Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30...
Transcript of Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30...
![Page 1: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/1.jpg)
Factored Customer Models for Agent-based Smart Grid Simulation
Prashant Reddy Manuela VelosoPrashant Reddy Manuela VelosoMachine Learning Department Computer Science Department
S h l f C t S iSchool of Computer ScienceCarnegie Mellon University
8th Annual CMU Conference on the Electricity IndustryMarch 14, 2012
![Page 2: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/2.jpg)
Smart Grid DistributionSmart Grid Distribution
Source: EPRI
2
![Page 3: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/3.jpg)
Distribution Grid ComplexityDistribution Grid Complexity
Source: IEEE
3
![Page 4: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/4.jpg)
OutlineOutline
1. Agent-based Smart Grid simulation Power Trading Agent Competition
2. Intermediary agent strategies Strategy learning for broker agents
Interactions of multiple learning broker agents
3. Factored customer models Timeseries simulation using Bayesian learning
Decision-theoretic demand side management
4
![Page 5: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/5.jpg)
OutlineOutline
1. Agent-based Smart Grid simulation Power Trading Agent Competition
2. Intermediary agent strategies Strategy learning for broker agents
Interactions of multiple learning broker agents
3. Factored customer models Timeseries simulation using Bayesian learning
Decision-theoretic demand side management
5
![Page 6: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/6.jpg)
Agent based Smart Grid SimulationAgent-based Smart Grid Simulation
Di t ib ti id d l d lti t t Distribution grid modeled as a multi-agent system Focus on emergent economics of self-interested behavior 1
Do not assume rationality nor determinismy Agents contributed by independent research teams Competitive benchmarking to drive innovation
6
1 Leigh Tesfatsion, ACE: A Constructive Approach to Economic Theory. Ch. 16, Handbook of Computational Economics, 2005.
![Page 7: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/7.jpg)
Agent based Smart Grid SimulationAgent-based Smart Grid Simulation
Di t ib ti id d l d lti t t Distribution grid modeled as a multi-agent system Focus on emergent economics of self-interested behavior 1
Do not assume rationality nor determinismy Agents contributed by independent research teams Competitive benchmarking to drive innovation
Power Trading Agent Competition (Power TAC) Annual tournament at major AI or MAS conferencej
Builds upon experience with other TAC domains Simulation platform available for offline research
Assumes liberalized retail markets Assumes liberalized retail markets Customers have choice of “broker agents”
7
1 Leigh Tesfatsion, ACE: A Constructive Approach to Economic Theory. Ch. 16, Handbook of Computational Economics, 2005.
![Page 8: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/8.jpg)
Power TAC ScenarioPower TAC Scenario
CompetitionParticipants
SimulationInfrastructure
http://www.powertac.org
8
p // p g
![Page 9: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/9.jpg)
9
![Page 10: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/10.jpg)
OutlineOutline
1. Agent-based Smart Grid simulation Power Trading Agent Competition
2. Intermediary agent strategies Strategy learning for broker agents
Interactions of multiple learning broker agents
3. Factored customer models Timeseries simulation using Bayesian learning
Decision-theoretic demand side management
10
![Page 11: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/11.jpg)
Strategy Learning for Broker AgentsStrategy Learning for Broker Agents
Reddy & Veloso. Strategy Learning for Autonomous Agents in Smart Grid Markets.Twenty-Second Intl Joint Conf on Artificial Intelligence (IJCAI) Barcelona 2011
11
Twenty Second Intl. Joint Conf. on Artificial Intelligence (IJCAI), Barcelona, 2011.
![Page 12: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/12.jpg)
Interactions of Multiple Learning Broker AgentsInteractions of Multiple Learning Broker Agents
Reddy & Veloso. Learned Behaviors of Multiple Autonomous Agents in Smart Grid Markets.Twenty-Fifth AAAI Conf on Artificial Intelligence San Francisco 2011
12
Twenty Fifth AAAI Conf. on Artificial Intelligence, San Francisco, 2011.
