number; performing matrix dot product operation on the preprocessed data set and the random matrix to obtain an m multiplied by n high-order matrix; repeating the above process for multiple times to obtain multiple groups of high-order matrixes, and splicing the high-order matrixes on a third dimension to form a three-dimensional matrix.
The step of extracting features of the high-order matrix by adopting a sliding window to obtain a feature matrix specifically comprises the following steps: setting the size of a sliding window; extracting statistical features from the data of each window; and repeating the process to finally obtain the statistical characteristics of each sensor on each time window.
The step of clustering the feature matrix of each underwater sensor to obtain a plurality of clustering centers specifically comprises the following steps: splicing the feature matrixes of all the sensors together, and carrying out a K-Means clustering algorithm; setting the number k of the clustering centers as a certain proportion of the total number of the sensors; and after the K-Means clustering iteration converges, K clustering centers and class clusters corresponding to the centers are obtained.
The step of comparing the clustering degree of each feature matrix and the corresponding clustering center to judge the sensors to be reduced and increased specifically comprises the following steps: calculating the Euclidean distance between each sensor feature matrix and the corresponding clustering center; setting a clustering degree threshold value, and if the distance is larger than the upper threshold value limit, marking the sensor as to-be-reduced; if the distance is smaller than the threshold lower limit, marking a sensor needing to be newly added in the cluster; and for the sensor needing to be newly added, calculating a new coordinate point by adopting an inverse distance interpolation method to serve as a newly added sensor point.
Further, the upper limit of the predetermined threshold is 0.8, and the lower limit is 0.4.
A second aspect of the present invention provides a computer readable storage medium, wherein the computer readable storage medium stores program instructions, and the program instructions are used to execute the underwater sensor data acquisition method.
A third aspect of the present invention provides an underwater sensor data acquisition system comprising the computer readable storage medium described above.
Compared with the prior art, the underwater sensor data acquisition method, medium and system provided by the invention have the beneficial effects that: redundant sensors in the densely distributed areas can be intelligently identified, the number of the sensors is reduced, and resources are effectively saved. The monitoring leak area can be intelligently found, the position of the newly added sensor is prompted, and the monitoring coverage is improved. The network layout can be continuously optimized according to the water environment change, rather than single static layout. The method solves the technical problems that in the current underwater sensor data acquisition process, sensors are deployed under water generally by means of manual experience, so that the acquisition range is small or too small due to the fact that sensors are deployed more densely, and the acquisition precision is insufficient.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, based on the embodiments of the invention, which are apparent to those of ordinary skill in the art without inventive faculty, are intended to be within the scope of the invention.
Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, based on the embodiments of the invention, which are apparent to those of ordinary skill in the art without inventive faculty, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
As shown in fig. 1, a flowchart of a method for collecting underwater sensor data according to a first aspect of the present invention is provided, the method includes the following steps:
s10, acquiring coordinates and real-time acquired data of a plurality of underwater sensors, including coordinates, water temperature and salinity, wherein the underwater sensors are underwater sensors of a salt depth meter;
s20, preprocessing the acquired data to obtain preprocessed data;
s30, processing the preprocessed data by adopting a random matrix to obtain a high-order matrix;
s40, carrying out feature extraction on the high-order matrix by adopting a sliding window to obtain a feature matrix;
s50, clustering the feature matrix of each underwater sensor to obtain a plurality of clustering centers, wherein the number of the clustering centers is smaller than that of the underwater sensors;
s60, comparing the clustering degree of each feature matrix with the clustering center corresponding to the feature matrix, and if the clustering degree is larger than the preset threshold upper limit, marking the underwater sensor corresponding to the feature matrix as a sensor to be reduced; if the clustering degree is smaller than the preset threshold lower limit, searching M optimal coordinates by adopting an interpolation method according to the coordinates of a plurality of underwater sensors corresponding to the clustering center, wherein M is larger than or equal to 1, and the preset threshold upper limit is 0.8 and the preset threshold lower limit is 0.4;
s70, sending the sensor to be reduced and the newly added underwater sensor coordinate set to operation and maintenance personnel;
and S80, the personnel to be operated and maintained adjusts the deployment of the underwater sensors according to the sensors to be reduced and the newly added underwater sensor coordinate set, and then re-acquires the data acquired by the plurality of underwater sensors in real time and sends the data to the personnel to be operated and maintained.
