Abstract—LABVIEW is a graphical programming language that has its roots in automation control and data acquisition. In this paper we have utilized this platform to provide a powerful toolset for process identification and control of nonlinear systems based on artificial neural networks (ANN). This tool has been applied to the monitoring and control of a lab-scale distillation column DELTALAB DC-SP. The proposed control scheme offers high speed of response for changes in set points and null stationary error for dual composition control and shows robustness in presence of externally imposed disturbance. Keywords—Distillation, neural networks, LABVIEW, monitoring, identification, control. I. INTRODUCTION NN is attractive due to its information processing characteristics such as nonlinearity, high parallelism, fault tolerance as well as capability to generalize and handle imprecise information [1]. Such characteristics have made ANN suitable for solving a variety of problems. The application of ANN in chemical engineering began with pioneering works [2], and in subsequent years the number of research publications on ANN in chemical engineering was steadily increased. Most of these publications cover five major areas: process control, dynamic modeling, forecasting fault diagnosis, and optimization. In the area of process control, ANN was applied through adaptive control or model-based control. By monitoring the on-line process data, ANN could be used to adjust controller parameter for optimal performance. Dynamic modelling using ANN was also well practising in process industries. By exploiting the relationship among the process variables, ANN model was developed as estimator and to be implemented in advance control techniques (soft sensors). Manuscript received March 31, 2008. This work was supported by CICYT project DPI05-08344. J. Fernandez de Canete is with the System Engineering and Automation Dpt. , University of Malaga, Spain (phone: 34-952-132887; fax: 34-952- 133361; e-mail: canete@isa.uma.es). P. Del Saz-Orozco is with the System Engineering and Automation Dpt., University of Malaga, Spain (phone: 34-952-131418; fax: 34-952-131413; e- mail: delsaz@isa.uma.es). S. Gonzalez-Perez is with the System Engineering and Automation Dpt., University of Malaga, Spain (phone: 34-952-131412; fax: 34-952-131413; e- mail: sgp@isa.uma.es). Similar to dynamic modelling, forecasting can also contribute in process industries by using prediction based on the history data. ANN was also useful in fault diagnosis since it has the ability to store knowledge about the process and learn from the quantitative historical fault information. ANN was implemented in plant optimization for optimal parameter searching to ensure process plant is always safe and productive. Distillation column is the most common unit operation in the chemical industry and understanding its behaviour has become the most challenging job for chemical engineers. As distillation column can be viewed as an integrated and complex system, the operation and control of column become very difficult. Basically, there are five basic variables required to be controlled to achieve efficient operation, composition of distillate stream, composition of bottom stream, liquid level of reflux drum, liquid level of base column and column pressure. Distillation dynamics and control were well studied in past decades especially for composition control [3]. In practice, on- line analyzer for composition is rarely used due to its costs and measurement delay. Therefore composition is often regulated indirectly using tray temperature close to product withdrawal location. In order to achieve control purpose, many manipulated variables could be used, such as reflux flow (L), distillate flow (D), bottom flow (B) and vapour flow (V). This gave rises to many control strategy with different combination of manipulated variables configurations [4] reported some practical configurations alternatives and their comment for the distillation control scheme mainly refers to composition control. However, using temperature for composition control is not always satisfactory. This is because by maintaining constant temperature does not guarantee the ability to maintain a constant composition. Since the temperature- composition relationship only hold when pressure is kept constant and fluctuations in the column pressure resulting from disturbances. Focusing on the distillation control problem, several control schemes based on knowledge of the plant neural model have been reported, such as predictive control, inverse model control and adaptive control [5]. LABVIEW is a powerful and versatile graphical programming environment that was developed primarily to Artificial Neural Networks for Identification and Control of a Lab-Scale Distillation Column using LABVIEW J. Fernandez de Canete, S. Gonzalez-Perez, and P. del Saz-Orozco A World Academy of Science, Engineering and Technology 47 2008 64