In June 2006, northeast Poland experienced a complete system failure in the form of a blackout that lasted a few hours. This major fault prompted action to find a permanent solution to a voltage control problem. Research and development work on this issue was conducted in Polish technical universities and foreign institutions before the Electrical Power Research Institute (EPRI) was able to offer a solution in June 2008.

For large transmission systems, voltage and reactive power control systems are key. An unavoidable parameter of system load flows is the presence of reactive power associated with the phase difference between voltage and current. The reactive power is partially compensated on the customer side while the remainder effectively loads the network. Supply contracts used by the industry do not specify the need for unity power factor.

Reactive power, used by transmission system operators (TSOs) to control the transmission line voltage, is additional to the load currents, increasing the voltage drop across network impedances. By adjusting the reactive power flow, the TSO can change voltage drop along the lines and the receiving end voltage, namely at the customer connection point. Therefore, the voltage at the customer connection is dependent on the voltage control used by the TSO between the source generator and the customer load.

As the majority of reactive power is generated by power plants, this method of voltage control has limitations. The generators are able to deliver adjustable leading and lagging reactive power without affecting fuel costs. The presence of reactive power effectively reduces the load transfer capability of the transmission system. However, because of the increased transmission line losses, it is not cost effective to transmit reactive power over long distances.

Northeast Poland has no significant generation plants; therefore, the energy (active power) and reactive power are delivered by other regions of the country. It is unfortunate the transmission lines from the main power plants are located some 1000 km (621 miles) southwest of the transmission system problems in northeast Poland. Voltage control is managed by capacitors and reactors installed at various system operational facilities along the transmission line route.

The Blackout

The Polish power system blackout occurred on June 26, 2006, when the predicted system morning peak demand was 18,200 MW, a demand much higher compared with the same month in previous years.

The demand was to be supplied from 75 generation units supported by a spinning reserve of 1350 MW, which included a second reserve of 237 MW, a minute reserve of 656 MW and a cold reserve of about 2600 MW.

In northeast Poland, there is a lack of grid-connected generation, the nearest power plant in the region being Ostroleka (3 × 200 MW), of which one unit was off-line for maintenance at the time of the blackout. In the early hours of June 26, there was the loss of a generator at Patnow Power Plant and, before noon, another four units, two in Kozienice and two in Laziska Power Plants, were disconnected from the system.

These five generators were the main source of supply to northeast Poland, so by 7 a.m., 570 MW of generation had been lost. Simultaneously, the predicted consumption was some 600 MW less than the system demand. By 1 p.m., there was a demand shortfall of 1100 MW. In the meantime, one generator in the Dolna Odra Power Plant was activated, but for technical reasons, the availability of the cold-reserve generation was delayed by six hours.

Adding to those issues, an unusual heat wave hit the country, causing deterioration of the operational conditions within the power plants. As a result of insufficient cooling water and above-average water temperatures, the generating capability of some power plants was reduced. This situation was mainly constricted to the power plants located in central and northern Poland. The loading on some of the transmission lines reached the acceptable load-transfer limits, which limited the use of available generation capacity from outside the region. The control of reactive power became critical.

The net effect of the demand for additional reactive power was to increase the load current and line voltage drops, reducing voltage at the customer terminals. Local control of voltage by means of auto transformers resulted in increasing line current and diminishing voltage levels, a deteriorating condition that continued until the low voltage caused the generators to shut down, leading to a complete system blackout.

Not the Usual Causes

This system blackout occurred without any unusual system equipment failures or adverse weather conditions such as lightning or a hurricane. The existing generation was sufficient to supply the system demand, and the transmission line capacity was much higher than required. The transmission system was equipped with automatic reactive power and a voltage control system that operated correctly. Nevertheless, the combined system operational characteristics were able to cause a major system blackout.

The reactive power compensation is location dependent, as the transmission of reactive power over long distances is not only uneconomical but also can be ineffective for voltage control. Reactive power flow involves the generation of additional reactive power. In normal operating conditions, the control system is able to manage and adjust the generation of reactive power in power plants to the voltage. If the location of reactive power sources is remote and inadequate, this form of system control can result in a system blackout.

To ensure the appropriate standards of system reliability are maintained, TSOs that do not own generation facilities have to achieve their voltage control task by contracted reserves and generation-load balancing. As customers are free to change their power demand at any time, the balancing requires a request for a change in the generation. In a market economy, all such contracts are profit oriented. On one hand, the reliability of supply in a market environment is more important because of customer expectations, but on the other hand, it is more difficult and more expensive to achieve.

The Search for a Solution

The events in June 2006 led to the availability of funds and a dramatic search to find a solution to avoid a recurrence. The Polish TSO was able to take the necessary steps to improve the control of voltages and reactive power in its transmission system. The search concluded when the TSO participated in a project led by EPRI. The project aimed to develop a highly automated method for the identification of areas prone to voltage instability in practical power system models.

For a wide range of system conditions and contingencies, the model sought had to identify regions, substations or buses at risk for voltage instability. The model also had to indicate areas where the existing generating capacity was evaluated as critical as well as areas where the level of reactive power resources required to maintain reliable system operation. The approach adopted for this research project was based on the power-voltage curve method combined with modal analysis.

The framework for this project can be summarized in the following tasks:

  • Define a system operating space based on a wide range of system load conditions, dispatch conditions and defined transactions (source-to-sink transfers).

  • Define a large set of contingencies that spans the range of credible contingencies.

  • Use power-voltage curve methodology; push the system through every condition, under all contingencies, until voltage instability is reached.

  • At the point of instability (the nose of the power-voltage curve for each case), perform modal analysis to determine the critical mode of instability as defined by a set of bus participation factors corresponding to the zero eigenvalue.

  • Store the results of the modal analysis in a database for analysis using data-mining techniques to identify the critical areas and track them throughout the range of system changes.

  • Establish the reactive reserve requirements for each identified critical area.

Results of the Research Project

The project for the Polish TSO started at the end of 2007, and a few months work with EPRI was needed to establish the right approach to apply the theoretical method for this particular system. TSO staff regularly attended EPRI project meetings, contributing to the development of the tool for the identification of voltage problems and lack of reactive power. Finally, in June 2008, the EPRI experts visited Warsaw, Poland, and the newly developed software was implemented on the Polish transmission system.

The computation was applied to a detailed model of the Polish power grid. As the result, several suspect areas were identified, and the reactive power needed in each of these areas for safe operation was evaluated. Based on the results of these calculations, the TSO began installing static VAR compensators. By August 2009, the TSO had commissioned five of the predicted seven reactive power compensators. The remaining two units will be installed in 2011. Since the compensators have been installed, operation of the transmission system has improved and no voltage problems have been experienced.

The framework developed in this research project has been successfully implemented in a pre-commercial-grade software tool and has been tested on several other large power systems with satisfactory results. To date, this software tool has been developed and used primarily for off-line planning studies, but in the future, it will be extended to on-line control center applications.

Grzegorz Blajszczak ( was awarded a master's degree from Warsaw University of Technology in 1984 and a PhD from the Hungarian Academy of Science in 1990. During his career, Blajszczak has been engaged on research on power quality at Warsaw University of Technology, Budapest Technical University and Rand Afrikaans University of Johannesburg. In 1994, he was appointed foreign relations manager in Energoprojekt-Warszawa, and in 1996, he joined Westinghouse Electric Europe as vice director. Blajszczak, who has been an IEEE member sice 1992, joined PSE Operator in 1999, where he is currently the supervisor for power quality.

Companies mentioned:

Electric Power Research Institute

PSE Operator