Long-term Wind Power Research Articles

Effective power dispatch capability decision method for a wind‐battery hybrid power system

With integration of a battery energy storage system (BESS) into the wind power system, the wind power variation can be mitigated to dispatch a constant power to the grid. To effectively supply the constant power dispatch to the grid with a limited BESS, the power dispatch capability should be defined primarily at beginning of each dispatching interval to cooperate with the transmission system operator. This study introduces a method to determine the power dispatch capability of a wind-battery hybrid power system with the minimal battery capacity. The power dispatch capability is decided under the condition that the battery power is maintained below its rating and that the state of charge of the battery is guaranteed within a safe range. For economic BESS operation, the battery capacity, including the power and energy ratings, is also determined cost-effectively to obtain the constant power dispatch. Based on the long-term wind power profile, the authors develop the determination process to minimise the battery capacity which ensures that the BESS sufficiently mitigates wind power variation to dispatch the constant power into grid under all wind conditions. To evaluate the proposed determination method, a case study is carried out with a 3-MW wind turbine generator and real wind speed data measured on Jeju Island in Korea.

Temporal Versus Stochastic Granularity in Thermal Generation Capacity Planning With Wind Power

We propose a stochastic generation expansion model, where we represent the long-term uncertainty in the availability and variability in the weekly wind pattern with multiple scenarios. Scenario reduction is conducted to select a representative set of scenarios for the long-term wind power uncertainty. We assume that the short-term wind forecast error induces an additional amount of operating reserves as a predefined fraction of the wind power forecast level. Unit commitment (UC) decisions and constraints for thermal units are incorporated into the expansion model to better capture the impact of wind variability on the operation of the system. To reduce computational complexity, we also consider a simplified economic dispatch (ED) based model with ramping constraints as an alternative to the UC formulation. We find that the differences in optimal expansion decisions between the UC and ED formulations are relatively small. We also conclude that the reduced set of scenarios can adequately represent the long-term wind power uncertainty in the expansion problem. The case studies are based on load and wind power data from the state of Illinois.

Development of Wind Power Data for Power Supply-demand Analysis and Analysis of Long-term Wind Power Variability

Wind power output varies over time under the influence of meteorological conditions. Understanding these variations is indispensable to further wind power integration and effective operation of the power system in the near future. In this study, we developed time-series wind power output data of 9 power systems in Japan in the 2010 and the 2011 fiscal year for power supply-demand analysis. We analyzed the variations in time windows of 20 minutes or more. Furthermore, we evaluated the effects of output increasing rate restriction control. We found that the large variations can be absorbed by the small energy loss, and that this increasing rate restriction control is very effective against infrequent large variations.

Long-Term Wind Power Prediction Based on Rough Set

In long-term wind power prediction, dealing with the relevant factors correctly is the key point to improve the prediction accuracy. This paper presents a prediction method with rough set analysis. The key factors that affect the wind power prediction are identified by rough set theory. The chaotic characteristics of wind speed time series are analyzed. The rough set neural network prediction model is built by adding the key factors as the additional inputs to the chaotic neural network model. Data of Fujin wind farm are used for this paper to verify the new method of long-term wind power prediction. The results show that rough set method is a useful tool in long-term prediction of wind power.

Closure to discussion on "A method for short-term wind power prediction with multiple observation points"

