CN104932027A - Reservoir classification method based on nuclear magnetic resonance logging - Google Patents

Abstract

The invention discloses a reservoir classification method based on nuclear magnetic resonance logging, comprising the steps of: obtaining a nuclear magnetic resonance transverse relaxation time T2 spectrum of a depth point to be classified of a reservoir to be classified; calculating the nuclear magnetic resonance porosity of the depth point to be classified of a reservoir to be classified in dependence on the nuclear magnetic resonance T2 spectrum; employing a doublet Gaussian density function to fit the nuclear magnetic resonance T2 spectrum to obtain parameters representing pore structure characteristics of the depth point to be classified of a reservoir to be classified; employing a cluster analysis method to clarify the depth point to be classified of a reservoir to be classified in dependence on the nuclear magnetic resonance porosity of the depth point to be classified of a reservoir to be classified and the parameters representing pore structure characteristics of the depth point to be classified of a reservoir to be classified; and determining the reservoir type of the reservoir to be classified in dependence on the classification result of the depth point to be classified of a reservoir to be classified. The technical scheme provides strong technical support for accurate divide and reasonable development of a reservoir type.

Description

Based on the Reservoir Classification method of nuclear magnetic resonance log

Technical field

The present invention relates to logging evaluation technical field in oil-gas exploration, particularly a kind of Reservoir Classification method based on nuclear magnetic resonance log.

Background technology

Compact reservoir poor properties, nonuniformity are strong, complex pore structure, and the Reservoir Classification method based on sample experiments analytical test result of coring can not meet the needs of compact reservoir classification.Nuclear magnetic resonance log can provide with the distribution of depth-logger continuous print reservoir pores, and being composed by nuclear magnetic resonance log T2 and extract RESERVOIR PORE STRUCTURE parameter, is the developing direction of Reservoir Classification.

At present, the method utilizing nuclear magnetic resonance log T2 to compose extraction RESERVOIR PORE STRUCTURE parameter is: nuclear magnetic resonance T 2 spectrum is converted to pseudo-capillary pressure curve, the general step of this method for transformation is: first, wellblock objective interval is cored in a large number, and capillary pressure curve and nuclear magnetic resonance T 2 spectrum are measured to sample experiments of coring; Secondly, utilize the transformation models such as linear approach, power function method, two-dimentional equal-area method, regional experience formula, the transformation model after scale, to model scale, is then used for non-cored interval by the capillary pressure curve obtained by core experiment and nuclear magnetic resonance T 2 spectrum; Finally, the pseudo-capillary pressure curve be converted to is utilized to extract parameter of pore structure, for Reservoir Classification.The method Problems existing is, when coring less or core sample does not do the experiment of pressure mercury and nuclear magnetic resonance experiment simultaneously, cannot set up transformation model accurately; In addition, even if obtain good transformation model by core experiment, but by models applying to non-cored interval time, can because the nonuniformity of reservoir, these transformation models are also inapplicable, cause Reservoir Classification effect undesirable.

At present, there is people abroad with bimodal Gaussian density function matching rock core capillary pressure curve, obtain and the closely-related parameter of rock pore structure, and by these parameters, rock is classified, but this method is composed not used for matching nuclear magnetic resonance log T2 and the research of Reservoir Classification.

Summary of the invention

Embodiments provide a kind of Reservoir Classification method based on nuclear magnetic resonance log, in order to improve the accuracy of Reservoir Classification, the method comprises:

The nuclear magnetic resonance T2 T2 obtaining the depth point to be sorted of reservoir to be sorted composes;

According to described nuclear magnetic resonance T 2 spectrum, calculate the NMR porosity of the depth point to be sorted of reservoir to be sorted;

Adopt nuclear magnetic resonance T 2 spectrum described in bimodal Gaussian density function matching, obtain the parameter of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted;

According to NMR porosity and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted, adopt the method for cluster analysis, classified in the depth point to be sorted of reservoir to be sorted;

According to the classification results of the depth point to be sorted to reservoir to be sorted, determine the Reservoir type of reservoir to be sorted.

In one embodiment, according to described nuclear magnetic resonance T 2 spectrum, calculate the NMR porosity of the depth point to be sorted of reservoir to be sorted, comprising:

By all for the nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted range value summations of layouting, obtain the NMR porosity of depth point to be sorted.

In one embodiment, according to the NMR porosity of the depth point to be sorted of following formulae discovery reservoir to be sorted:

$\mathrm{\φ}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i;$

Wherein, φ is the NMR porosity of the depth point to be sorted of reservoir to be sorted, and N is that the cloth of the nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted is counted, p _ifor the range value of nuclear magnetic resonance T 2 spectrum i-th component of the depth point to be sorted of reservoir to be sorted.

In one embodiment, adopt nuclear magnetic resonance T 2 spectrum described in bimodal Gaussian density function matching, obtain the parameter of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted, comprising:

The nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted is added up, obtains the actual measurement accumulation T2 spectrum of the depth point to be sorted of reservoir to be sorted;

Described in the cumulative distribution function matching of bimodal Gaussian density function, survey accumulation T2 spectrum, obtain six parameters of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted.

