The paper addresses the issue of apportioning of the cost of transmission losses to generators an... more The paper addresses the issue of apportioning of the cost of transmission losses to generators and demands in a multimarket framework. Line flows are unbundled using equivalent bilateral exchanges on a DC-network model and allocated to generators and demands. Losses are then calculated based on unbundled flows and straightforwardly apportioned to generators and demands. The proposed technique is particularly useful in a multimarket framework, where all markets have a common grid operator with complete knowledge of all network data, as is the case of the Brazilian electric-energy system. The methodology proposed is illustrated using the IEEE Reliability Test System and compared numerically with an alternative technique. Appropriate conclusions are drawn.
This article presents a load modeling application using real data measurements to estimate peak d... more This article presents a load modeling application using real data measurements to estimate peak demand reduction due to voltage control. The load models parameters estimates yield conservation voltage reduction indicators, that quantify the effectiveness of this smart grid approach designed to save energy and reduce power demand at peak times. The data analyzed were recorded from UK primary distribution substations, chosen in the scope of a research & development project. Stratified random sampling and statistical inference were used to extract information that gathered load composition differences due to the time of the day, the day of the week and the season, aiming the assessment of the confidence level of the peak demand reduction estimation.
The volume of Distributed Generation (DG) capacity connected to distribution networks is growing ... more The volume of Distributed Generation (DG) capacity connected to distribution networks is growing quickly worldwide. Due to the inherent variable nature of renewable sources, such as wind or hydro, or low-carbon technologies, such as CHP, contracts are based on energy produced rather than power availability. Consequently, DG plants are typically not dispatched. However, an adequate dispatch strategy of these generators could provide a number of benefits to distribution network operators (DNOs). This paper presents, from the economic perspective of a DNO, a comparison between the integration of DG plants with and without dispatch capabilities. A Genetic Algorithm (GA) is applied to define the dispatch strategy (i.e., generation pattern) in a way that the costs related to energy losses and network investment (related to reinforcements and OPEX) are minimised. Here, small hydro generators (less than 30-MW of capacity, run of river), which are very common in Brazil, are considered given its (relatively) more predictable nature. Multiple small hydro DG plants are investigated considering a horizon of a year represented in twelve demand scenarios. The proposed methodology was applied to the IEEE 34-bus test feeder. Results are presented and discussed.
22nd International Conference and Exhibition on Electricity Distribution (CIRED 2013), 2013
In Brazil, large consumers have different alternatives to contract energy. The type of contract a... more In Brazil, large consumers have different alternatives to contract energy. The type of contract and its parameters are important decision variables for consumers who aim to reduce their electricity bill. Here, it is proposed a model to assist large consumers to take these decisions. This model is based on a statistic auto regressive (AR) model to forecast monthly energy consumption and maximum demand in order to establish the best contract for the consumer. The performance of the forecast-based model is evaluated through statistical analysis and by comparison with a simpler method that considers the average values. Different contracting environments are investigated. Results show the importance of adequate forecasting in determining the most cost-effective contract strategy. I. NOMENCLATURE Contracted demand (kW). Tariff of contracted demand ($/kW). Peak energy consumption (MWh). Tariff of peak energy consumption ($/MWh). Off-peak energy consumption (MWh). Tariff of off-peak energy consumption ($/MWh). Penalty for exceeding contracted demand. Verified maximum monthly demand (kW).
