CN102254175B - Clustering analysis method for soil of different regions based on x-ray fluorescence spectrums - Google Patents

Abstract

The invention discloses a clustering analysis method for soil of different regions based on x-ray fluorescence spectrums and relates to the technical field of agriculture and environment. The method comprises the following steps of: 1, acquiring soil samples of the different regions; 2, acquiring the x-ray fluorescence spectrums of the soil samples; 3, pre-processing the x-ray fluorescence spectrums; and 4, qualitatively analyzing the pre-processed x-ray fluorescence spectrums by using a mode identification method to obtain the similarity of the soil samples of the different regions. The clustering analysis method for the soil of the different regions based on the x-ray fluorescence spectrums has the advantages that: soil clustering analysis based on the x-ray fluorescence spectrums is realized; research on the soil clustering analysis is enriched; and guidance significance is provided for measurement of soil heavy metal by using the x-ray fluorescence spectrums under the influence of soil parent materials of the different regions.

Description

Zones of different soil clustering method based on the x ray fluorescence spectra

Technical field

The present invention relates to the Agriculture and Environment technical field, particularly a kind of zones of different soil clustering method based on the x ray fluorescence spectra.

Background technology

Soil is all very complicated historical natural complexs of a kind of form and evolutionary process, is ingredient important in Agro-ecological System simultaneously.Soil has the heterogeneity of height because of the even impact of the factor such as mankind's activity of weathers different in soil-forming process, physics, chemistry, biology, matrix, landform, Soil ages on space distribution.The spatial distribution characteristic of soil attribute is the important evidence of agricultural land soil environmental improvement, farmland management and modern agriculture.Clustering methodology, is widely used in soil science research owing to soil assessment being provided the method for science as a kind of mode identification method in modern multivariate statistics.Feng Lixiao etc. select live acid, replacement acid, hydrolysis acid, activated aluminum, cation exchange capacity and degree of base saturation totally 6 factors as the index of cluster analysis, 20, Shaanxi yellowish soil and drab soil are carried out cluster analysis, obtain satisfied classification results.Wu Ke rather uses fuzzy clustering 5 diagnosis indexs of 10 soil in Transition Regions of Northern Subtropics in China-area, south, Henan is analyzed, and contrast with generation classification and diagnostic classification, comparative result shows, fuzzy cluster analysis is with that classification and diagnostic classification occur is substantially identical.

X ray fluorescence spectra technology is a kind of modern equipment analytical technology rapidly that is widely used, develops.It is wide that it can survey elemental range, comprises all elements of Z in the periodic table of elements 〉=3 (Li), can measure concentration range wide, and detection limit reaches 10 ^-5～10 ^-9The advantage of g/g, also have quick and precisely, simple to operate, can measure simultaneously the characteristics such as multiple element, be suitable for the mensuration of polytype solid-state and fluent meterial, and be easy to the robotization of Realization analysis process.The Wavelength dispersion type fluorescence spectrophotometer be analyzing crystal with the dispersion of the fluorescence light velocity after, measure feature x beam wavelength and the intensity of various elements, thereby measure the content of an element; And energy dispersion type fluorescence spectrophotometer is by the responsive semiconductor detector of high resolving power and multichannel analyzer, the x ray fluorescence of dispersion to be separated x-ray spectrum line by photon energy, measures each element according to the height of each element energy.At present, x ray fluorescence spectra technology mainly concentrates in the monitoring of soil and pollution of agricultural products, agricultural product the fields such as determination of trace elements at Application in agriculture.In addition, because this quasi-instrument pre-treatment means in the pedotheque testing process are simple, the analyzing and testing cost is low, the advantages such as detection time is short, a series of portable x ray fluorescence spectrometers have been developed both at home and abroad, thereby improved portability and the real-time of its application, more be conducive to carrying out of soil science correlative study.The x ray is measured particularly heavy metal element of metal elements in soil, and due to the impact that is subject to the factors such as soil types, matrix, Partial Elements may be interfered in the mensuration process, causes the mensuration of Partial Elements inaccurate.

