Number Of Input Variables Research Articles

Modelling and optimizing the impact resistance of engineered cementitious composites with Multiwalled carbon nanotubes using response surface methodology

Engineered Cementitious Composites (ECC) are highly regarded in construction owing to their tensile ductility and crack control capabilities, making them suitable for various structural applications. The accumulation of multi-walled carbon nanotubes (MWCNTs) further enhances their mechanical properties. However, there’s a significant knowledge gap concerning MWCNTs-ECC impact resistance. The objective of this study is to tackle the challenges associated with evaluating, optimizing, and predicting MWCNTs-ECC impact resistance to ensure its safe and widespread use in critical infrastructure by applying response surface methodology (RSM). Moreover, the 13 mixtures of ECC combined with several quantities of PVA fiber and MWCNTs as input elements were utilized to calculate the first (E1) and final (E2) impact energies. The findings demonstrated that the MWCNTs-ECC combinations’ impact resistance improved as the input ingredient concentrations increased. Besides, the optimum E1 and E2 of ECC combined with 1% of PVA fiber were noted by 1398 Joules and 12,956 Joules at 0.065% of MWCNTs on 28 days respectively. Furthermore, Response prediction models for E1 and E2 were created, and after being validated with an analysis of variance (ANOVA), it was determined that they had high R2 readings of 99.30% and 99.07%, correspondingly. The optimization process produced an ideal number of input variables for MWCNTs and PVA fiber, respectively, of 0.066% and 1%, with a desirability value of 100%. Moreover, it is recommended that the usage of 0.066% of MWCNTs in ECC combined with 1.0–1.50% PVA fiber provides optimum results for the construction industry.

An efficient single-loop strategy for time-variant reliability sensitivity analysis based on Bayes’ theorem

Time-variant reliability sensitivity analysis aims to measure the effect of input variables on the time-variant failure probability, which can then be used to guide structural design and optimization. The traditional methods for calculating the sensitivity index require a nested sampling procedure and their computational complexity depends on the number of input variables, making them computationally the most expensive. To address the above limitations, an efficient single-loop strategy is developed for time-variant reliability sensitivity (TRS) analysis. Based on Bayes' theorem, the TRS index is represented by the difference between the probability density function (PDF) and failure-conditional PDF of the variable. This derivation enables TRS analysis to be implemented using only a single set of samples, with computational costs that don’t depend on the number of input variables. Then, the sensitivity index is normalized to identify influential variables effectively. Several case studies involving numerical and engineering problems are employed to verify the feasibility and effectiveness of the proposed strategy. The direct Monte Carlo simulation is introduced to assess the accuracy of the obtained results. The results indicate that the proposed strategy provides acceptable sensitivity analysis for time-variant events, while greatly conserving computational resources and enhancing efficiency.

Explainable artificial intelligence for the interpretation of ensemble learning performance in algal bloom estimation.

Chlorophyll-a (Chl-a) concentrations, a key indicator of algal blooms, were estimated using the XGBoost machine learning model with 23 variables, including water quality and meteorological factors. The model performance was evaluated using three indices: root mean square error (RMSE), RMSE-observation standard deviation ratio (RSR), and Nash-Sutcliffe efficiency. Nine datasets were created by averaging 1hour data to cover time frequencies ranging from 1hour to 1month. The dataset with relatively high observation frequencies (1-24 h) maintained stability, with an RSR ranging between 0.61 and 0.65. However, the model's performance declined significantly for datasets with weekly and monthly intervals. The Shapley value (SHAP) analysis, an explainable artificial intelligence method, was further applied to provide a quantitative understanding of how environmental factors in the watershed impact the model's performance and is also utilized to enhance the practical applicability of the model in the field. The number of input variables for model construction increased sequentially from 1 to 23, starting from the variable with the highest SHAP value to that with the lowest. The model's performance plateaued after considering five or more variables, demonstrating that stable performance could be achieved using only a small number of variables, including relatively easily measured data collected by real-time sensors, such as pH, dissolved oxygen, and turbidity. This result highlights the practicality of employing machine learning models and real-time sensor-based measurements for effective on-site water quality management. PRACTITIONER POINTS: XAI quantifies the effects of environmental factors on algal bloom prediction models The effects of input variable frequency and seasonality were analyzed using XAI XAI analysis on key variables ensures cost-effective model development.

