The DoD desires prognostics and diagnostics tools to proactively address the degradation of microgrid equipment legacy infrastructure for installation energy. Specifically in this SBIR, the tools should be able to identify long-term generator and power distribution performance degradation. It is not cost-effective nor is the data available to develop physics-based performance models on the myriad of generator sizes, applications, distribution, and age. Hence, we focus on evidence-based quantitative methodologies. Our technical objectives for this feasibility study are to: 1. Collect operating data of a common diesel generator that is deployed across multiple configurations for both installation and operation energy needs. 2. Apply methods such maximum overlap discrete wavelet transforms for feature extraction and filtering to determine the influence of environmental factors, grid interconnections, maintenance protocols, and similar external factors to create normalized time series for modeling. 3. Develop non-linear diagnostic and prognostic models that identify change points for condition-based maintenance, determine remaining useful life, and uncertainty. 4. Define a dedicated software solution that can reside on a commercially available power control system. In keeping with the stated purpose of the SBIR program, this project will produce fundamental research that can be commercialized for government and private enterprise markets.
