Search

Soft sensors are mathematical models that estimate hard-to-measure variables given easy-to-measure ones. This field of study has given the industry a valuable tool to enable a better control of different plants and processes. With such models, quality indicators and other variables that usually demand costly or slow sensors can be predicted in real time. However, one important factor for application in industry is model interpretability. Moreover, regression models have an inherent trade-off between bias and variance, which is not considered by usual learning algorithms. Therefore, this work proposes a novel multi-objective least squares based on evolutionary feature selection and regularization, which incorporates feature selection and regularization, as well as the search for a better trade-off between bias and variance, using evolutionary multi-objective optimization algorithms. Experiments on two famous soft sensor benchmark problems, the debutanizer column and the sulfur recovery unit, indicate that the proposed algorithm can train more robust interpretable linear models than the commonly used partial least squares and least absolute shrinkage and selection operator. • Feature selection and regularization is addressed for soft sensor development. • A holistic multi-objective optimization design procedure is employed for the task. • An evolutionary multi-objective optimization algorithm builds stronger soft sensors. • The proposed method builds stronger interpretable models than other tested algorithms. • Case studies include a debutanizer column and a sulfur recovery unit. [ABSTRACT FROM AUTHOR]

Streamflow series forecasting composes a fundamental step in planning electric energy production for hydroelectric plants. In Brazil, such plants produce almost 70% of the total energy. Therefore, it is of great importance to improve the quality of streamflow series forecasting by investigating state-of-the-art time series forecasting algorithms. To this end, this work proposes the development of ensembles of unorganized machines, namely Extreme Learning Machines (ELMs) and Echo State Networks (ESNs). Two primary contributions are proposed: (1) a new training logic for ESNs that enables the application of bootstrap aggregation (bagging); and (2) the employment of multi-objective optimization to select and adjust the weights of the ensemble's base models, taking into account the trade-off between bias and variance. Experiments are conducted on streamflow series data from five real-world Brazilian hydroelectric plants, namely those in Sobradinho , Serra da Mesa , Jiraú , Furnas and Água Vermelha. The statistical results for four different prediction horizons (1, 3, 6, and 12 months ahead) indicate that the ensembles of unorganized machines achieve better results than autoregressive (AR) models in terms of the Nash–Sutcliffe model efficiency coefficient (NSE), root mean squared error (RMSE), coefficient of determination (R 2), and RMSE-observations standard deviation ratio (RSR). In such results, the ensembles with ESNs and the multi-objective optimization design procedure achieve the best scores. • Streamflow forecasting of five hydroelectric plants is addressed. • Ensembles of Echo State Networks and Extreme Learning Machines are employed. • A new Echo State Network training logic is proposed to enable bootstrap aggregation. • Multi-objective optimization is used for selecting and weighting base models. • Results confirm the advantages of using the proposed technique. [ABSTRACT FROM AUTHOR]

• A new taxonomy for multi-objective ensemble learning is proposed. • A holistic study on multi-objective ensemble learning is performed. • A collection of imbalanced data sets is used for comparison purposes. • An imbalanced real-world drinking-water quality anomaly detection is solved. • Results indicate the success of multi-objective ensemble learning. Ensemble learning methods have already shown to be powerful techniques for creating classifiers. However, when dealing with real-world engineering problems, class imbalance is usually found. In such scenario, canonical machine learning algorithms may not present desirable solutions, and techniques for overcoming this problem must be used. In addition to using learning algorithms that alleviate the imbalance between classes, multi-objective optimization design (MOOD) approaches can be used to improve the prediction performance of ensembles of classifiers. This paper proposes a study of different MOOD approaches for ensemble learning. First, a taxonomy on multi-objective ensemble learning (MOEL) is proposed. In it, four types of existing approaches are defined: multi-objective ensemble member generation, multi-objective ensemble member selection, multi-objective ensemble member combination, and multi-objective ensemble member selection and combination. Additionally, new approaches can be derived by combining the previous ones, such as multi-objective ensemble member generation and selection, multi-objective ensemble member generation and combination and multi-objective ensemble member generation, selection and combination. With the given taxonomy, two experiments are conducted for comparing (1) the performance of the MOEL techniques for generating and aggregating base models on several imbalanced benchmark problems and (2) the performance of MOEL techniques against other machine learning techniques in a real-world imbalanced drinking-water quality anomaly detection problem. Finally, results indicate that MOOD is able to improve the predictive performance of existing ensemble learning techniques. [ABSTRACT FROM AUTHOR]

