CN116337819B - Inversion method of water body chemical oxygen demand concentration - Google Patents
Inversion method of water body chemical oxygen demand concentration Download PDFInfo
- Publication number
- CN116337819B CN116337819B CN202310309703.1A CN202310309703A CN116337819B CN 116337819 B CN116337819 B CN 116337819B CN 202310309703 A CN202310309703 A CN 202310309703A CN 116337819 B CN116337819 B CN 116337819B
- Authority
- CN
- China
- Prior art keywords
- oxygen demand
- chemical oxygen
- water body
- water
- decision tree
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 104
- 239000000126 substance Substances 0.000 title claims abstract description 74
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 title claims abstract description 72
- 239000001301 oxygen Substances 0.000 title claims abstract description 72
- 229910052760 oxygen Inorganic materials 0.000 title claims abstract description 72
- 238000000034 method Methods 0.000 title claims abstract description 36
- 239000013598 vector Substances 0.000 claims abstract description 36
- 238000007637 random forest analysis Methods 0.000 claims abstract description 33
- 238000003066 decision tree Methods 0.000 claims description 39
- 238000012549 training Methods 0.000 claims description 26
- 238000001228 spectrum Methods 0.000 claims description 25
- 230000003595 spectral effect Effects 0.000 claims description 14
- 238000005070 sampling Methods 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 3
- 230000001351 cycling effect Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000013138 pruning Methods 0.000 claims description 3
- 238000000985 reflectance spectrum Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 19
- 230000007613 environmental effect Effects 0.000 abstract description 12
- 230000008859 change Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 239000013049 sediment Substances 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- IOVCWXUNBOPUCH-UHFFFAOYSA-M Nitrite anion Chemical compound [O-]N=O IOVCWXUNBOPUCH-UHFFFAOYSA-M 0.000 description 2
- UCKMPCXJQFINFW-UHFFFAOYSA-N Sulphide Chemical compound [S-2] UCKMPCXJQFINFW-UHFFFAOYSA-N 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- BAUYGSIQEAFULO-UHFFFAOYSA-L iron(2+) sulfate (anhydrous) Chemical compound [Fe+2].[O-]S([O-])(=O)=O BAUYGSIQEAFULO-UHFFFAOYSA-L 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000036284 oxygen consumption Effects 0.000 description 2
- 230000029058 respiratory gaseous exchange Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000003911 water pollution Methods 0.000 description 2
- 241000195493 Cryptophyta Species 0.000 description 1
- 206010021143 Hypoxia Diseases 0.000 description 1
- 241000192710 Microcystis aeruginosa Species 0.000 description 1
- 238000002835 absorbance Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000012851 eutrophication Methods 0.000 description 1
- 230000007954 hypoxia Effects 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 244000005700 microbiome Species 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000005416 organic matter Substances 0.000 description 1
- 230000029553 photosynthesis Effects 0.000 description 1
- 238000010672 photosynthesis Methods 0.000 description 1
- 230000001766 physiological effect Effects 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000010865 sewage Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002351 wastewater Substances 0.000 description 1
Classifications
-
- 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
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/55—Specular reflectivity
- G01N21/552—Attenuated total reflection
-
- 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
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/94—Investigating contamination, e.g. dust
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a water body chemical oxygen demand concentration inversion method, which comprises the following steps: s10, collecting water body data of a to-be-measured point, and encoding the water body data into a water body characteristic vector; s20, inputting the water feature vector into an improved random forest to obtain the water chemical oxygen demand concentration of the to-be-measured point. The inversion method of the water body chemical oxygen demand concentration reduces the influence of environmental meteorological factors on the inversion of the chemical oxygen demand concentration, has higher applicability, high precision, high flexibility and strong robustness, can be used as an auxiliary means for monitoring the water body chemical oxygen demand, improves the breadth and quality of monitoring the water body chemical oxygen demand, saves manpower and financial resources, and improves the efficiency.
Description
Technical Field
The invention belongs to the technical field of water quality monitoring, and particularly relates to a water body chemical oxygen demand concentration inversion method.
