During the deep peak shaving period, the boiler needs to operate at a lower load, so higher requirements were set for the boiler's stable combustion. In order to better evaluate and compare the... Show More

During the deep peak shaving period, the boiler needs to operate at a lower load, so higher requirements were set for the boiler's stable combustion. In order to better evaluate and compare the stable ignition capacity of bluff-body burners and slotted bluff-body burners, guidance and calculation support for the design of boiler deep peak shaving were needed. This study adopted the reflux heating chain ignition analysis method to study the differences in the steady combustion mechanism between the bluff-body burner and the slotted bluff-body burner. A thermal combustion experiment was conducted in employing the single angle coal powder combustion furnace. The experimental results were compared against the theoretical analysis results. In addition, a study was conducted on the application of slotted bluff-body burners in the deep peak shaving of a 330MW unit power plant boiler. The results showed that as long as the small slot flow can catch fire, the mixed temperature of the reflux fluid of the slotted bluff-body burner will be higher than that of the bluff body burner. This will enhance its steady combustion ability. The combustion test results were found to be consistent with the analysis results, indicating that when the slotted bluff-body burner is used on the 330MW unit, the boiler combustion is in good condition. In this case, it has a stable combustion capacity of 66MW (20% economic continuous rating) without oil injection at low loads. This study revealed the advantages of the slotted bluff-body burner in terms of its steady combustion mechanism compared to traditional bluff-body burners. It verified the feasibility for boiler deep peak shaving in practical applications. This is of great significance for improving the flexibility of coal-fired power generation units, enhancing the flexibility of the power system, and promoting carbon reduction.

To improve transportation efficiency, this paper analyzes the factors of transportation structure from the two levels of transportation—input and system output. An epsilon-based measure model of... Show More

To improve transportation efficiency, this paper analyzes the factors of transportation structure from the two levels of transportation—input and system output. An epsilon-based measure model of non-expected output is introduced, and the environmental benefits of transportation are considered. This model is used to analyze the regional transportation efficiency of 30 provinces and cities in China. Tobit regression and geographically weighted regression are applied to analyze the causes and spatial variation of differences in the efficiency of the transportation structure, and corresponding structural adjustment strategies are proposed. The results show that the regression coefficients of the share of secondary industry output in GDP, population density, and social fixed asset investment exert the most significant effects on transportation structure efficiency. The spatial distribution of sub-variable coefficients shows that spatial heterogeneity exists in the degree of influence of various socio-economic factors on the transportation structure efficiency in different regions.

To address issues related to the insufficient representation of text semantic information and the lack of deep fusion between internal modal information and intermodal information in current... Show More

To address issues related to the insufficient representation of text semantic information and the lack of deep fusion between internal modal information and intermodal information in current multimodal sentiment analysis (MSA) methods, a new method integrating multi-layer attention interaction and multi-feature enhancement (AM-MF) is proposed. First, multimodal feature extraction (MFE) is performed based on RoBERTa, ResNet, and ViT models for text, audio, and video information, and high-level features of the three modalities are obtained through self-attention mechanisms. Then, a cross modal attention (CMA) interaction module is constructed based on transformer, achieving feature fusion between different modalities. Finally, the use of a soft attention mechanism for the deep fusion of internal and intermodal information effectively achieves multimodal sentiment classification. The experimental results CH-SIMS and CMU-MOSEI datasets show that the classification results of proposed MSA method are significantly superior to other advanced comparative methods.

In this study, the authors perform slant correction on fixed-format dial images using the Hough transform, followed by their “scaling” using binary wavelets to achieve coarse segmentation for... Show More

In this study, the authors perform slant correction on fixed-format dial images using the Hough transform, followed by their “scaling” using binary wavelets to achieve coarse segmentation for characters or numbers. Simultaneously, they propose a binary threshold iteration method that accurately determines the position of each character or number even in the presence of adhesion or fragmentation, enabling precise segmentation. They employ their proposed approach to segment digits and characters displayed on 98 fixed-format dials. Experimental results demonstrate a recognition rate of 98.5% for both letters and numbers, highlighting significant practical value and real-world implications.

