Font Size: a A A

Studies On "Decentralized Control" System Of Chiller Plant

Posted on:2017-12-08Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y C DaiFull Text:PDF
GTID:1312330533455169Subject:Civil engineering
Abstract/Summary:PDF Full Text Request
Chiller plant system,which consists of chillers,pumps,cooling towers and some other electromechanical equipment,is the major component and the largest energy consumption part of the HVAC system in large public buildings.Thus,it is quite important for the HVAC system that the equipment of the chiller plant keep safe,stable and high efficiency operation,and auto control system of the chiller plant is an important technical method to achieve this goal.While the current control system,which is usually based on centralized hierarchical control structure,have the problems of poor practicality and bad generality in the process of real application,leading to the fact that the automation level for the chiller plant systems in real operation is generally low.However the current research focused oncontrol algorithm and control system related to chiller plant system does not pay significant attention to and not effectively solve the pain points.On basis of the technical framework of decentralized intelligent building platform,this paper presents a decentralized control system based on intelligent electromechanical equipment in the chiller plant.Designing some simple and direct principles followed by the connection of intelligent equipment of the chiller plant,the decentralized control system would be set up automatically by a plug-and-play approach.Developing a universal decentralized control algorithm which is independent of the specific chiller plant cinfiguratrion,the intelligent equipment would collaborate with each other and work together to achieve the control targets efficiently.By analyzing the operation demands of the equipment itself and the collaboration demands among different equipments,this paper defines some standardized information model for all kinds of intelligent equipments in the chiller plant and formulates the principles of organization and recognition for the intelligent equipments.Through the object oriented modular analysis,the intelligent equipments are devided into some independent equipment modules according to the function of different organization forms,meanwhile decoupling the optimal control problem of the whole chiller plant system.Two kinds of typical control problem are concluded by further analysis of the operation function inside and among the equipment modules,which are the distribution optimization problem(DTO)inside the equipment module and the diffusion optimization problem(DFO)among equipment omdules.The DTO problem which is the bottom level in the decentralized control system is mainly studied and a universal dencentralized algorithm is designed for the DTO problem based on mathematical analysis.The feasibility of this decentralized algorithm is verified through example applications of different equipment modules.And a continuous simulaiton verification is further applied on a typical chiller plant system model,which shows that solution of this decentralized algorithm is near optimal.For the DFO problem,a penalty method is introduced to conduct the equivalent transformation of decomposition.And a preliminary probe into this algorithm is conducted based on a similar model example.The decentralized algorithm for the DTO problem is developed and transplanted to the decentralized platform,and a hardware emulation environment is built for actual application test of this decentralized algorithm.Finally this decentralized control system,including the decentralized algorithm and the decentralized controller hardware,is applied on two practical application of engineering projects,which are the decentralized chiller module control and the decentralized pump “with different load” module control,demonstrating the features of plug-and-play,agile development,rapid deployment and near optimal control.
Keywords/Search Tags:chiller plant, intelligent device, decentralized control, distribution optimization problem, optimization algorithm
PDF Full Text Request
Related items