Lithology quantitative spectrum method for identifying lithology of tight reservoir
阅读说明：本技术 一种致密储层岩性识别的岩性定量谱方法 (Lithology quantitative spectrum method for identifying lithology of tight reservoir ) 是由 徐敬领 于 2020-07-09 设计创作，主要内容包括：本发明公开了一种致密储层岩性识别的岩性定量谱方法,包括：基于岩心、岩心分析资料及岩屑录井标定常规测井数据,找出对不同岩性敏感的测井数据,建立不同岩性的常规测井响应关系识别图版；根据不同岩性的测井响应关系与差异,结合数据分离原理,建立不同岩性的识别模型Q,实现岩性的有效区分；根据不同岩性的Q值分布区间不同,通过不同色系代表不同岩性,建立了岩性识别的单边谱和镜像谱,以直观、简洁、智能的方式实现岩性的识别与定量表征。其识别结果与岩心完全吻合,为之后致密储层参数的计算提供了直观、有效的参考依据。使用常规测井资料识别和表征岩性,极大的提高了测井解释的效率、降低了测井解释的成本。(The invention discloses a lithology quantitative spectrum method for identifying lithology of a tight reservoir, which comprises the following steps: calibrating conventional logging data based on the rock core, rock core analysis data and rock debris logging, finding out logging data sensitive to different lithologies, and establishing conventional logging response relation identification charts of different lithologies; according to the logging response relation and difference of different lithologies, and in combination with a data separation principle, establishing an identification model Q of different lithologies to realize effective differentiation of the lithologies; according to different Q value distribution intervals of different lithologies, different lithologies are represented through different color systems, a unilateral spectrum and a mirror image spectrum of lithology identification are established, and lithology identification and quantitative characterization are achieved in a visual, concise and intelligent mode. The recognition result is completely consistent with the core, and an intuitive and effective reference basis is provided for the calculation of the parameters of the compact reservoir later. The lithology is identified and represented by using the conventional logging information, the efficiency of logging interpretation is greatly improved, and the cost of logging interpretation is reduced.)
1. A lithology quantitative spectrum method for tight reservoir lithology identification is characterized by comprising the following steps:
step 1, firstly, calibrating lithology determined by rock core, rock core analysis data and rock debris logging data into a corresponding conventional logging profile, counting conventional logging response data of different lithologies, and finding out sensitive logging data of different lithologies, wherein the conventional logging response data comprises: natural Gamma Rays (GR), sound wave time difference (AC), Density (DEN), Compensation Neutrons (CNL) and Resistivity (RT), establishing logging response relation recognition charts with different lithologies, and preferably selecting conventional logging data sensitive to different lithologies according to the logging response relation recognition charts with different lithologies, wherein the conventional logging data comprises the following steps: natural gamma, sonic moveout and resistivity;
step 2, carrying out normalization on the natural Gamma (GR), the Resistivity (RT) and the acoustic time difference (AC) of logging data sensitive to different lithologies:
ACchinese angelica root-bark＝AC/ACmax(1)
GRChinese angelica root-bark＝GR/GRmax(2)
logRTChinese angelica root-bark＝logRT/logRTmax(3)
Wherein, ACChinese angelica root-barkIs the normalized value of the sonic moveout log data, and AC is the sonic moveout, unit: μ s/m; ACmaxTo address the maximum value of interval acoustic moveout, the unit: μ s/m; GRChinese angelica root-barkIs the normalized value of the natural gamma log data, GR is the natural gamma, unit: API, GRmaxFor the maximum value of the natural gamma of the treatment interval, the unit: an API; logRTChinese angelica root-barkTaking a logarithmic normalization value for resistivity logging data, wherein RT is resistivity, and the unit is as follows: omega m, RTmaxTo treat the maximum value of the interval resistivity, the unit: omega.m;
step 3, finding and finding out logging response characteristics and response differences of different lithologies, and identifying a cross plot according to logging response relations of different lithologies to obtain: (1) the logging responses of different lithologies have respective response intervals, although the response intervals are overlapped, the response intervals are obviously distinguished; (2) the logging response relation of different lithologies is obvious, namely the natural gamma of different lithologies is inversely related to the resistivity, the resistivity of different lithologies is reduced along with the increase of the natural gamma, the natural gamma of different lithologies is positively related to the time difference of sound waves, the time difference of sound waves of different lithologies is increased along with the increase of the natural gamma, and the time difference of sound waves of different lithologies is inversely related to the resistivity;
step 4, establishing recognition models of different lithologies of the compact reservoir to further distinguish the lithologies according to the logging response characteristics and response differences of the different lithologies, particularly response relations and differences of natural gammas, resistivity and acoustic time difference of the different lithologies, namely that the natural gammas of the different lithologies are in inverse correlation with the resistivity, the natural gammas of the different lithologies are in positive correlation with the acoustic time difference and the acoustic time difference of the different lithologies is in inverse correlation with the resistivity; establishing a plurality of lithological identification models Q according to the response relation and difference and by combining the principle of mathematical data difference1、Q2、Q3And Q4；