The following describes a specific embodiment of each of the steps described above:
s10, acquiring coordinates of a plurality of underwater sensors and data acquired in real time, wherein the coordinates, the water temperature, the salinity and the like can be included, and the coordinates, the water temperature, the salinity and the like can also include water pressure, water flow speed, water flow direction, chemical components of water and the like:
1) The spatial coordinates (x, y, z) of each sensor are pre-recorded according to the deployment location of the underwater sensor. The coordinates can be obtained through GPS positioning, and can also be manually measured and recorded;
2) Each underwater sensor collects water quality data at a current location, including water temperature and salinity, at certain time intervals (e.g., 1 minute). The water temperature can be measured by a temperature sensor, and the salinity can be measured by a conductivity sensor;
3) And matching the space coordinates of each sensor with water quality data of corresponding time points to form a data set (coordinates, time, water temperature and salinity). The data set is the data obtained in S10.
S20, preprocessing the acquired data to obtain preprocessed data:
1) Error and outlier data is removed: detecting whether the data of each sensor is in a normal range, such as water temperature is 0-35 ℃, salinity is 0-40ppt, and deleting the error data if the data is out of the normal range;
2) Interpolation fills in missing data: for the data which are not read by the individual sensors due to equipment faults, the data of the missing water temperature and salinity can be filled by carrying out spatial interpolation calculation through the readings of the adjacent sensors;
3) Smoothing the filtered noise data: and carrying out time series smoothing filtering on the data of each sensor to remove measurement noise. Algorithms such as a moving average method can be adopted;
4) Normalized data: the water temperature and salinity data are normalized and mapped to 0-1, so that the subsequent processing is facilitated;
and after pretreatment, a normalized and filtered complete data set is obtained, and each sensor has complete sequence data of four parameters of coordinates, time, water temperature and salinity.
S30, processing the preprocessed data by adopting a random matrix to obtain a high-order matrix:
1) Constructing a random matrix: generating a random matrix of m x n, where m is the time series length and n is the number of sensors. The values of the matrix elements are randomly distributed between 0 and 1;
2) Matrix operation: and performing matrix dot product operation on the preprocessed data set and the random matrix to obtain a m x n high-order matrix. The high-order matrix integrates time and space information;
3) Repeating the above process for several times to obtain multiple groups of high-order matrixes. And finally splicing the high-order matrixes on a third dimension to form a three-dimensional matrix. The three-dimensional matrix is the output result of S30;
the space-time correlation of the data can be enhanced by matrix operation, and the effect of subsequent feature extraction is improved.
S40, extracting features of the high-order matrix by adopting a sliding window to obtain a feature matrix:
1) Setting a sliding window size, for example, 30 (length of time) x 20 (number of sensors), the window sliding in the time direction;
2) Extracting statistical features such as mean, maximum, minimum, variance and the like of water temperature and salinity in each window from the data of each window;
3) And repeating the process to finally obtain the statistical characteristics of each sensor on each time window. The set of statistical features is a feature matrix of the sensor;
the sliding window is adopted to extract local statistical characteristics, so that time sequence information of data can be captured, and more distinguishing characteristics can be provided for subsequent clustering and analysis.
S50, clustering the feature matrix of each underwater sensor to obtain a plurality of clustering centers, wherein the number of the clustering centers is smaller than that of the underwater sensors:
1) Splicing the feature matrixes of all the sensors together, and carrying out a K-Means clustering algorithm;
2) The number k of cluster centers is set to be a certain proportion of the total number of sensors, for example 80% of the number of sensors. Through test and cross verification, selecting a proper k value;
3) And after the K-Means clustering iteration converges, K clustering centers and class clusters corresponding to the centers are obtained. Each cluster center can be regarded as feature overall statistics of all sensors in a class cluster;
4) Since k < total number of sensors, the multiple sensors converge to the same class of cluster. This may reduce redundant sensors.
