Search

The present research on photovoltaic (PV) forecasting is devoted to the use of spatial information about PV sites to improve the accuracy of the models, but most of the models have to increase their network complexity to learn the spatial dependence. In this paper, we propose to take advantage of the known geographic location information of PV sites and embed them directly into the input information of Decoder, which makes it easier for the model to focus its attention. We refer to this process as relevance amplification. Based on this, this paper proposes Relevance Amplification based DeepFEDformer (RAD-FEDformer), where DeepFEDformer adds multiple Multi-Layer Perceptron (MLP) layers to FEDformer to improve the perception of deep features. The Relevance Amplification Module (RAM) is designed to receive geographic correlation information as a way to enhance its influence on the Seasonality component of the Decoder input and improve the performance of the attention mechanism in the model. Using the power generation data from PV plants distributed in 11 regions of Belgium as our case study, we evaluated the performance of RAD-FEDformer in predicting PV data at 48/96/192 time steps into the future for each plant. Compared to other Transformer family models, RAD-FEDformer achieved SOTA results by demonstrating significant R 2 improvements of 16.70%, 51.45%, and 23.50% over FEDformer, along with an average MSE reduction of 16.9%. We designed Ablation Experiments to validate and compare the performance of MLPs of different sizes based on the ETTm1 dataset and discussed the performance of RAM and its effect on the attention mechanism based on the PV dataset. The results show that our model enhances the performance of the attention mechanism in Multi-Site PV prediction scenarios, and conclude that the optimization of RAM is more effective in longer sequence predictions. • Propose a module for embedding geographic information into input data with low overhead. • Improving the FEDformer model using Multi-layer Perception. • Discuss the performance of the model on multivariate prediction task based on ETT and PV datasets. • Compare several popular Transformer family forecasting models in the case study. [ABSTRACT FROM AUTHOR]

Accurate photovoltaic (PV) power forecast is crucial to power systems. Solar irradiance forecast is the fundamentals of PV power forecast. Current researches merely focus on improving the forecast accuracy. There hasn't been comprehensive study on the spatial-temporal characteristics of solar irradiance forecast errors. Thus, this paper analyzed the error characteristics of solar irradiance forecast provided by European Centre for Medium-Range Weather Forecasts (ECMWF), one of the most accurate numerical weather prediction (NWP) products. ECMWF's forecasts were compared with ERA5 solar irradiance reanalysis data to reveal the error distribution characteristics. Four-year (2017–2020) data from the geographical region bounded by 63°N, −126°W, 21°S, and 36°E (covering most parts of Europe and North America) were studied. Experiments show that solar irradiance forecast errors peak at noon of the local time. Furthermore, correlations between solar irradiance forecast errors and other weather variables were also revealed. Experiments show that solar irradiance forecast errors have negative correlation with low cloud cover forecast errors and positive correlation with air temperature forecast errors. The possible reasons for the correlation relationship were also analyzed in detail. This paper systematically reveals the spatial-temporal characteristics of solar irradiance forecast errors and provides a useful guideline for solar PV system operation. • Error distribution of ECMWF's solar irradiance forecast in Europe and North America is revealed. • Solar irradiance forecast errors in Europe are smaller than those in North America. • Solar irradiance forecast errors peak at noon of local time. • Solar irradiance forecast errors have slight seasonal and monthly differences. • Correlation between solar irradiance forecast errors and other weather variables is revealed. [ABSTRACT FROM AUTHOR]

Wind speed prediction is a crucial aspect in the utilization of wind energy. In this paper, a wind speed prediction model based on an outlier-robust ensemble deep random vector functional link network (ORedRVFL) and arithmetic optimization algorithm-optimized variational mode decomposition (AOA-VMD) is designed. First, the penalty factor and the number of mode decompositions of VMD are optimized using the AOA algorithm and the original data are decomposed using the optimized VMD. Then the decomposed data is predicted using the ensemble deep random vector functional link network (edRVFL) model. The edRVFL uses rich intermediate features for the final decision, which can make the final result closer to the real data. In order to strengthen the anti-interference ability to the outliers, this paper robustly improves the edRVFL model, and the improved model is called ORedRVFL. ORedRVFL reduces the impact of outliers by introducing regularization and norm to balance the relationship between training error and weights. The experiments have proved that the model proposed in this paper outperforms other models in terms of anti-interference ability and prediction accuracy. • An outlier-robust edRVRFL prediction model is developed. • Use VMD to decompose the data and enhance the model prediction capability. • Fuzzy entropy is used as a criterion for data aggregation. • AOA optimizes the VMD hyperparameters to enhance the decomposition effect of VMD. [ABSTRACT FROM AUTHOR]

With the integration of high proportion of distributed photovoltaic(PV), high-accuracy regional PV power forecasting technology can enhance the regional coordinated scheduling capability of the new power system. This paper proposes a regional PV power forecasting model based on an improved time-series dense encoder and graph attention network (ITDE-GAT), which takes into account the spatio-temporal correlations among the regional PV plants. Firstly, an improved complete ensemble empirical mode decomposition with adaptive noise(ICEEMDAN) is used to extract the clear-sky and fluctuation components from PV data. Secondly, the combined ITDE-GAT is applied to perform the regional PV power forecasting. Considering the static information, an improved dense encoder network (ITDE) is constructed to extract the temporal and spatial relationships of regional PV. Graph attention network (GAT) is then utilized to explore the spatial correlations among the regional PV. Finally, case study from two actual PV region datasets shows that the proposed model achieves higher forecasting accuracy and exhibits stronger generalization capabilities. The results demonstrate that compared to various advanced deep learning methods, the R2 evaluation metric of the approach proposed in this paper demonstrates, respectively, maximum improvements of 3.4 %, 6.5 %, and 7.8 % for the 1 h, 3 h, and 6 h ahead predictions. ICEEMDAN decomposition results can be interpreted efficiently. Graph attention networks can explore the spatial characteristics between PV plants. The ITDE model based on MLP can effectively improve the accuracy of PV multi-step prediction. The ITDE-GAT model has higher accuracy and stronger generalization ability. [ABSTRACT FROM AUTHOR]

