A new leak location method based on leakage acoustic waves for oil and gas pipelines

In order to study a new leak detection and location method for oil and natural gas pipelines based on acoustic waves, the propagation model is established and modified. Firstly, the propagation law in theory is obtained by analyzing the damping impact factors which cause the attenuation. Then, the dominant-energy frequency bands of leakage acoustic waves are obtained through experiments by wavelet transform analysis. Thirdly, the actual propagation model is modified by the correction factor based on the dominant-energy frequency bands. Then a new leak detection and location method is proposed based on the propagation law which is validated by the experiments for oil pipelines. Finally, the conclusions and the method are applied to the gas pipelines in experiments. The results indicate: the modified propagation model can be established by the experimental method; the new leak location method is effective and can be applied to both oil and gas pipelines and it has advantages over the traditional location method based on the velocity and the time difference. Conclusions can be drawn that the new leak detection and location method can effectively and accurately detect and locate the leakages in oil and natural gas pipelines.     

The method for leakage detection of urban natural gas pipeline based on the improved ITA and ALO

To solve the problems of the difficulty in early leakage monitoring and larger positioning error for urban hazardous chemicals pipelines, the optimized method based on the improved Inverse Transient Analysis (ITA) and Ant Lion Optimizer (ALO) was proposed. Firstly, based on the obtained experiment's results of leakage of natural gas in the non-metallic pipeline, the segment classification method was incorporated into the pressure gradient calculation. The modified method can adapt to the multi-node characteristics of urban pipe networks and help to obtain the preliminary positioning calculation results after optimization. Then the calculation results were embedded in the ITA calculation model. The input parameters of the gas pipeline such as boundary conditions, leakage rate and friction coefficient were used to establish the characteristic linear equations. Then the objective function of the least-squares criterion was defined, and the improved ITA model suitable for leakage detection of urban natural gas pipeline networks was constructed. Finally, the ALO was used to optimize the calculation process of the improved ITA model, and iteratively optimize the optimal friction coefficient and its corresponding minimum objective function (OF) value. As a result, a more precise location of the leakage source was calculated. The validation of the modified method is conducted by comparing the calculated values with the experiment's results. The results show that the method can accurately predict the location where the pipeline leakage occurs. The minimum error is 3.17%. Compared with the traditional ITA, this method not only accelerates the convergence speed of the objective function, but also improves the accuracy of location calculation.     

压力脉冲波法天然气管道堵塞检测研究

为快速检测出天然气管道堵塞位置、长度及堵塞程度,解决天然气输运管道中堵塞定位难题,探讨压力脉冲波法管道堵塞检测的可行性与效果.搭建长220 m的压力脉冲波堵塞检测实验系统,进行多组连续性长堵塞检测实验和水堵检测实验.结果表明:连续性长堵塞检测实验中堵塞前沿反射信号为负压波,堵塞后沿信号为正压波,且波形清晰可辩,其多组实...     

Experimental study on leak detection and location for gas pipeline based on acoustic method

The leak of gas pipelines can be detected and located by the acoustic method. The technologies of recognizing and extracting wave characteristics are summarized in details in this paper, which is to distinguish leaking and disturbing signals from time and frequency domain. A high-pressure and long distance leak test loop is designed and established by similarity analysis with field transmission pipelines. The acoustic signals collected by sensors are de-noised by wavelet transform to eliminate the background noises, and time-frequency analysis is used to analyze the characteristics of frequency domain. The conclusion can be drawn that most acoustic signals are concentrated on the ranges of 0-100 Hz. The acoustic signal recognition and extraction methods are verified and compared with others and it proves that the disturbing signals can be efficiently removed by the analysis of time and frequency domain, while the new characteristics of the accumulative value difference, mean value difference and peak value difference of signals in adjacent intervals can detect the leak effectively and decrease the false alarm rate significantly. The formula for leak location is modified with consideration of the influences of temperature and pressure. The positioning accuracy can be significantly improved with relative error between 0.01% and 1.37%.     

