Except for a few special applications (e.g. so-called pulse-combustors), combustion oscillations constitute an undesirable phenomenon which has an adverse impact on the service life, performance capability and environmental impact of a combustion system. There is therefore great interest in preventing instabilities of this kind or x96 if oscillation problems have already been occurring x96 to suppress these altogether or to reduce them to a tolerable level. The measures employed in this application can be sub-divided into two main groups: passive and active methods.

Passive methods

Broadly speaking, this applies to systems which do not possess any x93intelligencex94 in the form of a control system, i.e. whose influence on oscillations in a combustion system cannot be influenced and controlled by a specific system parameters. To put it in even more simple terms, passive methods are characterized by the fact that they do not require any power from an external source.

The following section provides a few examples of passive methods with a brief outline of their operating principle as well as their advantages and disadvantages:

• Damping elements and mufflers: Acoustic energy is dissipated in the form of heat. Do not prevent oscillations but can reduce the amplitudes of sound pressure. Cost-effective standard components, but often bulky and associated with additional pressure losses.
• Helmholtz and Lambda/4-resonators : Have an adverse impact on the resonance sound field of combustion oscillation, preventing this field from forming properly. Effectiveness restricted to certain frequencies, very large dimensions for low frequencies.
• "Baffles" : Prevent noise from spreading (e.g. in the combustion chamber) and therefore prevent the resonance sound field from forming. Simple design but vulnerable to burning out, and often unfavourable for the actual combustion process, additional pressure losses.
• "Wrong tuning" of the system : the eigenfrequencies of the combustion system are altered by geometrical changes. Without significant disadvantages for combustion management and efficiency, but usually very costly and elaborate in terms of design. Effectiveness also often restricted to individual frequencies, resulting in the risk of oscillations only altering their frequency.

Especially as a result of their restricted area of effectiveness and the disadvantages already given for passive methods, active methods constitute an interesting alternative, in particular for systems which operate under very changeable conditions.

Active methods

With active suppression of self-excited combustion oscillations, or "Active Instability Control (AIC)", the status of the system is monitored by sensors and is directed to a control system which x96 depending on the input signal x96 intervenes in system behaviour by means of an actuator, thus preventing oscillations from occurring. A simple example might look like this: A pressure sensor in the combustion system measures pressure fluctuations as they arise, from which the controller logic calculates an activation signal for a loudspeaker. Depending on this activation signal, the loudspeaker may generate an anti-sound field which is superimposed on the sound field in the system, thereby eliminating it.

Unfortunately, AIC is not as simple as this under practical conditions. Many different aspects need to be taken into account and the excitation mechanism underpinning the self-excited combustion oscillations needs to be investigated carefully before a suitable AIC strategy can be decided upon. Selection of sensors and actuators depends, in keeping with the correct controller structure, to a major extent on the combustion system involved (power, fuel, geometry), the frequencies involved and the existing sound fields. The additional cost associated with active instability control is however justified by the advantages this approach achieves:

• Low level of design intervention in the combustion system and not much space required. Implementation at a later date is relatively straightforward.
• Combustion power and combustion management are not adversely affected, no additional losses.
• Pollutant emissions and incompletely combusted components are actually reduced to some extent.
• The actual control process can respond flexibly to the actual system characteristics and changes within these. Intervention only takes place when and for as long as it is required.
• Following stabilization by the AIC, the controlled back actuator signal is lowered, causing a reduction in power intake by the actuator and therefore extending the service life of the actuator.