Summary form only given. The authors appreciate the valuable comments expressed by the discussers. Our comments are numbered to match the points made in the discussions. 1) It is true that time frames for wind power forecasting are very important for different applications. For example, very short-term wind power forecasting, in terms of seconds to minutes, is important for frequency control, electricity market clearing, and regulation actions. Short-term wind power forecasting, in terms of minutes to hours, is used for economic load dispatch planning and load increment/decrement decisions. Similarly, long-term wind power forecasting, in terms of hours to days, is required for other applications like unit commitment, maintenance scheduling to obtain optimal operating cost, reserve requirement decisions, prediction of power system congestion, and operational security in day-ahead electricity market. In particular, short-term wind power predictions are becoming increasingly important nowadays since the rise of the competitive electricity markets. In this paper, initially the authors have tested the algorithm only for some minutes ahead. However, the prediction for other time frames would be considered in the future. 2) As mentioned in the paper, the optimum values of the model parameters were found by numerical iterations. While making the empirical studies, it was observed that historical time interval during each exercise was different for different data ranges and times. Therefore, the optimal number of historical data points for the given data set was chosen based on manual trial-and-error methods. 3) First of all, we thank the discussers for their appreciation for the proposed idea. The accuracy improvement for wind power prediction was calculated by comparing the results of direction-dependent power curves with a single average power curve for the given data set. It depends on wind velocity as the discusser mentioned. As regards the correlation between the spread of data in Fig. 7 and the spread of data due to wind direction, actually Fig. 7 is shown as a general example to compare a manufacturer's power curve, as in Fig 6, and an actual power curve in a wind farm. In particular, unfortunately it is not the plot of same actual data used in the analyses. Therefore, specifically there is no relation of Fig. 7 with our direction-dependent power curves. As our analyses are based on numerical calculations, we have not worked out analytical relation yet to describe the wind power relation with wind speed and direction together. However, based on these preliminary results and discussers' comments, we will improve this idea of direction-dependent power curves with a generalized relation and conclusion. 4) The wind direction was not predicted using Persistence and Grey model. The proposed method has not been compared to other neural networks or Bayesian-based methods yet. We would consider to improve the proposed technique with automatic and adaptive tuning of the model parameters along with its comparison with other advanced prediction techniques. Once again, we are thankful to the discussers for their interest in our research paper and valuable feedback.

Data mining and wind power prediction: A literature review

Abstract Wind power generated by wind turbines has a non-schedulable nature due to the stochastic nature of meteorological conditions. Hence, wind power predictions are required for few seconds to one week ahead in turbine control, load tracking, pre-load sharing, power system management and energy trading. In order to overcome problems in the predictions, many different wind power prediction models have been used to achieve in the literature. Data mining and its applications have more attention in recent years. This paper presents a review study banned on very short-term, short-term, medium-term and long-term wind power predictions. The studies available in the literature have been evaluated and criticized in consideration with their prediction accuracies and deficiencies. It is shown that adaptive neuro-fuzzy inference systems, neural networks and multilayer perceptrons give better results in wind power predictions.

A new probabilistic method to estimate the long-term wind speed characteristics at a potential wind energy conversion site

This paper proposes the use of a new Measure–Correlate–Predict (MCP) method to estimate the long-term wind speed characteristics at a potential wind energy conversion site. The proposed method uses the probability density function of the wind speed at a candidate site conditioned to the wind speed at a reference site. Contingency-type bivariate distributions with specified marginal distributions are used for this purpose. The proposed model was applied in this paper to wind speeds recorded at six weather stations located in the Canary Islands (Spain). The conclusion reached is that the method presented in this paper, in the majority of cases, provides better results than those obtained with other MCP methods used for purposes of comparison. The metrics employed in the analysis were the coefficient of determination ( R 2) and the root relative squared error (RRSE). The characteristics that were analysed were the capacity of the model to estimate the long-term wind speed probability distribution function, the long-term wind power density probability distribution function and the long-term wind turbine power output probability distribution function at the candidate site.

Braking wind in Australia: A critical evaluation of the renewable energy target

This paper provides a critical evaluation of Australia’s new Renewable Energy Target (RET) program with respect to its capacity to support wind power development. Four structural flaws associated with the RET which undermine its effectiveness as a catalyst for technological change in the electricity sector are discussed: (i) the inclusion of waste coal mine gas (WCMG) as an eligible fuel source which acts as an indirect coal industry subsidy, (ii) program duration which is too short and ill-structured, (iii) a multiplier that is well-intended to support small-scale renewable technologies but which creates “phantom capacity”, and (iv) the capped target of 45,000 GWh which will stymie long-term wind power market investment. The paper concludes with recommendations which stress the importance of passing effective Carbon Pollution Renewable Scheme (CPRS) legislation to offset the weaknesses associated with the RET. If an effective CPRS cannot be implemented, the paper recommends that amendments be made to the RET to (i) remove WCMG from the list of approved alternative energy sources, and (ii) extend the RET targets to reach 120,000 GWh by 2030.

