Granted Patent B2

US 12,494,642 · App. 17/493,833 · Granted Dec 9, 2025

Inventors: Madhu Akhilesham (Pune, IN); Shikhar Kwatra (San Jose, CA); Venkata Vara Prasad Karri (Visakhapatnam, IN); Shailendra Moyal (Pune, IN); Akash U. Dhoot (Pune, IN)

Assignee: International Business Machines Corporation

H02J3/004H02J3/003H02J2300/20

Abstract

Claims (73)

1 . A computer-implemented process for distributing energy, the computer-implemented process comprising:

extracting signal data from audio signals captured by a sound transducer positioned within a determined area;

determining, based on the extracted signal data, a number of energy converters;

determining, based on the extracted signal data, an energy-capture position, within the determined area, for each energy converter of the determined number of energy converters;

converting, by the determined number of energy converters positioned within the determined area, sound energy to electrical energy;

classifying, with a device classifier engine, each device of a plurality of devices, wherein the classifying indicates an energy requirement of each device of the plurality of devices;

allocating, with an energy allocator engine, based on the classifying, the electrical energy to one or more devices of the plurality of devices; and

controlling, with a delivery controller, delivery of the allocated electrical energy to each device of the one or more devices using a switching mechanism to create a transmission channel for each device of the one or more devices.

2 . The computer-implemented process of claim 1 , further comprising:

predicting, based on the extracted signal data, a quantity of the electrical energy generated within a determined time by the converting.

3 . The computer-implemented process of claim 1 , wherein the energy-capture position is a position where greatest amount of the sound energy for the conversion into the electrical energy is captured.

4 . The computer-implemented process of claim 1 , further comprising:

predicting times for the conversion of the sound energy to the electrical energy based on recognizing patterns of sounds captured by one or more sound transducers during a determined time interval, wherein the one or more sound transducers include the sound transducer.

5 . The computer-implemented process of claim 1 , further comprising:

determining, based on a plurality of digital twin simulations, the number of energy converters and, for each energy converter of the determined number of energy converters, the energy-capture position within the determined area, wherein the plurality of digital twin simulations includes:

simulating the sound energy from different sounds occurring within differently dimensioned areas, and

simulating the converting of the sound energy generated by the different sounds to the electrical energy using various arrangements of different energy converters at different positions with a simulated area.

6 . The computer-implemented process of claim 1 , wherein

the classifying includes predicting, for each device of the plurality of devices, a time-based energy requirement based on patterns of energy consumption over a determined time interval.

7 . The computer-implemented process of claim 1 , wherein

the allocating includes storing a portion of the electrical energy based on at least one of a predicted event likely to affect future converting or user-specified input specifying a future need for the portion of the electrical energy.

8 . The computer-implemented process of claim 1 , further comprising:

positioning at least one energy converter of the determined number of energy converters within the determined area using a self-propelled vehicle and the sound transducer, wherein

the sound transducer is configured to determine an intensity of sounds occurring within the determined area at different locations during a determined time interval; and

selecting a location from the different locations based on learning the energy-capture position using reinforced learning that implements a gradient policy.

9 . The computer-implemented process of claim 1 , further comprising:

performing signal processing on the signal data to determine a frequency spectrum of each distinct sound of one or more distinct sounds identifiable from the audio signals;

mapping each distinct sound of the one or more distinct sounds to a unique frequency signature determined based on the frequency spectrum of each distinct sound of the one or more distinct sounds;

identifying, based on the unique frequency signature, one or more times during a determined time interval that each distinct sound of the one or more distinct sounds occurs within the determined area;

determining a quantity of the electrical energy producible by the converting of the sound energy generated by each distinct sound of the one or more distinct sounds; and

correlating the one or more distinct sounds with the one or more devices by matching the energy requirement of the one or more devices and the quantity of the electrical energy producible by the one or more distinct sounds, wherein the allocating is based at least in part on the correlating.

