CN117368141A - Perchlorate wastewater concentration intelligent detection method based on artificial intelligence - Google Patents
Perchlorate wastewater concentration intelligent detection method based on artificial intelligence Download PDFInfo
- Publication number
- CN117368141A CN117368141A CN202311668000.4A CN202311668000A CN117368141A CN 117368141 A CN117368141 A CN 117368141A CN 202311668000 A CN202311668000 A CN 202311668000A CN 117368141 A CN117368141 A CN 117368141A
- Authority
- CN
- China
- Prior art keywords
- curve
- fluctuation
- target
- target curve
- value
- 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
- VLTRZXGMWDSKGL-UHFFFAOYSA-M perchlorate Inorganic materials [O-]Cl(=O)(=O)=O VLTRZXGMWDSKGL-UHFFFAOYSA-M 0.000 title claims abstract description 53
- VLTRZXGMWDSKGL-UHFFFAOYSA-N perchloric acid Chemical compound OCl(=O)(=O)=O VLTRZXGMWDSKGL-UHFFFAOYSA-N 0.000 title claims abstract description 53
- 239000002351 wastewater Substances 0.000 title claims abstract description 40
- 238000001514 detection method Methods 0.000 title claims abstract description 22
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 21
- 238000001228 spectrum Methods 0.000 claims abstract description 74
- 238000002835 absorbance Methods 0.000 claims abstract description 45
- 238000000034 method Methods 0.000 claims abstract description 40
- 238000001914 filtration Methods 0.000 claims abstract description 29
- 238000010183 spectrum analysis Methods 0.000 claims abstract description 7
- 230000003595 spectral effect Effects 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000010606 normalization Methods 0.000 claims description 12
- 238000000354 decomposition reaction Methods 0.000 claims description 9
- 230000000694 effects Effects 0.000 abstract description 6
- 238000004458 analytical method Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004611 spectroscopical analysis Methods 0.000 description 3
- 230000000739 chaotic effect Effects 0.000 description 2
- 238000002329 infrared spectrum Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 238000001237 Raman spectrum Methods 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000002371 ultraviolet--visible spectrum Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 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
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3577—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing liquids, e.g. polluted water
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/148—Wavelet transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
- G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
Abstract
The invention relates to the technical field of physical analysis, in particular to an intelligent perchlorate wastewater concentration detection method based on artificial intelligence, which comprises the following steps: collecting a spectrum curve of perchlorate wastewater, decomposing the spectrum curve into a plurality of IMF component curves, obtaining a fluctuation coefficient of each component curve according to data differences among the curves, respectively equally dividing all the curves into a plurality of non-repeated curve segments, obtaining a total fluctuation duty ratio difference of each component curve according to the absorbance differences among the curve segments, thereby obtaining an optimal wavelet threshold value, obtaining a denoising curve by using a wavelet threshold denoising algorithm, obtaining a filtering denoising curve of the spectrum curve by using an EMD reconstruction algorithm, and obtaining the perchlorate concentration by using a spectrum analysis method. According to the invention, the filtering denoising effect is improved by self-adapting to the optimal wavelet threshold value of each component curve, so that the credibility of data on a spectrum curve is improved, and the accuracy of perchlorate concentration detection is improved.
Description
Technical Field
The invention relates to the technical field of physical analysis, in particular to an intelligent perchlorate wastewater concentration detection method based on artificial intelligence.
Background
Perchlorate is a special energetic chemical substance and is widely used in the fields of the munition industry, matches and firework manufacturing, etc. However, these related industrial processes can discharge related perchlorate-containing wastewater into the environment, which causes environmental pollution of water and soil, and affects the entire ecological environment, so that the concentration of the perchlorate wastewater needs to be detected before the perchlorate wastewater is discharged. The conventional approach is spectroscopic analysis, which determines the concentration by manually taking multiple samples of perchlorate in a wastewater environment and matching the same with known perchlorate concentration spectral data curves. However, when the spectrum sensor acquires spectrum data, noise interference factors exist in the data due to the limitation of the sensor, and a wavelet threshold denoising method is generally used for filtering and smoothing the spectrum curve to denoise.
The existing problems are as follows: because the noise caused by the equipment is a high-frequency fluctuation characteristic, the characteristic can lead the periodic composition of the whole original spectrum signal to be more chaotic after being overlapped and mixed with part of fluctuation in the original spectrum curve, and the influence after being overlapped in fluctuation with different fluctuation degrees is different, so that the fixed wavelet threshold value can show the problem of overlarge or overlarge small in each local area in the whole spectrum, further lead the spectrum curve finally obtained by the equipment to be inaccurate, and reduce the accuracy of concentration detection.
Disclosure of Invention
The invention provides an intelligent perchlorate wastewater concentration detection method based on artificial intelligence, which aims to solve the existing problems.