The smooth operation of the power system imposes high requirements on wind power forecasting. Given the forecasting challenges brought by the coexistence of strong and weak disturbances in wind power data, this paper proposes a hierarchical anti-disturbance mechanism to raise the accuracy of single-point wind power prediction. In the first-level disturbance processing, we prioritize the treatment of strong disturbances in wind power data through a dual-branch multi-source fusion prediction structure. The wind power subsequences with distinct time–frequency characteristics are divided into two sets, processed by the stationary set prediction branch and the drastic set prediction branch, respectively. For the drastic set, we incorporate microenvironmental multi-source data surrounding the wind turbine as input and propose a parallel-structured attention fusion module CA (Convolution attention fusion block) to provide effective data support for the prediction module, then extract rich fusion features through deep temporal LSTM algorithm and ultimately achieve fine-grained forecasting. For weak disturbances that have not been adequately addressed in the previous step, we conduct a secondary treatment through error correction strategy. By employing the hierarchical processing approach, this paper accomplishes a comprehensive handling of disturbances with different intensities. Experimental results on four data sets show that the proposed model DEWFM (Dual-branch and error correction wind power forecast) demonstrates remarkable performance advantages in comparing with the baseline model and other advanced models, the average values of MAE reaching 1.679, revealing the effectiveness of the proposed method in enhancing the precision of short-term wind power prediction. • Present a method to address the uncertainty in short-term wind power prediction. • A time–frequency hybrid prediction, differentially processing subsequences. • Enhanced LSTM with a multi-source data fusion mechanism for fine-grained prediction. • Error correction handles residual weak disturbances, enhancing prediction accuracy. • The performance evaluation shows the superiority of the proposed model. [ABSTRACT FROM AUTHOR]

Currently, wind power prediction has so many problems in the ultra-short-term time scale (0–4h), which is difficult to improve the deterministic prediction and probability prediction accuracy of the wind farm cluster because it can not fully explore the spatio-temporal correlation between the physical change process and the wind farm. In this paper, a method of ultra-short-term deterministic and probability prediction of wind farm cluster power based on Graph Convolutional Network (GCN) considering fluctuation correlation (FC) is proposed. Firstly, the adjacency matrix is constructed based on the power sequence fluctuation information of each wind farm, and the GCN is designed. Secondly, the spatio-temporal features of power and Numerical Weather Prediction (NWP) wind speed are extracted and fused based on the network model of dual-channel and dual-adjacency matrix. Thirdly, in order to effectively improve the prediction accuracy, a trend switching mechanism is designed based on the effective trend of NWP. When the fluctuation information of NWP is not accurate, the graph structure is constructed by the adjacency matrix based on geographical location to achieve effective prediction. Finally, the method proposed in this paper is applied to wind farm clusters in three provinces of China, compared with some commonly used methods, the average RMSE is reduced by 1.34 %, 1.62 %, 2.07 %, respectively, and the average CWC is reduced by 6.12 %, 4.49 %, 6.62 %, which verifies the effectiveness of this method. • A method for calculating FCC is proposed. • Combined with FCC, the graph structure of wind farm cluster is constructed. • A two-module GCN is designed. • A trend-switching mechanism is designed. [ABSTRACT FROM AUTHOR]

The randomness and uncertainty of charging behavior of EV users bring great challenges to the short-term prediction of charging load of EV charging stations. Aiming at the low precision of short-term load forecasting of EV charging station under the influence of many factors, this paper takes the optimization of EV charging loss as the objective of optimization, firstly, the time series attributes are combined with the nonlinear attributes, and PLSR and LightGBM are used to extract the attributes from EV charging load series, and the fusion prediction model is established. Secondly, the Bayesian optimization method is introduced to optimize hyper-parameters of the model, and the self-adaptive adjustment of the hyper-parameters are realized, which improves performance of the algorithm and the prediction precision. Finally, taking the actual load data of EV charging pile in a certain area as an example, the experimental analysis is carried out, the second-order cone optimal dispatching scheme is adopted to optimize the EV charging schedule. The results show that the EV prediction model constructed by this paper has higher prediction accuracy than other artificial intelligence prediction algorithms, and provides more accurate data for the second-order cone optimal scheduling scheme, the utility model effectively realizes the safe and stable operation of the power grid. • The accuracy of EV load forecasting is improved by using fusion algorithm. • The generalization ability of prediction model is improved by Bayesian optimization algorithm. • It provides an accurate data base for EV optimal scheduling. [ABSTRACT FROM AUTHOR]

The proper integration of wind power-driven grids relies heavily on a reliable balance between electricity production and demand. Therefore, accurate prediction is essential for planning and efficient operation of wind power systems to ensure their continuous supply. However, increasingly severe power quality disturbance (PQD) constantly disturbs this equilibrium, which affects the accuracy of wind power prediction to a large extent. For this purpose, this paper developed a novel optimization methodology to improve wind power prediction accuracy considering micro PQD dimension reduction and elimination for wind-storage integrated systems. A novel idea has been presented in this optimization methodology to eliminate the barrier of PQD and wind power prediction. In the micro aspect, a PQD dimension reduction and elimination strategy based on dynamic mode decomposition (DMD) and Wiener Filter (WF) is proposed to eliminate the autonomy of PQD. This elimination of autonomy allows the WF to get a higher signal-to-noise ratio (SNR). In the macro aspect, this paper takes PQD of different complexity into full consideration, and compares their effects on improving wind power prediction accuracy based on deep learning-based approaches. Through the experimental verification, it is confirmed that the proposed DMD-WF optimization method has demonstrated an effective dimension reduction and elimination of PQD. Moreover, it is found that the proposed PQD optimization method contributes to improve the deep learning-based prediction accuracy when PQD is more complex. The proposed methodology creates a novel perspective to improve the short-term wind power prediction accuracy, which provides a theoretical and methodological guidance for future development of large-scale integrated wind-storage systems. • Development of a novel methodology of PQD optimization and wind power prediction. • A Wiener Filter and dimension reduction-based PQD optimization method is proposed. • WOA-Attention-LSTM is used to analyze forecast accuracy based on PQD optimization. [ABSTRACT FROM AUTHOR]

When it comes to energy sources used in electricity production, the future forecasting of electricity production from renewable energy sources is highly important for both the success of technological advancements in the renewable energy field and energy security. To forecast electricity production from renewable energy sources reliably, it is necessary to accurately model the components of the relevant series. The central argument of this paper is that the various components derived from electricity production data, particularly the residual component, retain valuable predictive information despite their intricate and nonlinear nature. While linear modelling may be highly accurate initially, repeating residuals within linear structures is a discrepancy in terms of data type and methodology. In this paper, different types of hybrid models that combine a decomposition method and both machine learning and statistical approaches are suggested for forecasting electricity production from different energy sources. • Every individual component of a time series retains inherent predictive information. • To improve forecast accuracy, a two-tier approach is presented. • The multivariate models use inputs that can offer crucial information for accurate forecasts. • The use of a hybrid model leads to a significant improvement in forecasting accuracy. [ABSTRACT FROM AUTHOR]

