Browsing by Author "Montalvan Delgado, Joel Alejandro"

Now showing 1 - 2 of 2

Integrating artificial neural networks and cellular automata model for spatial-temporal load forecasting
(2023) Franco, John Fredy
The long-term distribution planning should include an understanding of consumer behavior and needs to develop strategic expansion alternatives that meet the future demand. The magnitude of growth along with the place where and when it will be developed are determined by the spatial load forecasting. Thus, this paper proposes a spatial-temporal load forecasting method to recognize and predict development patterns using historical dynamics and determine the development of consumers and electric load in small areas. An artificial neural network is integrated to a cellular automaton method to establish transition rules, based on land-use preferences, neighborhood states, spatial constraints, and a stochastic disturbance. The main feature is the incorporation of temporality, as well as taking advantage of geospatial-temporal data analytics to calibrate and validate a holistic and integral framework. Validation consists of measuring the spatial error pattern during the training and testing phase. The performance of the method is assessed in the service area of an Ecuadorian power utility. The knowledge extraction from large-scale data, evaluating the sensitivity of parameters and spatial resolution was carried out in reasonable times. It is concluded that adequate normalization and use of temporality in the spatial factors improve the error in the spatial-temporal load forecasting.
Proyección espacial de la demanda en la Empresa Regional Centrosur C.A, mediante métodos heurísticos
(Universidad de Cuenca, 2019-12-18) Montalvan Delgado, Joel Alejandro; Morales Jadan, Rommel Eduardo; Salgado Rodríguez, Modesto Enrique; Patiño Chitacapa, César Andrés
The next work presents the development and application of models focused on load forecasting through trend and simulation in order to solve the medium and large issues of the electric distribution system of the “Empresa Eléctrica Regional Centro Sur C.A”. The trend-focused model will use Matlab's tools such as Fuzzy Logic (FL) and Artificial Neuronal Networks (ANN) relating variables such as GNP, customers and population with energy consumption, making use of its records to project new consumption. The result of the trend model is compared with the load forecast made by the distribution company using the Holt Winter method. The simulation model will focus only on the urban areas of Cuenca for residential customers, using the programming language Python, to create a probability map by training a neural network that analyzes the evolution of spatial factors of proximity, environmental and local in a temporary way at the geographical grid level. To later disaggregate the global load forecasting of customers in each suitable grid categorized by means of the mathematical model known as Cellular Automata (CA), which is responsible for assigning new customers, and then converting this increment of customers into power demand for the load forecasting at 2033.