시스템 해석에 기초한 하수관망 오염 매카니즘과 관망 모니터링 및 이상진단

To effectively manage water resources and safeguard food security, it is essential to comprehend the impact of climate change and human activities on non-point source (NPS) pollution within agricultural watersheds. This study utilized SWAT+ to model the water balance and nutrient balance of the Xiaowei River Basin (XRB) in Shaanxi Province, China. Different scenarios were used to quantify the effects of human activities and climate on NPS pollution loads. Model validation achieved R2 values of 0.87 (streamflow) and 0.71 (total nitrogen load), indicating good performance. The baseline period (1998-2023) was divided into four evolution scenarios. Results showed climate dominated total nitrogen (TN) load contributions (average 93.6%), while human activities contributed 6.4%. However, human activities increased from 4.5% (1998-2003) to 9.7% (2018-2023), increasing TN load. TN loads decreased relative to the initial scenario, due to reduced precipitation. Future pollution loads were simulated using CMIP6 data (five GCMs) and projected population and LULC. Under SSP2-4.5, TN loads increased (2024-2040) and then decreased. Under SSP5-8.5, TN loads exhibited a consistent upward trend, driven by agricultural land expansion and reduced precipitation. Human activities' contribution is continually increasing. Projections indicate TN load under Best Management Practices (BMPs) is lower than that in other scenarios across all timeframes. Notably, in the long term (2071-2100), TN load under BMPs is lower than the baseline. Relevant decision-makers may consider implementing Best Management Practices (BMPs) such as precision fertilization and the establishment of vegetative buffer strips, which can help mitigate the effects of human activities.

In high-speed railway signaling systems, the network structure of the Train Control Center and Inter-locking Integrated System (TIS) is highly complex, with a large number of interfaces, numerous redundant channels, and forwarding components such as switches. These factors result in challenges such as insufficient accuracy, low efficiency, and poor real-time performance in terms of network monitoring and fault diagnosis. As the scale of railway yards continues to expand, these issues are becoming increasingly prominent. To address these challenges, this paper proposes a network fault propagation model based on the Fuzzy Broad Learning System (FBLS). By leveraging nonlinear transformations and feature mapping techniques, FBLS can efficiently extract and analyze network fault features, even with a relatively small amount of data. Experimental results show that the FBLS-based diagnostic model achieves higher accuracy and faster response speed in fault identification and propagation path analysis compared to traditional graph theory and fuzzy reasoning methods. Further comparisons with existing methods validate the advantages of FBLS in handling multi-source data, improving noise tolerance, and adapting to large-scale railway yard network systems, demonstrating its broad application prospects in railway signaling systems.

With the deepening advancement of smart manufacturing, regular fault diagnosis plays a crucial role in optimizing productivity and maintaining equipment safety and stability. Joint domain adaptation (JDA) techniques for fault diagnosis have garnered significant attention; however, their ability to characterize nonlinear, high-dimensional features remains limited. Additionally, low pseudo-label confidence levels result in insufficient distribution alignment, leading to domain drift that compromises cross-domain fault diagnosis performance. Based on this, this paper proposes a multi-kernel higher-order moment matching joint domain adaptive network (MKHoJDA) for cross-domain mechanical fault diagnosis. Firstly, a multi-scale dilatable residual convolution (MDRC) module is proposed to extract transferable features from both the source and target domains more accurately. Secondly, to address the limitation of traditional JDA in statistically measuring higher-order features, a multi-kernel higher-order moment matching (MKHoMM) statistical metric strategy is designed. Multiple kernel functions are combined to create a more expressive composite kernel, enhancing the power of feature mappings, reducing distribution discrepancies between domains. Additionally, to address the dependency of conditional distribution alignment (CDA) on pseudo-label accuracy, a prototype-aware pseudo-label feedback (PPF) mechanism was developed. This mechanism evaluates sample confidence using Jensen-Shannon (JS) divergence, selects high-confidence target domain samples for pseudo-label generation, and dynamically updates prototype centers. This approach effectively mitigates the negative impact of pseudo-labels while improving CDA performance. Finally, multiple cross-domain fault diagnosis experiments were conducted on the JNU and PU datasets. The average accuracy significantly surpassed that of other advanced comparative models, demonstrating that the proposed algorithm effectively mitigates the impact of data distribution differences and exhibits exceptional cross-domain fault diagnosis performance.