The aim is to develop soft sensors (SSs) to provide an estimation of the laundry moisture of clothes introduced in a household Heat Pump Washer–Dryer (WD-HP) appliance. The developed SS represents a cost-effective alternative to physical sensors, and it aims at improving the WD-HP performance in terms of drying process efficiency of the automatic drying cycle. To this end, we make use of appropriate Machine Learning models, which are derived by means of Regularization and Symbolic Regression methods. These methods connect easy-to-measure variables with the laundry moisture content, which is a difficult and costly to measure variable. Thanks to the use of SSs, the laundry moisture estimation during the drying process is effectively available. The proposed models have been tested by exploiting real data through an experimental test campaign on household drying machines. [ABSTRACT FROM AUTHOR]

In the last few years, machine learning techniques have been successfully applied to solve engineering problems. However, owing to certain complexities found in real-world problems, such as class imbalance, classical learning algorithms may not reach a prescribed performance. There can be situations where a good result on different conflicting objectives is desirable, such as true positive and true negative ratios, or it is important to balance model's complexity and prediction score. To solve such issues, the application of multi-objective optimization design procedures can be used to analyze various trade-offs and build more robust machine learning models. Thus, the creation of ensembles of predictive models using such procedures is addressed in this work. First, a set of diverse predictive models is built by employing a multi-objective evolutionary algorithm. Next, a second multi-objective optimization step selects the previous models as ensemble members, resulting on several non-dominated solutions. A final multi-criteria decision making stage is applied to rank and visualize the resulting ensembles. To analyze the proposed methodology, two different experiments are conducted for binary classification. The first case study is a famous classification problem through which the proposed procedure is illustrated. The second one is a challenging real-world problem related to water quality monitoring, where the proposed procedure is compared to four classical ensemble learning algorithms. Results on this second experiment show that the proposed technique is able to create robust ensembles that can outperform other ensemble methods. Overall, the authors conclude that the proposed methodology for ensemble generation creates competitive models for real-world engineering problems. • A new multi-objective optimization approach for ensemble generation is proposed. • The methodology is composed of two separate multi-objective problems. • A final multi-criteria decision making step is applied to select the final ensemble. • The proposed methodology is applied on a real-world competition problem. • Results show that the proposed methodology outperforms classical ensemble methods. [ABSTRACT FROM AUTHOR]

This paper proposes two novel machine learning algorithms, namely Random Vector Functional Link Forests and Extreme Learning Forests, to develop an improved unmanned aerial vehicles automatic target recognition system. Such models take advantage of the stochastic procedure followed by Random Forests, where random subsets of instances and features are selected to build diverse Decision Trees. However, different from the usual uni-variate split criterion from Decision Tree algorithms, we propose and employ the novel Random Vector Functional Link Tree or Extreme Learning Tree, where each decision split is performed using the fast non-linear mapping of multiple features provided by either Random Vector Functional Link or Extreme Learning Machines. To prove the efficacy of the novel algorithm, experiments are performed using 90 binary classification problems to compare the performance of the proposed algorithm against other state-of-the-art ensemble learning techniques. Statistical analysis indicates the success of the proposed algorithms in terms of both predictive performance and computational complexity. While the model with deeper trees outperforms classical ensembles in terms of predictive performance (1.41% error reduction) and has similar results to state-of-the-art ensemble models, the model with shallow trees outperforms all ensembles in terms of computational burden (at least 36% faster). Finally, the novel methods are applied to develop an automatic target recognition system for unmanned aerial vehicles, achieving a valuable trade-off in terms of area under receiver operating characteristic curve (0.9309), F 1 score (0.8190), accuracy (0.8646), and computational time (4.14 s). [ABSTRACT FROM AUTHOR]

The provision of clean and safe drinking water is a crucial task for water supply companies from all over the world. To this end, automatic anomaly detection plays a critical role in drinking water quality monitoring. Recent anomaly detection studies use techniques that focus on a single global objective. Yet, companies need solutions that better balance the trade-off between false positives (FPs), which lead to financial losses to water companies, and false negatives (FNs), which severely impact public health and damage the environment. This work proposes a novel dynamic multi-criteria ensemble selection mechanism to cope with both problems simultaneously: the non-dominated local class-specific accuracy (NLCA). Moreover, experiments rely on recent time series related classification metrics to assess the predictive performance. Results on data from a real-world water distribution system show that NLCA outperforms other ensemble learning and dynamic ensemble selection techniques by more than 15% in terms of time series related F 1 scores. As a conclusion, NLCA enables the development of stronger anomaly detection systems for drinking water quality monitoring. The proposed technique also offers a new perspective on dynamic ensemble selection, which can be applied to different classification tasks to balance conflicting criteria. Unlabelled Image • The solution for a real-world drinking water anomaly detection problem is presented. • Feature engineering and dynamic ensemble selection are explored to solve the task. • A novel multi-criteria dynamic ensemble selection algorithm is proposed. • The new algorithm outperforms all other tested dynamic ensemble selection methods. [ABSTRACT FROM AUTHOR]