Resilient decision systems and methods
Disclosed are systems and methods for utilizing a unique elastic command and control architecture to incorporate certain resiliency qualities in power grid management and outage mitigation.
1. A method for operating a multi-partner enclave in a power grid domain of nodes, comprising:
(i) selecting a partner node from a subset of the domain nodes based on proximity to the subset nodes and establishing a heuristic model in the partner node;
(ii) receiving non-operational data from the subset nodes in the partner node;
(iii) updating the heuristic model based on the received non-operational data;
(iv) receiving operational data from the subset nodes in the partner node;
(v) evaluating the received operational data within the partner node to identify an adjustment to an electrical property of the subset nodes by executing a compressed decision path comprising:
providing the operational data to the heuristic model to generate a heuristic vote;
providing the non-operational data to a Lower Ambiguity, Higher Uncertainty (LAHU) model to generate a Lower Ambiguity, Higher Uncertainty (LAHU) vote;
combining the heuristic vote and the Lower Ambiguity, Higher Uncertainty (LAHU) vote to generate a compressed decision vote; and
(vi) sending a command to implement the adjustment from the partner node to an electrical device in the subset nodes.
2. The method of claim 1 , further comprising introducing a headquarters node in communication with the partner node, wherein the partner node receives a command from the headquarters node.
3. The method of claim 1 , comprising identifying a key node in the subset nodes.
4. The method of claim 3 , wherein the electrical device is located at the key node and the partner node sends the command directly to the key node.
5. The method of claim 4 , wherein identifying a key node in the subset nodes comprises selecting the key node based on comparing the number of electrical customers supplied by the key node to the number of electrical customers supplied by the rest of the subset nodes.
6. The method of claim 4 , wherein identifying a key node in the subset nodes comprises selecting the key node based on a pre-prioritized set of electrical customers.
7. The method of claim 1 , wherein the power grid domain of nodes comprises a plurality of substations, comprising:
selecting distribution subsets of the domain nodes corresponding to the power grid components supplied by the substations; and
repeating steps (i) to (vi) for each distribution subset.
8. The method of claim 1 , wherein evaluating the received operational data within the partner node to identify an adjustment to an electrical property of the subset nodes comprises applying a voting algorithm.
9. The method of claim 8 , wherein the voting algorithm employs a comparator threshold.
10. The method of claim 8 , wherein evaluating the received operational data within the partner node to identify an adjustment to an electrical property of the subset nodes comprises selecting a decision path based on available decision cycle time.
11. The method of claim 1 , wherein the non-operational data comprises phasor measurement unit data.
12. The method of claim 1 , wherein the non-operational data comprises digital fault record data.
13. The method of claim 1 , wherein
combining the heuristic vote and the Lower Ambiguity, Higher Uncertainty (LAHU) vote to generate a compressed decision vote comprises comparing to an output of a historical model applied to the operational data and the non-operational data.
14. The method of claim 1 , wherein the operational data comprises a series input of operational state data; and wherein evaluating the received operational data within the partner node to identify an adjustment to an electrical property of the subset nodes comprises executing an uncompressed decision path, comprising:
providing the series input of operational state data to a historical baseline model to generate a historical vote;
providing the series input of operational state data and the non-operational data to a Higher Ambiguity, Lower Uncertainty (HALU) model to generate a Higher Ambiguity, Lower Uncertainty (HALU) vote;
combining the historical vote and the Higher Ambiguity, Lower Uncertainty (HALU) vote to generate an uncompressed decision vote.
15. The method of claim 14 , comprising selecting a decision output from among the uncompressed decision vote or the compressed decision vote, wherein the decision output is used to identify the adjustment.
16. The method of claim 1 , comprising, after sending the command to implement the adjustment, receiving a second set of non-operational data from the subset nodes in the partner node and updating the heuristic model based on the second set of non-operational data.
17. The method of claim 1 , comprising establishing a historical model in the partner node and updating the historical model based on the received operational data.
18. The method of claim 17 , comprising after sending the command to implement the adjustment, receiving a second set of operational data from the subset nodes in the partner node and updating the historical model based on the second set of operational data.
19. The method of claim 18 , comprising, after sending the command to implement the adjustment, receiving a second set of non-operational data from the subset nodes in the partner node and updating the heuristic model based on the second set of non-operational data.
20. The method of claim 19 , wherein evaluating the received operational data within the partner node to identify an adjustment to an electrical property of the subset nodes comprises selecting a decision path based on available decision cycle time.