Thresholds Of Back Propagation Neural Network Research Articles

Application of BP Neural Network Based on Genetic Algorithm Optimization in Evaluation of Power Grid Investment Risk

The artificial intelligence calculation method can effectively solve various nonlinear mapping relationships. The strength of these nonlinear solvers is exploited for the evaluation of power grid investment risk using back propagation (BP) neural network optimized by genetic algorithm. The mathematical model of the problem is constructed by selecting the transfer function of the neural network and defining the fitness function of genetic algorithm. BP neural network has good ability of self-learning, self-adaptation and generalization, which can overcome the drawbacks of traditional evaluation methods relying on experts' experience. For the characteristics of genetic algorithm global optimization, the genetic algorithm is used to optimize the weight and threshold of BP neural network, and BP neural network is trained to obtain the optimal evaluation model. The model fully exploits the local search ability of BP neural network and the global search ability of genetic algorithm. It has obtained good evaluation accuracy for the processing of multi-dimensional influence factor problem. And the model can be adapted to different power grids by changing the training data. However, the method cannot describe the specific relationship between each impact factor and the investment risk of the grid. The case study shows that the method can accurately and effectively evaluate power grid investment risk and improve the fault tolerance of the power grid investment risk evaluation.

Gesture Recognition Based on BP Neural Network Improved by Chaotic Genetic Algorithm

Aim at the defects of easy to fall into the local minimum point and the low convergence speed of back propagation (BP) neural network in the gesture recognition, a new method that combines the chaos algorithm with the genetic algorithm (CGA) is proposed. According to the ergodicity of chaos algorithm and global convergence of genetic algorithm, the basic idea of this paper is to encode the weights and thresholds of BP neural network and obtain a general optimal solution with genetic algorithm, and then the general optimal solution is optimized to the accurate optimal solution by adding chaotic disturbance. The optimal results of the chaotic genetic algorithm are used as the initial weights and thresholds of the BP neural network to recognize the gesture. Simulation and experimental results show that the real-time performance and accuracy of the gesture recognition are greatly improved with CGA.

Application of PSO-BP Neural Network Model on Identification of Human Dynamic Balance

Since the human dynamic balance system is nonlinear, common models can hardly characterize the adjustment process of human dynamic balance veritably. In this research, the particle swarm optimization (PSO) algorithm and back propagation (BP) neural network algorithm had been introduced to the identification of nonlinear system models for human dynamic balance. Firstly, in order to establish BP neural network based on prediction model about human dynamic balance system, torque and angle of human ankle for 50 healthy college students had been acquisited to act as input and output of nonlinear identification respectively in conditions of suffering random visual excitation; secondly, aiming at improving the convergence performance of conventional BP neural network, this article adopted PSO algorithm to optimize the initial weights and thresholds of BP neural network and then the prediction model based on PSO-BP algorithm had been built; finally, the predictive results before and after optimization had been compared. According to the results of simulation and quantitative comparison, the presented algorithm had accomplished modeling for the adjustment process of human dynamic balance, and it also showed that PSO optimization algorithm not only improved the performance of BP neural network, making it didn’t fall into local minimum easily, but also strengthened the generalization capability and enhanced prediction accuracy, making it more truly and accurately to characterize the adjustment process when human body maintains dynamic balance.

Fault diagnosis of PV array based on optimised BP neural network by improved adaptive genetic algorithm

The fault diagnosis model based on back propagation (BP) neural network is especially suitable for multi-fault and complex pattern recognition. However, the selection of initial weights and thresholds of BP neural network is lack of basis and it is easy to fall into the local optimum. Genetic algorithm (GA) can be used to optimise the initial weights and thresholds of the BP neural network. However, the selection process of GA based on roulette wheel selection is a random operation and the parameters of GA, the crossover probability and the mutation probability, are given a constant value which reduces the efficiency. Thus improved adaptive GA provides improvement in the selection, crossing and mutation process of GA, which prevents the neural network from falling into local optimum more effectively. Results of cases study show that this method can determine the fault types of photovoltaic array effectively.

Back propagation neural network with adaptive differential evolution algorithm for time series forecasting

The back propagation neural network (BPNN) can easily fall into the local minimum point in time series forecasting. A hybrid approach that combines the adaptive differential evolution (ADE) algorithm with BPNN, called ADE–BPNN, is designed to improve the forecasting accuracy of BPNN. ADE is first applied to search for the global initial connection weights and thresholds of BPNN. Then, BPNN is employed to thoroughly search for the optimal weights and thresholds. Two comparative real-life series data sets are used to verify the feasibility and effectiveness of the hybrid method. The proposed ADE–BPNN can effectively improve forecasting accuracy relative to basic BPNN, autoregressive integrated moving average model (ARIMA), and other hybrid models.

An Optimized Neural Network Classifier for Automatic Modulation Recognition

Automatic modulation recognition which is one of the key technologies in no-cooperative communications has extensive application prospects in civilian and military fields. The design of classifier played a decisive role in recognition results. The classifier based on back propagation (BP) neural network is better in the existing methods. However, the traditional back propagation neural network (BPNN) have some well-known disadvantages. This study investigates the design of a classifier for recognition of six common digital modulations. This classifier based on BP neural network trained by improved particle swarm optimization (PSO) which is applied as a local search algorithm to find the optimal weights and thresholds of BPNN. The simulation experiment results demonstrate that the proposed classifier has a higher recognition accuracy than other classifiers. DOI : http://dx.doi.org/10.11591/telkomnika.v12i2.3930

Modeling FOG Drift Using Back-Propagation Neural Network Optimized by Artificial Fish Swarm Algorithm

Based on the temperature drift characteristic of fiber optic gyroscope (FOG), a novel modeling and compensation method which integrated the artificial fish swarm algorithm (AFSA) and back-propagation (BP) neural network is proposed to improve the output accuracy of FOG and the precision of inertial navigation system. In this paper, AFSA is used to optimize the weights and threshold of BP neural network which determine precision of the model directly. In order to verify the effectiveness of the proposed algorithm, the predicted results of BP optimized by genetic algorithm (GA) and AFSA are compared and a quantitative evaluation of compensation results is analyzed by Allan variance. The comparison result illustrated the main error sources and the sinusoidal noises in the FOG output signal are reduced by about 50%. Therefore, the proposed modeling method can be used to improve the FOG precision.

A New Automatic Modulation Classifier Using Swarm Intelligence-Trained Neural Network

Monitoring modulation type of the detected signal is the most important intermediate step between signal detection and demodulation. The back propagation neural network (BPNN) was widely used in constructing modulated signal classifier in the field of automatic modulation classification (AMC). There are many visible features in the back propagation (BP) algorithm including adaptive learning, the ability of fault tolerant, etc. However, this algorithm has two main disadvantages, such as the slow convergence speed and easily falling into the local minimum. This paper presents a novel modulation classifier using BPNN trained with swarm intelligence algorithms (SIA), for the sake of overcoming these deficiencies. The initial weights and thresholds of BP neural network were optimized by SIA. As the SIA has an excellent global search property, this classifier can consume less training time and improve the automatic modulation type identification rate.