Artificial neural networks have been widely used to predict the mechanical properties of alloys in material research. This study aims to investigate the implicit relationship between the compositions and mechanical pr...
详细信息
Artificial neural networks have been widely used to predict the mechanical properties of alloys in material research. This study aims to investigate the implicit relationship between the compositions and mechanical properties of as-cast Mg-Li-AI alloys. Based on the experimental collection of the tensile strength and the elongation of representative Mg-Li-AI alloys, a momentum back-propagation (bp) neural network with a single hidden layer was established. Particle swarm optimization (PSO) was applied to optimize the bp model. In the neural network, the input variables were the contents of Mg, Li and AI, and the output variables were the tensile strength and the elongation. The results show that the proposed PSO-bp model can describe the quantitative relationship between the Mg-Li-AI alloy's composition and its mechanical properties. It is possible that the mechanical properties to be predicted without experiment by inputting the alloy composition into the trained network model. The prediction of the influence of AI addition on the mechanical properties of as-cast Mg-Li-AI alloys is consistent with the related research results.
The high inertia and long time-delay characteristics of main steam temperature control system in a thermal power plant will reduce the system control performance. In order to solve this problem, a genetic algorithm-ba...
详细信息
The high inertia and long time-delay characteristics of main steam temperature control system in a thermal power plant will reduce the system control performance. In order to solve this problem, a genetic algorithm-back propagation (GA-bp) optimised fuzzy neural network control strategy is proposed in this paper. Gauss function is chosen as membership function and fuzzy neural network is designed. GA combined with bp algorithm is chosen for the offline parameters optimisation of fuzzy neural network, and then bp algorithm is used for online parameters optimisation. GA-bp optimisation algorithm overcomes the shortcomings of GA algorithm or bp algorithm which is used to adjust the parameters of fuzzy neural network controller. The simulation experiment compared with cascade PID and fuzzy neural network is carried out. Simulation results show that the controller based on GA-bp optimised fuzzy neural network has faster response speed, smaller overshoot and error, better tracking performance, and reduces the lag effect of the control system under different load, working conditions and membership functions.
Aimed at the inadequacy of the standard bp algorithm, a near optimal learning rate bp algorithm (NOLRbp) is presented. Selecting the learning rate of the algorithm based on one-dimensional search algorithm of optimiza...
详细信息
ISBN:
(纸本)9783037852828
Aimed at the inadequacy of the standard bp algorithm, a near optimal learning rate bp algorithm (NOLRbp) is presented. Selecting the learning rate of the algorithm based on one-dimensional search algorithm of optimization theory avoids the blindness in determining the learning rate. Simulations show that the algorithm is superior to the standard bp algorithm (SDbp), momentum bp algorithm (MObp) and variable learning rate bp algorithm (VLbp).
Based on JS engine, this paper made a detailed analysis on the standard bp algorithm, bp algorithm with additional momentum item, bp algorithm of parameter to adjust and bp network based on LM algorithm. Finally, this...
详细信息
ISBN:
(纸本)9783037852989
Based on JS engine, this paper made a detailed analysis on the standard bp algorithm, bp algorithm with additional momentum item, bp algorithm of parameter to adjust and bp network based on LM algorithm. Finally, this paper gets the optimization algorithm and network by comparing in training step and diagnostic accuracy. In the end, the front panel of the diagnosis system is designed by MATLAB graphical user interface.
bp network is the most widely used of the neural net work model, but there are many problems of slow convergence speed and easily getting into the local minimum in the conventional bp algorithm. For this, an improved ...
详细信息
ISBN:
(纸本)9783037852828
bp network is the most widely used of the neural net work model, but there are many problems of slow convergence speed and easily getting into the local minimum in the conventional bp algorithm. For this, an improved algorithm is proposed. Momentum term is added, steepness factors are introduced and adaptive learning rate adjustment factor is added. In the Matlab platform simulations are carried out by each improvement methods on the same bp neural network. The results show that: Convergence of improved bp network is decreased from 1000 to 49 and the error is decreased from 10(-2) to 10(-6). The convergence speed has been significantly improved and the error has been decreased. Using the synthesis improvement method effect is obvious and it provides a good theoretical basis for the practical application.