Contributions to Higher Engineering Education, 2018
Experimenting is fundamental to the training process of all scientists and engineers. While exper... more Experimenting is fundamental to the training process of all scientists and engineers. While experiments have been traditionally done inside laboratories, the emergence of Information and Communication Technologies added two alternatives accessible anytime, anywhere. These two alternatives are known as virtual and remote labs, and are sometimes indistinguishably referred as online labs. Similarly to other instructional technologies, virtual and remote labs require some effort from teachers in integrating them into curricula, taking into consideration several factors that affect their adoption (i.e. cost) and their educational effectiveness (i.e. benefit). This chapter analyses these two dimensions and sustains the case where only through international cooperation it is possible to serve the large number of teachers and students involved in engineering education. It presents an example in the area of Electrical and Electronics Engineering, based on a remote lab named Virtual Instruments System in Reality, and it then describes how a number
2016 2nd International Conference of the Portuguese Society for Engineering Education (CISPEE), 2016
Experiments have been at the heart of scientific development and education for centuries. From th... more Experiments have been at the heart of scientific development and education for centuries. From the outburst of Information and Communication Technologies, virtual and remote labs have added to hands-on labs a new conception of practical experience, especially in Science, Technology, Engineering and Mathematics education. This paper aims at describing the features of a remote lab named Virtual Instruments System in Reality, embedded in a community of practice and forming the spearhead of a federation of remote labs. More particularly, it discusses the advantages and disadvantages of remote labs over virtual labs as regards to scalability constraints and development and maintenance costs. Finally, it describes an actual implementation in an international community of practice of engineering schools forming the embryo of a first world wide federation of Virtual Instruments System in Reality nodes, under the framework of a project funded by the Erasmus+ Program.
In this paper, a new load model is proposed for electric power distribution systems under varying... more In this paper, a new load model is proposed for electric power distribution systems under varying voltage conditions in order to estimate the energy losses and, consequently, optimize the energy losses in the system. The proposed methodology is based on the adjustment of the polynomial ZIP load parameters related to active and reactive power as a function of the static voltage variations measured at the substation. The ZIP parameters are determined using the Genetic Algorithm in order to minimize the error between the measured and estimated active and reactive power at the substation. The estimation of the energy losses during a period of analysis is determined by the computation of the power flow using the ZIP model of each load in the system and the substation voltages measured. The procedure requires information of the feeder topology, distribution lines, rated power of the transformers, and a database containing voltage and power measurements at the substation during the period ...
This article presents a load modeling application using real data measurements to estimate peak d... more This article presents a load modeling application using real data measurements to estimate peak demand reduction due to voltage control. The load models parameters estimates yield conservation voltage reduction indicators, that quantify the effectiveness of this smart grid approach designed to save energy and reduce power demand at peak times. The data analyzed were recorded from UK primary distribution substations, chosen in the scope of a research & development project. Stratified random sampling and statistical inference were used to extract information that gathered load composition differences due to the time of the day, the day of the week and the season, aiming the assessment of the confidence level of the peak demand reduction estimation.
Advances in Intelligent Systems and Computing, 2018
Laboratories are a fundamental part of engineering education due to the very nature of the engine... more Laboratories are a fundamental part of engineering education due to the very nature of the engineering profession. This is a characteristic of all engineering courses, though it may vary from one curriculum to the other and even in the same curriculum. This paper is dedicated to the analysis of different applications of a remote lab in an Electrical Engineering curriculum. If the types of experiments it offers are concerned, it is classified as an Electric and Eletronic Circuits lab; one in a set that also has Digital Electronics, Analog Electronics, Electrical Machines, Control Systems, etc. But it is not a traditional labit is a remote lab, with traditional components remotely accessed over the Internet. This work presents the preliminary results of the deployment of VISIRa remote lab for Electric and Electronic Circuits in some courses. It discusses the different course contexts and how the use of VISIR was adapted to each one. Results of students' opinions are presented and discussed.
This paper proposes a novel approach to achieve better results for the Non-intrusive load disaggr... more This paper proposes a novel approach to achieve better results for the Non-intrusive load disaggregation model for residential consumers. The proposed methodology considers two steps to face the problem. Firstly, periodical appliances are modelled by the Fourier series and extracted from the total power measured. Secondly, a Mixed Integer Linear Programming was proposed to disaggregate the remaining appliances. The paper presents results with a real case and indicates that the model can be useful for practical applications, such as the possibilty of the consumers of changing the modality of their tariff contract.
