Multi Parameter Estimation in an Induced Draft Cooling Tower Using Genetic Algorithm

Considering the need of performance control in engineering systems, this work presents a methodology to predict the controlling variables to control the performance of an induced draft cooling tower. At first, the set of experiments have been conducted with the variation of mass flow rate of water and air under identical ambient conditions. The experimental data for temperatures at different locations has been collected using data acquisition system (by National Instruments) in conjunction with LABVIEW™. Thereafter, relevant 3^rd order empirical correlations of range and approach have been developed using the experimental readings. Depending upon the pertinent requirement, it is required to operate the cooling tower at certain combination of mass flow rate of water and air to fulfill the required output. Based upon the user requirement, the correlations are further employed to construct relevant constraint functions using the least square technique. In order to meet a desired performance (say either a given range, approach or optimum operation) of the cooling tower, the retrieval of design variables (water and air flow rates) has been carried out using an inverse optimization methodology to ensure minimum power consumption. The Genetic Algorithm (GA) is used as an optimization algorithm that minimizes the objective function along with given constraint. The optimization algorithm simultaneously predicts the possible combination of mass flow rate of water and air (control or design variables) in order to meet the given requirement. Further, the methodology avoids multiple combinations of controlling variables that satisfies a particular requirement. Therefore, the user can select an optimum combination that results in minimum power consumption. Moreover, if the cost involved in the cooling tower is considered, it is directly proportional to the range (difference between water inlet and outlet temperatures), whereas, at the same time, the cost is inversely proportional to the approach (difference between outlet water temperature and inlet air wet bulb temperature). In many applications like HVAC (heating, ventilating and air conditioning), chillers, cold storage plants and many more, lower cooling water temperature (at system inlet) is preferable in order to enhance the system efficiency. On the other hand, lower water outlet temperature from the cooling tower for a given water inlet temperature (at tower inlet) means either high range of the tower or low approach, consequently increasing the tower operating cost. Therefore, in order to save the cost involved in cooling tower operation, a compromise between the range and the approach has to be maintained to achieve an optimum performance. So, this method can be also used to predict the optimum operating parameters ensuring the possible optimum performance from the cooling tower under a given set of operating conditions.