SPE Advanced Technology Series

Logging technology for determining production profiles through casing/tubing, the small diameter series of designing tools, and the advanced logging techniques is described. Applications of the logging technology are demonstrated using typical examples.

The choice of water alternate carbon dioxide flooding in Sanan was decided based on prefeasibility studies on EOR methods suitable for Putaohua reservoir in Sanan area of Daqing Oil Field. This process was first implemented on one pilot, before any industrial extension. This paper introduces the pilot test, evaluates the effects of CO2 drive and analyzes behavioral characteristics of the test. Introduction

An enhanced oil recovery (EOR) study by CO2 immiscible drive began in 1985 for the Sanan area of Daqing Oil Field. Chinese and French engineers worked on prefeasibility study successively, and determined a practicable project of CO2 drive.

Preflush water injection began in Oct. 1990, and water alternate gas injection started in July 1991 and ended in March 1993. During the test, production and injection wells were logged using "PLT," two time water-soluble tracer injection tests and one time gas tracer injection test were conducted. In addition, well temperature log, pressure measurement by high precision pressure gauge and sample analyses of oil, gas and water were made. In the course of evaluating the results, follow-up matching and result prediction of the CO2 drive were performed by numerical simulation.

General Conditions Of The Pilot Test Area

The pilot site is located in the S3-3045 well block within the first strip of the eastern transition zone in Sanan area, consisting of four inverted five-spot area patterns. C1-C4 are injection wells and C5-C13 are production wells. The spacing between injection well and production well is 106m, and area of the pilot is 90000m2. There are 6 equalizing wells beyond the test area, of which three are production wells, the others injection wells (Fig. 1).

Reservoir Characteristics. The objective layer PI2 is a river channel sand body deposited in fluvial delta environment. The PI2 is characteristic of both uniformity and combined rhythm. The horizontal distribution of the layer is steady, and vertical variation coefficient of heterogeneity is 0.61. The average parameters of the reservoir are as follows: sandstone thickness, 11.6m; effective thickness, 9.2m; air-permeability, 1628md; porosity, 0.276; reservoir temperature, 49 C; viscosity of oil at reservoir conditions, 9.8mPa s; mid-depth of the reservoir, 1139.5m.

Interpretation of data from interference test shows that there is a channel with high permeability between wells C1 and C2. The result of tracer detection also confirms the existence of the channel, and finds that communications between wells C9, C10 and C3, and between wells C4 and C5 are better. This suggests that the reservoir is seriously heterogeneous horizontally.

Development History. The Sanan area was put into production by waterflooding using four-spot area pattern with well spacing of 300m. S and P reservoirs were commingled for production.

Core data from inspection well in 1984 showed that flooded thickness of PI2 was up to 80-90%.

This paper presents a new method for modelling and operation optimization of large scale water injection systems. Systems of this kind usually consist of a great number of water injection stations, pipelines, valves, pipe fittings, and water injection wells, and is a multi-source multi-sink nonlinear time-varying system with thousands of variables. A properly simplified and sufficiently accurate mathematical model has been made by both the analytical method and the system identification method. A new operation optimization method has been developed for this system by the large scale system theory and the dynamic programming method. To solve this optimization problem, the concept of pressure valley in hydrodynamic system is presented for the first time in this paper. The water injection system is divided into a series of subsystems according to pressure valleys. It has been proved that system optimization on the whole is equivalent to optimization of every subsystem in it if coupling variables of the subsystems have the values resulting from the optimized system. Thus, the system optimization problem is decomposed into a series of subsystem optimization problems.

The above modeling method and operation optimization procedures are programmed in a software package. This package has been applied to an oilfield in Daqing. Electric power consumption for water injection decreased by 3-10%, and waterflood efficiency improved.