Remote Laboratories have become part of current teaching and learning, particularly in engineerin... more Remote Laboratories have become part of current teaching and learning, particularly in engineering. Their potential to aid students beyond their hands-on lab classes has been a matter of discussion in literature. Teachers and researchers are aware that the thorough analysis of both strengths and shortcomings of remote labs in didactical implementations may not only lead to the improvement of these resources but also of the pedagogical implications in engineering classes. The present study was carried out in a Higher Education Institution in Brazil in two different courses during three consecutive semesters where a remote lab (VISIR) addressing electric and electronic topics was implemented, yielding 471 students' academic results and opinions. These students' results (while using VISIR) cross-analysed with the course characteristics, reveal some factors teachers may tackle to foster student learning and motivation. The conclusions point to the need for VISIR interface modernization and showed it is more useful in basic courses than in more advanced ones, when dealing with classic lab experiments. Results also show that teachers' involvement plus their ability to brief students on VISIR's usefulness have a significant influence not only on students' performance but also on their perception of learning and satisfaction with the tool. In the analysed cases, the students with more learning needs seemed to be the ones who could benefit more from VISIR.
Because electricity generation from renewable energy is seasonal and intermittent, the risk assoc... more Because electricity generation from renewable energy is seasonal and intermittent, the risk associated with its cash flow volatility may discourage investments in this area. In this paper, a method called Energy Reallocation Mechanism is proposed to mitigate this problem. The idea, inspired by the current mechanism used in Brazil for Hydropower plants, consists of the creation of a renewable energy portfolio and the allocation of the quotas of this portfolio to each source. Because the portfolio cash flow is more stable than the individual cash flows, the financial risk of being in the portfolio is reduced. The proposed mechanism is based on a stochastic approach and it combines future scenarios of generation in order to produce future scenarios of cash flow. Beyond the Expected Value applied in the Brazilian mechanism, it also considers the Conditional Value at Risk to account for average and risk, respectively, associated with the generation profile of each source. Finally, based on the aforementioned, the Marginal Benefit method is applied to allocate the portfolio's quotas for the renewable sources guaranteeing that the solution provided is suitable under the cooperative game theory viewpoint. For didactic reasons a small renewable energy portfolio, formed by Wind Power, Small Hydro and Biomass, is used for comparing the proposed mechanism with the Shapley Value method and the current method applied in Brazil. The applicability for larger systems is demonstrated with real data from the Brazilian Power System.
The Brazilian Electricity Regulatory Agency (ANEEL) presented a proposal to revise the tariff str... more The Brazilian Electricity Regulatory Agency (ANEEL) presented a proposal to revise the tariff structure of distribution companies in Brazil. One of the main approved suggestions was to establish a mechanism called Tariff Flags, which aims to foster a Demand Response Program in Brazil via an increase in the energy tariff. In this work, the proposed mechanism is reviewed in detail and the expected results of its application are simulated and analyzed under different perspectives. This paper shows that the system operation directly impacts the Demand Response Program, since the spot prices will define which Tariff Flag should be triggered. In order to encompass and assess the main consequences of its application, this paper presents the expected effects on energy spot prices, system operating costs, probability of triggering each flag, investment recovery for utilities and finally, the impact for the final consumers. The case studies presented in this paper were developed using real information about the Brazilian electrical system for each economic sector and the price-demand elasticity is discussed using the literature for this application. Finally, some conclusions and guidelines are provided to improve the application of the mechanism.
