This work forecasts 〖CO〗_2 emissions from hybrid energy consumption in Michael Okpara University of Agriculture Umudike. By using primary consumption information from the university to establish carbon emission based data from grid electricity and fossil energy (Premium Motor Spirit and Automotive Gas Oil) use, the total amount of carbon emission from energy consumption of MOUAU in 7 months from January to July 2017 was found to be 705,505.01 kg, of which the carbon emission from fossil energy and grid electricity consumptions were 609,831.00 kg and 95,674.01 kg, with corresponding proportions of 86.43 % and 13.56 % respectively. The largest amount of carbon emission of 64,079.44 kg was from the Administrative Building with 56,808 kg generated and 7,217.44 kg from the grid. A backward propagation using Artificial Neural Network (ANN) was developed, trained for validation using the 〖CO〗_2 emission data of MOUAU in predicting the local contribution to global emission data. The results obtained showed that an ANN application is efficient and useful in solving climate pollution problems.
Key words: Carbon dioxide emissions, greenhouse gases, fossil energy, electricity generation
|
