Liberalisation and deregulation of the electricity market led to changes among the previously monopoly electric-power industry, as it was separated to a market and a government-regulated part. The reconstructed electric-power industry requires various utilities for generation, transmission and distribution of the electrical power to the consumers. The market-regulated part includes utilities for electrical-power generation, marketing and electricity trading, while the government-regulated part is responsible for proper power-system operation and control. Since the power-system operation characteristics generally do not correspond to the reconstructed electric-power sector, introduction of new ancillary services such as frequency and voltage control, scheduling and dispatch, loss compensation and load following was necessary. These services are coordinated by the transmission-system operator.
One of the main tasks of a power system is to supply all consumers with electrical power within defined reliability standards, including an uninterrupted power supply. Disturbances in power system can occur randomly at any time and may lead to power-system instability or even its collapse, where several consumers may be affected. The probability of disturbances can be minimized by proper power-system upgrades and expansions, by replacing or upgrading the existing equipment with a more reliable one, by providing an adequate diversity of the elements, by providing an adequate transmission capacity of the network, but nevertheless, the disturbances in the power system cannot be completely prevented. The reasons for this lies in the random occurrence of disturbances, the possible random peak loads in the system, the variable and uncertain power generation from the renewable energy sources, the delays in delivery, installation, replacement and maintenance of the equipment and in the unpredictable weather conditions.
One of the characteristics of a power system is its inability to efficiently accumulate the electrical energy. To achieve an uninterrupted power supply reaching a balance between the generation and the consumption of electricity is needed. To ensure energy balance, additional generation operating reserves above the expected demand load are required. The operating reserves can be utilized in case of load and generation mismatch, which enables the power system to deal with unexpected changes in load or generation. Achieving the system balance has become more difficult with the increasing number of renewable energy sources, as the generation of electrical power from most of these unconventional energy sources is intermittent by nature. Operating reserves are most often ensured by an installed cold-capacity reserve, which is not considered as part of a reliability evaluation, by potential-energy storage facilities, by pumped-storage hydro-power plants, by de-loaded wind-power plants or they can be imported. Some transmission-system operators suggest that for each MW of installed power from renewable energy sources another MW of backup must be available. However, introducing such measures may lead to very high expenses, which may not justify the benefits of the high reliability. On the other hand, operating reserves can also be defined by reliability analyses.
The doctoral dissertation is focused on a development of an analysis of power generation reliability level that will consider the impact of variable renewable energy sources in the observed power system for a reliable day-ahead operation. Additionally, common cause failures of several generating units are considered.
In the first part of the dissertation a theoretical background is given: the probability and reliability theory, power-generation reliability analysis, short-term forecasts of the expected demand load and the expected power generation from the renewable energy sources, ensuring an adequate level of operating reserves in the observed power systems and probabilistic risk assessment, where the common cause failures are discussed. The discussed background is basis for working in fields of power-generation reliability analysis and assessment.
In the second part of the thesis a new method for ensuring a reliable power generation is presented. The new method is based on an already well-established method for assessing the power-generation reliability, i.e., the Loss of Load Expectation (LOLE). Firstly, the definition of the index LOLE is expanded by the implementation of common cause failures of several generating units. The common cause failures are implemented using the Beta factor method, the Multiple Greek Letter method and the updated Multiple Greek Letter method. Additionally, an updated Multiple Greek Letter method is presented, which enables a more detailed definition of outage states of generating units, where several root causes and coupling mechanisms for one generating unit can be considered.
Secondly, the LOLE definition is improved by the implementation of the renewable energy sources, whose power generation is variable and uncertain. Their random failures due to mechanical errors and equipment malfunctions are also considered. The upgraded index LOLE is then used to evaluate the power-generation reliability in each hour of the following day separately, as the short-term forecasts of the power generation from the renewable energy sources are accurate only for a few hours or a day in advance. The obtained hourly values of index LOLE are then used to determine, what amount of additional operating reserve within every hour is required to satisfy the reliability criteria. In the third part of the doctoral thesis the results of the application of the new method are presented. The method was tested on a standard power system, i.e., the simplified 39-bus system of New England, and on a real power system, i.e., the Slovenian power system. The results show that the larger the system balance is (i.e., the greater the expected demand load reduced by the power generation from the renewable energy sources), the lower the power-generation reliability is. Therefore, the power-generation reliability in every hour depends on the expected demand load within the hour and on the forecasted power generation from the renewable energy sources.
Uncertainty of the renewable power generation also has a major impact on the power-generation reliability. The larger the forecast error for the power generation from the renewable energy sources is, the higher the value of index LOLE is. The larger the share of installed renewable energy sources is, the greater the impact of uncertainty of their generation on the power-generation reliability is. This represents the negative aspect of incorporating the renewable energy sources in power systems, as we cannot influence and precisely forecast their generation and can therefore not absolutely rely on their generation. Due to the uncertain generation from the renewable energy sources it is very difficult to accurately determine the power-generation reliability, which may consequently have a major impact on the determination of the appropriate level of additional operating reserve for ensuring a reliable day-ahead power-system operation.
The results also confirm that common cause failures of generating units have a major contribution to risk. If it is assumed, that two generating units can share the same cause of failure, the obtained power-generation reliability decreases and consequently a larger amount of additional operating reserve is required to satisfy the reliability criteria. The more generating units that are susceptible to common cause failures, the higher that their installed powers and unavailabilities are and the higher share of all their failures that are represented by the common cause failures, the lower the power-generation reliability is. Consequently, more additional operating reserve is suggested to achieve the desired level of reliability. This suggests that common cause failures of generating units need to be considered within reliability analysis, as they have a large impact on the power-generation reliability and on the amount of the required additional operating reserve. More, if common cause failures are considered according to the new Multiple Greek Letter method, outage states of generating units can be more accurately defined and a more detailed insight into reliability analysis is enabled.
By application of the presented method a reliable short-term electrical power supply to all consumers can be ensured; especially in modern power systems, where a large share of renewable energy sources is being installed. Additionally, the method takes into account the impact of simultaneous failures of several generating units caused by a common cause, such as the lightning strike. Consequently, the outage states of generating units can be more accurately defined, which enables the transmission-system operator to identify the critical failures of the generating units and to propose adequate measures in order to minimize the impact of common cause failures and therefore to ensure a higher power-generation reliability. The new method presents an original scientific contribution of the doctoral thesis, as it combines the impact of the variable and uncertain power generation from the renewable energy sources and the impact of simultaneous failures of several generating units caused by a common cause on the amount of the additional operating reserve, which is required for ensuring a reliable power system operation.
|