Abstract One of the main objectives of Demand Response Programs is to mitigate the effect of peak... more Abstract One of the main objectives of Demand Response Programs is to mitigate the effect of peak electricity demand by inducing consumers to shift or reduce their electricity consumption in peak times. In Brazil, big energy consumers should contract peak demand for the upcoming months with the utility distribution companies under a set of rules defined by the System Regulator. Based on this, the utility will have more information to reinforce the system accordingly. The challenge for consumers is to compute the best value of the peak demand contracted before the peak demand realization. One way to solve this problem is to simulate future scenarios of peak demand and optimize the value of the peak demand contracted in compliance with the current rules. In this paper, a support decision system is proposed to help consumers in this task. The proposal is divided into two parts, which are a statistical model, for estimating and simulating future scenarios of peak demand, and a stochastic optimization model, to compute the value of the peak demand contracted. In the first part, a Box & Jenkins model is used to estimate the parameters of the statistical model and to simulate it based on the historical electricity bill data. In the second part, a stochastic optimization model is applied using a convex combination of the Expected Value and Conditional Value-at-Risk, as the risk metrics for the cost uncertainty, in order to obtain the monthly values of peak demand contracted. The results achieved corroborate the importance of using an appropriate mathematical model to address the problem. To illustrate the proposed approach, a real case study for a big consumer in Brazil is presented.
A energia eólica vem se firmando de maneira concreta no Brasil. Incentivos foram dados a esta fon... more A energia eólica vem se firmando de maneira concreta no Brasil. Incentivos foram dados a esta fonte nos leilões do ACR. Contudo, o modelo de planejamento e simulação da operação vigente no Brasil desconsidera a incerteza na geração das fontes alternativas renováveis. Em aplicações, como a comercialização de energia proveniente dessas fontes, uma metodologia de simulação que produza cenários casados com grandezas sistêmicas torna-se necessária. Assim, este trabalho apresenta um modelo de geração de cenários de produção de fontes renováveis intermitentes que sejam dependentes dos cenários de variáveis sistêmicas, PLD, geração térmica, intercâmbios, etc., advindos do modelo de planejamento operativo.
2012 IEEE Power and Energy Society General Meeting, 2012
ABSTRACT Wind power generation plays an important role in most power systems worldwide. Despite t... more ABSTRACT Wind power generation plays an important role in most power systems worldwide. Despite that, only in 2009, wind farms appeared as a profitable and competitive investment option in Brazil. The Brazilian power system is mainly hydro based and, due to its many singularities, its coordination is still centralized. A collateral effect of such centralized dispatch coordination is that the system marginal costs, which determine the short-term energy prices, exhibit a highly volatile pattern. In this setting, selling energy through bilateral forward contracts backed up on intermittent generation profiles, such as wind farms and small-run hydros, exposes the generation company (Genco) to the so-called price-quantity risk. In this paper, we explore the well-known complementarity between these two renewable resources (wind and inflows) by proposing a statistical model capable to produce scenarios of renewable resources availability consistently with short-term price scenarios. Hence, we present a novel commercial model for a wind power producer based on a joint-selling strategy with a small-hydro Genco. In this model, the hydro Genco receives a surplus payment in comparison to the amount it would receive in the market and the wind Genco, the rest of the income. We show, by means of realistic data from the Brazilian power system, that such commercial model is able to mitigate the exposure to the short-term price and foster the wind power penetration into the contract market.
2013 IEEE Power & Energy Society General Meeting, 2013
ABSTRACT In this study, a fault location algorithm for electrical power systems is presented. The... more ABSTRACT In this study, a fault location algorithm for electrical power systems is presented. The proposal consists in an accurate impedance-based fault location algorithm by utilizing voltage and current measurements from local terminal and estimated remote source impedance. In order to evaluate the fault location algorithm a numerical analysis is made. Phase-to-ground faults are simulated in a 230 kV Brazilian transmission line using the software ATP/EMTP and MATLAB. The algorithm is tested considering the fault distance, fault resistance, and the variation of short-circuit capacity of remote system effects. The results show that the fault resistance and fault distance affect the estimated fault distance, but with minor errors. More, the proposed algorithm is not sensitive to the remote system short-circuit capacity variation.
