Design And Exploitation Of Measurement System For Water Lubricated Bearings And Seal Devices Test Bench

Water-lubricated bearings and sealing devices are important equipments for the normal operation of the ship’s tail shaft.Due to complicated working conditions,heavy load,poor cooling and lubrication,the bearing abrasion and seal failure sometimes occur.This has been one of the major projects of the industry.The optimized design and improvement of water-lubricated bearings and sealing devices require relevant test bench for data acquisition,calculation and analytical verification.Existing tail shaft test bench has a low-level of automation,and the data acquisition software code is solidified,which is still insufficient in terms of versatility,configuration and scalability.Especially for large shaft test benches,there are many monitoring object and complex module.The system is required to have excellent expandability and flexibility to meet the needs of the monitoring points increase and module adjustment in the future,and enhance the user’s ability to adjust the system.So,this paper focuses on the hardware design,data acquisition,abnormal data identification and cleaning of the test system based on water lubricated bearing and sealing device test bench.This paper firstly designed the hardware project of test system,including the selection of industrial computer,data acquisition card and various measuring sensors,and discussed the installation and measurement scheme of each sensor in detail.Aiming at the shortage of the leakage test,the leakage test container and the on-site measurement system with microprocessor are designed to achieve the accurate automatic test of the leakage.For traditional data acquisition software,data acquisition,computational and storage are closely correlative,resulting in inflexible code and software function modification is inconvenient.So,a configuration-based acquisition software design method is raised.The calculation rules and storage rules of each channel are refined as software configuration information,which is used to define the data acquisition process.At the same time,the user is allowed to modify the configuration information as need to enhance much flexibility and expandsibility performance of acquisition software.In addition,the further research on the expandsibility requirements of the acquisition software is carried out,and the design method of the acquisition software with expandsibility module based on.NET reflection technology is proposed.The key API types,methods,attributes and formal parameters in the vendor DLL library file are extracted into configuration information.The reflection technology is used to dynamically expand the new acquisition module according to the configuration,so as to increase and deploy the acquisition card of different manufacturers,and significantly improve the flexibility and adaptability of the acquisition software.In order to improve the quality of data acquisition,a sensor abnormal data recognition scheme based on interval estimation theory is designed.Combining the sliding window method,the median and variance of the data sequence are used to build confidence interval in order to identify random abnormal data.At the same time,the identification of the system abnormal data of the sensor is deeply studied.The mean,variance,skewness and kurtosis of the data sequence are calculated the change degree along with the magnitude and quantity of the sequence abnormal data by simulation.The mean and the skewness are used as the feature metrics after comparison.A recognition method of abnormal data based on DBSCAN clustering is proposed.By simulation,the method can identify the abnormal system data sequence of sensor.And analyze the influence of sample variance,sample length,abnormal size and quantity on clustering recognition results.In this way,the sensor system abnormal sequence identification scheme is designed.