Claims:We claim:
1. A system for optimizing electricity consumption using local data processing and deep learning techniques, the system (100) comprising:
a. one or more nodes (101) for collecting data for a pre-defined set of parameters from a pre-determined vicinity;
b. at least one processing unit (102) for processing the data received from the nodes (101) using an incremental deep learning technique;
c. a cloud network (103) for receiving notifications from the processing unit (102) upon the completion of a pre-defined task performed by the processing unit (102).

2. A system (100) as claimed in claim 1, wherein the nodes (101) are classified as:
device nodes for collecting data from a plurality of devices for a pre-defined set of parameters from a pre-determined vicinity; and
sensor nodes for collecting ambient data from a pre-determined vicinity.

3. A system (100) as claimed in claim 1, wherein data from the nodes (101) is transmitted to the processing unit (102) through a network communication protocol.

4. A system (100) as claimed in claim 1, wherein the processing unit (102) processes the data received from the nodes (101) and notifies the cloud network (103) upon the completion of a pre-defined task through a wired or wireless network infrastructure.

5. A method for optimizing electricity consumption using local data processing and deep learning techniques, the method (200) comprising the steps of:
a. collecting a plurality of data for a pre-defined set of parameters from a pre-determined vicinity by the nodes (101);
b. transmitting the collected data from the nodes (101) to the processing unit (102) through a network communication protocol;
c. processing the data received from the nodes (101) by the processing unit (102) using incremental deep learning techniques, wherein the processing unit (102) does not require a wired or wireless network infrastructure for data processing;
d. automating a pre-defined set of parameters in a pre-determined vicinity based on the data processed by the processing unit (102);
e. notifying the cloud network (103) by the processing unit (102) regarding the outcome of the processed data through a wired or wireless network infrastructure.

6. A method (200) as claimed in claim 5, wherein the processing unit (102) predicts the pattern of usage of a pre-defined set of parameters using incremental deep learning techniques and automates the pre-determined vicinity based on the outcome of the processing unit (102).
, Description:PREAMBLE TO THE DESCRIPTION:
[0001] The following specification particularly describes the invention and the manner in which it is to be performed:

DESCRIPTION OF THE INVENTION
Technical field of the invention
[0002] The present invention discloses a system and method for optimizing electricity consumption using local data processing and deep learning. The present invention particularly discloses a method for collecting a plurality of data from one or more nodes and locally processing the collected data in a processing unit, thereby making the system highly cost-effective.
Background of the invention
[0003] “Deep learning” refers to the usage of artificial neural networks for unsupervised learning using unstructured and/or unlabeled data. Existing technologies which employ deep learning techniques are highly dependent on a cloud network infrastructure for data processing. Consequently, this dependency results in several drawbacks such as increased usage of electricity, increased cloud costs and so on.
[0004] The Canadian Patent Number CA2831621C titled “A computer implemented electrical energy hub management system and method” discloses a system, computer program and method provided for enabling an energy hub for improved management and optimization of energy utilization (consumption, production and storage). In an embodiment, a computer-implemented energy hub management system comprises a micro energy hub configured to communicate with two or more energy components at a premise. An energy optimization engine has an energy component model for each energy component based on each energy component's operating characteristics, the energy optimization engine adapted to receive at least one input from the two or more energy components and an input from an external data source on any external energy utilization restrictions for the micro energy hub. In response to at least one input from the two or more energy components and any external energy utilization restrictions on the micro energy hub, the energy optimization engine is adapted to issue one or more control signals to at least one of the energy components at the premises to optimize energy utilization based on one or more optimization criteria.
[0005] Hence, there exists a need for a simple system for processing a plurality of data locally, thereby making the system highly cost-effective.
Summary of the invention:
[0006] The present invention overcomes the drawbacks of the prior art by disclosing a system and method for optimizing electricity consumption using local data processing and deep learning. The system comprises one or more nodes for collecting data for a pre-defined set of parameters from a pre-determined vicinity, wherein data from the nodes is transmitted to a processing unit through a network communication protocol. The processing unit locally processes the data received from the nodes using an incremental deep learning technique and notifies a cloud network upon the completion of a pre-defined task performed by the processing unit.
[0007] The method for optimizing electricity consumption using local data processing and deep learning comprises the steps of collecting a plurality of data from a pre-determined vicinity by the nodes and transmitting the collected data to the processing unit through a network communication protocol. The processing unit processes the data received from the nodes using incremental deep learning techniques. Further, the processing unit automates a pre-defined set of parameters in a pre-determined vicinity based on the data processed by the processing unit and notifies the cloud network regarding the outcome of the processed data through a wired or wireless network infrastructure.
[0008] The present invention provides a system and method for optimizing electricity consumption using local data processing and deep learning, wherein the system enables data obtained from one or more nodes to be processed locally which reduces the overall cost of operation in comparison to processing the data through the cloud network which is a highly expensive process. Additionally, the present invention reduces the consumption of electricity due to local data processing, thereby reducing the overall carbon footprint.
Brief description of the drawings:
[0009] The foregoing and other features of embodiments will become more apparent from the following detailed description of embodiments when read in conjunction with the accompanying drawings. In the drawings, like reference numerals refer to like elements.
[0010] FIG 1 illustrates a block diagram of a system for optimizing electricity consumption using local data processing and deep learning.
[0011] FIG 2 illustrates a method for optimizing electricity consumption using local data processing and deep learning.
Detailed description of the invention:
[0012] Reference will now be made in detail to the description of the present subject matter, one or more examples of which are shown in figures. Each example is provided to explain the subject matter and not a limitation. Various changes and modifications obvious to one skilled in the art to which the invention pertains are deemed to be within the spirit, scope and contemplation of the invention.
[0013] The present invention discloses a system and method for optimizing electricity consumption using local data processing and deep learning techniques. The system comprises one or more nodes for collecting data for a pre-defined set of parameters from a pre-determined vicinity, wherein the collected data is transmitted to a processing unit through a network communication protocol. The processing unit processes the data received from the nodes and predicts the pattern of usage of a pre-defined set of parameters using incremental deep learning techniques and automates the pre-determined vicinity based on the outcome of the processing unit. Further, the system comprises a cloud network for receiving notifications from the processing unit upon the completion of a pre-defined task performed by the processing unit.
[0014] FIG 1 illustrates a system for optimizing electricity consumption using local data processing and deep learning. The system (100) comprises one or more nodes (101) for collecting data for a pre-defined set of parameters from a pre-determined vicinity, wherein the nodes (101) are classified as device nodes for collecting data from a plurality of devices for a pre-defined set of parameters from a pre-determined vicinity and sensor nodes for collecting ambient data from a pre-determined vicinity. In one embodiment, the device nodes may collect data corresponding to the usage of a pre-determined device such as fan, light, air conditioner and so on by one or more users in a pre-defined vicinity. The sensor nodes employed in a pre-defined vicinity may collect data corresponding to a plurality of variable parameters such as temperature of the pre-defined vicinity.
[0015] Further, the system (100) comprises at least one processing unit (102) for processing the data received from the nodes (101) using an incremental deep learning technique, wherein data from the nodes (101) is transmitted to the processing unit (102) through a network communication protocol such as Message Queuing Telemetry Transport (MQTT) protocol. The processing unit (102) collects data from the nodes (101) and processes it locally without the usage of any wired or wireless network infrastructure. Further, the processing unit (102) predicts the pattern of usage of a pre-defined set of parameters using incremental deep learning techniques and automates the pre-determined vicinity based on the outcome of the processing unit (102). Further, the processing unit (102) sends notifications to a cloud network (103) whenever a pre-defined task performed by the processing unit (102) is completed, wherein the processing unit (102) sends notifications to the cloud network (103) using a wired or wireless network infrastructure.
[0016] For example, the pre-determined vicinity may be a room comprising a plurality of nodes (101) which are further classified into device nodes and sensor nodes. The device nodes collect data corresponding to the usage of a pre-determined appliance such as an air-conditioner which is employed by a user. The sensor nodes collect data corresponding to variable temperature of the room, wherein the data collected by the device nodes and sensor nodes are transmitted to the processing unit (102) using a network communication protocol such as the MQTT protocol. Further, the processing unit (102) locally processes the data transmitted by the nodes (101) and predicts the preferred pattern of usage of the air-conditioner using deep learning techniques. Consequently, the processing unit (102) changes the state of the air-conditioner as per the pattern of usage observed and predicted by the processing unit (102) and notifies the cloud network (103) whenever the state of the air-conditioner is changed by the processing unit (102).
[0017] FIG 2 illustrates a method for optimizing electricity consumption using local data processing and deep learning, wherein the method (200) comprises the steps of collecting a plurality of data for a pre-defined set of parameters from a pre-determined vicinity by the nodes (101) in step (201) and transmitting the collected data from the nodes (101) to the processing unit (102) through a network communication protocol in step (202). Further, the processing unit (102) locally processes the data received from the nodes (101) using incremental deep learning techniques without the usage of a wired or wireless network infrastructure for data processing in step (203).
[0018] The processing unit (102) predicts the pattern of usage of a pre-defined set of parameters using incremental deep learning techniques and automates the pre-determined vicinity based on the data processed by the processing unit (102) in step (204). Upon automating a set of pre-determined parameters in the pre-determined vicinity, the processing unit (102) notifies the cloud network (103) regarding the outcome of the processed data through a wired or wireless network infrastructure in step (205).
[0019] The present invention provides a system and method for optimizing electricity consumption using local data processing and deep learning, wherein the system enables data obtained from one or more nodes (101) to be processed locally which reduces the overall cost of operation in comparison to processing the data through the cloud network (103) which is a highly expensive process. Additionally, the present invention reduces the consumption of electricity due to local data processing, thereby reducing the overall carbon footprint.
[0020] While at least one exemplary embodiment has been presented in the foregoing detailed description, it should be appreciated that a vast number of variations exist.
Reference numbers:
Components Reference Numbers
System 100
Nodes 101
Processing unit 102
Cloud network 103