Accurate diagnosis of fracture geometry and conductivity is of great challenge due to the complex morphology of volumetric fracture network. In this study, a DNN (deep neural network) model was proposed to predict fra...Accurate diagnosis of fracture geometry and conductivity is of great challenge due to the complex morphology of volumetric fracture network. In this study, a DNN (deep neural network) model was proposed to predict fracture parameters for the evaluation of the fracturing effects. Field experience and the law of fracture volume conservation were incorporated as physical constraints to improve the prediction accuracy due to small amount of data. A combined neural network was adopted to input both static geological and dynamic fracturing data. The structure of the DNN was optimized and the model was validated through k-fold cross-validation. Results indicate that this DNN model is capable of predicting the fracture parameters accurately with a low relative error of under 10% and good generalization ability. The adoptions of the combined neural network, physical constraints, and k-fold cross-validation improve the model performance. Specifically, the root-mean-square error (RMSE) of the model decreases by 71.9% and 56% respectively with the combined neural network as the input model and the consideration of physical constraints. The mean square error (MRE) of fracture parameters reduces by 75% because the k-fold cross-validation improves the rationality of data set dividing. The model based on the DNN with physical constraints proposed in this study provides foundations for the optimization of fracturing design and improves the efficiency of fracture diagnosis in tight oil and gas reservoirs.展开更多
The Unified Power Quality Conditioner (UPQC) plays an important role in the constrained delivery of electrical power from the source to an isolated pool of load or from a source to the grid. The proposed system can co...The Unified Power Quality Conditioner (UPQC) plays an important role in the constrained delivery of electrical power from the source to an isolated pool of load or from a source to the grid. The proposed system can compensate voltage sag/swell, reactive power compensation and harmonics in the linear and nonlinear loads. In this work, the off line drained data from conventional fuzzy logic controller. A novel control system with a Combined Neural Network (CNN) is used instead of the traditionally four fuzzy logic controllers. The performance of combined neural network controller compared with Proportional Integral (PI) controller and Fuzzy Logic Controller (FLC). The system performance is also verified experimentally.展开更多
Random and fluctuating wind speeds make it difficult to stabilize the wind-power output,which complicates the execution of wind-farm control systems and increases the response frequency.In this study,a novel predictio...Random and fluctuating wind speeds make it difficult to stabilize the wind-power output,which complicates the execution of wind-farm control systems and increases the response frequency.In this study,a novel prediction model for ultrashort-term wind-speed prediction in wind farms is developed by combining a deep belief network,the Elman neural network,and the Hilbert-Huang transform modified using an improved particle swarm optimization algorithm.The experimental results show that the prediction results of the proposed deep neural network is better than that of shallow neural networks.Although the complexity of the model is high,the accuracy of wind-speed prediction and stability are also high.The proposed model effectively improves the accuracy of ultrashort-term wind-speed forecasting in wind farms.展开更多
A Raman spectroscopy method combined with neural network is used for the invasive and rapid detection of echinococcosis. The Raman spectroscopy measurements are performed on two groups of blood serum samples,which are...A Raman spectroscopy method combined with neural network is used for the invasive and rapid detection of echinococcosis. The Raman spectroscopy measurements are performed on two groups of blood serum samples,which are from 28 echinococcosis patients and 38 healthy persons,respectively. The normalized Raman reflection spectra show that the reflectivity of the echinococcosis blood serum is higher than that of the normal human blood serum in the wavelength ranges of 101—175 nm and 1 801—2 701 nm. Then the principal component analysis(PCA) and back propagation neural network(BPNN) model are used to obtain the diagnosis results. The diagnosis rates for healthy persons and echinococcosis persons are 93.333 3% and 90.909 1%,respectively,so the average final diagnosis rate is 92.121 2%. The results demonstrate that the Raman spectroscopy analysis of blood serum combined with PCA-BPNN has considerable potential for the non-invasive and rapid detection of echinococcosis.展开更多
基金supported by the National Natural Science Foundation of China(Grant No.52174044,52004302)Science Foundation of China University of Petroleum,Beijing(No.ZX20200134,2462021YXZZ012)the Strategic Cooperation Technology Projects of CNPC and CUPB(ZLZX 2020-01-07).
