Detection of Internal Defects in Concrete and Evaluation of a Healthy Part of Concrete by Noncontact Acoustic Inspection Using Normalized Spectral Entropy and Normalized SSE

Abstract

Non-destructive testing, with non-contact from a remote location, to detect and visualize internal defects in composite materials such as a concrete is desired. Therefore, a noncontact acoustic inspection method has been studied. In this method, the measurement surface is forced to vibrate by powerful aerial sound waves from a remote sound source, and the vibration state is measured by a laser Doppler vibrometer. The distribution of acoustic feature quantities (spectral entropy and vibrational energy ratio) is analyzed to statistically identify and evaluate healthy parts of concrete. If healthy parts in the measuring plane can be identified, the other part is considered to be internal defects or an abnormal measurement point. As a result, internal defects are detected. Spectral entropy (SE) was used to distinguish between defective parts and healthy parts. Furthermore, in order to distinguish between the resonance of a laser head and the resonance of the defective part of the concrete, spatial spectral entropy (SSE) was also used. SSE is an extension of the concept of SE to a two-dimensional measuring space. That is, based on the concept of SE, SSE is calculated, at each frequency, for spatial distribution of vibration velocity spectrum in the measuring plane. However, these two entropy values were used in unnormalized expressions. Therefore, although relative evaluation within the same measurement surface was possible, there was the issue that changes in the entropy value could not be evaluated in a unified manner in measurements under different conditions and environments. Therefore, this study verified whether it is possible to perform a unified evaluation for different defective parts of concrete specimen by using normalized SE and normalized SSE. From the experimental results using cavity defects and peeling defects, the detection and visualization of internal defects in concrete can be effectively carried out by the following two analysis methods. The first is using both the normalized SE and the evaluation of a healthy part of concrete. The second is the normalized SSE analysis that detects resonance frequency band of internal defects.

Study on movement measurement for internal defects in concrete structures by noncontact acoustic inspection method using correlation processing

“…Moreover, exploration experiments have been carried out in actual concrete structures such as tunnels, viaducts 21) and underground cavities, [22][23][24] and acoustic irradiation experiments using tiled outer wall specimens from a flying drone. 25,26) Furthermore, spatial spectral entropy (SSE) 22,27) can distinguish between resonance phenomena caused by internal defects and resonance phenomena of the LDV itself by using the planar vibration velocity distribution. A resonance judgment process 28) has also been devised to determine whether the emitted signal caused vibration.…”

An Internal Defect Detection Algorithm for Concrete Blocks Based on Local Mean Decomposition-Singular Value Decomposition and Weighted Spatial-Spectral Entropy

“…where Q f denotes the amplitude of the signal at frequency f. The SSE extends this analysis from independent signals to all measured signals in the two-dimensional scanning plane [42]. Figure 1 shows the conceptual diagram of SSE.…”

“…On the contrary, signals with spectral characteristics akin to white noise will have larger SE values. The SE is defined as follows [42].…”

“…For example, Yasuda et al used laser ablation of vibrated concrete and LDV to check the tunnel lining health and identify internal concrete defects based on the intrinsic frequency and damping ratio [37][38][39]. Sugimoto et al [40][41][42] proposed a spatial-spectral entropy-based defect imaging algorithm that uses the spatialentropy spectrum (SSE) to estimate the defect resonance frequency of the data collected via LDV, and displays the size and location of the defect based on the vibration modes at the defect resonance frequency.…”

Non-Destructive Inspection of Concrete Structures by Noncontact Acoustic Inspection Technology Using Acoustic Irradiation Induced Vibration

“…(STNB : Single ToNe Burst)波，時間周波数ゲート (13) ，振 動エネルギー比 (14) 及びスペクトルエントロピーを用いた欠 陥検出アルゴリズム (15) 及び高速計測を可能にするマルチ トーンバースト(MTNB : Multi ToNe Burst)波 (16) などが考 案された．実際に，トンネル，高架橋 (17) 及び地下空洞 (18) と いった実コンクリート構造物における探査実験，タイル外壁 供試体を用いた飛行中のドローンからの音波照射加振実 験 (19), (20) などが実施されている．更に面的な振動速度分布を 利用して内部欠陥に起因する共振現象と LDV 自体の共振現 象を識別することができる空間スペクトルエントロピー (SSE : Spatial Spectral Entropy) (21), (22) や送信した信号により 加振されたかどうか判定する共振判定処理 (23) 波で一つの周波数を使用する STNB 波 (13) ，1 回の送波で複 数の周波数を使用する MTNB 波 (16) 15), (17) 図 4 共振判定処理における受信波形分割の概念図 (23) 図 5 NCAI 法の基本セットアップ図 (26) テスト結果(検査者全員の結果が一致しない場合は△) ，下 段は NCAI 法による欠陥中心部における共振周波数の探査 結果例である(模擬欠陥の共振ピークは確認できるものの， 図 (15), (17) 図 13 N2U-Bridge における実験結果例 (15), (17) 163 図 14 高架橋での実験セットアップ図 (17) 図 15 高架橋床版のスキャンエリア図 (17) 図 16 NCAI 法による振動エネルギー比分布 (17) 図 17 叩き点検による検証実験 (17) 図 タイル外壁供試体の模擬欠陥配置図 (19), (20) 図 20 タイル外壁点検実験のセットアップ図 (20) 図 振動エネルギー比分布 (20) 5.5 民間マンションでの外壁点検実施例…”