电网稳态数据频率跟踪与在线压缩算法研究

发布时间：2018-06-06 06:48

本文选题：电力录波 + 数据无损压缩　；参考：《哈尔滨工业大学》2014年硕士论文

【摘要】：电网规模的扩大和智能电网的加速建设，使得电力系统故障变得日益复杂，同时造成需要实时记录并存储的电网数据量也越来越大。为满足电力系统故障分析、稳定性分析及电力系统安全等需求，电力录波装置应运而生。其主要用于实时采集并存储电网数据，在电网发生异常时，通过对这些数据分析以发现异常原因并实现对故障的定位，然而，长时间记录造成的海量数据不仅需要极大的存储空间，并且也不利于在后期对数据的提取与分析。因此，本文为提高电网数据压缩比设计了一种电网稳态频率跟踪算法，并设计了一种以实现在线数据压缩为目标的数据处理系统。首先，本文给出了电力录波装置的整体流程框图，对装置中各电路模块的功能做了详细的分析，并对各接口单元的通信速率及其对数据采样的影响进行了评估。进而，根据电网数据的特点，针对同步采样和异步采样得到的电网数据，在上位机上对比研究了Huffman编码和LZW编算法对电网数据的无损压缩比率。根据研究结果，进一步采用压缩效果较好的LZW算法在DSP平台中对电网数据进行压缩测试，并对比了其对多组不同数据样本的压缩效果。为实现同步采样以提高数据压缩比，设计了一种电网稳态频率跟踪算法。在此算法中，首先采用IIR组合滤波器从电网信号中提取基波信号，再采用扩展卡尔曼滤波算法对其进行频率跟踪，以得到电网预估频率完成同步采样。在电网信号发生异常时，对比采用单相电网模型的卡尔曼滤波算法和本文设计的频率跟踪算法的频率跟踪效果。论文最后将电网频率跟踪算法及LZW无损压缩算法移植到录波装置的DSP中，对硬件平台的数据获取、压缩及传输时间进行测试，通过精确的数据流实时性分析，配合FPGA数据获取及USB数据上传，综合评估了本文方法在DSP中进行数据处理的效果。
[Abstract]:With the expansion of power network scale and the accelerated construction of smart grid, the fault of power system becomes more and more complex, at the same time, the amount of power network data that needs to be recorded and stored in real time is more and more large. In order to meet the needs of power system fault analysis, stability analysis and power system safety, the power recording device emerged as the times require. It is mainly used to collect and store power network data in real time. In the event of power grid anomalies, we analyze these data to find the causes of the anomalies and realize the fault location. However, The large amount of data caused by long time records not only requires a lot of storage space, but also is not conducive to the extraction and analysis of the data in the later stage. Therefore, in order to improve the power network data compression ratio, a steady frequency tracking algorithm is designed, and a data processing system is designed to achieve on-line data compression. Firstly, the whole flow chart of the power recording device is given, the function of each circuit module in the device is analyzed in detail, and the communication rate of each interface unit and its influence on the data sampling are evaluated. Furthermore, according to the characteristics of power network data, the lossless compression ratio of Huffman coding and LZW coding algorithm to power network data is compared on the upper computer for synchronous sampling and asynchronous sampling. According to the research results, the LZW algorithm with good compression effect is further used to compress the power network data on the DSP platform, and the compression effect of the LZW algorithm on many different sets of data samples is compared. In order to realize synchronous sampling to improve data compression ratio, a steady frequency tracking algorithm is designed. In this algorithm, the IIR filter is used to extract the fundamental signal from the power system signal firstly, and then the extended Kalman filter algorithm is used to track the frequency to get the estimated frequency of the power network to complete the synchronous sampling. When the signal of the power network is abnormal, the effect of the Kalman filtering algorithm based on single-phase power network model and the frequency tracking algorithm designed in this paper is compared. Finally, the frequency tracking algorithm and the LZW lossless compression algorithm are transplanted to the DSP of the recording device. The data acquisition, compression and transmission time of the hardware platform are tested, and the real-time analysis of the accurate data flow is carried out. With the help of FPGA data acquisition and USB data upload, the effect of data processing in DSP is evaluated synthetically.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM712

【相似文献】