In our previous TerraBlog post, “How CCAs can model and deploy battery energy storage and DERs”, we shared a case study for using the Building Efficiency Optimization (BEO) software to reduce a community choice agency’s (CCAs) cost to service their customers. In this blog post we will share our next analysis and how it differs from the first.

The initial study assessed how an energy storage system would impact commercial NEM customers on the PG&E rate schedule A-6. We modelled a battery schedule focused on charging batteries with solar exports in the afternoon, and discharging the batteries between 4 and 9 pm, with the goal of reducing the excess generation credits (NEM payouts) for a CCA. In our newest analysis we assessed residential NEM customers with the same goal. A subset of the total residential NEM customers was chosen, customers on PG&E rate plan schedule E-6, since E-6 is a time-of-use (TOU) based rate schedule and customers on this rate schedule account for 27% of all NEM residential customers of the CCA being assessed.

Customers with a TOU rate schedule were assessed because rate schedules with strong differentials in time of use period pricing present the stronger opportunity for energy storage systems to produce value.

Once we selected the subset of customers, we proceeded with a K-means clustering analysis to identify customers with similar usage patterns. We took a two-step approach: first clustering customer load profiles for maximum load and then clustering for an average normalized load. Clustering for maximum load allowed us to visualize which customers had the largest differential between the peak and trough and highest peaks, as shown in the chart below.

The above cluster represents 170 of the initial 1347 customers on E-6. These customers were further clustered based on average normalized load. The average normalized load clustering allowed us to visualize which customers with these higher demands have the most solar exports. Three of the six new clusters had a significant amount of solar exports as can be seen in the following cluster graph. All three clusters had profiles similar to this one.

We then had a set of 98 customers that visually seem to be the best candidates for further evaluation. A few different battery operational strategies were assessed, all of which control for charging from solar exports, since the goal is to reduce the CCA’s NEM payout. The discharging strategies analyzed were as follows:

  1. Discharging between 4 and 9pm.
  2. Discharging when the rates are lowest, maximizing the NEM payout reduction
  3. Discharging when the tCO2/MWh values are the highest, minimizing 2030 Greenhouse Gas (GHG) emission content based on the Clean Net Short (CNS) 2030 table

All three of these strategies were then compared on efficacy of

  • Load shifting
  • Demand reduction
  • NEM payout reduction
  • GHG reduction across all of the CNS tables and the California Air Resources Board (CARB) unspecified power content method
  • CO2 savings per dollars of NEM payout reduction

Based on the net benefit across all of these metrics, the third operational strategy (Strategy 3) was found to produce the best results.

The 98 customers were then reduced to include only those customers with a NEM payout reduction of over $100 annually, since maximizing this reduction is the ultimate goal of the analysis. We were then left with 32 customers for whom it would be ideal to add a 10 kW/24 kWh, 2 hour energy storage system (sized on maximum load of each customer) under Strategy 3. Note, our analysis modeled a maximum discharge of 20 kWh from each battery to reserve capacity for resiliency and back-up power needs for the customer.

The following graphs represent load shifting and demand reduction performance of the batteries for the 32 customer set under Strategy 3. In the “Annual Total Load Shifting” graph, the grey lines represent the pre-storage monthly usage and the blue lines represent the post-storage monthly usage. A significant load shift performance is apparent from the reduced peak and trough.

Annual Total Load Shifting

In the following “Monthly Load Shifting” and “Monthly Max Demand Reduction” graphs, the green lines represent the pre-storage monthly usage/max demand and the blue lines are the post-storage monthly usage/max demand. In both graphs, the x-axis is hours of the day. In the “Monthly Load Shifting” graph, the y-axis is load. In the “Monthly Max Demand Reduction” graph, the y-axis is max demand. Again, substantial impacts of the batteries are apparent from the reduced peaks and troughs, in both the load shifting and demand reduction evaluations.

Monthly Load Shifting

Monthly Max Demand Reduction

Reducing the load troughs generates NEM payout savings for the CCA. Reducing the demand peaks may have an impact on a CCAs resource adequacy (RA) costs, which will be discussed in our next post, part two of the RA cost study.

Additionally, successfully shifting load has economic and GHG emission impacts for the CCA. The following tables display how GHG emission has been impacted by the addition of energy storage systems under two GHG emission evaluation methods:

  • the CARB unspecified power content method
  • the CNS tables

Method 1: CARB Unspecified Power Content

Month

Before Storage (tCO2)

After Storage (tCO2)

1

14.53

14.72

2

5.72

6.19

3

4.95

5.46

4

-0.48

-0.01

5

-3.26

-2.68

6

0.6

1.15

7

8.35

8.85

8

5.38

5.99

9

3.64

3.98

10

3.38

3.78

11

10.07

10.21

12

12.89

13.1

Total

65.77

70.74

Method 2: CNS Tables

Year

Before Storage (tCO2)

After Storage (tCO2)

2022

89.38

70.00

2026

82.09

67.27

2030

99.06

65.81

Under the CARB unspecified power method, the calculated GHG emissions increased with the addition of the energy storage systems. This is due to the fact that this methodology is a non-time-based calculation. This issue is explained in detail in our previous TerraBlog post, How battery energy storage systems can reduce CCAs GHG emissions.

There is, however, a reduction in projected GHG emissions when optimizing the battery charge and discharge cycles times using the CNS table values. For best results, the battery cycling is optimized around the 2030 CNS table.

The annual NEM payout reduction for this subset of 32 customers was found to be $3,984.85. This reduction represents on average an increase of $0.0219/kWh in customer bills, while providing resiliency and a reduced carbon footprint to these customers. Overall, a 34% reduction in GHG emissions, utilizing the 2030 CNS table, is accomplished.

The following table summarizes the results from the BEO software analysis between the two CCA customer populations:

 

CCA Customer Segment 1

CCA Customer Segment 2

Customer Type

Commercial

Residential

Starting Number of Customers

176

1347

Number of Customers in the Analysis

43

32

Total Annual Customer Usage Pre-DER Addition

-2,655,750 kWh

166,466 kWh

Total Annual Customer Usage Post-DER Addition

-2,461,187 kWh

178,874 kWh

Average Customer Bill Increase

$0.0181/kWh

$0.0219/kWh

Percentage of NEM Payout Reduction

5.5%

205%

Percentage of 2030 GHG Emissions Reduction

47%

34%

In conclusion, when assessed and deployed properly, DER programs have the potential of reducing CCA NEM payouts and GHG emissions, while simultaneously providing resiliency support for their customers. And the BEO software tool provides a crucial level of intelligence in developing that right strategy.

Our next post will assess how deploying an energy storage program can be utilized to reduce the resource adequacy costs for a CCA.

Stay in the know with our latest blog posts!


By submitting this form, you are consenting to receive marketing emails from: TerraVerde, 1100 Larkspur Landing Circle, Larkspur, CA, 94939, http://terraverde.energy. You can revoke your consent to receive emails at any time by using the SafeUnsubscribe® link, found at the bottom of every email. Emails are serviced by Constant Contact