Research and application of the method based on the displacement gradient method to implement the fragmented breakpoint at the end of the fault

: The limitation of seismic resolution makes the interpretation of fault end locations based on seismic data unclear, and often can only be predicted qualitatively by human, which makes the combination relationship of breakpoints near or below seismic resolution unclear. The breakpoint combination rate is generally low, which affects the development injection and recovery relationship and remaining oil distribution.


Introduction
The study of fault ends is actually a kind of subearthquake study, and the so-called sub-earthquake faults are small displacement faults below seismic resolution, including isolated small faults and large fault ends, which control the subsurface fluid flow and affect the nonhomogeneity of subsurface reservoirs in oil and gas fields as well as the effect of water injection and development [1][2], therefore, this study can not only improve the combination rate of block breaks and further clarify the tectonic features, but also Therefore, this study can not only improve the block breakpoint combination rate, further clarify the tectonic characteristics, but also improve the oil and gas field injection and extraction relationship, which is important to improve the development effect and optimize the well network deployment.
The study area is located in the northern part of Daqing Changyuan Sanan oil field, which is one of the most complex blocks on Daqing Changyuan.The fault combination rate of small faults in the study area is low, the overall fault combination rate of the block is 90.3%, but the fault combination rate of small faults below 3 meters is only 76.4%, which obviously lowers the overall accuracy of the block structure and affects the analysis of the development and injection relationship.How to improve the rate of fault combination is an important issue.Due to the limitation of seismic data resolution, fault ends are often predicted qualitatively by human, which makes most of the small fault points below 3 m uncertain in origin, and often treat them as scattered fault points within the neighboring fracture zone, and the cause is unclear.This paper quantitatively predicts the location of fault ends based on the displacement gradient method, and further attributes the fault end breakpoints correctly.The practical results show that the method can quantitatively predict the extension length of sub-seismic fault ends and effectively improve the combination rate of small breakpoints at fault ends.

Overview of the study area
The study area is located in the northern part of the Sanan oil field in Changquan, Daqing, with a development area of 41Km 2 and a stratigraphic dip angle of 5-23°.The study area is at the high point of the Changquan tectonics in Daqing, with complex fracture relationships (Figure 1).The 79 faults are developed on the top surface of the Sa II oil group, and the fault orientation is mainly north-east and south-west, with frequent mutual cutting.There are various combinations of fault planes and profiles, such as parallel, geese column and oblique, etc., and there are combinations of graben, basement, stacked tile and "Y" shape in the profile.The overall fault combination rate in the study area is 90.3%, of which 76.4% is below 3 meters, and there are many uncombined and scattered faults.It is important to understand the structural situation of this block, which is important for well network optimization, measures to explore potential and remaining oil analysis.

2 Based on the displacement gradient method to implement the end of the fault fragmentation breakpoint
The so-called displacement gradient method in this paper draws on numerical analysis methods [3] and the relative relationship between statistical fault extension length and fault distance size to establish a fault displacement gradient to estimate the length of the fault tail by plotting the relevant fault distance-distance curve, so as to quantitatively predict the length of the fault tail and its corresponding fault distance size (Figure 2).Seismic fault end point is the point where the fault displacement is lower than the seismic resolution, and the fault displacement at this point can be considered as the seismic resolution, so the displacement gradient can be obtained from the fault displacement map, and practice shows that for the prediction of sub-seismic faults at the end of faults, the displacement gradient method can be used for quantitative prediction based on seismic data.Using the seismic interpretation and logging data in the study area, the fault displacement gradient graph is established by counting the fault break distance, and then quantitatively predicting the extension length and break distance at the end of the fault based on the curve fitting formula, so that the suspected fault end break points can be combined.
The comparison method of "flat profile" is used to predict the tail of the fault quantitatively from the plane and profile respectively.In the plane, as shown in Figures 3 and 4, the location of the original interpreted end of fault No. 1 is shown as the red line, and the new extension location is predicted by making a displacement gradient graph of the fault and calculating the formula quantitatively based on the curve fitting formula as shown in the dashed line, but it is known that the fault is truncated by fault No. 2 based on the fault polygon, therefore, whether this prediction is reasonable, we further verified using logging data and found that N-A-2, the well, has a 4-meter uncombined scattered fault within the Sa II oil group, while a fault with a 3.1-meter break distance is predicted to exist at this location according to the displacement gradient map.It can be seen that the situation predicted by the displacement gradient method just happens to be very similar to the actual interpretation of this well N-A-2.The comprehensive comparison and verification of the above well seismic data leads to the conclusion that No. 1, after being truncated by No. 2 fault, does not end abruptly, but continues to develop for some distance through No. 2 fault, and also attributes the uncombined breakpoint with a break distance of 4 m in well N-A-2 to No. 1 fault.On the profile, as shown in Figures 5 and 26, the end of fault No. 4 is predicted to have developed a breakpoint with a break distance of 2 meters at a location with an extension length of 325 meters, as predicted by the quantitative prediction map of the fault end.By searching the log data, it was found that an uncombined scattered breakpoint with a break distance of 1.8 m was developed in the N-D-2 well within the Portuguese I formation, which further supports the accuracy of the prediction results, and therefore, the breakpoint was attributed to the end of the No. 4 fault.In the same way, by building the fault displacement gradient graphs one by one and quantitatively predicting the subseismic end extension length and break distance based on the curve fitting formula, the suspected fault end break points were compared with the results of this quantitative prediction, and a total of 16 suspected fault end break points were identified, of which the number of implemented combinations was 10 (Table 1), 7 combinations with break distance less than 3 m, and the other 6 were confirmed not to be fault end break points.

Conclusion
(1) The seismic resolution controls the interpretation of the end of the fault, because the fault development is gradually pointed out, so when the fault end break distance is smaller than the seismic resolution, it is not explained by seismic data, and the artificial qualitative judgment of the fault location is actually the location where the seismic resolution can be distinguished, and the location is usually recognized as the end of the fault, while the actual fault end extension should be longer.
(2) The practical application of the displacement gradient method proves that the method can quantitatively predict the extension length of the end of the fault and its corresponding fault distance size, so as to determine the location of the end of the fault, and the method can break through the traditional practice of qualitatively predicting the end of the fault due to the constraints of the resolution of seismic data.
(3) The well interpretation breakpoint data can verify the reliability of the interpretation results of the fault end prediction by using this method, and the relative error is generally less than 30% by comparing the actual breakpoint distance and the predicted breakpoint distance at the breakpoint, which has a high degree of confidence.
(4) By this method, the attribution of fault end breakpoints was further implemented in the study area, and 10 new breakpoints were combined, including 7 with a break distance of less than 3 meters, which improved the combination rate of breakpoints in the study area.

Figure 1
Figure 1 Polygon distribution of faults on the top surface of Sa Ⅱ in the study area

Figure 2
Figure 2 Schematic diagram of the method of calculating the tail length of a fault (Rotevatn and Fossen, 2011)

Figure 3 .
Figure 3. Fault polygons on the top surface of SaⅡ

Figure 5 Figure 6
Figure 5 Seismic profile of the over N-D-2 well

Table 1
Comparison of predicted results of fault ends and actual information of fault points