Wherein Q is1Identifying model 1, Q for lithology2Identifying model 2, Q for lithology3Identifying model 3, Q for lithology4For lithology recognition model 4, ACChinese angelica root-barkIs the normalized value of the sonic moveout log data, and AC is the sonic moveout, unit: μ s/m; ACmaxTo address the maximum value of interval acoustic moveout, the unit: μ s/m; GRChinese angelica root-barkIs the normalized value of the natural gamma log data, GR is the natural gamma, unit: an API; GRmaxFor the maximum value of the natural gamma of the treatment interval, the unit: an API; logRTChinese angelica root-barkTaking a logarithmic normalization value for resistivity logging data, wherein RT is resistivity, and the unit is as follows: omega.m; RT (reverse transcription)maxTo treat the maximum value of the interval resistivity, the unit: omega.m;
and 5, processing and calculating data according to the established lithology recognition model Q, so that Q values of different lithologies can be calculated, wherein the Q values are as follows: drawing a Q value distribution and recognition effect graph of each lithology according to the Q values of the limestone, the sandstone, the silty sand, the gray mudstone, the sandy mudstone and the mudstone, and obtaining distribution graphs of different lithology recognition models Q1, Q2, Q3 and Q4;
step 6, comparing the Q value distribution and the recognition effect graphs of different lithologies, finding out the graphs with independent and best-distinguished Q value distributions of different lithologies, and identifying the model Q which is most sensitive to different lithologies and has the best distinguishing degree4As a final lithology recognitionOther model parameters;
step 7, analyzing lithology recognition model parameter Q4Obtaining a Q value interval of each lithology: limestone Q value>1.2, the Q value of marl is 1.2-1.1, the Q value of sandstone is 1.1-1, the Q value of siltstone is 1-0.9, the Q value of gray mudstone is 0.9-0.8, the Q value of argillaceous siltstone is 0.8-0.6, the Q value of sandy mudstone is 0.6-0.5, and the Q value of mudstone is<0.5, the Q values of different lithologies are different in interval and size, the Q value of mudstone is the minimum, the Q value of limestone is the maximum, namely the Q value of a reservoir stratum with relatively good lithology is large, the mudstone is weakened, the visual effect is enhanced, and therefore lithology is effectively identified and divided;
step 8, in order to visually and effectively distinguish lithology in the well, the Q value intervals of different lithologies are changed according to color systems according to the size intervals and distribution of the Q values of different lithology identification model parameters, so that the Q value intervals of different lithologies are filled with different colors, and a lithology unilateral color system spectrum for lithology identification and a quantitative mirror image spectrum for lithology identification are respectively established; the two spectrograms are in perfect correspondence, and visual and quantitative identification and distinction of different lithologies of the compact reservoir are realized.
Technical Field
The invention relates to the technical field of oil and gas exploration and development, in particular to a lithology quantitative spectrum method for identifying lithology of a tight reservoir.
Background
The lithology determines the difference between the quality of a compact reservoir and a development scheme, the lithology is the basis of evaluation of the compact reservoir and development of an oil-gas reservoir, and the lithology identification is of great importance. The tight reservoir is different from the conventional reservoir, and the lithologic various mineral components of the tight reservoir are complex, including: various lithologies such as fine sandstone, siltstone, argillaceous sandstone, sandy mudstone, gray matter mudstone, marl rock, shale and the like, the mineral components are more complicated, and how to effectively identify the lithologies of compact reservoirs is a key problem to be solved urgently at present.
At present, methods for identifying lithology of tight reservoirs roughly have three types: 1) the lithology and mineral composition are determined by drilling and coring, core observation and core analysis tests, the method is most directly and effectively used for identifying the lithology, but the drilling and coring are expensive and limited, so the method is limited greatly. 2) Lithology can be identified by cuttings logging, but cuttings logging typically describes lithology every 1 or 2 meters, and the description is coarse and biased. 3) The lithology is identified by utilizing the electrical imaging logging information, the electrical imaging logging can image the underground stratum, the lithology identification is realized, but the imaging logging is expensive and limited in data, and the lithology is manually identified after the imaging logging graph is calibrated by a rock core, so that the imaging logging is expensive, time-consuming, labor-consuming and not intelligent in lithology identification.
