The workshop “Challenges of applied mathematics in the Brazilian energy industry” brought together professionals and experts from various companies to discuss how the technological transformations underway in the sector impact the mathematical models in use. The event held this Monday, in the auditorium of Imecc (Institute of Mathematics, Statistics and Scientific Computing) at Unicamp, also aimed to evaluate the possibilities of applied mathematics in the face of the so-called “green revolution”, led by the increasing penetration of sources of photovoltaic and wind energy.
Secundino Soares Filho, professor at the Faculty of Electrical and Computer Engineering (FEEC) whose line of research is mathematics applied to the operation of the electricity sector, questioned the operating paradigms prevalent in Brazil. “The first paradigm is that the electricity sector adopts stochastic models for system operation, without ever testing deterministic models, which in our tests proved to be almost as good as the first ones for systems with a single plant. For systems like the Brazilian one, which has 150 plants, stochastic models need to linearize the equations, while deterministic models can work in a non-linear way.”
Another paradigm listed by the Unicamp professor comes from the comparison of the non-linear deterministic model developed in his research with the linear stochastic model in force. “I will demonstrate that the performance of our model is much higher than that used by the ONS [National Electric System Operator]. And the third paradigm is in the policy adopted by the electricity sector of storing water at the beginning of the cascade (at the headwaters of the rivers), believing that that water will generate energy in all downstream plants. We have proven that it is better to store water in lower reservoirs, as storing water up there is a huge waste of energy. I defend the need to review these strategies.”
The workshop at Unicamp was promoted by the Center for Mathematical Sciences Applied to Industry (CeMEAI), which is based at USP in São Carlos and constitutes one of the Cepids (Research, Innovation and Dissemination Centers) financed by Fapesp. Its director, Professor José Alberto Cuminato, stated that CeMEAI's objective is to convince the industry that mathematical sciences can serve it very well. “This Cepid project is completing five years and we will submit the request for an extension for another three years at the end of the month. We basically have four areas of activity: optimization applied in operational research, computational flow mechanics, risk assessment and statistics, and computational intelligence and software engineering.”
“Impact of mathematical models on the regulation of hydroelectric generation” was the topic addressed by professor Claudia Sagastizabal, from Imecc, together with Bruno Goulart, a specialist in regulation at Aneel (National Electric Energy Agency). “I will try to put into intelligible language for mathematicians what will be presented by Goulart in relation to the gap between physical reality and the mathematical model. I say that the models are like maps, imperfect but useful, and that they can be improved with the work of mathematicians”, compares the Unicamp professor.
Claudia Sagastizabal said that it would show the impact on society of the lack of improvement in the model, simply because there is no human competence to run in parallel with what happens in reality. “An example is hydrological insurance, compensation that must be defined and attributed by the regulatory agency to hydroelectric plants. If the hydroelectric plant is unable to meet the stipulated energy production target – and we are in a very prolonged period of drought –, we have a difference that represents an injustice for the generator. So, the regulator needs to define the amount of compensation in volume and the price: this [mathematical] problem is very interesting.”
