Because of low volumetric sweep efficiency and large difference in injection and production rate of each zone, the recovery factor of water flooding is very low in some continental and heterogeneous sandstone oil fields, especially when the reservoir continuity is poor, permeability low and high permeability variation between different zones. Due to the amount of incremental producible oil is not large enough to further drill infill wells economically, drilling infill wells and performing polymer flooding simultaneously was proposed and a pilot test has been concluded. The paper introduces the laboratory studies on selection of polymer injection parameters and project design optimization. According to these studies, a pilot test of "Combining Small Well Spacing with Polymer Flooding" was conducted. The well spacing is 100 meters, the zones with a thickness of less than 1 meter and with a permeability rage of 5 to 100 md were combined together and put on development. Both lab studies and pilot results show that the volumetric sweep efficiency was greatly increased, especially after applying separate zone injection measures in polymer injectors based on injection profile surveys and tracer test data. The pilot test indicates that the technique of combining infill wells with polymer flooding is economically feasible with 10% OOIP incremental recovery at a production cost of $10/bbl. INTRODUCTION In continental, multilayer and heterogeneous sandstone oilfields, some reservoirs with poor connectivity and low permeability have lower recovery factors. The low degree of connection of layers between wells and inner-layer interferences lead to high watercut for waterflooding and poor effenciency for increasing the recoverable reserves of single wells. In order to further enhance the oil recovery and increase the oilfield's recoverable reserves, we drilled infill wellpatterns and performed polymerflooding simultaneously to improve the combined economic benefits of marginal reserovir. The main idea of the technology is to adjust the reservoir connectivity by drilling infill wellpatterns, and then inject a polymer solution. On the basis of waterflooding, polymerflooding could further increase some amount of recoverable reserves. The technology of combining dense wellpattern with polymerflooding will not only improve oil recovery but also obtain better economic benefits. Therefore, we implemented the pilot test in the oilfield and also applied separate-layer injection and stimulation measures during the test, which achieved good results with average incremental oil per day for a single well 1.83 times that of the original production, watercut decreased 10.2% and oil recovery increased more than 10%.

Being a maturing field that has been developed for over 40 years, Daqing Oilfield has nearly 40 years of water flooding development history. One of the major problems that Daqing Oilfield is currently faced with is severe casing damage of both producers and injectors. Factors influencing casing damage are numerous, heightening difficulties in prediction and comprehensive analysis. In the past, casing damage mechanism research emphasizes on casing strength in respect of well drilling, fault and shale in respect of geology, and water injection pressure in respect of reservoir. Yet the accuracy for trouble shooting based on single factor analysis is quite low. To conduct casing damage mechanism research, the influence of multi-factors on casing damage should be considered comprehensively to draw a correct conclusion. Therefore, research on multi-factor evaluation technology and its application in casing damaged wells is implemented. Three-layer feed forward neural network model is used to make the casing damage formation factors corresponding to the input vectors of the neural network. Factors causing casing damage such as water invasion in mudstone, pressure performance, geological features, casing performance, and stress field etc. are described as a series of real numbers between 0 and 1. More often than not, these factors are difficult to be described through mathematical methods, but a blurry value can be given by experts. Thus, a blurry vector for casing damage factor is composed of multi-factors, that is to say, the output of neural network is correspondent to the present working status of casings, which can be given by experts based on their experiences, and the output value is a description of the present working status of the casings. The given casing damage data is used as instructor signal to train the network, and following training, the network can perform the prediction of the working status of casings. Introduction Casing damage problem in Daqing Oilfield has become increasingly severe, resulting in immense economic loss1. The prediction research is an important approach to deal with casing damage. However, the analysis and prediction of casing damage is very difficult due to the numerous factors influencing casing damage. Some of the former prediction methods are mainly based on mechanics2, but due to complex geological conditions, using mathematical method alone is not sufficient to describe complex physical model of causes of casing damage. Generally, there are certain degrees of errors in building mathematical model and selecting boundary condition, so inevitably, software based on the mathematical model has many weaknesses such as system complexity or poor usability. Considering the problems mentioned above, an improved BP neural network (NN) is used to evaluate the casing damage to improve operability and practibility of the prediction software, thus contributing to prediction accuracy of over 70%. The precise prediction can provide guidance in taking efficient measures to avert casing damage, so the potential economic benefit is huge. Principle and improvement of BP neural network BP neural network is a three-layer feed forward neural network model3, whose basis idea is to constantly adjust and amend the joint weighted value through the back propagation of output error, so as to minimize the network error. The course of learning and training includes forward calculation and error back propagation. Generally, node function in the traditional BP neural network is Sigmoid. The modifier formulas of joint weighted value of the BP arithmetic can be expressed as below:

In the later stage of development, most oilfields faced high water cut issue after treatments. Meanwhile, how to effectively control oil wells water breakthrough after fracturing so as to increase oil production has become a big challenge for production engineers. This paper introduced a novel selective proppant which can effectively block water from entering well bore after fracturing. Based on "eggshell principle" and interfacial tension theory, this novel proppant was developed with a new monolayer strong hydrophobic interface treatment technique. This special processing technique could not only reduce the proppant's flow resistance of oil, but also increase its flow resistance of water. As the result, this novel proppant showed the characteristics of excellent lipophilicity and hydrophobicity as well. Experiments indicated that oil water diversion ratio of selective proppant was greater than 1.2, while the ordinary proppant was less than 0.7. The oil wettability index of selective proppant was greater than 0.8, while the ordinary proppant was less than 0.3. The main characteristics of this supporting agent was small apparent density, low broken rate, and high flow capacity. The proppant surface is oil wet, which has the characteristics of oil increasing precipitation. Experiments showed that the seepage velocity of water in selective proppant was obviously lower than in quartz sand, selective proppant with water resistance. Field test was carried out in 11 wells with selective proppant fracturing. A group of conventional fracturing wells were selected as contrasted wells with similar reservoir and operating conditions. After fracturing with novel proppant, the average water cut of producers was decreased by 7.2% per well, the average oil production was increased by 6.3 tons per well, and the average valid fracturing a period was 14.1 months. As for contrasted wells with conventional fracturing proppant, the average single well water cut decline was only 0.4%, the average single well daily oil 3.1 tons, an average period of only 5.8 months. The novel proppant showed remarkable advantages in improving oil production and decreasing water cut. This technique has a great future for mature oilfield development.