Beyond Hosting Capacity: Using Shortest Path Methods to Minimize Upgrade Cost Pathways

Nicolas Gensollen, Kelsey Horowitz, Bryan Palmintier, Fei Ding, Barry MatherNational Renewable Energy Laboratory (NREL)

WCPEC-7Waikoloa, HIJune 14th 2018

• Publ

• ication number or conference

NREL | 2

Problem description and formulation

Shortest Path Tool

Algorithm

Results

NREL | 3

Problem description and formulation

Shortest Path Tool

Algorithm

Results

NREL | 4

Target Hosting Capacity (THC)Options Beyond Hosting Capacity

• HC: Hosting Capacity

• THC: Target HC• Goal: Find the

“best” sequence of system upgrades to reach a target HC

• Best = Cheapest• Proactive vs.

reactive approach

NREL | 5

Problem formulation

State at t

State at t

State at t+1State at t+1Action

Weight = Action cost

Transition

NREL | 6

Problem formulation

HC=2C=0

HC=2C=0

Change LTC control for 5$

HC=4C=5

HC=4C=5

HC=6C=20HC=6C=20

HC=2C=0

HC=2C=0

Do nothing for 0$

Reconductoring for 20$

HC=8C=7

HC=8C=7

HC=9C=25HC=9C=25

HC=7C=22HC=7C=22

HC=2C=0

HC=2C=0

Do nothing for 0$

Reconductoring for 20$

Change Regulator settings for 2$

Change LTC control for 5$

Change Regulator settings for 2$H*=8

NREL | 7

Transition ModelTransition Model

State at t

State at t

State at

t+1

State at

t+1

Decisions available:

d0d1d2…

Decisions available:

d0d1d2…

d0d0

Sample deploym

ent 1, p

eak load

Sample deployment 1, min load

Pre-state t+1Pre-state t+1

Pre-state t+1Pre-state t+1

Pre-state t+1Pre-state t+1

Pre-state t+1Pre-state t+1

(…)Sample deployment K, min load

Sample deployment K, peak load

Worst case,Average case…

d1d1

(…)

K PV deployment samples2 Load conditions

NREL | 8

Problem description and formulation

Shortest Path Tool

Algorithm

Results

NREL | 9

Shortest Path Tool

• Written in Python• OpenDSS for co-

simulation• Generic

components• Cross-platform• Still early

development stage

NREL | 10

Problem description and formulation

Algorithm

Shortest Path Tool

Results

3

2

NREL | 11

Shortest Path Tool

• Stack of nodes to expand• Expand: Compute all child nodes• How to insert new nodes in the

stack?• Ranking function F = C + U• C: Cost to get to the node• U: Heuristic cost to go to the

destination from the node• Famous examples: Dijstra and A*

66PRIORITY

PRIORITY

44

11

55

33

NREL | 12

Dijstra (left) A* (right)

Source: Wikipedia

NREL | 13

Shortest Path ToolBranch and Bound

00 33

22

22

33

55

Cost = 2

Cost = 3

Cost = 1

Cost = 3

Cost = 2

Cost = 1

Cost = 2

3 + 1 > 3

3 + 2 > 3

2 + 1.5 > 3

2 + 0.5 < 3

Solution nodewith a cost of 3

These decisions can be removed

But this one has to be evaluated…

Cost = 1.5

Cost = 0.5

H0=1H*=6

H=4

H=5

H=3

H=6

H=5

NREL | 14

Problem description and formulation

Results

Shortest Path Tool

Algorithm

4

2

3

NREL | 15

Decisions and costs

• Change the setpoints of existing regulators ……• Change the setpoints of existing cap. Banks……• Change the LTC control settings ……………………..• Add new capacitor banks

– 600 kvar…………………………………………………………..– 900 kvar…………………………………………………………..– 1200 kvar…………………………………………………………

• Add new regulator …………………………………………• Remove existing capacitor………………………………

$2.5k/unit$7.2k/unit

$8k/unit

$11k/unit$14k/unit$32k/unit$55k/unit

$3k/unit

K. Horowitz, “Distribution system upgrade unit cost database,” National Renewable Energy Laboratory, Tech. Rep., 2017. [Online]. Available: https://www.nrel.gov/solar/distribution-grid-integration-unit-cost-database.html

NREL | 16

Results

• EPRI J1 Feeder• Decision set: 63

possibilities• OPT path ($13k):

– 1. LTC setpoint change– 2. Reg setpoint change– 3. Reg setpoint change

• Only 1195 nodes (0.0075%)

NREL | 17

Results

• All decisions at k=1 are LTC

• All decisions after are Reg setpoint modifications

• Diversity increases with k

• First decision is expensive: greedy cost minimizing will fail here

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

www.nrel.gov

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

This work was authored* by Alliance for Sustainable Energy, LLC, the manager and operator of the National Renewable Energy Laboratory for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding provided by U.S. Department of Energy Office of Energy Efficiency and Renewable Energy XXXX** Office. The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.

*Author: Insert the words “In part” if your research is funded by entities outside of DOE. The wording would say,This work was authored in part by the Alliance . . . .

**Author: insert applicable DOE office and program office, e.g., U.S. Department of Energy Office of Science or U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Solar Energy Technologies Office (spell out full office names; do not use initialisms/acronyms)].

Thank You