MKMunish Kumar

Munish Kumar

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Munish Kumar has a total of 19 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Forecasting ambient PM2.5 and PM10 concentrations in Hisar City through machine learning

    by Munish Kumar, Parveen Sihag, Sergij Vambol

    The accurate prediction of particulate matter concentrations serves as the foundation for effective air quality management. Hisar City, which resides in north-west India, experiences high levels of PM2.5 and PM10 pollution because of vehicular and industrial emissions, construction and farming activities, and weather conditions. The research tests machine learning models to forecast PM2.5 and PM10 levels in Hisar City. Historical air quality data together with weather data for a period of 2020-2024 were gathered and an exploratory data analysis was performed to study the seasonal behaviour of particulate matter for the representative location. For modelling analysis, the periodic data from the year 2020 to 2023 was used for model training and the data from the year 2024 was used for testing. The research used Support Vector Regression (SVM), Random Forest (RF), Gradient Boosting, AdaBoost and M5 regression to create and test multiple machine learning models. For model explainability, SHAP analysis along with the sensitivity were performed to study the role and contribution of weather variables on the PM concentration estimation. The researchers used correlation coefficient (CC), root mean squared error (RMSE), and mean absolute error (MAE) as statistical indicators to assess the model performance. The study results showed that ensemble-based models (Gradient Boosting and RF) provide better results than the other regression models for predicting PM concentrations of the city. Correlation coefficient of 0.73 and 0.668 was achieved by Gradient Boosting model for PM2.5 and PM10 prediction along with the least RMSE values during testing. The RF model achieved CC values of 0.724 and 0.674 for PM2.5 and PM10 prediction, respectively, during testing. The proposed method shows that machine learning techniques can be used to predict air quality reliably, which helps policymakers develop effective emergency response plans.

  • Performance evaluation of hybrid constructed wetlands for nitrogen removal and statistical approaches

    by Suresh Kumar, Vikramaditya Sangwan, Munish Kumar, Shweta Shweta, Shivani Shivani, Manoj Kumar, Surinder Deswal

    Nitrogen pollution in water bodies has become a pressing environmental and public health issue worldwide, demanding the implementation of effective nitrogen removal strategies. This research paper delves into the performance evaluation of hybrid constructed wetlands (HCWs) as a sustainable and innovative approach for nitrogen removal, employing a comprehensive year‐long dataset gathered from a practical setup. The study collected data under diverse operating conditions to investigate the effectiveness of HCWs in removing nitrogen. Results revealed that HCWs achieved nitrogen removal efficiencies ranging from 28% to 65%, influenced by temperature and hydraulic retention time. Optimal removal occurred at an average temperature of 28°C and a 4‐day hydraulic retention time. Notably, performance declined during colder periods, with temperatures below 15°C. The study also aims to predict nitrogen removal by three modeling techniques, that is, artificial neural networks (ANNs), support vector machines Pearson VII kernel function (SVM PUK), and multiple linear regression (MLR). Prediction has been done considering temperature (TEMP), hydraulic loading rate (HLR), initial concentration of chemical oxygen demand (COD) (CODin), initial concentration of total nitrogen (TNin), initial concentration of total phosphorous (TPin), and initial concentration of turbidity (TBin) as input parameters, whereas reduction of total nitrogen (RED TN) is regarded as output parameter. The performance of the soft computing techniques has been compared in terms of coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). The analysis revealed that the performance of the SVM (PUK) model (R2: 0.572, RMSE: 0.0359, MAE: 0.0294) for the prediction of TN reduction is superior followed by MLR (R2: 0.562, RMSE: 0.0365, MAE: 0.0294) and ANN (R2: 0.597, RMSE: 0.0377, MAE: 0.0301). The present study concludes that the treated effluent by the HCWs, using water hyacinth and water lettuce, is of fair quality, thus having potential application for the treatment of rice mill wastewater in warmer climates. Further, machine learning approaches employed in estimating the total nitrogen reduction by HCWs technology have shown promising applicability and utilization in such studies.Practitioner Points: Hybrid constructed wetlands (HCWs) are effective in removing nitrogen from wastewater. The performance of HCWs in nitrogen removal can vary due to physical, chemical, and biological processes. The performance of the HCWs highly depends on temperature and hydraulic retention time. Artificial neural networks (ANNs) and support vector machines (SVMs) provided better predictions of nitrogen removal with high accuracy and low root mean square error.


