Assessment of ANN and SVM models for estimating normal direct irradiation (Hb)

Abstract

a b s t r a c tThis study evaluates the estimation of hourly and daily normal direct irradiation (H b ) using machine learning techniques (ML): Artificial Neural Network (ANN) and Support Vector Machine (SVM). Time series of different meteorological variables measured over thirteen years in Botucatu were used for training and validating ANN and SVM. Seven different sets of input variables were tested and evaluated, which were chosen based on statistical models reported in the literature. Relative Mean Bias Error (rMBE), Relative Root Mean Square Error (rRMSE), determination coefficient (R 2 ) and ''d" Willmott index were used to evaluate ANN and SVM models. When compared to statistical models which use the same set of input variables (R 2 between 0.22 and 0.78), ANN and SVM show higher values of R 2 (hourly models between 0.52 and 0.88; daily models between 0.42 and 0.91). Considering the input variables, atmospheric transmissivity of global radiation (kt), integrated solar constant (H sc ) and insolation ratio (n/N, n is sunshine duration and N is photoperiod) were the most relevant in ANN and SVM models. The rMBE and rRMSE values in the two time partitions of SVM models are lower than those obtained with ANN. Hourly ANN and SVM models have higher rRMSE values than daily models. Optimal performance with hourly models was obtained with ANN4 h (rMBE = 12.24%, rRMSE = 23.99% and ''d" = 0.96) and SVM4 h (rMBE = 1.75%, rRMSE = 20.10% and ''d" = 0.96). Optimal performance with daily models was obtained with ANN2 d (rMBE = À3.09%, rRMSE = 18.95% and ''d" = 0.97) and SVM2 d (rMBE = 0.60%, rRMSE = 19.39% and ''d" = 0.97). ANN and SVM models improved H b estimations as compared with other results from the literature. SVM has better performance than ANN to estimate H b , and it should be the first option of choice.

Review on forecasting of photovoltaic power generation based on machine learning and metaheuristic techniques

“…A number of combinations have been used as hybrid methods by different researchers. These combinations are (i) genetic algorithms (GAs) and ANNs, (ii) fuzzy and ANNs, (iii) ANFIS, (iv) ANNs and physical model, (v) ANNs and ARMA, (vi) wavelets and ANNs, (vii) ANN and optimisation algorithms, (viii) WT and SVM, (ix) SVM and optimisation algorithms, and (x) seasonal auto-regressive integrated moving (SARIMA) and SVM [3,4,19,110]. From a comprehensive literature review, these combinations are described as follows.…”

Predictive Models for the Characterization of Internal Defects in Additive Materials from Active Thermography Sequences Supported by Machine Learning Methods

“…The MLP allows to represent some smooth measurable functional relationships between the inputs (predictors features) and the outputs (responses). MLP is a distributed, information processing system massively parallel and successfully applied for the generation of models to solve non-linear problems [ 39 , 40 ]. The processes are based on three different layers of neurons: input layers ( N neurons), hidden layers ( S neurons) and output layers ( L neurons), where each layer has a group of connected points (neurons).…”

Data mining tools in identifying the components of the microstructure of compacted graphite iron based on the content of alloying elements

“…As a model, they actually represent a black box. Their character is difficult to interpret for a man, so their use is limited to the implementation of computer tools, and the only knowledge that can be drawn is in the form of numerical results [43]. The situation is quite different in the case of decision trees (Table 8).…”