Evaluation of statistical and machine learning models using satellite data to estimate aboveground biomass: A study in Vietnam Tropical Forests

The combination of machine learning models with satellite imagery is becoming a popular data-modeling tool for biomass prediction, supporting land cover management. This study aims to select the most suitable model to estimate tropical forest aboveground biomass in Vietnam, helping to manage and monitor changes in biomass at regional and local scales. The study identified the optimal model for estimating forest aboveground biomass and minimizing the number of input variables while achieving satisfactory model performance. A total of 59 input variables, including topography, texture features, and vegetation indices, from satellite data were used in four non-parametric algorithms and a conventional parametric model, Artificial Neural Networks (ANN), Support Vector Machine (SVM), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Multiple Linear Regression (MLR) to predict biomass and evaluate changes aboveground biomass over 10 years in two tropical forests in Vietnam. The results indicated that all models had good estimation performance with R2 ranging from 0.615 to 0.754. For RF, MLR, and XGBoost, vegetation indices contributed the highest model weights, occupying 77.71% – 92.48%. For ANN and SVM, textural and topographic features were the majority of the model weights (73.74 – 96.36%). The RF model performed the best using 59 variables (R2 = 0.754, MAE = 78.5 Mg·ha−1, and %RMSE = 13.57%) and ten variables (R2 = 0.745, MAE = 85.8 Mg·ha−1, and %RMSE = 16.17%). The biomass map using the RF and ten variables achieved a good degree of fitting of 0.76, so it was suitable for managing and monitoring forest biomass in Vietnam. The results indicated a sharp decrease in the areas of dense and very dense forests from 2013 to 2021 and a gradual increase in 2023.

Understanding machine learning predictions of wastewater treatment plant sludge with explainable artificial intelligence.

This study investigates the use of machine learning (ML) models for wastewater treatment plant (WWTP) sludge predictions and explainable artificial intelligence (XAI) techniques for understanding the impact of variables behind the prediction. Three ML models, random forest (RF), gradient boosting machine (GBM), and gradient boosting tree (GBT), were evaluated for their performance using statistical indicators. Input variable combinations were selected through different feature selection (FS) methods. XAI techniques were employed to enhance the interpretability and transparency of ML models. The results suggest that prediction accuracy depends on the choice of model and the number of variables. XAI techniques were found to be effective in interpreting the decisions made by each ML model. This study provides an example of using ML models in sludge production prediction and interpreting models applying XAI to understand the factors influencing it. Understandable interpretation of ML model prediction can facilitate targeted interventions for process optimization and improve the efficiency and sustainability of wastewater treatment processes. PRACTITIONER POINTS: Explainable artificial intelligence can play a crucial role in promoting trust between machine learning models and their real-world applications. Widely practiced machine learning models were used to predict sludge production of a United States wastewater treatment plant. Feature selection methods can reduce the required number of input variables without compromising model accuracy. Explainable artificial intelligence techniques can explain driving variables behind machine learning prediction.

