CN117668472B - Island reef environment multi-parameter monitoring method and system - Google Patents

Island reef environment multi-parameter monitoring method and system Download PDF

Info

Publication number
CN117668472B
CN117668472B CN202410147279.XA CN202410147279A CN117668472B CN 117668472 B CN117668472 B CN 117668472B CN 202410147279 A CN202410147279 A CN 202410147279A CN 117668472 B CN117668472 B CN 117668472B
Authority
CN
China
Prior art keywords
data
dissolved oxygen
observation
data set
observation data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202410147279.XA
Other languages
Chinese (zh)
Other versions
CN117668472A (en
Inventor
彭勃
杨志勇
谢丰懋
王伟文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jinan University
Original Assignee
Jinan University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jinan University filed Critical Jinan University
Priority to CN202410147279.XA priority Critical patent/CN117668472B/en
Publication of CN117668472A publication Critical patent/CN117668472A/en
Application granted granted Critical
Publication of CN117668472B publication Critical patent/CN117668472B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/10—Pre-processing; Data cleansing
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N17/00—Investigating resistance of materials to the weather, to corrosion, or to light
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • 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
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Environmental Sciences (AREA)
  • Environmental & Geological Engineering (AREA)
  • Ecology (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

The invention relates to the technical field of digital data denoising, in particular to a multi-parameter monitoring method and system for island reef environment, comprising the following steps: acquiring each observation data set and each related data set corresponding to each dissolved oxygen data, and determining a first credibility and a second credibility of each observation data set based on data characteristics of each observation data set and each related data set; according to the noise representation value of each error covariance matrix, the first credibility and the second credibility of each observation data set, determining the noise representation value of each dissolved oxygen data, and further optimizing a Kalman filtering algorithm to obtain each denoised dissolved oxygen data; and monitoring the environment of the island to be monitored based on each denoised dissolved oxygen data. According to the invention, the noise representation value of each dissolved oxygen data is corrected, so that the denoising effect of the dissolved oxygen data is improved, and the accuracy of the sea island reef environment monitoring result is further improved.