In one embodiment, six parameters of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted are asked for according to following formula:

$F\left(A\right)={\left|\right|W_s[T{2}_{\mathrm{dist}\_\mathrm{cum}}\left(A\right)-{T2}_{\mathrm{dist}\_\mathrm{cum}\_m}]\left|\right|}_2^2;$

Wherein, A=[ω 1, lg μ 1, lg σ 1, ω 2, lg μ 2, lg σ 2] ^t, W _sfor data weighting matrix, this data weighting matrix when focusing on asking for large and small hole, can provide weights, T2 _{dist_cum}for the cumulative distribution function of bimodal Gaussian density function, T2 _{dist_cum_m}for the actual measurement accumulation T2 spectrum of depth point to be sorted;

By asking for the minimum value of F (A), obtain six parameter: ω 1, μ 1, σ 1, ω 2, μ 2, σ 2 of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted;

Wherein, ω 1 represents the pore volume fraction that aperture occupies; ω 2 represents the pore volume fraction that macropore occupies; μ 1 represents the average of the T2 spectrum that aperture is corresponding; μ 2 represents the average of the T2 spectrum that macropore is corresponding; σ 1 represents the standard deviation of the T2 spectrum that aperture is corresponding, represents the homogeneity of fine porosity; σ 2 represents the standard deviation of the T2 spectrum that macropore is corresponding, represents macroporous homogeneity.

In one embodiment, described bimodal Gaussian density function expression formula is:

$\begin{array}{c}p(\mathrm{lgx};\mathrm{\ω}1,\mathrm{lg\μ}1,\mathrm{lg\σ}1,\mathrm{\ω}2,\mathrm{lg\μ}2,\mathrm{lg\σ}2)\\ =\mathrm{\ω}1\frac{1}{\sqrt{2\mathrm{\π}}\mathrm{lg\σ}1}{e}^{-\frac{{(\mathrm{lgx}-\mathrm{lg\μ}1)}^2}{2{\left(\mathrm{lg\σ}1\right)}^2}}+\mathrm{\ω}2\frac{1}{\sqrt{2\mathrm{\π}}\mathrm{lg\σ}2}{e}^{-\frac{{(\mathrm{lgx}-\mathrm{lg\μ}2)}^2}{2{\left(\mathrm{lg\σ}2\right)}^2}}\end{array};$

Corresponding cumulative distribution function expression formula is:

$\begin{array}{c}p(\mathrm{lgx};\mathrm{\ω}1,\mathrm{lg\μ}1,\mathrm{lg\σ}1,\mathrm{\ω}2,\mathrm{lg\μ}2,\mathrm{lg\σ}2)\\ =\frac{\mathrm{\ω}1}{2}[1+\mathrm{erf}\left(\frac{(\mathrm{lgx}-\mathrm{lg\μ}1)}{\sqrt{2{\left(\mathrm{lg\σ}1\right)}^2}}\right)]+\frac{\mathrm{\ω}2}{2}[1+\mathrm{erf}\left(\frac{(\mathrm{lgx}-\mathrm{lg\μ}2)}{\sqrt{2{\left(\mathrm{lg\σ}2\right)}^2}}\right)]\end{array};$

Wherein, lgx represents the logarithm value of layouting that T2 composes;

Erf function is normally distributed error function:

$\mathrm{erf}\left(z\right)=\frac{2}{\sqrt{\mathrm{\π}}}{\∫}_0^z{e}^{-t^2}\mathrm{dt};$

And ω 1+ ω 2=1, ω 1>0, ω 2>0.

In one embodiment, according to the NMR porosity of the depth point to be sorted of reservoir to be sorted and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted, adopt the method for cluster analysis, classified in the depth point to be sorted of reservoir to be sorted, comprising:

K depth point is selected arbitrarily as cluster centre from the depth point to be sorted of reservoir to be sorted;

According to NMR porosity and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted, depth point to be sorted is categorized in the cluster the most close with described cluster centre;

To NMR porosity and the parameter computation of mean values of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted in each cluster, redefine cluster centre;

With the cluster centre redefined out, to the depth point to be sorted cluster again of reservoir to be sorted, until cluster centre no longer changes, or be less than given threshold value, complete the classification of the depth point to be sorted to reservoir to be sorted.

The technical scheme that the embodiment of the present invention provides, relative in prior art " with bimodal Gaussian density function matching rock core capillary pressure curve, obtain and the closely-related parameter of rock pore structure, and by these parameters, rock is classified " technical scheme, due to bimodal Gaussian density function matching nuclear magnetic resonance T 2 spectrum directly can be adopted, obtain the parameter of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted, again according to the NMR porosity of the depth point to be sorted of reservoir to be sorted and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted, adopt the method for cluster analysis, classified in the depth point to be sorted of reservoir to be sorted, finally, according to the classification results of the depth point to be sorted to reservoir to be sorted, determine the Reservoir type of reservoir to be sorted, improve the accuracy of Reservoir Classification.

