A load aggregator is a professional organization to develop power demand response, which integrates demand-side adjustable resources to participate in system operation and market transactions to gain profits. It is very important for further flexible management of demand-side resources and increasing demand-response benefit to realize personalized incentive customization of multiple types of demand-response resources. It is also the core business of load aggregators. In this paper, a general model of incentive price customization for load aggregators is established according to the principle of stackelberg game. Electric-vehicle users and building air-conditioning users are selected as the research object. Considering the impact of comfort price changing with load shedding, an optimization model of building air-conditioning users with different comfort prices and a model for optimizing the charge and discharge behavior of electric vehicles are constructed under the framework of stackelberg game model. Karush-Kuhn-Tucker condition, dual theorem and linear relaxation are used to derive a two-level mixed integer linear programming model with the minimum economic loss of load aggregator. The results show that the proposed model can be effectively used to customize dynamic incentive price of building air-conditioning users and electric-vehicle users for load aggregators, and provide new ideas for demand response.
With the continuous advancement of power system transformation, the regional integrated energy service provider (RIESP) has changed from a traditional top-down monopoly model without competition to a multi-party competition model. This paper comprehensively considers power generation cost of RIESP and user purchase satisfaction, and proposes a RIESP operation strategy model based on master-slave game. Firstly, the basic structure of RIESP and the two-step decision-making mechanism are introduced, and a master-slave game model is established, in which the maximum profit of RIESP is the leader objective function and the maximum user consumption surplus is the follower objective function. Then, the existence and uniqueness of the Stackelberg equilibrium solution of the model are proved, and the DE-PSO hybrid algorithm is used to solve it. Finally, through simulation examples, the effectiveness of the strategy proposed in this paper is verified, and both the revenue of RIESP and the surplus of user consumption are improved.
The limited regulation capacity in the integrated energy community is a key factor restricting the development of the community-based integrated (distributed) energy system. For that reason, a model for optimal sharing of day-ahead energy in multi-communities considering the matching degree of source and load curves among communities is proposed. Firstly, the relationship between energy supply structure and multi-energy flow is analyzed concretely in a single community. A mathematical model is built that includes electric vehicles and multiple energy conversion equipment, and an objective function is established that minimizes the cost of energy purchase, equipment operation and maintenance, and electric vehicle battery loss costs. Secondly, the comprehensive Spearman constant and Euclidean distance matching index is considered according to photovoltaic and load data among the communities, and the multi-community operation efficiency is optimized with the goal of minimizing the energy interaction cost. Finally, through a 3-community simulation system, the results show that the multi-energy sharing mode can effectively improve the overall economy of the system and the photovoltaic power consumption capacity. The introduction of matching indices in the model also improves the energy transmission efficiency, which verifies the rationality of the proposed model.
Integrated energy system with combined cool, heat and power system (IES-CCHP) is able to help power system to locally consume distributed wind and solar power, while satisfying charge demand of electric vehicles. However, uncertainties in the charge demand, wind and solar power significantly affect the economy of IES-CCHP. Therefore, this paper applies two-stage adjustable robust optimization to present day-ahead economic dispatch strategy for IES-CCHP. Day-ahead stage decides day-ahead dispatch strategy that can withstand the worst-case scenario before observing value of stochastic variables; real-time stage provides strategy for correcting the day-ahead strategy after confirming the stochastic variables. The objective is to minimize the costs of the two stages. Imprecise Dirichlet model is employed to dig historical data for constructing uncertainty set for describing stochastic variables. And then duality theory, big-M method, and column-and-constraint generation (C&CG) and so on, are applied to solve the presented two-stage model. Finally, experimental cases are carried out to demonstrate the effectiveness of model and method.
A large number of independent decision-making microgrids participate in the power market competition in the distribution network, which makes the traditional centralized trading face the problems of long decision-making time, high cost of trust and privacy security. In the face of this situation, a distributed transaction model of electricity in multi-microgrid applying smart contract technology is proposed. Firstly, a smart contract model of electricity trading negotiation market is constructed to protect the interests of users. Secondly, in order to realize safe operation of the distribution network, a network security constraint method suitable for smart contract is proposed. Finally, according to the transaction model, the off-chain scalability architecture of the blockchain is improved to ensure the computing capability of the blockchain, and ensure the security and traceability of all historical data storage. The method is tested on the IEEE 33-bus network, and the results show that the proposed distributed transaction model does not violate the network constraints, and the market participants can profit from the transaction.
