Modelling of Uncertain System: A comparison study of Linear and Non-Linear Approaches

Abba, S.I.; Gaya, Muhammad Sani; Yakubu, Mahadi Lawan; Zango, Muttaqa Uba; Abdulkadir, Rabiu Aliyu; Saleh, M.A.; Hamza, Amina; Abubakar, Ukashatu; Tukur, Abubakar Ibrahim; Wahab, Norhaliza Abdul

doi:10.1109/i2cacis.2019.8825085

Cited by 16 publications

(12 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The general structure of HW model composed of the application of a black-box process, which is developed purposely to predict the non-linearity [16]. The architecture of HW model consists of three blocks namely; static output non-linear blocks, a static input nonlinear block and a linear dynamic block [17,18]. HW mechanism involves the conversion of the nonlinear functions into linear input blocks, which are subsequently transferred back into the non-linear form again inform of output [17][18][19].…”

Section: Hammerstein-wiener Model (Hw)mentioning

confidence: 99%

Advanced chromatographic technique for performance simulation of anti-Alzheimer agent: an ensemble machine learning approach

et al. 2020

Self Cite

View full text Add to dashboard Cite

This work employs the application of three artificial intelligence (AI) techniques namely; support vector machine (SVM), Hammerstein-Wiener (HW) and multi-layer perceptron (MLP) for predicting the qualitative properties of an anti-Alzheimer agent using high-pressure liquid chromatography technique. The mobile phase (inform of acetonitrile and trifluoroacetic acid) and the column temperature was used as the predictors in modelling the maximum retention time (tR-max) and resolution (Resol.) as the output variables of the analyte. The measured and predicted values were checked using three performance indices including; Nash-Sutcliffe efficiency (NSE), correlation coefficient (CC) as the goodness of fits and a statistical error inform of root-mean-square error (RMSE). The results obtained demonstrated the promising ability of AI-based models in modelling the qualitative properties of the anti-Alzheimer agent. Observation of different outputs of the AI-based models at various time intervals showed the necessity of ensembling the outputs of the AI-based models. Therefore, simple average ensemble and support vector machine ensemble (SVM-E) were employed to enhance the performance skills of the simple models. The comparative performance of SVM-E inform of NSE indicated its ability in boosting and enhancing the performance skills of the single models SVM, MLP and HW models up to 5, 13 and 20% respectively in the testing stage for modelling tR-max.

show abstract

Section: Hammerstein-wiener Model (Hw)mentioning

confidence: 99%

Advanced chromatographic technique for performance simulation of anti-Alzheimer agent: an ensemble machine learning approach

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Generally, model efficiency performance should include at least one goodness of fit and at least one prediction error metrics 50 . Based on this determination coefficient (R 2 ), correlation coefficient (R), root mean square error (RMSE) and mean absolute percentage error (MAPE) are chosen as models appraisal metrics of the soft computing methods.…”

Section: Data Pre-processing Model Validation and Performance Metricsmentioning

confidence: 99%

Multi-Response Optimization and Prediction of Tribological behaviours of Filled Polytetrafluoroethylene Composites Using Taguchi-Deng and Evolutionary Harris Hawk’s Optimization Methodologies

