Water leak location based on improved dual-tree complex wavelet transform with soft thresholding de-noising

Highlights

• De-noising method based on dual-tree complex wavelet transform and soft thresholding.

• New leak localization algorithm with inclusion of wave dispersion phenomenon.

• Improved leak location through the de-noising method and localization scheme.

Abstract

Water leakage control emerges as a prime concern among researchers and water utility companies due to ever-increasing water loss level. Acoustic leak detection technique is a promising and widely used approach. Unfavorably, the deficiency of this method is that acoustic waves are interfered by undesirable noise and presence of multi-modal dispersive wave. The conventional correlation-based method assumes leak noise propagates as a single non-dispersive wave and this leads to unreliable detection method. This paper concerns on noise reduction through an improved de-noising method and leak localization by considering wave dispersion. Wavelet de-noising is a common de-noising method used in past leak detection works. But it has shortcomings of frequency aliasing and shift-variant due to DWT decomposition. Therefore, a shift-invariant Dual Tree Complex Wavelet Transform (DTCWT) is introduced here to substitute DWT for wavelet de-noising. DTCWT will decompose the signal into several bandwidths then soft thresholding is applied to remove noise by eliminating the irrelevant signals. Multilevel DTCWT decomposition aids to tackle the problem encountered by basic localization method because the wave velocity is evaluated based on the dominant frequency and dispersion curve. Experimental results show that the proposed de-noising method outperforms ordinary wavelet de-noising. Besides, the method effectively removes noise and makes the peak of cross-correlation function more pronounced and hence it is achievable to increase leak localization accuracy. The proposed method outweighs other methods by offering the lowest false trips occurrence and minimal localization error of 3.33%. A field test result proved that the method is able to locate the leak in a 235-meter-long, underground buried pipe with a minimal error of 2.63% which is 57% lower than the conventional method.

Introduction

Water is an irreplaceable natural resource that crucial for the survival of human being and other living organisms. Pipes are used extensively to transport water to consumers in residential areas, industrial areas, commercial areas and so on. Unfortunately, ageing pipes are prone to corrosion and leakages which cause an abundant amount of water lost in transit. The water lost known as non-revenue water (NRW) and it is present in a substantial amount which is typically 20 to 30% of water production [1]. Referring to the statistics presented by the National Water Services Commission [2], Malaysia’s NRW stood at 35.3% which is equivalent to 5.929 billion liters daily. NRW is mainly contributed by leakages [3]. And yet, the prime movers of leakages are due to aged and faulty pipes, corrosion, manufacturing errors and other external factors.

Leakages bring great menaces that can give effect to human’s lives and environment such as economic and natural resources loss [4]. They also lead to soil erosion, damage to foundations of properties and roads [5]. When a leak occurs, there is a pressure drop in the pipe system so the contaminants may intrude the pipe and causes public health issue [6]. The critical aftermath brought up by leakages is the unnecessary waste of water. This may account for water demand along with ever-increasing populations and urbanizations [7]. To this end, water leakage control has drawn attention and growing interest world widely with an aim to conserve water resource.

Under such circumstances, a great effort has been devoted to developing effective and reliable water leak detection method such as mass or volume balance method, negative pressure wave, fiber optic leak detection, acoustic method etc. Acoustic emission (AE) test is a non-destructive testing technique used for structural health monitoring. According to the ASTM E1316 Standard, AE defined as a phenomenon whereby transient waves generated due to the rapid release of energy within a material. When there are discontinuities in the material, it will emit wave which can be captured by the sensor. Then the signals are processed in a computer to perform leak detection and localization via a series of algorithmic procedures.

Detection of water leaks using acoustic method is a promising approach because it can provide a faster and real-time response [8] but it is limited to metallic pipelines only. Its performance is greatly reduced in plastic pipelines because the signals will experience larger attenuation in the plastic pipes. To date, recent works [9], [10], [11] which concerning hydrophone-based method show satisfactory leak-detection results in plastic pipes. Another demerit of AE method is due to the high occurrence of false alarm or missed detection. The signals are easily polluted by noise and this leads to inaccurately leak localization [12]. To resolve this problem, most of the pipeline operators will perform leak detection during night time because background noise from traffic and water usage is lowest.

Fig. 1 illustrates the concept of leak localization. The waves generated by leak noise will travel to both sensor 1 (S1) and sensor 2 (S2). The travelling time for both waves, t₁ and t₂ are different and leak can be located by means of computing the time difference of arrival (TDOA). Cross-correlation is a typical method used to compute TDOA between two signals by measuring signal similarity. There is a distinct peak at the time shift when two signals have a maximum similarity. Thereafter, the leak location can be calculated through Eq. (1) in combination with the knowledge of wave velocity.

$$d_1=0.5\left(L+vt\right)$$

where d₁ is leak location relative to the first sensor; L is the distance between two sensors; v is wave propagation velocity and t is a time delay.

From equation stated, it is clearly shown that accuracy of leak localization is highly related to two unknown values which are wave velocity, v and time delay, t. Similarly, these issues had been addressed in previous studies [13], [14], [15], [16], [17]:

• Traditional localization method assumes the wave propagation velocity is constant and ignores the fact that leakage-induced wave exhibits dispersive behavior. Hence, an accurate localization method that taking account of the dispersion phenomenon should be considered.

• In actual cases, the signals are easily polluted by noise which makes leakage-induced signals imperceptible and leads to imprecise time delay computation. The conventional filtering method is inapplicable to eliminate noise from wide-band acoustic signals. Therefore, a workable noise reduction method should be established.

In order to optimize acoustic leak detection practices, two aforementioned problems have to be solved. Hence, this paper introduces an improved noise reduction method which is derived from DTCWT and soft thresholding with the aim to increase leak positioning accuracy. In the meanwhile, the location scheme is based on frequency-varying velocity by taking consider of wave dispersion phenomenon.