CN115824993A - Method and device for determining chemical oxygen demand of water body, computer equipment and medium - Google Patents
Method and device for determining chemical oxygen demand of water body, computer equipment and medium Download PDFInfo
- Publication number
- CN115824993A CN115824993A CN202310107553.6A CN202310107553A CN115824993A CN 115824993 A CN115824993 A CN 115824993A CN 202310107553 A CN202310107553 A CN 202310107553A CN 115824993 A CN115824993 A CN 115824993A
- Authority
- CN
- China
- Prior art keywords
- water
- sample
- water sample
- waveform
- different
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- 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
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The embodiment of the invention provides a method, a device, computer equipment and a medium for determining the chemical oxygen demand of a water body, and relates to the technical field of water quality detection, wherein the method comprises the following steps: calculating similarity and clustering of water sample spectral data of different sample water bodies, dividing different water sample waveform categories, and further training and identifying a model; acquiring spectral data to be detected of a water body in an area to be monitored; determining the type of the waveform to be detected through an identification model according to the spectral data to be detected; determining a cod calculation model corresponding to the waveform type to be detected according to the corresponding relation between the waveform type and the cod calculation model; and according to the spectral data to be detected, obtaining the chemical oxygen demand of the water body of the area to be monitored through a cod calculation model corresponding to the waveform type to be detected. The method is beneficial to improving the accuracy of the calculation result of the chemical oxygen demand of the water body of the area to be monitored and reducing the workload and the cost for calculating the chemical oxygen demand of the water body of the area to be monitored.
Description
Technical Field
The invention relates to the technical field of water quality detection, in particular to a method and a device for determining the chemical oxygen demand of a water body, computer equipment and a medium.
Background
At present, a chemical method is mainly used for determining the cod (chemical oxygen demand) value in water, and the traditional chemical cod monitoring method has long analysis period and serious secondary pollution, and cannot meet the requirements of modern water quality monitoring on real-time performance and no secondary pollution. In the existing method for detecting water quality cod by using a spectrum method at home and abroad, the cod value is calculated by mainly measuring the ultraviolet absorbance at a single wavelength of 254nm, the method is rapid and convenient without pretreatment, but the deviation of the final measured value is large due to the diversity and complexity of factors influencing the water quality cod; and aiming at the pollution source calculation methods in different areas, the method has great difference, and a cod calculation model is reestablished in a new area by collecting water samples.
For water samples in different regions, the problem of limited adaptability and accuracy exists by using the same cod calculation model. However, the spectral cod calculation model is established in a new region, resampling and modeling are needed, the workload is large, and the cost is high.
Disclosure of Invention
In view of this, the embodiment of the invention provides a method for determining a chemical oxygen demand of a water body, so as to solve the technical problems of inaccuracy and large workload in cod detection in a new region in the prior art. The method comprises the following steps:
collecting water sample spectrum data of different sample water bodies through a standard station;
clustering water sample spectrum data of different sample water bodies to mark out different water sample waveform categories;
taking water sample spectrum data of different sample water bodies and corresponding water sample waveform categories as samples, and training a BP neural network to obtain an identification model;
acquiring spectral data to be detected of a water body in an area to be monitored;
determining the type of the waveform to be detected through the identification model according to the spectral data to be detected;
determining a cod calculation model corresponding to the waveform type to be detected according to the corresponding relation between the waveform type and the cod calculation model;
according to the spectral data to be detected, obtaining the chemical oxygen demand of the water body of the area to be monitored through a cod calculation model corresponding to the waveform type to be detected;
the water sample spectral data of different sample water bodies are clustered, and different water sample waveform categories are divided, and the method comprises the following steps:
calculating the similarity between water sample spectral data of different sample water bodies;
based on the similarity between the water sample spectrum data of different sample water bodies, clustering the water sample spectrum data of different sample water bodies, and dividing different water sample waveform categories;
the similarity between the water sample spectral data of different sample water bodies is calculated, and the method comprises the following steps:
calculating the correlation coefficient between the water sample spectral data of every two different sample water bodies by the following formula:
c=corr(x1,x2)
wherein c is a correlation coefficient, x1 and x2 represent two spectral curves corresponding to two different sample water bodies respectively, and corr () is a pearson correlation coefficient of the two spectral curves;
calculating the deviation between the water sample spectrum data of every two different sample water bodies according to the following formula:
d=x1-x2
wherein d is the deviation;
discretizing each deviation, dividing each discretized deviation into a plurality of intervals, and counting the proportion value of each interval through the following formula, wherein the proportion value is the proportion of the number of the deviations in the total number of the deviations, which is included in each interval, and the length of each interval is a preset length:
Pm=qm/sum(q)
wherein, pm represents the proportional value of the mth interval q, qm represents the number of deviations d contained in the mth interval q, and sum (q) represents the total number of deviations d;
calculating the information entropy of all the deviations according to the proportional values of the intervals by the following formula:
wherein, ent is information entropy, and M is the total number of intervals;
calculating the similarity between the water sample spectrum data of every two different sample water bodies according to the information entropy and the correlation coefficient between the water sample spectrum data of every two different sample water bodies by the following formula:
wherein w is the similarity between the water sample spectral data of every two different sample water bodies.
