Forecasting of Reservoir Inflow Using Machine Learning�Case Study: Klang Gate Dam Reservoir
Climate change is a long-term change in the ordinary weather conditions that affects the local, regional and global climates. One of the response solutions to overcome this phenomenon is by managing water resources more efficiently. The reservoir is a major infrastructure to achieve water resource m...
محفوظ في:
| المؤلفون الرئيسيون: | , , , , , |
|---|---|
| مؤلفون آخرون: | |
| التنسيق: | Conference Paper |
| منشور في: |
Springer Science and Business Media Deutschland GmbH
2023
|
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| الملخص: | Climate change is a long-term change in the ordinary weather conditions that affects the local, regional and global climates. One of the response solutions to overcome this phenomenon is by managing water resources more efficiently. The reservoir is a major infrastructure to achieve water resource management and therefore it requires accurate water resources forecasting. The SVR and MLPNN models are introduced as a solution to achieve an efficient reservoir inflow forecasting. There are many input parameters that influence reservoir water flow but the 3 most important parameters are storage level, rainfall, and evaporation that need to be fed into the two models. There have been various model parameters tested such as kernel types in the SVR and the number of hidden layers and neurons in the MLPNN. Both models have proven their ability but however, the MLPNN with two hidden layers and 4 neurons in each layer had outperformed the SVR after being tested using four different performance tests. � 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG. |
|---|