Normal Distribution Based Continuous Multi Ant-Colony Optimization And Its Application In Chemical Engineering

Since entering the 21st century, chemical industry is faced with pressures from economic, energy, environment and many other problems. Optimization technique to meet these challenges is an effective approach, which can be applied on any scales of the entire value chain in chemical industry. But chemical system is a typical complex system. As the scale of object problem becomes more and more large, of which the model structure becomes more and more complicated too, classical optimization method can't meet the demands of many practical problems. So the requirement for right efficient intelligent optimization methods has become increasingly urgent.Ant colony optimization (ACO) is recently proposed as a class of intelligent optimization methods. Its predominant distributed pattern of problem solving achieves great success in combinational problems, and brings extensively attentions of related research area. But to many practical engineering problems, they are usually expressed as continuous optimization problems. So, it is an imperative challenge on how to apply the basic ant colony optimization strategy to the problems solving in continuous space, which is the major work of this thesis.In nature, ant colony optimization is a unified searching algorithm framework based on the probability distribution model of solution space parameterization. The parameter is pheromone, and the solution set producing by ants can be considered as the samples for updating the probability distribution parameter. Hence, the model of pheromone distribution is the key factor in ant colony algorithm, which can thoroughly determine the behavior and distribution of ants. Therefore, the key issue for constructing high-performance continuous ant colony optimization is to design reasonable pheromone distribution model.Analyzing the pheromone distribution in biological model of ant colony foraging, we use normal distribution to simulate pheromone distribution andproposed a normal distribution of pheromone based continuous multi ant-colony algorithm, CMACO. Firstly, state transition of ant colony is implemented by stochastic sampling based on pheromone distribution function. Secondly, the distribution function is updated according to the quality of food source, by which the pheromone is updated. The iterative implementation can lead the ants to global optimal solution. Moreover, to improve optimization performance, multi-colonies strategy is introduced to balance the trade-off between global exploration and local exploitation ability based on group recruitment mechanism. Finally, CMACO was applied to several benchmark problems, and the results illustrate CMACO has well global optimization performance.To solve dynamic optimization problems efficiently, control vector parameterization approach is introduced to transform the original dynamic optimization to static optimization problems, which is directly solved by CMACO. The efficiency of this method was illustrated with two challenging problems for optimizing feed-rate of fed-batch bioreactors, and the results show CMACO has well global optimization ability and fast convergence speed.For complex phase equilibrium system, the Gibbs energy function has several local minima, so it's difficult to get the global minimum by the local optimization algorithms. CMACO was utilized to solve the phase equilibrium without reactions, and it need not considering the actual number and type of phases and needn't the derivative. The results show CMACO can find the global solution with high probability.In short, ACO was adapted elaborately for continuous optimization in this work, and proposed a normal distribution based continuous multi ant-colony optimization, which was applied on dynamic optimization problems of chemical processes and complex phase equilibrium calculation.