In this reported work, firstly, the artificial neural network (ANN) is taken as a target recognition algorithm and then jointly, the computational accuracy and an algorithm parameter (i.e. the number of hidden nodes) are optimised to minimise the overall energy consumption of ANN evaluations. This joint optimisation is motivated by the observation that both the computational accuracy and the algorithm parameter affect recognition accuracy and energy consumption. The evaluation shows that the jointly optimised computational accuracy and the algorithm parameter reduces the energy consumption of ANN evaluations by 79% at the same recognition target, compared with optimising only the algorithm parameter with precise computations. Furthermore, it is demonstrated that to evaluating ANNs with reduced computational accuracy, recognition accuracy is further improved by training the ANNs with reduced computational accuracy. This allows reduction of energy consumption by 86%.

No abstract

<p>In order to perform adequate water quality management, it is important to predict the water quality through measurement and data accumulation of the concentration of contaminants. However, daily measurement of water quality pollutant is unrealistic in practical aspect. In this study, the possibility of daily- or hourly-based water quality prediction through dissolved oxygen (DO) using RNN-LSTM (Recurrent Neural Network-Long Short-term Memory) algorithm, which is well-known for time-series learning, was performed. The research selected Bugok Bridge in Oncheon-stream, Busan, South Korea. Hourly-based DO, temperature, wind speed, relative humidity, rainfall data was collected at the target location and was converted to daily data. To forecast the DO concentration, TensorFlow, a deep learning open source library developed by Google, was utilized. Data of four years (2014-2017) was used for daily learning data and 2018 data was used for verification of the trained model. The performance with the adjusted number of hidden layers, number of repetitions, and the sequence length, as well as the accuracy of the model was analyzed. As a result of this research, it is proven that the performance of the prediction can be improved when weather data and large amount of data is available.</p>

No abstract