  • Application of Machine Learning Methods in Estimating the Oxygenation Performance of Various Configurations of Plunging Hollow Jet Aerators

    by Munish Kumar, N. K. Tiwari, Subodh Ranjan

    Plunging jet aerators are considered energetically attractive devices for oxygenation because of their good mixing characteristics and ease of construction and operation. In this mechanism of plunging jet aeration, the air-water interfacial area is increased by a free-falling jet impinging on the surface of a water pool. In this study, experimental data from various configurations of plunging hollow jet aerators are explored in formulating the correlations for predicting the values of volumetric oxygen transfer coefficient (KLa ) with the jet variables (discharge, jet thickness, jet velocity, jet length, depth of water pool, pipe outlet diameter, number of jets). Nonlinear regression modeling equations derived from dimensional and nondimensional data sets are compared with the neuro-fuzzy (ANFIS), support vector regression (SVM), artificial neural network (ANN), M5 tree (M5), and random forests (RF) methods. SVM models calibrated with both types of data sets provided better results when tested on the unseen data sets. Regression equations are also useful and give acceptable results. The SVM models and regression equations are further checked for effectiveness on the data set of past study on plunging hollow jets. The nondimensional form of the regression equation derived in the current study fits reasonably well when tested on the oxygenation data from previous work as compared to the other regression models. The sensitivity of the jet variables is also tested, which showed jet velocity and jet thickness as major contributing factors in oxygenating the aeration pools. An aeration device adds oxygen to water and wastewater during treatment to ensure that microorganisms have enough oxygen to oxidize organic pollutants. Aeration requirements for aeration tanks vary greatly depending on the organic load, and one or more hollow jet aerators can be fitted to accommodate the fluctuating load. Hollow jet aerators are well suited to use in packaged wastewater treatment plants. The annular opening of these aerators can be adjusted accordingly based on the varying load. High-velocity thinner jets are better for quickly renewing air-water films and generating shear and turbulence in the receiving pool, but wastewater should be well screened before aeration to avoid clogging of the small annular opening. The results of this study on the capacity of plunging hollow jet aerators of different configurations/arrangements at varying operating conditions were reflected by the dimensional as well as nondimensional relationships between oxygen transfer rate and jet equipment parameters. Implementing these relationships on previous jet aeration data indicates that the equations are accurate enough to be used to situations similar to those in this investigation. Without any prior experimental effort, these equations can be used to estimate the oxygen transfer expected in practical applications.


  • Predicting the infiltration characteristics for semi-arid regions using regression trees

    by Parveen Sihag, Munish Kumar, Saad Sh. Sammen

    Could advances in geoinformatics, irrigation management and climate adaptive agronomic practices ensure the sustainability of water supply in agriculture? This book comprises 33 chapters that contribute to a broad discussion and demonstration of state-of-the-art multifunctional role of water resources in agriculture. The aim of the book to provide insights into novel modelling (monitoring, analyzing/visualizing and prediction) approaches, irrigation management and agronomic practices to investigate the adaptability of water supply and crop production systems to changing environment. The book presents characteristic examples of new technologies and decision support systems (e.g., artificial intelligence/optimization modelling approaches, Big Geo data) in water efficiency at different levels, including: water supply hydraulic infrastructure systems, water retention measures, less exposed to evaporation and better adapted to infiltration, solutions to reduce water demand and developing techniques for reusing water. In Focus–a book series that showcases the latest accomplishments in water research. Each book focuses on a specialist area with papers from top experts in the field. It aims to be a vehicle for in-depth understanding and inspire further conversations in the sector.