Uncertainty quantification of acoustic metamaterial bandgaps with stochastic material properties and geometric defects

Acoustic metamaterials are a subject of increasing study and utility. Through designed combinations of geometries with material properties, acoustic metamaterials can be built to arbitrarily manipulate acoustic waves for various applications. Despite the theoretical advances in this field, however, acoustic metamaterials have seen limited penetration into industry and commercial use. This is largely due to the difficulty of manufacturing the intricate geometries that are integral to their function and the sensitivity of metamaterial designs to material batch variability and manufacturing defects. Capturing the effects of stochastic material properties and geometric defects requires empirical testing of manufactured samples, but this can quickly become prohibitively expensive with higher precision requirements or with an increasing number of input variables. This paper demonstrates how uncertainty quantification techniques, and more specifically the use of polynomial chaos expansions and spectral projections, can be used to greatly reduce sampling needs for characterizing acoustic metamaterial dispersion curves. With a novel method of encoding geometric defects in a 1D, interpretable, resolution-independent way, our uncertainty quantification approach allows for both stochastic material properties and geometric defects to be considered simultaneously. Two to three orders of magnitude sampling reductions down to ∼100 and ∼101 were achieved in 1D and 7D input space scenarios, respectively. Remarkably, this reduction in sampling was possible while preserving accurate output probability distributions of the metamaterial performance characteristics (bandgap size and location).

Developing a crash severity model based on multi objective evolutionary feature selection approaches

The main purpose of this article is assessing the capability of multiobjective optimisation evolutionary algorithms (MOEAs) along with machine learning method on predicting the crash injury severity with the approach of reducing input variables. Dataset of crashes which were occurred in Iran was analysed over the period of 2016–2020. The data were classified into two classes in terms of injury severity: fatal and non-fatal datasets. At first, general prediction models were created by inserting 31 available input variables into support vector machine (SVM) method based on two imbalanced and obtained balanced datasets. A cluster based under sampling technique was used to obtained balanced data. Then evolutionary algorithms of two objectives optimisation algorithms namely non-dominated sorting genetic algorithm (NSGA-II), multi-objective evolutionary algorithm based on decomposition (MOEA/D), and pareto envelope-based selection algorithm (PESA-II) along with SVM were used to select features that significantly affect the severity of crashes and developed the simplified models. Then the simplified developed model was compared to the model which was prepared with random forest (RF) method in feature selection. It is shown that the simplified model developed by combined MOEAs and SVM is more appropriate to accurately predict the crash severity, considering the number of input variables in comparison to RF.

Groundwater salinity modeling and mapping using machine learning approaches: a case study in Sidi Okba region, Algeria.

The groundwater salinization process complexity and the lack of data on its controlling factors are the main challenges for accurate predictions and mapping of aquifer salinity. For this purpose, effective machine learning (ML) methodologies are employed for effective modeling and mapping of groundwater salinity (GWS) in the Mio-Pliocene aquifer in the Sidi Okba region, Algeria, based on limited dataset of electrical conductivity (EC) measurements and readily available digital elevation model (DEM) derivatives. The dataset was randomly split into training (70%) and testing (30%) sets, and three wrapper selection methods, recursive feature elimination (RFE), forward feature selection (FFS), and backward feature selection (BFS) are applied to train the data. The resulting combinations are used as inputs for five ML models, namely random forest (RF), hybrid neuro-fuzzy inference system (HyFIS), K-nearest neighbors (KNN), cubist regression model (CRM), and support vector machine (SVM). The best-performing model is identified and applied to predict and map GWS across the entire study area. It is highlighted that the applied methods yield input variation combinations as critical factors that are often overlocked by many researchers, which substantially impacts the models' accuracy. Among different alternatives the RF model emerged as the most effective for predicting and mapping GWS in the study area, which led to the high performance in both the training (RMSE = 1.016, R = 0.854, and MAE = 0.759) and testing (RMSE = 1.069, R = 0.831, and MAE = 0.921) phases. The generated digital map highlighted the alarming situation regarding excessive GWS levels in the study area, particularly in zones of low elevations and far from the Foum Elgherza dam and Elbiraz wadi. Overall, this study represents a significant advancement over previous approaches, offering enhanced predictive performance for GWS with the minimum number of input variables.

A Machine Learning Model for Predicting In-Hospital Mortality in Chinese Patients With ST-Segment Elevation Myocardial Infarction: Findings From the China Myocardial Infarction Registry.

