Research On Application Of PSO Algorithm Based NMPC To District Heating Networks Control

With the advantages of energy-saving, protecting environment, economy andhigh-efficiency, District heating model has become the mainstream of the urban heating innorth China. District Heating Networks system consists of heat source, heat pipe networkand the user, Due to that it is a typical complex, multi-variable, nonlinear, time-varying,big lag system, the optimization and control problem of this system is more complicated.With the low-carbon economy, researching and developing the efficient control algorithmbecome a research hotspot, that can be applied to the Heating network adjustment to makesaving energy and reducing consumption as much as possible and meeting the demand ofheat users. In This paper, a new kind of algorithm on Application the Particle SwarmOptimization (Particle Swarm Optimization, PSO) based Nonlinear Model PredictiveControl (Nonlinear Model Predictive Control, the NMPC) to District Heating NetworksControl was put forward to achieve energy saving and consumption reducing.Firstly, the heating process mathematical model of the heating station was established,to get the transfer function model of quality and quantity adjustment was got bymechanism modeling method. Then the determined part of the model was got through thatthe model parameters identification was made using the field experiment data, and therandom part of the model was established by the ARIMA model (difference autoregressivemoving average model). Based on the heating station model, established the forecast modelof heating substation by the GPC (generalized predictive control), calculated and identifiedmodel parameters on-line to realize that the nonlinear model predictive control of theheating station in the heating process. Finally, for energy saving, designed the optimizationobjective function, solved the optimization of the problem heat supplying in the primarynetwork using the particle swarm (PSO) algorithm. Due to the uncertainty of Energytransmission model and environment information, heat load has uncertain and time-varyingcharacteristics. In This study, the predict heat load curve model was established which itused as a constraint condition, according to the weather forecast for outdoor temperature prediction curve. Then using the PSO algorithm, under the condition of meeting thedemand of user, the optimal value of total heat quantity was calculated to determine theheat source adjustment strategy and achieve the energy saving control purpose.In this Paper, by field experiment and the simulation results, we verified thenonlinear model predictive control effectiveness of the proposed heat station based ongeneralized predictive control method, and the correctness of the heat production strategyof the heat source based on the PSO algorithm, that has certain guiding significance for thework adjustment of the District heating.