Reduction of Higher Harmonics in the Signal Spectrum in Switched Mode Amplifiers Using Predistortions Generated by a Neural Network

The question about modern intelligent information processing methods usage for a ball mill filling level evaluation is considered. Vibration acceleration signal has been measured on a mill laboratory model for that purpose. It is made with accelerometer attached to a mill pin. The conclusion is made that mill filling level can not be measured with the help of such signal amplitude only. So this signal spectrum processed by a neural network is used. A training set for the neural network is formed with the help of spectral analysis methods. Trained neural network is able to find the correlation between mill pin vibration acceleration signal and mill filling level. Test set is formed from the data which is not included into the training set. This set is used in order to evaluate the network ability to evaluate the mill filling degree. The neural network guarantees no more than 7% error in the evaluation of mill filling level.

Download Full-text

A Method Based on Wavelet Neural Network for Power System Short Term Load Forecasting

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.521.303 ◽

2014 ◽

Vol 521 ◽

pp. 303-306 ◽

Cited By ~ 1

Author(s):

Hong Mei Zhong ◽

Jie Liu ◽

Qi Fang Chen ◽

Nian Liu

Keyword(s):

Neural Network ◽

Power System ◽

Bp Neural Network ◽

Load Forecasting ◽

Wavelet Neural Network ◽

Signal Spectrum ◽

Short Term ◽

Short Term Load Forecasting ◽

Load Sequence ◽

Daily Load

The short-term load of Power System is uncertain and the daily-load signal spectrum is continuous. The approach of Wavelet Neural Network (WNN) is proposed by combing the wavelet transform (WT) and neural network. By the WT, the time-based short-term load sequence can be decomposed into different scales sequences, which is used to training the BP neural network. The short-term load is forecasted by the trained BP neural network. Select the load of a random day in Lianyungang to study, according to the numerical simulation results, the method proves to achieve good performances.

Download Full-text

Fault Diagnosis of Water Pump Based on Probabilistic Neural Network

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.338.421 ◽

2011 ◽

Vol 338 ◽

pp. 421-424

Author(s):

Tie Jun Li ◽

Yan Chun Zhao ◽

Xin Li ◽

Cheng Shi Zhu ◽

Jian Rong Ning

Keyword(s):

Neural Network ◽

Fault Diagnosis ◽

Basic Principle ◽

Probabilistic Neural Network ◽

Back Propagation ◽

Vibration Signal ◽

Back Propagation Neural Network ◽

Signal Spectrum ◽

Water Pump ◽

Time Diagnosis

The basic principle of probabilistic neural network (PNN) is introduced, which is used in the fault diagnosis of water pump in this paper. The multiple and fractional frequencies in the fault vibration signal spectrum are taken as the feature vectors, and the samples of the fault are established. The probabilistic neural network is trained based on the symptom diagnosis. The result shows that probabilistic neural network can overcome the local optimization of back propagation neural network (BPNN) and meet the requirements for fast diagnosis and high precision diagnosis during fault diagnosis process, so probabilistic neural network can be used in the real time diagnosis, and the fault diagnosis based on probabilistic neural network is feasible.

Download Full-text

Neural network principle of implementation of digital filters

MATEC Web of Conferences ◽

10.1051/matecconf/201819303052 ◽

2018 ◽

Vol 193 ◽

pp. 03052 ◽

Cited By ~ 2

Author(s):

Sergey Morozov ◽

Gennady Makarov ◽

Konstantin Kuzmin

Keyword(s):

Neural Network ◽

Digital Filters ◽

Classical Method ◽

Signal Spectrum ◽

Neural Network Approach ◽

Frequency Responses ◽

The Neural Network ◽

Network Methods ◽

Element Base

Comparative evaluations of the frequency responses (FR) of two types of filters implemented by the classical and neural network methods are carried out. It is shown that the neural network principle of the implementation of digital filters can serve as an alternative to the classical method for specifically defined parameters of FR in the pass bands and attenuation bands of the frequencies of signal spectrum. The simplest method for calculating the parameters of the filters’ difference equations is the neural network approach, regardless of the type of classification of discrete and digital filters. The implementation of TM (transmultiplexer) on a digital element base requires the use of methods of filtering, modulating and demodulating signals that are largely different from traditional analog methods. The frequency responses of non-recursive types of filters presented in the paper are based on the property of the approximable function determined only in the pass bands and attenuation bands of the frequencies of signal spectrum.

Download Full-text

Tracking Error Estimate for Theodolite Based on General Regression Neural Network

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.472-475.1383 ◽

2012 ◽

Vol 472-475 ◽

pp. 1383-1387 ◽

Cited By ~ 1

Author(s):

Miao Li ◽

Hui Bin Gao

Keyword(s):

Neural Network ◽

Evaluation Method ◽

Tracking Error ◽

General Regression Neural Network ◽

Engineering Practice ◽

Accuracy Evaluation ◽

Tracking Accuracy ◽

Higher Harmonics ◽

Dynamic Target ◽

General Regression

To efficiently evaluate the tracking index of theodolite ,as well as avoiding influence resulted from higher harmonics when used opto dynamic target and use equivalent sine to evaluate the tracking index, in this paper, we resorted to General Regression Neural Network to do the new research on theodolite tracking accuracy evaluation method. Experimental results indicate that the method realized the equivalent sine to evaluate the tracking index of theodolite and avoided the higher harmonics influence with opto dynamic target. The research in this paper has important value to the engineering practice.

Download Full-text