The Cooperative Quantum-particle Swarm Algorithm And Its Application In The Energy Utilization Optimization Of The Steam Network

The steam network system is the important component of the public work, its safe and stable operation make the pledge that the enterprises can run safely and stably for a long time. The energy optimization of steam network is significant for reducing the ethylene energy consumption, improving the efficiency of the energy utilization, developing the low carbon industry and completing the nation task of energy saving and emission reduction.The energy utilization optimization of the steam network contains a large number of continuous variables (such as temperature, pressure, flow) and discrete variables (such as the off and on of the electric pumps). It belongs to problem of mixed integer nonlinear programming (MINLP). According to the problem, the cooperative quantum-particle swarm algorithm (CQGAPSO) is proposed to solve the problem of MINLP. The main contributions of this dissertation can be summarized as follows:Firstly, in this thesis according to MINLP, CQGAPSO is proposed. CQGAPSO divides the variables into two populations:continuous population and integer population, uses different algorithms to optimize the two populations, the two populations exchange the information by the cooperative mechanism. Test with optimization functions proved that CQGAPSO has the better optimization capacity.Then, because the steam flow is the key data of the energy optimization and it is hardly got, the soft-sensing of turbine steam flow based on the neutral network (NN) with switches is designed in this thesis, CQGAPSO is applied to tune both network structure and parameters of the NN simultaneously, The soft-sensing experimental results of turbine steam flow show that the soft-sensing model of turbine steam flow could obtain the better prediction accuracy and robustness than the all-connection NN and other models..Finally, according to the energy utilization optimization of the steam network, we use the operating cost of the steam network as the objective function, confirm the constraint condition based on the conservation of mass and energy, found the model of the steam network and optimize it by CQGAPSO, the result is good.