Prediction interval estimation for wind farm power generation forecasts using support vector machines

Accurate forecasting of wind power generation is quite an important as well as challenging task for the system operators and market participants due to its high uncertainty. It is essential to quantify uncertainties associated with wind power generation forecasts for their efficient application in optimal management of wind farms and integration into power systems. Prediction intervals (PIs) are well known statistical tools which are used to quantify the uncertainty related to forecasts by estimating the ranges of the future target variables. This paper investigates the application of a novel support vector machine based methodology to directly estimate the lower and upper bounds of the PIs without expensive computational burden and inaccurate assumptions about the distribution of the data. The efficiency of the method for uncertainty quantification is examined using monthly data from a wind farm in Australia. PIs for short term application are generated with a confidence level of 90%. Experimental results confirm the ability of the method in constructing reliable PIs without resorting to complex computational methods.

Utilising solar energy within conventional coal fired power stations

Although the thermodynamic advantages of using solar energy to replace the bled off steam in the regeneration system of Rankine cycle coal fired power stations has been proven theoretically, the practical techno/economic feasibility of the concept has yet to be confirmed relative to real power station applications. To investigate this concept further, computer modelling software “THERMSOLV” was specifically developed for this project at Deakin University, together with the support of the Victorian power industry and Australian Research Council (ARC). This newly developed software simulates the steam cycle to assess the techno/economic merit of the solar aided concept for various power station structures, locations and local electricity market conditions. Two case studies, one in Victoria Australia and one in Yunnan Province, China, have been carried out with the software. Chapter one of this thesis defines the aims and scope of this study. Chapter two details the literature search in the related areas for this study. The thermodynamic concept of solar aid power generation technology has been described in chapter three. In addition, thermodynamic analysis i.e. exergy/availability has been described in this chapter. The “Thermosolv” software developed in this study is detailed in chapter four with its structure, functions and operation manual included. In chapter five the outcomes of two case studies using the “Thermosolv” software are presented, with discussions and conclusions about the study in chapters 6 and 7 respectfully. The relevant recommendations are then made in chapter eight.

Multi-point and multi-level solar integration into a conventional coal-fired power plant

Solar-aided power generation (SAPG) is capable of integrating solar thermal energy into a conventional thermal power plant, at multi-points and multi-levels, to replace parts of steam extractions in the regenerative Rankine cycle. The integration assists the power plant to reduce coal (gas) consumption and pollution emission or to increase power output. The overall efficiencies of the SAPG plants with different solar replacements of extraction steam have been studied in this paper. The results indicate that the solar thermal to electricity conversion efficiencies of the SAPG system are higher than those of a solar-alone power plant with the same temperature level of solar input. The efficiency with solar input at 330 °C can be as high as 45% theoretically in a SAPG plant. Even the low-temperature solar heat at about 85 °C can be used in the SAPG system to heat the lower temperature feedwater, and the solar to electricity efficiency is nearly 10%. However, the low-temperature heat resource is very hard to be used for power generation in other types of solar power plants. Therefore, the SAPG plant is one of the most efficient ways for solar thermal power generation.

Too close for comfort : a fishway exit and a hydro-power station inlet

A major environmental issue for hydro-electric power generation is passage of fish through turbines, or entrainment onto trash racks. At Yarrawonga Weir, on the upper Murray River in south-eastern Australia, the positioning of a fish lock resulted in the potential for upstream migrating fish to be swept back into the adjacent power station by cross flows. In 2004, a 4.5-m long steel extension flume was attached to the exit to alleviate this problem. To determine the fate of native fish after exiting the extension flume, 72 individuals (305–1015 mm long) were implanted with radio-transmitters and released into the fish lock exit channel. In 2004 (power station inflows 10 300 ML day^–1), the majority of fish exited successfully (44 of 45) and only a single fish (2%) was entrained into the power station. In 2005 (power station inflows 12 000 ML day^–1), fish again exited successfully (26 of 27) but with a higher proportion entrained (5 of 27; 18%). This reduced success appeared to be related to strong transverse flows with high water velocities adjacent to the fish lock exit. The efficiency of fish passage at this site might be improved by altering water management strategies, integrating engineering and fish biology, and through field-testing of proposed solutions

Wind farm power uncertainty quantification using a mean-variance estimation method

This paper proposes an innovative optimized parametric method for construction of prediction intervals (PIs) for uncertainty quantification. The mean-variance estimation (MVE) method employs two separate neural network (NN) models to estimate the mean and variance of targets. A new training method is developed in this study that adjusts parameters of NN models through minimization of a PI-based cost functions. A simulated annealing method is applied for minimization of the nonlinear non-differentiable cost function. The performance of the proposed method for PI construction is examined using monthly data sets taken from a wind farm in Australia. PIs for the wind farm power generation are constructed with five confidence levels between 50% and 90%. Demonstrated results indicate that valid PIs constructed using the optimized MVE method have a quality much better than the traditional MVE-based PIs.