The invention discloses an artificial intelligence-based perchlorate wastewater concentration intelligent detection method, which adopts the following technical scheme:
the embodiment of the invention provides an intelligent perchlorate wastewater concentration detection method based on artificial intelligence, which comprises the following steps of:
collecting a spectrum curve of perchlorate wastewater, and decomposing the spectrum curve into a plurality of IMF component curves and a residual curve by using an EMD decomposition algorithm; the abscissa and ordinate of the data point on the curve are wave number and absorbance respectively;
recording any IMF component curve as a target curve; according to the data difference between the target curve and the spectrum curve, obtaining the wave number length, the fluctuation value and the original fluctuation value of each local extreme point on the target curve; obtaining the fluctuation coefficient of the target curve according to the wave number length, the fluctuation value and the original fluctuation value of all local extremum points on the target curve;
dividing the spectrum curve and all IMF component curves into a plurality of non-repeated curve segments; obtaining the fluctuation duty ratio difference of each curve segment of the target curve equal division according to the difference of the absorbance between the curve segments; obtaining the total fluctuation duty ratio difference of the target curve according to the fluctuation duty ratio difference of all curve segments equally divided by the target curve;
obtaining an optimal wavelet threshold corresponding to the target curve according to the fluctuation coefficient and the total fluctuation duty ratio difference of the target curve;
according to an optimal wavelet threshold corresponding to the target curve, a wavelet threshold denoising algorithm is used to obtain a denoising curve of the target curve; according to the residual curve and the denoising curves of all IMF component curves, using an EMD reconstruction algorithm to obtain a filtering denoising curve of a spectrum curve; and obtaining the perchlorate concentration in the perchlorate wastewater by using a spectral analysis method according to a filtering denoising curve of the spectral curve.
Further, according to the data difference between the target curve and the spectrum curve, the wave number length, the fluctuation value and the original fluctuation value of each local extremum point on the target curve are obtained, and the method comprises the following specific steps:
a first derivative method is used for respectively obtaining a spectrum curve and local extreme points on all IMF component curves;
obtaining the wave number range and the fluctuation value of each local extremum point according to the wave numbers and the absorbance of all the local extremum points on the target curve;
subtracting the minimum value from the maximum value in the wave number range of each local extreme point on the target curve, and recording the wave number length of each local extreme point;
and counting the absorbance of the local extremum points in the wave number range on the spectrum curve according to the wave number range of each local extremum point on the target curve, and recording the standard deviation of the absorbance of the local extremum points as the original fluctuation value of each local extremum point on the target curve.
Further, the obtaining the wave number range and the fluctuation value of each local extremum point according to the wave numbers and the absorbance of all the local extremum points on the target curve comprises the following specific steps:
counting wave numbers of two adjacent local extremum points of each local extremum point in all local extremum points on the target curve, and recording a range between the wave numbers of the two adjacent local extremum points as a wave number range of each local extremum point;
and calculating the difference value of the absorbance between each local extremum point and all adjacent local extremum points, and recording the average value of the absolute values of the difference values as the fluctuation value of each local extremum point.
Further, the specific calculation formula corresponding to the fluctuation coefficient of the target curve is obtained according to the wave number lengths, the fluctuation values and the original fluctuation values of all the local extreme points on the target curve, wherein the specific calculation formula comprises the following steps:
where a is the relief coefficient of the target curve,for the fluctuation value of the ith local extreme point on the target curve, < >>Is the mean value of the fluctuation values of all local extreme points on the target curve, +.>For the wavenumber length of the i-th local extreme point on the target curve, < >>Is the mean value of the wave number lengths of all local extreme points on the target curve, +.>For the original fluctuation value of the ith local extreme point on the target curve, N is the number of the local extreme points on the target curve, +.>Is a linear normalization function.
Further, the method for dividing the spectrum curve and all IMF component curves into a plurality of non-repeated curve segments comprises the following specific steps:
calculating the average value of wave number lengths of all local extreme points on a target curve, and recording the downward rounding value of the average value as the period of the target curve;
the least common multiple of the periods of all IMF component curves is recorded as a standard period;
on the horizontal axis, the spectrum curve and all IMF component curves are divided into a plurality of non-repeated curve segments by equally dividing the standard period into equal lengths.
Further, the method for obtaining the fluctuation duty ratio difference of each curve segment of the target curve equal division according to the difference of the absorbance between the curve segments comprises the following specific steps:
in the spectrum curve and all IMF component curves, the spectrum curve or any IMF component curve is recorded as a reference curve;
sequentially counting all curve segments on a reference curve according to the direction from small wave numbers to large wave numbers to obtain a curve segment sequence of the reference curve;
in a curve segment sequence of the spectrum curve, the j-th curve segment is marked as a standard curve segment;
in the curve segment sequence of the target curve, the j-th curve segment is marked as the target curve segment;
and obtaining the fluctuation duty ratio difference of the target curve segment according to the absorbance of all the data points on the standard curve segment and the target curve segment.
Further, the specific calculation formula corresponding to the fluctuation duty ratio difference of the target curve segment is obtained according to the absorbance of all the data points on the standard curve segment and the target curve segment, wherein the specific calculation formula is as follows:
wherein B is the fluctuation duty ratio difference of the target curve segment, C is the standard deviation of the absorbance of all data points on the standard curve segment,standard deviation of absorbance for all data points on the target curve segment, +.>Maximum value of absorbance for all data points on standard curve segment, +.>Is the minimum value in absorbance of all data points on the standard curve segment, +.>K is a preset exponential function adjustment value for an exponential function based on a natural constant, ++>Is a linear normalization function, and is an absolute value function.
Further, the specific calculation formula corresponding to the total fluctuation duty ratio difference of the target curve is obtained according to the fluctuation duty ratio difference of all curve segments equally divided by the target curve, wherein the specific calculation formula is as follows:
wherein the method comprises the steps ofFor the total fluctuation duty cycle difference of the target curve, +.>Fluctuation duty difference of the xth curve segment equally dividing the target curve, +.>Standard deviation of absorbance of all data points on the xth curve segment which is equal to the target curve, y is the number of curve segments which are equal to the target curve, +.>Is a linear normalization function.
