T 2331/10 (Operating wind turbines / GENERAL ELECTRIC COMPANY) 15-12-2017

Reasons for the Decision

1. Non-attendance at the oral proceedings

As announced in its response to the summons and the communication of the Board, the appellant did not attend the oral proceedings. In accordance with Rule 115(2) EPC and Article 15(3) RPBA, oral proceedings were held without the appellant. By deciding not to attend oral proceedings, the appellant has chosen not to take the opportunity to comment on the Board's opinion at the oral proceedings. In the present case, the duly summoned appellant has to be treated as relying only on its written case.

2. The invention

2.1 The invention is a wind farm monitoring and forecasting network (100) for the monitoring of a wind farm having a plurality of wind turbine generators (WTGs). The invention addresses the problem of handling deficiencies in electric power delivered to a local region where the wind farm is embedded due to variations in the electric power output of such wind farms, page 1, 2nd paragraph.

2.2 The wind farm monitoring and forecasting network (100) is coupled to a production forecasting system (200) which generates an electric power production forecast data signal (202) for the external electric power dispatchers. The forecasting is based on measurement data signals, which are the wind-farm's operating parameters (101, 107) and its meteorological conditions (105). This data is collected by a supervisory control and data acquisition system (SCADA, 110) and is transferred to the wind production forecasting system (200).

2.3 The wind production forecasting system (200) employs a regional weather forecast model (216) to generate wind farm specific meteorological forecast data signals (222). A turbine specific transfer function (232) then converts the wind-farm specific data into turbine-specific data and a power curve model (236) transfers this data into turbine-specific electric power production forecast signals (238). An availability model (240) with turbine-specific algorithms corrects the forecast based on historical performance data of the turbines to generate the production forecast (202).

3. Article 84 EPC (main request)

3.1 Claim 1 is not clear and is not supported by the description (Article 84 EPC) for the following reasons.

3.2 Firstly, the claim defines the electronic data communi-cation device (104) "for receiving measurement data signals (101, 105, 107) from monitoring devices of the WTGs and operational data signals (109) derived from an operational database (226)". The latter refers to wind production forecasting (WPF) database (226) which is part of wind production forecasting system (200).

According to the description, page 5, lines 7 to 21, and Figure 1, operational data signals (109) correspond to the measurement data signals (101, 105, 107) which are received by the electronic data communication device (104) and are retransmitted to SCADA (110). The SCADA system manipulates and stores this data (see page 6, third paragraph, to page 7, first paragraph), before transmitting it in terms of SCADA data signals (122) to wind production forecasting system (200). These SCADA data signals may include specific operational data which may then be stored in operational database (226), but there is no support that operational data is derived from the operational database (226) and sent back to the electronic data communication device (104).

3.3 Secondly, the wind production forecasting system (200), as presently claimed, comprises a "meteorological forecast input channel" which receives wide-area meteorological forecast data, which is received by a "resident narrow-area meteorological forecast module" to generate narrow-area meteorological forecast data. The narrow-area meteorological forecast data and the electric power generation availability of a "resident electric power generation availability module" are received by a "resident electric power production forecast module" which generates the electric power production forecast data signal (202).

3.4 At the same time, the claimed wind production foreca-sting system (200) comprises a "wind forecasting sub-system", an "electric power production forecast sub-system" and a "wind production forecasting database" which are coupled to each other, but not to the above-mentioned modules of the same system. These generate narrow-area wind turbine specific wind speed data signals from wide-area wind farm specific meteorological forecast wind speed data signals, but do not contribute to the generation of the electric power production forecast data signal (202).

3.5 Thus, the wind production forecasting system (200) comprises features which do not interact at all and, if so, they operate and exchange data in an unclear manner. It is not clear how the narrow-area meteorological forecast data and the electric power generation availability fit together to generate the final electric power production forecast signal (202). The claimed interaction and data exchange is unclear and also not supported by the description. There is no clear mapping between the claimed features and the description and figures.

3.6 The wind production forecasting system (200), as presently claimed, defines generating "at least one electric power production forecast data signal (202)" which is clearly unspecific for a particular turbine. The appellant's argument that there would be such a turbine specific signal was therefore not convincing.

3.7 Finally, another lack of support results from the definition of the forecasting system (200) in terms of "algorithms" which transmit, receive and process data signals. Such a system is not supported by the description, page 9, 2nd paragraph, to page 12, first paragraph, and page 15, second paragraph, which do not use the term "algorithm" to describe the system (200). The term "algorithm" is found only on pages 13 and 14, 3rd paragraph, but in combination with a method of operating the wind farm (102) which is different from system (200).

3.8 Accordingly, claim 1 is not clear and is not supported by the description (Article 84 EPC).

4. Article 84 EPC (auxiliary request)

4.1 Claim 1 of the auxiliary request further includes features from claims 2 to 4 of the main request. Therefore the same clarity objections raised against claim 1 of the main request apply to claim 1 of this request.

4.2 The claim defines the "wind forecasting sub-system" as comprising an "availability model including tunable parameters". The term "tunable parameters" is not clear, because any system comprises tunable parameters. This adds a further clarity problem. Moreover, the provision of an availability model is not technical.

4.3 Accordingly, claim 1 of the auxiliary request is not clear and is not supported by the description (Article 84 EPC).

5. Article 56 EPC (main request)

5.1 As far as can be understood from the present wording, at least claim 1 of the main request also does not involve an inventive step (Article 56 EPC).

5.2 A wind farm monitoring and forecasting network (100) are well-known in the art of wind farms, for example D1. The difference resides in the claimed forecasting production system (200) which converts wide-area forecast data into narrow-area forecast data to achieve a more accurate forecasting model. The distinguishing features refer to the modeling of data, the transformation of data signals by the use of algorithms and data models, which in themselves do not achieve a technical effect and cannot contribute to an inventive step.

5.3 This is part of a non-technical problem which is to produce a more accurate production forecast, which per se is an intellectual activity, following from T 49/99, (cf. decision, paragraphs 7 and 8). The technical part of the solution is self-evident; it is a straightforward implementation of a non-technical modelling and forecasting process using commonplace and well-known computer technology. D1 already discloses, at paragraph 60, the provision of forecasts based on wind speed, direction, and plan output.

5.4 Also, the predicted forecast data signal is not a physical variable of the underlying system, such as was the case in T 1227/05 where the electric voltage of the circuit was simulated. Recent decision T 625/11, following the line of T 1227/05, considered the determination of a "limited value representing a parameter" for the control of a nuclear power plant to confer technical character to the claim. The nature of the parameter was considered to be intrinsically linked to the functioning of the nuclear power plant and therefore technical. The Board is of the opinion that the situation of T 625/11 does not apply in the present case, because the production forecast data signal is not intrinsically linked to the functioning of the wind farm in order to control it.

5.5 The purpose of this forecast data signal appears to be rather to make sales of electric power generation with increased confidence (application, page 16, last paragraph), which is a non-technical objective. The Board is also of the view that there is no control of a "physical process" based on a mathematical model, as stated in T 953/94, because the production forecast is mentioned in the application to be used to facilitate revenue generation.