Background
Water is the source of life and the health of the water environment is closely related to the fate of humans. In recent years, with the rapid development of economy and the continuous increase of human activities, the water quality of inland water bodies in China is further deteriorated, the problem of water eutrophication is increasingly serious, the phenomena of water bloom outbreak, water body hypoxia and the like occur in some water bodies, the water quality is already in class V or class V, and the chemical oxygen demand is one of main pollution indexes. The chemical oxygen demand (CODcr) value can be used for indicating the amount of reducing substances such as organic matters, nitrite, sulfide, ferrous salt and the like in water and indicating the pollution degree of the reducing substances such as organic matters, nitrite, sulfide, ferrous salt and the like in water.
The large amount of urban domestic sewage and industrial and agricultural wastewater each year causes a large amount of reducing substances to directly or indirectly enter inland water bodies, so that the concentration of chemical oxygen demand in the water bodies is higher. The frequent occurrence of water pollution events seriously threatens the daily activities and life health of people, and water restoration and water environment protection are urgent. The chemical oxygen demand concentration is a main evaluation index of the water pollution degree, and the accurate acquisition of the chemical oxygen demand concentration can provide important technical support for the accurate control of pollutants and the restoration of the water.
At present, the monitoring of the chemical oxygen demand concentration of the water body has the modes of manual mobile monitoring, fixed site monitoring, remote sensing monitoring and the like. The manual mobile monitoring relies on field sampling and laboratory analysis, has large workload and low efficiency, can only monitor a limited area in a specific time period, and is difficult to reflect the water quality condition of the whole area; the fixed site monitoring can realize fixed-point real-time on-line monitoring through conversion of absorbance at a specific wavelength, and the requirement of large-range synchronous dynamic monitoring cannot be met; the remote sensing monitoring comprises monitoring means such as visible light, infrared spectrum and the like, and has the advantages of quick data acquisition, wide coverage range, synchronous observation and the like. The spectrum remote sensing can acquire abundant and fine spectrum information, is sensitive to the awareness of elements such as the chemical oxygen demand and the total nitrogen of the water body, has higher estimation precision, and becomes the most effective remote sensing monitoring means at present. The satellite spectrum remote sensing cost is higher, the resolution is lower, and the popularization and the application are difficult. The ground spectrum remote sensing has the characteristics of low cost, simple operation, less influence by weather, no need of atmospheric correction and the like, and is an effective acquisition means for quantitative research of water parameters.
Research shows that the change of wind speed breaks the balance relation between the sediment and water, so that the sediment is resuspended, and organic matters in the sediment are put into the water, so that the chemical oxygen demand concentration is influenced. The temperature change can influence the growth of algae, aquatic plants and microorganisms in the water body, and the growth and death of the substances can influence the change of the chemical oxygen demand concentration, so that the chemical oxygen demand seasonal change is also influenced to a certain extent. Besides the certain influence of temperature on the photosynthesis speed, the influence on the breathing oxygen consumption of aquatic organisms and the decomposition speed of organic matters in water is more remarkable, so that the oxygen consumption rate of the water body is restricted. The higher the water temperature, the stronger the physiological activity of the living beings, and the more oxygen is consumed by respiration; the faster the organic matter in the water is decomposed, the more dissolved oxygen is consumed.
However, the data used in the current remote sensing inversion of the chemical oxygen demand concentration is mostly limited to the water body reflection spectrum, the research on combining the environmental meteorological elements is very little, and the chemical oxygen demand concentration of the water body is influenced by the environmental meteorological elements and has obvious regional differences. Therefore, a method for remotely sensing the concentration of the chemical oxygen demand of the water body considering the influence of environmental meteorological elements is urgently needed.
Disclosure of Invention
Based on the existing problems, the invention provides an inversion method of the water body chemical oxygen demand concentration, which can be used as an auxiliary method of the traditional water quality monitoring means, saves labor and financial resources, and improves the breadth, quality and efficiency of water body chemical oxygen demand concentration monitoring.