Summary of the invention

The technical matters that (one) will solve

The technical problem to be solved in the present invention is: by the x ray fluorescence spectra, soil is carried out cluster analysis, realize the qualitative identification of zones of different soil.Can set up different rating models according to soil type based on this, to be more conducive to not to be subject to the factors such as soil types, matrix when the x ray is measured metal elements in soil to measuring the interference of process, improve the accuracy that Partial Elements is measured.

(2) technical scheme

For solving the problems of the technologies described above, the invention provides a kind of zones of different soil clustering method based on the x ray fluorescence spectra, comprise the following steps:

S1: the pedotheque that gathers zones of different;

S2: the x ray fluorescence spectra that gathers described pedotheque;

S3: described x ray fluorescence spectra is carried out pre-service;

S4: utilize the pretreated x ray fluorescence spectra of pattern-recongnition method qualitative analysis, obtain the similarity of described zones of different pedotheque.

Wherein, described step S2 specifically comprises:

To the pedotheque of zones of different carry out air-dry, grind, cross 100 order nylon mesh, compressing tablet pre-treatment;

The x ray fluorescence spectra of the pedotheque after the collection pre-treatment.

Wherein, in described step S3, pretreated mode comprises: one or more in signal correction, background removal, level and smooth, noise remove.

Wherein, in described step S4, the mode of pattern-recongnition method qualitative analysis comprises: one or more in principal component analysis (PCA), Euclidean distance method, method of fuzzy cluster analysis, Back Propagation nerual network technique.

Wherein, in described S4, method of fuzzy cluster analysis specifically comprises:

S4.1: the two-dimensional matrix of setting up fluorescence counting rate p.s. and energy value according to two sections of pedotheque different x ray fluorescence spectra data;

S4.2: (a) carries out standardization to numerical value in two-dimensional matrix according to formula;

${x^{\′}}_{\mathrm{ij}}=\frac{x_{\mathrm{ij}}-\stackrel{\‾}{x_j}}{s_j}$ (a), wherein: $\stackrel{\‾}{x_j}=\frac{1}{n}\underset{1}{\overset{n}{\mathrm{\Σ}}}{x}_{\mathrm{ij}},$ $s_j=\sqrt{\frac{\underset{1}{\overset{n}{\mathrm{\Σ}}}{(x_{\mathrm{ij}}-\stackrel{\‾}{x_j})}^2}{n-1}}$

N is the pedotheque number, and i is the pedotheque numbering, and j is x ray fluorescence spectra characteristic peak number;

S4.3: according to standardized value x ' _ij, calculate i and k similarity coefficient r ' that is classified between object with formula (b) _ik, set up similar matrix R ', make r _ik=0.5+0.5 * r ' _ik, set up fuzzy similarity matrix R;

${r^{\′}}_{\mathrm{ik}}=\frac{\underset{k=1}{\overset{n}{\mathrm{\Σ}}}{x^{\′}}_{\mathrm{ij}}{x^{\′}}_{\mathrm{kj}}}{\sqrt{\underset{k=1}{\overset{n}{\mathrm{\Σ}}}{x}_{\mathrm{ij}}^{\′2}}\sqrt{\underset{k=1}{\overset{n}{\mathrm{\Σ}}}{x}_{\mathrm{kj}}^{\′2}}}$ (b) $R^{\′}=\left(\begin{array}{cccc}{r^{\′}}_{\mathrm{1,1}}& {r^{\′}}_{\mathrm{1,2}}& ......& {r^{\′}}_{1,n}\\ {r^{\′}}_{\mathrm{2,1}}& {r^{\′}}_{\mathrm{2,2}}& ......& {r^{\′}}_{2,n}\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ {r^{\′}}_{n,1}& {r^{\′}}_{n,2}& ......& {r^{\′}}_{n,n}\end{array}\right)$

