Search

In recent years, support vector regression (SVR) models have been widely applied in short-term electricity load forecasting. A critical challenge when applying the SVR model is to determine the model for optimal hyperparameters, which can be solved using several optimization methods as the grid search algorithm. Another challenge that affects the response time and the precision of the SVR model is the normalization process of input data. In this paper, the grid search algorithm will be suggested based on data normalization methods including Z-score, min-max, max, decimal, sigmoidal, softmax, and then utilized to evaluate both the response time and precision. To verify the proposed methods, the actual electricity load demand data of two cities, including Queensland of Australia and Ho Chi Minh City of Vietnam, were utilized in this study. [ABSTRACT FROM AUTHOR]

In recent years, accurate electricity load forecasting has become increasingly essential for improving the management efficiency of power-generation systems. However, previously proposed hybrid models directly apply signal-processing technology in data preprocessing, resulting in poor efficiency in matching the electricity sequence characteristics. Moreover, most studies exhibit non-optimal performance in practical applications because they focus on forecasting accuracy and ignore forecasting stability. In this study, a combined framework that includes amodified noise processing strategy, multi-objective optimization algorithm, and deep neural network is proposed to solve the low prediction-accuracy problem in electricity load forecasting. The 30-minute real time data of electricity load from Queensland, Australia, are employed to verify the reliability of the proposed framework. The mean absolute percentage error values of the proposed framework in a multi-step prediction approached the values of MAPE 1−step = 0. 79 % , MAPE 2−step = 1. 13 % , and MAPE 3−step = 1. 50 % in Series 1, which significantly outperformed the existing contrast models. • A technology is proposed to combine meta-heuristic algorithm with de-noising method. • A modified algorithm is developed to quantitatively determine the degree of noise. • A multi-objective algorithm is designed to balance accuracy and stability. • A combined framework is designed to overcome the defects of individual model. [ABSTRACT FROM AUTHOR]

Short-term load forecasting (STLF) plays an irreplaceable role in the efficient management of electric systems. Particularly in the electricity market and industry, accurate forecasting could provide effective future plans and economic operations for operators of utilities and power systems. Hence, a more precise and stable load-forecasting model is essentially needed in the field of electricity load forecasting. To avoid the limitations of individual models, a new combined model is proposed. In this model, except for the multiple seasonal patterns used to reduce interferences from the original data, a new optimization algorithm is presented and applied to optimize the weight coefficients of the combined model based on non-positive constraint combination theory. To estimate the forecasting ability of the proposed combined model, half-hourly electricity power data from New South Wales, the State of Victoria and the State of Queensland in Australia were used in this paper as a case study. In the numerical experiments, compared with other six single models, the average mean absolute percent errors (MAPEs) of the combined model were reduced by 0.7138%, 1.0281%, 4.8394%, 0.9239%, 9.6316% and 7.3367%, respectively. [ABSTRACT FROM AUTHOR]

• Interval forecasting system is proposed to quantify the uncertainty information. • Data decomposition strategy is employed to process original data. • A novel method of determining weight to combine all individual models. • The Pareto optimal solution is theoretically proved. • Experiments demonstrate the availability and reliability of the proposed model. Electricity load prediction is of great significance to the development of the power market and stable operation of power systems. In recent years, scholars in this field have only considered point forecasting, which ignores the inevitable prediction bias and uncertain information. To fill this gap, this study proposes an interval prediction system consisting of an advanced data reconstruction strategy, a multi-objective optimization algorithm based on the theory of non-negative constraints, and an outstanding interval forecasting model fitted by the predicted fluctuation characteristics. Moreover, this study theoretically proves that the weight assigned by the optimization algorithm is the Pareto optimal solution. Empirical data with 30 min intervals from Queensland, Australia are selected as samples for research. The results not only demonstrate the superiority of the proposed model but also provide effective technical support for power grid operation and dispatch by quantifying changes in the prediction results caused by uncertainties. [ABSTRACT FROM AUTHOR]