Machine learning (ML) risk prediction models, although much more accurate than traditional statistical methods, are inconvenient to use in clinical practice due to their nontransparency and requirement of a large number of input variables. We aimed to develop a precise, explainable, and flexible ML model to predict the risk of in-hospital mortality in patients with ST-segment elevation myocardial infarction (STEMI). This study recruited 18,744 patients enrolled in the 2013 China Acute Myocardial Infarction (CAMI) registry and 12,018 patients from the China Patient-Centered Evaluative Assessment of Cardiac Events (PEACE)-Retrospective Acute Myocardial Infarction Study. The Extreme Gradient Boosting (XGBoost) model was derived from 9616 patients in the CAMI registry (2014, 89 variables) with 5-fold cross-validation and validated on both the 9125 patients in the CAMI registry (89 variables) and the independent China PEACE cohort (10 variables). The Shapley Additive Explanations (SHAP) approach was employed to interpret the complex relationships embedded in the proposed model. In the XGBoost model for predicting all-cause in-hospital mortality, the variables with the top 8 most important scores were age, left ventricular ejection fraction, Killip class, heart rate, creatinine, blood glucose, white blood cell count, and use of angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin II receptor blockers (ARBs). The area under the curve (AUC) on the CAMI validation set was 0.896 (95% CI 0.884-0.909), significantly higher than the previous models. The AUC for the Global Registry of Acute Coronary Events (GRACE) model was 0.809 (95% CI 0.790-0.828), and for the TIMI model, it was 0.782 (95% CI 0.763-0.800). Despite the China PEACE validation set only having 10 available variables, the AUC reached 0.840 (0.829-0.852), showing a substantial improvement to the GRACE (0.762, 95% CI 0.748-0.776) and TIMI (0.789, 95% CI 0.776-0.803) scores. Several novel and nonlinear relationships were discovered between patients' characteristics and in-hospital mortality, including a U-shape pattern of high-density lipoprotein cholesterol (HDL-C). The proposed ML risk prediction model was highly accurate in predicting in-hospital mortality. Its flexible and explainable characteristics make the model convenient to use in clinical practice and could help guide patient management. ClinicalTrials.gov NCT01874691; https://clinicaltrials.gov/study/NCT01874691.

Using ensemble machine learning and metaheuristic optimization for modelling the elastic modulus of geopolymer concrete

Geopolymer concrete emerges as a sustainable and durable alternative to conventional concrete, addressing its high carbon footprint and enhanced durability. The distinct properties of geopolymer concrete, governed by supplementary cementitious materials and alkaline activators, promise reduced environmental impact and improved structural resilience. However, its complex composition complicates the prediction of mechanical properties such as the elastic modulus, crucial for structural applications. This study introduces an innovative approach using the eXtreme Gradient Boosting (XGBoost) technique integrated with the multi-objective grey wolf optimizer to model the elastic modulus of geopolymer concrete. By dynamically selecting influential features and optimizing model accuracy, this methodology advances beyond traditional empirical models, which fail to capture the nonlinear interactions intrinsic to geopolymer concrete. Utilizing a comprehensive database gathered from extensive literature, 22 potential variables were examined that influence geopolymer concrete’s elastic modulus. After mitigating multicollinearity and optimizing hyperparameters via Bayesian optimization, six XGBoost models were developed with different combinations of input variables, revealing compressive strength and total water content as pivotal predictors. The findings illustrate the models’ precision, with the trade-off between prediction accuracy and model simplicity visualized through the relationship between the number of input variables and prediction error. The study culminates in a user-friendly graphical user interface that enables easy prediction of geopolymer concrete’s elastic modulus and fosters educational engagement. This interface, available online, underscores the practicality and accessibility of advanced machine learning predictions. Overall, this research not only provides a robust predictive framework for geopolymer concrete’s elastic modulus using optimized input variables but also enhances the understanding of its underlying determinants, contributing to the advancement of sustainable construction materials.

