Internet data center demand response optimization method based on multi-objective evolutionary algorithm

Abstract

The invention discloses an internet data center demand response optimization method based on a multi-objective evolutionary algorithm, and the method comprises the following steps: collecting the maximum value and the maximum workload capacity of each data center server, and collecting the total workload of front and rear end web portals; establishing an internet data center demand response model by taking data center profit maximization and customer service quality optimization as targets, and proposing a decision variable reduction method; providing a population-based method to find a finite solution set, namely a Pareto optimal solution. In the iteration process, the Pareto optimum is used for improving the quality of the solution. According to the algorithm, the convergence and diversity of the algorithm are ensured through operations such as variation, crossover, hybrid crossover, selection, archive updating and the like. According to the Internet data center demand response optimization method based on the multi-objective evolutionary algorithm, the feasibility of physical constraints and the expandability of the number of decision variables are effectively solved through the multi-objective evolutionary algorithm, and the internet data center demand response optimization method realizes good balance between data center profit and customer service quality.

Description

technical field [0001] The invention relates to the technical field of data processing, and more specifically, to a demand response optimization method for an Internet data center based on a multi-objective evolutionary algorithm. Background technique [0002] At present, big data is widely used in the power industry in the process of operation management, production marketing and analysis and decision-making. Building a high-performance Internet data center can provide a unified platform for the power industry, integrate analysis and processing data in a timely manner, and provide basic services for enterprise development. According to the characteristics of the uninterrupted operation of Internet data centers in the power industry, detailed anti-risk plans should be put forward in response to various risk failures that may occur. The uncertainty of user needs makes the service quality of Internet Data Center (IDC) particularly important. ; In addition, the final benefit of...

AI Technical Summary

Problems solved by technology

S. Bahrami formulated a "many-to-one" matching strategy, and proposed a distributed algorithm that guarantees convergence to a stable result by modeling real-time pricing and workload scheduling, but the algorithm is not in the modeling process Explicit consideration of Service-Level Agreements (Service-LevelAgreement, SLA), unable to highlight the importance of cost-effective QoS; M.Ghamkhari and H.Mohsenian-Rad, considering the trade-off between minimizing energy consumption and maximizing revenue, proposed The method of profit maximization is proposed, but this method only involves the convex optimization problem of one IDC, and cannot realize the optimization of multiple IDCs; in order to consider multiple IDCs in a multi-power market environment, L.Rao and H.Shao studied the work of IDC Load distribution or load balancing, poses a mixed integer programming problem, but only solves the energy consumption problem without introducing the constraints of QoS

Images