The paper addresses the issue of apportioning of the cost of transmission losses to generators an... more The paper addresses the issue of apportioning of the cost of transmission losses to generators and demands in a multimarket framework. Line flows are unbundled using equivalent bilateral exchanges on a DC-network model and allocated to generators and demands. Losses are then calculated based on unbundled flows and straightforwardly apportioned to generators and demands. The proposed technique is particularly useful in a multimarket framework, where all markets have a common grid operator with complete knowledge of all network data, as is the case of the Brazilian electric-energy system. The methodology proposed is illustrated using the IEEE Reliability Test System and compared numerically with an alternative technique. Appropriate conclusions are drawn.
This article presents a load modeling application using real data measurements to estimate peak d... more This article presents a load modeling application using real data measurements to estimate peak demand reduction due to voltage control. The load models parameters estimates yield conservation voltage reduction indicators, that quantify the effectiveness of this smart grid approach designed to save energy and reduce power demand at peak times. The data analyzed were recorded from UK primary distribution substations, chosen in the scope of a research & development project. Stratified random sampling and statistical inference were used to extract information that gathered load composition differences due to the time of the day, the day of the week and the season, aiming the assessment of the confidence level of the peak demand reduction estimation.
The volume of Distributed Generation (DG) capacity connected to distribution networks is growing ... more The volume of Distributed Generation (DG) capacity connected to distribution networks is growing quickly worldwide. Due to the inherent variable nature of renewable sources, such as wind or hydro, or low-carbon technologies, such as CHP, contracts are based on energy produced rather than power availability. Consequently, DG plants are typically not dispatched. However, an adequate dispatch strategy of these generators could provide a number of benefits to distribution network operators (DNOs). This paper presents, from the economic perspective of a DNO, a comparison between the integration of DG plants with and without dispatch capabilities. A Genetic Algorithm (GA) is applied to define the dispatch strategy (i.e., generation pattern) in a way that the costs related to energy losses and network investment (related to reinforcements and OPEX) are minimised. Here, small hydro generators (less than 30-MW of capacity, run of river), which are very common in Brazil, are considered given its (relatively) more predictable nature. Multiple small hydro DG plants are investigated considering a horizon of a year represented in twelve demand scenarios. The proposed methodology was applied to the IEEE 34-bus test feeder. Results are presented and discussed.
22nd International Conference and Exhibition on Electricity Distribution (CIRED 2013), 2013
In Brazil, large consumers have different alternatives to contract energy. The type of contract a... more In Brazil, large consumers have different alternatives to contract energy. The type of contract and its parameters are important decision variables for consumers who aim to reduce their electricity bill. Here, it is proposed a model to assist large consumers to take these decisions. This model is based on a statistic auto regressive (AR) model to forecast monthly energy consumption and maximum demand in order to establish the best contract for the consumer. The performance of the forecast-based model is evaluated through statistical analysis and by comparison with a simpler method that considers the average values. Different contracting environments are investigated. Results show the importance of adequate forecasting in determining the most cost-effective contract strategy. I. NOMENCLATURE Contracted demand (kW). Tariff of contracted demand ($/kW). Peak energy consumption (MWh). Tariff of peak energy consumption ($/MWh). Off-peak energy consumption (MWh). Tariff of off-peak energy consumption ($/MWh). Penalty for exceeding contracted demand. Verified maximum monthly demand (kW).