The embodiment of the invention also provides a device for determining the chemical oxygen demand of the water body, which is used for solving the technical problems of inaccuracy and large workload in cod detection in a new area in the prior art. The device includes:
the sample data acquisition module is used for acquiring water sample spectrum data of different sample water bodies through the standard station;
the clustering module is used for clustering the water sample spectrum data of different sample water bodies to mark out different water sample waveform categories;
the classification model training module is used for training a BP neural network to obtain a recognition model by taking water sample spectrum data of different sample water bodies and corresponding water sample waveform categories as samples;
the spectrum acquisition module is used for acquiring to-be-detected spectrum data of the water body in the to-be-monitored area;
the waveform determining module is used for determining the category of the waveform to be detected through the identification model according to the spectral data to be detected;
the model determining module is used for determining a cod calculation model corresponding to the waveform type to be detected according to the corresponding relation between the waveform type and the cod calculation model;
the calculation module is used for obtaining the chemical oxygen demand of the water body of the area to be monitored through a cod calculation model corresponding to the type of the waveform to be detected according to the spectral data to be detected;
the clustering module comprises:
the similarity calculation unit is used for calculating the similarity between the water sample spectral data of different sample water bodies;
the clustering unit is used for clustering the water sample spectral data of different sample water bodies based on the similarity between the water sample spectral data of different sample water bodies to mark out different water sample waveform categories;
the similarity calculation unit is used for calculating the correlation coefficient between the water sample spectral data of every two different sample water bodies according to the following formula:
c=corr(x1,x2)
wherein c is a correlation coefficient, x1 and x2 represent two spectral curves corresponding to two different sample water bodies respectively, and corr () is a Pearson correlation coefficient of the two spectral curves;
calculating the deviation between the water sample spectrum data of every two different sample water bodies according to the following formula:
d=x1-x2
wherein d is the deviation;
discretizing each deviation, dividing each discretized deviation into a plurality of intervals, and counting the proportion value of each interval through the following formula, wherein the proportion value is the proportion of the number of the deviations in the total number of the deviations, and the length of each interval is a preset length:
Pm=qm/sum(q)
wherein, pm represents the proportional value of the mth interval q, qm represents the number of deviations d contained in the mth interval q, and sum (q) represents the total number of deviations d;
calculating the information entropy of all the deviations according to the proportional value of each interval by the following formula:
wherein, ent is information entropy, and M is the total number of intervals;
calculating the similarity between the water sample spectral data of every two different sample water bodies according to the information entropy and the correlation coefficient between the water sample spectral data of every two different sample water bodies by the following formula:
wherein w is the similarity between the water sample spectral data of every two different sample water bodies.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can be run on the processor, wherein the processor realizes the arbitrary method for determining the chemical oxygen demand of the water body when executing the computer program so as to solve the technical problems of inaccuracy and large workload of cod detection in a new region in the prior art.
The embodiment of the invention also provides a computer readable storage medium, wherein a computer program for executing the method for determining the chemical oxygen demand of the water body is stored in the computer readable storage medium, so that the technical problems of inaccuracy and large workload in the cod detection of the new region in the prior art are solved.
Compared with the prior art, the beneficial effects that can be achieved by the at least one technical scheme adopted by the embodiment of the specification at least comprise: determining the category of the waveform to be detected of the spectral data to be detected of the water body of the area to be monitored through the identification model, further determining the cod calculation model corresponding to the category of the waveform to be detected of the spectral data to be detected of the water body of the area to be monitored according to the corresponding relation between the waveform category and the cod calculation model, and finally obtaining the chemical oxygen demand of the water body of the area to be monitored through the determined cod calculation model according to the spectral data to be detected. The COD calculation model corresponding to the category of the waveform to be detected of the spectral data to be detected can be directly adopted to calculate the COD of the water body aiming at the area to be monitored, and the category of the waveform to be detected of the spectral data to be detected is determined by the water body characteristic or characteristic of the area to be monitored, so that the COD calculation model which is adaptive to or matched with the water body of the area to be monitored can be adopted to calculate the COD of the water body of the area to be monitored, and the accuracy of the calculation result of the COD of the water body of the area to be monitored can be improved; meanwhile, the region to be monitored can be any region needing to detect the chemical oxygen demand, can be a new region without detecting the chemical oxygen demand or a region without establishing a cod calculation model, so that the cod calculation model corresponding to each waveform type can be transmitted in different regions to be monitored, the process of carrying out cod calculation model training by collecting data such as water samples and spectral data again in the region to be monitored is avoided, and the workload and the cost for calculating the chemical oxygen demand of the water body in the region to be monitored are reduced; in addition, in the process of training the recognition model, the similarity is calculated through the water sample spectrum data of the sample water body, then the water sample spectrum data of different sample water bodies are clustered based on the similarity, different water sample waveform categories are divided, and the accuracy of the recognition model can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart of a method for determining chemical oxygen demand of a water body according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a cod calculation model for training water samples according to an embodiment of the present invention;
FIG. 3 is a block diagram of a computer device according to an embodiment of the present invention;
fig. 4 is a block diagram of a device for determining chemical oxygen demand of a water body according to an embodiment of the present invention.
Detailed Description
The embodiments of the present application will be described in detail below with reference to the accompanying drawings.
