Papers by Keyword: Prediction

Abstract: Rock of Xujiahe formation in Sichuan basin is characterized by very low porosity and permeability, forecast the spatial distribution of fractured reservoir is the key of success in exploration. Combine logging and seismic data, analysis multi-attributes, optimizing post-stack seismic data processing method, we selected amplitude, average energy, chaotic reflection and amplitude weighted instantaneous frequency (AWIF) to study seismic response. The fractured reservoir have weak amplitude, low average energy, low AWIF and high chaotic reflection. Then with the aid of 3D visual technology, research data set, and recognize that some data, like amplitude and impedance etc, should emphases low value, others like chaotic reflection and auto-fault extraction(AFE) etc, should stand out high value. Base on these, form a suit of technology predicting fractured reservoir by using 3D visualization method. Take amplitude, chaotic reflection and AFE data as case, depict the spatial distribution of fractured reservoir. The results can clearly exhibit the spatial distribution of reservoir

Abstract: Reservoir immersion is one of the environmental geological problems in the construction of Yanqing new town. In this paper, the geological environment in Yanqing area was decribed, the causes of the immersion were analyzed, the immersion degree evaluation was determined according to the measured data, and the reservoir immersion range was predicted using groundwater dynamics, which provide the basis for the construction of Yanqing new town and disaster prevention and mitigation.

Abstract: Aimed at the fault diagnosis and prediction of automobile engine, firstly designed a framework structure of automobile engine fault diagnosis and prediction system, and built a hardware platform; Secondly adopted the genetic algorithm neural network to fault prediction and diagnosis reasoning; Finally after analyzing automobile exhaust components, engine vibration, engine abnormal sound parameters, inferred the appeared and impending fault of automobile then made the tips for users on the screen. The results show that the performance of system is well, the accuracy of diagnosis and prediction is 95% in different conditions of experiment and debugging.

Abstract: Remain useful life (RUL) prediction technology which is significant in the condition based maintenance (CBM) is a hot research topic nowadays. Rolling bearing is a basic component widely used in the mechanical industry, and its reliability affects the operation of rotating machinery. On the basis of traditional RUL technology for rolling bearing, a method named random average method (RAM) is introduced into RUL prediction and the implementation of it is instructed in detail via the processing of vibration data in full life of rolling bearing. Compared to traditional method, the proposed method based on RAM is better in both accuracy and timeliness.

Abstract: Traffic volume prediction has been an interesting topic for decades during which various prediction models have been proposed. In this paper, Kalman filtering (KF) model is applied to predict traffic volume because of its significance in continuously updating the state variable as new observations. In order to enhance the prediction accuracy, an improved KF model is developed based on the current and historical data. To validate the improved KF model, empirical analysis is conducted. The results show that the improved KF model has higher accuracy than the traditional one and is more reliable and powerful in traffic volume prediction.

Abstract: A novel nonlinear time series prediction method is proposed in this paper. This prediction method is based on the Multi-dimensional Taylor Network. The structure of the Multi-dimensional Taylor Network is introduced firstly. The Multi-dimensional Taylor Network provides a new method to predict the nonlinear time series. The prediction model based on the Multi-dimensional Taylor Network can realize the prediction of the nonlinear time series just with input-output data without the system mechanism, and it can describe the dynamic characteristics of the system. Finally, the new prediction method is applied in the structural vibration response prediction. Results indicate the validity and the better prediction accuracy of this method.

Abstract: Experimental studies on operating a marine diesel engine to determine the performance map under different working conditions need to consume a lot of money and labor. To solve this problem, a mathematical model based on Artificial Neural Networks (ANNs) combined genetic algorithms (GA) to predicate the performance emissions of the marine diesel engine is firstly reported in this paper. The predicted result showed that the network performance is sufficient for all target emission outputs. The input layer without transfer function consisted of 11 neurons is used, and output layer predicted 16 polycyclic aromatic hydrocarbons (PAHs). Electronic parameters such as VIC, SOI, CRP, NUN, VEO and VEC have influences on the PAHs emissions. The actual data obtained from the diesel is well agreed with the predicted data. The usage of ANNs is highly recommended to predict engine emissions instead of having to undertake complex and time-consuming experimental studies.

Abstract: In this paper there is presented a possibility to estimate, with a sufficient accuracy, contact forces on idler rollers of hexagonal idler housing in case of the pipe conveyor belt. This prediction is based on values of tension forces that are measured experimentally. The calculated regressive models are able to predict contact forces in individual positions of idler rollers with suitable correctness. The measurements of tension forces were carried out on a static model of a pipe conveyor. This static model is able to simulate real operational conditions of idler rollers, which guide the conveyor belt. In this way it is very simply possible to determine contact force values in the individual positions of idler rollers. Information about contact forces is useful for designers and for users of the conveyor belts, as well.

Abstract: In order to improve the prediction accuracy of busy telephone traffic, this study proposes a busy telephone traffic prediction method that combines wavelet transformation and least square support vector machine (lssvm) model which is optimized by particle swarm optimization (pso) algorithm. Firstly, decompose the pretreatment of busy telephone traffic data with mallat algorithm and get low frequency component and high frequency component. Secondly, reconfigure each component and use pso_lssvm model predict each reconfigured one. Then the busy telephone traffic can be achieved. The experimental results show that the prediction model has higher prediction accuracy and stability.

Abstract: The ultimate natural gas resources of Sichuan Basin were predicted by the application of hydrocarbon reservoir scale sequence method. The results revealed that a total of 236 gas reservoirs exist in Sichuan Basin, with total expected natural gas resources of 57718×10⁸m³.110 gas reservoirs are undiscovered, with geological reserves of 4.04×10¹²m³, accounting for70.0% of the total natural gas resources, among which, 17 undiscovered gas reservoirs have resources of more than 1000×10⁸m³, 13 undiscovered natural gas reservoirs have resources of 500-1000×10⁸m³ and 15 undiscovered gas reservoirs have resources of 100-500×10⁸m³. The undiscovered natural gas resources mainly lie in the following key exploration fields: Permian-Triassic reef flat reservoir of Huankaijiang-Liangping continental shelf and deep to ultra-deep layers of continental Xujiahe Formation in West Sichuan, medium to shallow layers of Northeast Sichuan and Central Sichuan focused on Xujiahe Formation, piedmont zone of Longmen Mountain and Micang-Daba Mountain, oolitic flat of Jialingjiang Formation of East Sichuan, South Sichuan and North Sichuan as well as five large sets of unconformity surfaces, namely top surface of Sinian Dengying Formation, top surface of Middle-Upper Ordovician, top surface of Middle Carboniferous, top surface of Permian Yangxin Series and top surface of Leikoupo Formation.