According to advantages of neural network and characteristics of operatingprocedures of engine, a new strategy is represented on the control of fuel injection and ignitiontiming of gasoline engine based on improved bp...
详细信息
According to advantages of neural network and characteristics of operatingprocedures of engine, a new strategy is represented on the control of fuel injection and ignitiontiming of gasoline engine based on improved bp network algorithm. The optimum ignition advance angleand fuel injection pulse band of engine under different speed and load are tested for the samplestraining network, focusing on the study of the design method and procedure of bp neural network inengine injection and ignition control. The results show that artificial neural network technique canmeet the requirement of engine injection and ignition control. The method is feasible for improvingpower performance, economy and emission performances of gasoline engine.
Back propagation (bp) algorithm is a very useful algorithm in many areas, but its leaning process is a very complicated non linear convergence process, in which, chaos often happens, and slow convergence speed and loc...
详细信息
Back propagation (bp) algorithm is a very useful algorithm in many areas, but its leaning process is a very complicated non linear convergence process, in which, chaos often happens, and slow convergence speed and local least often make it difficult for the non experts to use it widely, and an improved bp (Ibp) algorithm is therefore suggested to expedite the convergence speed. The algorithm can judge local least and take some steps automatically to jump out from the local least. Furthermore, this algorithm introduces the expert knowledge base. An Ibp based agile and current neural network (NN) constructed tool is designed. An initial NN can be constructed automatically using an expert knowledge base. And an Aitken’s Δ 2 process method is used to expedite the convergent speed for NN. Besides, the method of changing the parameter of Sigmoid function and increasing the hidden node is used to bring surge for NN to jump out from the local
In this paper, the Artificial Neural Network (ANN) is used to study the wave forces on a semi-circular breakwater. The process of establishing the network model for a specific physical problem is presented. Networks w...
详细信息
In this paper, the Artificial Neural Network (ANN) is used to study the wave forces on a semi-circular breakwater. The process of establishing the network model for a specific physical problem is presented. Networks with double implicit layers have been studied by numerical experiments. 117 sets of experimental data are used to train and test the ANN. According to the results of ANN simulation, this method is proved to have good precision compared with experimental and numerical results.
Application of ANN (Artificial neural network) to the electrical properties analysis of PZT is discussed in this paper. The same set of results of PZT samples were analyzed by a back-propagation (bp) network in compar...
详细信息
Application of ANN (Artificial neural network) to the electrical properties analysis of PZT is discussed in this paper. The same set of results of PZT samples were analyzed by a back-propagation (bp) network in comparison with a multiple nonlinear regression analysis (MNLR) model. The results revealed that the ANN model is much more accurate than MNLR model. The ANN approach also gave quite encouraging predictions for formulations not included in the train set samples, indicating that the bp network is a very useful and accurate tool for the properties analysis and prediction of multi-component solid solution piezoelectric ceramics. (C) 2003 Elsevier Science Ltd. All rights reserved.
On the basis of analyzing the essential of bp algorithm's learning process, the self-adapting slope function is used as the output function in neuron. The difference of the learning rate between the new bp algorit...
详细信息
On the basis of analyzing the essential of bp algorithm's learning process, the self-adapting slope function is used as the output function in neuron. The difference of the learning rate between the new bp algorithm and the standard bp algorithm is discussed. And based on it, how the rate difference is affected by the net inputs is analyzed. The conclusion in practical use is validated.
The greatest difficulty for bp network apply in practice is the low learning speed. In this paper, we analyze particularly the reason for the low speed of bp algorithm, bring forward self-adjust learning bp algorithm ...
详细信息
ISBN:
(纸本)0780378652
The greatest difficulty for bp network apply in practice is the low learning speed. In this paper, we analyze particularly the reason for the low speed of bp algorithm, bring forward self-adjust learning bp algorithm and the regulating policy of the hidden lay.
No lossless data compression method based on neural network is found before. A lossless compression method based on bp network for the long character-string of 0 and 1 is given by establishing specific mapping Y and s...
详细信息
ISBN:
(纸本)7563506861
No lossless data compression method based on neural network is found before. A lossless compression method based on bp network for the long character-string of 0 and 1 is given by establishing specific mapping Y and specific integer function and with the non-linear approximation capability of concrete three-layer bp network in this paper, and the compression & decompression algorithms of the lossless compression method are provided. Experiments show that the compression ratio of the lossless compression method is usually around 16/11 and the method can effectively compress the data which have been compressed by Huffman coding, arithmetric coding or dictionary coding.
An improved compound gradient vector based a NN online training weight update scheme is proposed in this paper. The convergent analysis indicates that because the compound gradient vector is employed during the weight...
详细信息
ISBN:
(纸本)0780378830
An improved compound gradient vector based a NN online training weight update scheme is proposed in this paper. The convergent analysis indicates that because the compound gradient vector is employed during the weight update, the convergent speed of the presented algorithm is faster than the bp algorithm. In this scheme an adaptive learning factor is introduced, in which the global convergence is obtained. and the convergent procedure on plateau and flat bottom area can speed up. Simulations have been conducted and the results demonstrate that the satisfactory convergent performance and strong robustness are obtained using the improved compound gradient vector NN online learning scheme for real time control.
Artificial Neural Networks (ANN) has many good qualities comparing with ordinary methods in Land Suitability Evaluation. Based on analysis of ordinary methods' limitations, some sticking points of bp model of ANN ...
详细信息
ISBN:
(纸本)0819451819
Artificial Neural Networks (ANN) has many good qualities comparing with ordinary methods in Land Suitability Evaluation. Based on analysis of ordinary methods' limitations, some sticking points of bp model of ANN used in land evaluation are discussed in detail, such as network structure, learning algorithm, etc. The land evaluation of Qionghai city is used as a case study. we know that ANN always can give more reasonable evaluation results from test.
The nonlinear compensation in the intelligent thermometer measurement instrumentation is introduced, and disadvantages of the compensation in hardware or software are discussed at present. The bp neural network method...
详细信息
ISBN:
(纸本)0780378652
The nonlinear compensation in the intelligent thermometer measurement instrumentation is introduced, and disadvantages of the compensation in hardware or software are discussed at present. The bp neural network method for the nonlinear compensation and its merits are proposed. The number of the hidden layer nodes and the improved algorithm are analyzed deeply, and the modeling process is presented in this paper. The example demonstrates that a more precise mathematical model of the thermocouple can be got based on the improved bp algorithm, which can be implemented by software, it is beneficial to improve the instrument accuracy.
暂无评论