2种基于模式识别的环状燃气管网泄漏检测方法

为了将模式识别技术应用于环状燃气管网泄漏检测并找到合适的特征提取方法,以天津城建大学实验室环状燃气管网泄漏为例,将实验的28种工况作为测试样本,与之对应的模拟工况作为训练样本,采用提取压力图像特征向量法和节点压力矩阵法分别进行环状燃气管网的泄漏检测,采用支持向量机分类器将2种方法获得的特征向量进行训练与分类检验,进而将其分类准确率进行对比分析。研究结果表明:该2种方法均可用于环状燃气管网泄漏检测,提取压力图像特征向量法因有效地降低了特征向量的维度和数据波动的干扰,其结果更优。结合SCADA和GIS系统,可将该法应用于实际水、气、油管网泄漏检测和定位,有助于降低成本,提高检测效果。     

煤与瓦斯突出预测的QGA-LSSVM模型

为快速、有效地对煤与瓦斯突出类型作出预测,运用灰色关联和因子分析模型对所选主要的判别指标进行分析提取,利用量子遗传算法(QGA)对最小二乘支持向量机(LSSVM)的参数作寻优处理,最终建立QGA-LSSVM煤与瓦斯突出预测模型。选取从砚石台矿区历史实测的数据,以96∶20的比例对该模型进行训练与测试,并将预测结果与其他预测模型的预测效果进行了比较。研究结果表明:对判别指标进行灰色关联分析可以有效去除对煤与瓦斯突出影响作用小的指标;用因子分析进行公共因子提取,可以有效减少数据信息冗余;利用QGA优化的LSSVM模型能使结果避免陷入局部最优解,用该模型可以有效预测煤与瓦斯突出类型,误判率为0。     

基于熵值法和支持向量机的煤矿应急救援能力评价模型

为增强煤矿应急救援能力,提高评价的精度,弥补单一评价方法的缺陷,构建了基于熵值法和支持向量机的煤矿应急救援能力评价模型。根据各评价指标数据所提供的信息量,运用熵值法设定指标权重,进一步将指标权重定量化、客观化;同时鉴于煤矿应急救援能力评价数据难收集、影响因素复杂多变的状况,结合具有处理非线性、小样本数据问题的支持向量机进行评价,保证了评价结果的准确性。实证分析表明,该模型的评价结果与煤矿实际应急救援能力非常接近,可用于实际的煤矿应急救援能力评价。     

基于稳态模型的井下抽采主管道泄漏检测方法研究*

为准确检测煤矿井下瓦斯抽采主管道泄漏位置,提出基于稳态模型的管道泄漏检测与定位方法,采用Comsol数值模拟及地面相似实验研究压力梯度法对煤矿井下抽采管道泄漏检测与定位的可行性及准确性。研究结果表明:管道未泄漏时,管内沿线压力呈均匀分布,当管道突发泄漏时,管内压力分布呈现明显弯折现象,弯折处即为管道漏点位置,并对管道阻力计算公式进行修正,提高了检测准确性;随着管道泄漏程度的加大,湍流效应显著增强,漏点处速度、压力产生明显突变,且当其他条件恒定时,随着管道泄漏孔径的扩大,管道的漏入量越大,定位的相对误差越小;在宏岩矿开展地面相似实验,实验结果绝对误差为4.5 m,相对误差为6%;在阳煤五矿井下8421抽采巷进行现场应用,绝对误差134 m,相对误差约7.95%。     

Correction model for CO detection in the coal combustion loss process in mines based on GWO-SVM

Accurate detection of CO gas is crucial to the prevention of coal combustion. Tuneable diode laser absorption spectroscopy (TDLAS) is a reliable method for CO detection during coal combustion. The influences of temperature and pressure cause changes in the line strength and linewidth of the index gases’ absorption spectra, leading to sizable measurement errors. To correct the distortion of the CO absorption spectrum caused by temperature and pressure fluctuation, a compensation model based on the grey wolf optimizer–support vector machine (GWO–SVM) was proposed. The results were compared with those of the single SVM, the back propagation neural network (BPNN), and multiple regression analysis (MRA). MRA was revealed to result in the lowest accuracy, which indicated that MRA is not ideal for compensation in TDLAS. The hyperparameter selection of the SVM had the disadvantages of randomness and blindness, which led to instability and large errors. The BPNN achieved better correction in the training stage, but severe overfitting occurred in the testing stage. The modified results revealed that the GWO–SVM model had higher accuracy and stability than the other models. It effectively inhibited the effects of temperature and pressure on the measured concentration and greatly improved the measurement accuracy. The equipment is thus suitable for CO gas detection with the aim to preventing coal combustion loss, and it can be further applied to loss prevention in other process fields.     

