An Accurate Leakage Localization Method for Water Supply Network Based on Deep Learning Network

In the water supply network, leakage of pipes will cause water loss and increase the risk of environmental pollution. For water supply systems, identifying the leak point can improve the efficiency of pipeline leak repair. Most existing leak location methods can only locate the leak point approximately to the node or pipe section of the pipe network, but cannot locate the specific location of the pipe section. This paper presents a framework for accurate location of water supply network leakage based on ResNet. The framework is to pinpoint leaks to specific locations along the pipeline. The leakage of two kinds of pipe networks is simulated. For a pipe network containing 40 pipes, the positioning accuracy of the pipe section is 0.94, and the MSE of the specific location of the leakage point is 0.000435. For the pipe network containing 117 pipes, the positioning accuracy of the pipe section is 0.91, and the MSE of the specific location of the leakage point is 0.0009177, and the leak location ability under different sensor arrangements is analyzed. Experiments verify the robustness and applicability of the framework.

Review of Research on System Thermal Inertia Modelling and Optimal Scheduling from the Perspective of Flexible Resources

The thermal inertia of the existing heating system is often considered an adverse factor, which will affect the operation of the system. However, under the perspective of resource flexibility, the thermal inertia of the system can effectively increase the flexibility of the system operation, significantly reduce the energy consumption and enhance the ability of energy supply and demand balance, and enhance the new energy integration, such as the wind power. Based on the flexible resources, it focuses on the study of the thermal inertia of the “network side” heating pipe network of the system and the optimal scheduling of the heating system. Combined with the thermal inertia of the pipe network, the operation characteristics of the power/heat output of the gas-steam combined cycle unit were analysed theoretically. On this basis, the optimal scheduling model of the system was established. Taking the energy supply system of an industrial park as an example, the model was verified to achieve a more stable power output effect of the unit.

Method of detection of district heating pipe network leakage using data monitoring of heat energy consumers

Currently, information systems to get data of metering devices are introduced to calculate the consumed thermal energy. The metering devices are installed at the thermal station of the consumers. However, the processing of these data is usually limited to the monthly data collection to calculate the payments and to monitor the output of the observed parameters beyond the established boundaries. The urgent issue is the possibility to use these data for the in-depth study of the processes, and, in particular, to detect district heating pipe network leakage. The authors have used both the methods and tools to model and analyze the operating modes of district heating pipe networks, methods to collect and monitor data of heat supply metering devices, methods to model district heating pipe networks in the geoinformation systems environment. The authors have proposed the method to detect the sections of the heat network where a heat medium leak has occurred. The difference of the method is the use only of the readings of the metering devices installed at consumers. The limitations of the application of the method and its implementation in geoinformation system environment are considered. An example is given to illustrate the possibility to detect the location of leakage based on the analysis of real data of the house heat metering devices collected during leakage and leakage elimination. Practical application of the developed method is discussed by the example of a real situation of leakage at the section of the heat network of the ISPU boiler house. The results obtained have confirmed the possibility to detect localization of leakage in heating networks based on the analysis of meter readings installed at consumers. The developed method can be applied in information systems to monitor the operating modes of district heating networks to search the places of accidents.

Application of networked water balance model in refined management of steel industrial park

Water shortages has become a major constraint on China's industrial development. Iron and steel industrial parks have a huge demand for water resources and complex production technologies. Therefore, it is very important to study the distribution, transfer and loss of water resources in industrial parks in order to improve the ability of refined water resources management. The purpose of this study is to reveal the water flow in industrial parks by using the principle of water balance and to provide a method for quantification and characteristic recognition of water resources in industrial production processes. In this research, an iron and steel industrial in North China was chosen as the case study. In order to calculate the water balance of the whole steel production processes, the industrial park was divided into 4 levels and 110 water units according to the pipe network system and production processes. Based on the results of multi-level and multi-node water balance, this paper analyzed the water intake structure and water consumption structure of industrial parks, and provided the methods to optimize the allocation of water resources and reduce the consumption of fresh water in industrial production process. The results of the study showed that the energy department accounted for 60.8% of the total water withdrawal of the industrial park. There was 6,249 m3/day of fresh water in the industrial park, which can be replaced by reclaimed water from urban sewage. Evaporation and pipe network leakage were the main water consumption factors in the steel park, which contribute 91.3% of the water consumption. Under the guidance of the research results, the evaporation water consumption of the industrial park was reduced by 8,412 m3/day, and pipe leakage was reduced by 600 m3/day. This article demonstrates the application of water balance principle in complex water use systems, which is helpful for water resources management based on water use process.