Accompanying drawing explanation

Accompanying drawing described herein is used to provide a further understanding of the present invention, forms a application's part, does not form limitation of the invention.In the accompanying drawings:

Fig. 1 is the schematic flow sheet based on the Reservoir Classification method of nuclear magnetic resonance log in the embodiment of the present invention;

Fig. 2 is the Reservoir Classification method based on nuclear magnetic resonance log in the embodiment of the present invention, to the classification results schematic diagram of reservoir to be sorted in A, B, C tri-mouthfuls of wells;

Fig. 3 is that the actual measurement T2 choosing depth point in the embodiment of the present invention composes and matching T2 genealogical relationship curve map;

Fig. 4 is that the actual measurement accumulation T2 spectrum choosing depth point in the embodiment of the present invention accumulates T2 genealogical relationship curve map with matching;

Fig. 5 is the method adopting cluster analysis in the embodiment of the present invention, to the schematic flow sheet classified in the depth point to be sorted of reservoir to be sorted.

Embodiment

For making the object, technical solutions and advantages of the present invention clearly understand, below in conjunction with embodiment and accompanying drawing, the present invention is described in further details.At this, exemplary embodiment of the present invention and illustrating for explaining the present invention, but not as a limitation of the invention.

Bimodal Gaussian density function can be used to characterize rock pore structure feature.The present invention utilizes bimodal Gaussian density function matching nuclear magnetic resonance T 2 spectrum, obtains six parameters characterizing pore structure of reservoir, in conjunction with NMR porosity, carries out Reservoir Classification with K mean cluster.Be described in detail below.

Fig. 1 is the schematic flow sheet based on the Reservoir Classification method of nuclear magnetic resonance log in the embodiment of the present invention, and as shown in Figure 1, the method comprises the steps:

Step 101: the nuclear magnetic resonance T2 T2 obtaining the depth point to be sorted of reservoir to be sorted composes;

Step 102: according to described nuclear magnetic resonance T 2 spectrum, calculates the NMR porosity of the depth point to be sorted of reservoir to be sorted;

Step 103: adopt nuclear magnetic resonance T 2 spectrum described in bimodal Gaussian density function matching, obtains the parameter of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted;

Step 104: according to NMR porosity and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted, adopt the method for cluster analysis, classified in the depth point to be sorted of reservoir to be sorted;

Step 105: according to the classification results of the depth point to be sorted to reservoir to be sorted, determine the Reservoir type of reservoir to be sorted.

The technical scheme that the embodiment of the present invention provides, relative in prior art " with bimodal Gaussian density function matching rock core capillary pressure curve, obtain and the closely-related parameter of rock pore structure, and by these parameters, rock is classified " technical scheme, due to bimodal Gaussian density function matching nuclear magnetic resonance T 2 spectrum directly can be adopted, obtain the parameter of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted, again according to the NMR porosity of the depth point to be sorted of reservoir to be sorted and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted, adopt the method for cluster analysis, classified in the depth point to be sorted of reservoir to be sorted, finally, according to the classification results of the depth point to be sorted to reservoir to be sorted, determine the Reservoir type of reservoir to be sorted, improve the accuracy of Reservoir Classification.

In one embodiment, above-mentioned steps 102 can comprise:

By all for the nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted range value summations of layouting, obtain the NMR porosity of depth point to be sorted.

During concrete enforcement, the embodiment of the present invention utilizes the T2 of nuclear magnetic resonance log to compose each component sum, and passing hole porosity scale, obtains NMR porosity φ.

In one embodiment, according to the NMR porosity of the depth point to be sorted of following formulae discovery reservoir to be sorted:

$\mathrm{\φ}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i;$

Wherein, φ is the NMR porosity of the depth point to be sorted of reservoir to be sorted, and N is that the cloth of the nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted is counted, p _ifor the range value of nuclear magnetic resonance T 2 spectrum i-th component of the depth point to be sorted of reservoir to be sorted.

In addition, can also by the NMR porosity of each depth point of following formulae discovery reservoir to be sorted:

φ=M (0); Wherein, φ is the NMR porosity of the depth point to be asked of reservoir to be sorted, and M (0) is the initial amplitude value of nuclear magnetic resonance original echo string.

In one embodiment, above-mentioned steps 103 can comprise:

The nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted is added up, obtains the actual measurement accumulation T2 spectrum of the depth point to be sorted of reservoir to be sorted;

Described in the cumulative distribution function matching of bimodal Gaussian density function, survey accumulation T2 spectrum, obtain six parameters of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted.

In one embodiment, six parameters of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted are asked for according to following formula:

$F\left(A\right)={\left|\right|W_s[T{2}_{\mathrm{dist}\_\mathrm{cum}}\left(A\right)-{T2}_{\mathrm{dist}\_\mathrm{cum}\_m}]\left|\right|}_2^2;$

Wherein, A=[ω 1, lg μ 1, lg σ 1, ω 2, lg μ 2, lg σ 2] ^t, W _sfor data weighting matrix, this data weighting matrix when focusing on asking for large and small hole, can provide weights, in the embodiment of the present invention, getting 1; T2 _{dist_cum}for the cumulative distribution function (accumulation T2 to be asked composes) of bimodal Gaussian density function, T2 _{dist_cum_m}for the actual measurement accumulation T2 spectrum of depth point to be sorted;

By asking for the minimum value of F (A), obtain six parameter: ω 1, μ 1, σ 1, ω 2, μ 2, σ 2 of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted; Adopt bimodal Gaussian density function matching nuclear magnetic resonance T 2 spectrum, concrete grammar composes based on asking for accumulation T2 to be asked and survey the least mean-square error accumulated T2 and compose, and method used is a kind of method asking lsqnonlin.

