Reservoir Design

Question 1

design based on …

Answer 1

• historical stream flows
• demand estimations
• probability of failure

Question 2

key design assumptions

Answer 2

• no drought more severe than historical data
• water demand is constant

Question 3

mass curve model design

Answer 3

gives a capacity that would start full and end empty over a critical period

considers the annual scale

Question 4

critical period

Answer 4

inflows are smaller than the demand

Question 5

rippl diagram
- details
- pros
- cons

Answer 5

• mass curve method
• empty at max deficit
• full again at second intersection
• CP from fist full to empty

Pros
- simple
- includes seasonality of inflow

Cons
- no evap
- no repeated failures
- no prob of failure

Question 6

sequent-peak algorithm
- details
- pros
- cons

Answer 6

• sum of deficit between supply and demand
• CP starts at -ve gradient

Pros
- variable demand

(same as rippl)
Cons
- no evap
- no repeated failures
- no prob of failure

Question 7

low flow period design

Answer 7

based on the Waitt curve, built on by the alexander method to give prob of failure

Question 8

waitt curve

Answer 8

• plots the minimum total flow over a duration (t)
• month scale
• CP measured from t=0

(same as rippl)
Pros
- simple
- includes seasonality of inflow

Cons
- no evap
- no repeated failures
- no prob of failure

Question 9

alexander method
- description
- application
- pros
- cons

Answer 9

Description:
plots qp(t) = p percentile flow for duration t
- assumes Xt independence therefore distribution of sum of Xt can be taken from that of Xt
—- gamma such that dist of X1 + X2 ~ T(2*aplpha, beta)
—- alpha = shape parameter
—- beta = scale parameter

Application:
- parameters approximated using max-likelihood method
- graph
—– D = demand/X_bar
—– a_hat = 1, thus C and CP can be augmented
—– p = 1/Tr

Pros:
- easy to use
- gives prob of failure

Cons:
- assumes gamma dist
- assumes independence, can calc correlation