With the continuous improvement of the guarantee mechanism of renewable energy power accommodation, it is an important link to lead the long-term development of renewable energy power accommodation to scientifically and rationally set the weight of renewable energy accommodation for each province and region. Applying multidisciplinary collaborative optimization theory, the actual grid structure and grid operation in the region is comprehensively considered in this paper. The influence of large amount of volatile renewable energy on the traditional controllable generating units is analyzed. Under the background of the implementation of carbon emission quotas, a method for calculating the optimal weight index of renewable energy is proposed. By setting up the system objective function of the optimal comprehensive cost and the parallel sub-disciplines under different consideration aspects, the global optimal solution is found by means of cyclic iteration of the system discipline and sub-discipline. The proposed model and algorithm are verified by an example of IEEE 118-node system and an actual power grid in a province.
With the current electric power reform and the continuous promotion of the electric power market, the development of virtual power plant has become a key part of them. It is also the main concentration part of the complex factors of market transactions, and the flexible load and clean energy also bring great challenges to the virtual power plant. Considering the operation characteristics of different types of users’ flexible load, this paper analyzes and models the characteristics of industrial, commercial and household loads, fully mobilizes some controllable resources within the virtual power plant to interact, and combines the operation characteristics of wind power, photovoltaic power and gas turbines to establish a purchase and sale optimization model for virtual power plant considering the deep interaction of diversified flexible loads. In addition, with the development of clean energy and the challenge of uncertainty brought by photovoltaic and wind power, a scenario data-driven distributional robust model with two-stage variable for power purchase and sale optimization of virtual power plant is proposed, which is solved by column and constraint generation algorithm. The effectiveness of the optimization model is verified by the analysis of examples.
Virtual power plant (VPP) technology is an effective way for renewable energy grid-connecting, and is helpful for demand-response resources such as flexible load and electric vehicle to participate in power market. This paper analyzes the characteristics, rights and responsibilities of each member in the virtual power plant,draws up the call contract based on price elasticity for each member, and designs the day-ahead and real-time power market transaction process including main energy and auxiliary services. According to the conditional value-at-risk (CVaR), the risk aversion model of virtual power plant is established, the Shapley value and marginal expected shortfall (MES) of each member are quantitatively analyzed, and the internal benefit distribution method of virtual power plant is given. An example shows that the virtual power plant can take care of the interests of all parties. Also in the example, this paper quantitatively analyzes the risk benefits under different market strategies. The results show the effectiveness of the proposed method.
Hybrid AC/DC microgrid can effectively coordinate the power distribution of microgrid through mutual power support between two sub-grids, and improve the system’s ability to suppress power fluctuations. In order to realize the reasonable mutual power support between the two sub-grids, this paper proposes a mutual power support strategy for hybrid AC/DC microgrid used in island operation mode. Firstly, in order to avoid the power loss caused by the frequent actions of inter-linking converter, a hierarchical control strategy is proposed, which includes power autonomy model and power interaction model, and the changing between system operating models is designed reasonably. Then the goal of mutual power support based on the conditions of the sub-grids and the state of charge (SOC) of the battery is proposed. Meanwhile, a power cooperation control algorithm, which gives consideration to both AC frequency and DC voltage, is designed to achieve the goal. A simulation model is built in PSCAD/EMTDC. The simulation results show that, with the mutual power support strategy, the AC and DC sub-grids can bear the power fluctuations of the system according to their own conditions. And reasonable SOC of the battery can be maintained.
With the acceleration of urbanization, the current urban power grid is faced with increasingly serious problem of operation block. The traditional control mode using load transfer for blocking control is less flexible, and the single control mode is not adaptive. According to characteristics of space-time distribution of transferrable load by energy storage, and considering the influence of operation strategy on the planning scheme, a two-layer planning model of energy-storage power station is proposed, which takes into account the transmission capacity of high-voltage distribution network (HVDN). The upper layer builds energy-storage planning model of investment income, and the lower layer builds more scenarios by giving priority to topology reconstruction, while energy-storage operation optimization model as complementary. Due to the mutual influence between operation control and planning configuration, the inner and outer layers of different time scales are optimized in the same frame. The cutting loads of the system nodes, the location, capacity and power of the energy-storage power station are used as the coupling variables to alternate iteration. The particle swarm optimization algorithm and CPLEX solver are used to alternate for solution. Finally, an example is given to demonstrate the effectiveness of the proposed model and the feasibility of the solution method.
