In this paper, the deterministic convergence of the batch back-propagation algorithm with penalty (bpAP) is proved under certain relaxed conditions for the activation function, the learning rate and the stationary poi...
详细信息
In this paper, the deterministic convergence of the batch back-propagation algorithm with penalty (bpAP) is proved under certain relaxed conditions for the activation function, the learning rate and the stationary point set of the error function. Both weak and strong convergence results are established. The boundedness of the weights in the training procedure is also proved in a simple and clear way. As a result, the usual requirements on the boundedness of the weights to guarantee the convergence are removed. Simulation results for an approximation problem are presented to support our theoretical findings. (C) 2014 Elsevier B.V. All rights reserved.
In this paper, a nonlinear intelligent observer design is applied for a class of nonlinear discrete-time flexible joint robot (DFJR) dynamic system based on artificial neural network (ANN). The DFJR system has a relat...
详细信息
In this paper, a nonlinear intelligent observer design is applied for a class of nonlinear discrete-time flexible joint robot (DFJR) dynamic system based on artificial neural network (ANN). The DFJR system has a relatively complex nonlinear dynamic and internal states' estimation of it poses a challenging robotic problem. Multilayer perceptron (MLP) is an important class of feed-forward ANNs that maps set of inputs onto a set of suitable outputs. The ANN under online learning is one of the artificial intelligence methods. Therefore, the MLP neural nonlinear observer is trained online and it is robust in the presence of external and internal uncertainties. The learning method of the intelligent observer is a simple back propagation (bp) algorithm and, furthermore, the learning method of estimation of the link positions and the velocities is bp-developed algorithm. Simulation results show promising performance of the proposed observer in the presence of measurement noise and parameters uncertainties.
In this paper an improved algorithm of bp neural network—— Levenberg-Marquardt(LM) algorithm is introduced, and the simulation predictions of oilfield cementing quality is done by using this method. Finally, a pract...
详细信息
In this paper an improved algorithm of bp neural network—— Levenberg-Marquardt(LM) algorithm is introduced, and the simulation predictions of oilfield cementing quality is done by using this method. Finally, a practical example verified the feasibility of the presented method.
The occurrence of coal mine disaster related with many environmental and social *** relationship between them was uncertainty,and was a kind of coupling relationship,and was *** was difficult to fit the relationship b...
详细信息
The occurrence of coal mine disaster related with many environmental and social *** relationship between them was uncertainty,and was a kind of coupling relationship,and was *** was difficult to fit the relationship between them using a mathematical *** also was the important reason that coal mine disaster was always hard to *** artificial neural network based on bp algorithm had highly nonlinear mapping *** could nonlinear map the relationship between the probability of coal mine disaster’s occurrence and its effect factors on the condition of building no complex mathematical *** then the probability of coal mine disaster could be predicted relatively *** provided technical support for prevention and management of coal mine disaster.
The occurrence of coal mine disaster related with many environmental and social factors. The relationship between them was uncertainty, and was a kind of coupling relationship, and was nonlinear. It was difficult to f...
详细信息
ISBN:
(纸本)9781612848334
The occurrence of coal mine disaster related with many environmental and social factors. The relationship between them was uncertainty, and was a kind of coupling relationship, and was nonlinear. It was difficult to fit the relationship between them using a mathematical model. This also was the important reason that coal mine disaster was always hard to predict. The artificial neural network based on bp algorithm had highly nonlinear mapping function. It could nonlinear map the relationship between the probability of coal mine disaster's occurrence and its effect factors on the condition of building no complex mathematical model. And then the probability of coal mine disaster could be predicted relatively accurately. It provided technical support for prevention and management of coal mine disaster.
This paper proposes a NARX (Nonlinear Auto-Regressive with Exogenous Inputs) model for the water distribution network real-time prediction and control. The model estimates the time-variable nodal demand equivalently b...
详细信息
Due to the explosive increment of data in big data era, it is a challenging task to analyze and extract meaningful data for users. Data needs to be timely operated because of the time sensitivity, so it faces enormous...
详细信息
ISBN:
(纸本)9781479920037
Due to the explosive increment of data in big data era, it is a challenging task to analyze and extract meaningful data for users. Data needs to be timely operated because of the time sensitivity, so it faces enormous pressure in storage and computing. To deal with the problem that it is hard to achieve valuable information from out-of-order streams over big data in short time, a model-matching algorithm based on improved bp (Back Propagation) is proposed. In the algorithm, the matching model is set dynamically. Information is extracted for users according to the order of data's arriving time. Furthermore, the algorithm parameters are automatically adjusted in the process of learning and matching. Accordingly, the responding speed of learning is accelerated and the time of matching reduces. In the simulation, a group of optimum parameters of improved bp are achieved by using self-adapting adjusting mechanism. The threshold (TH), connecting weight (CW) and learning rate (LR) are equal to 1.5, 3 and 1, respectively. We implement our model-matching algorithm on 10000 sets of out-of-order streams with these parameters. Results indicate that the proposed algorithm can obviously improve the accuracy and speed of matching and achieve better stability.
In order to solve the problem of garbage treatment effectively, a control method is proposed based on Takagi-Sugeno (T-S) fuzzy neural network model after analyzing the characteristics of garbage incinerators system a...
详细信息
ISBN:
(纸本)9781467368506
In order to solve the problem of garbage treatment effectively, a control method is proposed based on Takagi-Sugeno (T-S) fuzzy neural network model after analyzing the characteristics of garbage incinerators system and the main factors affecting combustion. A T-S fuzzy neural network model for garbage incinerators control is established which utilizes bp learning algorithm for data training. Then a simulation research is carried out to verify the feasibility and superiority. Results show that the T-S fuzzy neural network control can well track the input in a relative short time. The contrast analysis with conventional PID control and fuzzy control is done to show a better performance under the fuzzy neural network control. The fuzzy neural network control method can adapt to the complex garbage incineration process, which makes it high application value.
With the development of mobile Internet applications, the fourth generation mobile communication (4G) has been widely used, low-density parity-check (LDPC) codes have gradually become the first choice for 4G communica...
详细信息
ISBN:
(纸本)9781467397957
With the development of mobile Internet applications, the fourth generation mobile communication (4G) has been widely used, low-density parity-check (LDPC) codes have gradually become the first choice for 4G communication because of its superior performance. For the short of traditional encoding and belief propagation (bp) decoding algorithm, this paper adopts a method of parallel processing for LDPC encoding and decoding which is finally realized on the mobile terminal. On the CUDA platform, CPU scheduling and GPU parallelization are used to process a large number of repeated operations so that parallel encoding algorithm and heterogeneous parallel bp algorithm are achieved and the efficiency is significantly improved.
This paper presents a new concept of Space-Airborne Bi-Static Linear Array SAR (SA-BiLASAR) imaging using array antenna, which inherited the advantages of conventional linear array antenna 3-D SAR and Bi-Static SAR. F...
详细信息
ISBN:
(纸本)9781467372978
This paper presents a new concept of Space-Airborne Bi-Static Linear Array SAR (SA-BiLASAR) imaging using array antenna, which inherited the advantages of conventional linear array antenna 3-D SAR and Bi-Static SAR. Firstly, the geometrical model and the signal model are established. Secondly, a time domain back projection imaging algorithm is presented for SA-BiLASAR data processing. In the end, the feasibility of the SA-BiLASAR and the imaging algorithm arc demonstrated by simulated results.
暂无评论