No More Blackouts!

Alstom’s innovative grid control solutions-

To make the grid smarter, one of the most exciting challenges is designing and implementing completely new monitoring and control applications in order to detect and mitigate cascading outages. Blackouts occur when a series of unplanned events combine to stress the grid to breaking point. No longer able to remain intact, the grid splits into a number of “electrical islands”; some, or all, of these islands may collapse, thus causing a large-scale loss of electricity supply to customers.

The series of large-scale blackouts in the region, which occurred back in May, plunged the cities in the Philippines, Thailand, and Vietnam into darkness for varying periods of time. As is common in many blackouts, there were triggering events, followed by several compounding factors. In the case of Thailand, a combination of a lightning strike and a faulty high voltage cable, which were further exacerbated by dynamic effects in the power system, eventually resulted in a blackout across its 14 southern provinces including popular tourist hotspots like Hat Yai, Koh Samui, and Phuket.

Alstom’s risk mitigation and protection strategies

For more than a decade, Alstom, through its grid sector, has striven to mitigate the occurrences of disastrous blackouts. It has been developing powerful methods of analysis and robust applications, which are embedded into its grid control solutions. Its methodology encompasses a dual perspective:

Reactivity, derived from advanced online monitoring solutions, is targeted at instantaneously mitigating and fixing the effects of a local or regional disturbance.
Proactivity, an innovative offline application derived from advanced statistical methods of analysis, aims to characterise the actual capacity of the grid to resist a local or regional disturbance, at present, or in the years to come.

These complementary solutions that make extensive use of advanced data-mining analysis aim to provide grid operators and asset managers with an unprecedented understanding of the actual performance of their system, under the ever more demanding criteria of grid efficiency, system reliability, and risk management. The combination of both solutions is currently being implemented and deployed in a major 500/220 kV system in South America.

Online monitoring

“A main goal is to bridge the divide between fast local control (mainly protection) and slow wide area control (like regional voltage) to accomplish fast wide area control.” Professor A. Bose, Washington State University.

Grid analytics refers to the analysis processes that are applied to the information acquired from a power system. It is used to extract key performance and risk indicators, which are then applied to assess dynamic behaviour, stress, and measurable threats to the power system. These processes, applied in real time to incoming data, are employed for both real-time grid management and subsequent review and analysis. The concept is crucial in smart grid development.

Synchrophasor measurements are used to develop a robust infrastructure and applications are used to extract key information out of large quantities of measured data. This information is provided to users in real time in the control room and to analysts for post-event use.

In particular, measurement-based applications focus on extracting information on the dynamic performance of the grid, including oscillations, disturbances, and stability. The systems are used for increasing transfer levels and issuing early warnings of emerging threats to the stability of the grid. If large disturbances do occur, real-time information can be used to improve the response and restoration of the system.

While monitoring and real-time analysis can improve the supervision of a grid, wide area control solutions can also bolster the system’s defence against large disturbances. Compared to the traditional system integrity protection, the new approaches are more sensitive to the state of the grid, as they will limit their action to the extent that is needed, hence reducing the dangers associated with an overresponse. Synchrophasor-based defence schemes can thus be considered to offer a new generation of self-healing mechanisms for the grid.

Offline analysis

Early warning signs of degraded system performance, combined with clear visualisation of the stresses in the power system, are extremely helpful for averting a blackout, or reducing the impact of a large disturbance. Many large disturbances take time to develop to a stage in which a triggering event would result in a cascading failure. With the correct warning signs and restoration support in place, intervention at an early stage to rectify the problem is possible.

Improved models of the assets and their utilisation, coupled with the increased collection of ubiquitous measurements across the grid, should help to enhance the whole science of asset management. Offline analysis works by using large archives of measured data, or recorded incidents, to extract key information from behavioural patterns and threats.

Self-organised criticality

The topological description of the power system is enhanced with the introduction of a new vision that is derived from the “self-organised criticality” (SOC) concept. Originally known through the sandpile model, the real value of this concept stems from the fact that it offers a new interpretation of the overall behaviour of a power grid.

By taking into account both the power grid and the engineering actions that drive its operations, it is possible to quantify the maximum stress that a network can undergo and deduce all statistical properties in terms of technical and economic risks. As for thermal voltage drops, or stability limitations, it is necessary to introduce a new condition – the SOC limitation. It enables operators to quantify the marginal distribution of events against the identification of long-term, or persistent, correlation phenomena.

SOC can provide unprecedented features of statistical analysis capable of accounting for the true conditions of ageing, recurrent modes of failures, and consequences of inadequate maintenance policies. It also includes resilience to external factors such as seismic activity or human error.

The offline method of analysis is embedded into the power flow module of grid analysts. Through these analysis processes, analysts can obtain unpredicted information, in terms of actual resistance, monitor the degradation process, or even determine the risk of experiencing a cascading outage. Currently, no other conventional method of analysis has been able to perform any of the aforementioned functions.

Strategic decision-making

Alstom has studied a large number of power systems worldwide, ranging from small island networks to the largest interconnections. The experiences of different power systems have been valuable by helping to ground the solutions to physical problems, which can be applied to many different situations.

These two methods of online (synchrophasors) and offline (SOC) innovative monitoring solutions have been in place for several years at a major transmission system operator in South America for its 500/220 kV network. They provide the grid operator, asset managers, and market analysts with unprecedented understanding and reactivity to ensure grid efficiency and improve reliability performance criteria. They also solidly contribute to the strategic decisions of planning and maintenance strategies within the organisations. Thus, regional government bodies may want to consider employing these advanced solutions to ensure the stability of their power supply in the future.

About the Authors

Frédéric Heliodore