In the scene of multi-station (electric vehicle charging station, 5G communication base station, data center, distributed photovoltaic power station, energy storage station) integration, in order to solve the problem of size optimization and coordinated operation under the constraints of site conditions, functional integration, business requirements, etc. Firstly, the resource demand of each station and the possible functional fusion points among stations are analyzed. And on the basis of this, in order to improve the utilization rate of spare space in the existing substation, we take every station as configuration objects. The configuration principle of each station is defined. And a scheme of functional integration among stations is proposed. Then, to support the function integration scheme, the optimal capacity configuration model and the coordinated control strategy among stations are established. Finally, YALMIP/CPLEX is used to solve the optimization model, and the effectiveness of configuration method and coordinated control strategy have been verified by a typical example.
In the absence of power supply from grid, due to the limitation of the power supply capacity of the isolated microgrid and the uncertainty of the renewable energy output, the situation of insufficient power supply will occur. Making a reasonable power dispatch strategy can improve the electricity economy and customer satisfaction. This article focuses on the situation of limited electricity in the absence of power supply from grid. Taking multiple parks for power sharing as the research object, the grid reasonably allocates electricity to different types of loads in different parks, adjusts the working time of the loads, and establishes an optimal scheduling model of limited electricity for multiple parks to minimize power outage loss, configuration costs of energy storage, and customer dissatisfaction. Then, the non-dominated sorting genetic algorithm based on the elite strategy (NSGA-II) is applied to solve the model, and fuzzy membership degree method is used for the Pareto solution set to obtain the optimal solution. Finally, taking multiple parks in a certain area of Shanghai as an example, the scheme comparison verifies that the optimization strategy proposed in this paper can effectively improve the utilization rate of electric energy, reduce the loss of power outage and the cost of energy storage, and improve customer satisfaction as well.
Aiming at the demand of power grid historical data mining, a method for similarity matching of power grid operation section applying stacked automatic encoder (SAE) is proposed. According to the historical section information, combined with the characteristics of power grid operation, effective sample data is selected. The weights and bias parameters of the pre-training stage are obtained by using unlabeled valid sample data of power grid operation section and layer by layer automatic encoder (AE). Furthermore, in the parameter tuning stage, parameters of the whole network are fine-tuned by using labeled sample data, initialized weights and deviations to obtain a stacked self-coding network capable of mining the deep features of the running section. The proposed method establishes a nonlinear mapping relationship between historical operation section data and similarity measurement through the deep structure of SAE algorithm, and then obtains valuable historical information. IEEE 39-node system is used to verify the proposed method. The results show that the proposed method has higher matching accuracy than K-means algorithm, and the error rate decreases faster with the number of iterations.
In order to realize the full coverage collection of operation and maintenance data for the distributed photovoltaic station with the characteristics of many scattered and disordered points in a wide area, the best way is to configure a set of data acquisition devices for each distributed photovoltaic station. Therefore, it will also face the problems such as huge investment cost and heavy operation and maintenance tasks. This paper proposes a virtual acquisition method for distributed photovoltaic data applying the mixture of grey relational degree and BP neural network, and realizes the operation and maintenance data acquisition of the distributed photovoltaic station in the whole region under the condition of installing a small number of data acquisition devices. A distributed photovoltaic power station in a certain region of Jiangsu province is selected as the research object. Firstly, the historical operation and maintenance data of a distributed photovoltaic power station with data acquisition devices installed in the region are analyzed by using the grey relational theory, and the characteristic curves of irradiance and distributed photovoltaic output are obtained. Then, the correlation degree between the real-time daily irradiance information and historical irradiance data of the photovoltaic station to be virtually collected in the region is calculated, and the historical date with correlation degree above 0.9 is selected as the similar date, and then the BP neural network data virtual acquisition model is established according to the historical data of similar days, which is used to realize the virtual collection of distributed photovoltaic data in the region. Finally, the case verifies that the photovoltaic output power collected by this method has high precision and that the method can realize the virtual collection of photovoltaic output power data in the grid area.
Clustering of load data is an important foundation for analyzing electrical big data. Aiming at the difficulty of extracting sequential features of high-dimensional daily load data, and the reduction of accuracy of load clustering due to the separation of feature extraction and clustering processing, a deep embedding clustering method based on one dimensional convolutional auto-encoder (DEC-1D-CAE) is proposed for daily load data in this paper. Firstly, a one-dimensional convolutional auto-encoder is used to extract sequential features contained in the load curve. Then, a user-defined clustering layer is used for soft division of the extracted load feature vector. Finally, the Kullback-Leibler divergence (KLD) is used as loss function to jointly optimize convolutional auto-encoder and the clustering layer to obtain the clustering result. A numerical experiment were carried out and the results of the proposed method are better than K-means, 1D-CAE+K-means and DEC-1D-CAE on both Davies-Bouldin index (DBI) and Calinski-Harabasz index (CHI), which indicate that the proposed method can effectively improve the accuracy of daily load clustering.
