Nenad Bolf

Heat and mass transfer models in convection drying of clay slabs

Abstract Drying kinetics data for convection drying of industrially prepared roof tile clay slab were approximated with different mathematical models. Applied conventional model analysis enables evaluation of main transport properties: effective diffusion coefficient, mass and heat transfer coefficients, thermal conductivity, drying constant, and exponential model parameters. A neural network-based drying model was established using backpropagation algorithm for dynamics modelling of moisture content and temperature of thin clay sample. Obtained results confirm the assumption that both, the heat and mass transfers, are under external conditions. Very small values of Biot numbers confirm that fact. Drying air temperature and initial moisture content of clay strongly influence the drying kinetics and transport properties. The dependence between the drying air temperature and evaluated transport properties shows an exponential trend. Tomas and Skansi exponential model parameter, n , is independent from temperature. At lower values of initial moisture content of clay higher drying rates are achieved, which results with higher values of calculated transport properties. It was shown that neural network as an alternative method has potential for modelling the drying process and predicting drying dynamics based on experimental data.

SOFT SENSORS FOR SPLITTER PRODUCT PROPERTY ESTIMATION IN CDU

Soft sensor application for properties estimation of splitter bottom product in a crude distillation unit (CDU) is investigated. Based on continuous temperature, pressure, and flow measurements, two soft sensors are developed as estimators of the initial boiling point and end boiling point of splitter product. Soft sensor models are developed using multiple regression techniques and neural networks. After performing multiple linear regression analysis, it was concluded that linear models are not sufficiently accurate for the implementation in the real plant. Within multilayer perceptron (MLP) and radial basis function (RBF) neural networks, different learning algorithms are used (back propagation with variations of learning rate and momentum, conjugate gradient descent, Levenberg-Marquardt) as well as pruning and Weigend regularization techniques. Statistics and sensitivity analysis are provided for both models. Two developed soft sensors will be used as on-line estimators of heavy naphtha properties and for control purposes.

Soft sensors for crude distillation unit product properties estimation and control

Neural network-based soft sensors are developed for quality estimation of kerosene, a refinery crude distillation unit side product. Based on temperature and flow measurements two soft sensors serve as the estimators for the kerosene distillation end point (95%) and freezing point. The neural networks are trained by the adaptive gradient method using cascade learning. Research results show possibilities of applying soft sensors for refinery product quality estimation and inferential control as an alternative for process analyzers and laboratory assays.

Evaluating rebar corrosion damage in RC structures exposed to marine environment using neural network

Data on the effects of the environmental conditions, structure, and properties of concrete onto the degree of damage caused by chloride-induced steel corrosion have been gathered on three concrete structures in an Adriatic marine environment. The damages were classified into six categories based on the type of remedial work necessary. An artificial neural network for feature categorization was used as a tool for classification of the degree of damage. The model was successfully trained and validated for the range of data from investigated bridges. The interactions and sensitivities of the principal parameters were investigated. The model indicates that the exposure and microclimate conditions are rated higher than the porosity, strength, water–cement ratio, cement content, and cement type. The model could be useful for planning the maintenance of investigated structures and design of remedial works.

Software sensors for monitoring of a solid waste composting process

Process identification for composting of tobacco solid waste in an aerobic, adiabatic batch reactor was carried out using neural network-based models which utilized the nonlinear finite impulse response and nonlinear autoregressive model with exogenous inputs identification methods. Two soft sensors were developed for the estimation of conversion. The neural networks were trained by the adaptive gradient method using cascade learning. The developed models showed that the neural networks could be applied as intelligent software sensors giving a possibility of continuous process monitoring. The models have a potential to be used for inferential control of composting process in batch reactors.