The invention provides a water body chemical oxygen demand concentration inversion method, which comprises the following steps:
s10, collecting water body data of a to-be-measured point, and encoding the water body data into a water body characteristic vector;
s20, inputting the water feature vector into an improved random forest to obtain the water chemical oxygen demand concentration of the to-be-measured point.
Further, the improved training method of the random forest comprises the following steps:
s21, generating t training subsets from a sample training set by adopting a sampling technology, and generating a corresponding decision tree { h } by utilizing each training subset 1 (x),h 2 (x),…,h t (x) -these decision trees constitute a random forest;
s22, q features are selected from the water feature vector, optimal attributes are selected for splitting and growing according to a preset rule at each node of the decision tree, and pruning is not performed during the splitting;
s23, calculating each decision tree h i (x) Weights of (2);
s24, inputting training sample data X and the chemical oxygen demand concentration Y corresponding to the training sample data X into a random forest to obtain a chemical oxygen demand concentration prediction result, wherein the chemical oxygen demand concentration prediction result is expressed as follows:
in the formula, the prediction result of each decision treeMultiplying the corresponding weights respectively, and adding, preferably, normalizing Pi in the formula;
and S25, evaluating the performance of the random forest to obtain the optimal random forest.
Further, decision tree h i (x) The weight is the construction of decision tree h i (x) The sum of absolute values of correlation coefficients corresponding to all features of (a).
Further, the correlation coefficient is the pearson coefficient of the ith feature and the chemical oxygen demand concentration.
Further, in S25, the evaluating method includes:
steps S21-S24 are executed by cycling through t and q, the output predicted value of the chemical oxygen demand concentration is compared with the true value, and t and q which show the best performance are selected as parameters of the final random forest; wherein, t ranges from 1 to the number of samples p in the sample training set, and q ranges from 1 to the number of water feature vectors n.
Further, in step S10, the water body data includes water body reflectance spectra, meteorological data, and/or chemical oxygen demand concentration.
Further, in step S10, the water body reflection spectrum is screened.
Further, in step S10, a continuous projection algorithm is used to screen the reflection spectrum features of the water body.
Further, in step S10, a plurality of spectral curves are obtained at the same sample sampling point, and then an average value of the curves is taken as a final water body reflection spectral curve of the sample.
Further, in step S10, the spectral curves acquired at the same sample point are smoothed and denoised.
The inversion method of the water body chemical oxygen demand concentration has the following beneficial effects:
(1) The invention smoothes and normalizes the spectrum, reduces the influence of background noise on the inversion of the chemical oxygen demand, and the spectrum characteristic selection further screens the optimized wavelength according to the contribution value of the optimized wavelength to the chemical oxygen demand, eliminates insensitive wavelength and reduces the complexity of the model.
(2) Aiming at the problems that the current water quality monitoring is limited to the water body reflection spectrum and the influence of the environmental elements on inversion precision is ignored, the environmental elements are considered in the modeling process, the influence of the environmental elements on the inversion effect of the chemical oxygen demand concentration is solved, and the constructed model has the characteristics of high prediction precision, good adaptability to environmental differences and the like.
(3) Aiming at the problem that decision trees with different generalization capacities in the traditional random forest have the same weight, the method for improving the random forest is applied to the field of spectrum data processing, so that the prediction of the chemical oxygen demand concentration is realized, and the inversion precision and efficiency of the chemical oxygen demand concentration are improved.
(4) The method provided by the invention has the advantages of higher applicability, high precision, high flexibility and strong robustness, provides a new thought for monitoring the concentration of the chemical oxygen demand of the water body, and improves the breadth, quality and efficiency of the chemical oxygen demand monitoring.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and other drawings may be obtained according to these drawings without inventive effort for a person of ordinary skill in the art.
FIG. 1 is a flow chart of an inversion method according to an embodiment of the invention;
figure 2 is a schematic diagram of a training process for an improved random forest in accordance with one embodiment of 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 of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The inversion method of the water body chemical oxygen demand concentration, as shown in figure 1, comprises the following steps:
s10, collecting water body data of a to-be-measured point, and encoding the water body data into a water body characteristic vector;
s20, inputting the water feature vector into an improved random forest to obtain the water chemical oxygen demand concentration of the to-be-measured point.