Linear range extender (LRE) system is directly coupled with free piston engine and linear motor and has great application potential in extended-range electric vehicles. In this paper, an analytical and numerical combined method was adopted to investigate the effect of key design parameters on the design and performance prediction of an LRE system. The results showed that the energy input and consumption reach a balance state, the LRE system maintains a stable steady vibration. Increasing the system load will reduce the piston operating stroke. The peak operating velocity of the piston, the average operating velocity of the piston gradually decrease, the operating frequency and the output power decreases as the load increases. When S/B is 0.8, the system output power reaches 4.5 kW, and when S/B is 1.2, the output power is only 2.2 kW. When the mover mass is 3.0 kg, the system output power reaches 3.8 kW, and when the mover mass is 8.0 kg, the output power is only 2.4 kW. When the excess air ratio increases from 0.8 to 1.2, the peak piston speed was reduced from 6.79 m/s to 6.03 m/s, the average piston operating speed was reduced from 5.30 m/s to 4.96 m/s, and the system operating frequency was reduced from 51.2 Hz to 47.96 Hz. If the LRE system wants to have stronger resistance to load fluctuations, it should choose a larger S/B ratio and a relatively larger mover mass during design. In order to pursue higher power output, the S/B and mover mass must be relatively small. • An analytical and numerical combined method was adopted. • The qualitative analysis and oscillation stability analysis was conducted. • The LRE System vibration characteristics and instability characteristics were explored. • A basis reference for the prototype design and performance prediction. [ABSTRACT FROM AUTHOR]

Accurate wind power generation forecasting is of great significance to improve the operation of power system. Probabilistic forecasting has a higher application value in power grid because it can provide more abundant forecasting information than deterministic forecasting. In addition, multi-step forecasting can provide forecasting results in a longer time range, so that decision makers can make longer-term planning and strategic arrangements. In this paper, we propose a novel multi-step improved temporal convolutional network based on quadratic spline quantile function (MITCN-QSQF) for probabilistic wind power forecasting. First, we combine maximum information coefficient, Gaussian similarity and adaptive resample to propose an effective similar power generation feature extraction method (MGR) for power generation. Then the temporal convolutional network is improved to construct the multi-step time series forecasting model MITCN. By combining the proposed model and the powerful probabilistic forecasting method quadratic spline quantile function (QSQF), high-quality probabilistic forecasting of wind power is achieved. Through comprehensive simulations on an open-source dataset, the superiority and efficiency of the proposed method are verified. Compared with some advanced benchmarks, the proposed model can obtain more accurate deterministic and probabilistic forecasting results. • A similar power generation feature extraction method is proposed. • The novel Seq2Seq model MITCN is proposed for multi-step forecasting. • Combining MITCN with QSQF enables high-quality probabilistic forecasting. [ABSTRACT FROM AUTHOR]

The determination of the pressure-bearing performance of hydrogen storage tanks (HST) is integral to enhancing their operational safety and risk management capabilities. In this study, an inaugural development of low-cost critical failure pressure (BP c) prediction model for small samples was proposed, which was based on a combination of the fuzzy grey relational analysis (FGRA) and GM(1,N). Specifically, a total of 15 of BP c -related factors were proposed for analysis and projections based on the data from bonfire tests. Results from the FGRA indicated burst pressure was most closely linked to the initial filling pressure (R ij1 = 0.939) of tanks under fire scenario. The normal working pressure (R ij2 = 0.924) and the elasticity modulus (R ij3 = 0.871) also demonstrating significant impacts. Furthermore, three GM(1,N) prediction models which could estimate BP c with multi-factor coupling were developed. Increasing the number of input feature factors could enhance the predictive capability of GM model. The GM(1,16) had optimal prediction performance among the models, achieving a prediction accuracy of 99.8 %. As well, the mean absolute error (MAE) was 0.799 MPa while the mean absolute percentage error (MAPE) was 1.56 %. This paper offered a novel option for establishing the HST critical failure criterion safely and efficiently. • Obtained factors on design features, physicochemical, and external thermal load. • Clarified the main controlling factors which affected critical failure pressure. • Developed GM(1,N)-based models to predict critical failure pressure. • Suggestions on the safety of hydrogen storage tanks in service were proposed. [ABSTRACT FROM AUTHOR]

This paper investigates the relationship between unemployment and inflation in 29 African economies by re-examining the Phillips Curve hypothesis and evaluating the role of global oil prices in predicting inflation. Existing studies often use uniform-frequency variables, which can lead to biased results by omitting information from high-frequency series. We employ a mixed data sampling regression model to analyze inflation in selected African countries by regressing monthly inflation rates on annual unemployment rates and daily oil price returns. Our findings show that the Phillips Curve hypothesis holds in a few countries across Central Africa (Cameroon, the Central African Republic, Chad, and Republic of the Congo), East Africa (Rwanda), North Africa (Algeria), Southern Africa (Botswana and Malawi), and West Africa (Burkina Faso and Mauritania). In the other countries, the hypothesis does not hold, potentially indicating a predominance of cost-push inflation. Our analysis demonstrates that global oil prices impact inflation differently across these countries. Additionally, daily oil price data significantly enhance inflation forecasting compared to monthly data. Incorporating additional economic variables, such as exchange rates and interest rates, further improves forecast accuracy. We recommend targeted monetary policies to mitigate the impacts of oil price fluctuations and unemployment on inflation. • Reassesses traditional demand-side Phillips curve in Africa. • Integrates supply side, notably oil price, into Phillips curve. • MIDAS method exceeds uniform frequency in curve modeling. • Oil price inclusion improves curve's predictive capacity. • Results maintain robustness in all in/out-sample analyses. [ABSTRACT FROM AUTHOR]

In Cameroon, predicting CO 2 emissions from road transports is crucial since it contributes in formulating the government's energy plans and strategies. In this perspective, this paper proposes an optimized wavelet transform hausdorff multivariate grey model (OWTHGM(1,N)) that addresses some of the weaknesses of the classical GM(1,N) model such as inaccurate prediction and poor stability. In all, three improvements have been made to the classical GM(1,N): First, all inputs are filtered using the wavelet transform, thereby denoising variables that could hamper modelling; second, a new time response function is established using the Hausdorff derivative; and finally, the use of Rao's algorithm to optimise the model's parameters as well as the ξ -order accumulated value of the observation data described by the Hausdorff derivative. In order to demonstrate the effectiveness of OWTHGM(1,N), it is applied to predict CO 2 emissions from road transport. The new model produces predictions with 1.27 % MAPE and 79.983 RMSE, and is therefore more accurate than competing models. OWTHGM(1,N) could therefore serve a reliable forecasting tool and used to monitor the evolution of CO 2 emissions in Cameroon. The forecast results also serve as a sound foundation for the formulation of energy consumption strategies and environmental policies. • The OWTHGM(1,N) model enhances CO 2 predictions in Cameroon's road transport sector. • The WT filters inputs, improving accuracy by denoising variables in the model. • Hausdorff derivative captures non-linear trends in emissions data. • Rao's algorithm optimizes model parameters for reliable forecasting. • OWTHGM(1,N) provides reliable forecasts with a MAPE of 1.27 % and RMSE of 79.983 [ABSTRACT FROM AUTHOR]