To solve the problem of fault prediction and diagnosis of household air conditioning, an improved Bayesian network (BN) fault diagnosis model is proposed. Firstly, the orthogonal defect classification (ODC) is used to make statistics and analysis of air conditioning fault data, and the structure of BN fault diagnosis model is built based on the analysis results. Then, genetic algorithm (GA) is used to optimize the conditional probability of network nodes and determine the network parameters. Finally, the cooling and heating failure data of household air conditioning were taken as an example to diagnose. Compared with the traditional BN model, the accuracy of fault diagnosis increases from 81.13% to 92.83%, which verifies the effectiveness of the model.

To investigate the impact of well point layout in pipe networks on surface inundation and rainwater/sewage discharge during the reconstruction of rainwater and sewage shunting, a coupled simulation using the stormwater management model and the graphics processing unit-accelerated surface water flow and associated transport model was conducted to compare the overall impact of different new pipe network methods. The study assesses the effects of well point distances and pipe network diameters on the overall system in different rainfall return periods. Research shows that, irrespective of whether new rainwater pipes or new sewage pipes are implemented, the transformed schemes involving different well point distances and pipe diameters yield positive effects compared to the pretransformation state. The solution that uses original combined pipe as a sewage pipe and builds a new rainwater pipe is more effective in eliminating surface inundation. Notably, the approach utilizing the original combined pipe as a sewage pipe while introducing a new rainwater pipe proves more effective in mitigating surface inundation. Specifically, a pipe network with a well point distance of 30 m and a pipe diameter of 1,000 m demonstrates the most effective results in preventing inundation among the two transformation methods. By comprehensively considering the influence of well point layout in the pipe network across various rainwater and sewage diversion reconstruction strategies, this study aims to offer theoretical support for the revitalization of old city pipe networks.

This study proposes an algorithm for the design of a distributed protocol for detecting and diagnosing permanent faults in wireless image sensor networks (WISNs). This work explicitly takes into account the possibility of faults in different sections of the image sensor node and communication channel. The fault diagnosis is achieved by disseminating local decision made at each node. Theoretical analysis and simulation results show that high level (>0.95) of diagnosis accuracy can be achieved for a wide range of fault rates. Correctness and complexity proofs are provided, and they show that the proposed algorithm performs better than the existing protocols from both communication and energy complexity perspective. Analytical study shows that time complexity is O ( n c ) and message complexity O ( n ) for an n -node WISN where n c is the number of cluster heads.

The urban area of Wanzhou District in Chongqing frequently experiences flooding and sewage overflow pollution due to its steep topography, frequent and intense short-duration rainstorms, and inadequate drainage capacity. To address these issues, this study integrated sponge city concepts with low-impact development (LID) measures to establish a stormwater management framework for the Guoben Road catchment area. Hydrometeorological and drainage data were obtained from local monitoring stations. Utilizing high-resolution remote sensing imagery and a 30 m-resolution digital elevation model, 50 sub-catchments were delineated in ArcGIS. Based on regional soil information and national sponge city design guidelines, a Stormwater Management Model employing the Horton infiltration method was constructed. This model simulated runoff and sewer network performance under Chicago design storm conditions with return periods of 1, 3, 5, and 10 years. Results indicate that LID facilities (permeable pavements and rain gardens) substantially reduced runoff generation and enhanced drainage system efficiency. Across the four rainfall scenarios, the total decrease in runoff volume ranged from 69.3% to 79.3%, with runoff coefficients falling from 0.51–0.61 to 0.10–0.19. The decrease in outfall discharge ranged from 11.0% to 76.1%, with peak flow reduced by up to 70.4%, most notably during the 1-year return-period event. Mitigation effects diminished during higher-return-period storms, attributed to limited LID storage capacity and rapid soil saturation, which reduced infiltration and detention capabilities during intense rainfall events. Overall, this integrated LID scheme effectively enhanced infiltration capacity within the mountainous urban catchment, reduced peak flows, and alleviated sewer network overloading. This study provides technical support for stormwater risk management, sewage overflow pollution control, and sponge city planning in Wanzhou District, Chongqing, and other cities with similar topographic and hydrological characteristics.