The reinforcement rib design is one of the key parts in entire bottle design. This paper presents the rib performance prediction system based on the bp algorithm and the finite element analysis, which adopts the finit...
详细信息
ISBN:
(纸本)9783037852590
The reinforcement rib design is one of the key parts in entire bottle design. This paper presents the rib performance prediction system based on the bp algorithm and the finite element analysis, which adopts the finite element analysis results as its learning samples, sets up the rib performance prediction system with bp artificial neural network. The results show that the artificial neural network plays an important role in rib performance prediction;meanwhile it can guide the bottle design in practical terms.
A DG-based distributed fault diagnosis method based on bp neural network with dynamic adaptive fuzzy Petri nets is proposed to solve the problem that traditional fault diagnosis methods lead to complex matrix and swit...
详细信息
ISBN:
(纸本)9781509046584
A DG-based distributed fault diagnosis method based on bp neural network with dynamic adaptive fuzzy Petri nets is proposed to solve the problem that traditional fault diagnosis methods lead to complex matrix and switching *** this paper,the general fault diagnosis model is constructed,and the simplified model of protection information is processed in the form of *** the operation mode and protection are changed,the model need not be reestablished,and the logic of the protection circuit breaker error correction is used with high fault ***,bp algorithm is used to train the fuzzy parameters in the ***,simulation test is carried out for the distribution network with DG,which verifies the reliability and fastness of the method.
In this paper, we take the average impact value method as the evaluation of neural network variable correlation indicators, analysis the data provided by Professor *** and A. Morais from University of Minho(Portugal) ...
详细信息
ISBN:
(纸本)9781509046584
In this paper, we take the average impact value method as the evaluation of neural network variable correlation indicators, analysis the data provided by Professor *** and A. Morais from University of Minho(Portugal) using the MIVbp algorithm to filtrate 13 characterization factors to get 7 characterization parameters affect forest fires, construct the simulation model of the prediction of forest fires based on the support vector machine algorithm, using a test set for testing, the accuracy rate reached 91.89%.
Stop-and-hop (S&H) assumption is usually exploited by most multiple receiver synthetic aperture sonar (SAS) imaging geometries, which are the basis of the SAS image formation algorithms. It is a reasonable approxi...
详细信息
ISBN:
(纸本)9781509039449
Stop-and-hop (S&H) assumption is usually exploited by most multiple receiver synthetic aperture sonar (SAS) imaging geometries, which are the basis of the SAS image formation algorithms. It is a reasonable approximation for the target focusing with a slow sonar platform speed. However, it is not suitable for the systems with a fast speed at far range as this assumption would deteriorate the target focusing performances. This paper firstly analyzes the error of S&H approximation in the 2-D space domain, which shows the necessity of the compensation for this approximation error. Based on back projection (bp) algorithm, simulated data is processed with and without the S&H approximation, respectively. The experiments are carried out with two sonar platform velocities, i. e. slow and fast sonar platform velocities. The processing results further indicate that the S&H error should be compensated when the system is operated with a fast speed.
This paper predicts the temperature of optical transmitter and receiver through the bp algorithm based on MATLAB simulation *** inner temperature is affected by ambient temperature,heating power,air pressure,wrapping ...
详细信息
This paper predicts the temperature of optical transmitter and receiver through the bp algorithm based on MATLAB simulation *** inner temperature is affected by ambient temperature,heating power,air pressure,wrapping mode,whose stabilization needs corresponding adjustment of these *** the bp network structure with 4-8-5-1 is adopted,the Tansig function is applied in the implication layer,at the same time,the Purelin function is applied in the output *** last,the finite samples are trained and tested through the *** with the PID algorithm,the simulation result shows that there is a higher precision and faster convergence in the bp *** the meantime,the generalization ability of the bp network gives inner temperature when there are 4 inputs,which reflects the basic feature of the whole system.
暂无评论