The following description of the embodiments of the present application is provided by way of specific examples, and other advantages and effects of the present application will be readily apparent to those skilled in the art from the disclosure herein. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. The present application is capable of other and different embodiments and its several details are capable of modifications and/or changes in various respects, all without departing from the spirit of the present application. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In an embodiment of the present invention, a method for determining chemical oxygen demand of a water body is provided, as shown in fig. 1, the method includes:
step S101: collecting water sample spectrum data of different sample water bodies through a standard station;
step S102: clustering water sample spectrum data of different sample water bodies to mark out different water sample waveform categories;
step S103: taking water sample spectrum data of different sample water bodies and corresponding water sample waveform categories as samples, and training a BP neural network to obtain an identification model;
step S104: acquiring spectral data to be detected of a water body in an area to be monitored;
step S105: determining the type of the waveform to be detected through an identification model according to the spectral data to be detected;
step S106: determining a cod calculation model corresponding to the waveform type to be detected according to the corresponding relation between the waveform type and the cod calculation model;
step S107: according to the spectral data to be detected, obtaining the chemical oxygen demand of the water body of the area to be monitored through a cod calculation model corresponding to the waveform type to be detected;
in step S102, the clustering of the water sample spectral data of different sample water bodies to partition different water sample waveform categories includes:
calculating the similarity between water sample spectrum data of different sample water bodies;
based on the similarity between the water sample spectrum data of different sample water bodies, clustering the water sample spectrum data of different sample water bodies, and dividing different water sample waveform categories;
the similarity between the water sample spectral data of different sample water bodies is calculated, and the method comprises the following steps:
calculating the correlation coefficient between the water sample spectral data of every two different sample water bodies by the following formula:
c=corr(x1,x2)
wherein c is a correlation coefficient, x1 and x2 represent two spectral curves corresponding to two different sample water bodies respectively, and corr () is a pearson correlation coefficient of the two spectral curves;
calculating the deviation between the water sample spectrum data of every two different sample water bodies according to the following formula:
d=x1-x2
wherein d is the deviation;
discretizing each deviation, dividing each discretized deviation into a plurality of intervals, and counting the proportion value of each interval through the following formula, wherein the proportion value is the proportion of the number of the deviations in the total number of the deviations, which is included in each interval, and the length of each interval is a preset length:
Pm=qm/sum(q)
wherein, pm represents the proportional value of the mth interval q, qm represents the number of deviations d contained in the mth interval q, and sum (q) represents the total number of deviations d;
calculating the information entropy of all the deviations according to the proportional value of each interval by the following formula:
wherein, ent is information entropy, and M is the total number of intervals;
calculating the similarity between the water sample spectral data of every two different sample water bodies according to the information entropy and the correlation coefficient between the water sample spectral data of every two different sample water bodies by the following formula:
wherein w is the similarity between the water sample spectral data of every two different sample water bodies.
As can be seen from the process shown in fig. 1, in the embodiment of the present invention, the to-be-detected waveform type of the to-be-detected spectral data of the water body in the to-be-monitored area is determined by the identification model, the cod calculation model corresponding to the to-be-detected waveform type of the to-be-detected spectral data of the water body in the to-be-monitored area is determined according to the correspondence between the waveform type and the cod calculation model, and finally, the to-be-detected spectral data is input into the determined cod calculation model, so that the chemical oxygen demand of the water body in the to-be-monitored area can be obtained. The COD calculation model corresponding to the type of the waveform to be detected of the spectral data to be detected can be adopted to calculate the COD of the water body aiming at the area to be monitored, and the type of the waveform to be detected of the spectral data to be detected is determined by the water body characteristic or characteristic of the area to be monitored, so that the COD calculation model which is adaptive to or matched with the water body of the area to be monitored can be adopted to calculate the COD of the water body of the area to be monitored, and the accuracy of the calculation result of the COD of the water body of the area to be monitored can be improved; meanwhile, the transmission of the cod calculation model in different areas to be monitored is realized, the process of performing cod calculation model training by collecting water samples, spectral data and other data again in the areas to be monitored is avoided, and the workload and the cost for calculating the chemical oxygen demand of the water body in the areas to be monitored are reduced; in addition, in the process of training the recognition model, the similarity is calculated through the water sample spectrum data of the sample water body, then the water sample spectrum data of different sample water bodies are clustered based on the similarity, different water sample waveform categories are divided, and the accuracy of the recognition model can be improved.
In specific implementation, the region to be monitored can be any region in which chemical oxygen demand needs to be detected, can be a new region in which chemical oxygen demand is not detected or is unknown, or can be a region in which a cod calculation model is not established, that is, the region to be monitored can be a cod calculation model in which a water body is not yet adapted or matched, and a region in which a water sample cod calculation model is not established by collecting water samples and spectral data.
In the embodiment, in the process of training the recognition model, water sample spectrum data of different sample water bodies are acquired through a standard station, namely the water sample spectrum data of different sample water bodies are acquired through the standard station; for example, the spectral sensors can be placed at standard stations (national control stations, city control stations, etc.) to measure the water sample spectral data of different sample water bodies. The different sample water bodies refer to water body samples in different areas.
In specific implementation, in order to realize the classification of water sample waveforms, various clustering algorithms can be directly adopted to classify the water sample waveforms, and in order to further realize the accurate classification of the water sample waveforms, the method for classifying the water sample waveforms based on the similarity is provided, so that water sample spectrum data with similar waveforms are classified into the same water sample waveform class, for example, the similarity between water sample spectrum data of different sample water bodies is calculated; and based on the similarity between the water sample spectrum data of different sample water bodies, clustering the water sample spectrum data of different sample water bodies, and dividing different water sample waveform categories.
In specific implementation, in the process of calculating the similarity between the water sample spectral data of different sample water bodies, the distance between the water sample spectral data can be calculated to express the similarity, and in order to further improve the calculation precision of the similarity, the method for calculating the similarity between the water sample spectral data is further provided.