S60, comparing the clustering degree of each feature matrix with the clustering center corresponding to the feature matrix, and if the clustering degree is larger than the preset threshold upper limit, marking the underwater sensor corresponding to the feature matrix as a sensor to be reduced; if the clustering degree is smaller than the preset threshold lower limit, searching M optimal coordinates by adopting an interpolation method according to the coordinates of a plurality of underwater sensors corresponding to the clustering center, wherein M is larger than or equal to 1, and the M is used as a coordinate set of the newly added underwater sensor:
1) Calculating Euclidean distance between each sensor feature matrix and the corresponding clustering center, wherein the smaller the distance is, the higher the clustering degree is;
2) Setting a clustering degree threshold, and if the distance is larger than an upper limit threshold, marking the sensor as to-be-reduced; if the distance is smaller than the lower threshold value, marking a sensor needing to be newly added in the cluster;
3) For a cluster needing to be newly added, calculating M new coordinate points as newly added sensor points by adopting an inverse distance interpolation method according to the coordinates and the characteristic distances of the existing sensors in the cluster;
the step intelligently determines the sensors to be reduced and increased through the clustering degree.
S70, sending the sensor to be reduced and the newly added underwater sensor coordinate set to an operation and maintenance person:
1) Generating a list of sensor numbers to be reduced, and sending the list to operation and maintenance personnel;
2) Generating a newly added sensor coordinate list, giving specific position information of each coordinate, and sending the specific position information to operation and maintenance personnel;
3) After the operation and maintenance personnel receive the information, the operation and maintenance personnel can carry out manual confirmation, sample test verification if necessary, and then sensor deployment adjustment;
4) After the adjustment is completed, the system is informed to restart the acquisition, and one sensor optimization iteration is completed.
And S80, after the personnel to be operated and maintained adjusts the deployment of the underwater sensors according to the sensors to be reduced and the newly added underwater sensor coordinate sets, the data acquired by the plurality of underwater sensors in real time are acquired again and sent to the personnel to be operated and maintained, and the specific acquisition mode is the same as that of the step S10.
The following is one specific embodiment of the method of the present invention:
s10, acquiring coordinates of a plurality of underwater sensors and data acquired in real time
Let the total number of the underwater sensors beFirst->The coordinates of the individual sensors are +.>,/>；
Each sensor is at fixed time intervalsData acquisition is carried out (for example, 1 minute) to obtain the water temperature of the current positionAnd salinity->Wherein->Indicate->A plurality of time sample points;
then construct a coordinate-time-data matching relationship:
。
s20, preprocessing the acquired data 1) removing error data, and judging whether the water temperature and salinity data are in a normal range or not:if the data of the sample point is not in the range, marking the data of the sample point as error data, wherein the range is set by operation and maintenance personnel; 2) Interpolation filling is for the sensor->At time->Under the condition that the data is empty, filling is calculated by three-dimensional linear interpolation through water temperature and salinity of the proximity sensor>Wherein (1)>Representing sensor +.>At->A set of proximity sensors in the vicinity of time, +.>Representing the corresponding weight coefficient; 3) Filter smoothing one-dimensional gaussian filtering of the data for each sensor:wherein (1)>Representing a gaussian kernel->Controlling the smooth window size; 4) Normalization mapping of Water temperature and salinity data to [0, 1 ] by linear normalization]:/>。
S30, generating a high-order matrix by matrix operation to construct a random matrixThe elements follow a uniform distribution U (0, 1), wherein the time sequence length +.>Sensor number->. For each time point->Calculate>Wherein (1)>Representing a matrix dot product operation,/->Are respectively->Normalized water and salinity for all sensors at each time point; repeating the above process to calculate->Finally, the results of all time points are spliced to obtain a three-dimensional matrixWherein->Representing the number of matrices spliced.
S40, extracting characteristics of the sliding window, and setting the size of the sliding window asThe sliding step length is +.>. For each windowCalculating statistical characteristics:；
i.e. within each window, the mean and standard deviation of water and salinity are calculated, whereinIs the average value of water temperature->Is salinity mean value (L.) of (L)>Is water temperature standard deviation>Is the salinity standard deviation. And repeating the process to finally obtain the statistical characteristics of each sensor on all windows.