Description

Island reef environment multi-parameter monitoring method and system
Technical Field
The invention relates to the technical field of digital data denoising, in particular to a multi-parameter monitoring method and system for island reef environment.
Background
The optical fiber sensing technology is a measuring method based on the optical principle, and the sensor in the optical fiber can detect and monitor the change of environmental parameters. In marine environment, dissolved oxygen is an important water quality index, has important influence on biological ecological system and water body health condition, and the optical fiber sensing technology can be used for monitoring the dissolved oxygen in seawater in real time to provide high-precision and continuous data, and can provide valuable data support in the fields of marine research, marine ecological protection, aquaculture and the like. Regarding dissolved oxygen, the inflow of excessive nutrients (e.g., elemental phosphorus) into the seawater may cause algae to proliferate in large amounts, which consume oxygen during death and decay, resulting in a decrease in the oxygen concentration in the body of water, resulting in the formation of oxygen deficiency due to eutrophication, i.e., too little dissolved oxygen in the seawater.
The fiber optic sensor can be used to measure optical oxygenation (Optical Oxygenation) or blue carbon fixation (Blue Carbon Fixation) in the ocean, referring to the process of plants in the ocean to absorb carbon dioxide and release oxygen through photosynthesis; specifically, by coating a layer of special fluorescent material on the optical fiber sensor, when dissolved oxygen interacts with the fluorescent material, the change of a fluorescent signal is caused, and the dissolved oxygen amount in the seawater can be deduced by measuring the intensity change of the fluorescent signal; oxygen is a depolarizer, so that in general, the more the oxygen content in water is, the more serious the corrosion of the optical fiber sensor material is, namely the change condition of dissolved oxygen in seawater can also adversely affect the maintenance operation of the optical fiber sensor, and in order to monitor the environmental corrosion condition of the optical fiber sensor in real time, high-quality dissolved oxygen data needs to be obtained.
Because of uncertainty of environmental factors and instability of measuring equipment in severe environments, some noise may be contained in the acquired dissolved oxygen data, the existing method for denoising the dissolved oxygen data is a Kalman filtering algorithm, when the Kalman gain of each data point is calculated by the traditional Kalman filtering algorithm, correlation and variability between different measured values cannot be accurately described by a single error covariance matrix only according to an error covariance matrix of one measured noise, contribution degrees of different measured values cannot be accurately reflected, accuracy and reliability of Kalman gain values are easily affected, denoising effect of the dissolved oxygen data is poor, and accuracy of island reef environment monitoring is low, and accuracy of environment corrosion condition monitoring of an optical fiber sensor is also low.
Disclosure of Invention
In order to solve the technical problem that the accuracy of monitoring the island reef environment is low due to the fact that the denoising effect of the traditional Kalman filtering algorithm on dissolved oxygen data is poor, the invention aims to provide a multi-parameter monitoring method and system for the island reef environment, and the adopted technical scheme is as follows:
the embodiment of the invention provides a multi-parameter monitoring method for island reef environment, which comprises the following steps:
Acquiring dissolved oxygen data of sea water of an island to be monitored at each acquisition time in a current acquisition time period, and further acquiring an observation data set and a related data set of each historical time period corresponding to the dissolved oxygen data at each acquisition time; wherein the lengths of the respective history periods are different;
for dissolved oxygen data at any acquisition time, analyzing data distribution characteristics and data change characteristics of the observed data set per se according to the observed data set of each history period corresponding to the dissolved oxygen data, and determining first credibility of the observed data set of each history period;
analyzing the data characteristic differences among all the observation data sets according to the first credibility of the observation data sets of each history period, the number of data in the observation data sets and the related data sets, and determining the second credibility of the observation data sets of each history period;
determining a noise representation value of an error covariance matrix of each history period corresponding to dissolved oxygen data; determining a noise representation value of the dissolved oxygen data according to the noise representation value of the error covariance matrix of each history period corresponding to the dissolved oxygen data and the first credibility and the second credibility of the observation data set of each history period;
Optimizing a Kalman filtering algorithm through a noise representation value of the dissolved oxygen data at each acquisition time, and denoising all the dissolved oxygen data by utilizing the optimized Kalman filtering algorithm to obtain denoised dissolved oxygen data;
and monitoring the environment of the island to be monitored based on each denoised dissolved oxygen data.
Further, for any collection time, the observation data set is used for constructing an error covariance matrix corresponding to the dissolved oxygen amount data of the collection time, the observation data set is composed of a plurality of historical dissolved oxygen amount data closest to the dissolved oxygen amount data of the collection time, the historical dissolved oxygen amount data are dissolved oxygen amount data of a plurality of continuous historical collection times in any historical period before the collection time, and the related data set is composed of phosphorus element content data of a plurality of continuous historical collection times in any historical period before the collection time.
Further, the analyzing the data distribution characteristics and the data change characteristics of the observation data set according to the observation data set of each history period corresponding to the dissolved oxygen data, and determining the first credibility of the observation data set of each history period includes:
For an observation data set of any historical period corresponding to dissolved oxygen data, performing curve fitting on the observation data set to obtain an observation fitting curve, and determining the amplitude between two adjacent data points in the observation fitting curve; wherein, the horizontal axis of the observation fitting curve is the acquisition time, and the vertical axis is the observation data;
analyzing the change condition of the amplitude between two adjacent data points according to the amplitude between the two adjacent data points in the observation fitting curve, and determining the first abnormality degree of the observation data set; calculating a difference value between the median of the observation data set and the average value of the observation data set, and determining the difference value between the normalized median and the average value as a second abnormality degree of the observation data set;
determining the minimum preset number of the observation data set, and determining the average value of the difference values between the minimum preset number of the observation data and the average value of the observation data set as a third abnormality degree of the observation data set;
and calculating the product of the first abnormality degree, the second abnormality degree and the third abnormality degree of the observation data set, carrying out inverse proportion normalization processing on the product of the three abnormality degrees, and taking the numerical value after the inverse proportion normalization processing as the first credibility of the observation data set of the history period corresponding to the dissolved oxygen data.
Further, the analyzing the change condition of the amplitude between two adjacent data points according to the amplitude between two adjacent data points in the observation fit curve, and determining the first abnormality degree of the observation data set includes:
counting the times that the amplitude between two adjacent data points is larger than the average amplitude corresponding to the observation fitting curve, and recording the times as the first times; counting the times that the amplitude between two adjacent data points is larger than the average amplitude corresponding to the observation fit curve and the ordinate of the former data point is larger than the ordinate of the latter data point in the two data points corresponding to the amplitude, and recording the times as the second times; the ratio of the second times to the first times is used as the first degree of abnormality of the observation data set.
Further, the analyzing the data characteristic differences among all the observation data sets according to the first credibility of the observation data sets, the number of data in the observation data sets and the related data sets in each history period to determine the second credibility of the observation data sets in each history period includes:
determining an observation fitting curve and a correlation fitting curve of each history period, and further determining extreme points in the observation fitting curve and the correlation fitting curve of each history period; wherein the horizontal axis of the relevant fitting curve is the acquisition time, and the vertical axis is the phosphorus element content data;
Determining the reality degree of the observation data set of each historical period according to each extreme point in the observation fitting curve and the correlation fitting curve of the historical period of the same period;
and for the observation data set of any history period corresponding to the dissolved oxygen data, determining the second credibility of the observation data set of the history period according to the first credibility, the true degree and the data number of the observation data set of the history period and the first credibility, the true degree and the data number of the observation data sets of other history periods except the history period.
Further, the calculation formula of the true degree of the observation data set of each history period is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the degree of realism of the observation data set of the nth history period corresponding to the dissolved oxygen data, norm is a linear normalization function, +.>For the average time interval between all adjacent extreme points in the correlation fit curve of the nth history period corresponding to the dissolved oxygen data, +.>The average time interval between all adjacent extreme points in the fitted curve is observed for the nth history period corresponding to the dissolved oxygen data,for->Absolute value is determined for- >For->Absolute value is determined for->The number of extreme points in the correlation fitting curve of the nth history period corresponding to the dissolved oxygen data, < +.>And (3) the number of extreme points in the observation fitting curve of the nth historical period corresponding to the dissolved oxygen data, wherein n is the serial number of the historical period.
