Effective and timely health monitoring of wind turbine gearboxes and generators is essential to reduce the costs of operations and maintenance activities, especially offshore. This paper presents a scalable and lightweight convolutional neural network (CNN) framework using high-dimensional raw condition monitoring data for the automatic detection of multiple wind turbine electromechanical faults. The proposed approach leverages the potential of combining information from a variety of signals to learn features and to discriminate the types of fault and their severity. As a result of the CNN layers used to extract features from the signals, this architecture works in the time domain and can digest high-resolution multi-sensor data streams in real-time. To overcome the inherent black-box nature of AI models, this research proposes two interpretability techniques, multidimensional scaling and layer-wise relevance propagation, to analyse the proposed model's inner-working and identify the signal features relevant for fault classification. Experimental results show high performance and classification accuracies above 99.9% for all fault cases tested, demonstrating the efficacy of the proposed fault-detection system.

In-service turbine monitoring is essential for maximizing the wind energy contribution to the global energy budget. Measurement of turbine shaft torque under transient wind conditions is fundamental to develop reliable condition monitoring techniques. Contact based measurements bring their own disadvantages and non-contactless measurements have many potential advantages. However, their performance needs to be validated against standard methods. This paper focuses on the development of an enhanced transient Feature Selective Validation (FSV) techniques to undertake this analysis with an emphasis on transient data processing. The nature of FSV makes it a natural technique to consider for this problem space. Open questions have existed as to how transients should be dealt with in FSV. This paper overcomes the limitations of previous approaches for step-function transient comparison and presents analytical methods to ensure the transient feature itself is considered, irrespective of how much pre- and post- transient data happens to be included.

This paper presents a simplified automated fault detection scheme for wind turbine induction generators with rotor electrical asymmetries. Fault indicators developed in previous works have made use of the presence of significant spectral peaks in the upper sidebands of the supply frequency harmonics; however, the specific location of these peaks may shift depending on the wind turbine speed. As wind turbines tend to operate under variable speed conditions, it may be difficult to predict where these fault-related peaks will occur. To accommodate for variable speeds and resulting shifting frequency peak locations, previous works have introduced methods to identify or track the relevant frequencies, which necessitates an additional set of processing algorithms to locate these fault-related peaks prior to any fault analysis. In this work, a simplified method is proposed to instead bypass the issue of variable speed (and shifting frequency peaks) by introducing a set of bandpass filters that encompass the ranges in which the peaks are expected to occur. These filters are designed to capture the fault-related spectral information to train a classifier for automatic fault detection, regardless of the specific location of the peaks. Initial experimental results show that this approach is robust against variable speeds and further shows good generalizability in being able to detect faults at speeds and conditions that were not presented during training. After training and tuning the proposed fault detection system, the system was tested on “unseen” data and yielded a high classification accuracy of 97.4%, demonstrating the efficacy of the proposed approach.

In MW-sized wind turbines, the most widely-used generator is the wound rotor induction machine, with a partially-rated voltage source converter connected to the rotor. This generator is a significant cause of wind turbine fault modes. In this paper, a harmonic time-stepped generator model is applied to derive wound rotor induction generator electrical & mechanical signals for fault measurement, and propose simple closed-form analytical expressions to describe them. Predictions are then validated with tests on a 30 kW induction generator test rig. Results show that generator rotor unbalance produces substantial increases in the side-bands of supply frequency and slotting harmonic frequencies in the spectra of current, power, speed, mechanical torque and vibration measurements. It is believed that this is the first occasion in which such comprehensive approach has been presented for this type of machine, with healthy & faulty conditions at varying loads and rotor faults. Clear recommendations of the relative merits of various electrical & mechanical signals for detecting rotor faults are given, and reliable fault indicators are identified for incorporation into wind turbine condition monitoring systems. Finally, the paper proposes that fault detectability and reliability could be improved by data fusion of some of these electrical & mechanical signals.

Non-intrusive, reliable and precise torque measurement is critical to dynamic performance monitoring, control and condition monitoring of rotating mechanical systems. This paper presents a novel, contactless torque measurement system consisting of two shaft-mounted zebra tapes and two optical sensors mounted on stationary rigid supports. Unlike conventional torque measurement methods, the proposed system does not require costly embedded sensors or shaft-mounted electronics. Moreover, its non-intrusive nature, adaptable design, simple installation and low cost make it suitable for a large variety of advanced engineering applications. Torque measurement is achieved by estimating the shaft twist angle through analysis of zebra tape pulse train time shifts. This paper presents and compares two signal processing methods for torque measurement: rising edge detection and cross-correlation. The performance of the proposed system has been proven experimentally under both static and variable conditions and both processing approaches show good agreement with reference measurements from an in-line, invasive torque transducer. Measurement uncertainty has been estimated according to the ISO GUM (Guide to the expression of uncertainty in measurement). Type A analysis of experimental data has provided an expanded uncertainty relative to the system full-scale torque of ±0.30% and ±0.86% for the rising edge and cross-correlation approaches, respectively. Statistical simulations performed by the Monte Carlo method have provided, in the worst case, an expanded uncertainty of ±1.19%.

