consider the automatic construction technology of
large-scale power plants and their complex
constraints and the model adaptive technology to
adapt to different trading rules and needs, especially
for the high proportion of hydropower spot market,
the differences in the geographical location,
regulation performance, installed capacity, incoming
water size, water level status, unit characteristics of
the cascade hydropower stations, it will have a great
impact on model construction and adaptability. To
this end, the integration technology of high
proportion hydropower spot clearing model is
proposed to solve the problem of hydropower spot
market clearing model integration from two aspects:
the automatic description and integration of model
boundary conditions, especially the characteristics of
hydropower system, and the target and constraint
configurable component integration library. See the
following for the detailed integration technology。
1)Automatic description and fusion of hydropower
system characteristics
The main difference between hydropower system
and thermal power system is that they involve many
complex nonlinear hydraulic characteristics and are
uncertain. There are also large differences between
river basins and upstream and downstream power
stations. Especially for the high proportion of
hydropower market, spot clearing modelling is
required to realize the automatic description and
integration of these characteristics. Therefore, a
directed graph description technology for the
relative spatial position of cascade hydropower
stations is proposed. By constructing the directed
graph cascade topology structure, the topology and
time sequence relationship between hydropower
stations can be automatically analysed to adapt to
the differences and changes in the composition
structure of hydropower stations in different basins;
The hydraulic condition analysis and fusion
technology based on the automatic transaction
mechanism is proposed, which defines the
characteristics, frequency, mode and other
transaction rules of incoming water, output, water
level, flow and other data, and combined with
JAVA multithreading technology, the automatic
timing processing of different scales and different
data businesses is realized, and the data encoding
and matching technology is used to automatically
integrate contract type data, which significantly
improves the analysis accuracy of complex
hydraulic data; The automatic modelling technology
for the restricted area of complex units of SCUC and
SCED high head large capacity units is proposed.
The irregular multi restricted operation area
constraints of southwest giant hydropower units can
be automatically incorporated into the spot clearing
model by using polygonal subdivision linearization
approximation modelling, so as to avoid
unreasonable transaction results
.
2)Hydroelectric target and constraint configurable
component integration library
The dynamic changes of spot market clearing rules,
dispatching conditions of hydropower system and
operation demand put forward new requirements for
the construction of clearing model, and it is
necessary to build an extensible component
integration library to cope with the flexible changes
of clearing rules. The clearing model is divided into
two parts: objective function and constraint
condition, and a flexible configuration component
library are constructed respectively. Each
component library can be freely combined to form
an extensible integration library. The objective
function library is composed of various objective
functions facing different business needs, including
the optimal cost category, the largest clean energy
consumption category, such as the largest
hydropower consumption, the smallest risk category,
such as the minimum risk of abandoned water, the
minimum risk of power shortage, the constraint
library involves different classification standards,
including classification by power type, classification
by constraint characteristics, etc.
4.2 Integrated Technology of Large-
Scale Complex Optimization
Algorithm
The solution of the clearing model in the spot
electricity market itself is a complex optimization
problem, especially in the high proportion
hydropower system, which is recognized as one of
the most complex systems in the world. Due to the
nonlinearity of hydropower itself and the coupling
constraints between the cascades, the clearing
problem has the characteristics of non-convex,
nonlinear, discrete, high-dimensional, and is facing
outstanding difficulties in solving. The traditional
solution methods include dynamic programming
methods, intelligent optimization algorithms, etc.
However, as the scale of the problem increases, the
control conditions and constraints of the model
increase significantly, and the traditional dynamic
programming methods have serious "dimension
disaster" problem; Although intelligent optimization
algorithms such as genetic algorithm and simulated