Further, the calculation formula of the second credibility of the observation data set of the nth history period corresponding to the dissolved oxygen data is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the second confidence of the observation data set of the nth history period corresponding to the dissolved oxygen amount data, exp is an exponential function based on a natural constant, A is the number of history periods corresponding to the dissolved oxygen amount data, s is the serial number of each history period except the nth history period, and the like>First confidence,/for the observation data set of the nth history period corresponding to dissolved oxygen data>The degree of realism of the observation data set for the nth history period corresponding to the dissolved oxygen amount data, +.>First credibility of observation data set of the s-th history period except the n-th history period corresponding to dissolved oxygen amount data, +.>For the degree of realism of the observation data set of the s-th history period other than the n-th history period corresponding to the dissolved oxygen amount data,/the >For->Absolute value is determined for->The number of data in the observation data set of the nth history period corresponding to the dissolved oxygen amount data, < +.>The number of data in the observation data set of the s-th history period except the n-th history period corresponding to the dissolved oxygen amount data is +.>To pair(s)The absolute value is determined.
Further, the determining the noise representation value of the dissolved oxygen data according to the noise representation value of the error covariance matrix of each history period corresponding to the dissolved oxygen data, the first credibility and the second credibility of the observation data set of each history period includes:
for any historical period corresponding to dissolved oxygen data, taking the product of the first credibility and the second credibility of the observation data set of the historical period as a molecule of a ratio, taking the accumulated sum of the products of the first credibility and the second credibility of the observation data set of all the historical periods as the denominator of the ratio, and taking the ratio as the noise expression weight of the error covariance matrix of the historical period;
calculating the product of the noise representation value and the noise representation weight of the error covariance matrix of the historical period as the corrected noise representation value of the error covariance matrix of the historical period; and taking the accumulated sum of the corrected noise representation values of the error covariance matrix of each history period corresponding to the dissolved oxygen data as the noise representation value of the dissolved oxygen data.
Further, the monitoring of the environment of the island to be monitored based on each denoised dissolved oxygen data includes:
if any dissolved oxygen data after denoising is smaller than a preset dissolved oxygen standard, judging that the environment of the island to be monitored is abnormal, otherwise, judging that the environment of the island to be monitored is normal.
The embodiment of the invention also provides a multi-parameter monitoring system for the island reef environment, which comprises a processor and a memory, wherein the processor is used for processing the instructions stored in the memory so as to realize a multi-parameter monitoring method for the island reef environment.
The invention has the following beneficial effects:
the invention provides a sea island reef environment multi-parameter monitoring method and system, which are used for acquiring an observation data set of each historical period corresponding to dissolved oxygen data at each acquisition time, analyzing data characteristics of the observation data set of each historical period, determining a first credibility and a second credibility of the observation data set, and subsequently weighting noise performance of an error covariance matrix of each historical period based on the two credibility to acquire more accurate noise performance corresponding to each acquisition time; by combining the data characteristics of a plurality of observation data sets with different sizes, the corresponding error covariance matrix can be better adapted to the change of actual measurement noise, so that the accuracy and the robustness of the subsequently determined denoised dissolved oxygen data are improved; when the second credibility of the observation data sets of each historical period is analyzed, the relevant data sets of each historical period are also introduced, and according to the relation between the phosphorus element content data and the dissolved oxygen data change characteristics, the influence of false anomalies caused by noise can be effectively weakened, so that the analysis of the noise representation value of the dissolved oxygen data is more accurate, the denoising effect of a Kalman filtering algorithm on the dissolved oxygen data is further improved, the island reef environment monitoring accuracy is improved, and the monitoring accuracy of the environmental corrosion condition of the optical fiber sensor of the acquisition equipment is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a multi-parameter monitoring method for island reef environment in an embodiment of the present invention;
FIG. 2 is a flow chart of a first confidence process for determining a set of observed data in an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description is given below of the specific implementation, structure, features and effects of the technical solution according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The embodiment of the invention provides a multi-parameter monitoring method for island reef environment, which is shown in figure 1 and comprises the following steps:
s1, obtaining dissolved oxygen data of sea water of an island to be monitored at each acquisition time in a current acquisition time period, and further obtaining an observation data set and a related data set of each historical time period corresponding to the dissolved oxygen data at each acquisition time.
Firstly, obtaining dissolved oxygen data of sea water of an island to be monitored at each acquisition time in a current acquisition period.
In this embodiment, dissolved oxygen data of sea water near the island reef to be monitored at each collection time within the current collection period is collected by using dissolved oxygen collection equipment and a continuous flow system. The dissolved oxygen amount acquisition equipment can be an optical fiber sensor, the optical fiber sensor can realize real-time, continuous and noninvasive monitoring of the dissolved oxygen amount in the ocean, and valuable data support is provided in the fields of ocean research, ocean ecological protection, aquaculture and the like; the current acquisition period may be set to 30 minutes and the interval between adjacent acquisition instants may be set to 2 seconds, i.e. the acquisition frequency is 2 seconds 1 time. It should be noted that, the collection frequency of the embodiment is uniform when collecting different types of raw data in the seawater, and may be set to 2 seconds and 1 time.
And secondly, acquiring an observation data set and a related data set of each historical period corresponding to the dissolved oxygen data at each acquisition time.
And a first substep, obtaining an observation data set of each history period corresponding to the dissolved oxygen data at each acquisition time.
In order to ensure that the dissolved oxygen data at each acquisition time in the current acquisition period obtain more accurate noise representation values, so as to improve the denoising effect of a Kalman filtering algorithm on the dissolved oxygen data, an error covariance matrix of measurement noise with different magnitudes is established for each dissolved oxygen data, a data set for constructing the error covariance matrix is recorded as an observation data set, and each observation data set for constructing the error covariance matrix, corresponding to the dissolved oxygen data at each acquisition time, is required to be acquired.
In this embodiment, the process of obtaining the observation data set of each history period corresponding to the dissolved oxygen amount data at each collection time is consistent, and for convenience of description, taking the observation data set of each history period corresponding to the dissolved oxygen amount data at any collection time as an example for analysis, the specific implementation steps may include:
firstly, setting the number of maximum dissolved oxygen data to be used for constructing each error covariance matrix and the number of error covariance matrices corresponding to the dissolved oxygen data at the acquisition time; determining the number of dissolved oxygen data needed to be used for constructing error covariance matrixes with different sizes, and sequentially marking the number as the number of dissolved oxygen data The number of error covariance matrices is marked +.>Setting the number of error covariance matrixes>The number of dissolved oxygen data to be used when the error covariance matrix of 6,6 different sizes is equal to +.>Maximum data number->Can be set to 90, the number of differences between the number of dissolved oxygen data to be used in the adjacent error covariance matrix is equal, and the number of differences can beSet to 15; based on the maximum data number->And the difference number can be known>Equal to 75->Equal to 60->Equal to 45->Equal to 30->Equal to 15. Of course, the number of maximum data and the number of error covariance matrices corresponding to each dissolved oxygen amount data can be determined by the practitioner according to specific practical conditions, and the method is not particularly limited.
It should be noted that, the number of data in the error covariance matrix in the kalman filtering algorithm is not necessarily equal to the number of data in the corresponding observation data set, and the data in the error covariance matrix is not corresponding to each data in the observation data set.
Secondly, acquiring 90 historical dissolved oxygen data which are positioned before the acquisition time and are closest to the dissolved oxygen data at the acquisition time to form a data set, and recording the data set as an observation data set of the 1 st historical period; acquiring 75 historical dissolved oxygen data which are positioned before the acquisition time and are nearest to the dissolved oxygen data at the acquisition time, and forming a data set, wherein the data set is recorded as an observation data set of a 2 nd historical period; acquiring 60 historical dissolved oxygen data which are positioned before the acquisition time and are nearest to the dissolved oxygen data at the acquisition time, and forming a data set, namely an observation data set of a 3 rd historical period; according to the acquisition rule of the observation data sets, the observation data sets of the subsequent historical period are continuously and iteratively acquired until the number of the observation data sets is equal to the number of the error covariance matrices At this time, an observation data set of each history period corresponding to the dissolved oxygen amount data at the acquisition time is obtained.
It should be noted that, the 1 st historical period is from the first historical collection time to the last historical collection time corresponding to the 90 historical dissolved oxygen amount data, the collection time before the collection time is the historical collection time, the lengths of the historical periods corresponding to the collection time are different, the historical dissolved oxygen amount data is the dissolved oxygen amount data before the dissolved oxygen amount data at the collection time, and the historical collection time and the historical dissolved oxygen amount data are both data relative to the dissolved oxygen data at the collection time.
And a second sub-step of acquiring a relevant data set of each history period corresponding to the dissolved oxygen data at each acquisition time.
