Search

With the development of artificial intelligence,more and more financial practitioners apply deep reinforcement learning to financial trading strategies.However,It is difficult to extract accurate features due to the characteristics of considerable noise,highly non-stationary,and non-linearity of single-scale time series,which makes it hard to obtain high returns.In this paper,we extract a multi-scale feature matrix on multiple time scales of financial time series,according to the classic financial theory-Chan Theory,and put forward to an approach of multi-scale stroke deep deterministic policy gradient reinforcement learning model(MSSDDPG)to search for the optimal trading strategy.We carried out experiments on the datasets of the Dow Jones,S&P 500 of U.S. stocks, and China's CSI 300,SSE Composite,evaluate the performance of our approach compared with turtle trading strategy, Deep Q-learning(DQN)reinforcement learning strategy,and deep deterministic policy gradient (DDPG) reinforcement learning strategy.The result shows that our approach gets the best performance in China CSI 300,SSE Composite,and get an outstanding result in Dow Jones,S&P 500 of U.S., 10 pages,5 figures,to be published in 2022 8th International Conference on Computer Technology Applications conference

With the great integration of distributed generation and practical control strategies as demand response, household prosumers are willing to take part in the day-ahead dispatch of distribution network. However, managing enormous amount of household prosumers with individually low controllable power becomes a huge challenge to the distribution system operator. In this paper, we propose a virtual power plant model for aggregating household prosumers into a simplified dispatch-able model. The feasible region for aggregated power of the household prosumers is contracted to a pyramid by means of Minkowski Sum, which represents an operational region for a power plant, and this model is then uploaded to distribution system operator. By using the aggregated model, distribution system operator no longer need to communicate with each household prosumer and to process a lot of individual constraints, but only use the virtual power plant model in day-ahead dispatch optimization. We take heat pump, electric vehicle, energy storage and photovoltaic into consideration and simulations show the effectiveness of the proposed model.

The distributed generation connected to the microgrid is a key way to balance power supply and demand and to reduce the loss of long distance transmission. The capacity selection of distributed generation in microgrid affects the reliability and economy of power supply. Appropriate capacity of distributed generation is an important part to the efficient and reliable system operation. Based on these points, some research are taken in this article. First, boundary constraints are carried out for the capacity selection of the distributed generation, which are determined by the rated working current of the branch, the upper transformer capacity and the load. Then, based on the consideration of reliability and economy of power supply, the selection optimization are further proposed. Finally, based on the dynamic simulation platform., the microgrid simulation model with multiple distributed generations is carried out to analyse the capacity selection.

Vehicular Computer Clouds (VCC) would be expected to play a major role in the Intelligent Transportation System (ITS) in the near future. While it is clear that the VCC concept raise exceptional security challenges. Generally speaking, most attacks come from operators. So it is important to design the novel VCC security scheme based on operator's attributes. In this paper, we develop the Fuzzy Operator-Attribute-Based Signcryption (FOABS), a novel security mechanism that makes relations between the vehicle operator's right and these operation. FOABS leverages fuzzy Attribute-Based encryption to enable data encryption, access control, and signature for vehicular control information in VCC. FOABS provides confidentiality, authenticity, undeniability, identity authentication, and collusion resistance.