Prediction intervals for short-term wind farm power generation forecasts

Quantification of uncertainties associated with wind power generation forecasts is essential for optimal management of wind farms and their successful integration into power systems. This paper investigates two neural network-based methods for direct and rapid construction of prediction intervals (PIs) for short-term forecasting of power generation in wind farms. The lower upper bound estimation and bootstrap methods are used to quantify uncertainties associated with forecasts. The effectiveness and efficiency of these two general methods for uncertainty quantification is examined using twenty four month data from a wind farm in Australia. PIs with a confidence level of 90% are constructed for four forecasting horizons: five, ten, fifteen, and thirty minutes. Quantitative measures are applied for objective evaluation and unbiased comparison of PI quality. Demonstrated results indicate that reliable PIs can be constructed in a short time without resorting to complicate computational methods or models. Also quantitative comparison reveals that bootstrap PIs are more suitable for short prediction horizon, and lower upper bound estimation PIs are more appropriate for longer forecasting horizons.

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.

Incorporation of a hybrid cooling system in water cooled power stations to reduce water and energy consumptions

Water shortage is a major problem facing the power industry in many nations around the world. The largest consumer of water in most power plants is the wet cooling tower. To assist water and energy saving for thermal power stations using conventional evaporative wet cooling towers, a hybrid cooling system is proposed in this paper. The hybrid cooling system may consists of all or some of an air pre-cooler, heat pump, heat exchangers, and adsorption chillers together with the existing cooling tower. The hybrid cooling system described in the paper, consisting of a metal hydride heat pump operating in conjunction with the existing wet cooling tower, is capable of achieving water saving by reducing the temperature of warm water entering the cooling tower. Cooler inlet water temperatures effectively reduce the cooling load on existing towers. This will ultimately reduce the amount of water lost to the air by evaporation whilst still achieving the same cooling output. At the same time, the low grade waste energy upgraded by the metal hydride heat pump, in the process of cooling the water, can be used to replace the bleed of steam for the lower stage feed heaters which will increase overall cycle efficiency.

Hybrid cooling system for water conservation in conventional power stations

This thesis proposes the concept of a hybrid cooling system to minimise water consumption in thermal power stations. It identifies that a significant amount of water can be potentially saved by the proposed system. Additional energy input requirements were less than previous approaches considered.

Intelligent energy management for plug-in hybrid electric vehicles : the role of ITS infrastructure in vehicle electrification

The desire to reduce carbon emissions due to transportation sources has led over the past decade to the development of new propulsion technologies, focused on vehicle electrification (including hybrid, plug-in hybrid and battery electric vehicles). These propulsion technologies, along with advances in telecommunication and computing power, have the potential of making passenger and commercial vehicles more energy efficient and environment friendly. In particular, energy management algorithms are an integral part of plug-in vehicles and are very important for achieving the performance benefits. The optimal performance of energy management algorithms depends strongly on the ability to forecast energy demand from the vehicle. Information available about environment (temperature, humidity, wind, road grade, etc.) and traffic (traffic density, traffic lights, etc.), is very important in operating a vehicle at optimal efficiency. This article outlines some current technologies that can help achieving this optimum efficiency goal. In addition to information available from telematic and geographical information systems, knowledge of projected vehicle charging demand on the power grid is necessary to build an intelligent energy management controller for future plug-in hybrid and electric vehicles. The impact of charging millions of vehicles from the power grid could be significant, in the form of increased loading of power plants, transmission and distribution lines, emissions and economics (information are given and discussed for the US case). Therefore, this effect should be considered in an intelligent way by controlling/scheduling the charging through a communication based distributed control.

Power plant cooling technologies : hybrid cooling systems can save water

In addition to water saving, the hybrid cooling concept presented in this book also has the potential to improve energy efficiency and possibly reduce CO2 emission by recovering and upgrading the 'waste' energy from the cooling water stream.Yilmaz and Kouzani are at Deakin Uni, Hessami is at Monash Uni, Hu is at Uni of Adelaide.

