Anderson Rodrigo de Queiroz

Sharing cuts under aggregated forecasts when decomposing multi-stage stochastic programs

Abstract Sampling-based decomposition algorithms (SBDAs) solve multi-stage stochastic programs. SBDAs can approximately solve problem instances with many time stages when the stochastic program exhibits interstage dependence in its right-hand side parameters by appropriately sharing cuts. We extend previous methods for sharing cuts in SBDAs, establishing new results under a novel interaction between a class of interstage dependency models, and how they appear in the stochastic program.

Simulating Electricity Spot Prices in Brazil Using Neural Network and Design of Experiments

The electricity price has been one of the most important variables since the introduction of deregulation on the electricity sector. On this way, efficient forecasting methods of spot prices have become crucial to maximize the agent benefits. In Brazil the electricity price is based on the marginal cost provided by an optimization software (NEWAVE). Forecasting the operational marginal cost (OMC) and its volatility has been one big problem in the Brazilian market because of the computational time taken by this software. This work presents a fast and efficient model to simulate the OMC using DOE (design of experiments) and ANN (artificial neural networks) techniques. The paper proved that the combined techniques provided a promising result and may be applied to risk management and investment analysis.

Effects of wind penetration in the scheduling of a hydro-dominant power system

A computational model that is able to determine the optimal economic generation scheduling considering decisions in a system with hydro, thermal and wind power plants is presented. The algorithm is based on the class of sampling-based decomposition algorithms used to solve large-scale multi-stage stochastic optimization problems. A case study composed by several simulation runs of the model is presented and the results about wind power effects in the scheduling of power generators are discussed.

Estimation of electricity elasticity for demand rates and load curve in Brazil

In this paper a methodology is proposed to calculate the elasticity between demand profile and electricity price. This elasticity is essential in the tariff design process because it is the first way to represent the customer response about the price signal. In this calculation it is necessary to identify, by using a clustering analysis algorithm, typical load profiles among the consumers. The elasticity is then obtained by comparing the load profiles with tariff variation between two consecutive years. The focus is on the relative variation (relation between peak and off peak periods) of tariff and load among the daily 24 hours. This approach captures the customer willingness to change the load profile due to tariff signals. Some real examples are shown using the network of a distribution company in Brazil. The results obtained show that the elasticity does exist and it is different for each customer in many activity sectors.

The influence of climate change on hydro generation in Brazil

More than 85% of the total electricity generation in Brazil comes from Hydro Plants. The hydroelectricity is a good example of renewable resources which, in this case, depends greatly on the precipitation. An optimization program is used in order to determine the dispatches of the plants and to minimize the operational costs. Given a set of hydro plants it is possible to calculate the power system assured energy or the amount of energy available to supply the load at a deficit risk of 5%. However, climate change can cause great variation on the assured energy. The problem drastically aggravates for the new available hydro potentials located in the Amazon region which probably will be affected by the global warming. This paper provides the initial results of a research project sponsored by generation companies and the regulatory agency in Brazil.

From voltage level to locational pricing of distribution network: The Brazilian experience

The distribution systems in Brazil are currently priced based on voltage level and time usage. The tariff structure does not take into account the bus or region where the consumption takes place. Only transmission which is represented by lines and substations equal and above 230 kV has been using locational pricing since the end of 90s. The wheeling method is derived from the investment cost related price (ICRP) proposed in UK. So each bus at transmission grid has a nodal price to encompass transmission fixed costs. This locational approach does not continue through the distribution system which has a method that came from the 70s decade that still follows the idea of radial system, i.e., the costs increase as the voltage decreases. Although the peak and off-peak differentiation is considered, the effectiveness of it in terms of distribution operation and planning optimization is questioned. Today, the Brazilian regulators are working in improving the tariff structure including the locational signal and a better time-of-use approach taking into account the characteristics of each Brazilian region. Along with this improvement, the regulators are also studying how to add the new small generators that are spreading around the distribution grid. This paper tries to show how the current price is developed and the future aspects and solutions that the Brazilian regulators are facing in order to promote efficiency and sustainability of distribution energy systems.

US Energy-Related Greenhouse Gas Emissions in the Absence of Federal Climate Policy

The planned US withdrawal from the Paris Agreement as well as uncertainty about federal climate policy has raised questions about the countrys future emissions trajectory. Our model-based analysis accounts for uncertainty in fuel prices and energy technology capital costs and suggests that market forces are likely to keep US energy-related greenhouse gas emissions relatively flat or produce modest reductions: in the absence of new federal policy, 2040 greenhouse gas emissions range from +10% to -23% of the baseline estimate. Natural gas versus coal utilization in the electric sector represents a key trade-off, particularly under conservative assumptions about future technology innovation. The lowest emissions scenarios are produced when the cost of natural gas and electric vehicles declines, while coal and oil prices increase relative to the baseline.

Multi-stage stochastic optimization for a PV-storage hybrid unit in a household

In the face of increasing global energy supply challenges, renewable energy sources provide a cleaner and environmentally friendly energy alternative. To address the intermittency in PV power generation, battery storage can be used to store energy during lower demand periods. This requires the charging and discharging routine of the storage system to be controlled to achieve optimal economic benefits. In this paper, coordinated control between PV component and an accompanying storage unit is presented considering the stochastic nature of PV generation. The stochastic dual dynamic programming (SDDP) algorithm is employed to optimize the charge/discharge profiles with the goal to minimize the overall cost of satisfying the daily household load demand. The PV-storage hybrid unit can jointly contribute in reducing the consumer costs as shown through simulation analysis.

Thermal generation investment analysis using decision tools

In this paper we present an investment problem where a decision maker from a company has to decide on the best among four possible alternatives of power supply. Two of these alternatives are related to investment in a thermal generator to produce electricity for the company own use. In this framework there are many uncertainties that have strong influence on the net present values of each alternative. Some of these uncertainties are identified and modeled for the purposes of this work. It is important to mention that electricity is an important input to the production process and represents considerable costs for the company. In order to avoid unnecessary expenses a solid analysis is necessary. This work combines decision analysis tools, as the influence diagram and decision tree, with investment analysis to help the decision maker to select the best supply alternative for the company.

Determining the optimal transmission system usage contracts for a distribution company

Improvements in transmission and distribution networks can be noticed in most countries that had their system architecture changed by the deregulation process. In this new environment one of the biggest challenges is the transmission and distribution open access. In Brazil, the National Electricity Regulatory Agency has established that the monthly amount of transmission system usage contracted by a distribution company (DISCO) should be informed per connection point (one value for each point) between transmission and distribution network. The usage of the transmission assets are represented by the power flowing from the transmission system to the DISCO network. This implies in monthly charges at each border transformer (which represents a connection point) that the DISCO must pay to honor the contracts. If the DISCO exceeds the contract values by certain percentage, monetary penalties are incurred. The penalty costs can lead the DISCO to a more conservatory behavior at the time that the usage contracts are settled. On the other hand if the DISCO expects a low trend of its demand it has the possibility to contract less and save money. Determining the optimum amount to contract is a stochastic optimization problem because of future load uncertainties. This paper provides model formulations for the Transmission System Usage Problem with a real case study.