In step S10, the water body data includes water body reflectance spectra, meteorological data, and/or chemical oxygen demand concentration. The water body reflection spectrum comprises a plurality of wave bands, and the meteorological data comprise wind speed, temperature and the like. As described in the background, weather also has a large impact on the chemical oxygen demand of the water. The data is converted into vectors, such as by word2vec, one-hot coding, etc.
During inversion, the water body data comprise reflection spectrum and meteorological data, and during training, the water body data comprise reflection spectrum, meteorological data and chemical oxygen demand concentration, and the chemical oxygen demand concentration at the moment is obtained through analysis of a water sample in a laboratory.
The data set required for improved random forest training may be collected at a body of water, such as a lake. For example, the water body reflection spectrum, the environmental meteorological elements and the water sample of a plurality of sample points in the water body are collected, the water sample is sent to a laboratory for analysis on the same day to obtain the chemical oxygen demand concentration, preferably, a plurality of spectrum curves are obtained by the same sample sampling point, and then the average value of the curves is taken as the final water body reflection spectrum curve of the sample. Multiple spectral curves obtained from the same sample point may also be smoothed and denoised, for example, each spectral curve may be smoothed using the Savitzky-Golay smoothing method (although there are many methods for smoothing denoising, such as polynomial fitting).
In one embodiment, the water feature vector may also be normalized, e.g., scaled equally, i.e., the feature vector is converted to a range of [0,1] by a linear function, as follows:
wherein X is norm For the normalized eigenvectors, X is the water body eigenvector (including the reflection spectrum eigenvector and the meteorological eigenvector) and max respectively the maximum value of the characteristic vector, X min Is characterized byMinimum of the amount.
In one embodiment, where the number of spectral features is large, such as where the spectrum contains 270 bands, each band is a feature, such a calculation is large. The invention preferably screens the spectrum characteristics to reduce the calculated amount of the random forest in the later stage of the invention. The spectral feature selection uses a continuous projection algorithm, which is a forward loop selection method, starting from one wavelength and calculating its projection onto the unselected wavelengths each time, introducing the wavelength with the largest projection vector into the wavelength combination until the loop a times. Each newly selected wavelength has a minimal linear relationship with the previous one. Of which a spectral features are selected as the final spectral feature selection result. The continuous projection algorithm can also be implemented on the normalized result of the water spectrum.
The continuous projection algorithm steps are as follows:
s11: initializing, optionally selecting a column of vectors PS in a spectrum matrix j Denoted as PS k(0) (i.e., k (0) =j);
s12: the set S is defined as:i.e. column vectors not yet selected for the wavelength chain, respectively calculate PS j Projection vector to vector in S:
s13: recording the serial number of the maximum projection:
s14: taking the maximum projection as the projection vector of the lower wheel
S15: q=q+1, if q < n, go back to step 2 to continue projection.
All normalized characteristic water body reflection characteristic vectors PS and wind speed characteristic vectors U obtained by p sample points norm Temperature characteristic vector T norm Composition independent variableWherein p is the total number of sample points, and comprises m selected spectral features and two environmental features of temperature and wind speed, and n features in total. For convenience of representation, let n=m+2, then X be a two-dimensional matrix of p rows and n columns, X ij I.e. the j-th feature representing the i-th sample.
The continuous projection algorithm is an existing method and will not be described in detail.
In step S20, the water feature vector is formed by splicing the screened spectral feature vector, wind speed feature vector and temperature feature vector to form a feature vector X, and the number of the water feature vectors is set to n. The feature vector X and the chemical oxygen demand concentration Y are used as training sample sets to be input into an improved random forest (which can be composed of 100-200 decision trees, wherein the decision trees are composed of different features through various combinations, and the number of the decision trees can be adjusted according to actual conditions), and then the predicted chemical oxygen demand concentration of the water body is output.