Construction prediction is the key for the shape control of energy storage salt caverns, which benefits with the storage capacity and long-term operational safety. However, the conventional grid discretization methods using elastic grid could not accurately tracking the three-dimensional boundary movements of salt cavern. This paper introduced a novel construction prediction model of salt cavern using a Volume-of-fluid based method. The tracking of the salt caverns' dissolving boundary is successfully reconstructed and tracked by the fluid volume fractions (0-1) in structural grids. The model was validated by indoor experiment and field data. In the simulation of indoor experiment, the lateral expansion, which was difficult to reproduce in previous models, is successfully simulated. The volume of the simulation chamber is 392.8 ml with an error of 1.9 %. During the simulation of Volgograd horizontal cavern construction, the simulated cavern shape is close to the sonar detection, with an average error in radius about 3.6 %. And the brine-discharge concentration is consistent with the site monitoring, with an average error about 4.5 %. These results validate the capability of the proposed method in three-dimensional dynamic boundary tracking and shape prediction of salt cavern. • A salt cavern construction prediction model is introduced using a Volume-of-Fluid method. • Three dimentional simulation of the movement of dissolving salt cavern interfaces is achieved for the first time. • A demo C++ program has been developed for numerical implementation of the proposed model. • model's reliability is verified through simulations of an indoor experiment and a field project. [ABSTRACT FROM AUTHOR]

The uncertainties of renewable energies and loads in integrated energy system (IES) result in unstable operation and performance degradation. To consider the impact of the uncertainties, the scheduling strategies of IES with the prediction of uncertainties are required. This paper designs a hybrid solar and gas turbine IES, in which the thermal and electrical storage units are integrated to address the uncertainties. A robust real-time energy scheduling strategy is proposed according to the real-time load and the predicted loads in the future time, in which the errors of multi-step interval prediction of renewable sources and loads are combined. The multi-step prediction method based on gated recurrent unit and time classification is constructed for the interval prediction of uncertain sources and loads by analyzing the probability statistics of prediction errors. In the robust optimization, the penalty factors of time sequences are proposed and considered to decrease the influences of future predicted information on current real-time scheduling, in which the future uncertainties in larger interval with current time has less influence. The optimal results in the hybrid IES demonstrate that the daily operation cost is 13.67 % lower than the method that does not consider the prediction parameters. Compared to the traditional scheduling strategies, the proposed strategy declines the operation cost by 2.93 %. The analysis of the confidence level of predictions ranging from 60 % to 98 % illustrates that the operation cost under the penalty factor of 0.94 is the lowest when the confidence level is from 94 % to 96 %. • A robust scheduling strategy with multi-step interval prediction is proposed. • A multi-step interval prediction method is designed to show the uncertainties. • The impacts of time-sequences are integrated into optimization objective. • The confidence level of interval prediction is discussed on operational costs. • The proposed strategy declines the operation cost of IES by 2.93 %. [ABSTRACT FROM AUTHOR]

Accurate wind power forecasting is essential for (i) the management of wind energy, (ii) increasing the integration of generated power into the electrical grid, and (iii) enhancing maintenance efficiency. This paper proposes a novel hybrid model for short-term (1-hour ahead) wind power forecasting. The model integrates the echo-state network (ESN) architecture with empirical mode decomposition (EMD) to improve forecast accuracy. Unlike existing approaches that use separate models for each decomposed signal after EMD method, proposed model uses a single ESN structure to predict wind power by incorporating all decomposed signals and their past values. The main advantage of this architecture is that it eliminates the need to train multiple models, thereby streamlining the forecasting process. To evaluate the performance of the proposed model, one year of data from the West of Duddon Sands, Barrow, and Horns Power offshore wind farms is utilized. Firstly, the classical ESN model and the EMD-ESN hybrid model are compared for the three datasets. Then, a comprehensive evaluation is performed by comparing the results of the proposed model with commonly used standalone and hybrid forecasting models such as Multi-Layer Perceptron (MLP), Adaptive-Neuro Fuzzy Inference System (ANFIS), Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), EMD-LSTM, EMD-BiLSTM, Variational Mode Decomposition based ESN (VMD-ESN), and Wavelet Decomposition based ESN (WD-ESN). The results show that the EMD-ESN hybrid model outperforms implemented models in predicting wind power in all three datasets. Furthermore, the proposed EMD-ESN model for an onshore wind turbine power data in Germany is compared with the SOTA models, such as Transformer, Informer, Autoformer, and Graph Patch Informer (GPI), in the literature. This study highlights the superior predictive capabilities of proposed model, making it a valuable tool for enhancing the accuracy of wind power forecasts, thereby contributing to the reliable integration of wind energy into the power grid. [Display omitted] • This study applies EMD to enhance wind data and improve accuracy. • EMD aids forecasting by effectively preserving data characteristics. • ESN simplifies wind power forecasting with its unified structure. • The ESN model facilitates forecasts by integrating signals. • Hybrid EMD-ESN model improves onshore and offshore wind power forecasting accuracy. [ABSTRACT FROM AUTHOR]

In the context of prominent energy crisis, photovoltaic power (PV) generation has received increasing attention, then accurate PV generation forecasting is crucial for ensuring the smooth operation of power stations. However, existing research is insufficient in comprehensively analyzing the impact of PV generation volatility. To fill the gaps and enhance the prediction accuracy, this paper proposes a new hybrid forecasting method. We first introduce the Locally Weighted Scatterplot Smoothing (LOWESS) method to process the data and enhance the data stability, and use Pearson correlation coefficient (PCC) and Random Forests (RF) for feature selection to improve the quality of input data. Then we use Attention mechanism and Convolutional Neural Network (CNN) layer to optimize Bi-directional Gate Recurrent Unit (BiGRU) model and form a new hybrid model. Finally, based on the Bagging algorithm, we use ensemble learning to further optimize the hybrid BiGRU model to enhance the depth and performance. The proposed method is validated through case analysis results from two different locations, Xuhui District in Shanghai, China and the DKASC area in Alice Springs, Australia. The results demonstrate that, compared with other models, the developed method exhibits exceptional prediction performance and effectively enhances the accuracy of PV generation forecasting. Keywords : photovoltaic power generation; Locally Weighted Scatterplot Smoothing; feature selection; ensemble learning; Bi-directional Gate Recurrent Unit. • LOWESS smoothing technology is used to reduce the variability of PV generation data. • Pearson correlation coefficient and Random Forest are used to for extract feature. • Attention mechanism and CNN layer are used to optimize BiGRU model. • Two BiGRU layers are stacked to improve the depth and prediction performance. • Ensemble learning is introduced to further optimize the hybrid BiGRU model. [ABSTRACT FROM AUTHOR]