Contributions to Higher Engineering Education, 2018
Experimenting is fundamental to the training process of all scientists and engineers. While exper... more Experimenting is fundamental to the training process of all scientists and engineers. While experiments have been traditionally done inside laboratories, the emergence of Information and Communication Technologies added two alternatives accessible anytime, anywhere. These two alternatives are known as virtual and remote labs, and are sometimes indistinguishably referred as online labs. Similarly to other instructional technologies, virtual and remote labs require some effort from teachers in integrating them into curricula, taking into consideration several factors that affect their adoption (i.e. cost) and their educational effectiveness (i.e. benefit). This chapter analyses these two dimensions and sustains the case where only through international cooperation it is possible to serve the large number of teachers and students involved in engineering education. It presents an example in the area of Electrical and Electronics Engineering, based on a remote lab named Virtual Instruments System in Reality, and it then describes how a number
2016 2nd International Conference of the Portuguese Society for Engineering Education (CISPEE), 2016
Experiments have been at the heart of scientific development and education for centuries. From th... more Experiments have been at the heart of scientific development and education for centuries. From the outburst of Information and Communication Technologies, virtual and remote labs have added to hands-on labs a new conception of practical experience, especially in Science, Technology, Engineering and Mathematics education. This paper aims at describing the features of a remote lab named Virtual Instruments System in Reality, embedded in a community of practice and forming the spearhead of a federation of remote labs. More particularly, it discusses the advantages and disadvantages of remote labs over virtual labs as regards to scalability constraints and development and maintenance costs. Finally, it describes an actual implementation in an international community of practice of engineering schools forming the embryo of a first world wide federation of Virtual Instruments System in Reality nodes, under the framework of a project funded by the Erasmus+ Program.
In this paper, a new load model is proposed for electric power distribution systems under varying... more In this paper, a new load model is proposed for electric power distribution systems under varying voltage conditions in order to estimate the energy losses and, consequently, optimize the energy losses in the system. The proposed methodology is based on the adjustment of the polynomial ZIP load parameters related to active and reactive power as a function of the static voltage variations measured at the substation. The ZIP parameters are determined using the Genetic Algorithm in order to minimize the error between the measured and estimated active and reactive power at the substation. The estimation of the energy losses during a period of analysis is determined by the computation of the power flow using the ZIP model of each load in the system and the substation voltages measured. The procedure requires information of the feeder topology, distribution lines, rated power of the transformers, and a database containing voltage and power measurements at the substation during the period ...
This article presents a load modeling application using real data measurements to estimate peak d... more This article presents a load modeling application using real data measurements to estimate peak demand reduction due to voltage control. The load models parameters estimates yield conservation voltage reduction indicators, that quantify the effectiveness of this smart grid approach designed to save energy and reduce power demand at peak times. The data analyzed were recorded from UK primary distribution substations, chosen in the scope of a research & development project. Stratified random sampling and statistical inference were used to extract information that gathered load composition differences due to the time of the day, the day of the week and the season, aiming the assessment of the confidence level of the peak demand reduction estimation.
Advances in Intelligent Systems and Computing, 2018
Laboratories are a fundamental part of engineering education due to the very nature of the engine... more Laboratories are a fundamental part of engineering education due to the very nature of the engineering profession. This is a characteristic of all engineering courses, though it may vary from one curriculum to the other and even in the same curriculum. This paper is dedicated to the analysis of different applications of a remote lab in an Electrical Engineering curriculum. If the types of experiments it offers are concerned, it is classified as an Electric and Eletronic Circuits lab; one in a set that also has Digital Electronics, Analog Electronics, Electrical Machines, Control Systems, etc. But it is not a traditional labit is a remote lab, with traditional components remotely accessed over the Internet. This work presents the preliminary results of the deployment of VISIRa remote lab for Electric and Electronic Circuits in some courses. It discusses the different course contexts and how the use of VISIR was adapted to each one. Results of students' opinions are presented and discussed.
This paper proposes a novel approach to achieve better results for the Non-intrusive load disaggr... more This paper proposes a novel approach to achieve better results for the Non-intrusive load disaggregation model for residential consumers. The proposed methodology considers two steps to face the problem. Firstly, periodical appliances are modelled by the Fourier series and extracted from the total power measured. Secondly, a Mixed Integer Linear Programming was proposed to disaggregate the remaining appliances. The paper presents results with a real case and indicates that the model can be useful for practical applications, such as the possibilty of the consumers of changing the modality of their tariff contract.