Specifically, the process of calculating the similarity between the water sample spectral data may include the following steps:
1. preprocessing the spectral data of the water sample;
2. calculating a correlation coefficient c between water sample spectral data of every two different sample water bodies;
c=corr(x1,x2)
wherein x1 and x2 are two spectral curves respectively corresponding to two different sample water bodies, and corr () is a Pearson correlation coefficient of the two spectral curves;
3. calculating deviation d between water sample spectral data of every two different sample water bodies to obtain a plurality of deviations d;
d=x1-x2；
4. discretizing each deviation d, dividing each discretized deviation d into M intervals, wherein the length of each interval is 0.1, counting a proportion value Pm of each interval, wherein the proportion value Pm is the proportion of the number qm of the deviations d in each interval q in the total number of the deviations d, and the proportion value Pm of each interval forms a data set P;
P=[P1,P2,…Pm,…,PM]
Pm=qm/sum(q)
wherein, pm represents the proportional value of the mth interval q, qm represents the number of deviations d contained in the mth interval q, and sum (q) represents the total number of deviations d;
5. calculating the information entropy Ent of all the deviations d according to the proportional value Pm of each interval;
6. the similarity between the water sample spectrum data (curves) of every two different sample water bodies is w;
in specific implementation, after the similarity between the water sample spectral data of every two different sample water bodies is obtained through calculation, the water sample spectral data can be further clustered based on the similarity, for example,
constructing a matrix according to the similarity between the water sample spectral data of every two different sample water bodies (namely, the similarity between the water sample spectral data of every two different sample water bodies is one value of the matrix);
constructing a degree matrix according to the matrix;
subtracting the matrix from the degree matrix to obtain a Laplace matrix;
standardizing the Laplace matrix to obtain a standardized Laplace matrix;
calculating eigenvectors corresponding to the minimum k1 (k 1 is a positive integer) eigenvalues in the normalized Laplace matrix;
and forming a characteristic matrix by using the characteristic vectors, taking each row in the characteristic matrix as a sample to obtain a plurality of samples, and clustering by using a clustering model according to the plurality of samples to obtain different water sample waveform categories. Specifically, the clustering model can be implemented by using various clustering algorithms.
Specifically, the process of clustering water sample spectral data of different sample water bodies based on the similarity may include the following steps:
1. constructing a matrix S according to the similarity between the water sample spectral data of every two different sample water bodies;
2. constructing a degree matrix D according to the matrix S;
d is a diagonal matrix, only the main diagonal has value, and the value is the value of each vertex;
wherein, the first and the second end of the pipe are connected with each other,
is the value of the ith vertex on the main diagonal of the degree matrix D, the value of the ith vertex is the sum of the similarity of the ith piece of spectral data and all spectral data except the ith piece of spectral data,
and (3) representing the similarity between the ith piece of spectral data and the jth piece of spectral data, wherein n is the total amount of all spectral data, and n1 is the total amount of all spectral data except the ith piece of spectral data in n.
D=
3. Calculating a Laplace matrix L;
4. to pairThe Laplace matrix L is standardized to obtain a standardized Laplace matrix
；
5. Computing
The eigenvectors f corresponding to the minimum k1 eigenvalues in the matrix respectively;
6. standardizing a matrix formed by the feature vectors F according to rows to finally form an n multiplied by k1 dimensional feature matrix F;
7. and taking each row in the characteristic matrix F as a k 1-dimensional sample to obtain n samples in total, and clustering through a clustering model according to the n samples to obtain k2 water sample waveform categories of a clustering dimension. The specific value of the clustering dimension k2 can be determined according to specific classification requirements.
During specific implementation, after the water sample spectrum data of different sample water bodies are subjected to clustering treatment, the water sample spectrum data with similar waveforms are divided into the same water sample waveform category, and a set Label consisting of the water sample waveform categories corresponding to the water sample spectrum data (spectrum curves) of each sample water body is as follows:
wherein li represents the water sample waveform category of the ith water sample spectrum data, f (X) is a set of all water sample waveform categories, X is a water sample spectrum data set of a certain water sample waveform category, n2 is the quantity of water sample spectrum data included in a certain water sample waveform category,
xi represents the ith sample spectrum data, T represents transposition at the upper right corner,
，
the wavelength of the spectrum data of the xi water sample is jAbsorbance data.
In specific implementation, in order to further improve the accuracy of water sample spectral data clustering, before the clustering is performed by using the clustering model, the water sample spectral data may be preprocessed, which may include but is not limited to: SG smoothing data, constant offset cancellation, normalization.
Firstly, SG smoothing processing is carried out on water sample spectral data to obtain a smoothed spectral value.
The smoothing formula is:
wherein the content of the first and second substances,
is the absorbance under the j-dimensional wavelength of the ith sample spectral data after smoothing, N represents the total number of dimensions of the wavelength j,
showing the absorbance of the i-th water sample spectrum data in the j dimension,
showing the absorbance of the i-th water sample spectrum data under the j-t dimension,
showing the absorbance of the ith sample of water spectrum data in the dimension of j + t,
represents the j-dimensional weight of the ith water sample data,
is the average of the weights and is,
represents the weight of the ith water sample spectrum data in the j + t dimension,
weight of j-t dimension for representing ith water sample spectrum dataAnd (4) heavy.
During specific implementation, the cod calculation models corresponding to different water sample waveform categories can adopt the existing known cod calculation models, and a model library of the cod calculation models corresponding to the water sample waveform categories can be established. Specifically, for the different sample water bodies, after water sample waveform categories of different water sample spectral data are divided, a cod calculation model corresponding to the different water sample waveform categories can be obtained based on the water sample spectral data of the different water sample waveform categories and the corresponding chemical oxygen demand concentrations through training, for example,
collecting the chemical oxygen demand concentrations of different sample water bodies through a standard station, namely collecting the chemical oxygen demand concentrations of the different sample water bodies through the standard station; for example, a spectrum sensor can be placed in a standard station (national control station, municipal control station, etc.), and the spectrum sensor is used for measuring water sample spectrum data of different sample water bodies, and meanwhile, cod concentration data of different sample water bodies measured by a large station are captured, so that the corresponding relation between the water sample spectrum data of each sample water body and the cod concentration data is established;
and training a machine learning component to obtain a cod calculation model corresponding to each water sample waveform category by using the water sample spectral data of each water sample waveform category and the chemical oxygen demand concentration of the water sample corresponding to the water sample spectral data of each water sample waveform category as sample data of each water sample waveform category.