An approach for wind power integration using demand side resources

Renewable energy resources, especially wind power, are expected to provide a considerable portion of the world energy requirements in the near future. Large-scale wind power penetration impacts the electricity industry in many aspects and raises a number of technical challenges for the electricity network. A day-ahead network-constrained market clearing formulation is proposed which considers demand side resources. The proposed approach can provide flexible load profile and reduce the need for ramp up/down services by the conventional generators. This method can potentially facilitate a large penetration of wind power by shifting the wind power generation from the off-peak periods to the high-peak hours. The validity of the proposed approach has been verified using the IEEE 30 bus and 57 bus test systems.

Short-term load and wind power forecasting using neural network-based prediction intervals

Electrical power systems are evolving from today's centralized bulk systems to more decentralized systems. Penetrations of renewable energies, such as wind and solar power, significantly increase the level of uncertainty in power systems. Accurate load forecasting becomes more complex, yet more important for management of power systems. Traditional methods for generating point forecasts of load demands cannot properly handle uncertainties in system operations. To quantify potential uncertainties associated with forecasts, this paper implements a neural network (NN)-based method for the construction of prediction intervals (PIs). A newly introduced method, called lower upper bound estimation (LUBE), is applied and extended to develop PIs using NN models. A new problem formulation is proposed, which translates the primary multiobjective problem into a constrained single-objective problem. Compared with the cost function, this new formulation is closer to the primary problem and has fewer parameters. Particle swarm optimization (PSO) integrated with the mutation operator is used to solve the problem. Electrical demands from Singapore and New South Wales (Australia), as well as wind power generation from Capital Wind Farm, are used to validate the PSO-based LUBE method. Comparative results show that the proposed method can construct higher quality PIs for load and wind power generation forecasts in a short time.

Characteristics and applications of energy storage system to power network - A review

Maintaining reliability and stability of a power systems in transmission and distribution level becomes a big challenge in present scenario. Grid operators are always responsible to maintain equilibrium between available power generation and demand of end users. Maintaining grid balance is a bigger issue, in case of any unexpected generation shortage or grid disturbance or integration of any renewable energy sources like wind and solar power in the energy mix. In order to compensate such imbalance and to facilitate more renewable energy sources with the grid, energy storage system (ESS) started to be playing an important role with the advancement of the state of the art technology. ESS can also help to get reduction in greenhouse gas (GHG) emission by means of integrating more renewable energy sources to the grid. There are various types of Energy Storage (ES) technologies which are being used in power systems network from large scale (above 50MW) to small scale (up to 100KW). Based on the characteristics, each storage technology has their own merits and demerits. This paper carried out extensive review study and verifies merits and demerits of each storage technology and identifies the suitable technology for the future. This paper also has conducted feasibility study with the aid of E-SelectTM tool for various ES technologies in applications point of view at different grid locations. This review study helps to evaluate feasible ES technology for a particular electrical application and also helps to develop smart hybrid storage system for grid applications in efficient way.

An optimized mean variance estimation method for uncertainty quantification of wind power forecasts

A statistical optimized technique for rapid development of reliable prediction intervals (PIs) is presented in this study. The mean-variance estimation (MVE) technique is employed here for quantification of uncertainties related with wind power predictions. In this method, two separate neural network models are used for estimation of wind power generation and its variance. A novel PI-based training algorithm is also presented to enhance the performance of the MVE method and improve the quality of PIs. For an in-depth analysis, comprehensive experiments are conducted with seasonal datasets taken from three geographically dispersed wind farms in Australia. Five confidence levels of PIs are between 50% and 90%. Obtained results show while both traditional and optimized PIs are hypothetically valid, the optimized PIs are much more informative than the traditional MVE PIs. The informativeness of these PIs paves the way for their application in trouble-free operation and smooth integration of wind farms into energy systems. © 2014 Elsevier Ltd. All rights reserved.

