Alexandre Rasi Aoki

Centralized and decentralized approaches to demand response using smart plugs

Residential customers demand generally vary during the day, causing overloads in the distribution system and increasing energy costs. This paper investigates the use of centralized and decentralized scheduling algorithms to help customers adjusting their consumption pattern in demand response programs with Time-Of-Use Tariff. Two algorithms are proposed, one with centralized information from several smart plugs communicating with a Home Automation Controller, and another one, decentralized, running local on each smart plug. The simulations, that used real data from smart plug prototypes from a Brazilian home, indicate that the use of such algorithms can effectively help residential customers achieve Demand-Side Management goals.

Energy efficiency indicators assessment tool for the industry sector

The improvement of energy efficiency in industries generates positive impacts on the economy, society and environment. The energy efficiency indicators (EEI) are important to have the diagnosis of the energy efficiency at the industries facilities through the utilization of energy management software that uses EEI. It can also help the power utility companies to improve the expansion planning of their network. This paper presents a methodology to select EEI and the management system developed for the industries to be able to manage, compare and benchmark their EEI with the average at the same industry segment.

Load shedding through optimal power flow to support self-healing actions in distribution feeders

One of the main characteristics of the smart grid is the implementation and development of tools that allow the minimization of system vulnerability as well as an increase in security, reliability and power quality in distribution systems. This paper presents the development of a methodology to support self-healing through demand-side management (load shedding). This methodology allows the self-healing system to restore more loads when compared to traditional methods, through the indication of loads to shed in each bus of the distribution system, taken into account power quality constraints. The methodology was implemented using a Binary Particle Swarm Optimization (BPSO) to self-healing switching determination and Optimal Power Flow (OPF) for the load shedding scheme. The methodology proved to be able to restore more load in an optimal way, minimizing losses and assuring power quality.

Transforming continuous attributes using GA for applications of Rough Set Theory to control centers

One of the possible application of Rough Sets Theory (RST) is the knowledge extraction in databases. Also, RST is useful to develop models for decision-making. During both processes one of the steps is the transformation of attributes with continuous values in digital values. This transformation sometimes can lose information. This paper presents a method for this transformation using genetic algorithms (GA). GA is used to determine the cut-off points for each attribute, getting a consistent transformation. An application in Control Centers with real data is presented.

Applications of Genetic Algorithm in Power System Control Centers

Electric power systems are the structures that have the purpose of generating, transporting and distributing electricity. These structures are composed of energy generating plants, transmission lines, distribution systems and elevating and distribution substations. The biggest challenge of this type of system is that the electrical energy is not a material that can be stored. That is, so it is required by the load (which can be a driver in an industry or the mere light of a lamp in a residence) this energy must be generated, transported and delivered to the user instantly.

Extracting procedures and knowledge from residential device database

Smart grid is the focal subject in our days. Many developments and policies have been created to support this new operation and planning philosophy. In special for consumer load management, where new equipments, controls, fares, and operation policies are been built. This paper is another step in this direction. It presents a device with automatic control to minimize the refrigerator consumption during the peak load periods. This control is based on consumer habits during the peak load and the thermal limits of the refrigerator. An AI granular computation is used to understand the consumer habits. Studies with a refrigerator manufacturer had been made and a complete system is presented.

Building Integrated Photovoltaics for smart grids

This paper presents a case study of a Building Integrated Photovoltaics deployed in a parking lot. Computational models were developed in order to estimate the energy generation, and the usage of batteries for energy management. A developed energy management system, with an algorithm to work with Time-Of-Use tariffs, electric vehicles and storage systems is also described and simulated. The results achieved in a pilot project are also discussed.

Case study on distributed generation for electric vehicle and energy storage integration

This paper presents an integrated system of photovoltaic (PV), energy storage and a charging station for an electric vehicle (EV). The operation mode of the integrated system is controlled by an algorithm which is based on a new Brazilian tariff supported on different prices depending on the hour and the day of the week. It will also be presented a panorama of the distributed generation in Brazil, and the challenges faced during the project and construction of this system.

Energy management application for smart grids aiming at mobile device

The presence of electricity in various sectors of the economy allowed it to reach a level of importance, in which its rational consumption has become essential to maintain and increase its offer. This requires solutions (as mobile applications) that enable the costumer to know their instantaneous consumption and act on it. This paper presents methods for monitoring and controlling the energy consumption, which supported the development of an Android application (for mobile devices) to interact with a Smart Grid Network. It proposes methods for daily, weekly and monthly analysis, establishment of targets of energy consumption or targets of energy spending, and load control. The actual application are also presented along with its functional description. Through this application the costumer may visualize its energy consumption data and use this information to support decisions regarding the management, control or adequacy of their consumption habits.

A Fuzzy-QFD decision making approach for selecting industry energy efficiency indicators

The improvement of energy efficiency generates positive impacts on the society and environment, and to be able to have a diagnosis of energy efficiency it is necessary to use energy efficiency indicators (EEI). The use of energy management software that uses EEI can help the energy utility to improve the expansion planning of its electricity generation, transmission and distribution system. Furthermore the utility can create energy efficiency programs targeted to industry sectors where the EEI show a situation with improvement opportunities. This paper demonstrates an example through a computational simulation using the Fuzzy Quality Function Deployment (FQFD) to achieve the prioritization of EEI as inputs for an energy management software development. The simulation enabled to achieve prioritization of EEI in a situation where the incoming data for QFD were introduced as linguistic variables rather than numeric variables, building up an opportunity to use a hybrid FQFD approach.