In specific implementation, the process of training the cod calculation model corresponding to each water sample waveform category can be realized through the following steps:
dividing sample data of each water sample waveform category into a plurality of data, taking each data as a test set in sequence, and taking a plurality of data except the test set as a training set, so that a plurality of test sets can be formed, and a plurality of training sets can be formed;
training a plurality of base models by adopting each training set to obtain a plurality of initial cod calculation models, predicting the prediction result of each test set by using each initial cod calculation model, and forming the prediction result of each initial cod calculation model by using the prediction result of each test set;
and training a bp neural network by taking the test set prediction result of each initial cod calculation model as a sample, wherein the trained bp neural network and the plurality of initial cod calculation models form cod calculation models corresponding to each water sample waveform category.
Specifically, as shown in fig. 2, the process of training the cod computation model may include the following steps:
1. for a certain water sample waveform category, taking water sample spectral data of the same water sample waveform category and the chemical oxygen demand concentration of a sample water body corresponding to the water sample spectral data as sample data of the certain water sample waveform category, and dividing the sample data into multiple data, for example, dividing the sample data of the certain waveform category into 5 parts;
2. for each base model, sequentially taking each part of data as a test set, taking a plurality of parts of data except the test set as a training set corresponding to each test set, thus forming a plurality of test sets, forming a plurality of training sets, carrying out a plurality of times of training on each base model to obtain an initial cod calculation model, predicting the prediction result of each test set by using the initial cod calculation model, and forming the test set prediction result of each initial cod calculation model by using the prediction result of each test set;
for example, each of 5 samples is sequentially and respectively used as a test set, and other 4 samples are sequentially used as a training set, as shown in fig. 2, the 5 th sample is used as a test set, the test set is represented by Pr in fig. 2, meanwhile, the 1 st to 4 th samples are used as a training set, le in fig. 2 represents one sample in the training set, a training set training base model is used, and a prediction result Q5 of the test set is predicted by using the trained base model; then, taking the 4 th sample data as a test set, taking the 1 st to 3 rd sample data and the 5 th sample data as a training set, training a base model by using the training set, predicting a prediction result Q4 of the test set by using the trained base model, repeating the step until each sample data is predicted as the test set, training each base model to obtain an initial cod calculation model, and obtaining a test set prediction result corresponding to the initial cod calculation model: q1, Q2 … Q5.
3. Step 2 is performed on a plurality of base models (which may be different models and may respectively adopt different regression models), so as to obtain a plurality of initial cod calculation models, as shown in fig. 2, taking training of 5 base models as an example, 5 initial cod calculation models are obtained through training, and test set prediction results obtained by the five initial cod calculation models through cross validation respectively, where the test set prediction results Q1, Q2 … Q5 of each initial cod calculation model form T, that is, the test set prediction results corresponding to the 5 initial cod calculation models respectively are: t1, T2, T3, T4, T5.
4. And taking T1-T5 as a training set and a testing set of the next layer, training the bp neural network, and forming a cod calculation model corresponding to the waveform category of the certain water sample by the trained bp neural network and the plurality of initial cod calculation models.
In this embodiment, a computer device is provided, as shown in fig. 3, comprising a memory 301, a processor 302, and a computer program stored on the memory and executable on the processor, the processor implementing any of the above methods for determining chemical oxygen demand of a water body when executing the computer program.
In particular, the computer device may be a computer terminal, a server or a similar computing device.
In this embodiment, a computer readable storage medium is provided, which stores a computer program for executing any of the above-described methods for determining chemical oxygen demand of a water body.
In particular, computer-readable storage media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer-readable storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable storage medium does not include transitory computer readable media (transmyedia), such as modulated data signals and carrier waves.
Based on the same inventive concept, the embodiment of the invention also provides a device for determining the chemical oxygen demand of the water body, as described in the following embodiment. Because the principle of solving the problems of the determination device for determining the chemical oxygen demand of the water body is similar to that of the determination method for the chemical oxygen demand of the water body, the implementation of the determination device for the chemical oxygen demand of the water body can refer to the implementation of the determination method for the chemical oxygen demand of the water body, and repeated parts are not repeated. As used hereinafter, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Fig. 4 is a block diagram of a configuration of an apparatus for determining chemical oxygen demand of a water body according to an embodiment of the present invention, as shown in fig. 4, the apparatus includes:
the sample data acquisition module 401 is used for acquiring water sample spectrum data of different sample water bodies through a standard station;
the clustering module 402 is used for clustering the water sample spectrum data of different sample water bodies to mark out different water sample waveform categories;
the classification model training module 403 is configured to train a BP neural network to obtain an identification model by using the water sample spectrum data of different sample water bodies and corresponding water sample waveform categories as samples;
the spectrum acquisition module 404 is configured to acquire to-be-detected spectrum data of a water body in an area to be monitored;
a waveform determining module 405, configured to determine a category of a waveform to be detected through an identification model according to the spectral data to be detected;
the model determining module 406 is configured to determine, according to a correspondence between a waveform type and a cod calculation model, a cod calculation model corresponding to the waveform type to be detected;
a calculating module 407, configured to obtain, according to the spectral data to be detected, a chemical oxygen demand of the water body in the region to be monitored through a cod calculation model corresponding to the type of the waveform to be detected;
the clustering module comprises:
the similarity calculation unit is used for calculating the similarity between the water sample spectral data of different sample water bodies;
the clustering unit is used for clustering the water sample spectral data of different sample water bodies based on the similarity between the water sample spectral data of different sample water bodies to mark out different water sample waveform categories;
the similarity calculation unit is used for calculating the correlation coefficient between the water sample spectrum data of every two different sample water bodies according to the following formula:
c=corr(x1,x2)
wherein c is a correlation coefficient, x1 and x2 represent two spectral curves corresponding to two different sample water bodies respectively, and corr () is a pearson correlation coefficient of the two spectral curves;
calculating the deviation between the water sample spectrum data of every two different sample water bodies according to the following formula:
d=x1-x2
wherein d is the deviation;
discretizing each deviation, dividing each discretized deviation into a plurality of intervals, and counting the proportion value of each interval through the following formula, wherein the proportion value is the proportion of the number of the deviations in the total number of the deviations, which is included in each interval, and the length of each interval is a preset length:
Pm=qm/sum(q)
wherein, pm represents the proportional value of the mth interval q, qm represents the number of deviations d contained in the mth interval q, and sum (q) represents the total number of deviations d;
calculating the information entropy of all the deviations according to the proportional values of the intervals by the following formula:
wherein, ent is information entropy, and M is the total number of intervals;
calculating the similarity between the water sample spectral data of every two different sample water bodies according to the information entropy and the correlation coefficient between the water sample spectral data of every two different sample water bodies by the following formula:
wherein w is the similarity between the water sample spectral data of every two different sample water bodies.
