论文部分内容阅读
对基于神经网络方法的冷水机组故障监测效率取决于训练数据和被测数据的质量问题进行了研究。采用小波变换的方法剔除测量数据中的噪声,提高数据质量,从而提高冷水机组故障诊断效率。结果表明:采用小波变换使得各个水平故障的检测效率均得到提高,尤其水平一的故障检测效率提高明显。故障水平一检测率的提高能够及时的辨别冷水机组的故障,从而采用措施防止故障进一步恶化,对降低能源消耗、提高系统的可靠性以及保证室内舒适性具有重要的意义。通过利用ASHRAE Project提供的数据对故障诊断与检测(fault detection and diagnosis)策略进行验证,检测率明显提高。
The efficiency of fault detection of chiller based on neural network is determined by the quality of training data and measured data. Using wavelet transform method to eliminate the noise in the measurement data and improve the data quality, so as to improve the fault diagnosis efficiency of chillers. The results show that the detection efficiency of each level fault is improved by using wavelet transform, especially the level one fault detection efficiency is obviously improved. The improvement of the fault level-1 detection rate can identify the chiller fault in time, so measures can be taken to prevent the fault from further deterioration, which is of great significance for reducing energy consumption, improving system reliability and ensuring indoor comfort. By using the data provided by ASHRAE Project to validate the fault detection and diagnosis strategy, the detection rate is significantly improved.