Design of a Condition Monitoring System for Wind Turbines

Renewable energy is being adopted worldwide, and the proportion of offshore wind turbines is increasing. Offshore wind turbines operate in harsh weather conditions, resulting in various failures and high maintenance costs. In this paper, a condition diagnosis model for condition monitoring of an offshore wind turbine has been developed. The generator, main bearing, pitch system, and yaw system were selected as components subject to the condition monitoring by considering the failure rate and downtime of the wind turbine. The condition diagnosis model works by comparing real-time and predictive operating data of the wind turbine, and about four years of Supervisory Control and Data Acquisition (SCADA) data from a 2 MW wind turbine was used to develop the model. A deep neural network and an artificial neural network were used as machine learning to predict the operational data in the condition diagnosis model, and a confusion matrix was used to measure the accuracy of the failure determination. As a result of the condition monitoring derived by inputting SCADA data to the designed system, it was possible to maintain the failure determination accuracy of more than 90%. The proposed condition monitoring system will be effectively utilized for the maintenance of wind turbines.

Research and Application of Rod Pump Working Condition Diagnosis and Virtual Production Metering Based on Electric Parameters

The conventional configurations of pumping well IOT consist of electric parameter indicator and dynamometer. The current, voltage, power, and other electrical parameters are easy to access, low costs, stable, and acquired daily during pumping well operation. If the working condition diagnosis and virtual production metering of pumping well can be realized through electrical parameters, the utilization of dynamometers can be cancelled or reduced, which is of great significance to reduce the investment and improve the coverage of IOT in oil wells. The conventional methods of diagnosis and analysis based on electrical parameters and virtual production metering are lack of theoretical basis. The combination of deep learning technology of big data and traditional methods will provide solutions to solve related technical problems. Considering that there are many energy transmission segments from the motor to the downhole pump, the characteristics of the electric parameter curve are more sophisticated and difficult to identify compared with dynamometer card due to the influence of the unbalance, pump fullness, rod/tube vibration, wax deposition and leakage. The shape characteristics of the electric parameter curve of the pumping well are analyzed in the time domain and frequency domain, which provides the basis for further diagnosis, analysis and production measurement. In this paper, an integrated multi-model diagnosis method is proposed. For the working conditions with a large scale of samples, the electrical parameters are converted to dynamometer cards for diagnosis by using the deep learning technology of big data. For the working conditions with sparse samples, the machine learning model is used to diagnosis directly with electrical parameters. The deep learning electric parameter model for production measurement is established. Through the combination of the big data model of electric parameters to dynamometer card, 3D mechanical model of rod string, and big data model of plunger leakage coefficient, the virtual production metering function of pumping well based on electrical parameters is successfully realized. The diagnosis and virtual production metering method and software based on electrical parameters have been applied in many oilfields of CNPC. The accuracy of identifying the upper and lower dead points of electric parameters is 98.0%; the coincidence rate of working condition diagnosis under electrical parameters is 92.0%; the average error of virtual production metering with electric parameters is 13.4%. The dynamometer and gauging room have been canceled in the demonstration area. The application of electrical parameters to diagnose working conditions and meter the production of pumping wells is the key to the low-cost IOT construction. Traditional mathematical and physical methods are difficult to solve this problem, but the application of big data analysis technology could do the job successfully.

Liver haematoma in pregnancy: A rare complication of HELLP syndrome

Liver haematoma complicating pregnancy with HELLP syndrome is a rare but life-threatening condition. Diagnosis requires a high index of suspicion. Here a case of liver haematoma is presented in a 29 years multigravida at 34 weeks of pregnancy. Initially hypertension was not revealed since she had presented in shock. She had presented with on and off epigastric pain for many days. In context of haemoperitoneum in ultrasound, emergency laparotomy was done and liver haematoma diagnosed intraoperatively. Haematoma was managed with packing and second look laparotomy was done to remove the packs. Haematoma gradually resolved over period of months.

Practical experience condition diagnosis gained from oil-immersed power transformers voltage class 10 kV

Dissolved Gas Analysis (DGA) for oil samples has been the most widely used diagnosis tool for transformer condition assessment for many years. However, DGA use to oil-filled transformers with a voltage class up to 100 kV. The aim of this paper is to address the issue of DGA interpretation to oil-filled transformers with a voltage class of 10 kV. This paper will present DGA tests results from 57 power transformers and will propose a maintenance decision making procedure using the IEC 60599-2015 Ratio Method, IEEE Std C57.104-2008 include Dornenberg Ratio Method and Rogers Ratio Method, and Russian Std CTO 56947007-29.180.010.094-2011 and Russian Std RD 153-34.0-46.302-00.

