Architecture of Data Access Based on Power Dispatch Control Cloud

Xiaoming Zhou, Tao Lei, Dapeng Li, Tao XieAdvances 2016 · DOI

Abstract

This paper presents a novel data access architecture for power dispatch control clouds, addressing the security and efficiency limitations of traditional client/server models. By implementing a "client-proxy-proxy-server" (CPPS) framework, the system enables regional workstations to securely access centralized cloud resources and data from other regions without altering their local operating environments. This approach ensures robust data synchronization for real-time, model, and graph data while maintaining high security through proxy-based isolation and a layered software stack.

Introduction

In modern power systems, data from various regions—including real-time metrics, operational models, and graphical data—is synchronized to a centralized "Power Dispatch Control Cloud." While centralization enables comprehensive big-picture analysis, it often creates a "man-machine" bottleneck.

Traditionally, regional workstations access this cloud using a standard client/server (C/S) architecture. However, direct C/S connections over wide-area networks pose significant security risks for critical infrastructure. Furthermore, operators frequently face "data silos," where accessing resources from a neighboring region requires switching platform environments, an inefficient process during time-sensitive dispatching operations. This paper proposes a secure, transparent architecture to bridge these gaps.

The Client-Proxy-Proxy-Server (CPPS) Architecture

The core innovation is the Client-Proxy-Proxy-Server (CPPS) architecture. Instead of a direct connection, the data flow is broken down into several controlled stages to enhance security and flexibility.

The Client-Proxy-Proxy-Server (CPPS) architecture illustrating the flow from regional workstations to cloud resources through dual proxies.

As shown in the architecture, the system utilizes two primary proxy layers:

  1. Local Data Proxy: Deployed within the regional dispatching institution to handle local requests.
  2. Cloud Data Proxy (Load Balancing & Service Proxy): Located at the cloud center's boundary, these components act as gatekeepers, managing service routing and load distribution.

This "double-proxy" approach ensures that remote workstations never interact directly with the core database, providing a vital layer of insulation that protects sensitive grid data.

Layered Software Stack

The Power Dispatch Control Cloud is organized into five distinct layers to manage complexity and ensure operational security.

The five-layer software architecture of the Power Dispatch Control Cloud, highlighting the vertical security layer.
  • Resource Layer: The foundation where data resides, including distributed databases and file systems.
  • Virtualization Layer: Abstracts physical hardware into pools of computing, storage, and network resources.
  • Application Support Layer: Manages the core logic of the CPPS architecture, including load balancing and service proxies.
  • Application Layer: Performs high-level tasks such as network analysis, mass data processing, and state estimation.
  • Security Layer: A vertical component that handles identity authentication, authority management, and data security across all levels of the stack.

Key Insights and Results

The implementation of this architecture has led to several significant improvements in grid management:

  • Logical Unity vs. Physical Dispersion: The system can utilize physically separate cloud centers for disaster recovery while maintaining a logically unified service for the user.
  • Environment Stability: A workstation in "Region A" can pull resources from "Region B" via the cloud without the operator needing to log out or change their local platform environment. The proxy system handles the translation and retrieval seamlessly.
  • Investment Efficiency: Centralizing the Application Layer reduces the requirement for high-end hardware at every regional site, significantly lowering construction and maintenance costs.

Limitations and Future Work

While the architectural framework is robust, it currently serves as a high-level design. Future work should focus on quantifying the latency overhead introduced by the proxy chain and exploring the scalability limits regarding concurrent users and data throughput. Additionally, future research will likely investigate specific big data analysis and data mining algorithms within the Application Layer to further optimize grid dispatching based on cross-regional data.

Summary updated

Abstract

The data synchronization between all the regional dispatching system under the dispatch control center and power dispatch control cloud has been realized. The data information from each region such as real-time data, manipulate data, model data and graph data firstly is sent to the communication server for this region. Besides, the communication server sends this data to the dispatch control cloud center after some processing. In order to obtain the resource from cloud center, the workstations in each region have to access cloud center remotely. However, most process in man-machine remote access adopts client/server architecture, which affects the security of data access. To solve this problem and provide data proxy service in cloud center, a new architecture of data access based on power dispatch control cloud is proposed in this paper.The architecture of data access is a client-proxy-proxy-server architecture, which can not only ensure remote workstation accessing its own resource from the cloud center, but also ensure it obtaining resource of other regions from the cloud center with not switching the platform environment of smart grid dispatching and control system.
BibTeX
@inproceedings{zhou2016architecture,
  author = {Xiaoming Zhou and Tao Lei and Dapeng Li and Tao Xie},
  title = {{Architecture of Data Access Based on Power Dispatch Control Cloud}},
  booktitle = {Advances},
  year = {2016},
  doi = {10.2991/iccsae-15.2016.68},
}

Loading related content...

Figure