Biodiesel Production by Transesterification of Tallow Fat Using Heterogeneous Catalysis

Biodiesel is an eco-friendly alternative diesel fuel prepared from domestic renewable resources i.e. vegetable oils and animal fats. In this process, biodiesel is produced by transesterification of triglycerides present in animal fat or vegetable oils, by displacing glycerine with a low molar mass alcohol using homogeneous or heterogeneous catalysis. The resulting ester, after mixing with diesel fuels, has physicochemical properties similar to those of conventional fuels. In this work, the batch process of biodiesel production has been studied using tallow fat as raw material with methanol and a heterogeneous catalyst. The quality of the produced biodiesel was evaluated by the determination of important properties, such as viscosity, flash point, cetane number, oxidation stability, glycerine content, acid value, etc. The produced biodiesel was found to demonstrate fuel properties within the ranges recommended by the ASTM D6751.

Effect of Rice Husk Ash on High-Temperature Mechanical Properties and Microstructure of Concrete

Effects of rice husk ash (RHA) on the strength and temperature resistance of concrete were investigated. Different amounts of cement in concrete were replaced by RHA and fly ash (FA), used as mineral admixtures, under the condition of a constant binder content. The compressive strength and temperature resistance were tested at different temperatures. The results show that mixing concrete with the appropriate amounts of RHA can improve its compressive strength. At 800 °C, the strength is 50 % greater than that of normal concrete (NC). Thus, RHA can improve the strength and temperature resistance of concrete.

Study on the Synthesis and Corrosion Inhibition Performance of Mannich-Modified Imidazoline

A novel Mannich-modified imidazoline (MMI) as cationic emulsifier was synthesised for corrosion harm reduction, through three steps — acylation, cyclization, and Mannich reaction. The surface activity was characterized by determination of surface tensions and critical micelle concentration (CMC). The corrosion inhibition performance of five types of steels in the simulated corrosion solution in the presence of the MMI was investigated by static weight loss tests. The results showed that the MMI had good surface activities, with CMC of 19.8 μg g−1 and surface tension of 36.4 mN m−1. The corrosion test results indicated that the corrosion rates of different materials were decreased significantly, and degrees of corrosion inhibition were always higher than 80.0 %. The main inhibition mechanism was most likely due to the adsorption of the corrosion inhibitor on the steel surface, leading to the prevention of corrosion medium from the metal surface.

Development of Soft Sensors for Crude Distillation Unit Control

Abstract Soft sensors for distillation end point (D95) on-line estimation in crude distillation unit (CDU) are developed. Experimental data are acquired from the refinery distributed control system (DCS) and include on-line available continuously measured variables and laboratory assays. Soft sensors are developed using different linear and nonlinear identification methods. Additional laboratory data for model identification are generated by Multivariate Adaptive Regression Splines (MARSplines). The models are evaluated based on Route Mean Square (RMS), Absolute Error (AE), FIT and Final Prediction Error (FPE) criteria. The best results are achieved with Box Jenkins (BJ), Output Error (OE) and Hammerstein–Wiener (HW) model. Based on developed soft sensors it is possible to estimate fuel properties in continuous manner and apply inferential control. By real plant application of developed soft sensors considerable savings could be expected, as well as compliance with strict law regulations for product quality specifications.

Soft sensors model optimization and application for the refinery real-time prediction of toluene content

ABSTRACT Industrial facilities nowadays show an increasing need for continuous measurements, monitoring and controlling many process variables. The on-line process analyzers, being the key indicators of process and product quality, are often unavailable or malfunction. This paper describes development of soft sensor models based on the real plant data that could replace an on-line analyzer when it is unavailable, or to monitor and diagnose an analyzer’s performance. Soft sensors for continuous toluene content estimation based on the real aromatic plant data are developed. The autoregressive model with exogenous inputs, output error, the nonlinear autoregressive model consisted of exogenous inputs and Hammerstein–Wiener models were developed. In case of complex real-plant processes a large number of model regressors and coefficients need to be optimized. To overcome an exhaustive trial-and-error procedure of optimal model regressor order determination, differential evolution optimization method is applied. In general, the proposed approach could be, of interest for the development of dynamic polynomial identification models. The performance of the models are validated on the real-plant data.