$R=\left(\begin{array}{cccc}{\mathrm{<figure-callout\; id="1"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">r</figure-callout>}}_{\mathrm{1,1}}& {\mathrm{<figure-callout\; id="1"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">r</figure-callout>\; </figure-callout>}}_{\mathrm{1,2}}& ......& r_{1,\mathrm{<figure-callout\; id="2"\; label="n\; r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">n</figure-callout>}}& \\ r_{}{}_{\mathrm{2,1}}& {\mathrm{<figure-callout\; id="2"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">r</figure-callout>\; </figure-callout>}}_{\mathrm{2,2}}& ......& r_{2,n}\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ r_{n,1}& r_{n,2}& ......& r_{n,n}\end{array}\right)$ $R^{*}=\left(\begin{array}{cccc}{r^{*}}_{\mathrm{1,1}}& {r^{*}}_{\mathrm{1,2}}& ......& {r^{*}}_{1,n}\\ {r^{*}}_{\mathrm{2,1}}& {r^{*}}_{\mathrm{2,2}}& ......& {r^{*}}_{2,n}\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ {r^{*}}_{n,1}& {r^{*}}_{n,2}& ......& {r^{*}}_{n,n}\end{array}\right)$

S4.4: calculate R ²=R * R, R ⁴=R ²* R ²..., until R ^2k=R ^kTill, and fuzzy matrix R is transformed to fuzzy equivalent matrix R ^*

S4.5: with fuzzy equivalent matrix R ^*In r ^* _ikDescending arrangement is got λ value, successively as r ^* _ikGreater than λ, make r ^* _ikBe 1; As r ^* _ikLess than λ, make r ^* _ikBe 0, can obtain matrix R ^* _λAt matrix R ^* _λIn with delegation or samely classify 1 as and should divide same class into;

S4.6: repeating step S4.5, wherein get different λ values, progressively sort out, obtain cluster result.

Wherein, described x ray fluorescence spectra comprises: energy dispersion type x ray fluorescence spectra or Wavelength dispersion type x ray fluorescence spectra.

(3) beneficial effect

The research of soil cluster analysis has been enriched in soil cluster analysis based on the x ray fluorescence spectra disclosed by the invention, and the heavy metal-polluted soil of method measure to(for) the x ray fluorescence spectra that affected by the zones of different parent soil material has directive significance.

Description of drawings

Fig. 1 is a kind of based on x ray fluorescence spectra soil clustering method process flow diagram of the embodiment of the present invention;

Fig. 2 is the x ray fluorescence spectra figure of part pedotheque in the embodiment of the present invention;

Fig. 3 is the dendrogram that adopts the cluster analysis that Fuzzy Cluster Analysis method carries out zones of different soil in Fig. 1 method.

Embodiment

Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used for explanation the present invention, but are not used for limiting the scope of the invention.

As shown in Figure 1, a kind of based on x ray fluorescence spectra soil clustering method process flow diagram for the embodiment of the present invention comprises:

Step S101 gathers the zones of different pedotheque.Pedotheque will be able to fully represent the pedotheque of region and the type of soil, is seriously gathering pedotheque in the area away from pollution source and serious artificial interference as far as possible.

Step S102 gathers the x ray fluorescence spectra of pedotheque, and that pedotheque need to carry out is air-dry, grind, cross the pre-treatment work such as 100 order nylon mesh, compressing tablet.Then gather the x ray fluorescence spectra of the pedotheque after pre-treatment.But the collection collecting energy color dispersion-type x ray fluorescence spectra of x ray fluorescence spectra or Wavelength dispersion type x ray fluorescence spectra.

Step S103 carries out pre-service to the x ray fluorescence spectra.Comprise spectrum is carried out one or more in signal correction, background removal, level and smooth, noise remove.

Step S104, utilize the pretreated x ray fluorescence spectra of pattern-recongnition method qualitative analysis, obtain the similarity of described zones of different pedotheque, namely realized the cluster analysis of zones of different soil, refer to that specifically the close pedotheque of the same area or soil parent material will be divided into a class.Sorted soil is carried out the x ray measure particularly heavy metal element of metal elements in soil, improved accuracy of measurement.

The pattern-recognition way of qualitative analysis uses principal component analysis (PCA), Euclidean distance method, method of fuzzy cluster analysis, one or more in Back Propagation (BP) nerual network technique.

The below comes more detailed description the present invention with an object lesson, and wherein the pattern-recognition way of qualitative analysis is method of fuzzy cluster analysis.