The stability and reliability of the power system are of utmost significance in upholding the smooth functioning of modern society. Fault diagnosis and prediction represent pivotal factors in the... Show More

The stability and reliability of the power system are of utmost significance in upholding the smooth functioning of modern society. Fault diagnosis and prediction represent pivotal factors in the operation and maintenance of the power system. This article presents an approach employing graph neural network (GNN) to enhance the precision and efficiency of power system fault diagnosis and prediction. The system's efficacy lies in its ability to capture the intricate interconnections and dynamic variations within the power system by conceptualizing it as a graph structure and harnessing the capabilities of GNN. In this study, the authors introduce a substitution for the pooling layer with a convolution operation. A central role is played by the global average pooling layer, connecting the convolution layer and the fully connected layer. The fully connected layer carries out nonlinear computations, ultimately providing the classification at the top-level output layer. In experiments and tests, we verified the performance of the system.

This integration enables the system to collect and monitor information from remote sources efficiently. During the course of this research, a novel predictive PID approach was developed, splitting... Show More

This integration enables the system to collect and monitor information from remote sources efficiently. During the course of this research, a novel predictive PID approach was developed, splitting the control architecture into two tiers. The upper tier utilizes the extreme learning machine (ELM) as an intelligent predictive model, while the lower tier integrates an enhanced single-neuron adaptive predictive PID control algorithm, combining the strengths of ELM and PID control. The research findings suggest that the AI algorithm-based instrument automatic monitoring and control system holds significant promise. This technology has the potential to enhance production efficiency, reduce energy consumption, improve environmental monitoring, and provide superior safety and quality control.

Chinese financial institutions (CFIs) are increasingly embracing artificial intelligence (AI) for their financial decision-making driven by AI's capacity to mitigate risks and enhance efficiency and... Show More

Chinese financial institutions (CFIs) are increasingly embracing artificial intelligence (AI) for their financial decision-making driven by AI's capacity to mitigate risks and enhance efficiency and accuracy. However, there remain ethical challenges related to the integration of AI in financial decision-making. This study develops the AI ethics best practices model (AB-PraM) to mitigate ethical concerns and enhance the application of AI in financial decision-making. By employing a quantitative methodology, this research collected questionnaire data from 320 financial experts in CFIs. Structural equation modelling (SEM) was adopted to identify AI ethics best practices for the implementation of the AB-PraM. The findings of this research will mitigate AI ethics challenges in CFIs and provide a practical framework for transparent and accountable decision-making in alignment with ethical standards and regulations.

In order to solve the problems that most models are complex, time-consuming, and have difficulty in identifying image errors, an image identification and error correction method of test report based... Show More

In order to solve the problems that most models are complex, time-consuming, and have difficulty in identifying image errors, an image identification and error correction method of test report based on deep reinforcement learning and the internet of things platform in the smart lab was proposed. Firstly, a smart lab architecture was designed based on the internet of things platform, achieving efficient operation of the laboratory through cloud edge collaboration. Then, the depth separable convolution improved convolutional neural network is used to extract image features, and the features are input into bidirectional recurrent neural networks (BiLSTM) for analysis to complete image recognition. Finally, the ICNN-BiLSTM model is used as the agent of reinforcement learning, and image error correction is completed by identifying the distance between the image and the key points of the reference image. Based on the Python platform, the proposed method was experimentally demonstrated, and the results showed that its average error correction accuracy reached 96.75%, with a processing time of 15.37s.

Traditional methods often fall short in modeling the nonlinear, seasonally variable nature of urban water demand. This proposed solution is an integrated ARIMA-LSTM deep learning model, combining... Show More

Traditional methods often fall short in modeling the nonlinear, seasonally variable nature of urban water demand. This proposed solution is an integrated ARIMA-LSTM deep learning model, combining ARIMA's proficiency in linear trend and seasonal modeling with LSTM's strength in capturing nonlinear time dependencies. In these experiments, the authors trained and evaluated using daily water demand data from 2015 to 2020, with its performance validated for the year 2021. The ARIMA-LSTM model demonstrates promising results, outperforming individual models in terms of accuracy. In validation, it achieves a high coefficient of determination (R2) of 0.98 and a significantly low root mean square error (RMSE) of 2.94. These metrics indicate an excellent fit to the data and a high level of precision in its predictions. The significance of this research lies in its potential to advance the field of urban water demand forecasting, ultimately contributing to better water resource management and sustainability in urban areas.

