The appeal stems from an Opposition against an invention relating to noise and vibration in wind turbines by using the processing unit that analysed detected noise/vibrations with reference characteristics from simulations and offline testing done using the generator’s rotational frequency and torque as input.

The Opponent argued that the processing unit was not sufficiently described, and using the specific input for simulation was obvious.

The Board agreed with the Patentee and noted that the gaps in disclosure would be filled using the common knowledge of the skilled person. 

Calculating the reference characteristic of the noise/vibration component based on a simulation or offline test using a rotational frequency and the torque of the generator as input is considered technical. Since the prior art did not indicate that rotational frequency and torque are set, let alone used to calculate based on simulation or offline test, the subject-matter was considered inventive.

Here are the practical takeaways from the decision T 2048/21 of February 9, 2024, of the Technical Board of Appeal 3.5.02

Key takeaways

The person skilled in the art would be able to fill these gaps in the disclosure of the patent taking into account the level of common general knowledge of the person skilled in the art. The common general knowledge of the skilled person includes profound knowledge of the implementation details of the control system and its components.

The relevant information [in the application] is normally limited to a schematic representation of the required processing blocks. Unless otherwise described, their technical implementation requires the use of standard components and/or algorithms, the existence and operation of which are fully known to the skilled person, and to which the skilled person can also make any necessary adaptations, at least within the limits of what is generally known to them.

Calculating the reference characteristic of the noise/vibration component based on a simulation or offline test using a rotational frequency and a torque of the generator as input is technical

The invention

This invention addresses the problem of noise and vibration in wind turbines, particularly those using permanent magnet synchronous generators (PMSG). Conventionally, this problem may be approached in the machine design phase. However, due to manufacturing limitations, a noise/vibration-free machine may not be possible. The invention provides a feedback control method with respect to the d-component of the current, allowing the drive to operate at reduced noise and vibration levels. The invention uses a noise/vibration detection module to measure the noise and vibrations in the wind turbine. A processing unit to analyse the detected noise/vibrations to understand their actual characteristics and compare them with reference characteristics. Finally, the reference determination module determines the reference noise/vibration characteristics from simulations and offline testing using the generator’s rotational frequency and torque as input.

  • ”Claim

Is it patentable?

The Opponent argued insufficiency of disclosure ( Art 83 EPC), particularly the information about the module for calculation of the reference current. The Board noted as follows:

1.9 The board further agrees with the appellant to the extent that the patent does not contain a detailed description of exactly which processes are carried out in the noise/vibration processing module 423 and how it is configured. The patent merely discloses in paragraph [0093] that “a suitable signal processing algorithm may be required”.

1.11 However, in the board’s view, the appellant’s conclusion that the person skilled in the art would not have been able to fill these gaps in the disclosure of the patent does not appropriately take into account the level of common general knowledge of the person skilled in the art as defined in point 1.8 above. The common general knowledge of the skilled person includes profound knowledge of the implementation details of the control system and its components, including those of the processing means required to make measured or estimated signals comparable with other types of signals (such as the reference characteristic 429), as well as technical details of the control module 437 that are necessarily implemented to calculate the Id current command 407 (first type of d-component) on the basis of a difference signal 433.

1.12 Furthermore, it is noted that the level of detail which the appellant has objected as being lacking in the patent is not normally found in patents or patent applications in this technical field. This understanding is also supported by the disclosure of documents E7 and E11. The relevant information is normally limited to a schematic representation of the required processing blocks. Unless otherwise described, their technical implementation requires the use of standard components and/or algorithms, the existence and operation of which are fully known to the skilled person, and to which the skilled person can also make any necessary adaptations, at least within the limits of what is generally known to them.

The patent contains nothing that could lead the skilled person to assume that the components required to carry out the invention and their configuration in the present case involve non-standard means which the skilled person does not know or cannot (re)configure.

1.13 Consequently, it can be reasonably assumed that the skilled person in the relevant technical field of the present invention is sufficiently familiar with the technical details required to implement the relevant processing components in practice…

The Board then addressed the objection to the lack of inventive step over document E11 (EP 2 552 012 A1):

3.2 The subject-matter of claim 1 at least differs from document E11 in feature 1.3.2.1, according to which the reference determination module is adapted to calculate the reference characteristic of the noise/vibration component based on a simulation or offline test using a rotational frequency and a torque of the generator as input.

3.7 The objective technical problem arising on the basis of document E11 and in the light of the distinguishing feature was identified by the appellant as the implementation of the “specified values” mentioned in paragraph [0071] of document E11.

The Opponent argued that the feature was disclosed in document D7 (EP 2 485 388 A1), which was rejected by the Board. Subsequently, the Opponent alleged it was obvious in view of the common general knowledge. However, the Board sided with the Patentee:

3.12 The board agrees with the respondent that paragraph [0071], in view of the use of “specified values” not equal to zero, alone cannot be considered as sufficient motivation for the skilled person to provide for a calculation of reference characteristics according to feature 1.3.2.1. In particular, the respondent has rightly argued that the “specified values” to be used according to paragraph [0071] in E11 could also simply be known to the skilled person in view of the design of the respective generator and set accordingly. There is therefore not even a clear hint in E11 for the skilled person to consider the rotational frequency and a torque to set the “specified values”, let alone to carry out a calculation based on simulation or offline test.

Furthermore, the multi-step considerations put forward by the appellant, which allegedly lead the skilled person to feature 1.3.2.1 on the basis of document E11 in combination with the common general knowledge of the skilled person, already demonstrate that it was not possible for the skilled person to arrive at the subject-matter of claim 1 without further elaboration.

Moreover, the appellant’s argumentation is flawed by the fact that feature 1.3.2.1 does not claim a mere calculation of the reference characteristic “offline”, but by “offline test”, which is not the same thing. Thus, even if the reference characteristics (i.e. the non-zero “specified values” in paragraph [0071]) had been calculated offline, this does not mean that the skilled person would have calculated them on the basis of an offline test using the rotational frequency and torque of the generator.

Therefore, the Board agreed that the subject-matter was inventive over the cited document.

More information

You can read the full decision here: T 2048/21 of February 9, 2024, of the Technical Board of Appeal 3.5.02

Stay in the loop

Never miss a beat by subscribing to the email newsletter. Please see our Privacy Policy.

Privacy policy *
* = Required field

Please share this article if you enjoyed it!