Miscibility of linear low density polyethylene (LLDPE)-graft-poly(methyl methacrylate) with poly(vinylidene fluoride) (PVF2) and its application as compatibilizer LLDPE/PVF2 blends

Differential scanning calorimetry (DSC), Fourier-transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM), and X-ray photoelectron spectroscopy (XPS) were used to study the miscibility of blends of a graft copolymer of poly(methyl methacrylate) on linear low density polyethylene (LLDPE-g-PMMA, G-3) with poly(vinylidene fluoride)(b) (PVF2) and the compatibilization of blends of LLDPE/PVF2. The specific interaction between PMMA side chains and PVF2 in G-3/PVF2 binary blends is weaker than that between the homopolymers PMMA and PVF2. There are two states of PVF2 in the melt of a G-3/PVF2 (60/40, w/w) blend, one as pure PVF2 and the other interacting with PMMA side chains. The miscibility between PMMA side chains and PVF2 affects the crystallization of PVF2. LLDPE-g-PMMA was demonstrated to be a good compatibilizer in LLDPE/PVF2 blends, improving the interfacial adhesion and dispersion in the latter. Diffusion of PMMA side chains into PVF2 in the interfacial region reduces the crystallization rate and lowers the melting point (T-m) and the crystallization temperature (T-c) of PVF2 in the blends.

The complete amino acid sequence of growth hormone and partial amino acid sequence of prolactin and somatolactin from sea perch (Lateolabrax japonicus)

Growth hormone (GH), prolactin (PRL) and somatolactin (SL) were purified simultaneously under alkaline condition (pH 9.0) from pituitary glands of sea perch (Lateolabrax japonicas) by a two-step procedure involving gel filtration on Sephadex G-100 and reverse-phase high-performance liquid chromatography (rpHPLC). At each step of purification, fractions were monitored by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and by immunoblotting with chum salmon GH. PRL and SL antisera. The yields of sea perch GH, PRL and SL were 4.2, 1.0 and 0.28 mg/g wet tissue, respectively. The molecular weights of 19,200 and 20,370 Da were estimated by SDS-PAGE for sea perch GH and PRL, respectively. Two forms of sea perch SL were found: one (28,400 Da) is probably glycosylated, while the other one (23,200 Da) is believed to be deglycosylated. GH bioactivity was examined by an in vivo assay. Intraperitoneal injection of sea perch GH at a dose of 0.01 and 0.1 mug/g body weight at 7-day intervals resulted in a significant increase in body weight and length of juvenile rainbow trout. The complete sea-perch GH amino acid sequence of 187 residues was determined by sequencing fragments cleaved by chemicals and enzymes. Alignment of sea-perch GH with those of other fish GHs revealed that sea-perch GH is most similar to advanced marine fish, such as tuna, gilthead sea bream, yellowfin porgy, red sea bream, bonito and yellow tail with 98.4, 96.2%, 95.7%, 95.2%, 94.1% and 91% sequence identity, respectively. Sea-perch GH has low identity to Atlantic cod (76.5%), hardtail (73.3%), flounder (68.4%), chum salmon (66.3%), carp (54%) and blue shark (38%). Partial amino-acid sequences of 127 of sea-perch PRL and the N-terminal of 16 amino-acid sequence of sea-perch SL have been determined. The data show that sea-perch PRL has a slightly higher sequence identity with tilapia PRL( 73.2%) than with chum salmon PRL(70%) in this 127 amino-acid sequence. (C) 2001 Elsevier Science B.V. All rights reserved.

Molecular identification of 16 Porphyra lines using sequence-related amplified polymorphism markers

