No abstract

In the current electrical load profile analysis, considering the shortage of traditional methods on the typical load profile extraction of single consumers and the load profile feature extraction, this paper proposes an approach based on time series data mining. Firstly, this method reduces the dimension of the load profile of a single consumer based on the Piecewise Aggregate Approximation(PAA), and re-expresses the load profile of the consumer over a period based on the Symbolic Aggregate approXimation(SAX), representing the consumer's load profile with a symbolic string to extract the typical load profile. Then, combined with the load characteristic indices and time series-based features, the typical load profiles of different consumers are clustered based on the K-means algorithm to analyze the power consumption behaviors. Finally, this paper performs a case analysis with a UCI test data set, and the results show that the proposed approach can excavate typical power consumption behaviors of consumers and improve the electrical load profile analysis efficiency and the clustering quality.INDEX TERMS cluster analysis, electrical load profile analysis, Symbolic Aggregate approXimation, time series data mining.

Integration of electricity and gas distribution networks improves energy utilization and alleviates environmental pollution. In order to obtain a reasonable energy prices and optimize the day-ahead scheduling scheme under the premise of considering the interests of both supply and demand, an interactive equilibrium model considering electricity-gas energy distribution system and integrated load aggregators is established in this paper. In the upper level, the integrated energy distribution network (IEDN) maximizes the profit via solving the day-ahead optimal energy flow considering energy prices changing in the next day. In the lower level, integrated load aggregators (ILA) minimize charging costs, thus optimizing energy distribution based on integrated demand responses. By using the Karush-Kuhn-Tucker (KKT) optimality conditions and the duality theorem of linear programming, the master-slave game model is transformed into a single layer optimization problem. To solve this problem, second-order cone programming (SOCP) and penalty convex-concave procedure (PCCP) are used to deal with the non-convex functions of electricity-gas distribution system. Therefore, the proposed model is approximated by a mixed-integer second-order cone programming problem. Finally, case studies verify the effectiveness of the proposed model and method. INDEX TERMS Integrated energy distribution network, master-slave game, energy pricing, second-order constraint programming, integrated demand response. NOMENCLATURE A. SETS P2G,m,t dual variables of P2G power input limits λ max CHP,m,t , λ min CHP,m,t dual variables of CHP gas input limits λ max Fired,m,t , λ min Fired,m dual variables of boiler gas input limits λ in,max P,m,t , λ in,min P,m,t dual variables of power storage input limits λ out,max P,m,t , λ out,min P,m,t dual variables of power storage output limits λ S,max P,m,t , λ S,min P,m,t dual variables of power storage capacity limits λ in,max G,m,t , λ in,min G,m,t dual variables of gas storage input limits λ out,max G,m,t , λ out,min G,m,t dual variables of gas storage output limits λ S,max G,m,t , λ S,min G,m,t dual variables of gas storage capacity limits ρ the penalty factor of PCCP v c dynamic adjustment ratio of the penalty factor C. PARAMETERS

Temporal and spatial distribution (TSD) model presented in our previous work of air pollutants is an effective model in describing the increment ground level concentration caused by power generation. In this paper, the newly emerging temporal and spatial characteristics of power dispatch when incorporating the TSD model are studied. Firstly, a multi-objective optimization dispatching model for wind-thermalstorage generation system is proposed. In the time dimension, the model can coordinate multiple generation sources in the face of atmospheric condition variation. In the space dimension, the correlations between power plants location, pollutant diffusion paths and atmospheric boundary layers are considered. Secondly, chance constraints are adopted to address stochastic variables, while the stochastic formulation is transformed into a deterministic one based on wind power distribution. Then a multi-objective optimization method is employed to obtain a desired Pareto front. Case studies are carried out on modified IEEE 39-bus system and Guangdong grid system under four strategies, which validate the performance of the proposed model and the effectiveness of the strategy. INDEX TERMS Environmental economic dispatch, comprehensive pollution evaluation value, geographic grid, temporal and spatial characteristics, multi-objective optimization. NOMENCLATURE INDICES WEICONG WU received the B.S. degree in electrical engineering from Huaqiao University, Xiamen, China, in 2018. He is currently pursuing the M.S. degree in electrical engineering with the School of Electric Power Engineering. His research interest includes optimal operation of power systems.

Aiming at the problem of coordinating system economy, security and control performance in secondary frequency regulation of the power grid, a sectional automatic generation control (AGC) dispatch framework is proposed. The dispatch of AGC is classified as three sections with the sectional dispatch method. Besides, a hierarchical multi-agent deep deterministic policy gradient (HMA-DDPG) algorithm is proposed for the framework in this paper. This algorithm, considering economy and security of the system in AGC dispatch, can ensure the control performance of AGC. Furthermore, through simulation, the control effect of the sectional dispatch method and several AGC dispatch methods on the Guangdong province power grid system and the IEEE 39 bus system is compared. The result shows that the best effect can be achieved with the sectional dispatch method. INDEX TERMS automatic generation control; hierarchical multi-agent deep deterministic policy gradient; sectional AGC dispatch; reinforcement learning.