A back-propagation algorithm applied to tool wear monitoring

Purushothaman, S.; Srinivasa, Y.G.

doi:10.1016/0890-6955(94)90047-7

Cited by 60 publications

(24 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the study, a total of 262 data consisting of 244 data for training and 18 data for test were used. The input and output data used were normalized between 0 and 1 (Purushothaman and Srinivasa, 1994).…”

Section: Artificial Neural Networkmentioning

confidence: 99%

Determination of weed intensity in wheat production using image processing techniques

Ağın¹,

Taner²

2015

Anadolu J. Agr. Sci.

View full text Add to dashboard Cite

It is of great importance to precisely and carefully apply the minimum amount of chemicals as needed because agricultural chemicals negatively impact the human health, environment and balance in the nature and increase the production costs. In this study, it was aimed at determining the density of broad leaf weeds and contributing to the reduction of herbicide use in wheat grown fields. For this purpose, Image Processing Techniques were used in this study; and Artificial Neural Networks (ANN) and regression models were developed for determination of weeds. In the ANN model, Weed Covered Areas Acquired by Image Processing Techniques (WCAAIPT) was used as input parameter; and Actual Weed Covered Areas (AWCA) as output parameter. In the study, a total of 262 data consisting of 244 data for training and 18 data for test were used. In the ANN model, the structure of the network was designed in the form of 1-(9-5)-1, consisting of 1 input layer, 2 hidden layers and 1 output layer; and the number of neurons in the hidden layer were determined to be 9-5. Also, tansig was used in the first hidden layer, logsig in the second hidden layer; and purelin transfer functions were used in the output layer. In the ANN and Regression models, R 2 value of the ANN model was found to be 99% and the goodness of fit (U 2 ) to be 0.000436; whereas R 2 and U 2 values of the Regression model were found to be 95% and 0.008431, respectively. It was determined that the results obtained from the ANN model were in agreement with the experimental data. By the developed ANN model, it would be possible to design and manufacture agricultural machinery in order to determine the weed density and reduce the herbicide use.Buğday üretiminde yabancı ot yoğunluğunun görüntü işleme teknikleri kullanılarak belirlenmesi Anahtar Sözcükler: Artificial neural networks Image processing Weeds Winter wheat ÖZET Tarımsal ilaçların insan sağlığı, çevre ve doğal dengeyi olumsuz yönde etkilemeleri ve artan üretim maliyetleri nedeniyle hassas, dikkatli, en az ilaç kaybına neden olacak şekilde ve gerektiği kadar uygulanması oldukça önem arz etmektedir. Çalışmada, buğday ekili arazilerde geniş yapraklı yabancı ot yoğunluğunun tespit edilmesi ve herbisit kullanımının azaltılmasına katkıda bulunulması amaçlanmıştır. Bu amaçla araştırmada Görüntü İşleme Teknikleri kullanılmış ve yabancı ot tespitine yönelik olarak Yapay Sinir Ağları (YSA) ve regresyon modelleri geliştirilmiştir. YSA modelinde, Görüntü İşleme Teknikleri ile elde edilen yabancı ot alanları giriş ve gerçek yabancı ot alanları çıkış parametresi olarak değerlendirilmiştir. Çalışmada eğitim için 244 ve test için 18 veri olmak üzere toplam 262 veri kullanılmıştır. Ağın yapısı 1-(9-5)-1 olacak şekilde, 1 giriş katmanı, 2 gizli katman ve 1 çıkış katmanı olarak dizayn edilmiş ve gizli katmanların nöron sayıları 9-5 olarak belirlenmiştir. Ayrıca birinci gizli katmanda tansig, ikinci gizli katmanda logsig, çıkış katmanında ise purelin transfer fonksiyonları kullanılmıştır. YSA ve Regresyon modelleri değerlendi...

show abstract

Section: Artificial Neural Networkmentioning

confidence: 99%

Determination of weed intensity in wheat production using image processing techniques

Ağın¹,

Taner²

2015

Anadolu J. Agr. Sci.

