Ilmar Santos

Wind turbine condition monitoring based on SCADA data using normal behavior models. Part 1: System description

This paper proposes a system for wind turbine condition monitoring using Adaptive Neuro-Fuzzy Interference Systems (ANFIS). For this purpose: (1) ANFIS normal behavior models for common Supervisory Control And Data Acquisition (SCADA) data are developed in order to detect abnormal behavior of the captured signals and indicate component malfunctions or faults using the prediction error. 33 different standard SCADA signals are used and described, for which 45 normal behavior models are developed. The performance of these models is evaluated in terms of the prediction error standard deviations to show the applicability of ANFIS models for monitoring wind turbine SCADA signals. The computational time needed for model training is compared to Neural Network (NN) models showing the strength of ANFIS in training speed. (2) For automation of fault diagnosis Fuzzy Interference Systems (FIS) are used to analyze the prediction errors for fault patterns. The outputs are both the condition of the component and a possible root cause for the anomaly. The output is generated by the aid of rules that capture the existing expert knowledge linking observed prediction error patterns to specific faults. The work is based on continuously measured wind turbine SCADA data from 18 turbines of the 2 MW class covering a period of 30 months. The system proposed in this paper shows a novelty approach with regard to the usage of ANFIS models in this context and the application of the proposed procedure to a wide range of SCADA signals. The applicability of the set up ANFIS models for anomaly detection is proved by the achieved performance of the models. In combination with the FIS the prediction errors can provide information about the condition of the monitored components. In this paper the condition monitoring system is described. Part two will entirely focus on application examples and further efficiency evaluation of the system.

Using Data-Mining Approaches for Wind Turbine Power Curve Monitoring: A Comparative Study

Four data-mining approaches for wind turbine power curve monitoring are compared. Power curve monitoring can be applied to evaluate the turbine power output and detect deviations, causing financial loss. In this research, cluster center fuzzy logic, neural network, and k-nearest neighbor models are built and their performance compared against literature. Recently developed adaptive neuro-fuzzy-interference system models are set up and their performance compared with the other models, using the same data. Literature models often neglect the influence of the ambient temperature and the wind direction. The ambient temperature can influence the power output up to 20%. Nearby obstacles can lower the power output for certain wind directions. The approaches proposed in literature and the ANFIS models are compared by using wind speed only and two additional inputs. The comparison is based on the mean absolute error, root mean squared error, mean absolute percentage error, and standard deviation using data coming from three pitch regulated turbines rating 2 MW each. The ability to highlight performance deviations is investigated by use of real measurements. The comparison shows the decrease of error rates and of the ANFIS models when taking into account the two additional inputs and the ability to detect faults earlier.

Frequencies in the Vibration Induced by the Rotor Stator Interaction in a Centrifugal Pump Turbine

The highest vibration levels in large pump turbines are, in general, originated in the rotor stator interaction (RSI). This vibration has specific characteristics that can be clearly observed in the frequency domain: harmonics of the moving blade passing frequency and a particular relationship among their amplitudes. It is valuable for the design and condition monitoring to count on these characteristics. A CFD model is an appropriate tool to determine the force and its characteristics. However, it is time consuming and needs highly qualified human resources while usually these results are needed immediately and in situ. Then, it is useful to determine these characteristics in a simple, quick, and accurate method. At present, the most suitable method indicates a large amount of possible harmonics to appear, without indicating the relative importance of them. This paper carries out a theoretical analysis to predict and explain in a qualitative way these frequencies and amplitudes. The theoretical analysis incorporates the number of blades, the number of guide vanes, the RSI nonuniform fluid force, and the sequence of interaction. This analysis is compared with the method currently in use, and both methods are applied to a practical case. The theoretical analysis gives a resulting force over the pump turbine, which corresponds well to the measured behavior of a pump turbine in terms of its frequencies and the relationship between their amplitudes. A corrective action is proposed as a result of the analysis and after it is carried out in one of the units, the vibration levels are reduced. The vibration induced by the RSI is predicted considering the sequence of interaction and different amplitudes in the interactions between the same moving blade and different stationary blades, giving a different and original interpretation about the source of the vibration characteristics. A successful corrective action is proposed as a consequence of this new interpretation.

Tilting-Pad Journal Bearings With Electronic Radial Oil Injection

This paper gives a theoretical treatment of the problem of journal bearings modeling connected to electronic oil injection into the bearing gap. The feasibility of influencing the static behavior of hydrodynamic forces by means of such oil injection is investigated. The lubricant is injected into the bearing gap by two mechanisms of lubrication. the conventional hydrodynamic lubrication and through orifices distributed along the bearing surface (active lubrication in the radial direction). By controlling the pressure of the oil injection, it is possible to get large variations in the active hydrodynamic forces; such effects could be useful for reducing vibrations in rotating machines.

On the Adjusting of the Dynamic Coefficients of Tilting-Pad Journal Bearings

This paper gives a theoretical and experimental contribution to the problem of active modification of the dynamic coefficients of tilting-pad journal bearings, aiming to increase the damping and stability of rotating systems. The theoretical studies for the calculation of the bearing coefficients are based on the fluid dynamics, specifically on the Reynolds equation, on the dynamics of multibody systems and on some concepts of the hydraulics. The experiments are carried out by means of a test rig specially designed for this investigation. The four pads of such a bearing are mounted on four flexible hydraulic chambers which are connected to a proportional valve. The chamber pressures are changed by means of the proportional valve, resulting in a displacement of the pads and a modification of the bearing gap. By changing the gap, one can adjust the dynamic coefficients of the bearing. With help of an experimental procedure for identifying the bearing coefficients, theoretical and experimental results are co...