Accurate wind speed forecasting is of great significance for the utilization of Arctic wind energy resources. The traditional single model is difficult to fully depict the nonlinearity of wind speed and its wide range of variations. In this paper, a hybrid model is proposed for multi-step wind speed forecasting, which combined complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), convolutional neural network (CNN) and long-short term memory neural network (LSTM). The wind speed series is firstly decomposed into several intrinsic mode functions (IMF) by CEEMDAN to provide more stable data. Secondly, four-step-ahead forecasts are realized using well-tuned CNN-LSTM model for each IMF. Finally, the forecasted wind speed is obtained by reconstruction. The effectiveness and feasibility of the proposed method is validated based on thorough evaluation and step-by-step analysis. The RMSE of the proposed model is 0.4046 m/s, which are reduced by 58 % compared with 7 benchmark models. Furthermore, the average prediction interval of the proposed model is also reduced by 20 %, 16 % and 7 % compared to CEEMDAN-FCNN, CEEMDAN-CNN and CEEMDAN-LSTM respectively. The results prove that all three parts of the proposed model contribute to a better performance in wind speed forecasting. [Display omitted] • CEEMDAN can reduce the nonlinearity of wind speed data and unstable IMFs have a large impact on the wind speed forecasting. • Combining LSTM with CNN can enhance feature extraction and improve the accuracy of multi-step wind speed forecasting. • The proposed model achieves more accurate forecasting for catastrophe points which provides more critical information. [ABSTRACT FROM AUTHOR]

Crude oil plays a vital role in industrial and social development and has become an integral part of the economic development. However, influenced by policies, wars, etc., it is hard to capture the trend of complex and volatile crude oil price if only one model is used, which often leads to poor forecasts. To enhance the prediction accuracy and robustness of forecasting models, a novel combined forecasting method with time-varying weights, i.e., Jaynes weight hybrid (JWH) model incorporating the Shannon information entropy and several forecasting methods is proposed in this paper. In the selection of the baseline models, the autoregressive integrated moving average in classical statistical forecasting strategies, back propagation neural network, extreme learning machine in neural network and long short-term memory neural network in deep learning models are chosen to fit the crude oil price. Four datasets and three experiments are constructed to verify the prediction ability of the novel combined forecasting model. Empirical results are calculated by five measurement criteria, suggesting that the prediction accuracy of the novel combined method is significantly higher than several comparison models and the mean absolute percentage error of the model has arrived 2.81 % in detail. Particularly, the proposed model has achieved satisfactory performances in Covid-19 and the War in Ukraine, further verifying the robustness of the combined methodology. • A time-varying weight combined forecasting model is proposed for crude oil price. • The proposed model outperforms the comparison models. • The proposed model shows good robustness during special period: COVID-19. • Hypothesis testing is used to test the validity of the proposed combination model. [ABSTRACT FROM AUTHOR]

The rapid adoption of electric vehicles (EVs) has led to dramatic increase in charging demands that poses great challenges for efficient charging infrastructure rollout and operation. It is crucial to accurately assess charging demand in urban areas to optimize the siting and sizing of charging infrastructure. This paper proposes a novel urban charging load forecasting model for private passenger EVs based on massive operating data of EVs in Beijing. First, the characteristics of travel patterns for private passenger EVs, urban road network, functional area distribution and charging infrastructure distribution within the entire Beijing area are identified. Then a charging load forecasting model that can simultaneously simulate trip chains for EVs is constructed by considering the occupancy states of public charging piles and the interactions among different EVs. Finally, the effectiveness of the proposed charging load forecasting model is verified based on comprehensive test data. Our findings imply that the number of EVs at recharge and the charging power can be reliably predicted with the accuracy of over 84.73 % and 81.92 %, respectively. It provides the foundation for optimal charging infrastructure planning and charging scheduling. [ABSTRACT FROM AUTHOR]

Accurate prediction of energy prices is crucial to the development of energy security and environmental policies in various countries. This paper proposes a novel multi-step prediction hybrid model with genetic algorithm for variational mode decomposition, improved complete ensemble empirical modal decomposition with adaptive noise, bidirectional gated recurrent unit, temporal convolutional network, and multi-layer perceptron (GVMD-ICEEMDAN-BIGRU-TCN-MLP) for predicting carbon and natural gas futures prices. First genetic algorithm (GA) is used to fix the parameters of VMD model, the carbon and natural gas prices are decomposed into subsequences. Then the difference between the original series and the VMD after decomposition is further decomposed into subseries using ICEEMDAN. Next, the highest frequency series is predicted using the MLP model, and other subsequences are predicted using the BIGRU-TCN model. Finally, each predicted value is added linearly to determine the final result of steps 1, 3, and 5 of the entire forecasting process. According to the experimental results, it is shown that the model has lower prediction errors than the comparison model under mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE), coefficient of determination (R2), and modified Diebold-Mariano test (MDM). The good prediction results of the novel hybrid model are demonstrated in multi-step ahead integrated prediction experiments, especially in the experiments with 1-step ahead prediction, as well as in the experiments with varying training ratios. • GVMD-ICEEMDAN-BIGRU-TCN-MLP model is used to forecast energy prices. • ICEEMDAN is used to further decompose the residuals. • Hierarchical prediction based on subsequence features is better. • Multi-step forecasting integrated study of energy prices. [ABSTRACT FROM AUTHOR]

Current load forecasting methods struggle with the randomness and complexity of multiple loads in regional integrated energy systems, resulting in less accurate predictions. To address this problem, this paper proposes a novel two-phase decomposition hybrid forecasting model that combines complementary ensemble empirical mode decomposition, sample entropy, variational mode decomposition, and genetic algorithm-bidirectional long short term memory. The proposed approach begins by decomposing the load sequence into multiple intrinsic mode function components at different frequencies using complementary ensemble empirical mode decomposition. Afterward, the sample entropy values are calculated for each intrinsic mode function. The intrinsic mode function with the highest sample entropy value is subjected to a two-stage decomposition using the variational mode decomposition method. Through this technique, a series of stationary components is acquired. Subsequently, the bidirectional long short term memory model is optimized using the genetic algorithm, and the genetic algorithm-bidirectional long short term memory approach is employed to predict all the decomposed components. Finally, the prediction results are combined and reconstructed to obtain the final prediction value. Experimental results demonstrate the effective handling of non-stationary load sequences by the forecasting model, showcasing the highest level of accuracy in regional integrated energy system multiple loads forecasting. • CEEMD technology reduces the complexity of data samples. • The combine prediction model effectively improves the prediction accuracy. • Use grey relation analysis technique to determine the input of prediction model. • The improve sparrow optimization algorithm is utilized for parameter optimization. [ABSTRACT FROM AUTHOR]

