Search

In conventional mobile crowdsensing (MCS) applications, the crowdsensing server (CS-server) needs mobile users’ precise locations for optimal task allocation, which raises privacy concerns. This paper proposes a privacy-preserving task allocation framework (called P2TA) for edge computing enhanced MCS, focusing on optimize task acceptance rate while protecting participants’ privacy by introducing edge nodes. The basic idea is that edge nodes act as task assignment agents with privacy protection that prevents an untrusted CS-server from accessing a user’s private data. We begin with a thorough analysis of the limitations of typical task allocation and obfuscation schemes. On this basis, the optimization problem about location obfuscation and task allocation is formulated in consideration of privacy constraints, travel distance and impact of location perturbation. Through problem decomposition, the location obfuscation subproblem is modeled as a leader-follower game between the designer of location obfuscation mechanism and the potential attacker. Against inference attack with background knowledge, a genetic algorithm is introduced to initialize an obfuscation matrix. With the matrix, an edge node makes task allocation decisions that maximize task acceptance rate subject to differential and distortion privacy constraints. The effectiveness and superiority of P2TA compared to exiting task allocation schemes are validated via extensive simulations.

The power system rescheduling based on model-free methods has obvious defects in practical application, such as poor scenario transferability, long data training time, and waste of domain knowledge. To overcome the above defects, a transfer-reinforcement-learning-based rescheduling method of differential power grids considering security constraints is proposed. When constructing the Markov decision-making process of security-constrained rescheduling, both the off-limits of line power and node voltage are considered in the reward. The action space of deep reinforcement learning is narrowed by calculating the sensitivities of devices and mapped to control the active and reactive power regulating devices to reschedule active and reactive power simultaneously. According to the change degree of transfer object, the applications of transfer learning are divided into two scenarios. For the security-constrained rescheduling transfer scenario of different structures of the same power grid, a domain-adaption transfer learning method is formed, realizing good data adaptability after structure changes with the original model. Moreover, a policy-based transfer learning method is constructed for the security-constrained rescheduling transfer scenario of different power grids, enhancing the training speed and training effect of target power grid. Two standard systems and two actual power grids are utilized to verify the effectiveness of the method. For the actual power grids, the effects of the two scenarios are improved by 5.8% and 3.9% with transfer learning. Compared with other methods, this method not only has obvious advantages in transferability, but also has a shorter learning process and lower control cost.

To improve the scalability and elasticity of application deployment and operation in cloud computing environments, new architectures and techniques are developed and studied, e.g., microservice architecture, and Docker container. Especially, Docker container enables the sharing on operation system and supporting libraries, which is more lightweight, prompt and scalable than Hypervisor based virtualization. These features make it ideally suit for applications deployed in microservice architecture. However, existing models and schemes, which are mostly designed for Hypervisor based virtualization techniques, fall short to be efficiently used for Docker container based application deployment. To take the benefits of microservice architecture and Docker containers, we explore the optimization of application deployment in cloud data centers using microservice and Docker containers. Our goal is to minimize the application deployment cost as well as the operation cost while preserving service delay requirements for applications. In this paper, we first formulate the application deployment problem by examining the features of Docker, the requirements of microservice-based applications, and available resources in cloud data centers. We further propose a communication efficient framework and a suboptimal algorithm to determine the container placement and task assignment. The proposed algorithm works in a distributed and incremental manner, which makes it scalable to massive physical resources and diverse applications under the framework. We validate the efficiency of our solution through comparisons with three existing strategies in Docker Swarm using real traces from Google Cluster. The evaluation results show that the proposed framework and algorithm provide more flexibility and save more cost than existing strategies.

In view of the problem of manpower and time consumption caused by unsolvable power flow in current AC / DC hybrid power grid with new energy, a power flow state assessment and solvability recovery adjustment method based on graph convolutional recurrent neural network (GCRNN) and deep double Q-learning network (DDQN) is proposed. First, active power balance degree, reactive power balance degree and channel transmission margin are defined and weighted together to evaluate the solvability of power flow. Then, a power flow state evaluation model based on GCRNN is proposed by combining graph convolution with recurrent neural network, where the input is tangent vector of power flow equations, and the output is comprehensive evaluation index of solvability. Next, the state space, action space and multiple rewards of deep reinforcement learning are formed. By introducing the search direction, learning information extraction and search space constraints, an automatic adjustment method to restore power flow solvability is designed. Finally, the simulation results show that the proposed method performs well both in New England 39-bus standard system and actual power grid of China. Compared with the results obtained by other methods, the superiority of the proposed method is verified.