The production profile can be obtained with a continuous flow meter and a gradiomanometer. Slip and other factors in the vertical string for two phase (oil-water) flow, affect the data and make it very difficult to interpret. To solve this problem, studies were made in a simulation well of the relationship between the rotary speed of the spinner and the water holdup, and the ratio of the apparent velocity to the average fluid velocity and water holdup. Four relationship charts and interpretation programs were developed. A total model of the simulating experiment was built, which can interpret multiple zone oil/water production rates to draw their profile map, and quantitatively interpret oil bubbles moving upward, and interpret the reverse rotation of the spinner in the vertical string. The data obtained from 20 wells show that the method has high resolution so that low producing layers can be detected and quantitatively interpreted. Also, intra-formational interpretations for thick formations can be made. In conjunction with the water intake profile of the surrounding injection well, a plot of water-injection versus oil-production can also be drawn. Introduction

Daqing Oilfield has entered into the stage of high water cut. To maintain high and stable production rate, the reservoir producing situation must be known. A satisfactory fluid production profile can be obtained with a continuous flow meter and a gradiomanometer under the conditions of the two phase (oil-water) flow. Compared with inflatable and station measurements, continuous measurement has many advantages. The upper limit for flow rate is over 1000 m3/d. Logging speed is high and repeat runs can be made. The log data are reliable and can reflect the intra-formational changes of a thick formation. However, there is some slip between oil and water as the two-phase flow moves vertically, so changes in the flow behavior are very complicated and the data interpretation is very difficult. Thus, a study of interpretation methods of two-phase (oil-water) flow was conducted.

Test procedure and test result for oil water two phase flow

A test for two-phase (oil-water) flow was made in a simulation well. The test media are diesel fuel and water. The tools used in experiment are a spinner flowmeter and a gradiomanometer. Water holdup, Yw, is derived from an average density measured with the gradiomanometer and spinner count, CPS, that is obtained from a continuous flowmeter. Since Daqing oil field has entered into the stage of high water cut, the water holdup in the borehole is generally greater than 50%. Thus, a holdup of more than 40% was selected for the study. Experiments were conducted in two modes, station measurement and continuous measurement. For different water cuts, different flowrates were selected. 27,000 data points were obtained and four charts were generated in this experiment. To determine water cut, Kw, Figure 1 shows the relationship between the spinner count and different water cuts.

Figure 2 shows that the relationship between the ratio of the apparent velocity to the average fluid velocity and the water holdup. The velocity measured in the borehole is an apparent velocity, Va, that is the response to the flowrate of the fluid which is covered with spinner. With a measured water holdup, Yw, and apparent velocity, Va, average flowrate Ut1 can be found from Figure 2.

Figure 3 shows the relationship of the spinner count (CPS) and the average flow rate, Ut2, under the condition of the continuous measurement. The curved form and changing trend of Figure 3 is generally consistent with similar data obtained with station measurements. The average flowrate obtained from the stationary measurements is Ut3. The spinner response is a negative value when water cut is lower than 40% and flowrate is about 30m3/d. An average fluid velocity Ut can be derived by averaging over Ut1, Ut2, and Ut3.

Mathematical model and computerized interpretation program The figures mentioned above were generated with a computer using an empirical formula to fit the experimental data. The computer plotter can automatically plot the apparent velocity regression curve of every station and list the multiple or separate-zone oil and water production profiles. Based on the empirical formula, a mathematical model was build to simplify the multiple element regression as a single-element regression combining various related variables and using them to form new variables.

1. Modeling

g(Q,Yw,CPS)=0

the relationships among Q,Yw,Kw,CPS are as follows:

This technique including explosive reforming and explosive welding is mainly adopted for repairing downhole damaged casing in an oilfield. It is based on the principle of explosive action to restore a deformed casing and to weld the damaged section.

This paper describes development and applications of separate layer production technology, which facilitates full exploitation of the potential of each oil-bearing layer. The technology has evolved as water cut has increased in Daqing oilfield and focus has shifted to thinner, lower productivity layers. In Daqing field, 11 kinds of separate layer water injection strings, which can be divided into two main categories (hydraulic inflatable and compressional), 33 kinds of separate layer production (water shut off) strings divided into four main categories (including integral, slip support releasing, drill-able and pressure balance), four separate layer fracturing techniques (including sleeve separate layer, limited entry completion, multi-frac and positioning balance) have been developed.