Take pedotheque to amount to 77 parts from Heilongjiang Province and farmland, Qinghai Province, wherein the Soil In Heilongjiang Province sample is 38 parts, is numbered 1 to 38; 39 parts, Soil In Qinghai Province sample is numbered 39 to 77.Pre-determining under the precondition of soil region, adopting energy dispersion type x ray fluorescence spectra to carry out the soil cluster analysis, the result demonstration, two class soil can be according to correct the separating in zone.

1, collecting soil sample and processing, sample gathers respectively the typical soil from Heilongjiang Province and Qinghai Province, is respectively 38 parts and 39 parts.Belong to respectively unsaturated siallite and tierras on the order of soil.Sample is gathering away from pollution source and serious artificial interference serious area, and (0～20cm), soil sample grinds at room-dry, deposits after crossing 100 order nylon mesh screens to take topsoil soils.Wood, pottery or agate apparatus are all adopted in the processing such as air-dry, the pack of sample, pulverizing, grinding.

2, the collection of x ray fluorescence spectra uses the portable x ray fluorescence spectrometer of Niton 3t type to gather soil energy color dispersion-type x ray fluorescence spectra, and Fig. 2 is the x ray fluorescence spectra figure from the part pedotheque in Heilongjiang Province and Qinghai Province that collects.

3, spectrum pre-service is carried out to spectrum method that pre-service adopts and is selected from signal correction, background removal, level and smooth, noise remove etc.

4, pattern-recognition is chosen pedotheque x ray fluorescence spectra energy value and is carried out fuzzy cluster analysis at 3KeV～11KeV and two sections fluorescence spectrums of 13KeV～17KeV.The form of pedotheque x ray fluorescence spectra with 2-D data imported in EXCEL, set up the two-dimensional matrix of fluorescence counting rate p.s. (CPS/s) and energy value (KeV).Numerical value in two-dimensional matrix is carried out standardization according to formula (a).

(a), wherein

According to formula (b) compute matrix correlation coefficient r ' _ik, set up similar matrix R '.Make r _ik=0.5+0.5 * r ' _ik, set up fuzzy similarity matrix R.Fuzzy matrix R has reflexivity and symmetry, but does not generally have transitivity.In order to make fuzzy matrix obtain transitivity, need to calculate R ²=R * R, R ⁴=R ²* R ²..., until R ^2k=R ^kTill, fuzzy matrix R is transformed to fuzzy equivalent matrix R ^*

${r^{\&prime;}}_{\mathrm{ik}}=\frac{\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{x^{\&prime;}}_{\mathrm{ij}}{x^{\&prime;}}_{\mathrm{kj}}}{\sqrt{\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{x}_{\mathrm{ij}}^{\&prime;2}}\sqrt{\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{x}_{\mathrm{kj}}^{\&prime;2}}}$ (b) $R^{\&prime;}=\left(\begin{array}{ccccc}{r^{\&prime;}}_{\mathrm{1,1}}& {r^{\&prime;}}_{\mathrm{1,2}}& ...& ...& {r^{\&prime;}}_{\mathrm{1,77}}\\ {r^{\&prime;}}_{\mathrm{2,1}}& {r^{\&prime;}}_{\mathrm{2,2}}& ...& ...& {r^{\&prime;}}_{\mathrm{2,77}}\\ .& .& .& .& .\\ .& .& .& .& .\\ .& .& .& .& .\\ .& .& .& .& .\\ {r^{\&prime;}}_{\mathrm{77,1}}& {r^{\&prime;}}_{\mathrm{77,2}}& ...& ...& {r^{\&prime;}}_{\mathrm{77,77}}\end{array}\right)$