Wherein, ω 1 represents the pore volume fraction that aperture occupies; ω 2 represents the pore volume fraction that macropore occupies; μ 1 represents the average of the T2 spectrum that aperture is corresponding; μ 2 represents the average of the T2 spectrum that macropore is corresponding; σ 1 represents the standard deviation of the T2 spectrum that aperture is corresponding, and represent the homogeneity of fine porosity, the sorting of its value larger expression fine porosity network is poorer, and tortuosity is higher; σ 2 represents the standard deviation of the T2 spectrum that macropore is corresponding, and represent macroporous homogeneity, the sorting of its value larger expression macropore network is poorer, and tortuosity is higher.

In one embodiment, described bimodal Gaussian density function expression formula is:

$\begin{array}{c}p(\mathrm{lgx};\mathrm{\ω}1,\mathrm{lg\μ}1,\mathrm{lg\σ}1,\mathrm{\ω}2,\mathrm{lg\μ}2,\mathrm{lg\σ}2)\\ =\mathrm{\ω}1\frac{1}{\sqrt{2\mathrm{\π}}\mathrm{lg\σ}1}{e}^{-\frac{{(\mathrm{lgx}-\mathrm{lg\μ}1)}^2}{2{\left(\mathrm{lg\σ}1\right)}^2}}+\mathrm{\ω}2\frac{1}{\sqrt{2\mathrm{\π}}\mathrm{lg\σ}2}{e}^{-\frac{{(\mathrm{lgx}-\mathrm{lg\μ}2)}^2}{2{\left(\mathrm{lg\σ}2\right)}^2}}\end{array};$

Corresponding cumulative distribution function expression formula is:

$\begin{array}{c}p(\mathrm{lgx};\mathrm{\ω}1,\mathrm{lg\μ}1,\mathrm{lg\σ}1,\mathrm{\ω}2,\mathrm{lg\μ}2,\mathrm{lg\σ}2)\\ =\frac{\mathrm{\ω}1}{2}[1+\mathrm{erf}\left(\frac{(\mathrm{lgx}-\mathrm{lg\μ}1)}{\sqrt{2{\left(\mathrm{lg\σ}1\right)}^2}}\right)]+\frac{\mathrm{\ω}2}{2}[1+\mathrm{erf}\left(\frac{(\mathrm{lgx}-\mathrm{lg\μ}2)}{\sqrt{2{\left(\mathrm{lg\σ}2\right)}^2}}\right)]\end{array};$

Wherein, lgx represents the logarithm value of layouting that T2 composes, and erf function is normally distributed error function:

$\mathrm{erf}\left(z\right)=\frac{2}{\sqrt{\mathrm{\π}}}{\∫}_0^z{e}^{-t^2}\mathrm{dt};$

And ω 1+ ω 2=1, ω 1>0, ω 2>0.

In one embodiment, above-mentioned steps 104 can comprise:

K depth point is selected arbitrarily as cluster centre from the depth point to be sorted of reservoir to be sorted;

According to NMR porosity and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted, depth point to be sorted is categorized in the cluster the most close with described cluster centre;

To NMR porosity and the parameter computation of mean values of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted in each cluster, redefine cluster centre;

With the cluster centre redefined out, to the depth point to be sorted cluster again of reservoir to be sorted, until cluster centre no longer changes, or be less than given threshold value, complete the classification of the depth point to be sorted to reservoir to be sorted.

Be described with example more below, so that understand how to implement the present invention.

The Reservoir Classification method based on nuclear magnetic resonance log utilizing the embodiment of the present invention to provide, carried out Reservoir Classification to the San Koujing (A well, B well and C well) in somewhere, be divided three classes by reservoir, classification results as shown in Figure 2.Wherein, first is gamma ray curve, and second is actual measurement nuclear magnetic resonance T 2 spectrum, 3rd road is matching nuclear magnetic resonance T 2 spectrum, 4th road is NMR porosity, and the 5th road is aperture proportion, and the 6th road is macropore proportion, 7th road is the average that aperture T2 composes, 8th road is the average that macropore T2 composes, and the 9th road is the standard deviation that aperture portion T2 composes, and the tenth road is the standard deviation that macroperforation T2 composes, 10th is for one Reservoir Classification result, and the 12 road is well logging interpretation conclusion.

As can be seen from the classification results of Fig. 2, A well 1-3 reservoir is the Ith class reservoir, perforation formation testing day produce oil 31.03 tons.B well No. 4 to No. 6 reservoirs are I ~ II class reservoir, close examination to No. 4 and No. 6 floor perforations, day produce oil 13.35 tons.C well 7-12 reservoir is the IIIth class reservoir.C well 15-17 layer is I ~ II class reservoir, perforation formation testing day produce oil 12.33 tons.