Sequence-related amplified polymorphism (SRAP) is a novel molecular marker technique designed to amplify open reading frames (ORFs). The SRAP analytic system was set up and applied to Porphyra germplasm identification in this study for the first time. Sixteen Porphyra lines were screened by SRAP technique with 30 primer combinations. In the analysis, 14 primer combinations produced stable and reproducible amplification patterns in three repetitive experiments. Among the total 533 amplified fragments, 522 (98%) were polymorphic, with an average of 38 fragments for each primer combination, ranging in size from 50 to 500 bp. The 533 fragments were visually scored one by one and then used to develop a dendrogram with Unweighted Pair-Group Method Arithmetic Average (UPGMA), and the 16 Porphyra lines were divided into two major groups at the 0.68 similarity level. From the total 533 fragments, I I amplified by two primer combinations, ME1/EM1 and ME4/EM6, were used to develop the DNA fingerprints of the 16 Porphyra lines. The DNA fingerprints were then converted into binary codes, with I and 0 representing presence and absence of the corresponding amplified fragment, respectively. In the DNA fingerprints, each of the 16 Porphyra lines has its unique binary code and can be easily distinguished from the others. This is the first report on the development of SRAP technique and its utilization in germplasm identification of seaweeds. The results demonstrated that SRAP is a simple, stable, polymorphic and reproducible molecular marker technique for the classification and identification of Porphyra lines. (c) 2007 Elsevier B.V. All rights reserved.

Ulva and Enteromorpha (Ulvaceae, Chlorophyta) from two sides of the Yellow Sea: analysis of nuclear rDNA ITS and plastid rbcL sequence data

Ulvacean green seaweeds are common worldwide; they formed massive green tides in the Yellow Sea in recent years, which caused marine ecological problems as well as a social issue. We investigated two major genera of the Ulvaceae, Ulva and Enteromorpha, and collected the plastid rbcL and nuclear ITS sequences of specimens of the genera in two sides of the Yellow Sea and analyzed them. Phylogenetic trees of rbcL data show the occurrence of five species of Enteromorpha (E. compressa, E. flexuosa, E. intestinalis, E. linza and E. prolifera) and three species of Ulva (U. pertusa, U. rigida and U. ohnoi). However, we found U. ohnoi, which is known as a subtropical to tropical species, at two sites on Jeju Island, Korea. Four ribotypes in partial sequences of 5.8S rDNA and ITS2 from E. compressa were also found. Ribotype network analysis revealed that the common ribotype, occurring in China, Korea and Europe, is connected with ribotypes from Europe and China/Japan. Although samples of the same species were collected from both sides of the Yellow Sea, intraspecific genetic polymorphism of each species was low among samples collected worldwide.

Geophysical signatures associated with fluid flow and gas hydrate occurrence in a tectonically quiescent sequence, Qiongdongnan Basin, South China Sea

Interpretation of high-resolution two-dimensional (2D) and three-dimensional (3D) seismic data collected in the Qiongdongnan Basin, South China Sea reveals the presence of polygonal faults, pockmarks, gas chimneys and slope failure in strata of Pliocene and younger age. The gas chimneys are characterized by low-amplitude reflections, acoustic turbidity and low P-wave velocity indicating fluid expulsion pathways. Coherence time slices show that the polygonal faults are restricted to sediments with moderate-amplitude, continuous reflections. Gas hydrates are identified in seismic data by the presence of bottom simulating reflectors (BSRs), which have high amplitude, reverse polarity and are subparallel to seafloor. Mud diapirism and mounded structures have variable geometry and a great diversity regarding the origin of the fluid and the parent beds. The gas chimneys, mud diapirism, polygonal faults and a seismic facies-change facilitate the upward migration of thermogenic fluids from underlying sediments. Fluids can be temporarily trapped below the gas hydrate stability zone, but fluid advection may cause gas hydrate dissociation and affect the thickness of gas hydrate zone. The fluid accumulation leads to the generation of excess pore fluids that release along faults, forming pockmarks and mud volcanoes on the seafloor. These features are indicators of fluid flow in a tectonically-quiescent sequence, Qiongdongnan Basin. Geofluids (2010) 10, 351-368.