$R=\left(\begin{array}{ccccc}{\mathrm{<figure-callout\; id="1"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">r</figure-callout>}}_{\mathrm{1,1}}& {\mathrm{<figure-callout\; id="1"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">r</figure-callout>\; </figure-callout>}}_{\mathrm{1,2}}& ...& ...& r_{\mathrm{1,77}}\\ {\mathrm{<figure-callout\; id="2"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">r</figure-callout>}}_{\mathrm{2,1}}& {\mathrm{<figure-callout\; id="2"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">r</figure-callout>\; </figure-callout>}}_{\mathrm{2,2}}& ...& ...& r_{\mathrm{2,77}}\\ .& .& .& .& .\\ .& .& .& .& .\\ .& .& .& .& .\\ .& .& .& .& .\\ r_{\mathrm{77,1}}& {\mathrm{<figure-callout\; id="77"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="r"\; filenames="HDA0000054188570000021.png"\; state="\{\{state\}\}">r</figure-callout>\; </figure-callout>}}_{\mathrm{77,2}}& ...& ...& r_{\mathrm{77,77}}\end{array}\right)$ $R^{*}=\left(\begin{array}{ccccc}{r^{*}}_{\mathrm{1,1}}& {r^{*}}_{\mathrm{1,2}}& ...& ...& {r^{*}}_{\mathrm{1,77}}\\ {r^{*}}_{\mathrm{2,1}}& {r^{*}}_{\mathrm{2,2}}& ...& ...& {r^{*}}_{\mathrm{2,77}}\\ .& .& .& .& .\\ .& .& .& .& .\\ .& .& .& .& .\\ .& .& .& .& .\\ {r^{*}}_{\mathrm{77,1}}& {r^{*}}_{\mathrm{77,2}}& ...& ...& {r^{*}}_{\mathrm{77,77}}\end{array}\right)$

With fuzzy equivalent matrix R ^*In r ^* _ikDescending arrangement is got λ value, successively as r ^*i _kGreater than λ, make r ^* _ikBe 1; As r ^* _ikLess than λ, make r ^* _ikBe 0, can obtain matrix R ^* _λAt matrix R ^* _λIn with delegation or samely classify 1 as and should divide same class into; Get different λ values, progressively sort out, and then get λ, then carry out cluster, its step is fully the same.Progressively get λ value, progressively sort out, obtain at last cluster result.The dendrogram of above numerical operation and two places soil sample (Fig. 3) all uses Matlab 7.0 software computings.Found out by Fig. 3, carry out fuzzy cluster analysis from the x ray spectrum, pedotheque from Heilongjiang Province and Qinghai Province can very clearly be divided into two classes, the similarity of two places soil is 64.9%, namely when λ gets 0.649, all pedotheques are poly-is a class, with 64.9% similarity as two places soil.And two places separately the similarity of soil be respectively 75.8% and 85.3%.

Method of fuzzy cluster analysis in the present invention also can adopt principal component analysis (PCA), Euclidean distance method, Back Propagation nerual network technique, all as the input data with the x ray fluorescence spectra of soil, calculate similarity according to different separately modes at last, can reach same effect.

The research of soil cluster analysis has been enriched in soil cluster analysis based on the x ray fluorescence spectra disclosed by the invention, and the heavy metal-polluted soil of method measure to(for) the x ray fluorescence spectra that affected by the zones of different parent soil material has directive significance.

Above embodiment only is used for explanation the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; without departing from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.

Claims (4)

1. the zones of different soil clustering method based on the x ray fluorescence spectra, is characterized in that, comprises the following steps:

S1: the pedotheque that gathers zones of different;

S2: the x ray fluorescence spectra that gathers described pedotheque;

S3: described x ray fluorescence spectra is carried out pre-service;

S4: utilize the pretreated x ray fluorescence spectra of pattern-recongnition method qualitative analysis, obtain the similarity of described zones of different pedotheque, the mode of described pattern-recongnition method qualitative analysis comprises method of fuzzy cluster analysis, and wherein method of fuzzy cluster analysis specifically comprises:

S4.1: the two-dimensional matrix of setting up fluorescence counting rate p.s. and energy value according to two sections of pedotheque different x ray fluorescence spectra data;

S4.2: (a) carries out standardization to numerical value in two-dimensional matrix according to formula;

