A permutation entropy-based EMD–ANN forecasting ensemble approach for wind speed prediction

Assessment of hybrid machine learning algorithms using TRMM rainfall data for daily inflow forecasting in Três Marias Reservoir, eastern Brazil

This study investigates the application of the Gaussian Radial Basis Function Neural Network (GRNN), Gaussian Process Regression (GPR), and Multilayer Perceptron Optimized by Particle Swarm Optimization (MLP-PSO) models in analyzing the relationship between rainfall and runoff and in predicting runoff discharge. These models utilize autoregressive input vectors based on daily-observed TRMM rainfall and TMR inflow data. The performance evaluation of each model is conducted using statistical measures to compare their effectiveness in capturing the complex relationships between input and output variables. The results consistently demonstrate that the MLP-PSO model outperforms the GRNN and GPR models, achieving the lowest root mean square error (RMSE) across multiple input combinations. Furthermore, the study explores the application of the Empirical Mode Decomposition-Hilbert-Huang Transform (EMD-HHT) in conjunction with the GPR and MLP-PSO models. This combination yields promising results in streamflow prediction, with the MLP-PSO-EMD model exhibiting superior accuracy compared to the GPR-EMD model. The incorporation of different components into the MLP-PSO-EMD model significantly improves its accuracy. Among the presented scenarios, Model M4, which incorporates the simplest components, emerges as the most favorable choice due to its lowest RMSE values. Comparisons with other models reported in the literature further underscore the effectiveness of the MLP-PSO-EMD model in streamflow prediction. This study offers valuable insights into the selection and performance of different models for rainfall-runoff analysis and prediction.

Carbon price prediction based on multiple decomposition and XGBoost algorithm

Carbon trading is an effective way to limit global carbon dioxide emissions. The carbon pricing mechanisms play an essential role in the decision of the market participants and policymakers. This study proposes a carbon price prediction model, multi-decomposition-XGBOOST, which is based on sample entropy and a new multiple decomposition algorithm. The main steps of the proposed model are as follows: (1) decompose the price series into multiple intrinsic mode functions (IMFs) by using complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN); (2) decompose the IMF with the highest sample entropy by variational mode decomposition (VMD); (3) select and recombine some IMFs based on their sample entropy, and then perform another round of decomposition via CEEMDAN; (4) predict IMFs by XGBoost model and sum up the prediction results. The model has exhibited reliable predictive performance in both the highly fluctuating Beijing carbon market and the comparatively stable Hubei carbon market. The proposed model in Beijing carbon market achieves improvements of 30.437%, 44.543%, and 42.895% in RMSE, MAE, and MAPE, when compared to the single XGBoost models. Similarly, in Hubei carbon market, the RMSE, MAE, and MAPE based on multi-decomposition-XGBOOST model decreased by 28.504%, 39.356%, and 39.394%. The findings indicate that the proposed model has better predictive performance for both volatile and stable carbon prices.

Wavelet kernel least square twin support vector regression for wind speed prediction

Wind energy is a powerful yet freely available renewable energy. It is crucial to predict the wind speed (WS) accurately to make a precise prediction of wind power at wind power generating stations. Generally, the WS data is non-stationary and wavelets have the capacity to deal with such non-stationarity in datasets. While several machine learning models have been adopted for prediction of WS, the prediction capability of primal least square support vector regression (PLSTSVR) for the same has never been tested to the best of our knowledge. Therefore, in this work, wavelet kernel-based LSTSVR models are proposed for WS prediction, namely, Morlet wavelet kernel LSTSVR and Mexican hat wavelet kernel LSTSVR. Hourly WS data is gathered from four different stations, namely, Chennai, Madurai, Salem and Tirunelveli in Tamil Nadu, India. The proposed models' performance is assessed using root mean square, mean absolute, symmetric mean absolute percentage, mean absolute scaled error and R². The proposed models' results are compared to those of twin support vector regression (TSVR), PLSTSVR and large-margin distribution machine-based regression (LDMR). The performance of the proposed models is superior to other models based on the results of the performance indicators.

An innovative forecasting model to predict wind energy

In recent years, the global wind power construction is accelerating. Although wind power is a clean energy without pollution, its volatility and irregularity have a great impact on wind power integration. Therefore, scholars pay more and more attention to the ultra-short-term prediction of wind speed. At present, the popular wind speed prediction model usually combines wind speed decomposition algorithm, machine learning algorithm, and intelligent optimization algorithm. The general wind speed decomposition algorithm cannot use the information contained in the factors affecting wind speed. Besides, the current popular optimization algorithms, such as gray wolf optimization algorithm, have strong convergence and better optimization effect, but their structure is complex and their operation complexity is large. And the PSO algorithm has simple structure and fast operation speed. To solve the above problems, a novel combination prediction model is proposed in this paper. This model uses VMD to decompose the wind speed into high-frequency signal and low-frequency signal and then uses principal component analysis and spectral clustering to extract and classify the influencing factors. In addition, aiming at the problem of slow convergence speed in the later stage of PSO iteration, an adaptive improved PSO is proposed. Finally, this paper also designs a rolling train method to adjust the size of training samples. Through four experiments of wind speed series in two periods, it is proved that the combined prediction model proposed in this paper has the following advantages: the model fully extracts the information of wind speed and influencing factors; the improved PSO algorithm has better optimization effect; rolling training method effectively improves the prediction ability of the model; the combined forecasting model has good adaptability and competitiveness.