Remote Laboratories have become part of current teaching and learning, particularly in engineerin... more Remote Laboratories have become part of current teaching and learning, particularly in engineering. Their potential to aid students beyond their hands-on lab classes has been a matter of discussion in literature. Teachers and researchers are aware that the thorough analysis of both strengths and shortcomings of remote labs in didactical implementations may not only lead to the improvement of these resources but also of the pedagogical implications in engineering classes. The present study was carried out in a Higher Education Institution in Brazil in two different courses during three consecutive semesters where a remote lab (VISIR) addressing electric and electronic topics was implemented, yielding 471 students' academic results and opinions. These students' results (while using VISIR) cross-analysed with the course characteristics, reveal some factors teachers may tackle to foster student learning and motivation. The conclusions point to the need for VISIR interface modernization and showed it is more useful in basic courses than in more advanced ones, when dealing with classic lab experiments. Results also show that teachers' involvement plus their ability to brief students on VISIR's usefulness have a significant influence not only on students' performance but also on their perception of learning and satisfaction with the tool. In the analysed cases, the students with more learning needs seemed to be the ones who could benefit more from VISIR.
Because electricity generation from renewable energy is seasonal and intermittent, the risk assoc... more Because electricity generation from renewable energy is seasonal and intermittent, the risk associated with its cash flow volatility may discourage investments in this area. In this paper, a method called Energy Reallocation Mechanism is proposed to mitigate this problem. The idea, inspired by the current mechanism used in Brazil for Hydropower plants, consists of the creation of a renewable energy portfolio and the allocation of the quotas of this portfolio to each source. Because the portfolio cash flow is more stable than the individual cash flows, the financial risk of being in the portfolio is reduced. The proposed mechanism is based on a stochastic approach and it combines future scenarios of generation in order to produce future scenarios of cash flow. Beyond the Expected Value applied in the Brazilian mechanism, it also considers the Conditional Value at Risk to account for average and risk, respectively, associated with the generation profile of each source. Finally, based on the aforementioned, the Marginal Benefit method is applied to allocate the portfolio's quotas for the renewable sources guaranteeing that the solution provided is suitable under the cooperative game theory viewpoint. For didactic reasons a small renewable energy portfolio, formed by Wind Power, Small Hydro and Biomass, is used for comparing the proposed mechanism with the Shapley Value method and the current method applied in Brazil. The applicability for larger systems is demonstrated with real data from the Brazilian Power System.
The Brazilian Electricity Regulatory Agency (ANEEL) presented a proposal to revise the tariff str... more The Brazilian Electricity Regulatory Agency (ANEEL) presented a proposal to revise the tariff structure of distribution companies in Brazil. One of the main approved suggestions was to establish a mechanism called Tariff Flags, which aims to foster a Demand Response Program in Brazil via an increase in the energy tariff. In this work, the proposed mechanism is reviewed in detail and the expected results of its application are simulated and analyzed under different perspectives. This paper shows that the system operation directly impacts the Demand Response Program, since the spot prices will define which Tariff Flag should be triggered. In order to encompass and assess the main consequences of its application, this paper presents the expected effects on energy spot prices, system operating costs, probability of triggering each flag, investment recovery for utilities and finally, the impact for the final consumers. The case studies presented in this paper were developed using real information about the Brazilian electrical system for each economic sector and the price-demand elasticity is discussed using the literature for this application. Finally, some conclusions and guidelines are provided to improve the application of the mechanism.
