Water Resources Allocation in Colorado River Basin Based on Game Model and Genetic Algorithm

This article gives a water allocation method in complex scenarios to determine the best way to manage water use and electricity production, especially with continued drought in the Colorado River Basin. With water and electricity consumption data in recent years, we predicted future water consumption based on deep learning algorithms. This makes the input of the model meet future needs to a certain extent. The model is divided into the following two parts based on reasonable idealized assumptions: Initial allocation and internal allocation. Depending on the Game Model and Genetic Algorithm used, we name these steps the GMG model. This model takes the complete selfishness of each state as the starting point, the distribution situation and drawbacks at this time are given, and it is naturally extended to the situation where states cooperate and jointly pursue the greatest interests. Then, we analyzed the distribution steps, taking into account the principle of fairness, and finally determined the method. Finally, we made a quantitative comparison of the effects of the selfish model, the cooperative model, and the current situation, verifying the value of the model. Substituting a large amount of data into our model predicts at least a 20% improvement in overall benefits compared to the actual situation at the corresponding stage.