  • Soft computing based predictive modelling of oxygen transfer performance of plunging hollow jets

    by Munish Kumar, N. K. Tiwari, Subodh Ranjan

    Bubbles are formed as the plunging water jet passes through atmosphere and impinges on the surface of water in the pool, which increases the oxygen level of water. The amount of oxygen transferred into an aeration system can be altered by the flow characteristics of plunging jet. The present study models and simulates the oxygen transfer properties with basic flow characteristics of plunging hollow jets. Variables related to the jet included in this study are jet thickness, jet velocity, jet length, and depth of water. For the estimation of volumetric oxygen transfer coefficient ( KLa), modelling techniques such as multiple nonlinear regression (MNLR), artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), multivariate adaptive regression splines (MARS), and generalized regression neural network (GRNN) are used. The results are presented based on training and testing of the models. A nonlinear relationship proposed for the estimation of KLa is compared with the other soft computing-based approaches. The overall comparison of the results obtained from the application of modelling techniques in estimating the experimental data of plunging hollow jets yielded better prediction accuracy by ANFIS (bell-shaped membership function) as well as ANN as compared to MNLR, MARS, and GRNN.


  • Open Access

    Assessment of infiltration models developed using soft computing techniques

    by Parveen Sihag, Munish Kumar, Balraj Singh

    In this study, predicting ability of support vector machines (SVM), Gaussian process (GP), artificial neural network (ANN), and Random forests (RF) based regression approaches was tested on the infiltration data of soil samples having different compositions of sand, silt, clay, and fly ash. In addition to this, their performances were compared with the Kostiakov model (KM) and Philip’s model (PM). Dataset containing a total of 392 observations was collected from the experimental measurements of soil infiltration rate on different soil samples. Out of the total dataset, 272 recordings were randomly selected for training and the residual 120 observations were selected for validation of the developed models. Standard statistical parameters were used to measure the predicting ability of various developed models. The result suggests that the best performance could be achieved by Polynomial kernel function-based GP regression (GP_Poly) with coefficient of correlation values as 0.9824, 0.9863, Bias values as 0.0006, −2.3542, root-mean-square error values as 47.7336, 40.3026, and Nash Sutcliffe model efficiency values as 0.9655, 0.9727 using training and testing dataset, respectively. Furthermore, time is found as the most influencing input variable for predicting the infiltration rate when GP_Poly-based model is used to predict the infiltration rate.


  • Open Access

    Experimental study and modelling discharge coefficient of trapezoidal and rectangular piano key weirs

    by Munish Kumar, Parveen Sihag, N. K. Tiwari, Subodh Ranjan

    Crest length is an important parameter in influencing the discharge handling capacity of a weir. Nonlinear weirs with longer crests are cost effective alternatives for those existing dam structures which are more susceptible to failure due to loss of storage capacity by reservoir silting problem, and insufficiency of the structure in evacuating the updated flow due to the limited space. Piano key weir is a type of nonlinear weir designed in the form of piano keys, over-hanged from both the upstream and the downstream with sloping floors founded on a base or footprint. These weirs can be easily placed over gravity dams due to smaller footprint than labyrinth weirs. The present study’s focus is on the comparative analysis of identical configurations of trapezoidal and rectangular piano key (PK) weirs. The importance of (crest length to width) L/W ratio and weir height (P) in affecting the discharge efficiency of both types of PK weirs is investigated in the experimental study. Furthermore, soft computing approaches are applied to the current data set obtained from both types of weirs by considering discharge coefficient ( $$C_{\text{d}}$$Cd ) as a function of dimensionless geometric variables of PK weirs. The modelling performance of random forest regression and M5 tree approach is tested in order to estimate the values of discharge coefficient. The results conclude higher predictive accuracy of random forest model over M5 tree model.