Further, the method for obtaining the optimal wavelet threshold corresponding to the target curve according to the fluctuation coefficient and the total fluctuation duty ratio difference of the target curve comprises the following specific steps:
calculating a difference value of the total fluctuation duty ratio difference of a subtracted target curve, and recording a normalized value of the product of the difference value and the fluctuation coefficient of the target curve as a filtering adjustment coefficient of the target curve;
using a Vthreshold algorithm to obtain a wavelet threshold corresponding to the target curve;
and obtaining an optimal wavelet threshold corresponding to the target curve according to the wavelet threshold corresponding to the target curve and the filtering adjustment coefficient.
Further, the method for obtaining the optimal wavelet threshold corresponding to the target curve according to the wavelet threshold corresponding to the target curve and the filter adjustment coefficient comprises the following specific steps:
calculating the product of the filter adjustment coefficient of the target curve and q, then calculating the sum of the product of the filter adjustment coefficient of the target curve and q plus q, and recording the product of the sum and the wavelet threshold corresponding to the target curve as the optimal wavelet threshold corresponding to the target curve; and q is a preset constant.
The technical scheme of the invention has the beneficial effects that:
in the embodiment of the invention, a spectrum curve of perchlorate wastewater is collected, and the spectrum curve is decomposed into a plurality of IMF component curves and a residual curve. And obtaining the fluctuation coefficient of each IMF component curve according to the data difference between the curves. Dividing the spectrum curve and all IMF component curves into a plurality of non-repeated curve segments, and obtaining the total fluctuation duty ratio difference of each IMF component curve according to the absorbance difference between the curve segments. The optimal wavelet threshold corresponding to each IMF component curve is obtained, the noise-containing degree in each IMF component curve is analyzed, correction and adjustment are carried out by combining the differences between the component curves and the spectrum curve, the robustness is improved, the obtained wavelet threshold is ensured to be optimal, and the denoising effect is improved. And (3) using a wavelet threshold denoising algorithm to obtain a denoising curve of each IMF component curve, then using an EMD reconstruction algorithm to obtain a filtering denoising curve of the spectrum curve, thereby reducing the information loss degree after the component curve is denoised and inversely transformed into the spectrum curve, and finally using a spectrum analysis method to obtain the perchlorate concentration in the perchlorate wastewater. The invention improves the filtering denoising effect by self-adapting to the optimal wavelet threshold value of each IMF component curve, thereby improving the credibility of data on a spectrum curve and improving the accuracy of perchlorate concentration detection.
Drawings
In order to more clearly illustrate the embodiments of the invention 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, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of steps of an intelligent perchlorate wastewater concentration detection method based on artificial intelligence.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of the specific implementation, structure, characteristics and effects of the perchlorate wastewater concentration intelligent detection method based on artificial intelligence according to the invention by combining the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The specific scheme of the perchlorate wastewater concentration intelligent detection method based on artificial intelligence provided by the invention is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of an intelligent detection method for perchlorate wastewater concentration based on artificial intelligence according to an embodiment of the invention is shown, the method comprises the following steps:
step S001: collecting a spectrum curve of perchlorate wastewater, and decomposing the spectrum curve into a plurality of IMF component curves and a residual curve by using an EMD decomposition algorithm; the abscissa and ordinate of the data points on the curve are wavenumber and absorbance, respectively.
And collecting a spectrum curve of the perchlorate wastewater by using a spectrum sensor. Wherein the horizontal axis of the spectrum curve is wave number and the vertical axis is absorbance.
What needs to be described is: the sample is collected from the wastewater and needs to be subjected to a part of necessary pretreatment such as filtration dilution and the like, and then a proper spectrum technology such as ultraviolet visible spectrum, infrared spectrum, raman spectrum and the like is selected, wherein the infrared spectrum is used in the embodiment, and then the perchlorate concentration spectrum data curve in the wastewater is finally obtained through the technology and the subsequent part of pretreatment such as spectrum measurement and data treatment.
In perchlorate spectral data curves, noise due to interference with the device tends to appear as high frequency and is prevalent, i.e., noise can be seen in the spectral data curve as a constant fluctuation in all parts. In the spectrum curve acquired by this embodiment, different fluctuations are decomposed into different components, so after EMD decomposition, equipment noise mainly exists in a certain component, and different components are scattered and dispersed in components corresponding to different periods due to different fluctuation degrees in the original data curve, so for this case, it is necessary to analyze and quantify the noise-prone degree shown by each component after EMD decomposition, and correspondingly apply different filtering degrees to each component based on the feature, thereby achieving adaptive filtering.
And decomposing the spectrum curve by using an EMD decomposition algorithm to obtain a plurality of IMF component curves and a residual curve.
What needs to be described is: the EMD decomposition algorithm is a well-known technique, specific methods are not described herein, EMD refers to empirical mode decomposition, IMF refers to an intrinsic mode function, and the amount of data on each IMF component curve is the same as the amount of data on the spectral curve.
Step S002: recording any IMF component curve as a target curve; according to the data difference between the target curve and the spectrum curve, obtaining the wave number length, the fluctuation value and the original fluctuation value of each local extreme point on the target curve; and obtaining the fluctuation coefficient of the target curve according to the wave number length, the fluctuation value and the original fluctuation value of all local extreme points on the target curve.
It is then necessary to analyze the curve relief characteristics on each IMF component curve and to analyze the relief smoothness in the component as its noise level. The distribution fluctuation coefficient is obtained based on the distribution of extreme points in each component.
And (3) respectively obtaining a spectrum curve and local extreme points on all IMF component curves by using a first derivative method. The first derivative method is a known technique, and the specific method is not described here.
Any one IMF component curve is recorded as a target curve.
And counting wave numbers of two adjacent local extremum points of each local extremum point in all local extremum points on the target curve, and recording the range between the wave numbers as the wave number range of each local extremum point.