Abstract One of the main objectives of Demand Response Programs is to mitigate the effect of peak... more Abstract One of the main objectives of Demand Response Programs is to mitigate the effect of peak electricity demand by inducing consumers to shift or reduce their electricity consumption in peak times. In Brazil, big energy consumers should contract peak demand for the upcoming months with the utility distribution companies under a set of rules defined by the System Regulator. Based on this, the utility will have more information to reinforce the system accordingly. The challenge for consumers is to compute the best value of the peak demand contracted before the peak demand realization. One way to solve this problem is to simulate future scenarios of peak demand and optimize the value of the peak demand contracted in compliance with the current rules. In this paper, a support decision system is proposed to help consumers in this task. The proposal is divided into two parts, which are a statistical model, for estimating and simulating future scenarios of peak demand, and a stochastic optimization model, to compute the value of the peak demand contracted. In the first part, a Box & Jenkins model is used to estimate the parameters of the statistical model and to simulate it based on the historical electricity bill data. In the second part, a stochastic optimization model is applied using a convex combination of the Expected Value and Conditional Value-at-Risk, as the risk metrics for the cost uncertainty, in order to obtain the monthly values of peak demand contracted. The results achieved corroborate the importance of using an appropriate mathematical model to address the problem. To illustrate the proposed approach, a real case study for a big consumer in Brazil is presented.
A energia eólica vem se firmando de maneira concreta no Brasil. Incentivos foram dados a esta fon... more A energia eólica vem se firmando de maneira concreta no Brasil. Incentivos foram dados a esta fonte nos leilões do ACR. Contudo, o modelo de planejamento e simulação da operação vigente no Brasil desconsidera a incerteza na geração das fontes alternativas renováveis. Em aplicações, como a comercialização de energia proveniente dessas fontes, uma metodologia de simulação que produza cenários casados com grandezas sistêmicas torna-se necessária. Assim, este trabalho apresenta um modelo de geração de cenários de produção de fontes renováveis intermitentes que sejam dependentes dos cenários de variáveis sistêmicas, PLD, geração térmica, intercâmbios, etc., advindos do modelo de planejamento operativo.
2012 IEEE Power and Energy Society General Meeting, 2012
ABSTRACT Wind power generation plays an important role in most power systems worldwide. Despite t... more ABSTRACT Wind power generation plays an important role in most power systems worldwide. Despite that, only in 2009, wind farms appeared as a profitable and competitive investment option in Brazil. The Brazilian power system is mainly hydro based and, due to its many singularities, its coordination is still centralized. A collateral effect of such centralized dispatch coordination is that the system marginal costs, which determine the short-term energy prices, exhibit a highly volatile pattern. In this setting, selling energy through bilateral forward contracts backed up on intermittent generation profiles, such as wind farms and small-run hydros, exposes the generation company (Genco) to the so-called price-quantity risk. In this paper, we explore the well-known complementarity between these two renewable resources (wind and inflows) by proposing a statistical model capable to produce scenarios of renewable resources availability consistently with short-term price scenarios. Hence, we present a novel commercial model for a wind power producer based on a joint-selling strategy with a small-hydro Genco. In this model, the hydro Genco receives a surplus payment in comparison to the amount it would receive in the market and the wind Genco, the rest of the income. We show, by means of realistic data from the Brazilian power system, that such commercial model is able to mitigate the exposure to the short-term price and foster the wind power penetration into the contract market.
2013 IEEE Power & Energy Society General Meeting, 2013
ABSTRACT In this study, a fault location algorithm for electrical power systems is presented. The... more ABSTRACT In this study, a fault location algorithm for electrical power systems is presented. The proposal consists in an accurate impedance-based fault location algorithm by utilizing voltage and current measurements from local terminal and estimated remote source impedance. In order to evaluate the fault location algorithm a numerical analysis is made. Phase-to-ground faults are simulated in a 230 kV Brazilian transmission line using the software ATP/EMTP and MATLAB. The algorithm is tested considering the fault distance, fault resistance, and the variation of short-circuit capacity of remote system effects. The results show that the fault resistance and fault distance affect the estimated fault distance, but with minor errors. More, the proposed algorithm is not sensitive to the remote system short-circuit capacity variation.
Uploads
Papers by Delberis Lima