Phylogeography of the Qinghai-Tibetan Plateau endemic Juniperus przewalskii (Cupressaceae) inferred from chloroplast DNA sequence variation

The vegetation of the northeast Qinghai-Tibetan Plateau is dominated by alpine meadow and desert-steppe with sparse forests scattered within it. To obtain a better understanding of the phylogeography of one constituent species of the forests in this region, we examined chloroplast trnT-trnF and trnS-trnG sequence variation within Juniperus przewalskii, a key endemic tree species. Sequence data were obtained from 392 trees in 20 populations covering the entire distribution range of the species. Six cpDNA haplotypes were identified. Significant population subdivision was detected (G(ST) = 0.772, N-ST = 0.834), suggesting low levels of recurrent gene flow among populations and significant phylogeographic structure (N-ST > G(ST), P < 0.05). Eight of the nine disjunct populations surveyed on the high-elevation northeast plateau were fixed for a single haplotype (A), while the remaining, more westerly population, contained the same haplotype at high frequency together with two low frequency haplotypes (C and F). In contrast, most populations that occurred at lower altitudes at the plateau edge were fixed or nearly fixed for one of two haplotypes, A or E. However, two plateau edge populations had haplotype compositions different from the rest. In one, four haplotypes (A, B, D and E) were present at approximately equivalent frequencies, which might reflect a larger refugium in the area of this population during the last glacial period. Phylogenetic analysis indicated that the most widely distributed haplotype A is not ancestral to other haplotypes. The contrasting phylogeographic structures of the haplotype-rich plateau edge area and the almost haplotype-uniform plateau platform region indicate that the plateau platform was recolonized by J. przewalskii during the most recent postglacial period. This is supported by the findings of a nested clade analysis, which inferred that postglacial range expansion from the plateau edge followed by recent fragmentation is largely responsible for the present-day spatial distribution of cpDNA haplotypes within the species.

Molecular systematics of pikas (genus Ochotona) inferred from mitochondrial DNA sequences

The phylogenetic relationships among worldwide species of genus Ochotona were investigated by sequencing mitochondrial cytochrome b and ND4 genes. Parsimony and neighbor-joining analyses of the sequence data yielded congruent results that strongly indicated three major clusters: the shrub-steppe group, the northern group, and the mountain group. The subgeneric classification of Ochotona species needs to be revised because each of the two subgenera in the present classification contains species from the mountain group. To solve this taxonomic problem so that each taxon is monophyletic, i.e., represents a natural clade, Ochotona could be divided into three subgenera, one for the shrub-steppe species, a second for the northern species, and a third for the mountain species. The inferred tree suggests that the differentiation of this genus in the Palearctic Region was closely related to the gradual uplifting of the Tibet (Qinghai-Xizang) Plateau, as hypothesized previously, and that vicariance might have played a major role in the differentiation of this genus on the Plateau, On the other hand, the North American species, O. princeps, is most likely a dispersal event, which might have happened during the Pliocene through the opening of the Bering Strait. The phylogenetic relationships within the shrub-steppe group are worth noting in that instead of a monophyletic shrub-dwelling group, shrub dwellers and steppe dwellers are intermingled with each other. Moreover, the sequence divergence within the sister tars of one steppe? dweller and one shrub dweller is very low. These findings support the hypothesis that pikes have entered the steppe environment several times and that morphological similarities within steppe dwellers were due to convergent evolution. (C) 2000 Academic Press.

