Bialek T Weather Forecasting
Solar Visibility Thomas Bialek, PhD PE Chief Engineer
April 30, 2015
Daily Wind Power Generation Hourly Data – September 2010 250
Generation (MW)
200
150
100
50
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Hour Peak day of the month: September 27, 2010, Four days with loads over 3,700 MW shown as dashed lines
23
24
9/1/2010 9/2/2010 9/3/2010 9/4/2010 9/5/2010 9/6/2010 9/7/2010 9/8/2010 9/9/2010 9/10/2010 9/11/2010 9/12/2010 9/13/2010 9/14/2010 9/15/2010 9/16/2010 9/17/2010 9/18/2010 9/19/2010 9/20/2010 9/21/2010 9/22/2010 9/23/2010 9/24/2010 9/25/2010 9/26/2010 9/27/2010 9/28/2010 9/29/2010 9/30/2010 Average
2
Net Load and Flexible Capacity Needs – SDG&E Net Load
3
Proposed Generation Connecting to SDG&E Facilities
4
4
Integration of DER Solar & Electric Vehicle Customers
5
Forecasting Engine(s) Flow Prices Forecast
Customer Demand Forecast
Weather Forecast
Intermittent DER Forecast
A common, shared reference model is needed to align forecasts to our system topology
ISO Forecast & Schedule
DERMS
Integrated Forecasts
DMS
EMS
SDG&E Weather Network (150 Stations)
Solar Forecasting: Building On the Work Done in CSI III, Numerical Weather Prediction, Big Data Approach
Privileged & Confidential, Prepared at the Direction of Counsel
88
Drivers – Advanced DMS The limitations of DMS today Powerflow (PF) – Basis for DMS applications. If PF is incorrect, the following DMS applications are incorrect: • Fault Location Isolation Service Restoration (FLISR) • Suggested Switching (SS)
• Fault Location Analysis (FLA) • Feeder Load Management (FLM)
• Version 10: Power flow model does not utilize Distributed (DER’s) characteristics
Energy Resources
• Version 12: Power flow model will utilize DER’s characteristics
9
Questions ?
Thank you. Thomas Bialek Chief Engineer [email protected] www.sdge.com/smartgrid/
10
April 30, 2015
Daily Wind Power Generation Hourly Data – September 2010 250
Generation (MW)
200
150
100
50
0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Hour Peak day of the month: September 27, 2010, Four days with loads over 3,700 MW shown as dashed lines
23
24
9/1/2010 9/2/2010 9/3/2010 9/4/2010 9/5/2010 9/6/2010 9/7/2010 9/8/2010 9/9/2010 9/10/2010 9/11/2010 9/12/2010 9/13/2010 9/14/2010 9/15/2010 9/16/2010 9/17/2010 9/18/2010 9/19/2010 9/20/2010 9/21/2010 9/22/2010 9/23/2010 9/24/2010 9/25/2010 9/26/2010 9/27/2010 9/28/2010 9/29/2010 9/30/2010 Average
2
Net Load and Flexible Capacity Needs – SDG&E Net Load
3
Proposed Generation Connecting to SDG&E Facilities
4
4
Integration of DER Solar & Electric Vehicle Customers
5
Forecasting Engine(s) Flow Prices Forecast
Customer Demand Forecast
Weather Forecast
Intermittent DER Forecast
A common, shared reference model is needed to align forecasts to our system topology
ISO Forecast & Schedule
DERMS
Integrated Forecasts
DMS
EMS
SDG&E Weather Network (150 Stations)
Solar Forecasting: Building On the Work Done in CSI III, Numerical Weather Prediction, Big Data Approach
Privileged & Confidential, Prepared at the Direction of Counsel
88
Drivers – Advanced DMS The limitations of DMS today Powerflow (PF) – Basis for DMS applications. If PF is incorrect, the following DMS applications are incorrect: • Fault Location Isolation Service Restoration (FLISR) • Suggested Switching (SS)
• Fault Location Analysis (FLA) • Feeder Load Management (FLM)
• Version 10: Power flow model does not utilize Distributed (DER’s) characteristics
Energy Resources
• Version 12: Power flow model will utilize DER’s characteristics
9
Questions ?
Thank you. Thomas Bialek Chief Engineer [email protected] www.sdge.com/smartgrid/
10