A Novel Feature Selection Approach-Based Sampling Theory on Grapevine Images Using Convolutional Neural Networks

AbstractFeature selection, reducing number of input variables to develop classification model, is an important process to reduce computational and modeling complexity and affects the performance of image process. In this paper, we have proposed new statistical approaches for feature selection based on sample selection. We have applied our new approaches to grapevine leaves data that possess properties of shape, thickness, featheriness, and slickness that are investigated in images. To analyze such kind of data by using image process, thousands of features are created and selection of features plays important role to predict the outcome properly. In our numerical study, convolutional neural networks have been used as feature extractors and then obtained features from the last average pooling layer to detect the type of grapevine leaves from images. These features have been reduced by using our suggested four statistical methods: simple random sampling, ranked set sampling, extreme ranked set sampling, moving extreme ranked set sampling. Then, selected features have been classified with artificial neural network and we obtained the best accuracy of 97.33% with our proposed approaches. Based on our empirical analysis, it has been determined that the proposed approach exhibits efficacy in the classification of grapevine leaf types. Furthermore, it possesses the potential for integration into various computational devices.

Damage detection of truss structures using meta-heuristic algorithms and optimized group method of data handling surrogate model

Optimization-based methods are increasingly being implemented for structural damage detection (SDD) problems through the minimization of the objective functions based on vibration data. Meanwhile, the challenge of high computational cost hinders the applied use of SDD methods, especially upon addressing large-scale structures. In this study, a two-stage damage detection approach is proposed for damage localization as well as extent quantification which reduces the computational time of SDD problems. In the first stage of the proposed framework, suspected locations of damage are identified by employing a residual force vector. Then in the second stage, damage extent of already identified elements will be assessed by model updating method through three different optimization algorithms, namely particle swarm optimization (PSO), differential evolution (DE), and enhanced colliding bodies optimization (ECBO). To spare time-consuming modal analysis in each iteration of optimization algorithms, a PSO-based group method of data handling (GMDH) polynomial neural network (PNN) surrogate model is proposed. Performance of GMDH PNN strongly depends on the input variables, the number of input variables, and the order of the polynomial. The PSO-based GMDH PNN, developed in this study, employs PSO for the optimum selection of these parameters. Furthermore, the efficiency of Elemental Energy Quotient Indicator (EEQI) as a new damage index proposed in this paper is compared with the approach based on the natural frequency-based index. This paper also deals with the incomplete modal data measured from limited number of sensors. This problem is addressed by employing the model reduction technique and solving the SDD problem with modal information obtained from the reduced model. Numerical examples show that the suggested PSO-based GMDH surrogate model is an accurate tool for various types of SDD problems, and it can provide promising results with a significant reduction in computational time.

Comparative performance of intensive care mortality prediction models based on manually curated versus automatically extracted electronic health record data

IntroductionBenchmarking intensive care units for audit and feedback is frequently based on comparing actual mortality versus predicted mortality. Traditionally, mortality prediction models rely on a limited number of input variables and significant manual data entry and curation. Using automatically extracted electronic health record data may be a promising alternative. However, adequate data on comparative performance between these approaches is currently lacking. MethodsThe AmsterdamUMCdb intensive care database was used to construct a baseline APACHE IV in-hospital mortality model based on data typically available through manual data curation. Subsequently, new in-hospital mortality models were systematically developed and evaluated. New models differed with respect to the extent of automatic variable extraction, classification method, recalibration usage and the size of collection window. ResultsA total of 13 models were developed based on data from 5,077 admissions divided into a train (80%) and test (20%) cohort. Adding variables or extending collection windows only marginally improved discrimination and calibration. An XGBoost model using only automatically extracted variables, and therefore no acute or chronic diagnoses, was the best performing automated model with an AUC of 0.89 and a Brier score of 0.10. DiscussionPerformance of intensive care mortality prediction models based on manually curated versus automatically extracted electronic health record data is similar. Importantly, our results suggest that variables typically requiring manual curation, such as diagnosis at admission and comorbidities, may not be necessary for accurate mortality prediction. These proof-of-concept results require replication using multi-centre data.