What needs to be described is: since the first and last local extremum points on the target curve have only one adjacent local extremum point, the present embodiment makes the wavenumber ranges of the first and last local extremum points on the target curve be the wavenumber ranges of the second and penultimate local extremum points, respectively. And the spectral curve typically contains a superposition of multiple frequency components, there are multiple local extrema on the spectral curve and IMF component curves.
On the target curve, the maximum value minus the minimum value in the wave number range of each local extremum point is recorded as the wave number length of each local extremum point.
And calculating the difference value of the absorbance of each local extremum point and all adjacent local extremum points of the local extremum points in all local extremum points on the target curve, and recording the average value of the absolute values of the difference values as the fluctuation value of each local extremum point.
What needs to be described is: only one difference value corresponding to the first local extreme point and the last local extreme point on the target curve is needed, so that the average value is not needed to be taken.
And according to the wave number range of each local extreme point on the target curve, counting the absorbance of the local extreme point in the wave number range on the spectrum curve, and recording the standard deviation of the absorbance as the original fluctuation value of each local extreme point on the target curve.
What needs to be described is: the standard deviation represents the intensity of data change, the value range of the standard deviation is 0 to positive infinity, and when the number of local extreme points in the wave number range is counted to be smaller than 2 on the spectrum curve, the fact that no large data fluctuation exists in the wave number range on the spectrum curve is indicated, so that the corresponding original fluctuation value is the minimum standard deviation 0.
The calculation formula of the fluctuation coefficient A of the target curve is known as follows:
where a is the relief coefficient of the target curve,for the fluctuation value of the ith local extreme point on the target curve, < >>Is the mean value of the fluctuation values of all local extreme points on the target curve, +.>For the wavenumber length of the i-th local extreme point on the target curve, < >>Is the mean value of the wave number lengths of all local extreme points on the target curve, +.>Is the original fluctuation value of the ith local extreme point on the target curve. N is the number of local extremum points on the target curve. />Normalizing the data values to [0,1 ] as a linear normalization function]Within the interval.
What needs to be described is:representation ofThe difference between the fluctuation value and the distribution interval between the adjacent local extremum points on the target curve is quantized to obtain the noise characteristic of the component, the higher the characteristic is, the more orderly fluctuation characteristic of the equipment noise is obvious, the higher the corresponding degree of conforming to the fluctuation characteristic of the noise is, and the higher the filtering degree is adopted in the follow-up denoising of the component. The greater the degree of fluctuation in the spectral curve, the more chaotic it is with the periodic characteristics after noise superposition, and the higher the corresponding degree of characteristic loss of the device noise itself decomposed into each period after EMD decomposition is likely to be, so for this part of extreme points, the embodiment considers that it participates in the calculation of the distribution fluctuation coefficient of this component, the weight value should be relatively weaker, and therefore the difference is->Is the weight. To this end use->And represents the fluctuation coefficient of the target curve.
Step S003: dividing the spectrum curve and all IMF component curves into a plurality of non-repeated curve segments; obtaining the fluctuation duty ratio difference of each curve segment of the target curve equal division according to the difference of the absorbance between the curve segments; and obtaining the total fluctuation duty ratio difference of the target curve according to the fluctuation duty ratio difference of all curve segments equally divided by the target curve.
For fluctuation analysis of noise, only through the stability of extreme points in components, the stability of the extreme points cannot be completely used as a filtered adjustment value, and the duty ratio of the noise degree in a spectrum curve is considered, so that corresponding adjustment is performed, and the situation that the data information in the components with higher duty ratios of the noise features is damaged more seriously due to the fact that the noise features in some components occupy higher amounts in the spectrum curve but have the same filtering degree as other components is avoided.
And calculating the average value of the wave number lengths of all the local extreme points on the target curve, rounding down the average value, and recording the average value as the period of the target curve.
In the above manner, the period of each IMF component curve is obtained.
The least common multiple of the periods of all IMF component curves is recorded as the standard period.
What needs to be described is: the least common multiple is the smallest positive integer that can be divided simultaneously among two or more integers, which is known in the art.
On the horizontal axis, the spectrum curve and all IMF component curves are divided into a plurality of non-repeated curve segments by equally dividing the standard period into equal lengths.
What needs to be described is: if the length of the transverse axis corresponding to the last curve segment of the curve division is less than the standard period, the embodiment also considers that the curve segment is a curve segment.
Any one IMF component curve or spectrum curve is recorded as a reference curve in the spectrum curve and all IMF component curves.
And sequentially counting all curve segments on the reference curve according to the direction from small wave numbers to large wave numbers to obtain a curve segment sequence of the reference curve.
In the above manner, a sequence of curve segments of the spectral curve, a sequence of curve segments of each IMF component curve are obtained.
In the curve segment sequence of the spectrum curve, the j-th curve segment is marked as a standard curve segment.
In the curve segment sequence of the target curve, the j-th curve segment is marked as the target curve segment.
The calculation formula of the fluctuation duty ratio difference B of the target curve segment is known as follows:
wherein B is the fluctuation duty ratio difference of the target curve segment, C is the standard deviation of the absorbance of all data points on the standard curve segment,standard deviation of absorbance for all data points on the target curve segment, +.>Maximum value of absorbance for all data points on standard curve segment, +.>Is the minimum of the absorbance of all data points on the standard curve segment. />The present embodiment uses +.>The inverse proportion relation and normalization processing are presented, an implementer can set an inverse proportion function and a normalization function according to actual conditions, k is a preset exponential function adjusting value, and k is used for delaying the attenuation speed of the exponential function. />Normalizing the data values to [0,1 ] as a linear normalization function]Within the interval, || is an absolute value function. In this embodiment, k is set to 0.01, which is described as an example, and other values may be set in other embodiments, and this embodiment is not limited thereto.