The traditional random forest consists of a plurality of decision trees, and for the random forest { h } 1 (x),h 2 (x),…,h t (x) And the input vector X is arranged, each decision tree carries out result prediction on X relatively independently, after the random forest model obtains the predicted results of all decision trees, the predicted result of the whole model is given out through a set statistical rule, and generally, for regression problems, the random forest takes the average number of the predicted results given out by all decision trees as the final predicted result. The invention distributes the weight of the decision tree through the correlation coefficient index aiming at the unreasonable mechanism that decision trees with different prediction capabilities have the same weight, thereby improving the overall prediction accuracy of the model.
The training method of the improved random forest, as shown in fig. 2, comprises the following steps:
s21, generating from the sample training set by adopting sampling technologyt training subsets, each training subset is utilized to generate a corresponding decision tree { h } 1 (x),h 2 (x),…,h t (x) -these decision trees constitute a random forest;
s22, q features are selected from n feature vectors, optimal attributes are selected for splitting at each node of the decision tree according to a preset rule, each decision tree grows to the maximum extent, and pruning is not carried out during the process of splitting completely;
steps S21 and S22 are the same as the conventional random forest method, and will not be described here again.
S23, for each decision tree h i (x) Correlation coefficient R corresponding to the selected feature i ＝{R i1 ,R i2 ,…,R iq }，R i E, R, calculating a weight value:
wherein p is i To construct decision tree h i (x) Sum of absolute values of correlation coefficients corresponding to all features of R ij For decision tree h i (x) Correlation coefficient corresponding to the j-th feature. The training subsets used are different for each decision tree, and q features are used, but the features used for each decision tree are different, i.e. are equivalent to different sample combinations and feature combinations.
Wherein, the correlation coefficient Ri is the Pirson coefficient of the ith characteristic and the chemical oxygen demand concentration, and the calculation method comprises the following steps:
the total number of the samples is p, the characteristic number is m, the samples correspond to the chemical oxygen demand concentration Y, and the formula is adoptedWherein R is j For the j-th characteristic and the correlation coefficient of chemical oxygen demand,/->And->The mean value of the statistics of the j-th feature in X and Y is calculated to obtain a pearson correlation coefficient set R= { R 1 ,R 2 ,…,R m },m∈{1,2,…,n}。
S24, inputting training sample data X and the chemical oxygen demand concentration Y corresponding to the training sample data X into a random forest to obtain a chemical oxygen demand concentration prediction result, wherein the chemical oxygen demand concentration prediction result is expressed as follows:
in the formula, the prediction result of each decision treeRespectively multiplied by the corresponding weights p i And adding to obtain the final chemical oxygen demand concentration. Preferably, pi is normalized first.
S25, evaluating the model performance. The method comprises the following steps: each possibility is tried by cycling through t and q, comparing the output predicted value of the chemical oxygen demand concentration with the actual value, and selecting the parameter with the best performance as the final result (the adjustment optimization belongs to the prior art), wherein the range of t is 1 to the number of samples p in the sample training set, and the range of q is 1 to the number of water body characteristic vectors n.
In one embodiment, the index of the evaluation is a decision coefficient R 2 The calculation method comprises the following steps:
wherein: p represents the number of samples; y is i Representing the measured value of the chemical oxygen demand of the ith sample;representing an i-th sample chemical oxygen demand estimate; />The average of the measured values of chemical oxygen demand of p samples is shown.
Comparing the result of the invention with a traditional random forest chemical oxygen demand inversion model without considering environmental characteristics and a traditional random forest chemical oxygen demand inversion model with considering environmental characteristics, wherein the model precision is shown in table 1:
table 1 model accuracy comparison
Those of ordinary skill in the art will appreciate that: 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 or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions, which are defined by the scope of the appended claims.