Wind-thermal systems operation optimization considering emission problem

This paper proposes a hybrid computational framework based on Sequential Quadratic Programming (SQP) and Particle Swarm Optimization (PSO) to address the Combined Unit Commitment and Emission (CUCE) problem. By considering a model which includes both thermal generators and wind farms, the proposed hybrid computational framework can minimize the scheduling cost and greenhouse gases emission cost. The viability of the proposed hybrid technique is demonstrated using a set of numerical case studies. Moreover, comparisons are performed with other optimization algorithms. The simulation results show that our hybrid method is better in terms of the speed and accuracy. The main contribution of this paper is the development of a emission unit commitment model integrating with wind energy and combining the SQP and PSO methods to achieve faster and better performance optimization

Hybrid technique of ant colony and particle swarm optimization for short term wind energy forecasting

Wind farms are producing a considerable portion of the world renewable energy. Since the output power of any wind farm is highly dependent on the wind speed, the power extracted from a wind park is not always a constant value. In order to have a non-disruptive supply of electricity, it is important to have a good scheduling and forecasting system for the energy output of any wind park. In this paper, a new hybrid swarm technique (HAP) is used to forecast the energy output of a real wind farm located in Binaloud, Iran. The technique consists of the hybridization of the ant colony optimization (ACO) and particle swarm optimization (PSO) which are two meta-heuristic techniques under the category of swarm intelligence. The hybridization of the two algorithms to optimize the forecasting model leads to a higher quality result with a faster convergence profile. The empirical hourly wind power output of Binaloud Wind Farm for 364 days is collected and used to train and test the prepared model. The meteorological data consisting of wind speed and ambient temperature is used as the inputs to the mathematical model. The results indicate that the proposed technique can estimate the output wind power based on the wind speed and the ambient temperature with an MAPE of 3.513%.

Stability enhancement of DFIG wind turbine using LQR pitch control over rated wind speed

A novel pitch control design method is proposed for the doubly fed induction generator (DFIG) wind turbine (WT) using linear quadratic regulator (LQR). A seven-order model represents the DFIG WT which is linearized by truncated Taylor series expansion. A systematic approach is adopted to determine the weighting matrices in LQR design for the optimal solution. Simulations have been carried out to compare the performance of the proposed LQR pitch control method against a PI pitch control for small and large disturbances. It is shown that the proposed control method enhances low-voltage ride-through capability and improves system damping under large disturbances.

Static output feedback frequency stabilization of time-delay power systems with coordinated electric vehicles state of charge control

In this paper, for the first time, electric vehicles are used for both the primary and secondary frequency controls to support power plants to rapidly suppress fluctuations in the system frequency due to load disturbances. Via networked control and wide-area communication infrastructures, multiple interval time-varying delays exist in the communication channels between the control center, power plant, and an aggregation of electric vehicles. By coordinating batteries’ state of charge control, the behaviors of the vehicle owners and the uncertainties imposed by the changes of the batteries’ state of charge are taken intoconsideration. A power system model incorporating multiple time-varying delays and uncertainties is first proposed. Then, a robust static output feedback frequency controller is designed to guarantee the resulting closed-loop system stable with an H∞ attenuation level. By utilizing a novel integral inequality, namely refined-Jensen inequality, and an improved reciprocally convex combination, the design conditions are formulated in terms of tractable linear matrix inequalities which can be efficiently solved by various computational tools. The effectiveness of the proposed control scheme is verified by extensive simulations.

Mining small databases by collecting knowledge

Current data mining techniques may not be helpful for mining some companies/organizations such as nuclear power plants and earthquake bureaus, which have only small databases. Apparently, these companies/organizations also expect to apply data mining techniques to extract useful patterns in their databases so as to make their decisions. However, data in these databases such as the accident database of a nuclear power plant and the earthquake database in an earthquake bureau, may not be large enough to form any patterns. To meet the applications, we present a new mining model in this paper, which is based on the collecting knowledge from such as Web, journals, and newspapers.

Collecting quality data for database mining

Data collecting is necessary to some organizations such as nuclear power plants and earthquake bureaus, which have very small databases. Traditional data collecting is to obtain necessary data from internal and external data-sources and join all data together to create a homogeneous huge database. Because collected data may be untrusty, it can disguise really useful patterns in data. In this paper, breaking away traditional data collecting mode that deals with internal and external data equally, we argue that the first step for utilizing external data is to identify quality data in data-sources for given mining tasks. Pre- and post-analysis techniques are thus advocated for generating quality data.