SDDP
Energy Systems Operation Planning
Energy Systems Operation Planning
View the release notes for SDDP 17.3
SDDP is a stochastic dispatch model for electrical systems with detailed representation of transmission and fuel networks. It is used for long, medium and short-term operational studies and is highly flexible in terms of its temporal and spatial levels of detail. SDDP finds the optimal operation policy of systems containing several types of technologies.
Hydro plants
Variable production coefficient, evaporation, filtration, non-controllable spillage, etc.
Thermal plants
Unit commitment, combined-cycle power plants, nonlinear heat rate, fuel contracts.
Renewable modeling
Detailed modeling and production of synthetic scenarios for variable renewable energy sources with Time Series Lab.
Energy efficiency
Regulations to reduce energy waste and environmental impact by optimizing energy generation.
Demand side management
Response to price, signals by segment by system, area, or bus-level.
Reliability analysis
Stochastic reliability study focusing on renewable energy and storage using Coral.
Storage devices
Storage capacity, charge and discharge capacities, efficiencies, ramping constraints.
Operating reserves
Dynamic Probabilistic Reserve (DPR) based on renewable scenarios.
Maintenance
User-defined or optimized plant maintenance.
And much more...
The objective of SDDP is to minimize the sum of costs for purchase and transportation of fuels for thermal plants, pollutant emission costs, hydro and thermal plants' O&M costs, transmission wheeling rates, cost of energy not supplied and other penalties. In other words, the model calculates the least-cost operation policy of the system, taking all the aspects above into consideration.
SDDP is recognized as a best-in-class energy market simulation engine, providing leading analytical tools and support for professionals such as modelers, power producers, and analysts.
SDDP stands for Stochastic Dual Dynamic Programming, an algorithm developed by PSR in the 1980s for solving large-scale multi-stage optimization problems under uncertainty. It's not necessary to enumerate the combinations of reservoir levels and the future cost function approximation is made through a Benders decomposition scheme.
The SDDP algorithm has been extended to several areas and became the global industry standard, with more than 1.700 citations in the scientific literature. It has been successfully applied for more than thirty years for mid- and long-term optimal stochastic scheduling of very complex real systems with multi-scale storage and probabilistic modelings such as hydro inflows, renewable intermittency, demand, and fuel prices.
Download SDDP folder