Below for the depth point of 2266.569 meters, B well, the Reservoir Classification method based on nuclear magnetic resonance log provided by the invention is described in detail introduction.

(1) according to nuclear magnetic resonance log, the nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted is obtained.

(2) NMR porosity of a certain depth point (2266.569 meters) of reservoir to be sorted is calculated:

Following table 1 is the data of B well 2266.569 meters of depth points, as shown in table 1 below, and the cloth points N that this depth point T2 composes is 128, and the NMR porosity φ of this depth point obtained by the 2nd row read group total of following table 1 is: 7.308.

T2 layouts	Original T2 spectrum	The T2 spectrum that matching obtains	Actual measurement accumulation T2 spectrum	Matching accumulation T2 spectrum
					0.1	0	0.0001211	0	0.0001211
0.1095	0	0.0001968	0	0.0003179
					0.1199	0	0.0003144	0	0.0006323
0.1313	0	0.0004932	0	0.0011255
					0.1437	0	0.0007603	0	0.0018858
0.1573	0	0.0011515	0	0.0030373
					0.1723	0	0.0017135	0	0.0047508
0.1886	0	0.002505	0	0.0072558
					0.2065	0.0008913	0.0035981	0.0008913	0.0108539
0.2261	0.0036343	0.0050778	0.0045256	0.0159317
					0.2476	0.0069619	0.0070405	0.0114875	0.0229722
0.2711	0.0108678	0.0095909	0.0223553	0.0325631
					0.2968	0.0153308	0.0128364	0.0376861	0.0453995
0.3249	0.0203165	0.0168795	0.0580026	0.062279
					0.3558	0.02578	0.0218074	0.0837826	0.0840864
0.3895	0.0316679	0.0276808	0.1154505	0.1117672
					0.4265	0.03792	0.034521	0.1533705	0.1462882
0.467	0.0444711	0.0422978	0.1978416	0.188586
					0.5113	0.0512514	0.0509193	0.249093	0.2395053
0.5598	0.0581873	0.0602251	0.3072803	0.2997304
					0.6129	0.0652004	0.0699848	0.3724807	0.3697152
0.6711	0.0722061	0.0799025	0.4446868	0.4496177
					0.7348	0.0791117	0.0896291	0.5237985	0.5392468
0.8045	0.0858134	0.0987804	0.6096119	0.6380272
					0.8808	0.0921938	0.1069613	0.7018057	0.7449885
0.9644	0.0981214	0.1137938	0.7999271	0.8587823
					1.0559	0.1034501	0.118946	0.9033772	0.9777283

1.1561	0.108023	0.1221587	1.0114002	1.099887
					1.2658	0.1116773	0.1232677	1.1230775	1.2231547
1.3859	0.1142523	0.1222175	1.2373298	1.3453722
					1.5174	0.1155991	0.1190671	1.3529289	1.4644393
1.6614	0.1155904	0.1139846	1.4685193	1.5784239
					1.819	0.114131	0.1072338	1.5826503	1.6856577
1.9916	0.1111666	0.0991518	1.6938169	1.7848095
					2.1806	0.1066919	0.090123	1.8005088	1.8749325
2.3876	0.1007561	0.0805503	1.9012649	1.9554828
					2.6141	0.0934663	0.0708274	1.9947312	2.0263102
2.8622	0.0849875	0.061315	2.0797187	2.0876252
					3.1337	0.0755407	0.0523222	2.1552594	2.1399474
3.4311	0.0653956	0.0440954	2.220655	2.1840428
					3.7567	0.0548614	0.0368134	2.2755164	2.2208562
4.1131	0.0442738	0.0305891	2.3197902	2.2514453
					4.5034	0.0339796	0.0254767	2.3537698	2.276922
4.9308	0.0243212	0.0214815	2.378091	2.2984035
					5.3986	0.0156203	0.0185724	2.3937113	2.3169759
5.9109	0.0081643	0.0166942	2.4018756	2.3336701
					6.4718	0.0021945	0.0157786	2.4040701	2.3494487
7.0859	0	0.0157542	2.4040701	2.3652029
					7.7583	0	0.0165532	2.4040701	2.3817561
8.4944	0	0.0181163	2.4040701	2.3998724
					9.3004	0	0.0203944	2.4040701	2.4202668
10.183	0	0.0233492	2.4040701	2.443616
					11.149	0.0038015	0.0269518	2.4078716	2.4705678
12.207	0.0101026	0.0311798	2.4179742	2.5017476
					13.365	0.017764	0.0360144	2.4357382	2.537762
14.634	0.0265707	0.0414371	2.4623089	2.5791991
					16.022	0.0362939	0.0474257	2.4986028	2.6266248
17.543	0.046701	0.0539516	2.5453038	2.6805764
					19.207	0.057565	0.0609767	2.6028688	2.7415531
21.03	0.0686713	0.0684516	2.6715401	2.8100047
					23.025	0.0798217	0.0763136	2.7513618	2.8863183
25.21	0.090836	0.0844863	2.8421978	2.9708046
					27.602	0.1015509	0.0928792	2.9437487	3.0636838
30.221	0.1118187	0.1013886	3.0555674	3.1650724
					33.089	0.1215047	0.1098988	3.1770721	3.2749712
36.229	0.1304853	0.1182845	3.3075574	3.3932557
					39.667	0.138648	0.1264131	3.4462054	3.5196688
43.431	0.1458911	0.1341483	3.5920965	3.6538171
					47.552	0.1521267	0.1413536	3.7442232	3.7951707
52.064	0.1572826	0.1478961	3.9015058	3.9430668
					57.004	0.1613054	0.1536508	4.0628112	4.0967176
62.413	0.1641631	0.1585044	4.2269743	4.255222
					68.335	0.1658469	0.1623587	4.3928212	4.4175807
74.82	0.1663713	0.1651347	4.5591925	4.5827154
					81.919	0.1657739	0.1667742	4.7249664	4.7494896