Air pollution relevance analysis in the bay of Algeciras (Spain)

The aim of this work is to accomplish an in-depth analysis of the air pollution in the two main cities of the Bay of Algeciras (Spain). A large database of air pollutant concentrations and weather measurements were collected using a monitoring network installed throughout the region from the period of 2010-2015. The concentration parameters contain nitrogen dioxide (NO2), sulphur dioxide (SO2) and particulate matter (PM10). The analysis was developed in two monitoring stations (Algeciras and La Línea). The higher average concentration values were obtained in Algeciras for NO2 (28.850 µg/m3) and SO2 (11.966 µg/m3), and in La Línea for PM10 (30.745 µg/m3). The analysis shows patterns that coincide with human activity. One of the goals of this work is to develop a useful virtual sensor capable of achieving a more robust monitoring network, which can be used, for instance, in the case of missing data. By means of trends analysis, groups of equivalent stations were determined, implying that the values of one station could be substituted for those in the equivalent station in case of failure (e.g., SO2 weekly trends in Algeciras and Los Barrios show equivalence). On the other hand, a calculation of relative risks was developed showing that relative humidity, wind speed and wind direction produce an increase in the risk of higher pollutant concentrations. Besides, obtained results showed that wind speed and wind direction are the most important variables in the distribution of particles. The results obtained may allow administrations or citizens to support decisions.

Short-term wind speed prediction using hybrid machine learning techniques

Wind energy is one of the potential renewable energy sources being exploited around the globe today. Accurate prediction of wind speed is mandatory for precise estimation of wind power at a site. In this study, hybrid machine learning models have been deployed for short-term wind speed prediction. The twin support vector regression (TSVR), primal least squares twin support vector regression (PLSTSVR), iterative Lagrangian twin parametric insensitive support vector regression (ILTPISVR), extreme learning machine (ELM), random vector functional link (RVFL), and large-margin distribution machine-based regression (LDMR) models have been adopted in predicting the short-term wind speed collected from five stations named as Chennai, Coimbatore, Madurai, Salem, and Tirunelveli in Tamil Nadu, India. Further to check the applicability of the models, the performance of the models was compared based on various performance measures like RMSE, MAPE, SMAPE, MASE, SSE/SST, SSR/SST, and R². The results suggest that LDMR outperforms other models in terms of its prediction accuracy and ELM is computationally faster compared to other models.

Solar Power System Assessments Using ANN and Hybrid Boost Converter Based MPPT Algorithm

The load pressure on electrical power system is increased during last decade. The installation of new power generators (PGs) take huge time and cost. Therefore, to manage current power demands, the solar plants are considered a fruitful solution. However, critical caring and balance output power in solar plants are the highlighted issues. Which needs a proper procedure in order to minimize balance output power and caring issues in solar plants. This paper investigates artificial neural network (ANN) and hybrid boost converter (HBC) based MPPT for improving the output power of solar plants. The proposed model is analyzed in two steps, the offline step and the online step. Where the offline status is used for training various terms of ANNs in terms of structure and algorithm while in the online step, the online procedure is applied with optimum ANN for maximum power point tracking (MPPT) using traditional converter and hybrid converter in solar plants. Moreover, a detail analytical framework is studied for both proposed steps. The mathematical and simulation approaches show that the presented model efficiently regulate the output of solar plants. This technique is applicable for current installed solar plants which reduces the cost per generation.

Artificial Neural Networks, Sequence-to-Sequence LSTMs, and Exogenous Variables as Analytical Tools for NO2 (Air Pollution) Forecasting: A Case Study in the Bay of Algeciras (Spain)

This study aims to produce accurate predictions of the NO₂ concentrations at a specific station of a monitoring network located in the Bay of Algeciras (Spain). Artificial neural networks (ANNs) and sequence-to-sequence long short-term memory networks (LSTMs) were used to create the forecasting models. Additionally, a new prediction method was proposed combining LSTMs using a rolling window scheme with a cross-validation procedure for time series (LSTM-CVT). Two different strategies were followed regarding the input variables: using NO₂ from the station or employing NO₂ and other pollutants data from any station of the network plus meteorological variables. The ANN and LSTM-CVT exogenous models used lagged datasets of different window sizes. Several feature ranking methods were used to select the top lagged variables and include them in the final exogenous datasets. Prediction horizons of t + 1, t + 4 and t + 8 were employed. The exogenous variables inclusion enhanced the model's performance, especially for t + 4 (ρ ≈ 0.68 to ρ ≈ 0.74) and t + 8 (ρ ≈ 0.59 to ρ ≈ 0.66). The proposed LSTM-CVT method delivered promising results as the best performing models per prediction horizon employed this new methodology. Additionally, per each parameter combination, it obtained lower error values than ANNs in 85% of the cases.