![Page 13: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/13.jpg)
OutlineOutline
1. Agent-based Smart Grid simulation Power Trading Agent Competition
2. Intermediary agent strategies Strategy learning for broker agents
Interactions of multiple learning broker agents
3. Factored customer models Timeseries simulation using Bayesian learning
Decision-theoretic demand side management
13
![Page 14: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/14.jpg)
Factored Customer ModelsFactored Customer Models
P TAC i l d f d t l d t ti ti l d l Power TAC includes fundamental and statistical models Trade-off on behavioral accuracy vs. scalability
14
![Page 15: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/15.jpg)
Factored Customer ModelsFactored Customer Models
P TAC i l d f d t l d t ti ti l d l Power TAC includes fundamental and statistical models Trade-off on behavioral accuracy vs. scalability
Goals for statistical models:1. Representation
a. Represent diverse types of consumers and producersb. Represent varying levels of granularity
2 Automated learning2. Automated learning Learn parameters from “real-world” data
3. Facilitate agent algorithms Develop algorithms that can be applied in real-world
15
![Page 16: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/16.jpg)
Factored Customer Model RepresentationFactored Customer Model Representation
{ }N { }N { }MC = h{B i}Ni= 1,{Si}
Ni= 1,U i, B i = {O ij}
M ij= 1
16
![Page 17: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/17.jpg)
Factored Customer Model RepresentationFactored Customer Model Representation
{ }N { }N { }MC = h{B i}Ni= 1,{Si}
Ni= 1,U i, B i = {O ij}
M ij= 1
Factored Customer
Load BundleLoad Bundle
Load Originator
Load Originator
Tariff Subscriber
Load Bundle
Utility Optimizer
Load OriginatorLoad OriginatorTariff Subscriber
17
![Page 18: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/18.jpg)
Factored Customer Model RepresentationFactored Customer Model Representation
{ }N { }N { }MC = h{B i}Ni= 1,{Si}
Ni= 1,U i, B i = {O ij}
M ij= 1
Factored Customer
Load Bundle
Calendar- Time of day- Day of week
Tariff Terms- Fixed Payments- Variable Rates
Contract Time Load Bundle
Load Originator
Load Originator
Tariff SubscriberDay of week
- Month of year
Weather- Temperature
Cloud Cover
- Contract Time- Exit Penalties- Reputation
Load Bundle
Utility Optimizer
Load Originator
- Cloud Cover- Wind Speed- Wind Direction
Tariff Terms
Tariff Evaluation- Inertia- Rationality
Load OriginatorTariff Subscriber - Price Elasticity- Control Events
Load Origination- Reactivity- Receptivity- Rationality
18
![Page 19: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/19.jpg)
OutlineOutline
1. Agent-based Smart Grid simulation Power Trading Agent Competition
2. Intermediary agent strategies Strategy learning for broker agents
Interactions of multiple learning broker agents
3. Factored customer models Timeseries simulation using Bayesian learning
Decision-theoretic demand side management
19
![Page 20: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/20.jpg)
Timeseries Simulation using Bayesian LearningTimeseries Simulation using Bayesian Learning
Gi ll l f b d d t fit d l th t Given small samples of observed data, fit a model that can generate a long range time series forecast Use “similar” samples to improve the fitp p
20
![Page 21: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/21.jpg)
Timeseries Simulation using Bayesian LearningTimeseries Simulation using Bayesian Learning
Gi ll l f b d d t fit d l th t Given small samples of observed data, fit a model that can generate a long range time series forecast Use “similar” samples to improve the fitp p
ARIMA forecasting over long range is poor
Consumption
Time (Hours)
Consumption
21
Time (Hours)
![Page 22: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/22.jpg)
Bayesian Timeseries Simulation MethodBayesian Timeseries Simulation Method
Use similarUse similartraining data
Fit hierarchicalBayesian model
usingGibbs samplingG bbs sa p g
Latent factorLatent factoroptimization to boost fit
Forecast from small sample
22
![Page 23: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/23.jpg)
Boosted Bayesian Timeseries SimulationBoosted Bayesian Timeseries Simulation
Consumption
Time (Hours)
Consumption
Time (Hours)
23