The integration of smart grids with Advanced Metering Infrastructure (AMI) has bridged the realms of the Internet of Vehicles (IoV) and Electric Vehicles (EVs), yet challenges persist in managing EV Battery Range and charging infrastructure effectively. This paper presents an innovative IoV-based framework tailored for EVs, with a specific focus on forecasting electricity consumption in a Vehicle-to-Grid (V2G) scenario. By exploring the hurdles surrounding electric vehicle usage, the research lays the foundation for Electric Vehicles (EVs). The proposed IoV model optimizes IoT sensor utilization through a fog layer and employs a Back Propagation (BP) neural network for battery State of Charge (SoC) estimation, integrating Principal Component Analysis (PCA) for data dimensionality reduction. Leveraging substantial computing capabilities, the cloud layer predicts Electricity Consumption Data (ECD) associated with EVs in V2G scenarios. Performance evaluation metrics like Akaike Information Criteria (AIC), Mean Error (ME), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Squared Error (RMSE) are assessed across state-of-the-art forecasting algorithms. Incorporating EV potential into the optimal model reveals a significant 10% reduction in electricity demand. This research advances IoV-based frameworks, offering insights to enhance EV efficiency within the broader energy infrastructure. • An innovative Internet of Vehicles (IoV) framework for depicting electricity demand on the demand side in V2G scenarios. • In-depth experimentation and evaluation of the state-of-the-art approaches of clustering and forecasting. • The devised model successfully projected a 10% decrease in demand side for consumers. [ABSTRACT FROM AUTHOR]

In response to the high uncertainty associated with residential net load due to the coupling of distributed photovoltaic generation and user demand, this paper proposed a novel cluster-based stacking ensemble learning model for net load interval prediction. Firstly, the k-means algorithm is employed to discover the similarity in user electricity consumption patterns. Then, a RIME optimization algorithm with local enhancement (LRIME) is developed to optimize the parameters and weights of the base learners in stacking ensemble learning. Subsequently, base learners with strong predictive capabilities and significant diversity are chosen as the first-layer predictive models, extreme learning machine (ELM) is utilized as the second-layer predictive model, ultimately resulting in the proposed stacking ensemble learning prediction model. And utilizing the bootstrap method to fit the volatility of point predictions, different prediction intervals are obtained at varying confidence levels, aiming to quantify the integrated uncertainty in photovoltaic generation and load. Testing on the open Ausgrid electricity load data in Australia provided robust validation of the proposed method's effectiveness. In comparison with other outstanding prediction models, the proposed ensemble model can effectively capture the uncertainty in integrating photovoltaic generation and user load. • A RIME optimization algorithm with local enhancement (LRIME) is developed. • A stacking ensemble model is proposed for residential net load interval prediction. • The proposed model demonstrates higher forecasting accuracy and reliability. [ABSTRACT FROM AUTHOR]

Existing regression, tree-based and NN models either lacks probabilistic prediction, takes longer training time, have high computational requirements or sacrifice accuracy. This paper introduces a novel framework, (MAFS+ISTD+PGBM), specifically to overcome these limitations. First three challenges are addressed by integrating gradient boosting and quantile regression model. The key idea is to combine speed and scalability of gradient boosting with probabilistic capabilities of quantile regression, forming PGBM. However, the issue of mediocre accuracy still remained. To address this, two pre-processing techniques are introduced. MAFS utilizes statistical methods and knowledge-based analysis to identify the most relevant features, while ISTD extracts and eliminates trend and seasonality components, ensuring stationarity. After rigorous evaluations, (MAFS+ISTD+PGBM) emerges as the superior performer surpassing all existing models in terms of training time and accuracy with highest R2 score of 0.997 and low values across all error metrics. The proposed model took less than one-third of training time (∼ 15 min) compared to CNN-LSTM+attn., (∼ 48 min), the only model with comparable accuracy of proposed model. Thus, proposed approach shall be used to empower grid operators with highly accurate and cost-effective probabilistic forecasts which allows them to make informed decisions about system stability and optimize resource utilization, ensuring reliability and efficiency. • PGBM : Enabled existing gradient boosting model for probabilistic forecast. • MAFS : Improved accuracy of PGBM by identifying relevant input features. • ISTD : Improved accuracy of PGBM by extracting trend & multiple seasonalities. [ABSTRACT FROM AUTHOR]

To reflect the trend of electricity consumption (EC) in China's Pearl River Delta (PRD) under a balanced environment and economy, this paper proposes an EC forecasting framework from the perspective of optimal economic growth. The historical GDP statistics of the PRD are calibrated with data from the new "NPP-VIIRS-like" nighttime light dataset from 2000 to 2020, and two simulation scenarios of optimal economic growth are constructed according to the carbon emission reduction rate. Finally, based on the electricity data from 2000 to 2017, an error correction model is constructed to predict and compare the trend of EC in the PRD under different carbon emission scenarios. The results demonstrate the following. (1) The economic growth path under low-carbon constraints is more closely aligned with the actual economic development of the PRD. (2) The electricity demand required to sustain optimal economic growth in the PRD under low-carbon constraints is projected to reach saturation around 2037, approximately a decade earlier than the scenario without carbon constraints. The results of the projections are expected to guide future work on power system planning and economic development assessment in the context of carbon reduction. • The balance between environmental resources and economic development is considered. • The optimal economic growth path under the low-carbon goal is measured. • Calibration of regional statistics with the nighttime light dataset "NPP-VIIRS-like". • The error correction model is built as a short-term regulation mechanism for power prediction. • Predicted electricity demand for optimal economic growth in the PRD under different scenarios. [ABSTRACT FROM AUTHOR]

Accurate forecasting of significant wave heights (SWH) plays an important role in the generation and conversion of ocean wave energy, being a promising clean energy source, while a challenging task since the SWH is produced by a complex natural process. Additionally, some deep forecasting models integrate parallel and recursive structures via unexplained ways to present intelligent fitting, leading to insufficient inheritance of hybrid advantages. Since biological multisensory systems automatically cope with complex time series, this paper explores a b io-multisensory-inspired g ate-attention c oordination network (BGC-Net) for forecasting SWH. Specifically, inspired by the multisensory cognitive process, the BGC-Net includes sensory capture, brain memory, and forecasting components, and their collaboration offers a clear multi-level handling process. In internal implementations, through mimicking that multiple senses independently capture external stimuli, the first component adopts several activation functions for suppressing data fluctuations and detecting the complex nonlinear and linear features. Through mimicking that the left and right hemispheres interact and collaborate electrical impulses, the second component integrates both gate-driven recursive and self-attention-driven parallel structures to extract multivariate and temporal dependencies at the feature level. Through mimicking that a higher cerebral cortex fuses extracted information for the final cognition, the third component proposes three feature fusion ways of concatenate, attention, and gated fusion, and takes the best one to generate SWH forecasts. In summary, the BGC-Net not only has interpretability at the biological level but also effectively integrates the advantages of parallel and recursive structures. Five experiments and six analyses all indicate that the BGC-Net exceeds 17 baselines, has root mean squared error (RMSE) average improvements with 22.30% and 19.22% for two datasets, and provides references to support SWH forecasting. • A deep bionic model is explored to forecast significant wave heights. • The multi-level handling is inspired by the multisensory cognition process. • Multiple activation functions enhance the intelligent feature learning. • The advantages of recursive and parallel structures are integrated. • Five experiments and six analyses show its high performance. [ABSTRACT FROM AUTHOR]

