Performance Comparison of Supervised Machine Learning Algorithms for Multiclass Transient Classification in a Nuclear Power Plant

Abstract. Probability distribution functions (PDFs) of model inputs that affect the transport and dispersion of a trace gas released from a coastal California nuclear power plant are quantified using ensemble simulations, machine learning algorithms, and Bayesian inversion. The PDFs are constrained by observations of tracer concentrations and account for uncertainty in meteorology, transport, diffusion, and emissions. Meteorological uncertainty is calculated using an ensemble of simulations of the Weather Research and Forecasting (WRF) model that samples five categories of model inputs (initialization time, boundary layer physics, land surface model, nudging options, and reanalysis data). The WRF output is used to drive tens of thousands of FLEXPART dispersion simulations that sample a uniform distribution of six emissions inputs. Machine learning algorithms are trained on the ensemble data, and used to quantify the sources of ensemble variability and to infer, via inverse modeling, the values of the 11 model inputs most consistent with tracer measurements. We find a substantial ensemble spread in tracer concentrations (factors of 10 to 103), most of which is due to changing emissions inputs (about 80 %), though the cumulative effects of meteorological variations are not negligible. The performance of the inverse method is verified using synthetic observations generated from arbitrarily selected simulations. When applied to measurements from a controlled tracer release experiment, the most likely inversion results are within about 200 meters of the known release location, 5 and 50 minutes of the release start and duration times, respectively, and 22 % of the release amount. The inversion also estimates probabilities of different combinations of WRF inputs of matching the tracer observations.

Download Full-text

Bayesian inverse modeling of the atmospheric transport and emissions of a controlled tracer release from a nuclear power plant

Atmospheric Chemistry and Physics ◽

10.5194/acp-17-13521-2017 ◽

2017 ◽

Vol 17 (22) ◽

pp. 13521-13543 ◽

Cited By ~ 11

Author(s):

Donald D. Lucas ◽

Matthew Simpson ◽

Philip Cameron-Smith ◽

Ronald L. Baskett

Keyword(s):

Machine Learning ◽

Power Plant ◽

Nuclear Power Plant ◽

Inverse Modeling ◽

Nuclear Power ◽

Inverse Method ◽

Learning Algorithms ◽

Reanalysis Data ◽

Machine Learning Algorithms ◽

Start Time

Abstract. Probability distribution functions (PDFs) of model inputs that affect the transport and dispersion of a trace gas released from a coastal California nuclear power plant are quantified using ensemble simulations, machine-learning algorithms, and Bayesian inversion. The PDFs are constrained by observations of tracer concentrations and account for uncertainty in meteorology, transport, diffusion, and emissions. Meteorological uncertainty is calculated using an ensemble of simulations of the Weather Research and Forecasting (WRF) model that samples five categories of model inputs (initialization time, boundary layer physics, land surface model, nudging options, and reanalysis data). The WRF output is used to drive tens of thousands of FLEXPART dispersion simulations that sample a uniform distribution of six emissions inputs. Machine-learning algorithms are trained on the ensemble data and used to quantify the sources of ensemble variability and to infer, via inverse modeling, the values of the 11 model inputs most consistent with tracer measurements. We find a substantial ensemble spread in tracer concentrations (factors of 10 to 103), most of which is due to changing emissions inputs (about 80 %), though the cumulative effects of meteorological variations are not negligible. The performance of the inverse method is verified using synthetic observations generated from arbitrarily selected simulations. When applied to measurements from a controlled tracer release experiment, the inverse method satisfactorily determines the location, start time, duration and amount. In a 2 km × 2 km area of possible locations, the actual location is determined to within 200 m. The start time is determined to within 5 min out of 2 h, and the duration to within 50 min out of 4 h. Over a range of release amounts of 10 to 1000 kg, the estimated amount exceeds the actual amount of 146 kg by only 32 kg. The inversion also estimates probabilities of different WRF configurations. To best match the tracer observations, the highest-probability cases in WRF are associated with using a late initialization time and specific reanalysis data products.

Download Full-text

Performance Improvement of Decision Tree: A Robust Classifier Using Tabu Search Algorithm

Applied Sciences ◽

10.3390/app11156728 ◽

2021 ◽

Vol 11 (15) ◽

pp. 6728

Author(s):

Muhammad Asfand Hafeez ◽

Muhammad Rashid ◽

Hassan Tariq ◽

Zain Ul Abideen ◽

Saud S. Alotaibi ◽

...

Keyword(s):

Machine Learning ◽

Tabu Search ◽

Decision Tree ◽

Decision Trees ◽

Search Algorithm ◽

Learning Algorithms ◽

Performance Comparison ◽

Machine Learning Algorithms ◽

Supervised Machine Learning ◽

Tabu Search Algorithm

Classification and regression are the major applications of machine learning algorithms which are widely used to solve problems in numerous domains of engineering and computer science. Different classifiers based on the optimization of the decision tree have been proposed, however, it is still evolving over time. This paper presents a novel and robust classifier based on a decision tree and tabu search algorithms, respectively. In the aim of improving performance, our proposed algorithm constructs multiple decision trees while employing a tabu search algorithm to consistently monitor the leaf and decision nodes in the corresponding decision trees. Additionally, the used tabu search algorithm is responsible to balance the entropy of the corresponding decision trees. For training the model, we used the clinical data of COVID-19 patients to predict whether a patient is suffering. The experimental results were obtained using our proposed classifier based on the built-in sci-kit learn library in Python. The extensive analysis for the performance comparison was presented using Big O and statistical analysis for conventional supervised machine learning algorithms. Moreover, the performance comparison to optimized state-of-the-art classifiers is also presented. The achieved accuracy of 98%, the required execution time of 55.6 ms and the area under receiver operating characteristic (AUROC) for proposed method of 0.95 reveals that the proposed classifier algorithm is convenient for large datasets.

Download Full-text