Sanjeev S. Tambe

Hybrid process modeling and optimization strategies integrating neural networks/support vector regression and genetic algorithms: study of benzene isopropylation on Hbeta catalyst

This paper presents a comparative study of two artificial intelligence based hybrid process modeling and optimization strategies, namely ANN-GA and SVR-GA, for modeling and optimization of benzene isopropylation on Hbeta catalytic process. In the ANN-GA approach [Ind. Eng. Chem. Res. 41 (2002) 2159], an artificial neural network model is constructed for correlating process data comprising values of operating and output variables. Next, model inputs describing process operating variables are optimized using genetic algorithms (GAs) with a view to maximize the process performance. The GA possesses certain unique advantages over the commonly used gradient-based deterministic optimization algorithms. In the second hybrid methodology, a novel machine learning formalism, namely support vector regression (SVR), has been utilized for developing process models and the input space of these models is optimized again using GAs. The SVR-GA is a new strategy for chemical process modeling and optimization. The major advantage of the two hybrid strategies is that modeling and optimization can be conducted exclusively from the historic process data wherein the detailed knowledge of process phenomenology (reaction mechanism, rate constants, etc.) is not required. Using ANN-GA and SVR-GA strategies, a number of sets of optimized operating conditions leading to maximized yield and selectivity of the benzene isopropylation reaction product, namely cumene, were obtained. The optimized solutions when verified experimentally resulted in a significant improvement in the cumene yield and selectivity.

Prediction of all India summer monsoon rainfall using error-back-propagation neural networks

SummaryIn this paper, multilayered feedforward neural networks trained with the error-back-propagation (EBP) algorithm have been employed for predicting the seasonal monsoon rainfall over India. Three network models that use, respectively, 2, 3 and 10 input parameters which are known to significantly influence the Indian summer monsoon rainfall (ISMR) have been constructed and optimized. The results obtained thereby are rigorously compared with those from the statistical models. The predictions of network models indicate that they can serve as a potent tool for ISMR prediction.

NONLINEAR PH CONTROL

A simple new method for designing nonlinear IMC controlles for SISO systems has been developed. The method has been applied to the neutralization of a simulated strong acid-strong base system. The objective of the control effort in this instance is to maintain the effluent pH at 7.00 in the presence of disturbances. An examination of the results shows that the controller provides perfect set point compensation and excellent disturbance rejection. The results also show that to implement this type of controller for pH control a fast CPU with extended precision capabilities and a fast analog-to-digital converter would be required.

Artificial neural networks for prediction of mycobacterial promoter sequences

A multilayered feed-forward ANN architecture trained using the error-back-propagation (EBP) algorithm has been developed for predicting whether a given nucleotide sequence is a mycobacterial promoter sequence. Owing to the high prediction capability ( congruent with 97%) of the developed network model, it has been further used in conjunction with the caliper randomization (CR) approach for determining the structurally/functionally important regions in the promoter sequences. The results obtained thereby indicate that: (i) upstream region of -35 box, (ii) -35 region, (iii) spacer region and, (iv) -10 box, are important for mycobacterial promoters. The CR approach also suggests that the -38 to -29 region plays a significant role in determining whether a given sequence is a mycobacterial promoter. In essence, the present study establishes ANNs as a tool for predicting mycobacterial promoter sequences and determining structurally/functionally important sub-regions therein.

Genetic Programming Assisted Stochastic Optimization Strategies for Optimization of Glucose to Gluconic Acid Fermentation

This article presents two hybrid strategies for the modeling and optimization of the glucose to gluconic acid batch bioprocess. In the hybrid approaches, first a novel artificial intelligence formalism, namely, genetic programming (GP), is used to develop a process model solely from the historic process input‐output data. In the next step, the input space of the GP‐based model, representing process operating conditions, is optimized using two stochastic optimization (SO) formalisms, viz., genetic algorithms (GAs) and simultaneous perturbation stochastic approximation (SPSA). These SO formalisms possess certain unique advantages over the commonly used gradient‐based optimization techniques. The principal advantage of the GP‐GA and GP‐SPSA hybrid techniques is that process modeling and optimization can be performed exclusively from the process input‐output data without invoking the detailed knowledge of the process phenomenology. The GP‐GA and GP‐SPSA techniques have been employed for modeling and optimization of the glucose to gluconic acid bioprocess, and the optimized process operating conditions obtained thereby have been compared with those obtained using two other hybrid modeling‐optimization paradigms integrating artificial neural networks (ANNs) and GA/SPSA formalisms. Finally, the overall optimized operating conditions given by the GP‐GA method, when verified experimentally resulted in a significant improvement in the gluconic acid yield. The hybrid strategies presented here are generic in nature and can be employed for modeling and optimization of a wide variety of batch and continuous bioprocesses.