Accurate energy consumption prediction is pivotal in supporting efficient grid scheduling and enhancing renewable energy penetration. This paper proposes a novel hybrid prediction model called CEEMDAN-SE-EC-BiLSTM. The proposed model utilizes Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to decompose the original energy consumption data into Intrinsic Mode Functions (IMFs). Afterwards, Sample Entropy (SE) is employed to evaluate the complexity of the IMFs. The complexed IMFs are then transformed into simpler IMFs using a second decomposition. All IMFs are predicted using Bidirectional Long Short-Term Memory (BiLSTM) to attain the initial predicted sequence. To further refine the prediction, an Error Correction (EC) technique is introduced. This involves subtracting the initial predicted sequence from the original sequence to obtain the error sequence, which is then decomposed using CEEMDAN. Subsequently, the decomposed error sequence is predicted using BiLSTM to generate the error prediction sequence. Finally, the predicted results and the error prediction are combined to obtain the final energy forecast sequence. To verify the superior performance of the proposed method, the model is tested on five selected buildings and benchmarked against seven prediction models. The experimental results indicate that the proposed hybrid prediction model achieves the highest accuracy in building energy prediction. • A novel hybrid approach for building energy prediction is proposed. • The SE is used to evaluate the necessity for secondary decomposition. • The EC technique is employed to improve the prediction accuracy. • The performance of the proposed model is benchmarked against other models. [ABSTRACT FROM AUTHOR]

With the continuous global increase in installed wind power capacity and subsequent surge in power generation, the contradiction between the safe operation of the grid and the efficient consumption of new energy after large-scale grid connection has become increasingly prominent. This paper presents a hybrid model prediction method to further improve the stability and accuracy of wind power prediction. Firstly, variational modal decomposition (VMD) optimized by the coati optimization algorithm (COA) is employed to decompose original wind power, mitigating the non-stationary characteristics of the power sequence. Subsequently, permutation entropy (PE) is used to recombine the decomposed components, and the combined feature selection method is achieved by integrating the Spearman correlation coefficient (SCC) and the autocorrelation function (ACF). Then, the multivariate combined model is constructed, and the improved multi-objective coati optimization algorithm (IMOCOA) determines the weight coefficients of each model to enhance the performance of the hybrid model. Finally, research and analysis are conducted from multiple scenarios and time scales using historical operating data from a wind farm in Xinjiang. The experimental results show that the proposed prediction model effectively improves the accuracy and stability of the wind power prediction compared with other popular prediction models. • A new hybrid model is proposed to implement wind power prediction. • An improved multi-objective coati optimization algorithm is introduced. • A combined feature selection method is presented. • A new weight coefficients determination method is proposed. [ABSTRACT FROM AUTHOR]

In accordance with the emission regulations specified in GB17691-2018 ″Emission Limits and Measurement Methods for Heavy-Duty Vehicles (China VI)," this paper conducted actual on-road emission testing of a specific N 2 heavy-duty diesel vehicle using a vehicle-mounted portable emission measurement system (PEMS). Grey relational analysis and principal component analysis were employed to select input parameters and reduce the dimensionality of input parameters, thereby enhancing the quality of the input parameters. The main framework of the model utilized a Gated Recurrent Unit (GRU) neural network characterized by a small scale, high accuracy, and good generalization capability. Innovatively, a combination optimization algorithm incorporating Bayesian algorithm and k-fold cross-validation was used to optimize hyperparameters, improving the predictive accuracy of the model. The training process employed a learning rate adjustment strategy that combined cosine annealing and exponential decay, thereby improving the search performance during the neural network training process. The resulting BO-GRU network exhibited a root mean square error (RMSE) of 1.2352 (mg s−1) and a coefficient of determination (R2) of 0.9569 on the test set, indicating good predictive accuracy with a relatively small number of network parameters. This research provides a potential approach for the online precise monitoring of NOx emissions from heavy-duty diesel vehicles. • GRA and PCA minimized noise in input models, boosting parameter quality and model predictivity. • Bayesian optimization with K-fold cross-validation for hyperparameter tuning enhanced accuracy and training speed. • Cosine annealing and exponential decay improved training parameter search, enabling optimal parameter identification. [ABSTRACT FROM AUTHOR]

Natural gas, as a clean and low-carbon energy resource, assumes a vital role in facilitating the transformation of the Chinese energy structure. Effectively forecasting its consumption holds great practical implications for the high-quality social development of China. Therefore, a hybrid model for forecasting natural gas consumption (NGC) in China is developed in this paper. Firstly, the Logarithmic Mean Divisia Index (LMDI) method is adopted to decompose the influencing factors of the NGC in China from 1994 to 2020, which revealed that the energy structure effect and the economic development effect have a positive promotion on NGC, while the energy intensity effect manifests a significant inhibition. Subsequently, based on the contribution rate of each factor, the particle swarm optimization (PSO) algorithm to optimize the long and short-term memory neural network (LSTM) model is constructed for NGC forecasting. Compared to other benchmark models, the PSO-LSTM model demonstrated a significant improvement in predictive accuracy, showcasing its valuable application in NGC prediction. Finally, the PSO-LSTM model is employed to analyze the scenario prediction of NGC development from 2021 to 2035. The forecast result indicated that the future NGC in China will show a yearly growing trend, which may lead to a serious imbalance between the supply and demand of natural gas, so the Chinese government should fully consider the energy security issue when formulating relevant policies. • The LMDI method is applied to decompose the factors of natural gas consumption. • The energy consumption structure and economic development positively promote NGC. • The energy intensity effect negatively inhibits the development of NGC. • Scenario analysis of NGC is performed by constructing a PSO-LSTM model. [ABSTRACT FROM AUTHOR]

Accurate total wind power forecasting is essential to wind farms. Conventional methods predict each wind turbine separately, which fails to exploit the spatial correlation between turbines and causes error accumulation through summation. Additionally, numerical weather prediction (NWP) with limited spatial–temporal resolutions is difficult to assist in power forecasting effectively, resulting in decreased performance as the forecasting steps increase. To address these drawbacks, a novel "N encoder-1 decoder" multi-channel fusion group forecasting sequence-to-sequence (seq2seq) architecture with wind resource quality assisted spatial attention (WRQASA) and NWP correction (NWPC) is proposed in this paper. The encoders in the architecture extract the temporal features of each turbine, followed by WRQASA adaptively adjusting their impact on the decoder. NWPC constructs micro-scale meteorological data based on its large-scale counterpart. The decoder takes the output from WRQASA and NWPC, and predicts the total power of the turbine group. The model is evaluated on a real-world wind farm. Ablation studies revealed the effectiveness of the novel architecture, WRQASA and NWPC. In 12, 24, 36 and 72-h multi-step forecasting scenarios, the new method outperformed conventional methods, such as seq2seq, light gradient boosting machine and support vector regression, achieving a more effective prediction of total power. • A multi-channel seq2seq for wind farm's total power prediction is proposed. • Wind resource quality assists self-adjusting each turbine's impact on final output. • Downscaled weather prediction improves wind power forecasting performance. • The common issue of declining performance as prediction steps increase is tackled. [ABSTRACT FROM AUTHOR]