砂泥岩薄互层储层预测方法与应用研究

With the deeply development of exploration and development in petroleum in China, new increasing reserves are found in old oil fields and the verge of the old ones through re-study of geological property. It is more and more important to discovery and develop thin layer or thin inter-bedded layers reservoirs. All of the targets are thin sand-shale inter-bedded reservoirs and the core technology is reservoir predictions between wells in thin sand-shale inter-bedded layers. The continuity of the thin sand-shale inter-bedded layers in space or separating and heterogeneity is the key of reservoir geology research. The seismic reflection, high resolution analysis method and inversion method to thin sand-shale inter-bedded layers are thorough discussed and deeply studied in this paper to try to find the methods and resolutions of reservoir geology research. The below is followed. 1. Based on the pre-research of other people, five models are created: the sand sphenoid body, interlay sandstone and interlay shale of the equal thickness, interlay sandstone of the equal thickness and interlay shale of the unequal thickness, interlay sandstone of the unequal thickness and interlay shale of the unequal thickness, interlay sandstone of the changing thickness in sequence and interlay shale of the changing thickness in sequence. Then the study of the forward modeling are conducted on the thin layer and thin inter-bedded layers geological characters and seismic reflections including amplitude, frequency, phase, wave shape and time-frequency responding in the domains of time and frequency. The affect of petro-physics difference of layers, single thin layer thickness, thickness of inter-bedded, layer number of inter-bedded, incident wavelet domain frequency and types, sample interval to seismic reflection characters, frequency spectrum and time-frequency respond of reflectivity is theoretically discussed. 2. Qualitatively analyzing the sedimentary rhythm of the thin inter-bedded layers in vertical orientation and computing the single layer thickness or the average thickness with the method of generalized S transform. Identifying the reflecting interface or lithology interface using the amplitude value of amplitude spectrum domain frequency. 3. Based on the seismic respond of thin sand-shale inter-bedded layers, bring out the high resolution analysis method of seismic data in thin sand-shale inter-bedded layers using wavelet analysis and the idea of affecting low and high frequency with middle frequency. Then analyzing the effect to the method and testing some wavelets in the method. This method is applied to the theoretical models and the field data. 4. Bring forward one improved very fast simulated annealing method (IVFSA) to resolve the problem nonlinearity and multi-parameters of the inversion in thin inter-bedded layers. And IVFSA is more productive and higher precision than general ways. 5. New target constrained function is used in the inversion based on the property of the inversion in thin inter-bedded layers. 6. Making the full use of geological and logging information, IVFSA and the new function are applied in the non-linear inversion to improve reservoir prediction and evaluation in thin inter-bedded formations combined with the idea of logging and seismic inversion. This method was applied to the field data and got good results.

岩性油藏储层预测的地震理论与方法研究

At present the main object of the exploration and development (E&D) of oil and gas is not the structural oil-gas pools but the subtle lithological oil-gas reservoir. Since the last 90's, the ratio of this kind of pools in newly-added oil reserves is becoming larger and larger, so is the ratio in the eastern oilfields. The third oil-gas resource evaluation indicates the main exploration object of Jiyang depression is the lithological oil-gas pools in future. However, lack of effective methods that are applied to search for this kind of pool makes E&D difficult and the cost high. In view of the urgent demand of E&D, in this paper we deeply study and analyze the theory and application in which the seismic attributes are used to predict and describe lithological oil-gas reservoirs. The great results are obtained by making full use of abundant physics and reservoir information as well as the remarkable lateral continuity involved in seismic data in combination with well logging, drilling-well and geology. ①Based on a great deal of research and different geological features of Shengli oilfield, the great progresses are made some theories and methods of seismic reservoir prediction and description. Three kinds of extrapolation near well seismic wavelet methods-inverse distance interpolation, phase interpolation and pseudo well reflectivity-are improved; particularly, in sparse well area the method of getting pseudo well reflectivity is given by the application of the wavelet theory. The formulae for seismic attributes and coherent volumes are derived theoretically, and the optimal method of seismic attributes and improved algorithms of picking up coherent data volumes are put forward. The method of making sequence analysis on seismic data is put forward and derived in which the wavelet transform is used to analyze not only qualitatively but also quantitatively seismic characteristics of reservoirs.② According to geologic model and seismic forward simulation, from macro to micro, the method of pre- and post-stack data synthetic analysis and application is put forward using seismic in close combination with geology; particularly, based on making full use of post-stack seismic data, "green food"-pre-stack seismic data is as possible as utilized. ③ In this paper, the formative law and distributing characteristic of lithologic oil-gas pools of the Tertiary in Jiyang depression, the knowledge of geological geophysics and the feasibility of all sorts of seismic methods, and the applied knowledge of seismic data and the geophysical mechanism of oil-gas reservoirs are studied. Therefore a series of perfect seismic technique and software are completed that fit to E&D of different categories of lithologic oil-gas reservoirs. ④ This achievement is different from other new seismic methods that are put forward in the recent years, that is multi-wave multi-component seismic, cross hole seismic, vertical seismic, and time-lapse seismic etc. that need the reacquisition of seismic data to predict and describe the oil-gas reservoir. The method in this paper is based on the conventional 2D/3D seismic data, so the cost falls sharply. ⑤ In recent years this technique that predict and describe lithologic oil-gas reservoirs by seismic information has been applied in E&D of lithologic oil-gas reservoirs on glutenite fans in abrupt slop and turbidite fans in front of abrup slop, slump turbidite fans in front of delta, turbidite fans with channel in low slope and channel sanbody, and a encouraging geologic result has been gained. This achievement indicates that the application of seismic information is one of the most effective ways in solving the present problem of E&D. This technique is significant in the application and popularization, and positive on increasing reserves and raising production as well as stable development in Shengli oilfield. And it will be directive to E&D of some similar reservoirs