Statistical analysis of the physico–chemical data on the coastal waters of Cochin

Measurements of temperature, salinity, dissolved oxygen, nitrogen as ammonia, nitrate and nitrite, and phosphate along with chlorophyll were carried out at three stations on the coastal waters of Cochin, south west India, at two-levels of the water column over a period of five years. The data set has been factorised using principal component analysis (PCA) for extracting linear relationships existing among a set of variables. A graphical display of the scores generated from the PCA was done by means of boxplots and biplots, which helped in the interpretation of the data. The major factors conditioning the system are related to the input of fresh water from the estuary of the Periyar river and the high organic load of the bottom sediment in the coastal area which results in a reducing environment, as reflected in the parameters of dissolved oxygen, ammoniacal-nitrogen and nitrite-nitrogen. Another factor which contributes to the variation in the system is related to the unloading activity in the port area. The present approach presents a logical way to interpret the complex data of the physico-chemical measurements.

Counterpropagation neural networks for fault detection and diagnosis

This paper shows the application of a counterpropagation neural network (CPNN) to detect single faults and their magnitudes. The performance of CPNN has been evaluated by considering a variety of faults occurring in a nonisothermal continuous stirred tank reactor (CSTR). The results presented here indicate that CPNN provides an attractive alternative to error-back-propagation (EBP) networks due to its faster learning ability for fault detection and diagnosis.

Fischer–Tropsch synthesis with Co/SiO2–Al2O3 catalyst and steady-state modeling using artificial neural networks

This paper reports the results of an experimental study involving Fischer–Tropsch synthesis on Co/SiO2–Al2O3 catalyst. The objective of the study was to find the reaction conditions for achieving an optimal selectivity with respect to C+5-liquid hydrocarbons. The experimental data on reaction conversion and steady-state concentrations of different product species has been used to develop artificial neural-network-based models which are generic and can be used for predicting the reaction behaviour under different operating conditions.

Optimization of Continuous Distillation Columns Using Stochastic Optimization Approaches

The present work describes the use of two stochastic optimization formalisms, namely, genetic algorithms (GAs) and simultaneous perturbation stochastic approximation (SPSA), for the optimization of continuous distillation columns. Both the simple and azeotropic systems are considered in the analysis. In particular, for a specified degree of separation the problem of finding the optimal values of: (i) the number of stages, (ii) reflux ratio (entrainer quantity in the case of azeotropic distillation), (iii) feed location(s), have been addressed. The GA-based optimization has several attractive features such as: (i) convergence to the global rather than to a local minimum, (ii) the objective function need not satisfy smoothness, differentiability, and continuity criteria, (iii) robustness of the algorithm. The other optimization technique used in the study i.e., SPSA, is a rapid gradient-descent related method for multivariate optimization and is especially well-suited in situations where direct computation of the objective function gradient is not feasible, or the objective function measurements could be noisy. The feasibility of utilizing the GA and SPSA techniques has been demonstrated by considering the separation of three binary and two azeotropic systems of industrial relevance.

Application of artificial neural networks for prokaryotic transcription terminator prediction.

Artificial neural networks (ANN) to predict terminator sequences, based on a feed‐forward architecture and trained using the error back propagation technique, have been developed. The network uses two different methods for coding nucleotide sequences. In one the nucleotide bases are coded in binary while the other uses the electron—ion interaction potential values (EIIP) of the nucleotide bases. The latter strategy is new, property based and substantially reduces the network size. The prediction capacity of the artificial neural network using both coding strategies is more than 95%.