In order to distinguish false abnormality caused by noise influence from true abnormality caused by eutrophication, when analyzing the data characteristics of the observation data sets of the respective history periods corresponding to the dissolved oxygen amount data at each acquisition time, it is necessary to acquire the relevant data sets of the respective history periods corresponding to the dissolved oxygen amount data at each acquisition time, and the data directly associated with the presence of the dissolved oxygen amount data is phosphorus element content data. By combining the data characteristics of the phosphorus element content data at each acquisition time in the same historical period, the correlation between the observation data set and the related data set in the same historical period is analyzed, and the numerical accuracy of the second credibility is improved.
In this embodiment, the process of acquiring the relevant data set of each history period corresponding to the dissolved oxygen amount data at each acquisition time is consistent, and for convenience of description, the relevant data set of each history period corresponding to the dissolved oxygen amount data at any acquisition time is taken as an example to perform analysis, and the specific implementation steps may include:
firstly, determining an observation data set of each historical period corresponding to dissolved oxygen data at the acquisition time, and secondly, based on the historical period of each observation data set; and collecting phosphorus element content data at each collection time in the same historical period, forming a data set, and recording the data set as a related data set so as to obtain the related data set of each historical period corresponding to the dissolved oxygen data at the collection time.
As an example, assume a set of dissolved oxygen data isAcquiring observation data sets of 2 history periods corresponding to the 6 th acquisition time, wherein the observation data sets are respectively +.>And->The historical time periods corresponding to the two observation data sets are respectively from the 2 nd acquisition time to the 5 th acquisition time and from the 4 th acquisition time to the 5 th acquisition time; collecting phosphorus content data of each collecting time from the 2 nd collecting time to the 5 th collecting time to form a data set, and recording as Collecting the phosphorus content data of each collecting time from the 4 th collecting time to the 5 th collecting time to form a data set, which is marked as +.>At this time, relevant data sets of 2 historical periods corresponding to the 6 th acquisition time are obtained, which are respectively +.>And->。
S2, analyzing data distribution characteristics and data change characteristics of the observation data set according to the observation data set of each historical period corresponding to dissolved oxygen data at each acquisition time, and determining first credibility of the observation data set of each historical period.
The data reliability of the observation data set is quantified by analyzing the data sub-characteristics and the data change characteristics of the observation data set in each historical period corresponding to the dissolved oxygen data, so that the noise performance weight of each error covariance matrix corresponding to the dissolved oxygen data is determined. The oxygen dissolving amount data corresponding to the real anomaly and the false anomaly are extremely small values, and when the extremely small values appear in the combination of the observed data, the probability of noise in the observed data set is higher, the authenticity of all the oxygen dissolving amount data in the observed data set is poorer, and the reliability is lower. Therefore, the data anomaly condition of the observation data set is quantified by analyzing the extremely small degree of the observation data in the observation data set of each historical period, so as to determine the first credibility of the observation data set of each historical period corresponding to the dissolved oxygen data at each acquisition time.
The calculation process of the first credibility of the observation data set of each history period corresponding to the dissolved oxygen amount data at each collection time is consistent, taking the determination of the first credibility of the observation data set of any history period corresponding to the dissolved oxygen amount data at any collection time as an example, the flowchart of the process of determining the first credibility of the observation data set is shown in fig. 2, and the method comprises the following steps:
and a first step, performing curve fitting on the observation data set to obtain an observation fitting curve, and determining the amplitude between two adjacent data points in the observation fitting curve.
In this embodiment, curve fitting may be performed on each observation data in the observation data set by using a least square method, and the obtained fitting curve is referred to as an observation fitting curve, and the implementation process of the least square method is the prior art and will not be described in detail herein; from each data point in the observation fit curve, the amplitude between two adjacent data points is determined. The process of calculating the amplitude is the prior art, and is not included in the scope of the present invention, and will not be described in detail herein.
And each two adjacent data points in the observation fit curve correspond to one amplitude, and the amplitude can represent the data change degree between the two data points, namely the difference of the vertical coordinates between the two adjacent data points, and the larger the amplitude is, the larger the data change degree is. The horizontal axis of the observation fitting curve is the acquisition time, the vertical axis of the observation fitting curve is the observation data, and the observation data is the historical dissolved oxygen data of the historical acquisition time before the acquisition time.
And secondly, analyzing the change condition of the amplitude between two adjacent data points according to the amplitude between the two adjacent data points in the observation fitting curve, and determining the first abnormality degree of the observation data set.
In this embodiment, the number of times that the amplitude between two adjacent data points is greater than the average amplitude corresponding to the observation fit curve is counted as the first number of times; counting the times that the amplitude between two adjacent data points is larger than the average amplitude corresponding to the observation fit curve and the ordinate of the former data point is larger than the ordinate of the latter data point in the two data points corresponding to the amplitude, and recording the times as the second times; the ratio of the second times to the first times is used as the first degree of abnormality of the observation data set.
It should be noted that, the first degree of abnormality may represent a situation that the second number of times of observing the fitted curve is the duty ratio of the first number of times, and the larger the duty ratio of the first number of times in the second number of times is, the larger the amplitude between all two adjacent data points in the observed fitted curve is, the larger the average amplitude of the whole section of observed data curve is, and the more the number of times that the amplitude changes to be the decrease of the observed data from the previous acquisition time to the subsequent acquisition time is, the more the observed data with smaller abnormality in the observed data set will be, the greater the possibility that the observed data set is abnormal, and the first degree of abnormality is the greater.
And thirdly, calculating the difference value between the median of the observed data set and the average value of the observed data set, and determining the difference value between the normalized median and the average value as a second abnormality degree of the observed data set.
It should be noted that, the difference between the median and the average value may represent the offset condition of the observation data set, and when the median is greater than the average value, it indicates that the whole of the observation data set is left, and the whole of the observation data set is small; in order to accommodate the case where the difference between the median and the average is negative, normalization processing is required for the difference between the median and the average. The greater the left degree of the whole observation data set, the more observation data with abnormal changes in the observation data set, the greater the possibility of occurrence of abnormal conditions in the observation data set, and the greater the second abnormal degree.
And a fourth step of determining the minimum preset number of the observation data set, and determining the average value of the difference values between the minimum preset number of the observation data and the average value of the observation data set as a third abnormality degree of the observation data set.
In this embodiment, the preset number may be set to 5, and the practitioner may set the size of the preset number according to the specific practical situation. The third degree of abnormality may represent a difference between a plurality of smallest observed data in the observed data set and an average value corresponding to the observed data set, and the greater the third degree of abnormality, the greater the degree of data dispersion in the observed data set, the greater the possibility of occurrence of an abnormality.
And fifthly, calculating the product of the first abnormality degree, the second abnormality degree and the third abnormality degree of the observed data set, carrying out inverse proportion normalization processing on the product of the three abnormality degrees, and taking the numerical value after inverse proportion normalization processing as the first credibility of the observed data set of the historical period corresponding to the dissolved oxygen data.
The greater the degree of abnormality of the observation data set, the greater the possibility that noise exists in the observation data set, the lower the reliability of the observation data set, so that the product of the three degrees of abnormality is inversely proportional processed, thereby obtaining the first reliability of the observation data set.
As an example, the calculation formula of the first confidence of the observation data set of the nth history period corresponding to the dissolved oxygen amount data may be:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the first credibility of the observation data set of the nth history period corresponding to the dissolved oxygen data, exp is an exponential function based on a natural constant, +.>Fitting a second number of times of the curve for the observation of the nth history period corresponding to the dissolved oxygen amount data,/for the observation of the second history period corresponding to the dissolved oxygen amount data>Fitting a first number of times, +. >For the first degree of abnormality of the observed data set for the nth historical period corresponding to the dissolved oxygen data, norm is a linear normalization function, +.>A second degree of abnormality of the observation data set for the nth history period corresponding to the dissolved oxygen amount data, +.>A median value of the observation data set of the nth history period corresponding to the dissolved oxygen amount data,/for>For the average value of the observation data sets of the nth history period corresponding to the dissolved oxygen amount data, U is the serial number of the minimum preset number of the observation data sets of the nth history period corresponding to the dissolved oxygen amount data, U is the preset number, & lt/EN & gt>The (2) is the (u) minimum observation data of the observation data set of the (n) th history period corresponding to the dissolved oxygen data,>for->Absolute value is determined for->And a third degree of abnormality of the observation data set for the nth history period corresponding to the dissolved oxygen amount data.
