Application Of Gene Expression Programming Research Articles

Application of Gene Expression Programming (GEP) in Power Transformers Fault Diagnosis Using DGA

Accurate diagnosis of incipient faults in oil-filled power transformers is important in preventive maintenance of transformers. Dissolved gas analysis (DGA) is an effective tool to diagnose incipient transformer faults. The majority of the methods reported in literature to analyze DGA results lay more emphasis on user experience rather than mathematical formulation/justification. Furthermore, sometimes DGA results for a certain fault do not belong to any of the IEC/IEEE standard and cannot be categorized/diagnosed. To address these issues, we propose a new approach for DGA interpretation using gene expression programming (GEP). The proposed approach is employed for analysis of 552 DGA samples collected from transformers of Himachal Pradesh State Electricity Board, India, in conjunction with samples extracted from reliable literature. We use the aforementioned dataset to test and validate our proposed GEP model. We also compare the performance of our approach against other artificial intelligence-based techniques such as artificial neural network, fuzzy-logic, and support vector machine. Results and comparison against other soft computing approaches show relative superiority of GEP-based DGA interpretation in terms of classification accuracy.

Application of gene expression programming to predict daily dew point temperature

In the present research, gene expression programming (GEP) was used to estimate daily dew point temperature (Tdew) in Tabriz and Urmia which are located in the northwest of Iran, using the recorded data sets for the period of 1998–2012. Accordingly, three types of scenarios namely temperature-based, multiple parameters-based and periodicity-based were considered to develop the GEP models. Various combinations of meteorological parameters including minimum air temperature (Tmin), maximum air temperature (Tmax), mean air temperature (T), actual vapor pressure (ea), relative humidity (RH) and atmospheric pressure (Pa) were used as inputs in the first and second scenarios. Also, root mean square error (RMSE), mean absolute error (MAE), Nash-Sutcliffe efficiency (NSE) and coefficient of determination (R2) were employed to investigate the models’ accuracy. The results showed that the ea is the most effective parameter in estimating Tdew. In the second part of the present study, Tdew of each station was estimated using meteorological data of the adjacent station. It is found that the meteorological data of the adjacent station can be used to predict Tdew. Moreover, the results of validation stage generally confirmed the outcomes of testing stage.

GEP to predict characteristics of a hydraulic jump over a rough bed

This study presents new application of Gene Expression Programming (GEP) to predict characteristics of a hydraulic jump over a rough bed. Performances of the GEP model compared with traditional equations and common artificial intelligence techniques (ANN and SVR). Published data were used from the literature for the hydraulic jump characteristics over a rough bed. The prediction uncertainties of all of the models are quantified and compared. Results of this study showed that the GEP model is able to predict the characteristics of a hydraulic jump over a rough bed with acceptable precision (RMSE = 0.29, R2 = 0.975 for sequent depth ratio (y2y1) for test data and RMSE = 4.804, R2 = 0.8 for hydraulic jump length ratio Ljy1 for test data). Comparison of GEP and traditional equations demonstrate that GEP models can represent more accurate equations for practical applications (traditional equations performance RMSE = 0.338, R2 = 0.967 for sequent depth ratio and RMSE = 5.72, R2 = 0.712 for jump length ratio). Comparison of common artificial intelligence techniques (ANN and SVR) indicated that the performance of these models is slightly better than GEP model but application of GEP model due to derivation of explicit equations is easier for practical purposes.

Computer-aided modeling for predicting layer thickness of a duplex treated ceramic coating on tool steels

Five different tool steels (DIN 1.2080, 1.2210, 1.2344, 1.2510 and 1.3343) have been targeted for a duplex surface treatment consisted of nitriding followed by vanadium thermo-reactive diffusion (TRD). TRD process was performed in molten salt bath at 575, 650 and 725°C for 1 to 15h. A duplex ceramic coating of vanadium carbonitride (VCN) with a thickness up to 10.2μm was formed on tool steel substrates. Characterization of the ceramic coating by means of scanning electron microscopy (SEM) and X-ray diffraction analysis (XRD) indicated that the diffused compact and dense layers mainly consisted of V(C,N) and V2(C,N) phases. Layer thickness of duplex coating has been modeled by gene expression programming (GEP). Recently, application of GEP as a computer-aided technique has got appreciable attraction especially for modeling and to formulate engineering demands. For GEP approaches, chemical composition of steel substrates along with different bath and processing parameters totally composed of 17 different parameters were considered as inputs to establish mathematical correlations. Finally, the training and testing results in models have shown strong potential for predicting the layer thickness of duplex treated ceramic coating on tool steels.