In electricity price forecast, deep networks have shown promising due to their strong nonlinear modeling capability but corresponding design tricks are not convincible, reflecting the situation that biologically-inspired networks in time-series analysis are still not enough explored. To this end, this paper proposes a m ulti- s cale v isual-inspired network with three explored modules, namely MSV-Net, for electricity price forecasting. In particular, through simulating both the ocular globe and neurons in signal conversion and further enhancement, the multi-scale feature extraction module captures the overall and specific time-scale features via multiple convolutional kernels and especially highlights the core features via self-attention mechanisms. Through imitating the left and right hemispheres in analyzing and producing complementary visual information, the dual-stream collaboration module adopts the convolutional layer and gated recurrent unit to extract and integrate time-dependent relationships in parallel, thus improving the model stability. Through simulating the higher visual region in understanding complex environments, the forecasting module is introduced to forecast the electricity price. The layer-by-layer collaboration of the above three modules guarantees the effectiveness of the MSV-Net. Moreover, four comprehensive evaluation metrics are explored to assess the forecasting performance in peak, low, and normal samples. Experiments and discussions under two electricity price datasets show that the MSV-Net has satisfactory forecasting accuracy and stability, in improving average mean absolute percent error of 31.59 % and 25.15 % for two datasets compared with 12 baselines, and shows extensible in other energy forecasting fields. • Short-term electricity price series involves multi-time-scale features. • A multi-scale visual-inspired network (MSV-Net) is explored. • Intelligent feature extraction ensures the effectiveness of MSV-Net. • Four novel comprehensive evaluation metrics are proposed. • Its high performance is reflected via four experiments and six discussions. [ABSTRACT FROM AUTHOR]

Reliable electricity price forecasts are of great importance to operators and participants in power markets. However, due to mixing effects of various factors, electricity price fluctuations are too complex to extract hidden features for accurately modelling. Additionally, biologically-inspired ideas are promising in significantly improving the performance and rationality of deep forecasting networks. Based on the fact that the auditory system effectively handles complex sound signals with amplitude, frequency and wavelength characteristics, this paper explores an auditory-circuit-motivated deep network with three following modules for forecasting short-term electricity prices. Specifically, the coding module imitates that left and right ears receive the sound and convert it into electric signals in parallel, and codes the input electricity prices into multiple valuable features, improving the feature integrity and model stability. The analysis module imitates that the left and right hemispheres integrate and handle electric signals to produce higher auditory information, and captures nonlinear and short-term dependencies via stacking convolutional and gating operations. The forecasting module imitates that the high-level brain region pays attention to the external environment based on higher auditory information, and generates final forecasts via the attention mechanism. Based on the cooperation of the above three modules, the proposed deep network mimics the flowchart, structures and functions, thereby inheriting superior handling capability of the auditory circuit. Experiment results under two real-world sets show that the proposed deep network is superior to 13 baselines, and improves the mean absolute error by averages of 14.7 % and 20.6 % in normal parts, 12.9 % and 23.4 % in high-fluctuation parts, and 16.1 % and 16.2 % in peak-low parts, which is an effective supplementary model for electricity price forecasting. • An auditory-circuit-motivated deep network is developed. • Forecasting applications are performed for short-term electricity prices. • Flowchart, structures, and functions of the auditory circuit are imitated. • Gating and attention mechanisms improve the forecasting accuracy. • Six experiments show high performance of the bionic deep network. [ABSTRACT FROM AUTHOR]

An accurate and reliable wind power prediction model has important significance for the operation of power systems and large-scale grid connection. This paper proposes a hybrid deep learning model, CEEMDAN-SE-TR-BiGRU-Attention, for high and low frequency wind power prediction by combining complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), sample entropy (SE), Transformer (TR) and bidirectional gated recurrent unit with attention mechanism (BiGRU-Attention). Firstly, the CEEMDAN decomposes the original wind power sequence into multiple sub-modes and a residual, and the sample entropy of each sub-sequence is calculated by restructuring the sequence, which can effectively alleviate the impact of the original non-stationary series on the accuracy and computational complexity. Next, the reconstructed sequences are further divided into high and low frequency sequences according to the sample entropy value of the original sequence. The Transformer and BiGRU-Attention models are respectively applied to the prediction of high frequency and low frequency sequences according to the characteristics of each sequence. Finally, the predicted values of all components are superimposed to obtain the final prediction results. Experiments are carried out on four datasets with different seasons, and different models are compared to illustrate the effectiveness and superiority of the proposed model. The experimental results show that the proposed model achieves better prediction accuracy. • A prediction model is proposed by combining CEEMDAN, SE, TR and BiGRU-Attention. • Transformer and BiGRU-Attention are established for high and low frequency sequences. • Extensive experiments show that the proposed model has better prediction accuracy. [ABSTRACT FROM AUTHOR]

Accurate regional distributed PV power forecasting provides data support for power grid management and optimal operation. Distributed PV has the characteristics of large quantity, small capacity and difficulty in obtaining meteorological data. Statistical upscaling method is commonly used to forecast regional power. However, the current research ignores how to reasonably divide the sub-regions with similar output characteristics and mine the spatial and temporal correlation between different sub-regions. Therefore, this paper proposes a short-term regional distributed PV power forecasting method based on sub-region division considering spatio-temporal correlation. Firstly, the representative power plant is selected after dividing the sub-region by the AP clustering algorithm. Then, the GCN is used to extract spatial correlation features, and the LSTM is used to extract the evolution features of dynamic spatial correlation features, and the power forecasting models of representative plants in different weather types are established. Finally, the data integrity and similarity of the sub-region are scored, and the upscaling weight is determined to realize the power forecasting of the whole region. The distributed PV power generation data of Pingshan County, Hebei Province, China is used for simulation test. The results show that the forecasting method proposed has higher forecasting accuracy than the traditional model. • The spatio-temporal correlation between distributed PV power plants is studied. • Introduce how to divide appropriate distributed PV sub-regions. • A power forecasting method considering spatio-temporal correlation is proposed. • Sub-regional data evaluation improves the forecasting accuracy of regional PV. [ABSTRACT FROM AUTHOR]