In a calculation formula of the first credibility, the first degree of abnormality, the second degree of abnormality and the third degree of abnormality all show negative correlation with the first credibility, and the larger the three degrees of abnormality are, the smaller the credibility of the observation data set is, namely the smaller the first credibility is; the first confidence level may characterize a degree of normality of data change features and data distribution features of the observation data set itself, the degree of normality referring to a likelihood that there is no real, spurious anomaly data in the observation data set.
It should be noted that, when the first reliability is analyzed, the whole observation data set is analyzed from three different angles to show extremely small degree, namely, the first abnormality degree, the second abnormality degree and the third abnormality degree are determined, so that the numerical accuracy of the first reliability can be effectively improved.
S3, analyzing the data characteristic differences among all the observation data sets according to the first credibility of the observation data sets, the number of data in the observation data sets and the related data sets in each historical period, and determining the second credibility of the observation data sets in each historical period corresponding to the dissolved oxygen data at each acquisition time.
It should be noted that, the first reliability only analyzes the data features of the observation data set itself, and only performs subsequent analysis of the noise expression weight based on the first reliability, if the entire observation data set presents abnormality, the noise expression weight calculated subsequently will lack certain reliability and accuracy. In order to improve the numerical accuracy of the noise performance weights, the second credibility of the observation data sets of each history period is determined by analyzing the data characteristic differences between the observation data sets of each history period and the other observation data sets except for the observation data sets.
For convenience of description, taking the second credibility of the observation data set of any history period corresponding to the dissolved oxygen amount data of any collection time as an example, the calculation process of the second credibility of the observation data set of each history period corresponding to the dissolved oxygen amount data of each collection time is consistent, the method comprises the following steps:
and in the first step, determining an observation fitting curve and a correlation fitting curve of the historical period, and further determining extreme points in the observation fitting curve and the correlation fitting curve of the historical period.
In this embodiment, a least square method may be used to perform curve fitting on each observation data in the observation data set in the history period first, so as to obtain a fitted curve corresponding to the observation data set, that is, an observation fitted curve; and then, performing curve fitting on the relevant data set in the history period by using a least square method to obtain a fitting curve corresponding to the relevant data set, wherein the fitting curve corresponding to the relevant data set is used as a relevant fitting curve. Wherein the horizontal axis of the relevant fitting curve is the acquisition time, and the vertical axis is the phosphorus element content data; the curve fitting process and the determination process of the extreme points are both the prior art and are not within the scope of the present invention, and will not be described in detail herein.
The data set is converted into a curve so as to be convenient for determining each extreme point on the curve, wherein the extreme points comprise a maximum value point and a minimum value point, the extreme points can be used for analyzing the similarity of data change between the observation fit curve and the relevant fit curve, and the more similar the data change is, the higher the correlation between the observation data set and the relevant data set corresponding to the history period is.
And secondly, determining the reality degree of the observation data set of the historical period according to each extreme point in the observation fitting curve and the correlation fitting curve of the historical period of the same period.
It should be noted that, for the observation data set and the related data set in the same history period, if the probability that the observation data in the observation data set generates noise data generated by uncontrollable factors such as equipment and environment is smaller, the data changes of the two data sets are illustrated to show stronger similarity, and the stronger the authenticity of the observation data set in the history period is, the greater the degree of authenticity of the calculated observation data set is.
Regarding the authenticity of the observation data set, any of the observation data sets of the history period may have false abnormal data generated by uncontrollable factors and true abnormal data generated by eutrophication, firstly, the abnormal data means that the dissolved oxygen amount data in the sea water is too low; secondly, eutrophication means that excessive nutrient substances flow into the seawater to cause mass propagation of algae, so that the dissolved oxygen in the seawater is too low; then, the actual abnormal data can be reflected to some extent by the phosphorus element content data of the same history period, and the related data of the dissolved oxygen amount refers to the phosphorus element content data. When calculating the Kalman gain of dissolved oxygen data at each acquisition time, the interference caused by uncontrollable factors should be eliminated as much as possible, namely, the interference of false abnormal data is eliminated, and the real abnormal data is the data actually generated and measured by the seawater itself, and can be used for calculating the Kalman gain.
In this embodiment, by analyzing the difference between the average time interval between all the adjacent extremum points in the observation fit curve of the same history period and the average time interval between all the adjacent extremum points in the correlation fit curve, the degree of similarity of the data fluctuation range between the observation data set and the correlation data set of the same history period is quantified, and the greater the degree of similarity of the data fluctuation range, the greater the authenticity of the observation data set of the history period; and quantifying the similarity degree of the data fluctuation times between the observation data set and the related data set in the same historical period by analyzing the number difference of extreme points between the observation fit curve and the related fit curve in the same historical period, wherein the greater the similarity degree of the data fluctuation times is, the stronger the authenticity of the observation data set in the historical period is.
As an example, the calculation formula of the degree of reality of the observation data set of the nth history period corresponding to the dissolved oxygen amount data may be:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the degree of realism of the observation data set of the nth history period corresponding to the dissolved oxygen data, norm is a linear normalization function, +.>For the average time interval between all adjacent extreme points in the correlation fit curve of the nth history period corresponding to the dissolved oxygen data, +. >For the average time interval between all adjacent extreme points in the observation fit curve of the nth history period corresponding to the dissolved oxygen data, +.>For->Absolute value is determined for->For->Absolute value is determined for->The number of extreme points in the correlation fitting curve of the nth history period corresponding to the dissolved oxygen data, < +.>And (3) the number of extreme points in the observation fitting curve of the nth historical period corresponding to the dissolved oxygen data, wherein n is the serial number of the historical period.
In the calculation formula of the degree of realism,degree of difference in data fluctuation range between the observation data set and the related data set, which can represent the same history period, +.>The larger the data fluctuation range of the two data sets is, the more dissimilar the data fluctuation ranges are, the larger the observed data sets in the history period are affected by noise, and the smaller the reality degree is;degree of difference in the number of data fluctuations between the observation data set and the related data set, which may represent the same history period, +.>Smaller (less)>The closer to 1, the smaller the difference in the number of data fluctuations of the two data sets, the smaller the influence of noise on the observed data set of the history period, and the greater the degree of reality.
It should be noted that, by calculating the actual degree of the observation data set in each history period, the correlation between the observation data set in the same history period and the related data set can be quantified, and the stronger the correlation is, the more likely the abnormal data in the observation data set in the history period is caused by real eutrophication, the less the influence of uncontrollable factors is, which is helpful to further correct the first credibility of the observation data set in each history period and improve the accuracy of the data feature analysis of the observation data set itself.
And thirdly, determining the second credibility of the observation data set of the historical period according to the first credibility, the actual degree and the data number of the observation data set of the historical period and the first credibility, the actual degree and the data number of the observation data set of each historical period except the historical period.
In this embodiment, the greater the sum of the differences between the first credibility of the observation data set of the history period and the first credibility of the observation data set of each history period other than the history period, the weaker the similarity between the observation data set of the history period and the observation data set of each history period, the greater the probability that the observation data set of the history period has noise points, and the smaller the corresponding second credibility; the stronger the correlation between the observed data set and the related data set of the same history period is, the less the observed data set of the history period has the possibility of noise data generated by uncontrollable factors such as equipment, environment and the like, the greater the corresponding authenticity is; the larger the data number difference between the observation data set of the history period and the observation data sets of other certain history periods is, the smaller the reference value of the observation data set of the other certain history period to the observation data set of the history period is, and the smaller the authenticity of the first credibility of the observation data set of the history period is.
As an example, the calculation formula of the second confidence of the observation data set of the nth history period corresponding to the dissolved oxygen amount data may be:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the second confidence of the observation data set of the nth history period corresponding to the dissolved oxygen amount data, exp is an exponential function based on a natural constant, A is the number of history periods corresponding to the dissolved oxygen amount data, s is the serial number of each history period except the nth history period, and the like>First confidence,/for the observation data set of the nth history period corresponding to dissolved oxygen data>The degree of realism of the observation data set for the nth history period corresponding to the dissolved oxygen amount data, +.>First credibility of observation data set of the s-th history period except the n-th history period corresponding to dissolved oxygen amount data, +.>For the degree of realism of the observation data set of the other s-th history period except the n-th history period corresponding to the dissolved oxygen amount data,for->Absolute value is determined for->The number of data in the observation data set of the nth history period corresponding to the dissolved oxygen amount data, < +.>The number of data in the observation data set of the s-th history period except the n-th history period corresponding to the dissolved oxygen amount data is +. >For->The absolute value is determined.