From Darcy's law of two phases (oil and water), an analytic expression for oil and water vertical cross flow including gravity cross flow, capillary cross flow and additional cross flow under the effect of cyclic waterflooding is developed. Based on function J, this paper analyzes the vertical cross flow characteristics of capillary and its influence on production, explains the essential distinction between cyclic waterflooding and conventional waterflooding. Introduction

At the end of 1950's, a Former Soviet Union expert, M.R. Surkuchef, advanced the concept of cyclic waterflooding for the first time. He thought that production process could be improved by changing the type of water-injection and establishing non-steady state artificially in the reservoir. Based on this theory, the Former Soviet Union has conducted cyclic waterflooding in Pokorove oilfield and some other oil fields for field test and commercial production since 1964. All obtained better production results. The field test showed that cyclic waterflooding can improve recovery rate by 3 -10% compared with conventional waterflooding.

Since 1965, the Former Soviet Union focused on cyclic waterflooding theory research. Among those theories the most influential one was the mathematical model set up by O.I. Chenkowa et al. of the Former Soviet petroleum Scientific Institute in 1970's. The problem was that when cyclic waterflooding was conducted, this model could only predict the degree and influential factor of its improvement, by estimating the flow rate through porous medium between high and low permeability layers in vertical heterogeneous reservoir on the base of water stagnant coefficient concept; but could not explain the mechanism of cyclic waterflooding.

Through laboratory experiments, W.G. Ogengianiyanches et al. believes that the physical nature of improving production by cyclic waterflooding is that when injection is halted, capillary cross flow becomes important so more crude oil is displaced from layers with low permeability and the water saturation of high permeability watered-out layers is reduced. Thus, the conclusion that cylic waterflooding is more effective in water-wet reservoirs.

Field test and laboratory numerical simulation data show that cyclic waterflooding in oil-wet reservoirs can also be effective. This paper discusses the mechanism of cyclic waterflooding on the basis of seepage flow dynamics.

The Way of Oil & Water Vertical Cross Flow Under Cyclic Waterflooding

Set up a coordinate system, with the vertical downward direction as Z-axis. Suppose there are only two phases (oil and water) existing in the reservoir. Based on Darcy's Law, oil and water vertical movement equations can be written as:

(1)

(2)

Analyses of seepage flow and numerical simulation show that during the half cycle when injection is halted or reduced, another pressure differential between the high- and low-permeability layers besides capillary pressure and gravity potential is created. This kind of pressure differential inherent in unconventional water injection is called as additional pressure differential. In contrast, during the half cycle of reinjection or greater injection, the pressure buildup is quick in high permeability layers and slow in low permeability layers, creating a reverse pressure differential (Fig. 1). Then

(3)

Table (1) (2) and (3) as simultaneous equations and let

Then we have:

(4)

The paper summarizes the methods and effects of cyclic waterflooding in the southern oilfields of the Daqing Placanticline, discusses the feasibility and necessity of this process in lower permeability reservoirs, and analyzes systematically the main factors which impact the results of cyclic flooding. Production performance is compared and contrasted for fields developed by both conventional and cyclic waterflooding. Based on field case histories, the impact of cyclic waterflooding on performance at different water cuts is evaluated. It is demonstrated that cyclic flooding should be implemented at low water cuts, but is beneficial at all development stages. Cyclic waterflooding is an effective measure to enlarge sweep volume and enhance conformance efficiency and oil recovery. The practices described can provide guidance for cyclic waterflooding in similar oilfields.

Since 1980, the oil production of Daqing oilfield has been at high water-cut stage. The rate of oil production has naturally decreased. Research was undertaken to maintain and enhance fracturing effects on oil production, including strengthening the preparation for fracturing and management after fracturing. Properly selecting the wells and zones for fracturing is important. Pre-treatment of injection wells and restoration of reservoir pressure can improve the results of fracture treatments in producing wells. Fracturing techniques were also developed for multiple thin intervals. This paper uses examples from the Lamadian Oilfield of Daqing.

Based on lab data this paper summarizes differences of petrophysical parameters of sandstones in Daqing Oil Field which were not watered out before waterflooding and were watered out after waterflooding. Pore throat dimensions, relative permeability, and wettability change with prolonged waterflooding. Using internal waterflooding at an early stage is a method of maintaining the reservoir energy and achieving long-term high and stable production. Injecting a great quantity of water for a long time is an effective means of enhancing oil recovery of reservoirs for Daqing Sandstone Oil Field.