Energy Central Professional


Article: How artificial intelligence can give reliability to a clean and renewable energy matrix

CE Noticias Financieras  


    ARTHUR OLIVEIRA - Electricity sector specialist, executive manager for Data & Analytics NTT DATA

    The pressure to avoid the major impacts of climate change has called for greater use of alternative energy sources, such as wind and solar, which emit far fewer greenhouse gases than fossil fuel-based sources. One of the biggest challenges for the energy transition is linked to the intermittency of light and wind, which can cause disruptions in energy supply. The construction of a reliable energy matrix involves guaranteeing the uninterrupted process of energy generation and distribution. People and companies cannot afford to suffer from a lack of supply. This is why, in several countries, the solution was to invest in coal-fired thermoelectric power plants, which can be quickly turned on to guarantee supply.

    A key tool available to companies in the sector to meet these challenges is new digital technology. More specifically, artificial intelligence. There are numerous practical applications of artificial intelligence and machine learning techniques using data for more accurate forecasts of wind supply and solar incidence. Today AI is able to go beyond predictive analysis (forecasts) and can do prescriptive analysis. That is, it takes into account the correlation of a range of information, statistics, and historical data to determine practical actions to reduce errors and prediction, making the model even more assertive.

    In a practical case to determine actions to avoid solar and wind energy shortages, the first step is to determine the variables that will be measured. For example, solar flux density, wind speed, temperatures, and measurement of energy use by consumers. The second step is to create the algorithm with the calculations. From the monitoring of the environment and climate changes, done by digital sensors, data is generated in real time. The information collected is treated and stored, and then used by the algorithm.

    With this, the AI algorithm is able to determine when there may be shortages, based on consumption and on the time of year with less sun and wind. Also, based on this data, the algorithm preemptively recommends actions such as the best time to do energy storage (solar), predict the increased capacity of the systems, and the appropriate time to use more solar, wind, or both combined generation capacity.

    There are other uses. For example, AI can also be used for analysis and monitoring of the power transmission grid, performing analysis between the energy generated and the energy consumed in order to find points of technical losses in the distribution grid. In the case of non-technical losses, it is possible to identify possible frauds by creating a repository of patterns of typical cases by different characteristics: customer types, measuring point type, contracted power, tariff and geographical areas from the historical data of the cases.

    This process occurs in a cyclical manner. With each new cycle, the AI learns something new from the newly generated data, reducing errors and increasing the accuracy of predictions and recommendations. This is done by a subset of AI called machine learning, which aims to learn from the data and increasingly improve the results of the answers autonomously for most cases, requiring human intervention for new variables.

    Brazil is very well positioned to become one of the largest global producers of clean and renewable energy. The country has a good insolation rate and stable wind, especially in the Northeast Region. According to data from the Energy Research Company (EPE), almost 85% of the Brazilian electricity matrix comes from renewable sources. The use of artificial intelligence is the missing point for us to have a reliable energy matrix that presents fewer risks of shortages or rationing.


Copyright © 1996-2022 by CyberTech, Inc. All rights reserved.
Energy Central® and Energy Central Professional® are registered trademarks of CyberTech, Incorporated. Data and information is provided for informational purposes only, and is not intended for trading purposes. CyberTech does not warrant that the information or services of Energy Central will meet any specific requirements; nor will it be error free or uninterrupted; nor shall CyberTech be liable for any indirect, incidental or consequential damages (including lost data, information or profits) sustained or incurred in connection with the use of, operation of, or inability to use Energy Central. Other terms of use may apply. Membership information is confidential and subject to our privacy agreement.