In the calculation formula of the second confidence level,can represent the corrected first credibility, corresponding to the observation data set of the nth history period corresponding to the dissolved oxygen data,>the difference of the corrected first credibility between the observed data set of the nth history period corresponding to the dissolved oxygen amount data and the observed data set of the other s history period except the nth history period can be characterized, wherein the larger the difference of the corrected first credibility is, the smaller the similarity between the observed data set of the nth history period and the observed data set of the s history period is, the lower the credibility of the observed data set of the nth history period is, and the higher the possibility of false abnormal data exists；/>Can be characterized by the reference value of the observation data sets of the nth history period except the nth history period corresponding to the dissolved oxygen amount data, the smaller the difference of the data numbers of the two observation data sets is, the calculatedThe greater the reference value of (c).
S4, determining a noise representation value of an error covariance matrix of each historical period corresponding to dissolved oxygen data at each acquisition time; and determining the noise representation value of the dissolved oxygen data at each acquisition time according to the noise representation value of the error covariance matrix of each history period corresponding to the dissolved oxygen data at each acquisition time and the first credibility and the second credibility of the observation data set of each history period.
In this embodiment, the larger the first reliability of the observation data set corresponding to the dissolved oxygen data at any acquisition time, the smaller the influence of uncontrollable factors such as equipment and environment on the error covariance matrix constructed by the observation data set is, the higher the data accuracy is, and the larger the weight of the noise of the error covariance matrix constructed by the observation data set when the noise is weighted should be; similarly, the greater the second reliability of the observation data set, the stronger the similarity between the error covariance matrix constructed by the observation data set and the error covariance matrix constructed by other observation data sets corresponding to the dissolved oxygen data at the acquisition time, and the less likely noise occurs in the observation data set, the greater the weight of the noise of the error covariance matrix constructed by the observation data set when weighted should be.
Firstly, determining an error covariance matrix of each historical period corresponding to dissolved oxygen data at each acquisition time, and then determining a noise representation value of the error covariance matrix of each historical period. The error covariance matrix of each history period can be constructed by the observation data set corresponding to the history period, the observation data set of each history period has the corresponding error covariance matrix, and the dissolved oxygen data of each acquisition time can correspond to a plurality of error covariance matrices with different sizes. The determining process of the error covariance matrix and the noise performance value is the prior art, and can be determined in the implementation process of the Kalman filtering algorithm, and will not be described in detail.
The calculation process of the noise representation value of the dissolved oxygen amount data at each acquisition time is consistent, and for convenience of description, taking the determination of the noise representation value of the dissolved oxygen amount data at any acquisition time as an example, the method comprises the following steps:
in the first step, for any historical period corresponding to dissolved oxygen data, taking the product of the first credibility and the second credibility of the observation data set of the historical period as a molecule of a ratio, taking the sum of the products of the first credibility and the second credibility of the observation data set of all the historical periods as a denominator of the ratio, and taking the ratio as the noise performance weight of an error covariance matrix of the historical period.
A second step of calculating the product of the noise representation value of the error covariance matrix of the history period and the noise representation weight as a corrected noise representation value of the error covariance matrix of the history period; and taking the accumulated sum of the corrected noise representation values of the error covariance matrix of each history period corresponding to the dissolved oxygen data as the noise representation value of the dissolved oxygen data.
As an example, the calculation formula of the noise expression value of the dissolved oxygen amount data at the kth acquisition time may be:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->The noise representation value of the dissolved oxygen data at the kth acquisition time is represented by A, which is the number of historical time periods corresponding to the dissolved oxygen data,/day >Noise representation values of the error covariance matrix for the nth history period of the kth acquisition instant,/-, for the error covariance matrix>A first confidence level of the observation data set of the nth history period corresponding to the dissolved oxygen data of the kth acquisition time,/for the kth acquisition time>A second confidence level of the observation data set of the nth history period corresponding to the dissolved oxygen data of the kth acquisition time,/day>Is the sum of the products of the first credibility and the second credibility of the observation data sets of all the history periods corresponding to the dissolved oxygen data at the kth acquisition time,noise performance weight of error covariance matrix of nth history period corresponding to dissolved oxygen data of kth acquisition time,/th acquisition time>And (3) representing the value of the corrected noise of the error covariance matrix of the nth historical period corresponding to the dissolved oxygen data at the kth acquisition time.
It should be noted that, when calculating the noise representation value of the dissolved oxygen amount data at each acquisition time, an error covariance matrix of a plurality of measurement noises with different magnitudes is used, and the first credibility and the second credibility of the observation data set in each history period for constructing the error covariance matrix are quantized, and finally the weight of the noise representation of the error covariance matrix is determined, which is helpful to better adapt to the change of the actual measurement noises, so as to improve the accuracy of the denoised dissolved oxygen amount data obtained later.
S5, optimizing a Kalman filtering algorithm through a noise representation value of the dissolved oxygen data at each acquisition time, and denoising all the dissolved oxygen data by utilizing the optimized Kalman filtering algorithm to obtain denoised dissolved oxygen data; and monitoring the environment of the island to be monitored based on each denoised dissolved oxygen data.
In this embodiment, based on the noise representation value of the dissolved oxygen amount data at each acquisition time determined in this embodiment, the calculation result of the kalman gain in the process of denoising the dissolved oxygen amount data at each acquisition time in the current acquisition period by using the kalman filtering algorithm is improved, so that an optimized kalman filtering algorithm corresponding to the dissolved oxygen amount data can be obtained, denoising is performed on all the dissolved oxygen amount data by using the optimized kalman filtering algorithm, and each denoised dissolved oxygen amount data can be obtained. The implementation process of the kalman filtering algorithm is the prior art and is not within the scope of the present invention, and will not be described in detail here.
The optimized Kalman filtering algorithm can improve the numerical accuracy and reliability of Kalman gain values in the process of denoising the dissolved oxygen data, so that the denoising effect of the dissolved oxygen data is improved, and the dissolved oxygen data with higher data quality is obtained.
After denoising dissolved oxygen data at each acquisition time in the current period, monitoring the environment of the island to be monitored, wherein the method specifically comprises the following steps: if any dissolved oxygen data after denoising is smaller than a preset dissolved oxygen standard, judging that the environment of the island to be monitored is abnormal, wherein the environment abnormality refers to that seawater near the island to be monitored presents the dissolved oxygen data to be too low, otherwise, judging that the environment of the island to be monitored is normal. The preset dissolved oxygen standard may be set to 6 mg, and the practitioner may set the preset dissolved oxygen standard according to specific practical conditions, without specific limitation. And then based on each dissolved oxygen data after denoising, the damage condition of the optical fiber sensor material of the data acquisition equipment in the seawater environment can be analyzed, if the damage condition is serious, for example, when the dissolved oxygen data after denoising is more than 9 mg, the optical fiber sensor material at the current acquisition time is corroded, the maintenance work of the optical fiber sensor needs to be carried out in time, otherwise, the maintenance work of the optical fiber sensor does not need to be carried out in the current acquisition period. The damage degree of different materials is different, so that the detailed implementation process of analyzing the damage condition of the optical fiber sensor material based on dissolved oxygen data is not developed, and an operator can carry out maintenance work of the optical fiber sensor according to the specific actual condition.
In the relationship between the dissolved oxygen concentration and the degree of corrosion of the equipment, the concentration of dissolved oxygen increases with the concentration, and the corrosion rate increases, but when the concentration reaches a certain limit, high oxygen causes the oxide to become a passivation film, thereby reducing the corrosion rate.
So far, the embodiment completes the monitoring process of the island reef environment to be monitored in the current acquisition period.
The embodiment of the invention also provides a multi-parameter monitoring system for the island reef environment, which comprises a processor and a memory, wherein the processor is used for processing the instructions stored in the memory so as to realize the multi-parameter monitoring method for the island reef environment.
According to the embodiment of the invention, the error covariance matrixes of the measurement noise with different sizes are established for the dissolved oxygen data at each acquisition time, the weight of the noise of the quantized error covariance matrix is represented when the noise is weighted, the noise representation value of the dissolved oxygen data at each acquisition time is determined, the numerical accuracy of the Kalman gain value in the Kalman filtering algorithm implementation process is improved, the accuracy of the denoised dissolved oxygen data is further improved, and the reliability of the island reef environment monitoring result is enhanced.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention and are intended to be included within the scope of the invention.