What needs to be described is:the larger the difference of standard deviation between the target curve corresponding to the jth curve segment and the data on the spectral curve is, the more relative components are represented, the component is noise or normal fluctuation, the duty ratio of fluctuation characteristics of the component in the spectral curve is relatively higher, and the degree of constraint on the filtering degree is higher when the part of information of the component is filtered so as to prevent the information from losing too much. However, for different curve segments, when there are extreme points in the spectrum curve where the difference is very large, the noise is relatively weaker in the part, so that for all components, the weight participation of the curve segment in the overall fluctuation duty ratio difference needs to be weakened. Thus->Is->Is, i.e. +.>The larger the adjustment weight, the smaller the adjustment weight, thereby representing the fluctuation duty ratio difference of the target curve segment by the product of the adjustment weight and the adjustment weight.
And obtaining the fluctuation duty ratio difference of each curve segment in the curve segment sequence of the target curve according to the mode.
However, this embodiment focuses on the noise level in each component, so that further, when the obtained fluctuation duty ratio difference is applied, the difference cannot be directly accumulated, but the fluctuation duty ratio difference of the whole component is obtained by taking the level of the fluctuation in each curve segment as the salient level of the component in the component.
From this, the difference in the total fluctuation ratio of the target curveThe calculation formula of (2) is as follows:
wherein the method comprises the steps ofFor the total fluctuation duty cycle difference of the target curve, +.>Fluctuation duty difference of the xth curve segment equally dividing the target curve, +.>The standard deviation of the absorbance of all data points on the xth curve segment which is the equal division of the target curve, and y is the number of curve segments which are the equal division of the target curve. />Normalizing the data values to [0,1 ] as a linear normalization function]Interval ofAnd (3) inner part.
What needs to be described is: this example considers that while the fluctuation duty cycle difference is relatively higher, it should also be based on the specific magnitude of its standard deviation, i.e., the degree of prominence, in its own component, each curve segment being of its own absorbance of all data points, the higher its standard deviation being indicative of that portion being more sensitive in its overall component to the constraints derived from the fluctuation duty cycle difference, and therefore it needs to take up a higher weight, and therefore also usesRepresenting the total fluctuation duty cycle difference of the target curve.
Step S004: and obtaining an optimal wavelet threshold corresponding to the target curve according to the fluctuation coefficient and the total fluctuation duty ratio difference of the target curve.
The calculation formula of the filter adjustment coefficient P of the target curve for adjusting the self filter size is thus:
wherein P is the filter adjustment coefficient of the target curve, A is the fluctuation coefficient of the target curve,is the total fluctuation duty cycle difference of the target curve. />Normalizing the data values to [0,1 ] as a linear normalization function]Within the interval.
What needs to be described is: the greater a indicates a relatively higher noise level in the target curve and thus a relatively higher degree of filtering. WhileThe larger the target curve, the higher the degree of filtering, the higher the damage to the original signal, and therefore the constraint is required, whether the fluctuations are characterized by noise or the original signal. Therefore, A and->And the normalized value of the product of (c) represents the filter adjustment coefficient of the target curve.
And (5) using a Vthreshold algorithm to obtain a wavelet threshold corresponding to the target curve.
What needs to be described is: the Chinese name of the Vthreshold algorithm is V-shaped soft and hard threshold method, which is a wavelet threshold denoising algorithm and is a known technology, and the specific method is not described here.
From this, the calculation formula of the optimal wavelet threshold F corresponding to the target curve is known as follows:
where F is the optimal wavelet threshold corresponding to the target curve,and P is a filtering adjustment coefficient of the target curve, and q is a preset constant. In this embodiment, Q is set to 0.5, which is described as an example, and other values may be set in other embodiments, and this embodiment is not limited thereto. Thereby making->The value of (2) is in the range of 0.5 to 1.
Step S005: according to an optimal wavelet threshold corresponding to the target curve, a wavelet threshold denoising algorithm is used to obtain a denoising curve of the target curve; according to the residual curve and the denoising curves of all IMF component curves, using an EMD reconstruction algorithm to obtain a filtering denoising curve of a spectrum curve; and obtaining the perchlorate concentration in the perchlorate wastewater by using a spectral analysis method according to a filtering denoising curve of the spectral curve.
And filtering and denoising the target curve by using a wavelet threshold denoising algorithm according to the optimal wavelet threshold F corresponding to the target curve to obtain a denoising curve of the target curve. The wavelet threshold denoising algorithm is a well-known technique, and a specific method is not described here.
And obtaining a denoising curve of each IMF component curve according to the mode.
And obtaining a filtering denoising curve of the spectrum curve by using an EMD reconstruction algorithm according to the denoising curves and residual curves of all the IMF component curves. The EMD reconstruction algorithm is a well-known technique, and a specific method is not described herein.
And obtaining the perchlorate concentration in the perchlorate wastewater by using a spectral analysis method according to a filtering denoising curve of the spectral curve. The spectroscopic analysis is a well-known technique, and a specific method is not described herein.
What needs to be described is: spectroscopic analysis is to measure the absorption characteristics of light of a specific wavelength using a perchlorate solution. Spectral data can be acquired using a spectrometer or a portable spectrometer and by comparison with a sample of known concentration, a relationship between spectrum and concentration is established to infer the concentration of the sample of unknown concentration.