10 . A system, comprising:

a processor configured to:

extract signal data from audio signals captured by a sound transducer positioned within a determined area;

determine, based on the extracted signal data, a number of energy converters;

determine, based on the extracted signal data, an energy-capture position, within the determined area, for each energy converter of the determined number of energy converters;

convert, based on the determined number of energy converters positioned within the determined area, sound energy to electrical energy;

classify each device of a plurality of devices, wherein the classification indicates an energy requirement of each device of the plurality of devices;

allocate, based on the classification, the electrical energy to one or more devices of the plurality of devices; and

control delivery of the allocated electrical energy to each device of the one or more devices using a switching mechanism to create a transmission channel for each device of the one or more devices.

11 . The system of claim 10 , wherein the processor is further configured to:

predict, based on the extracted signal data, a quantity of the electrical energy generated within a determined time by the conversion.

12 . The system of claim 10 , wherein the processor is further configured to:

predict times for the conversion of the sound energy to the electrical energy based on recognition of patterns of sounds captured by one or more sound transducers during a determined time interval, wherein the one or more sound transducers include the sound transducer.

13 . The system of claim 10 , wherein the processor is further configured to:

determine, based on a plurality of digital twin simulations, the number of energy converters and, for each energy converter of the determined number of energy converters, the energy-capture position within the determined area, wherein the plurality of digital twin simulations includes:

simulation of the sound energy from different sounds occurring within differently dimensioned areas, and

simulation of the conversion of the sound energy generated by the different sounds to the electrical energy using various arrangements of different energy converters at different positions with a simulated area.

14 . The system of claim 10 , wherein

the classification includes prediction, for each device of the plurality of devices, of a time-based energy requirement based on patterns of energy consumption over a determined time interval.

15 . The system of claim 10 , wherein

the allocation includes storage of a portion of the electrical energy based on at least one of a predicted event likely to affect future conversion or user-specified input that specifies a future need for the portion of the electrical energy.

16 . The system of claim 10 , wherein the processor is further configured to:

position at least one energy converter of the determined number of energy converters within the determined area using a self-propelled vehicle and the sound transducer, wherein

the sound transducer is configured to determine an intensity of sounds that occur within the determined area at different locations during a determined time interval; and

select a location from the different locations based on learning the energy-capture position using reinforced learning that implements a gradient policy.

17 . The system of claim 10 , wherein the processor is further configured to:

perform signal processing on the signal data to determine a frequency spectrum of each distinct sound of one or more distinct sounds identifiable from the audio signals;

map each distinct sound of the one or more distinct sounds to a unique frequency signature determined based on the frequency spectrum of each distinct sound of the one or more distinct sounds;

identify, based on the unique frequency signature, one or more times during a determined time interval that each distinct sound of the one or more distinct sounds occurs within the determined area;

determine a quantity of the electrical energy producible by the conversion of the sound energy generated by each distinct sound of the one or more distinct sounds; and

correlate the one or more distinct sounds with the one or more devices by a match of the energy requirement of the one or more devices and the quantity of the electrical energy producible by the one or more distinct sounds, wherein the allocation is based at least in part on the correlation.

18 . A computer program product, the computer program product comprising:

one or more computer-readable storage media and program instructions collectively stored on the one or more computer-readable storage media, the program instructions executable by a processor to cause the processor to initiate operations including:

extracting signal data from audio signals captured by a sound transducer positioned within a determined area;

determining, based on the extracted signal data, a number of energy converters;

determining, based on the extracted signal data, an energy-capture position, within the determined area, for each energy converter of the determined number of energy converters;

converting, based on the determined number of energy converters positioned within the determined area, sound energy to electrical energy;

classifying each device of a plurality of devices, wherein the classifying indicates an energy requirement of each device of the plurality of devices;

allocating, based on the classifying, the electrical energy to one or more devices of the plurality of devices; and

controlling delivery of the allocated electrical energy to each device of the one or more devices using a switching mechanism to create a transmission channel for each device of the one or more devices.

19 . The computer program product of claim 18 , wherein the program instructions are executable by the processor to cause the processor to initiate the operations further including:

predicting, based on the extracted signal data, a quantity of the electrical energy generated within a determined time by the converting.

Assignments (3)

Continuity (1)

References Cited (28)