국가바람지도 및 지리정보시스템 기반의 해상풍력단지 입지전략 연구

This study has analyzed the scale, location, resource potential and feasibility of offshore wind farm scientifically and systematically based on the national wind map and GIS. For long-term wind power development, this study pursues siting strategy building, selection of target area and deciding development priority as well as the presenting a basis for assessment that are necessary for policy decision making by making theme layers under GIS environment. According to the analysis after organizing technological development by stages, even if only the most suitable sites are developed among the area of offshore wind farm candidates that can be developed under the current technological standard, it has been evaluated as being able to develop about 3 times of the wind power dissemination target until 2012. It is expected that about 5% of territorial water area can be developed in a short-term future while the southern offshore area possessing relatively favorable wind resource than the western offshore has been identified as the most feasible site. While about 23% of territorial water area has been classified as potential area for offshore wind farm development in a long-term future, even Jeju Island and offshore of Ulsan possessing excellent wind resource have been analyzed as feasible sites. The feasibility assessment of offshore wind lam development established by this study is expected to assist national strategy building for accomplishing the wind power dissemination target.

Long-Term Wind Speed and Power Forecasting Using Local Recurrent Neural Network Models

This paper deals with the problem of long-term wind speed and power forecasting based on meteorological information. Hourly forecasts up to 72-h ahead are produced for a wind park on the Greek island of Crete. As inputs our models use the numerical forecasts of wind speed and direction provided by atmospheric modeling system SKIRON for four nearby positions up to 30 km away from the wind turbine cluster. Three types of local recurrent neural networks are employed as forecasting models, namely, the infinite impulse response multilayer perceptron (IIR-MLP), the local activation feedback multilayer network (LAF-MLN), and the diagonal recurrent neural network (RNN). These networks contain internal feedback paths, with the neuron connections implemented by means of IIR synaptic filters. Two novel and optimal on-line learning schemes are suggested for the update of the recurrent network's weights based on the recursive prediction error algorithm. The methods assure continuous stability of the network during the learning phase and exhibit improved performance compared to the conventional dynamic back propagation. Extensive experimentation is carried out where the three recurrent networks are additionally compared to two static models, a finite-impulse response NN (FIR-NN) and a conventional static-MLP network. Simulation results demonstrate that the recurrent models, trained by the suggested methods, outperform the static ones while they exhibit significant improvement over the persistent method.

Locally recurrent neural networks for long-term wind speed and power prediction

The paper deals with a real-world application, the long-term wind speed and power forecasting in a wind farm using locally recurrent multilayer networks as forecast models. To cope with the complexity of the process and to improve the performance of the models, a class of optimal on-line learning algorithms is employed for training the locally recurrent networks based on the recursive prediction error (RPE) algorithm. A global RPE algorithm is devised and three local learning algorithms are suggested by partitioning the GRPE into a set of sub-problems at the neuron level to reduce computational complexity and storage requirements. Experimental results on the wind prediction problem demonstrate that the proposed algorithms exhibit enhanced performance, in terms of convergence speed and the accuracy of the attained solutions, compared to conventional gradient-based methods. Furthermore, it is shown that the suggested recurrent forecast models outperform the atmospheric and time-series models.

Wind resource potentials at Quetta

Hourly wind data of Quetta airport (Samungli) for 1984 and 1985 are analysed on the basis of speed frequency curves, diurnal variations and of the speed duration curves. In this paper, we conclude that the speed frequency curves give more reliable information as compared to speed duration curves, from the view-point of the wind energy utility, i.e. the estimation of the average power output. However, wind-tunnel testing is required to observe wind flow over the topography of Quetta and to estimate the longterm wind power. We conclude, in addition that the location of data collection as well as the accuracy of input wind data becomes critical to wind resource assessment; which required computer simulation of wind data over at least a period of 10 years by numerical analysis and mathematical models.