tailieunhanh - Expert Systems for Human Materials and Automation Part 8

Tham khảo tài liệu 'expert systems for human materials and automation part 8', kỹ thuật - công nghệ, cơ khí - chế tạo máy phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | Interface Layers Detection in Oil Field Tanks A Critical Review 201 Hence for a time delay less than a threshold twater thliquid d where d is the distance between the sensor and reflector the type of liquid being sensed by the actual sensor corresponds to water. Otherwise in case the time delay is greater than oil thliquid d then the liquid is either emulsion or oil depending on the number of pulses being collected . emulsion for less than 3 pulses oil otherwise . Finally in case no echo is detected then the corresponding phase corresponds to foam or gas. Note that the thresholds twater thliquid d and toil thliquid d . according to Section a and Figure for an operation temperature ranging from 20 0C to 70 0C setting twater and toil to 140 qs and toil 150 ps respectively is reasonable for thliquid d were selected in such a way that the classification is independent of the temperature. The same procedure is done for all sensors of the device to provide the water-cut profile of the column. This algorithm which has been coded in assembly and implemented into the transmitter has the advantage of being simple and does not require complicated hardware. However it is not capable to provide the water-cut value. b. A neural network-based algorithm for water-cut computation The second algorithm dedicated for water-cut computation is based on a feed forward neural network with backpropagation training. The motivation of using neural network is due to the fact that the elements of the database as shown in Figures 3 5 and 6 are not linear and depend on several variables . temperature and flow rate . The topology that gave satisfactory results was input layer of dimension 6 one hidden layer with 6 neurons and the output layer with 1 neuron for the water-cut value Figure 23 . This network demonstrated to be robust enough to determine the water-cut value within relatively low computation time. The first layer contains the six input variables peak to peak voltage delay

TỪ KHÓA LIÊN QUAN
crossorigin="anonymous">
Đã phát hiện trình chặn quảng cáo AdBlock
Trang web này phụ thuộc vào doanh thu từ số lần hiển thị quảng cáo để tồn tại. Vui lòng tắt trình chặn quảng cáo của bạn hoặc tạm dừng tính năng chặn quảng cáo cho trang web này.