Determination of Optimal Location and Capacity of Detention Facilities

Low impact development (LID) techniques are increasingly used as a stormwater management strategy to maintain the hydrological conditions of developed areas and mitigate the negative consequences of stormwater runoff and nonpoint source pollution. Although LID techniques have been commonly used in moderate to humid areas, further information is needed on their effectiveness in semi-arid and arid regions. This study aims to examine and compare the effectiveness of LID techniques in arid and semi-arid climate conditions. First, a comprehensive study was conducted to rank LID techniques based on literature reviews while also incorporating different stakeholder priorities. Then, the city of Varamin, Tehran, Iran, was chosen as a low slope arid and semi-arid study region to assess the applicability of the best three high-ranked LID techniques using the storm water management model (SWMM). The results indicated that rainwater harvesting is the most effective technique in terms of stormwater quality and quantity management. The implementation of the rainwater harvesting technique across the case study is likely to decrease the average discharge, peak discharge, total volume, total runoff, and total suspended solids (TSS) by respectively 31.2%, 12.7%, 40.71%, 40.77% and 37.91%. Besides, rainwater tanks were projected to provide the water demands of home gardens during the five dry months, in addition to other domestic needs for water conservation objectives. The application of LID techniques in such water-limited regions can be useful for both stormwater management and water conservation.

Uncertainty assessment in baseflow nonpoint source pollution prediction: The impacts of hydrographic separation methods, data sources and baseflow period assumptions

Rapid urbanization has caused widespread environmental pollution in Africa, restricting sustainable development programs. Today, non-point source pollution has become a major threat to many African cities. In this study, we have classified urban land use followed by the identification of urban functional areas based on multi-source data. We have then evaluated non-point source pollution based on nitrogen and phosphorus loads at a spatial scale in Mwanza city (Tanzania). We have estimated the spatial distribution of total nitrogen (TN) and total phosphorus (TP) loads at a scale of 1 square kilometer grid. The results show that ① the built-up area of the city was identified as 28.9 km2, accounting for 6.8% of the total area of the city. The area of informal settlements was 56.9 km2, accounting for 13.4% of the total area of the city. About 90.38% of 2.41×105 kg of the total TN load and 87.23% of 4.00×104 kg of the total TP load in the city were derived from the informal settlements alone in 2020. ② The average load intensities of TN and TP in Mwanza city were 6.90 kg·hm-2·yr-1 and 1.14 kg·hm-2·yr-1, respectively. The average TN and TP intensities in Mwanza city were slightly higher than the average TN and TP load intensities in the Lake Victoria basin. However, given the high population density of the Mwanza city area, the average TN and TP intensities were found to be consistent. ③ The TN and TP pollution loads were relatively concentrated in the inner urban area, including the lower reaches of the Mirongo River basin, the eastern side of the Nyamagana Peninsula, and the lakeside area extending north and south from the peninsula, and gradually decreased toward the city’s outskirts. We propose that the identified riparian areas in our study should be prioritized as potentially key construction zones for pollution treatment plant facilities and urban renewal pilot programs for restraining the widespread informal settlements. This study contributes to the development of an innovative urban non-point source pollution control measure, which may be applicable to low-income countries in Africa and elsewhere.

Faults in the optical network environment propagate quickly to different parts of the network raising a large number of alarms. This paper describes two new fault diagnosis techniques for single fault diagnosis in an optical network. Both techniques operate under the same premises: when a fault occurs, a central manager detects the fault and through tests and correlation of information identifies the failure. The performance of the proposed schemes is compared both analytically and by simulation to each other and other proposed schemes for fault diagnosis in the optical network environment.

Effectiveness and resilience of BMPs to watershed climate adaptation considering the uncertainty of hydrological model and GCMs

With the development of big data, artificial intelligence and other technologies, data-driven aviation equipment fault diagnosis and prediction technology has gradually become a research hotspot in the aviation field. Many typical intelligent algorithm models have been applied to this field. However, limited by the airborne embedded computing environment, there are still some problems in the deployment of intelligent prediction models represented by deep neural networks on aircraft. This paper summarizes and analyzes the research and application of typical deep neural networks such as convolutional neural networks in the field of aircraft fault diagnosis and prediction. Facing the airborne embedded environment, the current difficulties in deploying the deep neural network algorithm model in the airborne environment are analyzed. The development direction of the application of fault prediction and diagnosis algorithms represented by neural networks in the future is discussed.

Spatiotemporal variation in nitrogen loads and their impacts on river water quality in the upper Yangtze River basin