Claims (8)
1. A method for inverting the chemical oxygen demand concentration of a body of water, the method comprising:
s10, collecting water body data of a to-be-measured point, and encoding the water body data into a water body characteristic vector; the water body data comprises reflection spectrum and meteorological data;
s20, inputting the water feature vector into an improved random forest to obtain the water chemical oxygen demand concentration of the to-be-measured point; in the improved random forest of the present invention,
for each decision tree h i (x) Decision tree h i (x) The weight is the construction of decision tree h i (x) The sum of absolute values of pearson coefficients corresponding to all features of (a);
the result of chemical oxygen demand concentration prediction is expressed as:
in the formula, the prediction result of each decision treeRespectively multiplied by the corresponding weights p i And after addition, t is the number of decision trees.
2. The inversion method of claim 1, wherein the improved training method of random forests comprises:
s21, generating t training subsets from a sample training set by adopting a sampling technology, and generating a corresponding decision tree { h } by utilizing each training subset 1 (x),h 2 (x),…,h t (x) -these decision trees constitute a random forest;
s22, q features are selected from the water feature vectors, optimal attributes are selected for splitting at each node of the decision tree according to a preset rule, the water feature vectors grow to the maximum extent, and pruning is not performed during the splitting;
s23, calculating each decision tree h i (x) Weight p of (2) i The method comprises the steps of carrying out a first treatment on the surface of the Decision tree h i (x) The weight is the construction of decision tree h i (x) The sum of absolute values of pearson coefficients corresponding to all features of (a);
s24, inputting training sample data X and the chemical oxygen demand concentration Y corresponding to the training sample data X into a random forest to obtain a chemical oxygen demand concentration prediction result, wherein the chemical oxygen demand concentration prediction result is expressed as follows:
in the formula, the prediction result of each decision treeMultiplying the corresponding weights respectively and adding;
s25, evaluating the performance of the random forest to obtain an optimal random forest; the evaluation method comprises the following steps: steps S21-S24 are executed by cycling through t and q, the output predicted value of the chemical oxygen demand concentration is compared with the true value, and t and q which show the best performance are selected as parameters of the final random forest; wherein, t ranges from 1 to the number of samples p in the sample training set, and q ranges from 1 to the number of water feature vectors n.
3. Inversion method according to claim 1, wherein in step S10 the water body data comprises water body reflectance spectra, meteorological data and/or chemical oxygen demand concentration.
4. An inversion method according to claim 3, wherein in step S10, the reflection spectrum of the water body is screened.
5. An inversion method according to claim 3, wherein in step S10, the water reflection spectrum characteristics are screened using a continuous projection algorithm.
6. The inversion method according to claim 1, wherein in step S10, a plurality of spectral curves are obtained at the same sample sampling point, and then an average value of the curves is taken as a final water reflection spectral curve of the sample.
7. The inversion method according to claim 6, wherein in step S10, the spectral curves obtained at the same sample point are smoothed and denoised.
8. Inversion method according to claim 2, characterized in that p is set in S24 i Normalization was performed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310309703.1A CN116337819B (en) | 2023-03-27 | 2023-03-27 | Inversion method of water body chemical oxygen demand concentration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310309703.1A CN116337819B (en) | 2023-03-27 | 2023-03-27 | Inversion method of water body chemical oxygen demand concentration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116337819A CN116337819A (en) | 2023-06-27 |
| CN116337819B true CN116337819B (en) | 2023-10-27 |
Family
ID=86878443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310309703.1A Active CN116337819B (en) | 2023-03-27 | 2023-03-27 | Inversion method of water body chemical oxygen demand concentration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116337819B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108038351A (en) * | 2017-12-06 | 2018-05-15 | 武汉大学 | A kind of chlorophyll computational methods based on remote sensing image and water ecological model |