The lithology of the tight reservoir is various, the mineral composition is complicated, the lithology is recognized by simply adopting the rock core and rock core analysis data 1), although the coring data is accurate, the coring data is effective after all, the lithology recognition cannot be carried out comprehensively, the method is not advisable, 2) the rock debris logging is adopted for lithology recognition, 1 m of lithology is recognized, the recognition is too coarse and not fine, the depth error exists, and 3) the imaging logging data is adopted for recognizing the lithology, so that the efficiency is low, the cost is high, and the data are limited. Therefore, how to utilize the conventional logging information to carry out quick and effective identification on lithology, the identification precision is improved, and the efficiency of logging interpretation is further improved, and the method has great significance.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a lithology quantitative spectrum method for identifying lithology of a tight reservoir, and solves the defects in the prior art.
In order to realize the purpose, the technical scheme adopted by the invention is as follows:
a lithology quantitative spectrum method for tight reservoir lithology identification comprises the following steps:
ACchinese angelica root-bark＝AC/ACmax(1)
GRChinese angelica root-bark＝GR/GRmax(2)
logRTChinese angelica root-bark＝logRT/logRTmax(3)
Wherein, ACChinese angelica root-barkIs the normalized value of the logging data of the acoustic time difference, and AC is the acoustic time difference (unit: mu s/m); ACmaxThe maximum value (unit: mu s/m) of the acoustic time difference of the well section is treated; GRChinese angelica root-barkIs the normalized value of natural gamma log data, GR is the natural gamma (unit: API), GR is the natural gammamaxIs the maximum value (unit: API) of the natural gamma of the processing well section; logRTChinese angelica root-barkTaking the normalized value of the logarithm of the resistivity logging data, wherein RT is the resistivity (unit: omega. m), and RT ismaxThe maximum value (unit: Ω · m) of the resistivity of the treatment interval.
Wherein Q is1Identifying
And 5, processing and calculating data according to the established lithology recognition model Q, so that Q values of different lithologies can be calculated, wherein the Q values are as follows: and drawing a Q value distribution and recognition effect graph of each lithology according to the Q values of the limestone, the sandstone, the silty sand, the gray mudstone, the sandy mudstone and the mudstone, and obtaining a distribution graph of different lithology recognition models Q1, a distribution graph of different lithology recognition models Q2, a distribution graph of different lithology recognition models Q3 and a distribution graph of different lithology recognition models Q4.
Compared with the prior art, the invention has the advantages that:
1. the method uses the conventional logging information to identify the lithology of the tight reservoir, greatly improves the efficiency and the precision of logging interpretation and reduces the cost of logging interpretation.
2. The invention uses the conventional logging information to establish the logging response relation recognition charts with different lithologies, so that sensitive logging data with different lithologies are firstly obtained, the logging responses with different lithologies have respective response intervals, the logging response relations are obvious, namely the natural gamma and the resistivity of different lithologies (limestone, sandstone, silt, gray mudstone, sandy mudstone and mudstone) are in inverse correlation, the natural gamma and the sonic time difference of different lithologies are in a positive correlation rule, and the sonic time difference and the resistivity of different lithologies are in inverse correlation, thereby laying a foundation for the quantitative recognition of the lithologies.
3. The lithology recognition model Q is established by using the acoustic wave time difference of different lithologies and the sensitive response relation and difference between natural gamma and resistivity, and guiding basis is provided for quantitative lithology recognition of the tight reservoir and calculation of other parameters.
4. The invention establishes the lithology spectrum capable of visually identifying the lithology, has very wide application prospect and provides a basic and effective evaluation standard for parameter calculation and reservoir evaluation of a compact reservoir later.