In one embodiment, the clustering unit is used for constructing a matrix according to the similarity between the water sample spectral data of every two different sample water bodies; constructing a degree matrix according to the matrix; subtracting the matrix from the degree matrix to obtain a Laplace matrix; standardizing the Laplace matrix to obtain a standardized Laplace matrix; calculating eigenvectors corresponding to the minimum k1 eigenvalues in the normalized Laplace matrix; and forming a characteristic matrix by using the characteristic vectors, taking each row in the characteristic matrix as a sample to obtain a plurality of samples, and clustering by using a clustering model according to the plurality of samples to obtain different water sample waveform categories.
In one embodiment, the above apparatus further comprises:
the cod calculation model training module is used for acquiring the chemical oxygen demand concentrations of different sample water bodies through a standard station; and training a machine learning component to obtain a cod calculation model corresponding to each water sample waveform category by using the water sample spectral data of each water sample waveform category and the chemical oxygen demand concentration of the water sample corresponding to the water sample spectral data of each water sample waveform category as sample data of each water sample waveform category.
In one embodiment, the cod computing model training module is specifically configured to divide sample data of each water sample waveform category into multiple pieces of data, sequentially use each piece of data as a test set, and use multiple pieces of data except the test set as a training set; training a plurality of base models by adopting each training set to obtain a plurality of initial cod calculation models, predicting the prediction result of each test set by using each initial cod calculation model, and forming the prediction result of each initial cod calculation model by using the prediction result of each test set; and training a bp neural network by taking the test set prediction result of each initial cod calculation model as a sample, wherein the trained bp neural network and the plurality of initial cod calculation models form cod calculation models corresponding to each water sample waveform category.
The embodiment of the invention realizes the following technical effects: determining the cod calculation model corresponding to the to-be-detected waveform type of the to-be-detected spectral data of the water body of the region to be monitored according to the corresponding relation between the waveform type and the cod calculation model, and finally inputting the to-be-detected spectral data into the determined cod calculation model to obtain the chemical oxygen demand of the water body of the region to be monitored. The COD calculation model corresponding to the type of the waveform to be detected of the spectral data to be detected can be directly adopted to calculate the COD of the water body aiming at the area to be monitored, and the type of the waveform to be detected of the spectral data to be detected is determined by the water body characteristic or characteristic of the area to be monitored, so that the COD calculation model which is adaptive to or matched with the water body of the area to be monitored can be adopted to calculate the COD of the water body of the area to be monitored, and the accuracy of the calculation result of the COD of the water body of the area to be monitored can be improved; meanwhile, the cod calculation model can be transmitted in different areas to be monitored, the process of performing cod calculation model training by collecting water samples, spectral data and other data again in the areas to be monitored is avoided, and the workload and the cost for calculating the chemical oxygen demand of the water body in the areas to be monitored are reduced.
It will be apparent to those skilled in the art that the modules or steps of the embodiments of the invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, embodiments of the invention are not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the embodiment of the present invention by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A method for determining the chemical oxygen demand of a water body is characterized by comprising the following steps:
collecting water sample spectrum data of different sample water bodies through a standard station;
clustering water sample spectrum data of different sample water bodies to mark out different water sample waveform categories;
taking water sample spectrum data of different sample water bodies and corresponding water sample waveform categories as samples, and training a BP neural network to obtain an identification model;
acquiring spectral data to be detected of a water body in an area to be monitored;
determining the type of the waveform to be detected through the identification model according to the spectral data to be detected;
determining a cod calculation model corresponding to the waveform type to be detected according to the corresponding relation between the waveform type and the cod calculation model;
according to the spectral data to be detected, obtaining the chemical oxygen demand of the water body of the area to be monitored through a cod calculation model corresponding to the waveform type to be detected;
the water sample spectral data of different sample water bodies are clustered, and different water sample waveform categories are divided, and the method comprises the following steps:
calculating the similarity between water sample spectral data of different sample water bodies;
based on the similarity between the water sample spectrum data of different sample water bodies, clustering the water sample spectrum data of different sample water bodies, and dividing different water sample waveform categories;
the similarity between the water sample spectral data of different sample water bodies is calculated, and the method comprises the following steps:
calculating the correlation coefficient between the water sample spectral data of every two different sample water bodies by the following formula:
c=corr(x1,x2)
wherein c is a correlation coefficient, x1 and x2 represent two spectral curves corresponding to two different sample water bodies respectively, and corr () is a pearson correlation coefficient of the two spectral curves;
calculating the deviation between the water sample spectrum data of every two different sample water bodies according to the following formula:
d=x1-x2
wherein d is the deviation;
discretizing each deviation, dividing each discretized deviation into a plurality of intervals, and counting the proportion value of each interval through the following formula, wherein the proportion value is the proportion of the number of the deviations in the total number of the deviations, which is included in each interval, and the length of each interval is a preset length:
Pm=qm/sum(q)
wherein, pm represents the proportional value of the mth interval q, qm represents the number of deviations d contained in the mth interval q, and sum (q) represents the total number of deviations d;
calculating the information entropy of all the deviations according to the proportional value of each interval by the following formula:
wherein, ent is information entropy, and M is the total number of intervals;
calculating the similarity between the water sample spectral data of every two different sample water bodies according to the information entropy and the correlation coefficient between the water sample spectral data of every two different sample water bodies by the following formula:
wherein w is the similarity between the water sample spectral data of every two different sample water bodies.