中国南方热带地区田洋玛珥湖24万年以来的气候变化及湖泊演变

A continuous long (224m) and high-resolution core TY2 was recovered from paleo-maar-lake Tianyang, tropical South China. Based on the diatom records of the upper 130-m core, this paper focuses on exploring climate change and the lake evolution history in tropical South China during the past 240ka. The most typical and unique characteristics of the diatom assemblages is that, Aulacoseira granulata was dominant or absolutely dominant species (80-90%) during most parts of the 130-m core, while Cyclotella stelligera var. tenuis and Fragilaria construens var. venter were subdominant species in only limited parts of the lower and upper core, respectively. Time scale is always the biggest problem for the study of TY2 core, so although diatom is seldom used for establishing time scale, here we attempt this by correlating the diatom-reconstructed temperature sequence with the time scale of ODP core 806B from Equatorial Western Pacific. Verified by the few most reliable ages from TY2 core and the parallel core TYl, a rather reasonable and reliable time scale was established. 01S 7/6 falls at the depth of 100m (ca. 194kaBP), OIS 6/5 at 75m (ca. 132kaBP), OIS 5/4 at 46m (ca. 75kaBP), OIS 4-3 at 35m (ca. 60kaBP). Qualitative and quantitative environmental reconstructions are made on the basis of diatom assemblage ecotype and EDDI dataset. Correlation of diatom-reconstructed temperature and moisture changes of Core TY2 with pollen-reconstructed temperature and rainfall sequence of Core TYl proves that the results are quite consistent in most periods. Thus the reconstruction results from diatom are quite reliable, and probably have a much higher resolution than pollen results. Combined with lithological and magnetic susceptibility variations, the diatom analysis reveals that, the general climate in tropical South China during the past 240ka was warm and wet. On the time scale of glacial-interglacial, warm and wet, cool and dry are not always synchronous. It was relatively warm-wet during the penultimate interglacial, cool-dry during the penultimate glacial, warm-dry during the last interglacial, and cooler-drier during the last glacial. In contrast, on the time scale of subglacial-subinterglacial scales, warm and dry, cool and wet corresponds very obviously, showing very clear 21-23 ka precession cycle. Analysis also shows that, the water of Tianyang paleo-maar-lake was generally warm, turbulent, turbid, meso-trophic, slightly alkaline, low conductivity and fresh during the past 240 kaBP, with small variations in some parts. Tianyang paleolake experienced shallow to semi-deep lake in OIS7d, open shallow lake in OIS7c-OIS5b, shallow coastal lake in OIS5a-OIS4c, swamp in OIS4b, and then completely dried up in OIS3c. The lake evolution was mainly controlled by temperature and precipitation changes in tropical China. While temperature and precipitation changes were probably controlled by the migrations of monsoon rainband and the evaporation rate, which was in turn controlled by the evolution of East Asian monsoon. Therefore, when the summer monsoon was strongest the climate was warm-dry, when stronger the climate was warm-wet; when the winter monsoon was strongest the climate was cool-dry, stronger cool-wet. This mechanism caused the warm-dry sub interglacial and cool-wet subglacial climate in the tropical South China.