Study on Construction Resource Optimization and Uncertain Risk of Urban Sewage Pipe Network

With considering sewage pipe network upgrading projects in the “villages” in cities, the optimization of construction resources and the assessment of delay risks could be achieved. Based on the schedule-cost hypothetical theory, the mathematical model with constraint indicators was established to obtain the expression of optimal resource input, and conclude the method to analyze the schedule uncertainties. The analysis showed that cyclical footage of pipe could be regarded as a relatively fixed value, and the cost can be regarded as a function that depending on the number of working teams. The optimal number of teams and the optimal schedule occurred when the minimum total cost achieved. In the case of insufficient meteorological data, the Monte Carlo simulation method and uncertainty analysis method can be applied to assess the impact of rainfall on the total construction period, correspondingly the probability of such risk could be derived. The calculation showed that the risk of overdue completion varied significantly according to the construction starting time. It was necessary to take rainfall risk into consideration and make corresponding strategies and measures.

Approximate approach for improving pressure attenuation accuracy during hydraulic transients

The quasi-steady friction model is generally adopted in water hammer simulation in pipe network systems, which cannot accurately reflect the attenuation of pressure, while the existing unsteady friction model is challenging to use in complex pipe network systems. In this study, a convenient method for treating the friction term is proposed based on the Moody diagram. The attenuation process of water hammer pressure can be accurately reflected by reading the relationship curve between Reynolds number and the Darcy friction factor in the pipeline transient process. Combined with the classical water hammer experiment and the long pipe valve closing experiment in our laboratory, the accuracy of this model is verified, and the influence of absolute roughness (e) and Reynolds number (Re) on the model was analyzed as well. The results show that the pressure attenuation using the Method of Characteristics (MOC) and the proposed friction model has a good agreement with the experimental data. The absolute roughness has little influence on the results in hydraulically smooth pipe, while the minimum Reynolds number has a significant influence. When selecting the minimum Reynolds number, 2% ∼ 5% of the initial flow rate is recommended for calculation.

Analysis of Mixed Water Source for Urban Underground Water Pipe Leakage by Using Water Geochemistry Model Calculation

China is a country short of water resources, and the leakage of urban water pipe network not only aggravates the current situation of water shortage, but also causes major accidents such as ground collapse, so it is of great significance to study the discrimination of urban underground pipe leakage. In this paper, the conventional ions and hydrogen and oxygen isotopes of water samples are determined by ion chromatograph and inductively coupled plasma mass spectrometer, and the characteristic factors are selected by cluster analysis and principal component analysis, and the mixed water discrimination model based on conventional ions is established According to the difference of hydrogen and oxygen isotope content between buried pipe water and groundwater, a discrimination model based on hydrogen and oxygen isotope is established, and the two models are combined to discriminate the leakage of buried pipe. The results show that, in terms of conventional ion content characteristics, the water in the pipe network is high in K++Na+ and Cl−, while the shallow groundwater near the pipe network is low in K++Na+ and Cl−, and the accuracy of the discriminant model based on conventional ions reaches 87.5%. In the aspect of hydrogen and oxygen isotope content characteristics, the water in the pipe network is closer to the precipitation line than the shallow groundwater, and establishing a discriminant model based on hydrogen and oxygen isotope can determine the leakage of buried pipes. This study provides a scientific basis for judging the leakage of urban underground pipes.

Application of Geological Radar Method in Exploration of Reservoir Complex Pipelines

Geodesic radar method is a non-destructive geophysical method to detect underground geological structure based on the differences in conductivity, magnetism and dielectric constants of different substances. This method has the advantages of economy, fast, continuous, accurate, non-destructive, non-contact, etc. It is applicable to both metal and non-metal pipelines, and it is the preferred tool for underground pipeline detection. In this paper, the geological radar method is used to detect the complex pipe network around Daning Reservoir in detail by laying out multiple detection lines. It provides basic support for the later pipe network rectification. The research results can provide experience for the detection of the complex pipe network in the urban reservoir.