Page 43 - Mathematical Models and Algorithms for Power System Optimization

Daily Economic Dispatch Optimization With Pumped Storage Plant 33

approximately half the number of integer variables can be reduced, which can also greatly
reduce the calculation burdens.
(5) If the pumped storage power station must run at the peak load time when the peak-valley
difference is large, it is possible to reasonably arrange the pumping and generating
conditions of the pumped storage power station by setting the virtual cost coefficients in
advance.
(6) If the pumped storage plant must run to satisfy water storage balancing constraints when
the peak-valley difference is small (despite the fact that only the thermal power unit can
achieve peak regulation), a virtual cost coefficient is set in advance to realize the
optimized operation of pumped storage plant at all times, taking into account the capacity
of the pumped storage power station shared with Hong Kong.
(7) If the regional tie line is needed to satisfy a certain amount of transmission power, then the
generated output constraint for each area can be set so that the transmission power of the
regional tie line can be automatically controlled to satisfy the requirement of stable
operation of the power system.

2.5.4 Description of the Output Results

Output results include the following:
(1) Optimization calculation results concerning pumping and generating for the pumped
storage plant in each time period.
(2) Optimization calculation results concerning water balancing and reservoir capacity
constraint for the pumped storage plant.
(3) Optimization results of generated output of each unit in each time period.
(4) Optimization results of generated output of each power plant in each time period.
(5) Optimization calculation results of daily generated energy of each power plant.
(6) Optimization results of generated output of each area in various time periods.
(7) Load curves upon optimized operation of the pumped storage plant.

2.6 Implementation

2.6.1 Concrete Expression of Objective and Constraint Functions for a
Small Scale System

Before optimization calculation on full-scale systems, it is better to verify the proposed model
and calculation method using a small-scale one, to confirm whether the cost coefficient setting,
variables, and constraint conditions are right or not.

First, a simplified mathematical testing model is derived for the small-scale system to give a
clear view about the proposed model. The testing system consists of one area, one thermal