South Sup: Demand Response Community Meal Mini-Pilot at Southside Community Center

 

EXECUTIVE SUMMARY

The goal of South Sup was to explore how a community center could act as an aggregator of residential demand response, by providing a community meal and shifting cooking and occupancy based energy load. This was motivated by a study conducted at EcoVillage Ithaca where it was found that their community meals saved an average of 20% (0.686 kW) electric demand per household participating, during a 4-10 PM time window. The key to achieving aggregate savings is having many households attend, the more that attend, the higher the energy savings. Currently, NYSEG demand response programs require at least 50 kW of demand reduction for a customer to participate. It is estimated that it would take around 75 households participating in a community meal to achieve 50 kW of demand reduction. The target for South Sup was 25 households participating per meal as a beginning which could be built up to higher participation over time with additional funding. A question was whether households would attend meals held on short notice (under 48 hours). The short notice mimics the short time that demand response programs communicate the need for demand reduction to program participants.

Three meals were held at Southside Community Center, one a kick-off and then two short notice meals. The short notice meals had 15 and 21 households in attendance respectively. The attendance was a good initial showing and close to our target. The feedback from participants was generally positive, with most respondents providing positive feedback about the short notice. For these meals to achieve closer to 72 households in attendance more resources and time are needed to promote social media, deliver fliers, and conduct inter-personal communication and outreach. If South Sup were expanded to hold more meals over a longer period of time, it could continue to build on the brand that has been established and it could reach a higher number of households. This project was a collaboration between Quinn Energy, Learn@EcoVillage, and the Southside Community Center.

Keywords: Community Meal, Demand Response, Residential Energy

Full Document: Download

 

A Case Study of the Measured Demand Savings of Community Meals

 

ABSTRACT

Energy savings can be achieved without energy being the primary motivation. This case study explores how residents in EcoVillage at Ithaca (EVI) share in regular community dinners that achieve energy and power demand savings. The primary motivators for the community dinners are based on the desire for social interaction and access to prepared, local, affordable, and healthy meals – but the energy and demand savings are real. Cooking is one of the largest power demand loads in EVI homes. Household electric energy data at thirty-second intervals, for thirty households, were analyzed. Cooking behavior trends were identified and the impact of community meals was measured. This case study highlights the potential role of community meals in achieving demand reduction, deep energy savings and zero net energy (ZNE) for residential neighborhoods.

This poster was presented at the ACEEE Behavior, Energy, and Climate Change Conference 2016.

Keywords: Community Meal, Demand Management, Energy Savings, M&V

Full Document: Download

 

Discovering the Energy Story

Part 2:
Multi-Load Characterization

EXECUTIVE SUMMARY

Power distribution systems provide power to multiple loads at the same time. Because of this, characterizing the relationships of multiple loads can be of more interest to system operators and designers than characterizing an individual load. Of particular interest is identifying and estimating the system peak, which drives capacity requirements. 

This white paper is the second part of a series called Discovering the Energy Story. In Part 1: Load Characterization, methods for single-load power analysis were described and demonstrated using an example residential load. This second paper looks at the multi-load power analysis of a residential neighborhood. It applies approaches defined in Part 1 and expands to multi-load specific characterization.

This paper includes load statistics and Coincident Load analysis for the thirty (30) households in the neighborhood. Coincident Load represents the summation of individual loads during the same instance in time. The average individual load levels are compared to Coincident Load levels. Key metrics are assessed with an emphasis on the comparison of Peak to Average Ratios.

The Peak to Average Ratio significantly decreased as the number of residences that were summed in the Coincident Load increased. Peak to Average Ratio values were: 24 for the individual residence average, 6 for the energy center average, and 3 for the entire neighborhood. In Part 1: Load Characterization, the Peak to Average Ratio for the example residence was 41. The high variance in Peak to Average Ratio reflects a low Coincidence Factor for the neighborhood residences of 0.2. The low Coincidence Factor and high individual residence Peak to Average Ratios highlight how critical the selection and application of load data is in characterizing a system node.

Keywords: multi-load, characterization, coincident load, power, residential, analytics

Full Document: Download


Discovering the Energy Story

Part 1:
Load Characterization

EXECUTIVE SUMMARY

Smart meter and related technologies are generating a lot of energy consumption data. Load characterization techniques can harvest valuable information from the data. This white paper provides a description of load characterization approaches and then applies them to a residential household data set. The example data is a thirty-second interval electrical power data set. The data was collected over a four-week period of time in February and March 2016, in Ithaca, NY.

The load data was summarized using common load level and load shape metrics. Graphical representations of daily profiles and load duration profiles were demonstrated. The impact of appropriate sample rates was explored. The load was modeled using curve fitting and probability analysis. The tradeoff and challenge of retaining data fidelity while reducing historic data storage requirements was explored. The probability analysis was found to best represent the data while reducing data storage requirements by a factor of 10-20x.

Keywords: load characterization, power data, energy analytics

Full Document: Download