Claims (4)

1. The island reef environment multi-parameter monitoring method is characterized by comprising the following steps of:
acquiring dissolved oxygen data of sea water of an island to be monitored at each acquisition time in a current acquisition time period, and further acquiring an observation data set and a related data set of each historical time period corresponding to the dissolved oxygen data at each acquisition time; wherein the lengths of the respective history periods are different;
for dissolved oxygen data at any acquisition time, analyzing data distribution characteristics and data change characteristics of the observed data set per se according to the observed data set of each history period corresponding to the dissolved oxygen data, and determining first credibility of the observed data set of each history period;
analyzing the data characteristic differences among all the observation data sets according to the first credibility of the observation data sets of each history period, the number of data in the observation data sets and the related data sets, and determining the second credibility of the observation data sets of each history period;
determining a noise representation value of an error covariance matrix of each history period corresponding to dissolved oxygen data; determining a noise representation value of the dissolved oxygen data according to the noise representation value of the error covariance matrix of each history period corresponding to the dissolved oxygen data and the first credibility and the second credibility of the observation data set of each history period;
Optimizing a Kalman filtering algorithm through a noise representation value of the dissolved oxygen data at each acquisition time, and denoising all the dissolved oxygen data by utilizing the optimized Kalman filtering algorithm to obtain denoised dissolved oxygen data;
monitoring the environment of the island to be monitored based on each denoised dissolved oxygen data;
the analyzing the data distribution characteristics and the data change characteristics of the observation data set according to the observation data set of each history period corresponding to the dissolved oxygen data, and determining the first credibility of the observation data set of each history period comprises the following steps:
for an observation data set of any historical period corresponding to dissolved oxygen data, performing curve fitting on the observation data set to obtain an observation fitting curve, and determining the amplitude between two adjacent data points in the observation fitting curve; wherein, the horizontal axis of the observation fitting curve is the acquisition time, and the vertical axis is the observation data;
analyzing the change condition of the amplitude between two adjacent data points according to the amplitude between the two adjacent data points in the observation fitting curve, and determining the first abnormality degree of the observation data set; calculating a difference value between the median of the observation data set and the average value of the observation data set, and determining the difference value between the normalized median and the average value as a second abnormality degree of the observation data set;
Determining the minimum preset number of the observation data set, and determining the average value of the difference values between the minimum preset number of the observation data and the average value of the observation data set as a third abnormality degree of the observation data set;
calculating the product of the first abnormality degree, the second abnormality degree and the third abnormality degree of the observation data set, carrying out inverse proportion normalization processing on the product of the three abnormality degrees, and taking the numerical value after the inverse proportion normalization processing as the first credibility of the observation data set of the history period corresponding to the dissolved oxygen data;
the method for determining the first abnormality degree of the observation data set includes the steps of:
counting the times that the amplitude between two adjacent data points is larger than the average amplitude corresponding to the observation fitting curve, and recording the times as the first times; counting the times that the amplitude between two adjacent data points is larger than the average amplitude corresponding to the observation fit curve and the ordinate of the former data point is larger than the ordinate of the latter data point in the two data points corresponding to the amplitude, and recording the times as the second times; taking the ratio of the second times to the first times as a first degree of abnormality of the observation data set;
The analyzing the data characteristic differences among all the observation data sets according to the first credibility of the observation data sets, the number of data in the observation data sets and the related data sets in each history period, and determining the second credibility of the observation data sets in each history period comprises the following steps:
determining an observation fitting curve and a correlation fitting curve of each history period, and further determining extreme points in the observation fitting curve and the correlation fitting curve of each history period; wherein the horizontal axis of the relevant fitting curve is the acquisition time, and the vertical axis is the phosphorus element content data;
determining the reality degree of the observation data set of each historical period according to each extreme point in the observation fitting curve and the correlation fitting curve of the historical period of the same period;
for the observation data set of any history period corresponding to the dissolved oxygen data, determining the second credibility of the observation data set of the history period according to the first credibility, the true degree and the data number of the observation data set of the history period and the first credibility, the true degree and the data number of the observation data sets of other history periods except the history period;
The calculation formula of the actual degree of the observation data set of each history period is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the degree of realism of the observation data set of the nth history period corresponding to the dissolved oxygen data, norm is a linear normalization function, +.>For the average time interval between all adjacent extreme points in the correlation fit curve of the nth history period corresponding to the dissolved oxygen data, +.>For the average time interval between all adjacent extreme points in the observation fit curve of the nth history period corresponding to the dissolved oxygen data, +.>To pair(s)Absolute value is determined for->For->Absolute value is determined for->The number of extreme points in the correlation fitting curve of the nth history period corresponding to the dissolved oxygen data, < +.>The number of extreme points in the observation fitting curve of the nth historical period corresponding to the dissolved oxygen data is counted, and n is the serial number of the historical period;
the calculation formula of the second credibility of the observation data set of the nth historical period corresponding to the dissolved oxygen data is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the second confidence of the observation data set of the nth history period corresponding to the dissolved oxygen amount data, exp is an exponential function based on a natural constant, A is the number of history periods corresponding to the dissolved oxygen amount data, s is the serial number of each history period except the nth history period, and the like >First confidence,/for the observation data set of the nth history period corresponding to dissolved oxygen data>Observation number for nth history period corresponding to dissolved oxygen amount dataBased on the degree of realism of the collection->First credibility of observation data set of the s-th history period except the n-th history period corresponding to dissolved oxygen amount data, +.>For the degree of realism of the observation data set of the other s-th history period except the n-th history period corresponding to the dissolved oxygen amount data,for->Absolute value is determined for->The number of data in the observation data set of the nth history period corresponding to the dissolved oxygen amount data, < +.>The number of data in the observation data set of the s-th history period except the n-th history period corresponding to the dissolved oxygen amount data is +.>To pair(s)Obtaining an absolute value;
the determining the noise representation value of the dissolved oxygen data according to the noise representation value of the error covariance matrix of each history period corresponding to the dissolved oxygen data and the first credibility and the second credibility of the observation data set of each history period comprises the following steps:
for any historical period corresponding to dissolved oxygen data, taking the product of the first credibility and the second credibility of the observation data set of the historical period as a molecule of a ratio, taking the accumulated sum of the products of the first credibility and the second credibility of the observation data set of all the historical periods as the denominator of the ratio, and taking the ratio as the noise expression weight of the error covariance matrix of the historical period;
Calculating the product of the noise representation value and the noise representation weight of the error covariance matrix of the historical period as the corrected noise representation value of the error covariance matrix of the historical period; and taking the accumulated sum of the corrected noise representation values of the error covariance matrix of each history period corresponding to the dissolved oxygen data as the noise representation value of the dissolved oxygen data.
2. The island reef environment multi-parameter monitoring method according to claim 1, wherein for any collection time, the observation data set is used for constructing an error covariance matrix corresponding to dissolved oxygen data of the collection time, the observation data set is composed of a plurality of historical dissolved oxygen data closest to the dissolved oxygen data of the collection time, the plurality of historical dissolved oxygen data are dissolved oxygen data of a plurality of continuous historical collection times in any historical period before the collection time, and the related data set is composed of phosphorus element content data of a plurality of continuous historical collection times in any historical period before the collection time.
3. The method for monitoring the environment of the island according to claim 1, wherein the monitoring the environment of the island to be monitored based on each denoised dissolved oxygen amount data comprises:
If any dissolved oxygen data after denoising is smaller than a preset dissolved oxygen standard, judging that the environment of the island to be monitored is abnormal, otherwise, judging that the environment of the island to be monitored is normal.
4. An island reef environment multi-parameter monitoring system comprising a processor and a memory, wherein the processor is configured to process instructions stored in the memory to implement an island reef environment multi-parameter monitoring method of any one of claims 1 to 3.
CN202410147279.XA 2024-02-02 2024-02-02 Island reef environment multi-parameter monitoring method and system Active CN117668472B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410147279.XA CN117668472B (en) 2024-02-02 2024-02-02 Island reef environment multi-parameter monitoring method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410147279.XA CN117668472B (en) 2024-02-02 2024-02-02 Island reef environment multi-parameter monitoring method and system