This paper presents a discussion of the development of wind energy generation in the United Kingdom and the challenges faced by the wind industry including reliability, performance and condition monitoring, particularly in the offshore environment. The worldwide installed capacity of offshore wind has now risen to over 7 GW, with an ever increasing deployment rate of new assets. About 90% of the global currently installed capacity is in Northern Europe, with the United Kingdom having the world's largest share at 4 GW. Capacity factor data from UK offshore wind farms is presented, providing an insight into the current performance of large Round 2 offshore wind farms compared to the earlier Round 1 farms and to onshore farms. The data reveal that the United Kingdom's Round 2 offshore farms are achieving an average monthly capacity factor of 38.3% with a peak value of 75.8%. The older Round 1 farms have a lower average capacity factor of 33.6% while large onshore farms with capacities above 100 MW have achieved 25.6%. Offshore wind turbine performance has improved over time, and the industry is applying the learning from early experiences to achieve better performances at the more recently installed farms. Despite these improvements in turbine availability, the cost of energy from wind, particularly offshore, remains too high for it to be a commercially viable form of generation without subsidies. Reducing the cost of energy from wind to economically sustainable levels is the most important challenge facing the industry today. Operation and maintenance costs constitute up to 30 % of the total cost of energy from wind in large farms. The industry must overcome the challenges associated with improving component reliability and the development and adoption by operators of appropriate condition monitoring systems and maintenance strategies, in order to reduce costs to sustainable levels. Research and development work carried out with these goals in mind is also reviewed in the paper.

Wind Turbine (WT) global installed capacity is expected to increase from 318GW to 596GW between 2013 and 2019, with an increasing proportion being from offshore wind farms. With up to 70% of Operations and Maintenance (O&M) costs coming from unplanned maintenance, the adoption of cost effective condition monitoring (CM) techniques is crucial for competitive development of offshore wind. Monitoring the torque of a WT can provide much information about the WT's health and it has been shown to be successful in the detection of faults in the main drive train components. Although WT torsional effects are important, torque measurement on such a large, low speed, inaccessible machine is practically and logistically difficult, although it is possible using costly specialised intrusive in-line equipment. This paper presents the development of a non-intrusive method for monitoring the drive train torque using timing differences between optical probe measurements along a shaft. An algorithm has been developed and initially verified using a simulated WT for speed and torque data. The algorithm torque was accurate to within ±3% of the input. The initial performance of the proposed technique has been successfully tested experimentally under both steady and transient torque conditions. Experimental results show good agreement between the algorithm predictions and the measurements. The proposed algorithm successfully detects changes in shaft speed and torque, with the torque mean percentage error within 16-25%. Once implemented on a WT drive train, the proposed non-intrusive method can overcome the majority of problems limiting the industrial application of CM systems (CMSs) based on shaft torque measurements.

Wind turbine condition monitoring is gaining in importance as operators and developers move towards larger, further offshore, less accessible wind farms. However, condition monitoring is made highly challenging by the variable speed, variable load nature of wind turbines. Electrical faults, in the form of brush gear or slip ring damage contribute significantly to downtime yet operators have little experience of detecting these faults in the field. This paper subjects a physical test rig to rotor electrical unbalances and applies a frequency tracking algorithm to mechanical and electrical monitoring signals to compare the sensitivity of the various signals. The results shows that the total electrical power signal gives the clearest response to rotor electrical unbalance, as expected, but that torque measurements could be a viable alternative. Speed signal analysis showed much lower sensitivity to unbalance than torque or power measurements.

Improving the availability of wind turbines is critical for minimising the cost of wind energy, especially offshore. The development of reliable and cost-effective gearbox condition monitoring systems (CMSs) is of concern to the wind industry, because the gearbox downtime has a significant effect on the wind turbine availabilities. Timely detection and diagnosis of developing gear defects is essential for minimising an unplanned downtime. One of the main limitations of most current CMSs is the time consuming and costly manual handling of large amounts of monitoring data, therefore automated algorithms would be welcome. This study presents a fault detection algorithm for incorporation into a commercial CMS for automatic gear fault detection and diagnosis. Based on the experimental evidence from the Durham Condition Monitoring Test Rig, a gear condition indicator was proposed to evaluate the gear damage during non-stationary load and speed operating conditions. The performance of the proposed technique was then successfully tested on signals from a full-size wind turbine gearbox that had sustained gear damage, and had been studied in a National Renewable Energy Laboratory's (NREL) programme. The results show that the proposed technique proves efficient and reliable for detecting gear damage. Once implemented into the wind turbine CMSs, this algorithm can automate the data interpretation, thus reducing the quantity of the information that the wind turbine operators must handle.