Misfire Detection in Spark Ignition Engine Using Transfer Learning.

Misfire detection in an internal combustion engine is an important activity. Any undetected misfire can lead to loss of fuel and power in the automobile. As the fuel cost is more, one cannot afford to waste money because of the misfire. Even if one is ready to spend more money on fuel, the power of the engine comes down; thereby, the vehicle performance falls drastically because of the misfire in IC engines. Hence, researchers paid a lot of attention to detect the misfire in IC engines and rectify it. Drawbacks of conventional diagnostic techniques include the requirement of high level of human intelligence and professional expertise in the field, which made the researchers look for intelligent and automatic diagnostic tools. There are many techniques suggested by researchers to detect the misfire in IC engines. This paper proposes the use of transfer learning technology to detect the misfire in the IC engine. First, the vibration signals were collected from the engine head and plots are made which will work as input to the deep learning algorithms. The deep learning algorithms have the capability to learn from the plots of vibration signals and classify the state of the misfire in the IC engines. In the present work, the pretrained networks such as AlexNet, VGG-16, GoogLeNet, and ResNet-50 are employed to identify the misfire state of the engine. In the pretrained networks, the effect of hyperparameters such as back size, solver, learning rate, and train-test split ratio was studied and the best performing network was suggested for misfire detection.

ANN modelling of sediment concentration in the dynamic glacial environment of Gangotri in Himalaya.

The present study explores for the first time the possibility of modelling sediment concentration with artificial neural networks (ANNs) at Gangotri, the source of Bhagirathi River in the Himalaya. Discharge, rainfall and temperature have been considered as the main controlling factors of variations in sediment concentration in the dynamic glacial environment of Gangotri. Fourteen feed forward neural networks with error back propagation algorithm have been created, trained and tested for prediction of sediment concentration. Seven models (T1-T7) have been trained and tested in the non-updating mode whereas remaining seven models (T1a-T7a) have been trained in the updating mode. The non-updating mode refers to the scenario where antecedent time (previous time step) values are not used as input to the model. In case of the updating mode, antecedent time values are used as network inputs. The inputs applied in the models are either the variables mentioned above as individual factors (single input networks) or a combination of them (multi-input networks). The suitability of employing antecedent time-step values as network inputs has hence been checked by comparative analysis of model performance in the two modes. The simple feed forward network has been improvised with a series parallel non-linear autoregressive with exogenous input (NARX) architecture wherein true values of sediment concentration have been fed as input during training. In the glacial scenario of Gangotri, maximum sediment movement takes place during the melt period (May-October). Hence, daily data of discharge, rainfall, temperature and sediment concentration for five consecutive melt periods (May-October, 2000-2004) have been used for modelling. High Coefficient of determination values [0.77-0.88] have been obtained between observed and ANN-predicted values of sediment concentration. The study has brought out relationships between variables that are not reflected in normal statistical analysis. A strong rainfall: sediment concentration and temperature: sediment concentration relationship is shown by the models which are not reflected in statistical correlation. It has also been observed that usage of antecedent time-step values as network inputs does not necessarily lead to improvement in model performance.

Spatio-temporal variation in the hydrochemistry of Tawa River, Central India: effect of natural and anthropogenic factors.

Tawa River is the biggest left bank tributary of the Narmada, the largest west-flowing river of the Indian peninsula. Central India enjoys a tropical climate, is highly urbanized, and the river flow is mostly controlled by monsoon; a large part of the population depend on rivers for their livelihood. Spatial and temporal variations in the hydrochemistry of the Tawa River were studied based on seasonal sampling along the course of the river and its tributaries. The study is important because not much data exist on small size rivers and the river processes spell out correctly in smaller basins. The monsoon season accounts for more than 70% of river water flow. The basin is characterized by silicate lithology; however, water chemistry is controlled by carbonate-rich soils and other weathering products of the silicate rocks, as indicated by the high (Ca + Mg)/(Na + K) ratios (>3.8). The values of the Na-normalized ratios of Ca(2+), Mg(2+), and HCO3(-) suggest that both the carbonate and silicate lithology contribute to the hydrochemistry. On average, 42% of HCO3(-) in the Tawa River water is contributed by silicate weathering and 58% from carbonate lithology. The water remains undersaturated with respect to calcite during the monsoon and post-monsoon seasons and supersaturated during the pre-monsoon season. A significant influence of mining in the basin and other industrial units is observed in water chemical composition.

Trace metals biogeochemistry of Kumaun Himalayan Lakes, Uttarakhand, India.

The increasing urbanization, along with tourism, has posed a major threat to the Kumaun Himalayan Lakes, Uttarakhand, India. The total metal concentration in the water, interstitial water, and sediments along with the metal fractionation studies were carried out to understand the remobilization of the trace metals from the sediments of the lakes. The high concentration of the metals in the water column of the lakes generally decreases with depth and the metals release from the sediment is mainly due to the prevalence of anoxic condition at the sediment-water interface and sediment column. The sediment shows that metals Fe and Cr are derived from detrital source, whereas Co, Ni, and Zn are derived mainly from the organic matter dissolution. The sparse correlation of the trace metals with Ti shows most of the metals have chiefly re-precipitated from the water column. The metals speciation studies also supports that metals experience a high rate of anoxic dissolution and their precipitation onto the sediments are determined by the sediment composition and organic matter content. The high concentration of manganese in the interstitial water in the lakes indicates dissolution of organic matter. The released manganese is adsorbed/precipitated as carbonate phase (Nainital Lake) and oxide pahse (in other lakes). The study shows that the trace metals are regenerated from the sediments due to oxyhydroxide dissolution and organic matter decomposition.