Working-condition diagnosis of a beam pumping unit based on a deep-learning convolutional neural network

At present, beam pumping units are the most extensively-applied component in rod pumping systems, and the analysis of the indicator diagram of a rod pump is an important means of judging its downhole working condition. However, the synthetic study and judgment of the indicator diagram by manual means has a low efficiency, large error, and poor immediacy, and it is difficult to apply the conclusions in time and accurately to adjust the operating parameters of the pumping units. Moreover, expert systems rely on expert experience and conventional machine learning requires manual pre-selection of geometric features such as moments and vector curves, which will reduce the accuracy of recognition when similar indicator diagrams appear. To solve the above technical defects, in this paper, a deep-learning convolutional neural network (CNN) is proposed using the CNN model based on AlexNet. The automatic recognition of the indicator diagram is thus realized, and, on the basis of previous studies, this model simplifies the structure of the model and takes into account 15 common downhole working conditions of the pumping unit. In this model, the batch normalization (BN) layer is used to replace the local response normalization (LRN) and dropout layers and all kinds of indicator diagrams are put into the same model frame for automatic identification. The experimental application of the measured data shows that the model not only has a short training time, but also has a working-condition diagnosis accuracy of 96.05%, which can solve the deficiencies and defects of artificial identification, expert systems, and conventional machine learning to a certain extent. A deep-learning CNN can provide a new reference for fast working-condition diagnosis of indicator diagram, making indicator-diagram judgment timely and accurate, and thus it is possible to provide a direct basis for parameter adjustment of pumping units.

Autonomous Analysis of Infrared Images for Condition Diagnosis of HV Cable Accessories

Infrared thermography has been used as a key means for the identification of overheating defects in power cable accessories. At present, analysis of thermal imaging pictures relies on human visual inspections, which is time-consuming and laborious and requires engineering expertise. In order to realize intelligent, autonomous recognition of infrared images taken from electrical equipment, previous studies reported preliminary work in preprocessing of infrared images and in the extraction of key feature parameters, which were then used to train neural networks. However, the key features required manual selection, and previous reports showed no practical implementations. In this contribution, an autonomous diagnosis method, which is based on the Faster RCNN network and the Mean-Shift algorithm, is proposed. Firstly, the Faster RCNN network is trained to implement the autonomous identification and positioning of the objects to be diagnosed in the infrared images. Then, the Mean-Shift algorithm is used for image segmentation to extract the area of overheating. Next, the parameters determining the temperature of the overheating parts of cable accessories are calculated, based on which the diagnosis are then made by following the relevant cable condition assessment criteria. Case studies are carried out in the paper, and results show that the cable accessories and their overheating regions can be located and assessed at different camera angles and under various background conditions via the autonomous processing and diagnosis methods proposed in the paper.

Language and Communication Impact of Hypertension: A Qualitative Study

Background. Hypertension (HTN) is the second main source of outpatient morbidity in Ghana, and the understanding of a disease is necessary for its prevention and management. Language and communication are contributing factors to HTN in Ghana. No studies have been conducted to assess knowledge/awareness of HTN (in the context of its understanding) among students in Ghana. Following a local name for HTN in Ghana, researchers interviewed students through a focus group to assess their understanding/perception (meaning, cause, and prevention) of the disease. Available literature has concerned itself with clients’ knowledge of their condition (diagnosis) rather than their comprehension of the true nature of what HTN is. The objective of this study is to assess the knowledge/awareness of HTN in the context of its understanding of the meaning, perception, causes, and prevention of hypertension among students of Ghana’s Senior High School (Second Cycle). Semistructured interviews with the use of the theme lists were employed. Focus group conversations and interviews were held in the local Akan (Twi) language, which was later translated, interpreted, and analyzed. Overall, 25 second-cycle students participated. 60% were between 15 and 17 years, 24% were ≥18 years, and 16% were <15 years of age. Males were 44% and females were 56%. Students gave diverse perceptions of their knowledge of HTN. The local language’s translation of HTN has influenced and affected its meaning/understanding among some, thus affecting their perception of causes and prevention.

Integrated Multiple-Defect Detection and Evaluation of Rail Wheel Tread Images using Convolutional Neural Networks

The wheel-rail interface is regarded as the most important factor for the dynamic behaviour of a railway vehicle, affecting the safety of the service, the passenger comfort, and the life of the wheelset asset. The degradation of the wheels in contact with the rail is visibly manifest on their treads in the form of defects such as indentations, flats, cavities, etc. To guarantee a reliable rail service and maximise the availability of the rolling-stock assets, these defects need to be constantly and periodically monitored as their severity evolves. This inspection task is usually conducted manually at the fleet level and therefore it takes a lot of human resources. In order to add value to this maintenance activity, this article presents an automatic Deep Learning method to jointly detect and classify wheel tread defects based on smartphone pictures taken by the maintenance team. The architecture of this approach is based on a framework of Convolutional Neural Networks, which is applied to the different tasks of the diagnosis process including the location of the defect area within the image, the prediction of the defect size, and the identification of defect type. With this information determined, the maintenancecriteria rules can ultimately be applied to obtain the actionable results. The presented neural approach has been evaluated with a set of wheel defect pictures collected over the course of nearly two years, concluding that it can reliably automate the condition diagnosis of half the current workload and thus reduce the lead time to take maintenance action, significantly reducing engineering hours for verification and validation. Overall, this creates a platform or significant progress in automated predictive maintenance of rolling stock wheelsets.