Hybrid deep model for brain age prediction in MRI with improved chi-square based selected features

Ageing and its related health conditions bring many challenges not only to individuals but also to society. Various MRI techniques are defined for the early detection of age-related diseases. Researchers continue the prediction with the involvement of different strategies. In that manner, this research intends to propose a new brain age prediction model under the processing of certain steps like preprocessing, feature extraction, feature selection, and prediction. The initial step is preprocessing, where improved median filtering is proposed to reduce the noise in the image. After this, feature extraction takes place, where shape-based features, statistical features, and texture features are extracted. Particularly, Improved LGTrP features are extracted. However, the curse of dimensionality becomes a serious issue in this aspect that shrinks the efficiency of the prediction level. According to the “curse of dimensionality,” the number of samples required to estimate any function accurately increases exponentially as the number of input variables increases. Hence, a feature selection model with improvement has been introduced in this paper termed an improved Chi-square. Finally, for prediction purposes, a Hybrid classifier is introduced by combining the models like Bi-GRU and DBN, respectively. In order to enhance the effectiveness of the hybrid method, Upgraded Blue Monkey Optimization with Improvised Evaluation (UBMOIE) is introduced as the training system by tuning the optimal weights in both classifiers. Finally, the performance of the suggested UBMIOE-based brain age prediction method was assessed over the other schemes to various metrics.

원전의 안전계통 설정치 불확도 평가에서의 중심극한정리 타당성 확인

In evaluating the uncertainty of nuclear power plant safety system setpoints, the central limit theorem(CLT) is often used when combining the uncertainties of each element. However, depending on the distribution characteristics of individual elements and the number of elements to be combined, there are cases where the central limit theorem cannot be applied. International standards are being revised to reflect this. In this paper, we describe various methodologies for calculating safety system setpoints and confirm whether the combined components follow the central limit theorem according to the distribution characteristics of the individual elements. The Monte Carlo simulation method is used as a basic tool to confirm this, and normality is verified using KS test in Python program library. Additionally, the variance characteristics of each distribution are examined and differences from the uniform distribution are analyzed. As a result of the analysis, in order for the combined component to follow the central limit theorem, if the distribution is concentrated on both sides compared to the uniform distribution, the number of input variables needs to be four or more. That is, when the number of input element uncertainty of the safety system setpoint exceeds 4, the combined uncertainty was evaluated to have normality regardless of the distribution of input element uncertainty. Therefore, it was confirmed that the method of applying CLT to calculate the coverage factor and then using it to obtain the expanded uncertainty was valid. In the case of a distribution concentrated toward the center compared to a uniform distribution, the number of variables less than 3 was required. The U type distribution, which is used to evaluate the uncertainty of wireless communication, showed characteristics closer to a uniform distribution than an inverted triangular distribution. Also, in the case of variance, it was found that the variance of the uniform distribution was larger than that of the linear distribution, confirming that this conformed to the maximum entropy principle.

Technology independent optimization when implementing sparse systems of disjunctive normal forms of Boolean functions in ASIC