89.692	0.1641132	0.1672429	4.8890796	4.9167325
					98.203	0.1614656	0.1665309	5.0505452	5.0832634
107.52	0.1579222	0.1646531	5.2084674	5.2479165
					117.72	0.1535847	0.161649	5.3620521	5.4095655
128.9	0.1485612	0.1575812	5.5106133	5.5671467
					141.13	0.1429627	0.152533	5.653576	5.7196797
154.52	0.1368997	0.1466059	5.7904757	5.8662856
					169.18	0.1304788	0.139916	5.9209545	6.0062016
185.23	0.1238009	0.1325901	6.0447554	6.1387917
					202.81	0.1169592	0.1247622	6.1617146	6.2635539
222.05	0.1100383	0.116569	6.2717529	6.3801229
					243.12	0.1031129	0.1081463	6.3748658	6.4882692
266.19	0.0962482	0.0996249	6.471114	6.5878941
					291.45	0.0894996	0.0911281	6.5606136	6.6790222
319.11	0.0829132	0.0827685	6.6435268	6.7617907
					349.39	0.0765264	0.0746459	6.7200532	6.8364366
382.54	0.0703687	0.0668459	6.7904219	6.9032825
					418.84	0.064462	0.059439	6.8548839	6.9627215
458.58	0.0588222	0.0524804	6.9137061	7.0152019
					502.1	0.0534594	0.0460098	6.9671655	7.0612117
549.74	0.0483788	0.0400527	7.0155443	7.1012644
					601.9	0.0435816	0.0346211	7.0591259	7.1358855
659.02	0.0390657	0.0297152	7.0981916	7.1656007
					721.55	0.0348263	0.0253247	7.1330179	7.1909254
790.02	0.0308563	0.0214308	7.1638742	7.2123562
					864.98	0.0271468	0.0180078	7.191021	7.230364
947.06	0.0236878	0.0150249	7.2147088	7.2453889
					1036.9	0.0204684	0.0124477	7.2351772	7.2578366
1135.3	0.017477	0.0102399	7.2526542	7.2680765
					1243	0.0147016	0.0083643	7.2673558	7.2764408
1361	0.0121303	0.0067841	7.2794861	7.2832249
					1490.1	0.009751	0.0054637	7.2892371	7.2886886
1631.5	0.007552	0.0043692	7.2967891	7.2930578
					1786.4	0.0055217	0.0034694	7.3023108	7.2965272
1955.9	0.0036489	0.0027354	7.3059597	7.2992626
					2141.5	0.0019229	0.0021416	7.3078826	7.3014042
2344.7	0.0003335	0.0016648	7.3082161	7.303069
					2567.1	0	0.0012851	7.3082161	7.3043541
2810.7	0	0.0009849	7.3082161	7.305339
					3077.4	0	0.0007496	7.3082161	7.3060886
3369.5	0	0.0005665	7.3082161	7.3066551
					3689.2	0	0.0004251	7.3082161	7.3070802
4039.2	0	0.0003167	7.3082161	7.3073969
					4422.5	0	0.0002343	7.3082161	7.3076312
4842.2	0	0.0001721	7.3082161	7.3078033
					5301.6	0	0.0001255	7.3082161	7.3079288
5804.7	0	0.0000909	7.3082161	7.3080197
					6355.5	0	0.0000654	7.3082161	7.3080851

6958.6	0	0.0000467	7.3082161	7.3081318
					7618.9	0	0.0000331	7.3082161	7.3081649
8341.8	0	0.0000233	7.3082161	7.3081882
					9133.3	0	0.0000163	7.3082161	7.3082045
10000	0	0.0000113	7.3082161	7.3082158

Table 1

(3) adopt nuclear magnetic resonance T 2 spectrum described in bimodal Gaussian density function matching, for each depth point to be sorted, obtain the parameter of the pore structure characteristic characterizing this depth point to be sorted:

First, the nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted is added up, obtain the actual measurement accumulation T2 spectrum of this depth point to be sorted.

During concrete enforcement, shown in as above table 1 the 4th arranges, T2 spectrum nuclear magnetic resonance log obtained is cumulative, obtains actual measurement accumulation T2 and composes T2 _{dist_cum_m}.

