【導(dǎo)讀】時，燃油保持正常的流動性是至關(guān)重要的。該傳感器能進行流體檢測，能根據(jù)化學(xué)規(guī)律對污染物。潤滑系統(tǒng)和預(yù)防污染的總體看法以及測試結(jié)果。了為期6個月測試，并在石油的變質(zhì)界限下驗證了多個污染物的分類。組成部分以及運行是否正常，使根據(jù)診斷和預(yù)測剩余使用壽命去維修成為可能。CBM減少了生命周期維護成本，改善了系統(tǒng)的安全性，并增加了運行的可預(yù)見性。常的共同原因之一，因此潤滑劑的質(zhì)量監(jiān)測是對CBM系統(tǒng)比較理想的補充。其誤差可能的來源主要包括取樣位置、容器交叉污染以及測試方法的準(zhǔn)確。器避免不必要的系統(tǒng)磨損，使維護時間間隔最佳化，并能盡早處理設(shè)備問題。技術(shù)通過減少設(shè)備的停機時間，以及降低運營成本使利益最大化。至20微米會造成的60％的發(fā)動機磨損。感的，而這些污染普遍存在于柴油機系統(tǒng)中并嚴(yán)重地降低潤滑劑的工作性能。包含CAN通信協(xié)議的支持能簡化并整合現(xiàn)有的檢測控制系統(tǒng)。概率的每個特征向量。

　　

【正文】 uses the remaining 30% for testing. Figure 3 shows a confusion matrix for the classifier. In a confusion matrix, the rows indicate the actual classification of the system (based on estimated contamination) and the columns indicate the classifier’s result. For an ideal classifier, the confusion matrix should have all instances lie on the diagonal. Misclassifications appear as deviations from the diagonal. As seen by the dominance of diagonal instances in Figure 3, the classifier predicts most of the classes very accurately. For example, the SOS classifies Virgin Oil (row 1) with 95% accuracy. However, for high fuel contamination (row 10), the classifier achieved only 61% accuracy. This is largely due to a limited sensitivity to fuel contamination, which is inherent in the current design。 therefore, high levels of other types of contamination can overwhelm the fuel contamination classification. Figure 3: Confusion Matrix Showing Results with Correct and Incorrect Classifications Labbased Verification of SOS Results The project sponsors conducted a sensor verification test in which they contaminated the test stand with known as well as unknown levels of fuel, water, and soot contaminantion, sampled oil from the test stand at regular intervals, and dispatched the samples to a laboratory for analysis. Testing ran over a 2day period and the test stand was drained, cleaned and refilled at the start of each day. The resulting laboratory analysis reports confirmed the ability of the Smart Oil SensorTM to detect contamination and trend multiple contamination level simultaneously. The independent analysis lab performed gas chromatography according to ASTM D3524 to measure fuel dilution. Figure 4 shows that the actual fuel levels measured by the lab fall between the fuel dilution bounds predicted by the sensor with a high level of accuracy and consistency. Note: Samples labeled as set ‘A’ denote day 1 of testing and set ‘B’ denote day two. The actual contamination levels that fall outside of the predicted bounds were never off by more than one class. SOS results trend very well with the laboratory results and are well within the margin of error (2% per ASTM specification) of the lab analysis method. Figure 4: Fuel Dilution Level Verification using Gas Chromatography Two independent laboratories performed water concentration analyses using a coulometric KarlFischer test (ASTM D6304). Figure 5 shows the actual water contaminant level detected by the labs and the bounds of the SOS classification. The plot highlights the variability that can occur between analysis labs. The SOS classifications trend well with both lab reports。 however, conclusions on measurement accuracy are limited by lab inconsistency. The discrepancy between the labs is most likely due to improper correction of the measurement for zinc dialkyl dithio phosphate (ZDDP) interference. Because the SOS relies on laboratory analysis to verify classifier measurements, selection of analysis methods and laboratory are critical to the overall accuracy of the sensor. The authors also employed two laboratories to estimate the soot levels in the oil samples (one using Wilkes Soot meter and the other using FTIR analysis) and pared these to the SOS predicted results. Both labs actually reported significantly lower soot levels than were actually added. Figure 5: Water Contamination Verification using Karl Fischer Titration Future Work While the sensor demonstrated the capability of identifying and trending multiple contaminants in a test stand environment, the authors are addressing several issues that will improve the mercial success of the sensor. Improvements such as expanding the interrogation frequency range, improving temperature pensation algorithms, and increasing the sensor’s sensitivity to particular contaminants will enhance classifier accuracy and resolution. Further improvements such as reducing electronics size, improving temperature limitations, and decreasing sensor head size will allow the sensor implementation in a much broader array of application. The authors have already extended their initial work by employing the oil sensing capability in the applications shown in Table 1. Moving beyond diesel lubes, the sensor performs extremely well while monitoring water content and lubricant quality within gearbox systems. ‘Realworld’ application testing of the sensor in diesel and gearbox applications will be used to further verify the sensor’s capabilities and identify potential limitations. Conclusion In this paper, the authors describe the Smart Oil SensorTM technology that employs broadband spectroscopy approaches, electrochemical techniques, and advanced multisensor data fusion methods to present a near realtime, inline oil analysis device. The results from the 6 months of continuous testing and data analysis demonstrate the flexibility and robustness of the classifier and highlight the accuracy it can achieve. Testing has also identified areas of improvement that need addressing to improve classifier accuracy and resolution. The laboratory analysis of oil samples corroborates the assertion that the Smart Oil SensorTM is able to detect and track levels of oil contamination due to fuel, water, and soot. With further testing, the SOS can be adapted to a wide array of other possible contaminations and fluid types. The authors have also provided additional installation and test applications in which they have integrated the oil quality sensor for further development and validation. These evaluations, in concert with dedicated laboratory groundtruth data collection, will provide the means to evolve the technology and demonstrate its ability to track and predict oil contamination in such environments. Fundamentally, the SOS technology provides a