Wind turbine condition monitoring based on SCADA data using normal behavior models. Part 2: Application examples

This paper is part two of a two part series. The originality of part one was the proposal of a novelty approach for wind turbine supervisory control and data acquisition (SCADA) data mining for condition monitoring purposes. The novelty concerned the usage of adaptive neuro-fuzzy interference system (ANFIS) models in this context and the application of a proposed procedure to a wide range of different SCADA signals. The applicability of the set up ANFIS models for anomaly detection was proven by the achieved performance of the models. In combination with the fuzzy interference system (FIS) proposed the prediction errors provide information about the condition of the monitored components. Part two presents application examples illustrating the efficiency of the proposed method. The work is based on continuously measured wind turbine SCADA data from 18 modern type pitch regulated wind turbines of the 2 MW class covering a period of 35 months. Several real life faults and issues in this data are analyzed and evaluated by the condition monitoring system (CMS) and the results presented. It is shown that SCADA data contain crucial information for wind turbine operators worth extracting. Using full signal reconstruction (FSRC) adaptive neuro-fuzzy interference system (ANFIS) normal behavior models (NBM) in combination with fuzzy logic (FL) a setup is developed for data mining of this information. A high degree of automation can be achieved. It is shown that FL rules established with a fault at one turbine can be applied to diagnose similar faults at other turbines automatically via the CMS proposed. A further focus in this paper lies in the process of rule optimization and adoption, allowing the expert to implement the gained knowledge in fault analysis. The fault types diagnosed here are: (1) a hydraulic oil leakage; (2) cooling system filter obstructions; (3) converter fan malfunctions; (4) anemometer offsets and (5) turbine controller malfunctions. Moreover, the graphical user interface (GUI) developed to access, analyze and visualize the data and results is presented.

THD Analysis in Tilting-Pad Journal Bearings Using Multiple Orifice Hybrid Lubrication

Tilting pad journal bearings (TPJB) using multiple orifice hybrid lubrication are analyzed applying a thermohydrodynamic (THD) theory. Adiabatic boundary conditions are adopted, and a two-dimensional model is used to represent the fluid flow behavior in the bearing gap. The influence of operational conditions on the temperature distribution and on the bearing load capacity is discussed and compared to theoretical and experimental values for a conventional hydrodynamic case (without radial oil injection). To improve the cooling effects, as well as rotor attitudes, the best location for orifices is the area near the pad edges.

Compensation of Cross-Coupling Stiffness and Increase of Direct Damping in Multirecess Journal Bearings Using Active Hybrid Lubrication: Part I—Theory

Fluid film forces are generated in hydrostatic journal bearings by two types of lubrication mechanisms: the hydrostatic lubrication in the bearing recesses and hydrodynamic lubrication in the bearing lands, when operating in rotation. The combination of both lubrication mechanisms leads to hybrid journal bearings (HJB). When part of hydrostatic pressure is also dynamically modified by means of hydraulic control systems, one refers to the active lubrication. The main contribution of the present theoretical work is to show that it is possible to reduce cross-coupling stiffness and increase the direct damping coefficients by means of the active lubrication, what leads to rotor-bearing systems with larger threshold of stability.

Frequency Response Analysis of an Actively Lubricated Rotor/Tilting-Pad Bearing System

In the present paper the dynamic response of a rotor supported by an active lubricated tilting-pad bearing is investigated in the frequency domain. The theoretical part of the investigation is based on a mathematical model obtained by means of rigid body dynamics. The oil film forces are inserted into the model by using two different approaches: (a) linearized active oil film forces and the assumption that the hydrodynamic forces and the active hydraulic forces can be decoupled, and (b) equivalent dynamic coefficients of the active oil film and the solution of the modified Reynolds equation for the active lubrication. The second approach, based on the equivalent dynamic coefficients, leads to more accurate results because it includes the frequency dependence of the active hydraulic forces. Theoretical and experimental results reveal the feasibility of reducing resonance peaks by using the active lubricated tilting-pad bearing. By applying a simple proportional controller it is possible to reach 30% reduction of the resonance peak associated with the first rigid body mode shape of the system. One of the most important consequences of such a vibration reduction in rotating machines is the feasibility of increasing their operational range by attenuating resonance peaks and reducing vibration problems.

Contribution to experimental validation of linear and non-linear dynamic models for representing rotor–blade parametric coupled vibrations

Abstract This work makes a theoretical and experimental contribution to the problem of rotor–blades dynamic interaction. A validation procedure of mathematical models is carried out with the help of a simple test rig, built by a mass–spring system attached to four flexible rotating blades. With this test rig, it is possible to highlight some dynamic effects and experimentally simulate the structural behaviour of a windmill in two dimensions (2-D model). Only lateral displacement of the rotor in the horizontal direction is taken into account. Gyroscopic effect due to angular vibrations of the rotor is eliminated in the test rig by attaching the rigid rotor to a flexible foundation. The blades are modelled as Euler–Bernoulli beams. Using three different approaches to describe the beam deformation one achieves: (a) a linear model; (b) a linear beam model with second order terms; (c) a fully non-linear model. Tip masses at the end of the blades emphasize the coupling between the dynamic and elastic terms. The shape functions are chosen in order to reduce the mathematical model, so that only the first bending mode of the beam is taken into account. The resulting equations of motion have five degrees of freedom and illustrate linear, non-linear and time-dependent terms in a very transparent way. Although neither gyroscopic effect due to rotor angular vibrations nor higher blade mode shapes are considered in the analysis, the equations of motion of the rotor–blades system are still general enough for the purpose of the work: validation of different linear and non-linear models with time dependent (periodic) coefficients. Experiments are carried out in the time and frequency domains while the rotor operates with different constant angular velocities.