S50, the clustering center is used for generating clusters to splice the characteristic matrixes of all the sensors into a large matrix in the row direction. K-Means clustering algorithm is adopted for the pair +.>Clustering, setting the clustering number->Wherein->Is the number of sensors->Is a proportional coefficient of (2); iteratively optimizing an objective function:
；
wherein the method comprises the steps ofRepresenting a cluster center set, +.>Indicate->Class sample, I/O>Indicate->A cluster center of the class; after iteration convergence, a cluster center is obtained>。
S60, determining and adjusting the sensor to calculate each sample according to the clustering degreeAnd corresponding cluster center->Is a euclidean distance of (2):judging whether the sensor needs to be adjusted according to the distance threshold value:wherein->For a preset distance thresholdDefault toThe method comprises the steps of carrying out a first treatment on the surface of the For class cluster requiring sensor addition +.>Calculating new sensor coordinates:wherein (1)>Is a distance-based weight coefficient.
Specifically, the principle of the invention is as follows:
1) Mapping the sensing data to a high-dimensional space by utilizing a random matrix algorithm, and enhancing the space-time relevance of the data;
2) Capturing statistical characteristics of data on a time sequence by a sliding window method;
3) And classifying the sensors according to the feature similarity by using a clustering algorithm. The cluster center represents typical region features;
4) The clustering density is measured according to the sample distance. If the distance is greater than the threshold value, redundancy is indicated, and if the distance is less than the threshold value, insufficient SAMPLE monitoring is indicated;
5) And outputting a sensor proposal for reducing the redundant area, and outputting a new sensor coordinate according to the characteristic distribution in the cluster for the insufficient monitoring area.
A second aspect of the present invention provides a computer readable storage medium, wherein the computer readable storage medium stores program instructions, and the program instructions are used to execute the underwater sensor data acquisition method.
A third aspect of the present invention provides an underwater sensor data acquisition system comprising the computer readable storage medium described above.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (10)
1. The underwater sensor data acquisition method is characterized by comprising the following steps of:
s10, acquiring coordinates and real-time acquired data of a plurality of underwater sensors, including coordinates, water temperature and salinity, wherein the underwater sensors are underwater sensors of a salt depth meter;
s20, preprocessing the acquired data to obtain preprocessed data;
s30, processing the preprocessed data by adopting a random matrix to obtain a high-order matrix;
s40, carrying out feature extraction on the high-order matrix by adopting a sliding window to obtain a feature matrix;
s50, clustering the feature matrix of each underwater sensor to obtain a plurality of clustering centers, wherein the number of the clustering centers is smaller than that of the underwater sensors;
s60, comparing the clustering degree of each feature matrix with the clustering center corresponding to the feature matrix, and if the clustering degree is larger than the preset threshold upper limit, marking the underwater sensor corresponding to the feature matrix as a sensor to be reduced; if the clustering degree is smaller than the preset threshold lower limit, searching M optimal coordinates by adopting an interpolation method according to the coordinates of a plurality of underwater sensors corresponding to the clustering center, wherein M is larger than or equal to 1, and the M is used as a coordinate set of the newly added underwater sensor;
s70, sending the sensor to be reduced and the newly added underwater sensor coordinate set to operation and maintenance personnel;
and S80, the personnel to be operated and maintained adjusts the deployment of the underwater sensors according to the sensors to be reduced and the newly added underwater sensor coordinate set, and then re-acquires the data acquired by the plurality of underwater sensors in real time and sends the data to the personnel to be operated and maintained.
2. The method for acquiring the data of the underwater sensors according to claim 1, wherein the steps of acquiring the coordinates and the data acquired in real time of the plurality of underwater sensors, including the coordinates, the water temperature and the salinity, specifically include: according to the deployment position of the underwater sensor, the space coordinate of each sensor is recorded in advance; each underwater sensor collects water quality data of the current position at a certain time interval, including water temperature and salinity; and matching the space coordinates of each sensor with water quality data of corresponding time points to form a data set.
