Crude Oil Blending Problems Based On Swarm Intelligence Optimization Algorithms

To save costs and increase procurement diversity,refiners often mix one or more crude oils to replace the target,the technique known as crude oil blending.At present,scholars at home and abroad have proposed some methods to solve the crude oil blending problem,but they are all aimed at solving several crude oil blending problems with fixed proportion,and ignore the optimization problem of crude oil selection.Swarm intelligence optimization algorithm have the advantages of strong searching ability,fast solving speed and high usability,and is widely used in the optimization problems of industrial scheduling,artificial intelligence and other fields.Aiming at the characteristics of crude oil blending problem,this paper applies swarm intelligence optimization algorithm to crude oil blending problem.The main research contents are as follows:(1)A grey wolf optimization algorithm based on levy flight(LGWO)is proposed.The grey wolf optimization algorithm(GWO)is a typical swarm intelligence optimization algorithm.In the later stage of iteration,the standard GWO algorithm is easy to fall into local optimization,losing the initiative of the population and affecting the optimization accuracy of the algorithm.In order to solve these problems,this paper introduces levy flight with strong randomness,irregular searching and discontinuity in GWO algorithm.In order to detect the performance of the proposed algorithm,compared with other typical swarm intelligence optimization algorithms in 12 test functions,the results shows that the global optimization ability of LGWO algorithm is stronger.(2)A crude oil selection and blending optimization model is established,and the method of solving the model by LGWO algorithm is given.A new coding scheme is designed to deal with the continuous and binary variables in the model,and the proportion of mixed crude oil can be determined.In order to ensure the impartiality of the experiment,compared with the cuckoo search algorithm and particle swarm optimization algorithm which have strong robust on the crude oil data.,The similarity between the mixed crude oil and arget crude oil obtained by the LGWO reached 99.7%.the results show that the grey wolf optimization algorithm based on levy is an effective method to solve the crude oil blending problems.To sum up,this paper applies the gray Wolf optimization algorithm based on levy's flight to the oil blending problems,which can provide a reference for refineries to find alternative crude oil and other mixed materials,and is of great significance for theeconomic development of oil refining enterprises.