公开(公告)号：US20200050178A1

公开(公告)日：2020-02-13

申请号：US16654978

申请日：2019-10-16

Applicant: Google LLC

Inventor： Jim Gao ,  Christopher Gamble ,  Amanda Gasparik ,  Vedavyas Panneershelvam ,  David Barker ,  Dustin Reishus ,  Abigail Ward ,  Jerry Luo ,  Brian Kim ,  Mark Schwabacher ,  Stephen Webster ,  Timothy Jason Kieper ,  Daniel Fuenffinger ,  Zakerey Bennett

IPC: G05B19/4155 ,  G06N20/00

Abstract: Methods, systems, apparatus and computer program products for implementing machine learning within control systems are disclosed. An industrial facility setting slate can be received from a machine learning system and a determination can be made as to whether to adopt the settings in the industrial facility setting slate. The machine learning model can be a neural network, e.g., a deep neural network, that has been trained, e.g., using reinforcement learning to predict a data setting slate that is predicted to optimize an efficiency of a data center.

公开(公告)号：US12261440B2

公开(公告)日：2025-03-25

申请号：US18032871

申请日：2022-03-18

Applicant: Google LLC

Inventor： Vasileios Kontorinis ,  Dustin Reishus

IPC: G06F1/18 ,  G06F1/26 ,  H02J1/14 ,  H02J3/14 ,  H02J3/28

Abstract: Current imbalance may be detected and components reactively moved to correct the current imbalance. The components, such as rectifiers, machines, etc., may be moved from the most loaded phase to the least loaded phase. The imbalance may be detected at one or more power distribution units. Rebalancing may be performed using a model which preserves the number of components per rack, while limiting per-rack phase imbalance and minimizing imbalance among phases. Once the rebalancing has been computed, instructions for moving components according to the rebalancing may be generated.

公开(公告)号：US20230396066A1

公开(公告)日：2023-12-07

申请号：US18032871

申请日：2022-03-18

Applicant: Google LLC

Inventor： Vasileios Kontorinis ,  Dustin Reishus

IPC: H02J3/28 ,  G06F1/18 ,  H02J3/14 ,  H02J1/14 ,  G06F1/26

CPC classification number: H02J3/28 ,  G06F1/189 ,  H02J3/14 ,  H02J1/14 ,  G06F1/263 ,  H02J2310/16

Abstract: Current imbalance may be detected and components reactively moved to correct the current imbalance. The components, such as rectifiers, machines, etc., may be moved from the most loaded phase to the least loaded phase. The imbalance may be detected at one or more power distribution units. Rebalancing may be performed using a model which preserves the number of components per rack, while limiting per-rack phase imbalance and minimizing imbalance among phases. Once the rebalancing has been computed, instructions for moving components according to the rebalancing may be generated.

公开(公告)号：US11264803B1

公开(公告)日：2022-03-01

申请号：US17242915

申请日：2021-04-28

Applicant: Google LLC

Inventor： Vasileios Kontorinis ,  Dustin Reishus

IPC: G06F1/18 ,  H02J3/28 ,  H02J3/14 ,  H02J1/14 ,  G06F1/26

Abstract: Current imbalance may be detected and components reactively moved to correct the current imbalance. The components, such as rectifiers, machines, etc., may be moved from the most loaded phase to the least loaded phase. The imbalance may be detected at one or more power distribution units. Rebalancing may be performed using a model which preserves the number of components per rack, while limiting per-rack phase imbalance and minimizing imbalance among phases. Once the rebalancing has been computed, instructions for moving components according to the rebalancing may be generated.

公开(公告)号：US20240094692A1

公开(公告)日：2024-03-21

申请号：US17891255

申请日：2022-08-19

Applicant: Google LLC

Inventor： Vasileios Kontorinis ,  Peter Eldridge Bailey ,  Dustin Reishus ,  Claus Congcui Zheng ,  Alejandro Lameda Lopez

IPC: G05B19/042 ,  H02J3/46

CPC classification number: G05B19/042 ,  H02J3/46 ,  G05B2219/2639

Abstract: The present disclosure provides for dynamically deactivating rectifiers to force remaining rectifiers to operate at or near their peak power efficiency. Rectifiers, for example rectifiers on racks of a data center, may operate according to an efficiency curve, based on its current load. Instead of distributing an AC power load across more rectifiers that operate sub-optimally on their efficiency curve, aspects of the disclosure provide for automatically deactivating some rectifiers by lowering voltage set-points. As power load to a rack decreases, the voltage of the current to a rectifier with a reduced voltage set-point falls below the set-point and turns off. Power is automatically redistributed to the remaining active rectifiers. The redistribution increases the power load onto the remaining rectifiers, allowing the rectifiers to perform more efficiently in converting AC power to DC power.

公开(公告)号：US11809164B2

公开(公告)日：2023-11-07

申请号：US17681652

申请日：2022-02-25

Applicant: Google LLC

Inventor： Jim Gao ,  Christopher Gamble ,  Amanda Gasparik ,  Vedavyas Panneershelvam ,  David Barker ,  Dustin Reishus ,  Abigail Ward ,  Jerry Luo ,  Brian Kim ,  Mark Schwabacher ,  Stephen Webster ,  Timothy Jason Kieper ,  Daniel Fuenffinger ,  Zakerey Bennett

IPC: H02J3/46 ,  G06Q10/04 ,  G06Q10/10 ,  G05B19/4155 ,  G06N20/00

CPC classification number: G05B19/4155 ,  G06N20/00 ,  G05B2219/40499

Abstract: Methods, systems, apparatus and computer program products for implementing machine learning within control systems are disclosed. An industrial facility setting slate can be received from a machine learning system and a determination can be made as to whether to adopt the settings in the industrial facility setting slate. The machine learning model can be a neural network, e.g., a deep neural network, that has been trained, e.g., using reinforcement learning to predict a data setting slate that is predicted to optimize an efficiency of a data center.

公开(公告)号：US20220179401A1

公开(公告)日：2022-06-09

申请号：US17681652

申请日：2022-02-25

Applicant: Google LLC

Inventor： Jim Gao ,  Christopher Gamble ,  Amanda Gasparik ,  Vedavyas Panneershelvam ,  David Barker ,  Dustin Reishus ,  Abigail Ward ,  Jerry Luo ,  Brian Kim ,  Mark Schwabacher ,  Stephen Webster ,  Timothy Jason Kieper ,  Daniel Fuenffinger ,  Zakerey Bennett

IPC: G05B19/4155 ,  G06N20/00

Abstract: Methods, systems, apparatus and computer program products for implementing machine learning within control systems are disclosed. An industrial facility setting slate can be received from a machine learning system and a determination can be made as to whether to adopt the settings in the industrial facility setting slate. The machine learning model can be a neural network, e.g., a deep neural network, that has been trained, e.g., using reinforcement learning to predict a data setting slate that is predicted to optimize an efficiency of a data center.