Drawings
FIG. 1 is a flow chart of a lithology quantitative spectroscopy method of the present invention;
FIG. 2 is a cross-sectional view of log response for different lithologies;
FIG. 3 is a cross plot of natural Gamma (GR) and Resistivity (RT) log response for different lithologies;
FIG. 4 is a cross plot of natural Gamma (GR) and sonic time difference (AC) log responses for different lithologies;
FIG. 5 is a cross plot of acoustic moveout (AC) and Resistivity (RT) log response for different lithologies;
FIG. 6 is a graph of the distribution and recognition effect of different lithology recognition model Q1 values;
FIG. 7 is a graph of the distribution and recognition effect of different lithology recognition model Q2 values;
FIG. 8 is a graph of the distribution and recognition effect of different lithology recognition model Q3 values;
FIG. 9 is a graph of the distribution and recognition effect of different lithology recognition model Q4 values;
FIG. 10 is a quantitative spectral profile for tight reservoir lithology identification.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings by way of examples.
The invention relates to calibration of conventional logging data based on core and core analysis data and rock debris logging, wherein lithology determined by core and core analysis and rock debris logging is calibrated on a conventional logging curve as shown in figure 2, conventional logging response data of different lithology is counted, logging data sensitive to different lithology is found, and first, a conventional logging response relation identification chart of different lithology is established, and the logging response relation identification chart of different lithology is reflected as shown in figures 3, 4 and 5. According to the logging response relations and differences of different lithologies, namely that the natural gamma and the resistivity of different lithologies are in inverse correlation, the natural gamma and the sound wave time difference are in positive correlation, and the sound wave time difference and the resistivity are in inverse correlation, the data separation principle is combined, then the identification models Q of different lithologies are established, effective distinguishing of the lithologies is achieved, as shown in the figures 6, 7, 8 and 9, the effects of the different lithologies are reflected, the Q4 model is found to have the best lithology identification effect through comparison, and the Q4 model is selected as the final lithology identification model. And finally, according to different Q value distribution intervals of different lithologies, representing different lithologies through different color systems, establishing a unilateral spectrum and a mirror spectrum for lithology identification, wherein as shown in FIG. 10, the actual identification application effect of different lithologies is reflected, and the lithology identification and quantitative characterization are realized in a visual, concise and intelligent manner.
As shown in fig. 1, the steps of the present invention are as follows:
ACchinese angelica root-bark＝AC/ACmax(1)
GRChinese angelica root-bark＝GR/GRmax(2)
logRTChinese angelica root-bark＝logRT/logRTmax(3)
Wherein, ACChinese angelica root-barkIs the normalized value of the logging data of the acoustic time difference, and AC is the acoustic time difference (unit: mu s/m); ACmaxThe maximum value (unit: mu s/m) of the acoustic time difference of the well section is treated; GRChinese angelica root-barkIs the normalized value of natural gamma log data, GR is the natural gamma (unit: API), GR is the natural gammamaxIs the maximum value (unit: API) of the natural gamma of the processing well section; logRTChinese angelica root-barkTaking the normalized value of the logarithm of the resistivity logging data, wherein RT is the resistivity (unit: omega. m), and RT ismaxThe maximum value (unit: Ω · m) of the resistivity of the treatment interval.
And 4, establishing an identification model of the different lithologies of the compact reservoir to further distinguish the lithologies according to the logging response characteristics and the response differences of the different lithologies, particularly response relations and differences among the natural gamma and resistivity, the natural gamma and acoustic wave time difference and the resistivity of the different lithologies. According to the response relation and the difference,establishing a plurality of lithologic identification models Q1, Q2 and Q by combining the principle of mathematical data difference3And Q4。
Wherein Q is1Identifying
And 5, processing and calculating data according to the established lithological identification model Q to calculate Q values of different lithologies, and drawing a Q value distribution and identification effect graph of each lithology according to the Q values of different lithologies (limestone, sandstone, silty sand, grey mud rock, sandy mud rock and mudstone), namely the distribution of the different lithological identification model Q1 is shown in figure 6, the distribution of the different lithological identification model Q2 is shown in figure 7, the distribution of the different lithological identification model Q3 is shown in figure 8, and the distribution of the different lithological identification model Q4 is shown in figure 9.
And 8, in order to intuitively and effectively distinguish lithology in the well, changing the Q value intervals of different lithologies according to color systems according to the size intervals and distribution of the Q values of different lithology identification model parameters, filling different colors into the Q value intervals of different lithologies, and respectively establishing a lithology unilateral color system spectrum (a first row in a graph 10) for lithology identification and a quantitative mirror image spectrum (a second row in the graph 10) for lithology identification. The two spectrograms are in perfect correspondence, and visual and quantitative identification of different lithologies of the compact reservoir is realized.
- 上一篇：一种医用注射器针头装配设备
- 下一篇：一种强剥蚀区的金矿勘探方法