Port Throughput Forecast Model Based on Adam Optimized GRU Neural Network
Xiubin Chen, Lei Huang · 2020
The forecast of port throughput can not only provide a basis for the business plan of the enterprise, but also provide a reference for the construction of transportation facilities in the city where the port is located. Gated Recurrent Unit (GRU), as the latest Long Short-Term Memory (LSTM) variant, solves the complementary redundancy of the original input gate and forget gate. In order to predict data more accurately, this paper proposes a port throughput forecast model based on Adam optimized GRU neural network (Adam-GRU). Combining the effective gradient optimization algorithm Adam with GRU can make the forecasting process more efficient. Taking the throughput data of Guangzhou Port Group G Port Affairs Company as an example, this paper compares the Adam-GRU model with other proven traditional forecasting methods, including Back Propagation (BP) Neural Network, Recurrent Neural Network (RNN) and LSTM. The experimental results show that Adam-GRU performs better under the evaluation of various performance indicators, reflecting clear practicality and innovation.