The Central Mechanical Engineering Research Institute (also known as CSIR-CMERI Durgapur or CMERI Durgapur) is a public engineering research and development institution in Durgapur, West Bengal, India. It is a constituent laboratory of the Indian Council of Scientific and Industrial Research (CSIR). This institute is the only national level research institute in the field of mechanical engineering in India.The CMERI was founded in February 1958 under the endorsement of the CSIR. It was founded to develop national mechanical engineering technology, particularly in order to help Indian industries. During its first decade, the CMERI mainly focused its efforts towards national technology and import substitution. Currently, the Institute is making R&D efforts in the front-line areas of research such as Robotics, Mechatronics, Microsystem, Cybernetics, Manufacturing, Precision agriculture, Embedded system, Near net shape manufacturing and Biomimetics. Besides conducting research, the Institute works towards different R&D based mission mode programs of country to provide suitable technological solutions for poverty alleviation, societal improvement, energy security, food security, aerospace, mining, automobile and defense.
This study investigates the influence of laser power, scan speed, and hatch spacing on the mechanical and tribological behavior of Direct Metal Laser Sintering (DMLS)-fabricated SS316L using a Taguchi L27 experimental design combined with Grey Relational Analysis (GRA). The results show that increasing laser power and reducing hatch spacing significantly improved densification, achieving a relative density above 99
This study presents an integrated artificial neural network (NN)—based predictive modelling and meta-heuristic optimisation framework for multi-response optimisation of wire electro-discharge machining (Wire EDM) parameters during machining of Hastelloy C-276. Due to its high strength and low thermal conductivity, Hastelloy C-276 is difficult to machine using conventional techniques, necessitating advanced optimisation strategies. Six process parameters—pulse-on time, pulse-off time, arc-on time, arc-off time, wire feed, and servo voltage—were investigated using a Taguchi L27 orthogonal array. Four key performance responses were evaluated: material removal rate (MR), kerf width (WK), surface roughness (S), and recast layer thickness (RLT). An NN model was developed to predict machining responses, and its weights were optimised using four meta-heuristic algorithms: Genetic Algorithm (G-A), Particle Swarm Optimisation (P-S-O), Grey Wolf Optimisation (G-W-O), and Whale Optimisation Algorithm (W-O-A). Model performance was assessed using R2, RMSE, and MAPE metrics. Among the hybrid models, NN-G-W-O demonstrated superior predictive accuracy and convergence stability. Comparative analysis with static multi-objective decision-making techniques (M-O-O-R-A and AHP) indicated improved robustness of the hybrid optimisation framework within the investigated parameter space. Sensitivity analysis and Pareto optimisation further clarified parameter influence and trade-offs among responses. The proposed framework provides a reliable and systematic approach for multi-response optimisation in advanced machining applications.
Hydrogen inevitably proves to be a promising alternative fuel source. In this era of climatic crisis, hydrogen extraction in the purest form of energy is crucial from natural sources such as biomass. Thermo-chemical conversion of biomass is a potential route of hydrogen production from biomass. Gasification is mostly studied by the researchers for hydrogen-rich syngas production from biomass, but it suffers from the problem of producing a significant amount of impurities along with syngas. To overcome this, two-stage pyrolysis reforming is gaining interest since the last decade. Although the two-stage pyrolysis reforming process gives a better yield of hydrogen, it is costlier as reforming is done at higher temperature, and steam is added to the process. Therefore, the present investigation is focused on enhancing the production of syngas and elevating the hydrogen concentration in syngas from single-stage pyrolysis of rice husk by altering the pyrolysis condition and using suitable catalyst without addition of steam. The results obtained from the experiments reveal that slow pyrolysis in isothermal conditions is suitable for higher syngas yield. The presence of catalyst increases the overall production of pyrolysis vapor and hydrogen percentage in syngas. The pyrolysis vapor production increases from 57.20 to 71.04%, and hydrogen production increases from 15.31 to 40.24% while Ni-zeolite catalyst is used. Thus, single-stage catalytic slow pyrolysis of biomass could be a promising cost-effective alternative for cleaner hydrogen production from biomass.
A phosphor-in-glass (P-i-G) exhibiting orange-red emission, composed of Sm3+ ion-doped CaAl4O7, was synthesized utilizing a multicomponent tellurite glass system via melting quenching technique. Absence of sharp Bragg peaks indicates the amorphous nature of the fabricated P-i-G. Fourier transform infrared spectroscopy (FTIR) and Raman studies, confirmed the existence of characteristic vibrational modes corresponding to the B-O and Te–O bonds. Through the optical microscopy and scanning electron microscope (SEM), the presence of phosphor material was observed inside the glass matrix. Strong emission bands of Sm3+ ions were identified from the P-i-G sample using the photoluminescence (PL) measurement. The P-i-G demonstrates significant thermal and power stabilities, maintaining 65.5
The detailed characterization of solid carbonaceous fuels (SCFs) is crucial for optimal production of second-generation fuels through the thermochemical conversion process. Hence, a comprehensive machine learning (ML) modelling approach is presented for accurately predicting the higher heating value (HHV) of various solid carbonaceous fuels (SCFs). To encompass a wide and diverse range of SCFs, 3771 samples comprising 16 distinct fuel types, categorised into 5 major fuel classes, are used here. Six supervised ML models are used for predicting the HHV and their performance is evaluated. The multilayer perceptron (MLP) has emerged as the most effective model with the highest prediction accuracy among all the ML models. The study proposed a novel two-stage method for predicting the HHV from easily accessible proximate analysis (PA) data. Here, ultimate analysis (UA) data are first estimated from experimentally measured PA data, and then, HHV is predicted from the combination of PA and estimated UA data. The two-stage model shows high accuracy (R2 = 0.99, MAE = 0.48, and MSE = 0.68) compared to the previous studies reported in the literature. The robustness and generalisation of the model are validated through bootstrapping-based uncertainty analysis. Shapley additive explanations (SHAP) is employed to assess the global and local interpretability by analysing the summary plot, violin plot, and force plot. The applied interpretable ML framework driven by a large and diverse dataset offers a scientifically robust and cost-effective solution for accurately estimating the HHV of a wide range of SCFs from easily available PA data.