黄骅坳陷高尘头地区储层非均质性及成藏研究

Gaochentou region is located in the southwest direction of Gaochentou village in Huanghua city of Hebei province. In regionally structural position, It lies in Qikou sag In the middle part of Huanghua depression, which belongs to the east part of the south Dagang structure zone in the middle part of Huanghua depression. Its' very beneficial at regional structure in Gaochentou , and It becomes the advantage area for oil and gas gathered and preserved, Sandstone reservoir of Dongying Formation is main bearing bed .Dongying Formation in Gaochentou region of Huanghua depression is consisted of set of mudstone and sandstone interbeds by deposited delta fades . Dongying Formation can be divided into 3 members from above to below: the first member of Dongying Formation (FMDF), the second member of Dongying Formation (SMDF), and third member of Dongying Formation (TMDF). The lithology of the upper part of FMDF was consisted of mostly middle-grained and fine-grained sandstone, and it is small for the oil-bearing area of the sand bodies .The lithology of the lower part is coarse-grained sandstone bodies which are well connected between sandstone bodies of wells, and the lower part was main bed of oil production in Dongying Formation; SMDF and TMDF are consisted of larger scale set of mudstone, in which the sandbodies are lenticular and pinch out quickly, and the lithology was mostly fine sandstone and silt stone, in which there are little oil and gas .Because the reservoirs in this area are largely influenced by the factors such as lithology, fault and others, and the reservoirs have the strong,heterogeneity , there exists the problem of oil-down and water-up for vertical distribution of oil and gas bearing. It is not very clearly for the three dimension distribution of sandstone , and the geology researchs is not enough. So, it can't satisfy the need of further development and production for Gaochentou oilfield.Having the key problem of oil-down and water-up and the mechanism of the reservoir for Gaochentou area, There are as follow study works, the first, is study of the high-resolution correlation of sequence stratigraphy and sedimentary microfacies. Dongying Formation was divided into three parasequence sets and each parasequence set was divided into different amount of parasequences. FMDF, as the main oil and gas producing bed, can be divided into seven parasequences. Oil and gas are discovered in six parasequences except the seventh. On the basis of study of sedimentary microfacies, the sediments of Dongying Formation are considered deposited mainly in delta front subfacies. The microfacies types of Dongying Formation are sub-water distirbutary channel, sub-water natural bank, inter distributary channel bay, distributary channel mouth dam, and delta front mat sand.Seismic facies analysis and logging-constrained inversion technique were applied by Author for transverse prediction of sandstone reservoir. Having 4 modes of interwell single sandbodies correlation technique, Author have described distribution characteristics of sandbodies, and established geological reservoir model of Gaochentou reservoir.Author presented that the reservoirs characteristic have very strong heterogeneity ,and In the section of sandstone interlayed with mudstone,the folium sandstone interlayed with each other, and the wedge shaped sandbodies pinched out in the mudstone. So the pinch-out up sandstone trap and lenticular sandstone trap are easily formed. They are most small scale overlying pinches out in the place of slope. This article applies the concept of deep basin oil to resolve reasonably the problem of which the oil is below the water in Gaochentou area. Combined with the study of sedimentary facies, reservoir and other aspects, the mechanism and patterns of deep basin oil are studied on the basis of characteristics in Gaochentou area.On the basis of the above study, the mechanism of the oil and gas' migration and accumulation in isotropic sandstone and heterogeneous sandstone are thoroughly analyzed through experiments on physical modeling. Experiments on physical modeling show that the discrepancy between sand layers with different permeability and thickness has important influence on the direction, path, and injection layer of oil's migration. At the beginning of the injection of oil and gas in high permeability sand layer, the pressure is low, the migration resistance is small, and the oil and gas are more easily displacing the water in sand. So it can act as good transformation layer or reservoir. But at the beginning of the injection of oil and gas in sand layer with low permeability, the pressure is high, the migration resistance is big, and the oil and gas are more difficultly displacing the water in sand. So it can only act as bad or worse transformation layer or reservoir. Even if it cannot act as transformation layer or reservoir, it can act as water layer or dry layer. The discrepancy between sand layers on permeability and thickness can make discrepancy in injection of oil and gas between different layers. Consequently it leads to small amount of oil and gas injection in sand layers with low permeability. Ultimately it affects the oil's accumulation and distribution in different sand layers.At Last, combining analysis of the structure and pool forming condition, The thesis has established models of reservoir formation to predict the advantage distribution of oil and gas bearing , and put forward the prospective target It is not only of theoretical signification for explosion and importance, but also has realistic value in guiding the progressive petroleum exploration and exploitation.