Then, described in the cumulative distribution function matching of bimodal Gaussian density function, survey accumulation T2 spectrum, obtain six parameters of the pore structure characteristic characterizing this depth point to be sorted.

During concrete enforcement, be respectively according to six parameter ω 1, μ 1, σ 1, ω 2, the μ 2, σ 2 choosing depth point (2266.569 meters) that above-described embodiment matching obtains: 0.317187,1.2661,1.978065,0.68044,88.85704,2.93846.

In the embodiment of the present invention, the matching accumulation T2 choosing depth point (2266.569 meters) composes as above shown in table 1 the 5th row, and Fig. 3 is that the actual measurement T2 choosing depth point in the present embodiment composes and matching T2 genealogical relationship curve map; Fig. 4 is that the actual measurement accumulation T2 spectrum choosing depth point in the present embodiment accumulates T2 genealogical relationship curve map with matching; From the graph of relation of Fig. 3 and Fig. 4, the actual measurement T2 choosing depth point (2266.569 meters) compose compose with matching T2, survey accumulate T2 compose and matching to accumulate T2 spectrum correlation fine.

(4) according to NMR porosity and six parameters of depth point to be sorted, adopt the method for cluster analysis, treat depth of assortment point and classify:

The NMR porosity that above-described embodiment obtains the depth point of 2266.569 meters, B well is: 7.308, six parameter ω 1, μ 1, σ 1 that matching obtains, ω 2, μ 2, σ 2 is respectively: 0.317187,1.2661,1.978065,0.68044,88.85704,2.93846.According to the Reservoir Classification method that the embodiment of the present invention provides, each depth point to be sorted of A, B, C well, capital obtains factor of porosity as the depth point of 2266.569 meters, above-mentioned B well and six parameters, then, adopt the method for cluster analysis, according to NMR porosity and six parameters of the depth point each to be sorted of reservoir to be sorted, classified in the depth point each to be sorted of reservoir to be sorted.

Fig. 5 is the method adopting cluster analysis in the embodiment of the present invention, to the schematic flow sheet classified in the depth point each to be sorted of reservoir to be sorted, as shown in Figure 5, specifically comprises the steps:

K depth point is selected arbitrarily as cluster centre from the depth point to be sorted of reservoir to be sorted;

According to NMR porosity and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted, depth point to be sorted is categorized in the cluster the most close with described cluster centre;

To NMR porosity and the parameter computation of mean values of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted in each cluster, redefine cluster centre;

With the cluster centre redefined out, to the depth point to be sorted cluster again of reservoir to be sorted, until cluster centre no longer changes, or be less than given threshold value, complete the classification of the depth point to be sorted to reservoir to be sorted.

(5) last, according to the classification results of the depth point to be sorted to reservoir to be sorted, determine the Reservoir type of reservoir to be sorted:

Following table 2 is the Reservoir Classification result of B well 4-6 layer, and as can be seen from the classification results shown in following table 2 and Fig. 2, the classification results of each depth point of B well 4-6 reservoir is I ~ II class reservoir, closes examination to No. 4 and No. 6 floor perforations, day produce oil 13.35 tons.

Table 2

The Reservoir Classification method based on nuclear magnetic resonance log that the embodiment of the present invention provides, can reach following Advantageous Effects:

Application the inventive method achieves the effective classification to reservoir, there is the advantages such as simple, directly perceived, discrimination is high, good reliability, have obvious practical application effect, the Reservoir levels for fine and close oil and gas reservoir divides and reasonable effective exploitation provides strong technical support.

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the embodiment of the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (7)

1., based on a Reservoir Classification method for nuclear magnetic resonance log, it is characterized in that, comprising:

The nuclear magnetic resonance T2 T2 obtaining the depth point to be sorted of reservoir to be sorted composes;

According to described nuclear magnetic resonance T 2 spectrum, calculate the NMR porosity of the depth point to be sorted of reservoir to be sorted;

Adopt nuclear magnetic resonance T 2 spectrum described in bimodal Gaussian density function matching, obtain the parameter of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted;

According to NMR porosity and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted, adopt the method for cluster analysis, classified in the depth point to be sorted of reservoir to be sorted;

According to the classification results of the depth point to be sorted to reservoir to be sorted, determine the Reservoir type of reservoir to be sorted.

2., as claimed in claim 1 based on the Reservoir Classification method of nuclear magnetic resonance log, it is characterized in that, according to described nuclear magnetic resonance T 2 spectrum, calculate the NMR porosity of the depth point to be sorted of reservoir to be sorted, comprising:

By all for the nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted range value summations of layouting, obtain the NMR porosity of depth point to be sorted.

3., as claimed in claim 2 based on the Reservoir Classification method of nuclear magnetic resonance log, it is characterized in that, the NMR porosity according to the depth point to be sorted of following formulae discovery reservoir to be sorted:

$\mathrm{\φ}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i;$

Wherein, φ is the NMR porosity of the depth point to be sorted of reservoir to be sorted, and N is that the cloth of the nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted is counted, p _ifor the range value of nuclear magnetic resonance T 2 spectrum i-th component of the depth point to be sorted of reservoir to be sorted.