3. The method for collecting data of an underwater sensor according to claim 1, wherein the step of preprocessing the collected data to obtain preprocessed data comprises: detecting whether the data of each sensor is in a normal range, and removing errors and outlier data; for data which are not read by the individual sensors due to equipment faults, carrying out spatial interpolation calculation by the readings of the adjacent sensors, and filling the missing data; carrying out time series smoothing filtering on the data of each sensor to remove measurement noise; the water temperature and salinity data are normalized and mapped to between 0 and 1.
4. The method for collecting data of an underwater sensor according to claim 1, wherein the step of processing the preprocessed data by using a random matrix to obtain a high-order matrix comprises: constructing an m×n random matrix, wherein m is the time sequence length and n is the sensor number; performing matrix dot product operation on the preprocessed data set and the random matrix to obtain an m multiplied by n high-order matrix; repeating the above process for multiple times to obtain multiple groups of high-order matrixes, and splicing the high-order matrixes on a third dimension to form a three-dimensional matrix.
5. The method for collecting data of an underwater sensor according to claim 1, wherein the step of extracting features of the high-order matrix by using a sliding window to obtain the feature matrix specifically comprises: setting the size of a sliding window; extracting statistical features from the data of each window; and repeating the process to finally obtain the statistical characteristics of each sensor on each time window.
6. The method for collecting data of underwater sensors according to claim 1, wherein the step of clustering the feature matrix of each underwater sensor to obtain a plurality of cluster centers specifically comprises: splicing the feature matrixes of all the sensors together, and carrying out a K-Means clustering algorithm; setting the number k of the clustering centers as a certain proportion of the total number of the sensors; and after the K-Means clustering iteration converges, K clustering centers and class clusters corresponding to the centers are obtained.
7. The method for collecting data of an underwater sensor according to claim 1, wherein the step of comparing the clustering degree of each feature matrix and the corresponding clustering center to determine whether the sensor is to be reduced or added specifically comprises: calculating the Euclidean distance between each sensor feature matrix and the corresponding clustering center; setting a clustering degree threshold value, and if the distance is larger than the upper threshold value limit, marking the sensor as to-be-reduced; if the distance is smaller than the threshold lower limit, marking a cluster of the corresponding class with a sensor to be newly added; and for the sensor needing to be newly added, calculating a new coordinate point by adopting an inverse distance interpolation method to serve as a newly added sensor point.
8. An underwater sensor data acquisition method as in any of claims 1-7, characterized in that the predetermined threshold has an upper limit of 0.8 and a lower limit of 0.4.
9. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein program instructions, which when run, are adapted to perform a method of underwater sensor data acquisition as claimed in any one of claims 1-8.
10. An underwater sensor data acquisition system comprising the computer readable storage medium of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410223731.6A CN117807469B (en) | 2024-02-29 | 2024-02-29 | Underwater sensor data acquisition method, medium and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410223731.6A CN117807469B (en) | 2024-02-29 | 2024-02-29 | Underwater sensor data acquisition method, medium and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117807469A true CN117807469A (en) | 2024-04-02 |
| CN117807469B CN117807469B (en) | 2024-05-17 |
Family
ID=90434822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410223731.6A Active CN117807469B (en) | 2024-02-29 | 2024-02-29 | Underwater sensor data acquisition method, medium and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117807469B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10180339B1 (en) * | 2015-05-08 | 2019-01-15 | Digimarc Corporation | Sensing systems |
| CN111062739A (en) * | 2019-11-18 | 2020-04-24 | 中国电力科学研究院有限公司 | Intelligent power consumption behavior recommendation method and system based on random matrix decomposition |
| CN112054557A (en) * | 2020-09-07 | 2020-12-08 | 上海交通大学 | Distribution network topology change type identification method based on random matrix theory |