陆相砂泥岩层序的高精度检测方法研究及应用

Ordos Basin is a typical cratonic petroliferous basin with 40 oil-gas bearing bed sets. It is featured as stable multicycle sedimentation, gentle formation, and less structures. The reservoir beds in Upper Paleozoic and Mesozoicare are mainly low density, low permeability, strong lateral change, and strong vertical heterogeneous. The well-known Loess Plateau in the southern area and Maowusu Desert, Kubuqi Desert and Ordos Grasslands in the northern area cover the basin, so seismic data acquisition in this area is very difficult and the data often takes on inadequate precision, strong interference, low signal-noise ratio, and low resolution. Because of the complicated condition of the surface and the underground, it is very difficult to distinguish the thin beds and study the land facies high-resolution lithologic sequence stratigraphy according to routine seismic profile. Therefore, a method, which have clearly physical significance, based on advanced mathematical physics theory and algorithmic and can improve the precision of the detection on the thin sand-peat interbed configurations of land facies, is in demand to put forward.Generalized S Transform (GST) processing method provides a new method of phase space analysis for seismic data. Compared with wavelet transform, both of them have very good localization characteristics; however, directly related to the Fourier spectra, GST has clearer physical significance, moreover, GST adopts a technology to best approach seismic wavelets and transforms the seismic data into time-scale domain, and breaks through the limit of the fixed wavelet in S transform, so GST has extensive adaptability. Based on tracing the development of the ideas and theories from wavelet transform, S transform to GST, we studied how to improve the precision of the detection on the thin stratum by GST.Noise has strong influence on sequence detecting in GST, especially in the low signal-noise ratio data. We studied the distribution rule of colored noise in GST domain, and proposed a technology to distinguish the signal and noise in GST domain. We discussed two types of noises: white noise and red noise, in which noise satisfy statistical autoregression model. For these two model, the noise-signal detection technology based on GST all get good result. It proved that the GST domain noise-signal detection technology could be used to real seismic data, and could effectively avoid noise influence on seismic sequence detecting.On the seismic profile after GST processing, high amplitude energy intensive zone, schollen, strip and lentoid dead zone and disarray zone maybe represent specifically geologic meanings according to given geologic background. Using seismic sequence detection profile and combining other seismic interpretation technologies, we can elaborate depict the shape of palaeo-geomorphology, effectively estimate sand stretch, distinguish sedimentary facies, determine target area, and directly guide oil-gas exploration.In the lateral reservoir prediction in XF oilfield of Ordos Basin, it played very important role in the estimation of sand stretch that the study of palaeo-geomorphology of Triassic System and the partition of inner sequence of the stratum group. According to the high-resolution seismic profile after GST processing, we pointed out that the C8 Member of Yanchang Formation in DZ area and C8 Member in BM area are the same deposit. It provided the foundation for getting 430 million tons predicting reserves and unite building 3 million tons off-take potential.In tackling key problem study for SLG gas-field, according to the high-resolution seismic sequence profile, we determined that the deposit direction of H8 member is approximately N-S or NNE-SS W. Using the seismic sequence profile, combining with layer-level profile, we can interpret the shape of entrenched stream. The sunken lenticle indicates the high-energy stream channel, which has stronger hydropower. By this way we drew out three high-energy stream channels' outline, and determined the target areas for exploitation. Finding high-energy braided river by high-resolution sequence processing is the key technology in SLG area.In ZZ area, we studied the distribution of the main reservoir bed-S23, which is shallow delta thin sand bed, by GST processing. From the seismic sequence profile, we discovered that the schollen thick sand beds are only local distributed, and most of them are distributary channel sand and distributary bar deposit. Then we determined that the S23 sand deposit direction is NW-SE in west, N-S in central and NE-SW in east. The high detecting seismic sequence interpretation profiles have been tested by 14 wells, 2 wells mismatch and the coincidence rate is 85.7%. Based on the profiles we suggested 3 predicted wells, one well (Yu54) completed and the other two is still drilling. The completed on Is coincident with the forecastThe paper testified that GST is a effective technology to get high- resolution seismic sequence profile, compartmentalize deposit microfacies, confirm strike direction of sandstone and make sure of the distribution range of oil-gas bearing sandstone, and is the gordian technique for the exploration of lithologic gas-oil pool in complicated areas.