2. The method for determining chemical oxygen demand of water body according to claim 1, wherein the clustering the water sample spectral data of different sample water bodies based on the similarity between the water sample spectral data of different sample water bodies to classify different water sample waveform types comprises:
constructing a matrix according to the similarity between the water sample spectral data of each two different sample water bodies;
constructing a degree matrix according to the matrix;
subtracting the matrix from the degree matrix to obtain a Laplace matrix;
standardizing the Laplace matrix to obtain a standardized Laplace matrix;
calculating eigenvectors corresponding to the minimum k1 eigenvalues in the normalized Laplace matrix;
and forming a characteristic matrix by using the characteristic vectors, taking each row in the characteristic matrix as a sample to obtain a plurality of samples, and clustering by using a clustering model according to the plurality of samples to obtain different water sample waveform categories.
3. The method of determining chemical oxygen demand of a body of water of claim 1, further comprising:
collecting the chemical oxygen demand concentrations of different sample water bodies through a standard station;
and training a machine learning component to obtain a cod calculation model corresponding to each water sample waveform category by using the water sample spectral data of each water sample waveform category and the chemical oxygen demand concentration of the water sample corresponding to the water sample spectral data of each water sample waveform category as sample data of each water sample waveform category.
4. The method for determining chemical oxygen demand of water body according to claim 3, wherein the step of training the machine learning component to obtain the cod calculation model corresponding to each water sample waveform category by using the water sample spectrum data of each water sample waveform category and the water sample spectrum data of each water sample waveform category corresponding to the chemical oxygen demand concentration of the water sample as the sample data of each water sample waveform category comprises the following steps:
dividing sample data of each water sample waveform category into a plurality of data, taking each data as a test set in sequence, and taking a plurality of data except the test set as a training set;
training a plurality of base models by adopting each training set to obtain a plurality of initial cod calculation models, predicting the prediction result of each test set by using each initial cod calculation model, and forming the prediction result of each initial cod calculation model by using the prediction result of each test set;
and training a bp neural network by taking the test set prediction result of each initial cod calculation model as a sample, wherein the trained bp neural network and the plurality of initial cod calculation models form cod calculation models corresponding to each water sample waveform category.
5. A device for determining chemical oxygen demand of a water body, comprising:
the sample data acquisition module is used for acquiring water sample spectrum data of different sample water bodies through the standard station;
the clustering module is used for clustering the water sample spectrum data of different sample water bodies to mark out different water sample waveform categories;
the classification model training module is used for training a BP neural network to obtain a recognition model by taking water sample spectrum data of different sample water bodies and corresponding water sample waveform categories as samples;
the spectrum acquisition module is used for acquiring to-be-detected spectrum data of a water body in a to-be-monitored area;
the waveform determining module is used for determining the category of the waveform to be detected through the identification model according to the spectral data to be detected;
the model determining module is used for determining a cod calculation model corresponding to the waveform type to be detected according to the corresponding relation between the waveform type and the cod calculation model;
the calculation module is used for obtaining the chemical oxygen demand of the water body of the area to be monitored through a cod calculation model corresponding to the type of the waveform to be detected according to the spectral data to be detected;
the clustering module comprises:
the similarity calculation unit is used for calculating the similarity between the water sample spectral data of different sample water bodies;
the clustering unit is used for clustering the water sample spectral data of different sample water bodies based on the similarity between the water sample spectral data of different sample water bodies to mark out different water sample waveform categories;
the similarity calculation unit is used for calculating the correlation coefficient between the water sample spectral data of every two different sample water bodies according to the following formula:
c=corr(x1,x2)
wherein c is a correlation coefficient, x1 and x2 represent two spectral curves corresponding to two different sample water bodies respectively, and corr () is a pearson correlation coefficient of the two spectral curves;
calculating the deviation between the water sample spectrum data of every two different sample water bodies according to the following formula:
d=x1-x2
wherein d is the deviation;
discretizing each deviation, dividing each discretized deviation into a plurality of intervals, and counting the proportion value of each interval through the following formula, wherein the proportion value is the proportion of the number of the deviations in the total number of the deviations, which is included in each interval, and the length of each interval is a preset length:
Pm=qm/sum(q)
wherein, pm represents the proportional value of the mth interval q, qm represents the number of deviations d contained in the mth interval q, and sum (q) represents the total number of deviations d;
calculating the information entropy of all the deviations according to the proportional value of each interval by the following formula:
wherein, ent is information entropy, and M is the total number of intervals;
calculating the similarity between the water sample spectral data of every two different sample water bodies according to the information entropy and the correlation coefficient between the water sample spectral data of every two different sample water bodies by the following formula:
wherein w is the similarity between the water sample spectral data of every two different sample water bodies.
6. A computer apparatus comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements the method of determining chemical oxygen demand of a water body of any one of claims 1 to 4.
7. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method for determining chemical oxygen demand of a water body according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310107553.6A CN115824993B (en) | 2023-02-14 | 2023-02-14 | Method and device for determining water body chemical oxygen demand, computer equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310107553.6A CN115824993B (en) | 2023-02-14 | 2023-02-14 | Method and device for determining water body chemical oxygen demand, computer equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115824993A true CN115824993A (en) | 2023-03-21 |
| CN115824993B CN115824993B (en) | 2023-07-18 |
Family
ID=85521151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310107553.6A Active CN115824993B (en) | 2023-02-14 | 2023-02-14 | Method and device for determining water body chemical oxygen demand, computer equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115824993B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117517230A (en) * | 2024-01-03 | 2024-02-06 | 北京英视睿达科技股份有限公司 | Water quality monitoring method, device, equipment and medium based on ultraviolet-visible spectrum |
| CN117686447A (en) * | 2024-01-31 | 2024-03-12 | 北京英视睿达科技股份有限公司 | Water quality monitoring method, device, equipment and medium based on multichannel model |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106990060A (en) * | 2017-03-24 | 2017-07-28 | 四川碧朗科技有限公司 | Water quality index monitor, cloud data center and system, Forecasting Methodology and water sample recognition methods |
| CN107665288A (en) * | 2016-07-30 | 2018-02-06 | 复凌科技（上海）有限公司 | A kind of water quality hard measurement Forecasting Methodology of COD |
| CN109214421A (en) * | 2018-07-27 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of model training method, device and computer equipment |
| CN113077019A (en) * | 2021-06-07 | 2021-07-06 | 芯视界（北京）科技有限公司 | Pollution type identification method and device and storage medium |
| CN114368795A (en) * | 2021-12-31 | 2022-04-19 | 天健创新（北京）监测仪表股份有限公司 | Online black and odorous water body multi-mode identification method and system |
| CN114548282A (en) * | 2022-02-23 | 2022-05-27 | 杭州师范大学 | Sorting method and device based on fluorescence spectrum and readable storage medium |
| WO2022148252A1 (en) * | 2021-01-11 | 2022-07-14 | 江西师范大学 | Method and apparatus for measuring concentration of dissolved organic carbon in water |
-
2023
- 2023-02-14 CN CN202310107553.6A patent/CN115824993B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107665288A (en) * | 2016-07-30 | 2018-02-06 | 复凌科技（上海）有限公司 | A kind of water quality hard measurement Forecasting Methodology of COD |
| CN106990060A (en) * | 2017-03-24 | 2017-07-28 | 四川碧朗科技有限公司 | Water quality index monitor, cloud data center and system, Forecasting Methodology and water sample recognition methods |
| CN109214421A (en) * | 2018-07-27 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of model training method, device and computer equipment |
| WO2022148252A1 (en) * | 2021-01-11 | 2022-07-14 | 江西师范大学 | Method and apparatus for measuring concentration of dissolved organic carbon in water |
| CN113077019A (en) * | 2021-06-07 | 2021-07-06 | 芯视界（北京）科技有限公司 | Pollution type identification method and device and storage medium |
| CN114368795A (en) * | 2021-12-31 | 2022-04-19 | 天健创新（北京）监测仪表股份有限公司 | Online black and odorous water body multi-mode identification method and system |
| CN114548282A (en) * | 2022-02-23 | 2022-05-27 | 杭州师范大学 | Sorting method and device based on fluorescence spectrum and readable storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 吕蒙 等: "基于水样类型识别的光谱COD测量方法" * |
| 闻兵工 等: "基于光谱曲线整体相似性测度的匹配分类" * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117517230A (en) * | 2024-01-03 | 2024-02-06 | 北京英视睿达科技股份有限公司 | Water quality monitoring method, device, equipment and medium based on ultraviolet-visible spectrum |
| CN117686447A (en) * | 2024-01-31 | 2024-03-12 | 北京英视睿达科技股份有限公司 | Water quality monitoring method, device, equipment and medium based on multichannel model |
| CN117686447B (en) * | 2024-01-31 | 2024-05-03 | 北京英视睿达科技股份有限公司 | Water quality monitoring method, device, equipment and medium based on multichannel model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115824993B (en) | 2023-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115824993B (en) | Method and device for determining water body chemical oxygen demand, computer equipment and medium | |
| CN110634080B (en) | Abnormal electricity utilization detection method, device, equipment and computer readable storage medium | |
| CN109444813B (en) | RFID indoor positioning method based on BP and DNN double neural networks | |
| CN112508105B (en) | Fault detection and retrieval method for oil extraction machine | |
| CN108241901B (en) | Transformer early warning evaluation method and device based on prediction data | |
| CN107301617B (en) | Method and equipment for evaluating quality of waste gas monitoring data | |
| CN107203467A (en) | The reference test method and device of supervised learning algorithm under a kind of distributed environment | |
| CN112348290A (en) | River water quality prediction method, device, storage medium and equipment | |
| CN117196353A (en) | Environmental pollution assessment and monitoring method and system based on big data | |
| CN115047064A (en) | Pipeline defect quantification method, processor and pipeline defect quantification device | |
| CN102081045A (en) | Structural damage identification method based on laser television holographic technique | |
| CN103196356A (en) | Oil tube defect quantitative recognition method based on support vector machine | |
| Awawdeh et al. | Application of outlier detection using re-weighted least squares and R-squared for IoT extracted data | |
| CN117040983B (en) | Data sharing method and system based on big data analysis | |
| CN116466058B (en) | Water quality detection data processing method, water quality evaluation system, equipment and medium | |
| CN110046651A (en) | A kind of pipeline conditions recognition methods based on monitoring data multi-attribute feature fusion | |
| CN112906672A (en) | Steel rail defect identification method and system | |
| CN111786999A (en) | Intrusion behavior detection method, device, equipment and storage medium | |
| CN114091160B (en) | Intelligent detection method and system for bridge structure damage | |
| CN113740295B (en) | Method and device for detecting and analyzing finished oil additive based on near infrared modeling | |
| CN111062118B (en) | Multilayer soft measurement modeling system and method based on neural network prediction layering | |
| CN107506824B (en) | Method and device for detecting bad observation data of power distribution network | |
| CN113128158A (en) | Coupling coefficient identification method of wireless power transmission system | |
| Wang et al. | Cross-vendor and cross-state analysis of gps probe data latency | |
| CN113139673A (en) | Method, device, terminal and storage medium for predicting air quality |
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 |