| CN109409441A (en) * | 2018-11-16 | 2019-03-01 | 福州大学 | Based on the coastal waters chlorophyll-a concentration remote sensing inversion method for improving random forest |
| CN111797186A (en) * | 2020-06-17 | 2020-10-20 | 南开大学 | Method for inverting COD (chemical oxygen demand) of water body by remote sensing |
| CN112816421A (en) * | 2021-01-25 | 2021-05-18 | 中国科学院南京地理与湖泊研究所 | Land-based remote sensing monitoring method for nutritive salt and chemical oxygen demand of water body |
| CN115690632A (en) * | 2022-11-09 | 2023-02-03 | 中科合肥智慧农业协同创新研究院 | Water environment monitoring method for inland river water body |
-
2023
- 2023-03-27 CN CN202310309703.1A patent/CN116337819B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108038351A (en) * | 2017-12-06 | 2018-05-15 | 武汉大学 | A kind of chlorophyll computational methods based on remote sensing image and water ecological model |
| CN109409441A (en) * | 2018-11-16 | 2019-03-01 | 福州大学 | Based on the coastal waters chlorophyll-a concentration remote sensing inversion method for improving random forest |
| CN111797186A (en) * | 2020-06-17 | 2020-10-20 | 南开大学 | Method for inverting COD (chemical oxygen demand) of water body by remote sensing |
| CN112816421A (en) * | 2021-01-25 | 2021-05-18 | 中国科学院南京地理与湖泊研究所 | Land-based remote sensing monitoring method for nutritive salt and chemical oxygen demand of water body |
| CN115690632A (en) * | 2022-11-09 | 2023-02-03 | 中科合肥智慧农业协同创新研究院 | Water environment monitoring method for inland river water body |
Non-Patent Citations (1)
| Title |
|---|
| 基于机器学习的水体化学需氧量高光谱反演模型对比研究;王春玲等;光谱学与光谱分析;第42卷(第8期);第2353-2358页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116337819A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110309985B (en) | Crop yield prediction method and system | |
| CN110969346B (en) | River basin water ecological function partition treatment demand evaluation method based on index screening | |
| Silsbe et al. | Variability of chlorophyll a and photosynthetic parameters in a nutrient‐saturated tropical great lake | |
| CN109828089A (en) | DBN-BP-based water quality parameter nitrous acid nitrogen online prediction method | |
| CN110569605B (en) | NSGA 2-ELM-based non-glutinous rice leaf nitrogen content inversion model method | |
| CN112666328B (en) | Chlorophyll a remote sensing inversion method and device based on genetic neural network model | |
| CN114239422B (en) | Method for improving marine chlorophyll a concentration prediction accuracy based on machine learning | |
| CN110069032A (en) | A kind of eggplant greenhouse intelligent checking system based on wavelet neural network | |
| CN114511550A (en) | Poyang lake wetland vegetation net primary productivity remote sensing estimation method | |
| CN113049499A (en) | Indirect remote sensing inversion method for water total nitrogen concentration, storage medium and terminal equipment | |
| CN115951025B (en) | Inversion method for total phosphorus concentration of water body | |
| CN116337819B (en) | Inversion method of water body chemical oxygen demand concentration | |
| Liu et al. | Chlorophyll a estimation in lakes using multi-parameter sonde data | |
| CN116338819A (en) | Water dissolved oxygen concentration prediction system | |
| CN116432849A (en) | Leaf area index prediction method based on primary productivity data assimilation | |
| CN116223756B (en) | Method for generating water body nitrogen content prediction model | |
| CN115080905A (en) | Remote sensing inversion method for chlorophyll a concentration of plateau lake | |
| Markos et al. | Ecophysiological modelling of leaf level photosynthetic performance for three Mediterranean species with different growth forms | |
| CN114864015A (en) | Water eutrophication detection method, device, equipment and storage medium | |
| CN113155751A (en) | Remote sensing identification method for POC vertical structure type of eutrophic lake | |
| CN115078261A (en) | Inland lake total phosphorus concentration remote sensing monitoring method based on random forest model | |
| CN108734375B (en) | Method for estimating contribution rate of salvage engineering to lake aquatic vegetation group change | |
| Yamamoto et al. | Three-layered Feedforward artificial neural network with dropout for short-term prediction of class-differentiated Chl-a based on weekly water-quality observations in a eutrophic agricultural reservoir | |
| Gu et al. | Monitoring the chlorophyll fluorescence parameters in rice under flooding and waterlogging stress based on remote sensing | |
| Ma et al. | Crop Water Requirement Prediction Method Based on EEMD-Attention-LSTM Model |
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 |