Modelling and power quality analysis of a grid-connected solar PV system

Increased concern about global warming coupled with the escalating demand of energy has driven the conventional power system to be more reliable one by integrating Renewable Energies (RE) in to grid. Over the recent years, integration of solar PV forming a gridconnected PV is considered as one of the most promisingtechnologies to the developed countries like Australia to meet the growing demand of energy. This rapid increase in grid connected photovoltaic (PV) systems has made the supply utilities concerned about the drastic effects that have to be considered on the distribution network in particular voltage fluctuations, harmonic distortions and the Power factor for sustainable power generation. However, irrespective of thefact that the utility grid can accommodate the variability of load or irregular solar irradiance, it is essential to study the impact of grid connected PV systems during higher penetration levels as the intermittent nature of solar PV adversely effects the grid characteristics in meeting the load demand. Hence, keeping this in track, this paper examines the grid-connected PV system considering a residential network of Geelong region (38◦.09' S and 144◦.21’ E) and explores the level of impacts considering summer load profile with a change in the level of integrations. Initially, a PV power system network model is developed in Matlab-Simulink environment and the simulations are carried out to explore the impacts of solar PV penetration at low voltage distribution network considering power quality (PQ) issues such as voltage fluctuations, harmonics distortion at different load conditions.

Maximum power point tracking of partial shaded photovoltaic array using an evolutionary algorithm: a particle swarm optimization technique

 Partial shading is one of the unavoidable complications in the field of solar power generation. Although the most common approach in increasing a photovoltaic (PV) array’s efficiency has always been to introduce a bypass diode to the said array, this poses another problem in the form of multi-peaks curves whenever the modules are partially shaded. To further complicate matters, most conventional Maximum Power Point Tracking methods develop errors under certain circumstances (for example, they detect the local Maximum Power Point (MPP) instead of the global MPP) and reduce the efficiency of PV systems even further. Presently, much research has been undertaken to improve upon them. This study aims to employ an evolutionary algorithm technique, also known as particle swarm optimization, in MPP detection. VC 2014 Author(s).

A novel fuzzy multi-objective framework to construct optimal prediction intervals for wind power forecast

The forecasting behavior of the high volatile and unpredictable wind power energy has always been a challenging issue in the power engineering area. In this regard, this paper proposes a new multi-objective framework based on fuzzy idea to construct optimal prediction intervals (Pis) to forecast wind power generation more sufficiently. The proposed method makes it possible to satisfy both the PI coverage probability (PICP) and PI normalized average width (PINAW), simultaneously. In order to model the stochastic and nonlinear behavior of the wind power samples, the idea of lower upper bound estimation (LUBE) method is used here. Regarding the optimization tool, an improved version of particle swam optimization (PSO) is proposed. In order to see the feasibility and satisfying performance of the proposed method, the practical data of a wind farm in Australia is used as the case study.

A novel distributed algorithm for power loss minimizing in Smart Grid

Because power generation of renewable resources are unstable and demands of the customers are time-varying, the supply power and demands of the customers are always unequal. To meet the demands of the customers, power is transmitted from primary power generation to secondary power generation. It will cause high power loss. To solve this problem, a distributed algorithm is proposed in this paper. By using the algorithm, the micro-grids are able to exchange power with their neighbors so as to minimize the total power losses of the smart grid. Moreover, communication overhead (bandwidth) is reduced, comparing with centralized algorithm. Through computer simulations, we demonstrate that the proposed algorithm can lead to near-optimal result for alleviating the average power loss per micro-grid and reduce the communication overhead significantly in contrast with the centralized approach.

Simulation and hardware implementation of new maximum power point tracking technique for partially shaded PV system using hybrid DEPSO method

In photovoltaic (PV) power generation, partial shading is an unavoidable complication that significantly reduces the efficiency of the overall system. Under this condition, the PV system produces a multiple-peak function in its output power characteristic. Thus, a reliable technique is required to track the global maximum power point (GMPP) within an appropriate time. This study aims to employ a hybrid evolutionary algorithm called the DEPSO technique, a combination of the differential evolutionary (DE) algorithm and particle swarm optimization (PSO), to detect the maximum power point under partial shading conditions. The paper starts with a brief description about the behavior of PV systems under partial shading conditions. Then, the DEPSO technique along with its implementation in maximum power point tracking (MPPT) is explained in detail. Finally, Simulation and experimental results are presented to verify the performance of the proposed technique under different partial shading conditions. Results prove the advantages of the proposed method, such as its reliability, system-independence, and accuracy in tracking the GMPP under partial shading conditions.