| CN115761550A (en) * | 2022-12-20 | 2023-03-07 | 江苏优思微智能科技有限公司 | Water surface target detection method based on laser radar point cloud and camera image fusion |
| CN115776724A (en) * | 2023-02-10 | 2023-03-10 | 南京航空航天大学 | Sensor node layout optimization method and system for electromagnetic spectrum mapping |
| CN116227363A (en) * | 2023-04-25 | 2023-06-06 | 湖南省水务规划设计院有限公司 | Flood early warning method based on sensor distribution optimization |
| CN117609831A (en) * | 2023-11-21 | 2024-02-27 | 国网宁夏电力有限公司电力科学研究院 | Transformer partial discharge detection method, medium and system based on random matrix |
-
2024
- 2024-02-29 CN CN202410223731.6A patent/CN117807469B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10180339B1 (en) * | 2015-05-08 | 2019-01-15 | Digimarc Corporation | Sensing systems |
| CN111062739A (en) * | 2019-11-18 | 2020-04-24 | 中国电力科学研究院有限公司 | Intelligent power consumption behavior recommendation method and system based on random matrix decomposition |
| CN112054557A (en) * | 2020-09-07 | 2020-12-08 | 上海交通大学 | Distribution network topology change type identification method based on random matrix theory |
| CN115761550A (en) * | 2022-12-20 | 2023-03-07 | 江苏优思微智能科技有限公司 | Water surface target detection method based on laser radar point cloud and camera image fusion |
| CN115776724A (en) * | 2023-02-10 | 2023-03-10 | 南京航空航天大学 | Sensor node layout optimization method and system for electromagnetic spectrum mapping |
| CN116227363A (en) * | 2023-04-25 | 2023-06-06 | 湖南省水务规划设计院有限公司 | Flood early warning method based on sensor distribution optimization |
| CN117609831A (en) * | 2023-11-21 | 2024-02-27 | 国网宁夏电力有限公司电力科学研究院 | Transformer partial discharge detection method, medium and system based on random matrix |
Non-Patent Citations (2)
| Title |
|---|
| CHEN YANG ET AL.: "Sensor placement algorithm for structural health monitoring with redundancy elimination model based on sub-clustering strategy", MECHANICAL SYSTEMS AND SIGNAL PROCESSING, no. 124, 10 February 2019 (2019-02-10), pages 369 - 387 * |
| 林超等: "基于随机矩阵理论的 WSN 异常节点定位算法", 计算机工程, vol. 46, no. 1, 31 January 2020 (2020-01-31), pages 157 - 163 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117807469B (en) | 2024-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107169426B (en) | Crowd emotion abnormality detection and positioning method based on deep neural network | |
| CN108230302B (en) | Detection and disposal method for marine organism invading from cold source sea area of nuclear power plant | |
| CN112735094B (en) | Geological disaster prediction method and device based on machine learning and electronic equipment | |
| KR20180057785A (en) | A system of measuring fish number for image analysis and that of measure method | |
| CN113610381B (en) | Water quality remote real-time monitoring system based on 5G network | |
| CN105379186A (en) | Determining response similarity neighborhoods | |
| CN116881749B (en) | Pollution site construction monitoring method and system | |
| CN115240060B (en) | Starfish disaster early warning method and system | |
| CN113574480B (en) | Device for predicting equipment damage | |
| KR102582180B1 (en) | Small river smart prediction management system using accumulated data and its method | |
| CN110717496B (en) | Complex scene tree detection method based on neural network | |
| CN112507861A (en) | Pedestrian detection method based on multilayer convolution feature fusion | |
| CN117473351B (en) | Power supply information remote transmission system based on Internet of things | |
| CN106935038B (en) | Parking detection system and detection method | |
| CN117009900A (en) | Internet of things signal anomaly detection method and system based on graph neural network | |
| CN117171533B (en) | Real-time acquisition and processing method and system for geographical mapping operation data | |
| CN117807469B (en) | Underwater sensor data acquisition method, medium and system | |
| CN116881640A (en) | Method and system for predicting core extraction degree and computer-readable storage medium | |
| CN116563889A (en) | Device and method for estimating weight of laying hen based on machine vision | |
| CN114898100A (en) | Point cloud data extraction method, device, system, equipment and storage medium | |
| CN116861204B (en) | Intelligent manufacturing equipment data management system based on digital twinning | |
| CN117253141B (en) | Method and device for determining sample data of forest investigation sample and electronic equipment | |
| CN118347471B (en) | Water conservancy monitoring method and system based on multi-sensor fusion technology | |
| CN118504847B (en) | Intelligent beam field management method and system based on digital twin technology | |
| CN117540272A (en) | Target event detection method and device based on optical fiber sensing |
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 |