As the basic guarantee for the reliability and economic operations of state grid corporations, power load prediction plays a vital role in power system management. To achieve the highest possible prediction accuracy, many scholars have been committed to building reliable load forecasting models. However, most studies ignore the necessity and importance of data preprocessing strategies, which may lead to poor prediction performance. Thus, to overcome the limitations in previous studies and further strengthen prediction performance, a novel short-term power load prediction system, VMD-BEGA-LSTM (VLG), integrating a data pretreatment strategy, advanced optimization technique, and deep learning structure, is developed in this paper. The prediction capability of the new system is evaluated through simulation experiments that employ the real power data of Queensland, New South Wales, and South Australia. The experimental results indicate that the developed system is significantly better than other comparative systems and shows excellent application potential. [ABSTRACT FROM AUTHOR]

Short-term power load forecasting occupies an important position in improving the operating efficiency and economic effects of power system. Aiming at improving forecast performance, a substantial number of load forecasting models are proposed. However, most of the previous studies ignored the limitations of individual prediction models and the necessity of data preprocessing, resulting in low forecast accuracy. In this study, a novel hybrid model which combines data preprocessing technology, individual forecasting algorithm and weight determination theory is successfully presented for obtaining higher accuracy and better forecasting ability. Among this model, the data preprocessing stage first uses a novel combination data preprocessing method , which overcomes the shortcomings of single preprocessing methods. In addition, a combined forecasting mechanism composed of RBF, GRNN and ELM is proposed using the weight determination theory , which exceeds the limits of individual prediction models and improves prediction accuracy. For the sake of assessing the availability of the proposed hybrid model, three datasets of half-hour power load of Queensland, South Australia and Victoria in Australia are selected in this study. The final experimental results show that the proposed model not only can approximate the actual power load very well, but also can be used as a helpful tool for power grid planning and dispatching. • Two preprocessing techniques, CEEMD and SSA, are used in two stages. • Multi-objective evolutionary algorithm MOGWO is used to decide the weighting factor of combined model. • Based on the three neural networks, the prediction accuracy of power load is improved. • A comprehensive assessment of the composite model is carried out to evaluate its forecasting performance. • The proposed new hybrid model provides powerful technical support for power network dispatch management. [ABSTRACT FROM AUTHOR]

As energy saving becomes more and more popular, electric load forecasting has played a more and more crucial role in power management systems in the last few years. Because of the real-time characteristic of electricity and the uncertainty change of an electric load, realizing the accuracy and stability of electric load forecasting is a challenging task. Many predecessors have obtained the expected forecasting results by various methods. Considering the stability of time series prediction, a novel combined electric load forecasting, which based on extreme learning machine (ELM), recurrent neural network (RNN), and support vector machines (SVMs), was proposed. The combined model first uses three neural networks to forecast the electric load data separately considering that the single model has inevitable disadvantages, the combined model applies the multi-objective particle swarm optimization algorithm (MOPSO) to optimize the parameters. In order to verify the capacity of the proposed combined model, 1-step, 2-step, and 3-step are used to forecast the electric load data of three Australian states, including New South Wales, Queensland, and Victoria. The experimental results intuitively indicate that for these three datasets, the combined model outperforms all three individual models used for comparison, which demonstrates its superior capability in terms of accuracy and stability. [ABSTRACT FROM AUTHOR]

The feed-in tariff (FIT) scheme for small photovoltaic (PV) systems is fast approaching its closure or expiration in many countries. One of the issues after the closure of the FIT will be to determine the selling price of the surplus PV generation, for which valuing the surplus PV generation is essential. This study extracted issues significant to valuing electricity from PV systems by using quantitative content analysis in four states in Australia, where the local economic regulator conventionally estimates the value. Through quantitative content analysis, the region-specific issues relevant to each state were extracted. For example, forecast of wholesale price is the most significant issue in Victoria, where the methodology was not clear and the avoided wholesale cost tends to be overestimated. Similarly, in New South Wales and Queensland, merit order effect was the most significant issue among the stakeholders, where PV installation was higher than that in the other states. Thus, the proposed analysis method can be used to extract the issues relevant to valuing the electricity from PV generation. [ABSTRACT FROM AUTHOR]