Objectives. The problem of choosing the best methods and programs for circuit implementation as part of digital ASIC (Application-Specific Integrated Circuit) sparse systems of disjunctive normal forms (DNF) of completely defined Boolean functions is considered. For matrix forms of sparse DNF systems, the ternary matrix specifying elementary conjunctions contains a large proportion of undefined values corresponding to missing literals of Boolean input variables, and the Boolean matrix specifying the occurrences of conjunctions in DNF functions contains a large proportion of zero values.Methods. It is proposed to investigate various methods of technologically independent logical optimization performed at the first stage of logical synthesis: joint minimization of systems of functions in the DNF class, separate and joint minimization in classes of multilevel representations in the form of Boolean networks and BDD representations using mutually inverse cofactors, as well as the division of a system of functions into subsystems with a limited number of input variables and the method of block cover of DNF systems, focused on minimizing the total area of the blocks forming the cover.Results. When implementing sparse DNF systems of Boolean functions in ASIC, along with traditional methods of joint minimization of systems of functions in the DNF class, methods for optimizing multilevel representations of Boolean function systems based on Shannon expansions can be used for technologically independent optimization, while separate minimization and joint minimization of the entire system as a whole turn out to be less effective compared with block partitions and coatings of the DNF system and subsequent minimization of multilevel representations. Schemes obtained as a result of synthesis using minimized representations of Boolean networks often have a smaller area than schemes obtained using minimized BDD representations.Conclusion. For the design of digital ASIC, the effectiveness of combined approach is shown, when initially the block coverage programs of the DNF system is used, followed by the use of programs to minimize multilevel block representations in the form of Boolean networks minimized based on Shannon expansion.

Odor Impact Assessment via Dispersion Model: Comparison of Different Input Meteorological Datasets

Dispersion modeling is a useful tool for reproducing the spatial–temporal distribution of pollutants emitted by industrial sites, particularly in the environmental odor field. One widely used tool, accepted by regulatory agencies for environmental impact assessments, is the CALPUFF model, which requires a large number of input variables, including meteorological and orographical variables. The reliability of model results depends on the accuracy of these input variables. The present research aims to discuss a comparative study of odor dispersion modeling by initializing the CALMET meteorological processor with different input data: surface and upper air observational meteorological data, 3D prognostic data, and a blend of prognostic and measured data. Two distinct sources (a point and an area source) and two different simulation domains in Cuba and Italy are considered. The analysis of results is based on odor impact criteria enforced in some Italian regions by computing the 98th percentile of odor peak concentrations on an annual basis. For the area source, simulation results reveal that the ‘OBS’ and ‘HYBRID’ modes are largely comparable, whereas prognostic data tend to underestimate the odor concentrations, likely due to a reduced percentage of wind calms. For point sources, different input meteorological settings provide comparable results, with no significant differences.

Genetic programming expressions for effluent quality prediction: Towards AI-driven monitoring and management of wastewater treatment plants

Continuous effluent quality prediction in wastewater treatment processes is crucial to proactively reduce the risks to the environment and human health. However, wastewater treatment is an extremely complex process controlled by several uncertain, interdependent, and sometimes poorly characterized physico-chemical-biological process parameters. In addition, there are substantial spatiotemporal variations, uncertainties, and high non-linear interactions among the water quality parameters and process variables involved in the treatment process. Such complexities hinder efficient monitoring, operation, and management of wastewater treatment plants under normal and abnormal conditions. Typical mathematical and statistical tools most often fail to capture such complex interrelationships, and therefore data-driven techniques offer an attractive solution to effectively quantify the performance of wastewater treatment plants. Although several previous studies focused on applying regression-based data-driven models (e.g., artificial neural network) to predict some wastewater treatment effluent parameters, most of these studies employed a limited number of input variables to predict only one or two parameters characterizing the effluent quality (e.g., chemical oxygen demand (COD) and/or suspended solids (SS)). Harnessing the power of Artificial Intelligence (AI), the current study proposes multi-gene genetic programming (MGGP)-based models, using a dataset obtained from an operational wastewater treatment plant, deploying membrane aerated biofilm reactor, to predict the filtrated COD, ammonia (NH4), and SS concentrations along with the carbon-to-nitrogen ratio (C/N) within the effluent. Input features included a set of process variables characterizing the influent quality (e.g., filtered COD, NH4, and SS concentrations), water physics and chemistry parameters (e.g., temperature and pH), and operation conditions (e.g., applied air pressure). The developed MGGP-based models accurately reproduced the observations of the four output variables with correlation coefficient values that ranged between 0.98 and 0.99 during training and between 0.96 and 0.99 during testing, reflecting the power of the developed models in predicting the quality of the effluent from the treatment system. Interpretability analyses were subsequently deployed to confirm the intuitive understanding of input-output interrelations and to identify the governing parameters of the treatment process. The developed MGGP-based models can facilitate the AI-driven monitoring and management of wastewater treatment plants through devising optimal rapid operation and control schemes and assisting the plants’ operators in maintaining proper performance of the plants under various normal and disruptive operational conditions.