Publications (2)

Publication Number Publication Date
CN117668472A CN117668472A (en) 2024-03-08
CN117668472B true CN117668472B (en) 2024-04-05

Family

ID=90086725

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410147279.XA Active CN117668472B (en) 2024-02-02 2024-02-02 Island reef environment multi-parameter monitoring method and system

Country Status (1)

Country Link
CN (1) CN117668472B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109984070A (en) * 2019-05-22 2019-07-09 海南海博海洋产业发展有限公司 A kind of basalt fibre system can shunt the built-in imitative reef deep-sea culture net cage of lifting
WO2021072890A1 (en) * 2019-10-18 2021-04-22 平安科技（深圳）有限公司 Traffic abnormality monitoring method and apparatus based on model, and device and storage medium
CN112710356A (en) * 2020-12-30 2021-04-27 四川工商学院 Leech culture environment monitoring system
CN116541732A (en) * 2023-07-05 2023-08-04 山东金叶智能设备有限公司 Meteorological monitoring system based on ultrasonic data and optimization algorithm
CN117126717A (en) * 2023-05-26 2023-11-28 生态环境部华南环境科学研究所(生态环境部生态环境应急研究所) Intelligent sampling equipment and sampling method for plankton monitoring in seaweed bed ecosystem

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109984070A (en) * 2019-05-22 2019-07-09 海南海博海洋产业发展有限公司 A kind of basalt fibre system can shunt the built-in imitative reef deep-sea culture net cage of lifting
WO2021072890A1 (en) * 2019-10-18 2021-04-22 平安科技（深圳）有限公司 Traffic abnormality monitoring method and apparatus based on model, and device and storage medium
CN112710356A (en) * 2020-12-30 2021-04-27 四川工商学院 Leech culture environment monitoring system
CN117126717A (en) * 2023-05-26 2023-11-28 生态环境部华南环境科学研究所(生态环境部生态环境应急研究所) Intelligent sampling equipment and sampling method for plankton monitoring in seaweed bed ecosystem
CN116541732A (en) * 2023-07-05 2023-08-04 山东金叶智能设备有限公司 Meteorological monitoring system based on ultrasonic data and optimization algorithm

Also Published As

Publication number Publication date
CN117668472A (en) 2024-03-08

Similar Documents

Publication Publication Date Title
WO2016179864A1 (en) Fresh water acute standard prediction method based on metal quantitative structure-activity relationship
CN117457096B (en) Dynamic monitoring and adjusting system for simulating carbon dioxide dissolution in ocean acidification device
CN117195135B (en) Water pollution anomaly traceability detection method and system
CN117390569B (en) Water environment monitoring method based on data processing
CN118035780B (en) River state evaluation method and system based on multidimensional sensor
CN114355846B (en) Fault diagnosis method for papermaking sewage treatment process based on SBR simulation model
CN117668472B (en) Island reef environment multi-parameter monitoring method and system
CN114037202A (en) Multi-index-associated river water quality evaluation method, system, equipment and storage medium
CN112147202B (en) Electrochemical-based soil nitrate nitrogen real-time detection system and method
CN112711052A (en) GNSS coordinate sequence step detection improvement method and system based on continuous t inspection
CN110736712A (en) method for estimating photosynthetic effective radiation attenuation coefficient of lake
CN214122000U (en) Floating algae detection equipment
CN113836813B (en) Blast furnace tuyere water leakage detection method based on data analysis
CN113570265A (en) Offshore eutrophication evaluation system
You et al. Handling sparse longitudinal data with irregular missing data-analysis of fecal coliform bacteria data
CN112683860A (en) Floating algae detection equipment and detection method
CN114330951A (en) River and lake water quality safety assessment method based on fuzzy comprehensive evaluation
CN117030966B (en) Ecological environment intelligent monitoring method and system based on big data
CN109632042A (en) A kind of high-precision water level detecting system
CN116307361B (en) Quick adjustment and control method for aquaculture environmental factors
CN117290559B (en) Water treatment agent content monitoring method and system
CN116539831B (en) Water environment data monitoring processing method based on big data analysis
CN118258976B (en) Sea area water quality monitoring method and system based on Beidou short message
CN116908398A (en) Device and method for detecting transparency of water body by using turbidity and phosphate sensor
CN118518841B (en) Detection method for pollutants in aquatic products

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