Reliable models that can forecast energy demand (G) are needed to implement affordable and sustainable energy systems that promote energy security. In particular, accurate G models are required to monitor and forecast local electricity demand. However, G forecasting is a multivariate problem, and thus models must employ robust pattern recognition algorithms that can detect subtle variations in G due to causal factors, such as climate variables. Therefore, this study developed an artificial neural network (ANN) model that used climatic variables for 6-hour (h) and daily G forecasting. The input variables included the six most relevant climate variables from Scientific Information for Land Owners (SILO) and 51 Reanalysis variables obtained from the European Centre for Medium-Range Weather Forecast (ECMWF) models. This information was used to forecast G data obtained from the energy utility (Energex) at 8 stations in southeast Queensland, Australia, by utilizing statistically significant lagged cross-correlations of G with its predictor variables. The developed ANN model was then benchmarked against multivariate adaptive regression spline (MARS), multiple linear regression (MLR), and autoregressive integrated moving average (ARIMA) models using various statistical metrics, such as relative root-mean square error (RRMSE %). Additionally, this study developed a hybrid ANN model by combining the forecasts of the ANN, MARS, and MLR models. The bootstrap (B) technique was also used with the hybrid ANN model, creating the B-hybrid ANN, to estimate the forecast uncertainty. According to both forecast horizons, the results indicated that the ANN model was more accurate than the ARIMA, MARS, and MLR models for G forecasting. Furthermore, the hybrid ANN was the most accurate model developed in this research study. For example, at the best site (Redcliffe), the hybrid ANN model generated an RRMSE of 3.85% and 4.37% for the 6-h and daily horizons, respectively. This study found that an ANN model could be used for accurately forecasting G over multiple horizons in southeast Queensland. • Importance of external datasets (climate and atmospheric Reanalysis) to forecast energy demand. • Developing a hybrid ANN model by combining the outputs of three models. • Estimating the forecast uncertainty using a bootstrap method. • Low and high-dimensional models. • Energy security studies should explore ANN models trained with climate variables. [ABSTRACT FROM AUTHOR]

Accurate electrical load forecasting always plays a vital role in power system administration and energy dispatch, which are the foundation of the smooth operation of the national economy and people's daily life. Thinking from this vision, many scholars have made great efforts to seek suitable optimization algorithms to improve the performance of existing forecasting algorithm. However, most of the studies ignore the inherent disadvantages of single optimization algorithm, which leads to sub-optimal forecasting performance. Therefore, a novel electric load forecasting system was successfully proposed in this paper by the combination of data preprocessing, hybrid optimization algorithms, and several single classical forecasting methods, which successfully overcomes the defects of single traditional forecasting models and achieves higher forecasting accuracy than that of single model optimization. Besides, the 30 min interval data of Queensland, Australia from March to April is used as illustrative examples to evaluate the performance of the developed model. The results of tests demonstrate that the proposed hybrid model can better approximate the actual value, and it can also be employed as a useful tool for smart grids dispatching planning. • Develop a novel power load forecasting system for grid safety dispatch. • Extract important information effectively from the original load time series. • Creatively propose two hybrid optimization methods to optimize forecasting models. • Set up a comprehensive evaluation system to verify the effectiveness of the models. [ABSTRACT FROM AUTHOR]

Electricity load forecasting plays an essential role in improving the management efficiency of power generation systems. A large number of load forecasting models aiming at promoting the forecasting effectiveness have been put forward in the past. However, many traditional models have no consideration for the significance of data preprocessing and the constraints of individual forecasting models. Moreover, most of them only focus on the forecasting accuracy but ignore the forecasting stability, resulting in nonoptimal performance in practical applications. This paper presents a novel hybrid model that combines an advanced data preprocessing strategy, a deep neural network, and an avant-garde multi-objective optimization algorithm, overcoming the defects of traditional models and thus improving the forecasting performance effectively. In order to evaluate the validity of the proposed hybrid model, the electricity load data sampled in 30-min intervals from Queensland, Australia are used as a case to study. The experiments show that the new proposed model is obviously superior to all other traditional models. Furthermore, it provides an effective technical forecasting means for smart grid management. [ABSTRACT FROM AUTHOR]