Global Optimization Method And App Development Of Energy-saving Operation Of Chiller Room Based On I~2B Architecture

The chiller room system composed of chillers,chilled water pumps,cooling water pumps,cooling towers and other equipment bears most of the energy consumption of the central air conditioning system.The optimal control of the chiller room is the key to the efficient and energy-saving operation of the central air conditioning system.However,in actual projects,due to its own limitations,the traditional centralized tree-structured central air-conditioning automatic control system has led to many disadvantages such as complicated network configuration,blocked underlying information,and difficulty in upgrading and transformation.It is difficult to meet the new requirements of modern building intelligence.In order to change this situation,Tsinghua University proposed the Insect Intelligent Building(I~2B)network platform technology.Under the I~2B architecture,each electromechanical device or building space unit is connected to an computing process node(CPN),replacing the tree network in the traditional centralized architecture with CPN's centerless and flat networking.In the context of I~2B decentralized architecture,this paper aims at the problem of parallel equipment allocation optimization and global diffusion optimization in the field of collaborative optimization of chiller room equipment.First of all,the key equipment of the chiller room is modeled and the parameters are identified online in order to lay the foundation for algorithm analysis and simulation.Furthermore,for the optimization of parallel equipment allocation in chiller room,an optimization algorithm of parallel equipment allocation in chiller room based on particle swarm optimization is proposed.In the simulation experiment of multiple working conditions,compared with the common control strategy in the centralized architecture,the algorithm has better energy saving effect under the premise of meeting the constraint amount,and has the characteristics of fast convergence speed and high solution accuracy.Based on the parallel equipment allocation optimization problem,the scope of system optimization is expanded,and the global diffusion optimization problem of the chiller room is proposed.And the problem is further decomposed into two sub-problems:chilled water system diffusion optimization and cooling water system diffusion optimization.For the diffusion optimization problem of chilled water system,this paper proposes a global diffusion multi-objective optimization algorithm for chilled water system based on NSGA-?algorithm.In the simulation experiment,when the system uses the same equipment start-stop combination,the algorithm saves 8.43%on average than the common control strategy under the traditional architecture.For the cooling water system optimization problem,the cooling tower outlet water temperature is further optimized on the basis of the cooling tower's"uniform water distribution,synchronous frequency conversion"strategy.In multiple working condition simulation experiments,the algorithm can find an approximate optimal solution to the problem and have a good global optimization ability.In order to realize the engineering application of the above algorithm,this paper discusses the implementation of distributed algorithm under the actual I~2B architecture based on the distributed environment virtual software system DSP and building virtual simulation platform.Completed the preliminary development of APP for the parallel equipment allocation optimization algorithm and global diffusion optimization algorithm of the chiller room,and conducted the distributed verification and semi-physical simulation of the algorithm using the APP simulation platform under the I~2B architecture.The simulation results show that the group control APPs of chiller room can successfully find the near optimal solution of the problem under the I~2B architecture,which can meet the engineering application requirements.