Machine learning technology for predicting intraocular lens power. Diagnostics data generalization

Backgraund. Implantation of recent IOLs allows ophthalmologists to effectively solve the problems of surgical rehabilitation of patients with cataracts. The degree of improving the patient's visual functions is directly dependent on the accuracy of the preoperative calculation of the optical power of IOLs. The most famous formulas to calculate this indicator are SRK II, SRK/T, Hoffer-Q, Holladay II, Haigis, Barrett. All of them work well for an "average patient", however, they are not adequate enough at the boundaries of input variables ranges.
 Aims. To study the possibility of using mathematical models obtained as a result of deep learning of artificial neural networks (ANN models) to generalize data and predict the optical power of modern intraocular lenses.
 Materials and methods. ANN models were trained on large-scale samples, including depersonalized data for patients of ophthalmology clinic. Data provided in 2021 by ophthalmologist K.K. Syrykh, and reflect the results of both preoperative and postoperative observation of patients. The source file used to build the ANN model includes 455 records - 26 columns of input factors and one column for the output factor - calculating IOL (diopters). To conveniently build ANN models, we used a simulator program, previously developed by the authors.
 Results. The resulting models, in contrast to the traditionally used formulas, reflect the regional specificity of patients to a much greater extent. They also make it possible to retrain and optimize the structure based on newly received data, which allows taking into account the non-stationarity of the object. A distinctive feature of such ANN models in comparison with the well-known formulas SRK II, SRK/T, Hoffer-Q, Holladay II, Haigis, Barrett, widely used in the surgical cataract treatment in ophthalmology, is their ability to take into account a significant number of recorded input quantities, which makes it possible to reduce the mean relative error in calculating the optical power of IOL from 10–12% to 3.5%.
 Conclusions. This work shows the fundamental possibility of generalizing a significant amount of empirical data on calculating the optical power of the IOL using training ANN models that have a significantly larger number of input variables than when using traditional formulas and methods. The results obtained allow, in principle, to build an intelligent expert system with a continuous flow of new data from a source and a step-by-step retraining of the ANN model.

UNFIS: A Novel Neuro-Fuzzy Inference System with Unstructured fuzzy rules

An important constraint of Fuzzy Inference Systems (FIS) is their structured rules. Indeed, each rule should evaluate all input variables in its antecedent part. Thus, the length of all fuzzy rules and the number of input variables are equal. However, in many decision-making problems evaluating some conditions on a limited set of input variables is sufficient to decide properly (unstructured rules). Consequently, the FIS’s performance, generalizability, and interpretability are restricted by the aforementioned limitation. This study provides a neuro-fuzzy inference system that can create each fuzzy rule taking into account various sets of input variables in order to handle this problem. To realize this capability, a new fuzzy selector neuron with an adaptive parameter is proposed that can select input variables in the antecedent part of each fuzzy rule. Moreover, the consequent part of the Takagi–Sugeno–Kang FIS is properly changed to consider only the selected set of input variables. To learn the parameters of the proposed architecture, the Levenberg–Marquardt (LM) method is used for regression problems. For classification applications, a trust-region-based learning method (General quasi-Levenberg–Marquardt (GqLM)) is proposed to minimize cross-entropy in multiclass problems. The performance of the proposed method is compared with some related previous approaches in some real-world classification and regression problems. Based on these comparisons, the proposed method has better or very close performance with a parsimonious structure consisting of unstructured fuzzy rules.