The present invention has been completed.
In summary, in the embodiment of the present invention, a spectrum curve of perchlorate wastewater is collected, and the spectrum curve is decomposed into a plurality of IMF component curves and a residual curve. And obtaining the fluctuation coefficient of each IMF component curve according to the data difference between the curves. Dividing the spectrum curve and all IMF component curves into a plurality of non-repeated curve segments, and obtaining the total fluctuation duty ratio difference of each IMF component curve according to the absorbance difference between the curve segments. The optimal wavelet threshold corresponding to each IMF component curve is obtained, a wavelet threshold denoising algorithm is used for obtaining a denoising curve of each IMF component curve, and an EMD reconstruction algorithm is used for obtaining a filtering denoising curve of a spectrum curve, so that the concentration of perchlorate in perchlorate wastewater is obtained by using a spectrum analysis method. According to the invention, the optimal wavelet threshold value of each IMF component curve is self-adaptive, and the filtering denoising effect is improved, so that the credibility of data on a spectrum curve is improved, and the accuracy of perchlorate concentration detection is improved.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. The perchlorate wastewater concentration intelligent detection method based on artificial intelligence is characterized by comprising the following steps of:
collecting a spectrum curve of perchlorate wastewater, and decomposing the spectrum curve into a plurality of IMF component curves and a residual curve by using an EMD decomposition algorithm; the abscissa and ordinate of the data point on the curve are wave number and absorbance respectively;
recording any IMF component curve as a target curve; according to the data difference between the target curve and the spectrum curve, obtaining the wave number length, the fluctuation value and the original fluctuation value of each local extreme point on the target curve; obtaining the fluctuation coefficient of the target curve according to the wave number length, the fluctuation value and the original fluctuation value of all local extremum points on the target curve;
dividing the spectrum curve and all IMF component curves into a plurality of non-repeated curve segments; obtaining the fluctuation duty ratio difference of each curve segment of the target curve equal division according to the difference of the absorbance between the curve segments; obtaining the total fluctuation duty ratio difference of the target curve according to the fluctuation duty ratio difference of all curve segments equally divided by the target curve;
obtaining an optimal wavelet threshold corresponding to the target curve according to the fluctuation coefficient and the total fluctuation duty ratio difference of the target curve;
according to an optimal wavelet threshold corresponding to the target curve, a wavelet threshold denoising algorithm is used to obtain a denoising curve of the target curve; according to the residual curve and the denoising curves of all IMF component curves, using an EMD reconstruction algorithm to obtain a filtering denoising curve of a spectrum curve; and obtaining the perchlorate concentration in the perchlorate wastewater by using a spectral analysis method according to a filtering denoising curve of the spectral curve.
2. The method for intelligently detecting the concentration of the perchlorate wastewater based on artificial intelligence according to claim 1, wherein the method for obtaining the wave number length, the fluctuation value and the original fluctuation value of each local extremum point on the target curve according to the data difference between the target curve and the spectrum curve comprises the following specific steps:
a first derivative method is used for respectively obtaining a spectrum curve and local extreme points on all IMF component curves;
obtaining the wave number range and the fluctuation value of each local extremum point according to the wave numbers and the absorbance of all the local extremum points on the target curve;
subtracting the minimum value from the maximum value in the wave number range of each local extreme point on the target curve, and recording the wave number length of each local extreme point;
and counting the absorbance of the local extremum points in the wave number range on the spectrum curve according to the wave number range of each local extremum point on the target curve, and recording the standard deviation of the absorbance of the local extremum points as the original fluctuation value of each local extremum point on the target curve.
3. The method for intelligently detecting the concentration of the perchlorate wastewater based on artificial intelligence according to claim 2, wherein the wave number range and the fluctuation value of each local extremum point are obtained according to the wave numbers and the absorbance of all the local extremum points on a target curve, and the method comprises the following specific steps:
counting wave numbers of two adjacent local extremum points of each local extremum point in all local extremum points on the target curve, and recording a range between the wave numbers of the two adjacent local extremum points as a wave number range of each local extremum point;
and calculating the difference value of the absorbance between each local extremum point and all adjacent local extremum points, and recording the average value of the absolute values of the difference values as the fluctuation value of each local extremum point.
4. The intelligent perchlorate wastewater concentration detection method based on artificial intelligence according to claim 1, wherein the specific calculation formula corresponding to the fluctuation coefficient of the target curve is obtained according to the wave number length, the fluctuation value and the original fluctuation value of all local extremum points on the target curve:
where a is the relief coefficient of the target curve,for the fluctuation value of the ith local extreme point on the target curve, < >>Is the mean value of the fluctuation values of all local extreme points on the target curve, +.>For the wavenumber length of the i-th local extreme point on the target curve, < >>Is the mean value of the wave number lengths of all local extreme points on the target curve, +.>For the original fluctuation value of the ith local extreme point on the target curve, N is the number of the local extreme points on the target curve, +.>Is a linear normalization function.
5. The intelligent perchlorate wastewater concentration detection method based on artificial intelligence according to claim 1, wherein the method is characterized in that the spectral curve and all IMF component curves are equally divided into a plurality of non-repeated curve segments, respectively, and comprises the following specific steps:
calculating the average value of wave number lengths of all local extreme points on a target curve, and recording the downward rounding value of the average value as the period of the target curve;
the least common multiple of the periods of all IMF component curves is recorded as a standard period;
on the horizontal axis, the spectrum curve and all IMF component curves are divided into a plurality of non-repeated curve segments by equally dividing the standard period into equal lengths.