View full text Add to dashboard Cite

show abstract

“…Effectively, reference (Elanayar et al, 1990) utilised this experiment to relate tool wear, surface finish, and cutting forces and reported a high but unquantified achievement rate. Reference (Purushothaman and Srinivasa, 1994) approach and methodology is similar to that of (Rangwala and Dornfield, 1987), but differs in terms of sensor signal inputs, data processing. Reference (Purushothaman and Srinivasa, 1994) argues that there is no need to use dimensionally selected features, instead, they use the amplified components of the cutting force signal as sampled, together with the three cutting parameters (Rangwala andDornfield, 1987, Dornfeld, 1990).…”

mentioning

confidence: 99%

“…Reference (Purushothaman and Srinivasa, 1994) approach and methodology is similar to that of (Rangwala and Dornfield, 1987), but differs in terms of sensor signal inputs, data processing. Reference (Purushothaman and Srinivasa, 1994) argues that there is no need to use dimensionally selected features, instead, they use the amplified components of the cutting force signal as sampled, together with the three cutting parameters (Rangwala andDornfield, 1987, Dornfeld, 1990). Reference (Tanner and Loh, 1994) utilises the cutting forces as their sensor signal inputs with an additional component depiction the occurrence of tool breakage to monitor the cutting process.…”

mentioning

confidence: 99%

“…The same workpiece material, feed-rate and depth of cut are utilised in both test sets. Force-wear relationships partly based on the same principle as reported in (Purushothaman andSrinivasa, 1994, Elanayar et al, 1990) are employed to select the input coefficients for training the network. Results showed it to be capable of close flank wear estimation but poor crater wear estimation.…”

mentioning

confidence: 99%

“…Thus, the final classification results would not be entirely correct. References (Purushothaman and Srinivasa, 1994, Elanayar et al, 1990, Elanyar and Shin, 1995, Yao and Fang, 1993) also utilise the cutting force signals. The net output gave an indication of the product surface finish established earlier as representative of the tool state.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Rapid design of tool-wear condition monitoring systems for turning processes using novelty detection

Al-Azmi

Al‐Habaibeh

Redgate

2009

IJMTM

View full text Add to dashboard Cite

Condition monitoring systems of manufacturing processes have been recognised in recent years as one of the key technologies that provide the competitive advantage in many manufacturing environments. It is capable of providing essential means to reduce cost, increase productivity, improve quality and prevent damage to the machine or work-piece. Turning operations are considered one of the most common manufacturing processes in industry. It is used to manufacture different round objects such as shafts, spindles and pins. Despite recent development and intensive engineering research, the development of tool wear monitoring systems in turning is still on-going challenge. In this paper, force signals are used for monitoring tool wear in a feature fusion model. A novel approach for the design of condition monitoring systems for turning operations using novelty detection algorithm is presented. The results prove that the developed system can be used for rapid design of condition monitoring systems for turning operations to predict tool-wear.

show abstract

Modeling and prediction of weld shear strength in friction stir spot welding using design of experiments and neural network

Külekçi

Eşme²,

et al. 2011

Materialwissenschaft Werkst

View full text Add to dashboard Cite

Friction Stir Spot Welding (FSSW) is a kind of the friction stir welding (FSW) process, creates a spot, lap-weld without bulk melting work materials. The tensile shear strength of the FSSW welded joints mainly depends on the pin height, tool rotation and welding time. In the present study, two of the techniques, namely factorial design and neural network (NN) were used for modeling and predicting the tensile shear strength of EN AW 5005 aluminum alloy. Tensile shear strength was taken as a response variable measured after welding pin height, tool rotation and welding speed were taken as input parameters. Relationships between tensile shear strength and welding parameters have been investigated. The level of importance of the FSSW parameters on the tensile shear strength was determined by using the analysis of variance method (ANOVA). The mathematical relation between the tensile shear strength and FSSW welding parameters were established by regression analysis method. This mathematical model may be used in estimating the tensile shear strength of FSSW joints without performing any experiments. Finally, predicted values of tensile shear strength by techniques, NN and regression analysis, were compared with the experimental results and their nearness with the experimental values assessed. Results show that, NN is a good alternative to empirical modeling based on full factorial design.Keywords: Friction stir welding / neuronal network / modeling / design of experiment / Schlüsselwörter: Reibrührschweißen / neuronales Netzwerk / Modellierung / statistische Versuchsplanung /

show abstract

A back-propagation algorithm applied to tool wear monitoring

Cited by 60 publications

References 9 publications

Determination of weed intensity in wheat production using image processing techniques

Determination of weed intensity in wheat production using image processing techniques

Rapid design of tool-wear condition monitoring systems for turning processes using novelty detection

Modeling and prediction of weld shear strength in friction stir spot welding using design of experiments and neural network

Contact Info

Product

Resources

About