An Energy Optimization Model (EOM) to reduce mobile service providers network costs: A multi-objective optimization approach

Abstract

This paper presents an Energy Optimization Model (EOM) for Mobile Service Providers (MSP) that enables the optimization of power efficiency and the integration of optimum renewable and clean energy sources into the mobile network. The model approach drives both Operational Expenditure (OPEX) and Capital Expenditure (CAPEX) reduction via re-engineering power provisioning and site solutions. The features for the selection of Base Transceiver Station (BTS) type based on low power consumption are discussed within the context of establishing a direct impact on the site construction, power source type and dimensioning, and hence, the network cost structure. The K-mean clustering algorithm is used to cluster sites based on those features. HOMER® software was used to optimize the solution within each cluster of sites. This follows a multi-objective optimization function with power saving and CO2 emission as the dominant target factors and the cost, OPEX, Operation & Maintenance (O&M) as constraints. We focus on the lowest power optimization in this paper. Network energy optimization for between clusters (intra-cluster) is performed. Both Traffic and power profile data from Zain-Sudan MSP during the two month period from May-to-July 2013, is obtained and used as input to the EOM model. The optimum parameters for the set of solutions are then determined for deployment under budget & cost constraints. Renewable energy power generation profile for solar and wind from Laqawa site in the South-West of Sudan is used. Results indicate the effectiveness of the EOM model in finding optimum solutions per cluster of sites while facilitating for multi-objective optimization formulation across geographical regions and site types.