4. as claimed in claim 1 based on the Reservoir Classification method of nuclear magnetic resonance log, it is characterized in that, adopt nuclear magnetic resonance T 2 spectrum described in bimodal Gaussian density function matching, obtain the parameter of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted, comprising:

The nuclear magnetic resonance T 2 spectrum of the depth point to be sorted of reservoir to be sorted is added up, obtains the actual measurement accumulation T2 spectrum of the depth point to be sorted of reservoir to be sorted;

Described in the cumulative distribution function matching of bimodal Gaussian density function, survey accumulation T2 spectrum, obtain six parameters of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted.

5. as claimed in claim 4 based on the Reservoir Classification method of nuclear magnetic resonance log, it is characterized in that, ask for six parameters of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted according to following formula:

$F\left(A\right)={\left|\right|W_s[T2\mathrm{dist}\_\mathrm{cum}\left(A\right)T2\mathrm{dist}\_\mathrm{cun}\_m]\left|\right|}_2^2;$

Wherein, A=[ω 1, lg μ 1, lg σ 1, ω 2, lg μ 2, lg σ 2] ^t, W _sfor data weighting matrix, T2dist_cum is the cumulative distribution function of bimodal Gaussian density function, and T2dist_cum_m is the actual measurement accumulation T2 spectrum of depth point to be sorted;

By asking for the minimum value of F (A), obtain six parameter: ω 1, μ 1, σ 1, ω 2, μ 2, σ 2 of the pore structure characteristic of the depth point to be sorted characterizing reservoir to be sorted;

Wherein, ω 1 represents the pore volume fraction that aperture occupies; ω 2 represents the pore volume fraction that macropore occupies; μ 1 represents the average of the T2 spectrum that aperture is corresponding; μ 2 represents the average of the T2 spectrum that macropore is corresponding; σ 1 represents the standard deviation of the T2 spectrum that aperture is corresponding, represents the homogeneity of fine porosity; σ 2 represents the standard deviation of the T2 spectrum that macropore is corresponding, represents macroporous homogeneity.

6., as claimed in claim 5 based on the Reservoir Classification method of nuclear magnetic resonance log, it is characterized in that,

Described bimodal Gaussian density function expression formula is:

$\begin{array}{c}p(\mathrm{lgx};\mathrm{\ω}1,\mathrm{lg\μ}1,\mathrm{lg\σ}1,\mathrm{\ω}2,\mathrm{lg\μ}2,\mathrm{lg\σ}2)\\ =\mathrm{\ω}1\frac{1}{\sqrt{2\mathrm{\π}}\mathrm{lg\σ}1}{e}^{{-\frac{(\mathrm{lgx}-\mathrm{lg\μ}1)}{2{\left(\mathrm{lg\σ}1\right)}^2}}^2}+\mathrm{\ω}2\frac{1}{\sqrt{2\mathrm{\π}}\mathrm{lg\σ}2}{e}^{{-\frac{(\mathrm{lgx}-\mathrm{lg\μ}2)}{2{\left(\mathrm{lg\σ}2\right)}^2}}^2}\end{array}$

Corresponding cumulative distribution function expression formula is:

$\begin{array}{c}p(\mathrm{lgx};\mathrm{\ω}1,\mathrm{lg\μ}1,\mathrm{lg\σ}1,\mathrm{\ω}2,\mathrm{lg\μ}2,\mathrm{lg\σ}2)\\ =\mathrm{\ω}1[1+\mathrm{erf}\left(\frac{(\mathrm{lgx}-\mathrm{lg\μ}1)}{\sqrt{2{\left(\mathrm{lg\σ}1\right)}^2}}\right)]+\mathrm{\ω}2[1+\mathrm{erf}\left(\frac{(\mathrm{lgx}-\mathrm{lg\μ}2)}{\sqrt{2{\left(\mathrm{lg\σ}2\right)}^2}}\right)]\end{array};$

Wherein, lgx represents the logarithm value of layouting that T2 composes;

Erf function is normally distributed error function:

$\mathrm{erf}\left(z\right)=\frac{2}{\sqrt{\mathrm{\π}}}{\∫}_0^z{e}^{{-t}^2}\mathrm{dt};$

And ω 1+ ω 2=1, ω 1>0, ω 2>0.

7. as claimed in claim 1 based on the Reservoir Classification method of nuclear magnetic resonance log, it is characterized in that, according to the NMR porosity of the depth point to be sorted of reservoir to be sorted and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted, adopt the method for cluster analysis, classified in the depth point to be sorted of reservoir to be sorted, comprising:

K depth point is selected arbitrarily as cluster centre from the depth point to be sorted of reservoir to be sorted;

According to NMR porosity and the parameter of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted, depth point to be sorted is categorized in the cluster the most close with described cluster centre;

To NMR porosity and the parameter computation of mean values of the pore structure characteristic of the depth point to be sorted of sign reservoir to be sorted of the depth point to be sorted of reservoir to be sorted in each cluster, redefine cluster centre;

With the cluster centre redefined out, to the depth point to be sorted cluster again of reservoir to be sorted, until cluster centre no longer changes, or be less than given threshold value, complete the classification of the depth point to be sorted to reservoir to be sorted.