文摘Accurate diagnosis of fracture geometry and conductivity is of great challenge due to the complex morphology of volumetric fracture network. In this study, a DNN (deep neural network) model was proposed to predict fracture parameters for the evaluation of the fracturing effects. Field experience and the law of fracture volume conservation were incorporated as physical constraints to improve the prediction accuracy due to small amount of data. A combined neural network was adopted to input both static geological and dynamic fracturing data. The structure of the DNN was optimized and the model was validated through k-fold cross-validation. Results indicate that this DNN model is capable of predicting the fracture parameters accurately with a low relative error of under 10% and good generalization ability. The adoptions of the combined neural network, physical constraints, and k-fold cross-validation improve the model performance. Specifically, the root-mean-square error (RMSE) of the model decreases by 71.9% and 56% respectively with the combined neural network as the input model and the consideration of physical constraints. The mean square error (MRE) of fracture parameters reduces by 75% because the k-fold cross-validation improves the rationality of data set dividing. The model based on the DNN with physical constraints proposed in this study provides foundations for the optimization of fracturing design and improves the efficiency of fracture diagnosis in tight oil and gas reservoirs.
文摘The Unified Power Quality Conditioner (UPQC) plays an important role in the constrained delivery of electrical power from the source to an isolated pool of load or from a source to the grid. The proposed system can compensate voltage sag/swell, reactive power compensation and harmonics in the linear and nonlinear loads. In this work, the off line drained data from conventional fuzzy logic controller. A novel control system with a Combined Neural Network (CNN) is used instead of the traditionally four fuzzy logic controllers. The performance of combined neural network controller compared with Proportional Integral (PI) controller and Fuzzy Logic Controller (FLC). The system performance is also verified experimentally.
基金This study was supported by the Research and Application of Key Technologies in the Design of Large Onshore Smart Wind Power Base(Grant No.XBY-ZDKJ-2020-05)the Scientific Research Project of the China Electric Power Construction Corporation:Research and Application of Key Technologies in the Design of an Onshore Smart Wind Power Base(Grant No.DJ-ZDXM-2020-52)+2 种基金the Danish Energy Agency(Grant No.64013-0405)the Fundamental Research Funds for the Central Universities(Grant No.B210201018)the Jiangsu Province Policy Guidance Program(Grant No.BZ2021019).
文摘Random and fluctuating wind speeds make it difficult to stabilize the wind-power output,which complicates the execution of wind-farm control systems and increases the response frequency.In this study,a novel prediction model for ultrashort-term wind-speed prediction in wind farms is developed by combining a deep belief network,the Elman neural network,and the Hilbert-Huang transform modified using an improved particle swarm optimization algorithm.The experimental results show that the prediction results of the proposed deep neural network is better than that of shallow neural networks.Although the complexity of the model is high,the accuracy of wind-speed prediction and stability are also high.The proposed model effectively improves the accuracy of ultrashort-term wind-speed forecasting in wind farms.
基金supported by the National Natural Science Foundation of China(No.61308120)the Doctor Startup Project of Xinjiang University(No.BS120122)+1 种基金the Young Talents Project in Xinjiang Uygur Autonomous Region(No.2013731003)the Xinjiang Science and Technology Project(Nos.201412107 and 2014211B003)
文摘A Raman spectroscopy method combined with neural network is used for the invasive and rapid detection of echinococcosis. The Raman spectroscopy measurements are performed on two groups of blood serum samples,which are from 28 echinococcosis patients and 38 healthy persons,respectively. The normalized Raman reflection spectra show that the reflectivity of the echinococcosis blood serum is higher than that of the normal human blood serum in the wavelength ranges of 101—175 nm and 1 801—2 701 nm. Then the principal component analysis(PCA) and back propagation neural network(BPNN) model are used to obtain the diagnosis results. The diagnosis rates for healthy persons and echinococcosis persons are 93.333 3% and 90.909 1%,respectively,so the average final diagnosis rate is 92.121 2%. The results demonstrate that the Raman spectroscopy analysis of blood serum combined with PCA-BPNN has considerable potential for the non-invasive and rapid detection of echinococcosis.