煤与瓦斯突出预测的支持向量机(SVM)模型

基于支持向量机(SVM)分类算法,考虑影响煤与瓦斯突出的主要因素,建立了煤与瓦斯突出预测的SVM模型。该模型选取开采深度、瓦斯压力、瓦斯放散初速度、煤的坚固性系数以及地质破坏程度5个指标作为模型输入量,同时将煤与瓦斯突出程度划分为无突出、小型突出、中型突出和大型突出4个等级,进而使其评判结果更为细化。以实测数据作为学习样本进行训练,建立相应判别函数对待判样本进行预测。通过算例分析,表明该模型的方法对煤与瓦斯突出预测的合理性与有效性,可以在实际工程中推广。     

基于链式多重插补的WOA-ELM煤与瓦斯突出预测模型

为了提高缺失数据下煤与瓦斯突出预测准确率,提出1种基于链式支持向量机多重插补(MICE＿SVM)的鲸鱼优化算法(WOA)-极限学习机(ELM)预测模型,以淮南朱集矿区为例,选取5个煤与瓦斯突出影响指标作为模型特征,采用提出的MICE＿SVM算法插补突出事故数据中缺失值,利用WOA优选ELM输入层权值及隐含层阈值,构建煤与瓦斯突出预测模型,将插补后数据用于WOA-ELM模型的训练与测试,并与其他模型的预测效果对比。研究结果表明：MICE＿SVM插补前、后的有突出数据预测准确率分别为83.02%,90.41%,MICE＿SVM显著提高了有突出预测准确率,对无突出和整体的预测准确率提高不明显;数据插补后WOA优化ELM对无突出、有突出和整体的预测准确率分别为97.94%,96.25%,96.48%,较优化前分别提高了5.79%,5.84%,5.55%,数据插补后WOA-ELM为最佳预测模型。     

矿井突水水源识别的RS-LSSVM模型

为了对矿井突水水源进行准确、高效的判别,综合考虑水化学特征,选取Ca~(2+),Mg~(2+),K~++Na~+,HCO-3,SO2-4,Cl~-和总硬度7个指标的质量浓度(mg/L)作为矿井突水水源的最初判别指标。利用粗糙集(RS)理论的属性约简来筛选水化学特征指标,用以作为水源识别的核心判别指标,建立基于RS的矿井突水水源识别的最小二乘支持向量机(LSSVM)模型。选用约简处理后的13组煤矿数据对模型进行训练,再用训练好的模型对另外12组突水数据进行水源判别,并与未进行属性约简的LSSVM模型及Fisher判别分析法、随机森林方法进行对比。结果表明,利用属性约简方法可以很好地排除原始数据中的冗余信息干扰,因而能有效判别矿井突水水源,使矿井突水水源模型的误判率降低至0;而且指标约简过程可以降低LSSVM运算的复杂度,也能够提高判别效率。     

Pipeline leakage detection and isolation: An integrated approach of statistical and wavelet feature extraction with multi-layer perceptron neural network (MLPNN)

Leakage diagnosis of hydrocarbon pipelines can prevent environmental and financial losses. This work proposes a novel method that not only detects the occurrence of a leakage fault, but also suggests its location and severity. The OLGA software is employed to provide the pipeline inlet pressure and outlet flow rates as the training data for the Fault Detection and Isolation (FDI) system. The FDI system is comprised of a Multi-Layer Perceptron Neural Network (MLPNN) classifier with various feature extraction methods including the statistical techniques, wavelet transform, and a fusion of both methods. Once different leakage scenarios are considered and the preprocessing methods are done, the proposed FDI system is applied to a 20-km pipeline in southern Iran (Goldkari-Binak pipeline) and a promising severity and location detectability (a correct classification rate of 92%) and a low False Alarm Rate (FAR) were achieved.