System and method for learning the structure of deep convolutional neural networks
US-2019042911-A1 · Feb 7, 2019 · US
US11268994B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-11268994-B2 |
| Application number | US-202016827092-A |
| Country | US |
| Kind code | B2 |
| Filing date | Mar 23, 2020 |
| Priority date | Mar 25, 2019 |
| Publication date | Mar 8, 2022 |
| Grant date | Mar 8, 2022 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A method of determining the topology of a portion of the electrical distribution grid such as a feeder based on voltage and current measurements. The method employs Bayesian-based topology detection, where voltage and current data is measured at switches in the feeder at numerous sample times. A voltage variance is calculated for each switch, along with a voltage correlation coefficient for each pair of switches. Voltage variance is used to establish a preliminary position of a switch in the feeder, where switches closest to the substation source are expected to exhibit the least variance, and vice versa. The voltage correlation coefficient is used to provide a first determination of whether two switches are in series with each other in a parent-child relationship, and a current magnitude comparison is used to confirm or refute the preliminary parent-child relationship between the two nodes.
Opening claim text (preview).
What is claimed is: 1. A method for determining a topology of a distribution grid feeder, said method comprising: providing voltage data and current data, measured at a plurality of nodes in the feeder at a plurality of sample times, to a computer having a processor and memory; calculating, using the computer, a voltage variance for each of the nodes and a correlation coefficient for pairs of child nodes and parent nodes of the plurality of nodes based on the voltage data for the plurality of sample times; assigning all nodes to a set of undiscovered nodes and creating an empty set of discovered parentless nodes; identifying a first node as a node from the set of undiscovered nodes with a highest voltage variance; designating the first node as a leaf node; determining if a second node exists in the set of discovered parentless nodes, where the second node has a highest correlation coefficient with the first node and the second node has a lower average current than the first node; when the second node exists, designating the first node as a parent of the second node, removing the second node from the set of discovered parentless nodes and returning to determining if a second node exists; and when the second node does not exist, removing the first node from the set of undiscovered nodes, adding the first node to the set of discovered parentless nodes and returning to identifying a first node. 2. The method according to claim 1 wherein each of the nodes is a switch configured for fault location, isolation and service restoration (FLISR). 3. The method according to claim 2 wherein each of the switches includes current and voltage sensors, a processor with memory in communication with the sensors, a communication module configured to communicate with the computer, and an actuator configured to open or close the switch upon receiving a command from the processor. 4. The method according to claim 1 wherein the feeder is a three phase network and the voltage and current data are measured on any one of the phases. 5. The method according to claim 1 wherein the feeder includes a source connected to one of the nodes and one or more open tie switches downstream of a different ones of the nodes, where each of the open tie switches separate the feeder from an adjacent feeder. 6. The method according to claim 5 wherein the source is a voltage-reducing substation connected to a power transmission grid. 7. The method according to claim 1 wherein the source is connected to a remaining discovered parentless node when the topology is finally determined. 8. The method according to claim 1 wherein the variance is calculated as a square of a standard deviation of the voltage data for each of the nodes. 9. The method according to claim 1 wherein the correlation coefficient is calculated as a covariance of the voltage data for each pair of the nodes divided by a product of standard deviations of the voltage data for both nodes in the pair. 10. The method according to claim 1 further comprising using the topology of the feeder for grid operational control. 11. The method according to claim 1 further comprising re-determining the topology of the feeder after a fault location, isolation and service restoration (FLISR) event. 12. An electrical network system with self-topology determination capability, said system comprising: a plurality of nodes in the network system, each of the nodes measuring voltage data and current data at a plurality of sample times; and a computer having a processor and memory, said computer being in communication with the plurality of nodes and receiving the voltage data and the current data, where the computer is configured with an algorithm performing steps of; calculating a voltage variance for each of the nodes and a correlation coefficient for pairs of child nodes and parent nodes of the plurality of nodes based on the voltage data for the plurality of sample times, assigning all nodes to a set of undiscovered nodes and creating an empty set of discovered parentless nodes, identifying a first node as a node from the set of undiscovered nodes with a highest voltage variance, designating the first node as a leaf node, determining if a second node exists in the set of discovered parentless nodes, where the second node has a highest correlation coefficient with the first node and the second node has a lower average current than the first node, when the second node exists, designating the first node as a parent of the second node, removing the second node from the set of discovered parentless nodes and returning to determining if a second node exists, and when the second node does not exist, removing the first node from the set of undiscovered nodes, adding the first node to the set of discovered parentless nodes and returning to identifying a first node. 13. The system according to claim 12 wherein each of the nodes is a switch configured for fault location, isolation and service restoration (FLISR), and where each of the switches includes current and voltage sensors, a processor with memory in communication with the sensors, a communication module configured to communicate with the computer, and an actuator configured to open or close the switch upon receiving a command from the processor. 14. The system according to claim 12 wherein the network system is a three phase distribution feeder and the voltage and current data are measured on any one of the phases. 15. The system according to claim 14 wherein the feeder includes a source connected to one of the nodes and one or more open tie switches downstream of a different ones of the nodes, where each of the open tie switches separate the feeder from an adjacent feeder. 16. The system according to claim 12 wherein the variance is calculated as a square of a standard deviation of the voltage data for each of the nodes. 17. The system according to claim 12 wherein the correlation coefficient is calculated as a covariance of the voltage data for each pair of the nodes divided by a product of standard deviations of the voltage data for both nodes in the pair. 18. A computer program for determining a topology of a distribution grid feeder, said computer program comprising instructions for causing a computer to perform steps including: receiving and storing voltage data and current data measured at a plurality of nodes in the feeder at a plurality of sample times; calculating a voltage variance for each of the nodes and a correlation coefficient for pairs of child nodes and parent nodes of the plurality of nodes based on the voltage data for the plurality of sample times; assigning all nodes to a set of undiscovered nodes and creating an empty set of discovered parentless nodes; identifying a first node as a node from the set of undiscovered nodes with a highest voltage variance; designating the first node as a leaf node; determining if a second node exists in the set of discovered parentless nodes, where the second node has a highest correlation coefficient with the first node and the second node has a lower average current than the first node; when the second node exists, designating the first node as a parent of the second node, removing the second node from the set of discovered parentless nodes and returning to determining if a second node exists; and when the second node does not exist, removing the first node from the set of undiscovered nodes, adding the first node to the set of discovered parentless nodes and returning to identifying a first node. 19
Monitoring network conditions, e.g. electrical magnitudes or operational status · CPC title
Grid-level management of power transmission or distribution systems, e.g. load flow analysis or active network management · CPC title
by using digital technique · CPC title
Arrangements for monitoring electric power systems, e.g. power lines or loads; Logging · CPC title
Circuit arrangements for AC mains or AC distribution networks · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.