A fan regulation method and system based on machine learning algorithm

Abstract

The invention relates to the technical field of server heat dissipation, and provides a fan regulation method and a system based on a machine learning algorithm. The method comprises the following steps: selecting three of the particles generated by a particle swarm algorithm, and assigning values of the selected three particles to Kp, Ki and Kd respectively; Run PID control algorithm, the systemperformance indexes of PID closed-loop control are calculated, Taking the calculated system performance index as a benchmark, judging whether the system performance requirements of the PID control algorithm are met or not, if so, the parameters corresponding to the three selected particles are taken as the optimal parameters of the PID control algorithm, If you do not return to execute the selection of three of the particles generated from the particle swarm algorithm, The selected three particles are assigned to the parameters Kp, Ki and Kd of the PID controller respectively, so as to realizethe intelligent PID control of the server fan, avoid time-consuming and labor-consuming manual adjustment, and ensure that each component of the server works in a safe temperature range.

Description

Technical field [0001] The invention relates to the technical field of server heat dissipation, in particular to a fan control method and system based on a machine learning algorithm. Background technique [0002] With the development of new technologies such as cloud computing and big data, the bandwidth and capacity requirements for data storage are becoming higher and higher, and the computing speed and computing capacity of the processor are also increasing, resulting in the temperature of various components such as memory and hard disks. The heat dissipation of electronic devices has also been soaring, and the heat dissipation of electronic devices has become a very scorching problem at present, and the requirements of society on power consumption are getting lower and lower, and energy saving is currently a mainstream trend. How to effectively reduce the high temperature of various electronic components and reduce the power consumption of the fan wall, and to meet the optim...

AI Technical Summary

Problems solved by technology

[0005] Specifically, the technical problem to be solved by the present invention is: to provide a fan control method based on a machine learning algorithm, which aims to solve the problem of manually performing a large number of tests at different ambient temperatures in the prior art to find the most possible It is time-consuming and labor-intensive, and technical problems that cannot guarantee the best performance

Images