Energy Efficiency State Identification for Cogeneration Units Based on Benchmark Value

Abstract

In China, cogeneration units predominantly employ a flexible operation mechanism. However, it is possible that this could lead to a decline in performance and an increase in energy consumption. This paper introduces a methodology that utilizes the data mining technique to ascertain the benchmark value section of the energy efficiency status index for cogeneration units. The equal interval division method is utilized for the purpose of categorizing the operating conditions. The Gaussian mixture model is utilized to ascertain the benchmark value section in relation to the fluctuating operating conditions by estimating the probability of historical data. The methodology is verified by utilizing historical data from a functioning cogeneration unit. The findings suggest that the unit’s total heat consumption can be decreased by 32.5–50 kJ·(kW·h)−1 when compared to the design-based approach.