Application of gene expression programming and neural networks to predict adverse events of radical hysterectomy in cervical cancer patients.

The aim of this article was to compare gene expression programming (GEP) method with three types of neural networks in the prediction of adverse events of radical hysterectomy in cervical cancer patients. One-hundred and seven patients treated by radical hysterectomy were analyzed. Each record representing a single patient consisted of 10 parameters. The occurrence and lack of perioperative complications imposed a two-class classification problem. In the simulations, GEP algorithm was compared to a multilayer perceptron (MLP), a radial basis function network neural, and a probabilistic neural network. The generalization ability of the models was assessed on the basis of their accuracy, the sensitivity, the specificity, and the area under the receiver operating characteristic curve (AUROC). The GEP classifier provided best results in the prediction of the adverse events with the accuracy of 71.96 %. Comparable but slightly worse outcomes were obtained using MLP, i.e., 71.87 %. For each of measured indices: accuracy, sensitivity, specificity, and the AUROC, the standard deviation was the smallest for the models generated by GEP classifier.

Application of gene expression programming for proton–proton interactions at large hadrons collider

This paper describes how to use gene expression programming (GEP) as an evolutionary computational optimization approach. GEP, as a machine learning technique is usually used for modeling physical phenomena by discovering a new function. In case of modeling the p–p interactions at large hadrons collider (LHC) experiments, GEP is used to simulate and predict the number of charged particles multiplicity $$ \left\langle n \right\rangle $$ and total cross-section, σ T , as a function of total center-of-mass energy, $$ \sqrt s $$ . Considering the discovered function for $$ \left\langle n \right\rangle $$ ( $$ \sqrt s $$ ), the general trend of the predicted values shows good agreement at LHC [predicted values are 31.3251, 32.8638 and 35.3520 at $$ \sqrt s $$ = 8 TeV, $$ \sqrt s $$ = 10 TeV and $$ \sqrt s $$ = 14 TeV respectively]. The discovered function, trained on experimental data of particle data group shows a good match as compared with the other models. The predicted values of cross section at $$ \sqrt s $$ = 8, 10 and 14 TeV are found to be 101.0417, 105.0690 and 111.3407 mb respectively. Moreover, those predicted values are in good agreement with those reported by Block, Cudell and Nakamura.

RETRACTED ARTICLE: Application of gene expression programming to predict the compressive damage of lightweight aluminosilicate geopolymer

In the present study, compressive strength of lightweight aluminosilicate geopolymers produced by fine fly ash and rice husk bark ash together with palm oil clinker (POC) aggregates has been modeled by gene expression programming. To build the model, training and testing by using experimental results from 144 specimens were conducted. The used data in the models are arranged in a format of six input parameters that cover the quantity of fine POC particles, the quantity of coarse POC particles, the quantity of FA + RHBA mixture, the ratio of alkali activator to ashes mixture, the age of curing, and the test trial number. According to these input parameters, in the gene expression programming models, the compressive strength of each specimen was predicted. The best value of R2 and the minimum values of root mean square error (RMSE) and absolute percentage error (MAPE) are 0.9669, 2.583, and 1.984, respectively, all in training phase. The minimum value of R2 and the maximum values of RMSE and MAPE are 0.9456, 3.067, and 2.356, respectively, all in testing phase. The training and testing results in the models have shown a strong potential for predicting the compressive strength of the lightweight geopolymer specimens in the considered range and one may predict them with a tiny error.