Enrichment and fractionation of heavy metals in bed sediments of River Narmada, India.

A metal fractionation study on bed sediments of River Narmada in Central India has been carried out to examine the enrichment and partitioning of different metal species between five geochemical phases (exchangeable fraction, carbonate fraction, Fe/Mn oxide fraction, organic fraction and residual fraction). The river receives toxic substances through a large number of tributaries and drains flowing in the catchment of the river. The toxic substances of particular interest are heavy metals derived from urban runoff as well as municipal sewage and industrial effluents. Heavy metals entering the river get adsorbed onto the suspended sediments, which in due course of time settle down in the bottom of the river. In this study fractionation of metal ions has been carried out with the objective to determine the eco-toxic potential of metal ions. Although, in most cases (except iron) the average trace/heavy metal concentrations in sediments were higher than the standard shale values, the risk assessment code as applied to the present study reveals that only about 1-3% of manganese, <1% of copper, 16-19% of nickel, 4-20% of chromium, 1-4% of lead, 8-13% of cadmium and 1-3% of zinc exist in exchangeable fraction and therefore falls under low to medium risk category. According to the Geo-accumulation Index (GAI), cadmium shows high accumulation in the river sediments, rest of other metals are under unpolluted to moderately polluted class.

Heavy metals fractionation in Ganga River sediments, India.

The Ganga River is the largest river in India which, originates in the Himalayas and along with the Brahmaputra River, another Himalayan river, transports enormous amounts of sediments from the Indian sub-continent to the Bay of Bengal. Because of the important role of river sediments in the biogeochemical cycling of elements, the Ganga river sediments, collected from its origin to the down stretches, were studied in the present context, to assess the heavy metals associated with different chemical fractions of sediments. The fractionation of metals were studied in the sediments using SM&T protocol for the extraction of heavy metals and geo-accumulation index (GAI) (Muller, Schwermetalle in den sedimenten des rheins - Veranderungen seit. Umschau, 79, 778-783, 1979) and Metal Enrichment Factor (MEF) in different fractions were calculated. As with many river systems, residual fractions constitute more than 60% of total metals, except Zn, Cu and Cr. However, the reducible and organic and sulfide components also act as major sinks for metals in the down stretches of the river, which is supported by the high GAI and MEF values. The GAI values range between 4 and 5 and MEF exceed more than 20 for almost all the locations in the downstream locations indicating to the addition of metals through urban and industrial effluents, as compared to the low metals concentrations with less GAI and MEF in the pristine river sediments from the rivers in Himalayas.

Preconcentration of traces of gold, silver and palladium on activated carbon and its determination in geological samples by flame AAS after wet ashing.

A simple, rapid and cost effective preconcentration method is described for the determination of traces of gold (Au), silver (Ag) and palladium (Pd) in geological samples by flame atomic absorption spectrophotometry (AAS). The method is based on sorption of analytes (Au, Ag and Pd) on powdered activated carbon (AC) at pH approximately 1 in hydrochloric acid (HCl) medium. The samples are decomposed by aqua regia - HCl treatment. The quantitative sorption (better than 92%) of analytes is obtained on AC (in absence of complexing agent), by simply manipulating optimal conditions. The unique feature of the method is, the analytes are recovered quantitatively from AC by oxidizing and completely solubilising the carbon using concentrated nitric acid (HNO(3)) and perchloric acid (HClO(4)) i.e. by wet ashing. The method of wet ashing has several advantages over conventional dry ashing. The accuracy of the method is evaluated by analysing, five Canada centre for mineral and energy technology (CANMET) standards; MA-3, MA-1b, FER-1, SU-1A, and CPB-1. In addition, ASK-3 and one inhouse standard, kolar gold field (KGF) samples was also analysed. As no standard for Pd is available, its accuracy was evaluated by standard addition method. The method was applied on numerous geological samples for the determination of Au, Ag and Pd down to 0.1 ppm (based on 10 g sample) within +/-10% R.S.D. (n=5). The method could easily be adopted by any laboratory as the inputs are minimal (AC), inexpensive and easily available.

Fractionation of heavy metals and phosphorus in suspended sediments of the Yamuna river, India.

The Yamuna river is the largest tributary of the Ganges river system. It originates in the Himalayas and flows through a varied geological terrain encompassing a large basin area. Metals Fe, Mn, Pb, Zn, Cu in different chemical fractions of suspended sediments such as exchangeable, carbonates, Fe-Mn oxides, organics and residual fractions were studied. Phosphorus associated with different chemical forms are discussed. The metals are mostly associated with residual fractions in the sediments followed by organics, Fe-Mn oxides, exhangeable and carbonates. Intensive use of chemical fertilizers and pesticides in agriculture in the basin affects the high inorganic phosphorus content in sediments.

Spectrophotometric method for the determination of vanadium in uranium rich hydrogeochemical samples using pyridyl azo resorcinol (PAR).

A simple and precise method has been developed for the determination of traces of vanadium(V), using 4-(2 pyridyl azo) resorcinol, in natural water samples, containing very high concentrations of uranium. CDTA-pyrophosphate buffer has been used for masking interferants, including uranium which otherwise interferes above 125 ppb. The reaction of vanadium with PAR in the presence of buffer requires a waiting period of 45 min. The Sandell sensitivity of the method is 0.003 microg/ml, at 545 nm at an optimum pH of 6.5 +/- 0.2. The precision of the method is +/- 15% at the 100 ppb level of vanadium(V). The method has been successfully applied to a number of natural water samples during hydrogeochemical exploration.