6. The intelligent perchlorate wastewater concentration detection method based on artificial intelligence according to claim 1, wherein the obtaining of the fluctuation duty ratio difference of each curve segment of the target curve equal division according to the difference of absorbance between the curve segments comprises the following specific steps:
in the spectrum curve and all IMF component curves, the spectrum curve or any IMF component curve is recorded as a reference curve;
sequentially counting all curve segments on a reference curve according to the direction from small wave numbers to large wave numbers to obtain a curve segment sequence of the reference curve;
in a curve segment sequence of the spectrum curve, the j-th curve segment is marked as a standard curve segment;
in the curve segment sequence of the target curve, the j-th curve segment is marked as the target curve segment;
and obtaining the fluctuation duty ratio difference of the target curve segment according to the absorbance of all the data points on the standard curve segment and the target curve segment.
7. The intelligent perchlorate wastewater concentration detection method based on artificial intelligence according to claim 6, wherein the specific calculation formula corresponding to the fluctuation duty ratio difference of the target curve segment is obtained according to the absorbance of all data points on the standard curve segment and the target curve segment, and is as follows:
wherein B is the fluctuation duty ratio difference of the target curve segment, C is the standard deviation of the absorbance of all data points on the standard curve segment,for the target curve segmentStandard deviation of absorbance of all data points, +.>Maximum value of absorbance for all data points on standard curve segment, +.>Is the minimum value in absorbance of all data points on the standard curve segment, +.>K is a preset exponential function adjustment value for an exponential function based on a natural constant, ++>Is a linear normalization function, and is an absolute value function.
8. The intelligent perchlorate wastewater concentration detection method based on artificial intelligence according to claim 1, wherein the specific calculation formula corresponding to the total fluctuation duty ratio difference of the target curve is obtained according to the fluctuation duty ratio difference of all curve segments equally divided by the target curve:
wherein the method comprises the steps ofFor the total fluctuation duty cycle difference of the target curve, +.>Fluctuation duty difference of the xth curve segment equally dividing the target curve, +.>Standard deviation of absorbance of all data points on the xth curve segment which is equal to the target curve, and y is the curve segment which is equal to the target curveQuantity of->Is a linear normalization function.
9. The method for intelligently detecting the concentration of the perchlorate wastewater based on artificial intelligence according to claim 1, wherein the method for obtaining the optimal wavelet threshold corresponding to the target curve according to the fluctuation coefficient and the total fluctuation duty ratio difference of the target curve comprises the following specific steps:
calculating a difference value of the total fluctuation duty ratio difference of a subtracted target curve, and recording a normalized value of the product of the difference value and the fluctuation coefficient of the target curve as a filtering adjustment coefficient of the target curve;
using a Vthreshold algorithm to obtain a wavelet threshold corresponding to the target curve;
and obtaining an optimal wavelet threshold corresponding to the target curve according to the wavelet threshold corresponding to the target curve and the filtering adjustment coefficient.
10. The intelligent perchlorate wastewater concentration detection method based on artificial intelligence according to claim 9, wherein the obtaining the optimal wavelet threshold corresponding to the target curve according to the wavelet threshold corresponding to the target curve and the filter adjustment coefficient comprises the following specific steps:
calculating the product of the filter adjustment coefficient of the target curve and q, then calculating the sum of the product of the filter adjustment coefficient of the target curve and q plus q, and recording the product of the sum and the wavelet threshold corresponding to the target curve as the optimal wavelet threshold corresponding to the target curve; and q is a preset constant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311668000.4A CN117368141B (en) | 2023-12-07 | 2023-12-07 | Perchlorate wastewater concentration intelligent detection method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311668000.4A CN117368141B (en) | 2023-12-07 | 2023-12-07 | Perchlorate wastewater concentration intelligent detection method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117368141A true CN117368141A (en) | 2024-01-09 |
| CN117368141B CN117368141B (en) | 2024-03-01 |
Family
ID=89394906
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311668000.4A Active CN117368141B (en) | 2023-12-07 | 2023-12-07 | Perchlorate wastewater concentration intelligent detection method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117368141B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117589741A (en) * | 2024-01-18 | 2024-02-23 | 天津博霆光电技术有限公司 | Indocyanine green intelligent detection method based on optical characteristics |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6311130B1 (en) * | 1996-08-12 | 2001-10-30 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Computer implemented empirical mode decomposition method, apparatus, and article of manufacture for two-dimensional signals |
| CN103901115A (en) * | 2014-04-15 | 2014-07-02 | 南昌航空大学 | Ultrasonic coarse grain material detection method based on EMD (empirical mode decomposition) and wavelet threshold denoising |
| CN109270345A (en) * | 2018-07-20 | 2019-01-25 | 国网江苏省电力有限公司常州供电分公司 | A kind of detection method of mains by harmonics signal |
