# US 2026/0089876 A1 · Fan Control Setting Generation Using Machine Learning

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0089876 A1
- Document ID: US20260089876A1
- Status: Published application
- Kind: A1
- Application number: 18/897,885
- Filed: 2024-09-26
- Published: 2026-03-26
- Earliest claimed: 2024-09-26
- Assignees: NetApp, Inc.
- Applicants: NetApp, Inc.
- Inventors: Ravi Kumar Choubey; Kalpesh Goyal; Parthiban D P; Maheswari A
- Web page: https://patentgrep.com/patent/US20260089876A1

## CPC

- H05K 7/20209: {Thermal management, e.g. fan control}
- G06F 1/206: {comprising thermal management}
- G06N 20/00: Machine learning
- H05K: PRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The disclosure describes system, devices, and methods for fan speed control. In an example implementation, a method for operating a computer-implemented service is provided. The method includes obtaining sensor data from one or more sensors in a data storage environment. The sensor data includes temperature data associated with storage devices in the data storage environment. The method also includes providing an input (e.g., the sensor data) to a machine learning model trained to predict fan control settings of a fan in the data storage environment, determining the fan control setting based on an output from the machine learning model, and controlling the fan based on the fan control setting.

## Claims

### Claim 1

A controller, comprising:
a sensor interface configured to couple to one or more sensors in a data storage environment;
a processing device configured to execute a machine learning model trained to predict fan control settings of fans in the data storage environment, wherein the machine learning model is trained based on training data comprising test states and corresponding fan control settings determined by iteratively changing the fan control settings for given test states until reaching threshold fan control settings for the given test states; and
a fan control interface configured to couple to the fans in the data storage environment;
wherein the sensor interface is configured to:
obtain sensor data from the one or more sensors, wherein the sensor data comprises temperature states associated with storage devices in the data storage environment; and
provide an input to the machine learning model, wherein the input comprises the sensor data;
wherein the processing device is configured to determine, via the machine learning model, a fan control setting of a fan in the data storage environment based on the input; and
wherein the fan control interface is configured to:
obtain the fan control setting from the processing device; and
control the fan based on the fan control setting.

### Claim 2 (depends on 1)

The controller of claim 1, wherein the input further comprises a power savings state and a risk threshold state.

### Claim 3 (depends on 1)

The controller of claim 1, wherein the temperature states comprises temperature values of the storage devices in the data storage environment, processing devices in the data storage environment, and power management units in the data storage environment.

### Claim 4 (depends on 3)

The controller of claim 3, wherein the sensor data further comprises load states associated with the processing devices in the data storage environment, wherein the load states comprise indications of a given loads of processing devices relative to load capacities of the processing devices.

### Claim 5 (depends on 3)

The controller of claim 3, wherein the sensor data further comprises an ambient temperature state of the data storage environment.

### Claim 6 (depends on 1)

The controller of claim 1, wherein the fan control setting comprises a pulse-width modulation (PWM) duty cycle value with which to control a speed of the fan.

### Claim 7 (depends on 1)

The controller of claim 1, wherein to obtain the sensor data from the one or more sensors in the data storage environment, the sensor interface is configured to obtain the sensor data from the one or more sensors via an inter-integrated circuit interface.

### Claim 8

A computing apparatus comprising:
one or more computer-readable storage media; and
program instructions stored on the one or more computer-readable storage media executable by a processing device that, based on being read and executed by the processing device, direct the processing device to:
obtain sensor data from one or more sensors in a data storage environment, wherein the sensor data comprises temperature data associated with storage devices in the data storage environment;
provide an input to a machine learning model trained to predict fan control settings of a fan in the data storage environment, wherein the input comprises the sensor data;
determining a fan control setting based on an output from the machine learning mode; and
control the fan based on the fan control setting.

### Claim 9 (depends on 8)

The computing apparatus of claim 8, wherein the temperature data comprises temperatures of the storage devices in the data storage environment, processing devices in the data storage environment, and power management units in the data storage environment.

### Claim 10 (depends on 9)

The computing apparatus of claim 9, wherein the sensor data further comprises load data associated with the processing devices in the data storage environment.

### Claim 11 (depends on 9)

The computing apparatus of claim 9, wherein the sensor data further comprises ambient temperature data of the data storage environment.

### Claim 12 (depends on 8)

The computing apparatus of claim 8, wherein the fan control setting comprises a pulse-width modulation (PWM) duty cycle value with which to control a speed of the fan.

### Claim 13 (depends on 9)

The computing apparatus of claim 9, wherein the input further comprises a power savings metric and a risk threshold metric, and wherein to determine the fan control setting of the fan, the machine learning model is configured to determine the fan control setting based on further on the power savings metric and the risk threshold metric.

### Claim 14 (depends on 8)

The computing apparatus of claim 8, wherein to obtain the sensor data from the one or more sensors in the data storage environment, the sensor interface is configured to obtain the sensor data from the one or more sensors via an inter-integrated circuit interface.

### Claim 15

A method, comprising:
obtaining sensor data from one or more sensors in a data storage environment, wherein the sensor data comprises temperature data associated with storage devices in the data storage environment;
providing an input to a machine learning model trained to predict fan control settings of a fan in the data storage environment, wherein the input comprises the sensor data;
determining the fan control setting based on an output from the machine learning model; and
controlling the fan based on the fan control setting.

### Claim 16 (depends on 15)

The method of claim 15, wherein the temperature data comprises temperatures of the storage devices in the data storage environment, processing devices in the data storage environment, and power management units in the data storage environment.

### Claim 17 (depends on 16)

The method of claim 16, wherein the sensor data further comprises load data associated with the processing devices in the data storage environment.

### Claim 18 (depends on 16)

The method of claim 16, wherein the sensor data further comprises ambient temperature data of the data storage environment.

### Claim 19 (depends on 15)

The method of claim 15, wherein the fan control setting comprises a pulse-width modulation (PWM) duty cycle value with which to control a speed of the fan.

### Claim 20 (depends on 16)

The method of claim 16, wherein the input further comprises a power savings metric and a risk threshold metric, and wherein determining the fan control setting is further based on the power savings metric and the risk threshold metric.
