| Along with the popularity of distributed energy generations and hybrid energy appliances,the integrated energy systems combining various kinds of energy have drawn more and more attention.Integrated energy systems are with three characteristics.Firstly,the systems are more complex in structure and larger in scale.The complex network topology of the system with heterogenous components makes it a hierarchical structure.Secondly,it is no longer a simple individual but composed of different market participants,who compete and corporate with others in an energy market.Finally,its complex structure leads to its complex dynamics.System dynamics of multi-modes including electrical dynamic process,hydraulic dynamic process and thermal dynamic process,make it difficult to analysis one certain kind of dynamics of the system.In order to solve these problems,it is important to establish a generic model of integrated energy systems,which taking into consideration of the analysis of energy systems with various physical properties,multi-market entities,and multi-time scales.In this dissertation,our work subjects to solve the problem and can be mainly divided into four aspects:(1)This dissertation proposes an individual-based model(IBM).The individual-based model decouples complex system into independent individuals according to heterogenous physical characteristics.The model can be fully described by individual models and system models.The former is formalized with a quintuple set comprised of input,knowledge,state,function,and output sets.The latter is formalized with the set of sub-individuals and their connections.Specific physical characteristics and evolution mechanisms of one individual are described in the function set.The evolution of an individual is guided by the function set and based on the input of the external environment and its knowledge set,that is,the individual has the ability to make independent decisions based on environmental information.Additionally,the individuals can interact with others by input and output sets in a uniform manner.(2)This dissertation proposes an individual-based distributed parallel(IBDP)optimization method for solving integrated energy systems.Firstly,four kinds of basic modelling units are proposed for modelling integrated energy systems.Then we inherit the device level individuals from the basic units,which further composite the low level system individuals.Additionally,the low level system individuals make up high level system individuals.Based on the three-level hierarchical models,we decouple the integrated energy system from the device level and lowsystem level.Accordingly,the large complex optimization problem is decomposed into several individual subproblems which can be solved separately.The solutions obtained by the proposed individual-oriented distributed parallel optimization method are compared with that obtained by centralized solution in the high level system,and prove the validity of the proposed method.(3)This dissertation proposes a multi-individual interactive method(MIIM)for optimization of multiple market participants in integrated energy systems.In an open up energy market,system components are operated by different individuals and are involved in the market competition.Each individual in the market objects to maximize its own benefits and interacts with others through input and output sets.In order to protect the information privacy and independent decision making of each individual,and make full use of the limited mutual information to guide individual decisions to approach the maximum of social welfare,we set two types of information in the system.One type is local information which is only visible within the individual,whose role is to guarantee the information privacy and independent operation of one individual.The other type is the global information which is visible through the individuals,whose role is to transmit environmental boundary information and guide individual optimization.In the proposed method,we guide the individual optimization through the energy pricing mechanisms and energy demands.The individuals participate in an iterative process,updating local and global information in each iteration step until the system equilibrium is achieved.Addition to that,the models have also taken into consideration the part-load performance of generation units and flexible cooling energy demands.Simulation studies are carried out in several integrated energy systems.Furthermore,we apply the proposed method in Guangzhou Higer Education Mega Center to optimize the operation strategy of the energy supplier,while the investment feasibility of the optimized scheme is discussed.(4)This dissertation proposes an individual-based multi-time scale dynamic analysis(MTSDA)method for simulating multi-mode dynamics of integrated energy systems.Integrated energy systems are typical multi-mode systems,in which the dynamic process of electrical power systems is extremely fast,while that of thermal energy system is rather slower.Individual-based models are applied to establish the dynamic models of different individuals in the system.In their function sets,both quantified system states and unquantified rules,such as the evolutionary rules and system disturbance information,can be fully expressed.Each individual develops in its own mode and interacts with others at specific time points.In this way,complex dynamics of an integrated energy system are described accurately and the dynamic process of the system in each mode can be conveniently extracted to analyze the system stability.The proposed method is applied to simulate the multi-time scale dynamics of system under conditions of four types of disturbances and faults,which compose of the load variation and the three phase short-circuit fault in electrical power system,and the load variation and pipeline outage in thermal energy system.The results show that the method can significantly reduce computation time and interactions between individuals compared with traditional equal-step method. |
