Topic requirements
1. Establish a model to propose a set of fixed water supply and demand conditions, and analyze the operation of two dams (Lake Powell and Lake Mead): when the water level of Lake Mead is M and the water level of Lake Powell is P
1. How much water should be withdrawn from each lake to meet the specified demand?
2. If there is no additional water supply, how long will it take for the demand to be met?
3. Over time, how much additional water must be provided to ensure that fixed demands are met.
2. Use the model of problem 1 to propose the best way to resolve the competing interests between agriculture, industry, residential use, and electricity production. State the criteria for addressing interests.
1. How to distribute the water of the two lakes, resolving the competing interests between agriculture, industry, housing and electricity? and propose the best method.
2. Specify what criteria to use to solve the benefits.
3. Build a model to solve, what should be done if there is not enough water to meet the demand for water and electricity?
4. Explain the problem under the following conditions
1. The demand for water and electricity in agriculture, industry, and housing will change over time. What happens when population, agriculture, and industry increase or decrease?
2. What will happen when the proportion of renewable energy exceeds the expected value?
3. What happens when additional water and energy saving measures are implemented?
5. Write an article of one to two pages, representing the solution and putting forward your views, which is suitable for publication on "Drought and Thirst".
Summary:
In recent years, the western United States has suffered an unprecedentedly severe drought, and Lake Powell and Lake Mead are no longer full of water. How to rationally allocate hydropower to surrounding agriculture, industry and residence has become an urgent problem to be solved, which is the main purpose of this paper. First, we start with what is normal as usual when the water supply is plentiful. Due to the complexity of reality, we abstract the entire water distribution process into a strict and orderly system. Every parameter has been carefully considered and several appropriate constraints have been formulated in order to get as close as possible to the real situation. Secondly, in order to accurately measure the profit and loss caused by water shortage and power generation in the five states, we adopted many effective methods, such as the partition coefficient method, to construct a set of suitable functions. To typically reflect the severe impact of water scarcity on each state, we use an Analysis Hierarchy Process (AHP) to help us work across the board.
Secondly, we carefully evaluated various reliable statistical data such as water source and hydropower ratio, and proposed an optimized power supply strategy under the premise of giving priority to water use. Based on this, we decided to adopt a bi-level programming algorithm, which considers the losses caused by insufficient water supply in the upper layer and the benefits of remaining power generation in the lower layer. It is worth mentioning that we use particle swarm optimization (PSO) algorithm to find the optimal solution for each layer in each iteration. Our model has the advantage of strong applicability. When complex situations change, we just need to adjust some parameters accordingly to draw reliable conclusions. For example, the occurrence of drought means that the initial water level of the reservoir will drop, and the change is within the control range of our model. In dealing with this situation, we used ARIMA based time series analysis to predict future monthly water level changes in both reservoirs, which allowed us to evaluate the situation more
Keywords: double-layer programming, particle swarm optimization, evaluation measures, reservoir operation
See the comment area for the link to the complete paper~