| CN110507293A (en) * | 2019-07-26 | 2019-11-29 | 中国电子科技集团公司第三十八研究所 | A kind of ultra-broadband wall-through radar human body respiration and heartbeat detecting method and system |
| CN110688964A (en) * | 2019-09-30 | 2020-01-14 | 哈尔滨工程大学 | Wavelet threshold and EMD combined denoising method based on sparse decomposition |
| CN113241087A (en) * | 2021-05-10 | 2021-08-10 | 青岛科技大学 | Underwater sound signal denoising method based on combination of improved VMD and improved wavelet threshold method |
| CN116148393A (en) * | 2023-03-14 | 2023-05-23 | 中国国检测试控股集团股份有限公司 | Method for analyzing odor contribution by determining relative threshold value of odor of substance based on GC-O-MS |
| CN116611609A (en) * | 2023-04-18 | 2023-08-18 | 中国人民解放军海军工程大学 | Equipment stability state prediction and evaluation method based on monitoring parameters |
| CN116992293A (en) * | 2023-09-26 | 2023-11-03 | 北京豪迈生物工程股份有限公司 | Intelligent data processing method for chemiluminescent instrument |
-
2023
- 2023-12-07 CN CN202311668000.4A patent/CN117368141B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6311130B1 (en) * | 1996-08-12 | 2001-10-30 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Computer implemented empirical mode decomposition method, apparatus, and article of manufacture for two-dimensional signals |
| CN103901115A (en) * | 2014-04-15 | 2014-07-02 | 南昌航空大学 | Ultrasonic coarse grain material detection method based on EMD (empirical mode decomposition) and wavelet threshold denoising |
| CN109270345A (en) * | 2018-07-20 | 2019-01-25 | 国网江苏省电力有限公司常州供电分公司 | A kind of detection method of mains by harmonics signal |
| CN110507293A (en) * | 2019-07-26 | 2019-11-29 | 中国电子科技集团公司第三十八研究所 | A kind of ultra-broadband wall-through radar human body respiration and heartbeat detecting method and system |
| CN110688964A (en) * | 2019-09-30 | 2020-01-14 | 哈尔滨工程大学 | Wavelet threshold and EMD combined denoising method based on sparse decomposition |
| CN113241087A (en) * | 2021-05-10 | 2021-08-10 | 青岛科技大学 | Underwater sound signal denoising method based on combination of improved VMD and improved wavelet threshold method |
| CN116148393A (en) * | 2023-03-14 | 2023-05-23 | 中国国检测试控股集团股份有限公司 | Method for analyzing odor contribution by determining relative threshold value of odor of substance based on GC-O-MS |
| CN116611609A (en) * | 2023-04-18 | 2023-08-18 | 中国人民解放军海军工程大学 | Equipment stability state prediction and evaluation method based on monitoring parameters |
| CN116992293A (en) * | 2023-09-26 | 2023-11-03 | 北京豪迈生物工程股份有限公司 | Intelligent data processing method for chemiluminescent instrument |
Non-Patent Citations (3)
| Title |
|---|
| KEEGAN J. MOORE等: "Wavelet-bounded empirical mode decomposition for measured time series analysis", 《MECHANICAL SYSTEMS AND SIGNAL PROCESSING》, pages 14 - 29 * |
| ROBERTAS DAMASEVICIUS等: "IMF mode demixing in EMD for jitter analysis", 《JOURNAL OF COMPUTATIONAL SCIENCE》, pages 240 - 252 * |
| 牛阔;张朝霞;王娟芬;杨玲珍;: "改进EMD与小波阈值相结合的光生混沌信号降噪", 现代电子技术, no. 17, pages 53 - 58 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117589741A (en) * | 2024-01-18 | 2024-02-23 | 天津博霆光电技术有限公司 | Indocyanine green intelligent detection method based on optical characteristics |
| CN117589741B (en) * | 2024-01-18 | 2024-04-05 | 天津博霆光电技术有限公司 | Indocyanine green intelligent detection method based on optical characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117368141B (en) | 2024-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117368141B (en) | Perchlorate wastewater concentration intelligent detection method based on artificial intelligence | |
| CN116660198B (en) | Carbon nanomaterial slurry performance test method based on spectral analysis | |
| CN111238814B (en) | Rolling bearing fault diagnosis method based on short-time Hilbert transform | |
| CN106685478B (en) | Frequency hopping signal parameter estimation method based on signal time-frequency image information extraction | |
| CN107179310B (en) | Raman spectrum characteristic peak recognition methods based on robust noise variance evaluation | |
| CN109187392B (en) | Zinc liquid trace metal ion concentration prediction method based on partition modeling | |
| CN105116442A (en) | Lithologic oil-gas reservoir weak-reflection seismic signal reconstruction method | |
| CN113484257B (en) | Water quality concentration calculating system and method based on neural network and full spectrum absorbance | |
| CN116842326B (en) | Livestock-raising water quality monitoring system based on optical means | |
| CN116818739A (en) | Indocyanine green detection method based on optics | |
| CN105005978A (en) | Spectrum real-time filtering method based on Savitzky-Golay filter parameter optimization | |
| CN112613443A (en) | Robustness communication radiation source intelligent identification method based on deep learning | |
| CN111198167B (en) | Method for detecting concentration of metal ions in solution, terminal device and storage medium | |
| CN110632386B (en) | Solar radio interference filtering method, readable storage medium and electronic equipment | |
| CN103674251B (en) | The identification of cosmic rays interference in Raman spectrum and removing method | |
| CN115221927A (en) | Ultraviolet-visible spectrum dissolved organic carbon detection method | |
| CN114417937A (en) | Deep learning-based Raman spectrum denoising method | |
| CN113686811A (en) | Spectral data processing method based on double sensors | |
| CN115078281B (en) | Water body substance component detection and calculation method based on picture spectral similarity | |
| CN116380869A (en) | Raman spectrum denoising method based on self-adaptive sparse decomposition | |
| CN107727634A (en) | A kind of laser Raman spectroscopy solution spectrum processing method | |
| CN113297722A (en) | Coastal soil salinity assessment method and system | |
| CN112697746A (en) | Spectrum information-based sweet potato powder alum content detection method | |
| CN112229817A (en) | Method for establishing soda saline-alkali soil heavy metal quantitative inversion model | |
| CN116861167B (en) | FBG spectrum cyclic denoising method based on deep learning |
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 |