Method and apparatus for a self driving silent cooling control system

Patent metadata

Field	Value
Publication number	US-12457706-B2
Application number	US-202318104234-A
Country	US
Kind code	B2
Filing date	Jan 31, 2023
Priority date	Jan 31, 2023
Publication date	Oct 28, 2025
Grant date	Oct 28, 2025

Abstract

An information handling system (IHS) includes a hardware processor, memory device, and a power management unit (PMU) to provide power to the processor and memory device. The IHS further includes a fan and fan motor operatively coupled to a motor driver hardware operated by the hardware processor and an ion emitter/collector blower cooling system including an ion emitter/collector blower, the ion emitter/collector blower including an ion emitter and an ion collector placed in fluidic series with the fan and an ionic driving circuit operatively coupled to the ion emitter via a high voltage to ionize gases at the ion emitter/collector blower to create charged ions that generate an airflow along a voltage field to and through the ion collector. The IHS further includes a hardware embedded controller to execute code instructions of a self-driving cooling system reinforcement learning agent of an ion emitter/controller blower and fan control system, the execution of the self-driving cooling system reinforcement learning agent to adjust and initiate operation of the ion emitter/collector blower and fan based on metrics including operating characteristics of the information handling system from user workload, and based on operational states of the information handling system as determined in a learned policy to operate the ion emitter/collector blower and fan.

First claim

What is claimed is:

1 . An information handling system comprising:
a hardware processor; a memory device; a power management unit (PMU) to provide power to the processor and memory device; a fan and fan motor operatively coupled to a motor driver hardware operated by the hardware processor; an ion emitter/collector blower cooling system including:
an ion emitter/collector blower, the ion emitter/collector blower including an ion emitter and an ion collector placed in fluidic series with the fan;
an ionic driving circuit operatively coupled to the ion emitter via a high voltage to ionize gases at the ion emitter/collector blower to create charged ions that generate an airflow along a voltage field to and through the ion collector; and a hardware embedded controller to execute code instructions of a self-driving cooling system reinforcement learning agent of an ion emitter/controller blower and fan control system, the execution of the self-driving cooling system reinforcement learning agent to adjust and initiate operation of the ion emitter/collector blower and the fan based on metrics including operating characteristics of the information handling system from user workload, and based on operational states of the information handling system as determined in a learned policy to operate the ion emitter/collector blower and the fan.

2 . The information handling system of claim 1 further comprising:
the embedded controller to execute the code instructions of the self-driving cooling system reinforcement learning agent to prioritize the operation of the ion emitter/collector blower over initialization of the fan based on an operating mode of the information handling system being set to a quiet mode.

3 . The information handling system of claim 1 further comprising:
the embedded controller executing of the self-driving cooling system reinforcement learning agent by the embedded controller to determine a weighted total reward metric resulting from detected positive metrics and detected negative metrics resulting from actions executed by the embedded controller and the ion emitter/controller blower and fan control system to operate the ion emitter/collector blower and the fan for feedback into the self-driving cooling system reinforcement learning agent to update adjustments to the ion emitter/collector blower or the fan operation.

4 . The information handling system of claim 1 , wherein the execution of the self-driving cooling system reinforcement learning agent by the embedded controller subtracts a weighted negative metrics resulting from actions executed by the embedded controller to operate the ion emitter/collector blower and the fan from a weighted positive metrics resulting from actions executed by the embedded controller to operate the ion emitter/collector blower and the fan to determine feedback to a reinforcement learning algorithm for control of the ion emitter/collector blower and the fan.

5 . The information handling system of claim 1 further comprising:
the embedded controller executing the ion emitter/collector blower and fan control system to further detect an operating mode, the operating mode including a performance mode requiring an increased level of processor resource utilization, and where the ion emitter/collector blower has reached a maximum operating capacity, the self-driving cooling system reinforcement learning agent initiates setting a fan speed of the fan to a mid-range airflow.

6 . The information handling system of claim 5 further comprising:
the self-driving cooling system reinforcement learning agent initiates the ion emitter/collector blower and fan control system to control the fan speed from the mid-range airflow to a maximum airflow based on temperature feedback from a thermistor in the chassis.

7 . The information handling system of claim 1 further comprising:
the self-driving cooling system reinforcement learning agent initiates the ion emitter/collector blower and fan control system to receive data describing a detected operating mode as not being performance mode and where the ion emitter/collector blower has reached a maximum operating capacity, the ion emitter/collector blower and fan control system then sets a fan speed of the fan to a low-range airflow.

8 . An ion emitter/collector blower and fan cooling system for an information handling system comprising:
an ion emitter/collector blower, the ion emitter/collector blower including an ion emitter and an ion collector placed in fluidic series with a cooling fan; an ionic driving circuit operatively coupled to the ion emitter via a high voltage to ionize gases at the ion emitter/collector blower to create charged ions that generate an airflow along a voltage field to and through the ion collector; and an embedded controller to execute code instructions of a self-driving cooling system reinforcement learning agent and an ion emitter/collector blower and fan control system, the execution of the self-driving cooling system reinforcement learning agent to initiate operation of the ion emitter/collector blower and the cooling fan to based on feedback metrics including operating characteristics of the information handling system of usage workload and operational states of the information handling system iteratively input into a reinforcement learning algorithm to adjust the ion emitter/collector blower and fan control system control of the ion emitter/collector blower and of the cooling fan.

9 . The ion emitter/collector blower and fan cooling system of claim 8 further comprising:
the embedded controller to execute code instructions of the self-driving cooling system reinforcement learning agent to prioritize the operation of the ion emitter/collector blower over initialization of the fan based on an operating mode of the information handling system being set to a quiet mode.

10 . The ion emitter/collector blower and fan cooling system of claim 8 further comprising:
the embedded controller executing the self-driving cooling system reinforcement learning agent by the embedded controller to determine: a weighted rewards metric resulting from detected positive metrics and detected negative metrics resulting from actions executed by the embedded controller and the ion emitter/controller blower and fan control system to operate the ion emitter/collector blower and the cooling fan for feedback into the self-driving cooling system reinforcement learning agent to update adjustments to ion emitter/collector blower or the cooling fan operation.

11 . The ion emitter/collector blower and fan cooling system of claim 8 , wherein the execution of the self-driving cooling system reinforcement learning agent by the embedded controller subtracts weighted negative metrics resulting from actions executed by the embedded controller to operate the ion emitter/collector blower and the cooling fan from weighted positive metrics resulting from actions executed by the embedded controller to operate the ion emitter/collector blower and the cooling fan to determine feedback to a reinforcement learning algorithm for control of the ion emitter/collector blower and the cooling fan.

12 . The ion emitter/collector blower and fan cooling system of claim 8 further comprising:
the embedded controller executing the ion emitter/collector blower and fan control system to detect an operating mode of the information handling system is a performance mode requiring an increased level of processor resource utilization and to detect that the ion emitter/collector blower has reached a maximum operating capacity, and the self-driving cooling system reinforcement learning agent initiating setting

Assignees

• Dell Products Lp

Inventors

• Vichare Nikhil M
• North Travis C
• Markow Mitchell A

Classifications

• H05K7/20209 Primary

  Thermal management, e.g. fan control · CPC title