Ibrahim

Çamur

Savas

et al. 2022

Preprint

View full text Add to dashboard Cite

Filled polymer composites are widely used for automobile, structural and aerospace components owing to their exceptional combination of high specific stiffness and strength. This study presents Taguchi-Deng optimization of tribological parameters such as load, grit size, distance and speed as well as prediction of tribological behaviours of carbon-filled and bronze-filled polytetrafluoroethylene (PTFE) composites using pin on disk configuration. A plan of experiments based on Taguchi L27 (43) orthogonal array (OA) was designed to collect data in a controlled manner. The Taguchi L27 (43) was hybridized with Deng model to produce grey relational grades (GRG) for the multiple response optimization. Analysis of variance (ANOVA) was executed to establish the parameters affecting GRG of the composites. For the prediction of the tribological behaviours of the composites namely coefficient of friction (µ) and specific wear rate (Ks), support vector regression (SVR) was coupled with novel Harris Hawks’ optimization (HHO) and swarm particle optimization (PSO) forming SVR-HHO and SVR-PSO models respectively, were employed. Prediction accuracy of the models were appraised using coefficient of determination (R2), correlation coefficient (R), root mean square error (RMSE) and mean absolute percentage error (MAPE). GRG results revealed that optimum parameters which reduced tribological behaviours were factor combination L3G1D3S3. ANOVA for GRG reveled that grit size with 68.57% ranks as the most influential parameter followed by load with 20.57%, followed by distance having a contribution of 7.78% and finally speed with least contribution of 3.38% for minimum tribological loss. Validation performed using optimum parameters revealed an enhancement of 55% in GRG. Prediction accuracy of the single model increase to 19.50% and 57.08% on the average for hybrid µ and Ks models, respectively. Furthermore, SVR-HHO model indicated the higher prediction accuracy of the tribological behaviours of filled PTFE composites as compared to SVR-PSO model. These findings concluded these metaheuristic models are promising in predicting tribological behaviours of filled PTFE composites and thus can serve as a guide in the design and development of tribological materials.

show abstract

“…The degree of uncertainty and randomness that builds the stochastic process of an AR model makes it commonly used in time series simulations [6,32,33]. Base on the prior variables value knowledge, the AR model forecasts the value of the future.…”

Section: Autoregressive Modelmentioning

confidence: 99%

Effluents quality prediction by using nonlinear dynamic block-oriented models: a system identification approach

Abba¹,

Abdulkadir²,

Gaya³

et al. 2021

DWT

View full text Add to dashboard Cite

The dynamic and complex municipal wastewater treatment plant (MWWTP) process should be handled efficiently to safeguard the excellent quality of effluents characteristics. Most of the available mathematical models do not efficiently capture the MWWTP process, in such cases, the datadriven models are reliable and indispensable for effective modeling of effluents characteristics. In the present research, two nonlinear system identification (NSI) models namely; Hammerstein-Wiener model (HW) and nonlinear autoregressive with exogenous (NARX) neural network model, and a classical autoregressive (AR) model were proposed to predict the characteristics of the effluent of total suspended solids (TSS eff ) and pH eff from Nicosia MWWTP in Cyprus. In order to attain the optimal models, two different combinations of input variables were cast through auto-correlation function and partial auto-correlation analysis. The prediction accuracy was evaluated using three statistical indicators the determination coefficient (DC), root mean square error (RMSE) and correlation coefficient (CC). The results of the appraisal indicated that the HW model outperformed NARX and AR models in predicting the pH eff , while the NARX model performed better than the HW and AR models for TSS eff prediction. It was evident that the accuracy of the HW increased averagely up to 18% with regards to the NARX model for pH eff . Likewise, the TSS eff performance increased averagely up to 25% with regards to the HW model. Also, in the validation phase, the HW model yielded DC, RMSE, and CC of 0.7355, 0.1071, and 0.8578 for pH eff , while the NARX model yielded 0.9804, 0.0049 and 0.9902 for TSS eff , respectively. For comparison with the traditional AR, the results showed that both HW and NARX models outperformed in (TSS eff ) and pH eff prediction at the study location. Hence, the outcomes determined that the NSI model (i.e., HW and NARX) are reliable and resilient modeling tools that could be adopted for pH eff and TSS eff prediction.

show abstract

Modelling of Uncertain System: A comparison study of Linear and Non-Linear Approaches

Cited by 16 publications

References 22 publications

Advanced chromatographic technique for performance simulation of anti-Alzheimer agent: an ensemble machine learning approach

Advanced chromatographic technique for performance simulation of anti-Alzheimer agent: an ensemble machine learning approach

Multi-Response Optimization and Prediction of Tribological behaviours of Filled Polytetrafluoroethylene Composites Using Taguchi-Deng and Evolutionary Harris Hawk’s Optimization Methodologies

Effluents quality prediction by using nonlinear dynamic block-oriented models: a system identification approach

Contact Info

Product

Resources

About