減少雜訊影響之電力暫態信號監測方法

Power Transient Signal Monitoring Methods with Noise Effects Reduced

廖炯州(Chiung-Chou Liao)

指導教授 ： 楊宏澤

中原大學/工學院/電子工程研究所/博士(2001年)

https://doi.org/10.6840/cycu200100136

摘要

近年來，由於精密電子儀器與非線性負載的增加，使得電力暫態信號現象廣泛的受到重視。電力暫態信號現象的發生通常包含許多複雜的原因。為了提升電力品質並找尋電力暫態信號現象的發生原因，無可避免的需廣泛與長時間的監測電力信號。然而，如將所有的信號皆記錄下來加以分析，則儲存大量波形資料的成本勢必相當的昂貴。所以，若能偵測信號受干擾的時間點，儲存並分析這些受干擾的信號前後時間點的波形即可達成同樣的成效。
為偵測信號受干擾的時點，小波轉換的方法常被用來分析電力系統信號的擾動現象。小波轉換具有在時域與頻域多重解析的能力。在高頻時，經小波轉換後信號擁有較高的時域解析度；較低頻時，卻擁有較高的頻率解析度。由於多重解析的特性，使得小波轉換非常適合於偵測不同現象的電力系統暫態。藉由觀察高頻尺度時間軸的小波轉換係數值，可輕易偵測信號中發生暫態現象的時間點。藉由此種方法，受干擾的信號則可從大量信號中擷取出，以達到節省資料儲存空間與成本的目的。
在實際的應用上，偵測是否有電力暫態信號現象發生的過程經常因雜訊的影響而使選定小波轉換係數值臨界值以偵測是否有暫態信號的發生變的相當困難。亦即，小波轉換的偵測能力會因雜訊的影響而大幅降低。為增強以小波為基礎的電力暫態信號監測方法的能力，本文提出兩種去雜訊的方法，用以偵測含有雜訊環境中電力信號是否有暫態現象發生。此兩種去雜訊的方法，臨界值皆可依據背景雜訊值的大小，自動調整設定值。因此，經由所提出之去雜訊的演算法後，可恢復小波轉換在偵測電力信號暫態現象是否發生與發生時間點的能力。
為測試本文提出的去雜訊方法提昇電力暫態信號監測方法偵測準確率，本文利用電磁暫態程式模擬電力系統中多種不同型態的暫態干擾波形資料，以及收集實際現場量測波形資料，作為測試波形，同時，為測試本監測方法對雜訊多寡的容忍程度，本文亦模擬不同程度的訊雜比信號，測試本系統是否依然可成功偵測是否有擾動現象發生，以觀察所使用之方法在實測資料上的通用性。經測試後，使用本文所提出除雜訊方法的監測方法，其偵測電力暫態信號的能力與雜訊忍受度可獲得不錯的結果。

關鍵字

小波轉換 ； 除雜訊方法 ； 電力暫態信號監測方法

並列摘要

Recently, power transient signal phenomena have received wide attention due to increasing use of delicate electronic devices and non-linear power loads. The occurrence of a power transient signal phenomenon usually involves complicated causes. As a result, to upgrade the power quality, monitoring the power signals in a wide range and in a long run is evitable to find out the real causes of the power transient signal phenomenon. However, if all the signals under monitoring are recorded for analysis, the volume of the data would be prohibitively large. Therefore, only the portion of the signals covering the disturbances should be saved and analyzed for further investigation of the power transient signal phenomena.

To detect and localize of the disturbances of signal, the wavelet Transform (WT) approach is used in the analysis of power system disturbances. Multi-resolutions of both the time and the frequency are prepared in the WT approach: greater resolution in time is provided for high frequency components of a signal, and greater resolution in frequency for low frequency components. With the multi-resolution characteristics, the WT is considered very adequate for the analysis of the power system transients due to various disturbances. By observing the time-evolving coefficients of the higher-frequency scale of the signal after WT, the time points the transient event takes place in the signal can be easily detected. By means of this way, the disturbances can be identified out of the voluminous data recorded and a great deal of memory and cost can be saved.

In practical applications, the process of monitoring power transient signals often confuses the transient signals with the noises riding on the signals. Consequently, the threshold is difficult to give for detecting the existence of transient signals. Effectiveness of the WT techniques is, therefore, deteriorated greatly by the noises that ride on the signals. To enhance the capability of the WT-based power transient signal monitoring methods, this dissertation proposes two de-noising approaches to detecting transient disturbances in a noisy environment. In both proposed de-noising approaches, the thresholds for eliminating the influences of noises are determined adaptively according to the background noises. Through the proposed de-noising methods for the power transient signal monitoring methods, the abilities of the WT in detecting and localizing the disturbances can hence be restored.

To test the effectiveness of the developed de-noising schemes in enhancing the accuracy of monitoring power transient signal disturbances, employed were diverse data obtained from the EMTP for the main transient disturbances in the power methods. Simultaneously, to prove the noise tolerance of the proposed monitoring methods, this dissertation also added different degrees of noises to the simulated signals. The main purpose is to examine if the proposed monitoring methods can still detect and locate the disturbances. Besides, the data collected from actual field were also used as the testing data to verify the proposed methods in practical applications Using the de-noise approaches proposed in this dissertation, remarkable efficiency of monitoring the power transient signal phenomena and high tolerance to the noises are proved.

並列關鍵字

Wavelet Transform ； Power Transient Signal Monitoring Methods ； De-noising Scheme

參考文獻

[2]L. Cristaldi and A. Ferrero, “A Method and Related Digital Instrument for the Measurement of the Electric Power Quality,” IEEE Trans. on Power Delivery, Vol. 10, No. 3, pp. 1183-1189, July 1995.

[3]J.W. Gray and F.J. Haydock, “Industrial Power Quality Considerations When Installing Adjustable Speed Drive Systems,” IEEE Trans. on Industry Applications, Vol. 32, No. 3, pp. 646-652, May-June. 1996.

[4]W.E. Reid, “Power Quality Issues-Standards and Guidelines,” IEEE Trans. on Industry Applications, Vol. 32, No. 3, pp. 625-632, May-June 1996.

[5]R.C. Lentz, F.J. Mercede, and J.N. Mercede, “A Student Design Project to Improve Power Quality for a Commercial Facility,” IEEE Trans. on Power Systems, Vol. 10, No. 1, pp. 3-10, Feb. 1995.

[6]G.T. Heydt, Electric Power Quality, Circle Publications, Indiana, 1994.