${x^{\&prime;}}_{\mathrm{ij}}=\frac{x_{\mathrm{ij}}-\stackrel{\&OverBar;}{x_j}}{s_j}---\left(a\right),$ Wherein: $\stackrel{\&OverBar;}{x_j}=\frac{1}{n}\underset{1}{\overset{n}{\mathrm{\&Sigma;}}}{x}_{\mathrm{ij}},$ $s_j=\sqrt{\frac{\underset{1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_{\mathrm{ij}}-\stackrel{\&OverBar;}{x_j})}^2}{n-1}}$

N is the pedotheque number, and i is the pedotheque numbering, and j is x ray fluorescence spectra characteristic peak number;

S4.3: according to standardized value x' _ij, calculate i and k similarity coefficient r' that is classified between object with formula (b) _ik, set up similar matrix R', make r _ik=0.5+0.5 * r' _ik, set up fuzzy similarity matrix R;

${r^{\&prime;}}_{\mathrm{ik}}=\frac{\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{x^{\&prime;}}_{\mathrm{ij}}{x^{\&prime;}}_{\mathrm{kj}}}{\sqrt{\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{x^{\&prime;}}_{\mathrm{ij}}^2}\sqrt{\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{x^{\&prime;}}_{\mathrm{kj}}^2}}---\left(b\right)$

$R^{\&prime;}=\left(\begin{array}{cccc}{r^{\&prime;}}_{\mathrm{1,1}}& {r^{\&prime;}}_{\mathrm{1,2}}& ......& {r^{\&prime;}}_{1,n}\\ {r^{\&prime;}}_{\mathrm{2,1}}& {r^{\&prime;}}_{\mathrm{2,2}}& ......& {r^{\&prime;}}_{2,n}\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ {r^{\&prime;}}_{n,1}& {r^{\&prime;}}_{n,2}& ......& {r^{\&prime;}}_{n,n}\end{array}\right)$ $R=\left(\begin{array}{cccc}{r}_{\mathrm{1,1}}& r_{\mathrm{1,2}}& ......& r_{1,n}\\ r_{\mathrm{2,1}}& r_{\mathrm{2,2}}& ......& r_{2,n}\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ r_{n,1}& r_{n,2}& ......& r_{n,n}\end{array}\right);$

S4.4: calculate R ²=R * R, R ⁴=R ²* R ²..., until R ^2k=R ^kTill, and fuzzy matrix R is transformed to fuzzy equivalent matrix R*:

$R^{*}=\left(\begin{array}{cccc}{r^{*}}_{\mathrm{1,1}}& {r^{*}}_{\mathrm{1,2}}& ......& {r^{*}}_{1,n}\\ {r^{*}}_{\mathrm{2,1}}& {r^{*}}_{\mathrm{2,2}}& ......& {r^{*}}_{2,n}\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ .& .& .& .\\ {r^{*}}_{n,1}& {r^{*}}_{n,2}& ......& {r^{*}}_{n,n}\end{array}\right);$

S4.5: with fuzzy equivalent matrix R ^*In r ^* _ikDescending arrangement is got λ value, successively as r ^* _ikGreater than λ, make r ^* _ikBe 1; As r ^* _ikLess than λ, make r ^* _ikBe 0, can obtain matrix R ^* _λ, at matrix R ^* _λIn with delegation or samely classify 1 as and should divide same class into;

S4.6: repeating step S4.5, wherein get different λ values, progressively sort out, obtain cluster result.

2. the zones of different soil clustering method based on the x ray fluorescence spectra as claimed in claim 1, is characterized in that, described step S2 specifically comprises:

To the pedotheque of zones of different carry out air-dry, grind, cross 100 order nylon mesh, compressing tablet pre-treatment;

The x ray fluorescence spectra of the pedotheque after the collection pre-treatment.

3. the zones of different soil clustering method based on the x ray fluorescence spectra as claimed in claim 1, is characterized in that, in described step S3, pretreated mode comprises: one or more in signal correction, background removal, level and smooth, noise remove.

4. the zones of different soil clustering method based on the x ray fluorescence spectra as described in claim 1 ~ 3 any one, is characterized in that, described x ray fluorescence spectra comprises: energy dispersion type x ray fluorescence spectra or Wavelength dispersion type x ray fluorescence spectra.

Images