![Page 24: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/24.jpg)
Boosted Bayesian Forecasting Accuracy Boosted Bayesian Forecasting Accuracy
24
![Page 25: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/25.jpg)
Boosted Bayesian Forecasting Accuracy Boosted Bayesian Forecasting Accuracy
25
![Page 26: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/26.jpg)
OutlineOutline
1. Agent-based Smart Grid simulation Power Trading Agent Competition
2. Intermediary agent strategies Strategy learning for broker agents
Interactions of multiple learning broker agents
3. Factored customer models Timeseries simulation using Bayesian learning
Decision-theoretic demand side management
26
![Page 27: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/27.jpg)
Decision theoretic DSMDecision-theoretic DSM
M lti l d i i ki i t di i Multi-scale decision-making in two dimensions1. Temporal: Metering period vs. tariff contract period2. Contextual: Individual load vs. bundle/customer/co-op/ / p
27
![Page 28: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/28.jpg)
Decision theoretic DSMDecision-theoretic DSM
M lti l d i i ki i t di i Multi-scale decision-making in two dimensions1. Temporal: Metering period vs. tariff contract period2. Contextual: Individual load vs. bundle/customer/co-op/ / p
argm axyt
U S (pt,yt,UN (yt))
argm axz2 Z t0
U S� (P
zt0,Yt0,U
N� (Yt0))
Self Utility Price Profile Load Profile Neighborhood Utility
28
![Page 29: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/29.jpg)
Decision theoretic DSMDecision-theoretic DSM
M lti l d i i ki i t di i Multi-scale decision-making in two dimensions1. Temporal: Metering period vs. tariff contract period2. Contextual: Individual load vs. bundle/customer/co-op/ / p
argm axyt
U S (pt,yt,UN (yt))
P b bili ti lti tt ib t tilit d l
argm axz2 Z t0
U S� (P
zt0,Yt0,U
N� (Yt0))
Probabilistic multi-attribute utility model:
P r(z) =eλ U
S⌧(z)
PZ eλ U S
⌧(z)
29
![Page 30: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/30.jpg)
Decision theoretic DSMDecision-theoretic DSM
M lti l d i i ki i t di i Multi-scale decision-making in two dimensions1. Temporal: Metering period vs. tariff contract period2. Contextual: Individual load vs. bundle/customer/co-op/ / p
argm axyt
U S (pt,yt,UN (yt))
P b bili ti lti tt ib t tilit d l
argm axz2 Z t0
U S� (P
zt0,Yt0,U
N� (Yt0))
Monte Carlo Sampling Probabilistic multi-attribute utility model:
o e Ca o Sa p g
P r(z) =eλ U
S⌧(z)
PZ eλ U S
⌧(z)
30
![Page 31: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/31.jpg)
Peak Shifting (Herding) BehaviorPeak Shifting (Herding) Behavior
31
Ramchurn, et al. Agent-Based Control for Decentralised Demand Side Management in the Smart Grid. Autonomous Agent and Multi-Agent Systems (AAMAS), 2011.
![Page 32: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/32.jpg)
Household Demand ShiftingHousehold Demand Shifting
B d d t f G ’ M R i j t Based on data from Germany’s MeRegio project
nC
onsu
mpt
ion
C
Time (Hours)
32
( )
![Page 33: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/33.jpg)
Household Demand ShiftingHousehold Demand Shifting
L i d t i f 10% Lower variance and cost savings of ~10%
Balancing
Combined
Temporal
g
Original
Consumption
33
![Page 34: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/34.jpg)
DSM Deployment OptionsDSM Deployment Options
ARM C t A8 d Zi b S CARM Cortex-A8 and Zigbee SoC
34
![Page 35: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/35.jpg)
ConclusionConclusion
S Summary Versatile customer model representation Decision-theoretic algorithms for DSMg Bayesian learning algorithms for timeseries simulation
F t W k Future Work Customer type-specific factor modeling Non-cooperative decision-making modelsg
Participating in Power TACHosted at AAMAS Valencia June 2012 Hosted at AAMAS, Valencia, June 2012
More information at http://www.powertac.org
35
![Page 36: Factored Customer Models for Agent-based Smart Grid ...electriconf/2012/slides/Section...2015/03/30 · 1. Temporal: Metering period vs. tariff contract period 2. Contextual: Individual](https://reader035.fdocuments.us/reader035/viewer/2022071420/61193e2904bb0002f9385642/html5/thumbnails/36.jpg)
Factored Customer Models for Agent-based Smart Grid Simulation
Prashant Reddy Manuela [email protected] [email protected]
School of Computer SciencepCarnegie Mellon University