黄骅坳陷大港探区油气资源潜力与资源结构预测

Facing the problems that Dagang region of Huanghua Depression has high exploration degree and its remaining resource potential and structure are not clear, the theory of Petroleum Accumulation System (PAS) is applied to divide and evaluate the oil/gas systems quantitatively. Then, the petroleum accumulation systems are taken as units to forecast and analyse the oil/gas resources and their structure using statistical methods of sampling analysis of discovery process model and generalized pareto distribution model. The exploration benefit of the unit is estimated using exploration simulation methods. On the basis of the above study, the resource potential of Huanghua Depression is discussed.Huanghua Depression can be diveded into four petroleum accumulation systems, i.e. North PAS5 Middle Qibei PAS, Middle Qinan PAS and South PAS. Each PAS can be diveded futher into several sub- PASs. Using the basic princple of Analytical Hierarchy Process, the method of quantitative evaluation of PAS is established. Then the elements and maturity of PAS are evaluated quantitatively.Taking migration and accumulation units and sub-PASs as prediction units, sampling analysis of discovery process model and generalized pareto distribution model are applied comparatively to forecast the resource structure of eight migration and accumulation units in six PASs of medium-high exploration degree. The results of these two methods are contrasted and analyzed. An examination of X2 data of these two models from exploration samples shows that generalized pareto distribution model is more effective than sampling analysis of discovery process model in Huanghua Depression. It is concluded that minimum and maximum size of reservoir and discovery sequence of reservoirs are the sensitive parameters of these two methods.Aiming at the difficult problem of forecast in low exploration degree, by analysis of relativity between resource parameters and their possible influential geological factors, forecast models for resource parameters were established by liner regressing. Then the resource structure is forecasted in PASs of low exploration degree.Based on the forecast results, beginning with the analysis of exploration history and benefit variation, the exploration benefit variation of the above PASs is fitted effectively using exploration simulation method. The single well exploration benefit of remaining oil resource is also forecasted reasonably.The results of resource forecast show that the total oil resources ofHuanghua Depression amount to 2.28 b illion ton. By the end o f 2 003, the accumulative total proved oil reserve is 0.90 billion ton and the remaining oil resources is 1.38 billion ton. The remaining oil resource is concentrated in Kongdian-Dengmingshi, Banqiao-Beidagang, Qidong-Yangerzhuang and Baidong-Qizhong sub-PASs.

Generalized Low-Rank Approximations

We study the frequent problem of approximating a target matrix with a matrix of lower rank. We provide a simple and efficient (EM) algorithm for solving {\\em weighted} low rank approximation problems, which, unlike simple matrix factorization problems, do not admit a closed form solution in general. We analyze, in addition, the nature of locally optimal solutions that arise in this context, demonstrate the utility of accommodating the weights in reconstructing the underlying low rank representation, and extend the formulation to non-Gaussian noise models such as classification (collaborative filtering).

Learning World Models in Environments with Manifest Causal Structure

This thesis examines the problem of an autonomous agent learning a causal world model of its environment. Previous approaches to learning causal world models have concentrated on environments that are too "easy" (deterministic finite state machines) or too "hard" (containing much hidden state). We describe a new domain --- environments with manifest causal structure --- for learning. In such environments the agent has an abundance of perceptions of its environment. Specifically, it perceives almost all the relevant information it needs to understand the environment. Many environments of interest have manifest causal structure and we show that an agent can learn the manifest aspects of these environments quickly using straightforward learning techniques. We present a new algorithm to learn a rule-based causal world model from observations in the environment. The learning algorithm includes (1) a low level rule-learning algorithm that converges on a good set of specific rules, (2) a concept learning algorithm that learns concepts by finding completely correlated perceptions, and (3) an algorithm that learns general rules. In addition this thesis examines the problem of finding a good expert from a sequence of experts. Each expert has an "error rate"; we wish to find an expert with a low error rate. However, each expert's error rate and the distribution of error rates are unknown. A new expert-finding algorithm is presented and an upper bound on the expected error rate of the expert is derived.