Optimal sampling plan for clean development mechanism energy efficiency lighting projects

Abstract

Clean development mechanism (CDM) project developers are always interested in achieving required measurement accuracies with the least metering cost. In this paper, a metering cost minimisation model is proposed for the sampling plan of a specific CDM energy efficiency lighting project. The problem arises from the particular CDM sampling requirement of 90% confidence and 10% precision for the small-scale CDM energy efficiency projects, which is known as the 90/10 criterion. The 90/10 criterion can be met through solving the metering cost minimisation problem. All the lights in the project are classified into different groups according to uncertainties of the lighting energy consumption, which are characterised by their statistical coefficient of variance (CV). Samples from each group are randomly selected to install power meters. These meters include less expensive ones with less functionality and more expensive ones with greater functionality. The metering cost minimisation model will minimise the total metering cost through the determination of the optimal sample size at each group. The 90/10 criterion is formulated as constraints to the metering cost objective. The optimal solution to the minimisation problem will therefore minimise the metering cost whilst meeting the 90/10 criterion, and this is verified by a case study. Relationships between the optimal metering cost and the population sizes of the groups, CV values and the meter equipment cost are further explored in three simulations. The metering cost minimisation model proposed for lighting systems is applicable to other CDM projects as well.

Optimal metering plan for measurement and verification on a lighting case study

“…The idea is to reduce the M&V metering cost for lighting EE projects by optimally deciding the required confidence and precision levels in different reporting years over the projects' crediting period. For instance, studies [34] and [36] present a longitudinal MCM model by incorporating a liner lamp population decay model that is widely used in CDM lighting projects. The longitudinal MCM model provided in [34] and [36] is further improved in [5], which provides more detailed discussions on the lamp population decay models, weighted measurement impacts, and price inflations of the metering devices.…”

“…For instance, studies [34] and [36] present a longitudinal MCM model by incorporating a liner lamp population decay model that is widely used in CDM lighting projects. The longitudinal MCM model provided in [34] and [36] is further improved in [5], which provides more detailed discussions on the lamp population decay models, weighted measurement impacts, and price inflations of the metering devices. The longitudinal MCM models in [34], [36], and [5] are applicable to lighting projects with homogeneous lighting population that shares the same energy usage patterns and population decay dynamics.…”

“…[35] has proposed a spatial metering cost minimisation (MCM) model to balance the sampling uncertainties across lighting groups. The idea in [35] is to minimise the sample sizes and metering cost for CDM lighting EE projects by assigning optimal confidence and precision levels to the lighting groups with different energy consumption uncertainties. The model in [35] is applicable and useful in optimising the M&V metering plan, but lacks of considerations on lighting population decay dynamics over the projects' life cycle.…”

“…The idea in [35] is to minimise the sample sizes and metering cost for CDM lighting EE projects by assigning optimal confidence and precision levels to the lighting groups with different energy consumption uncertainties. The model in [35] is applicable and useful in optimising the M&V metering plan, but lacks of considerations on lighting population decay dynamics over the projects' life cycle. In practice, the lamp population will decay due to the lamp breakage, theft or other unpredicted damages.…”

“…Ref. [35] has proposed a spatial metering cost minimisation (MCM) model to balance the sampling uncertainties across lighting groups. The idea in [35] is to minimise the sample sizes and metering cost for CDM lighting EE projects by assigning optimal confidence and precision levels to the lighting groups with different energy consumption uncertainties.…”