Bitcoin is a digital currency system built on the foundation of fairness. However, some malicious miners, driven by their own interests, employ unfair tactics such as selfish mining to compete... Show More

Bitcoin is a digital currency system built on the foundation of fairness. However, some malicious miners, driven by their own interests, employ unfair tactics such as selfish mining to compete, which disregard the legitimate miners' investments in computational power and energy consumption. In order to assess the efficiency of the Bitcoin system in real-time and promptly detect malicious miners in the network, this paper proposes a data collection framework called BitTrace, which addresses the issues of low efficiency, lack of timeliness, and data loss in traditional data collection frameworks. BitTrace enables real-time collection and analysis of the blockchain formation process, storing it as structured data. Furthermore, the paper discusses factors that influence the efficiency of data collection and proposes a topological control scheme based on the DPC algorithm to enhance the integrity and efficiency of data collection. Researchers can explore various research areas and applications, such as selfish mining detection and legitimate mining strategy research.

Existing models still exhibit a deficiency in capturing more detailed contextual information when processing architectural images. This paper introduces a model for architectural image segmentation... Show More

Existing models still exhibit a deficiency in capturing more detailed contextual information when processing architectural images. This paper introduces a model for architectural image segmentation and retrieval based on an image segmentation network. Primarily, spatial attention is incorporated into the U-Net segmentation network to enhance the extraction of image features. Subsequently, a dual-path attention mechanism is integrated into the U-Net backbone network, facilitating the seamless integration of information across different spaces and scales. Experimental results showcase the superior performance of the proposed model on the test set, with average dice coefficient, accuracy, and recall reaching 94.67%, 95.61%, and 97.88%, respectively, outperforming comparative models. The proposed model can enhance the U-Net network's capability to identify targets within feature maps. The amalgamation of image segmentation networks and attention mechanisms in artificial intelligence technology enables precise segmentation and retrieval of architectural images.

Aimed at the problem of large errors in traditional power cable temperature measurement methods, a method based on edge computing and radio frequency identification (RFID) is proposed. Firstly, a... Show More

Aimed at the problem of large errors in traditional power cable temperature measurement methods, a method based on edge computing and radio frequency identification (RFID) is proposed. Firstly, a RFID electronic label design scheme was constructed utilizing the radio frequency signal between alternating electromagnetic fields to achieve the communication and information identification between two devices. Then, the design of power cable edge intelligent terminal is analyzed from the aspects of hardware and software. On this basis, the early warning criterion formula of power cable temperature fault state is abstracted by using edge computing algorithm. Based on edge computing and RFID, the corresponding temperature measurement method is proposed. Finally, the proposed power cable temperature measurement method is compared with other three methods through simulation experiments. The results indicate that the temperature measurement error of the proposed method is the minimum under different currents and in cross-sectional areas. Compared to the other three methods, the maximum improvement is 5.15% and 7.92%, while the minimum improvement is 1.05% and 1.45%, which outperforms the comparable algorithms in terms of performance.

Multi-depot vehicle routing problem with time windows (MDVRPTW) is a valuable practical issue in urban logistics. However, heuristic methods may fail to generate high-quality solutions for massive... Show More

Multi-depot vehicle routing problem with time windows (MDVRPTW) is a valuable practical issue in urban logistics. However, heuristic methods may fail to generate high-quality solutions for massive problems instantly. Thus, this article presents a novel reinforcement learning algorithm integrated with a multi-head attention mechanism and a local search strategy to solve the problem efficiently. The routing optimization was regarded as a vehicle tour generation process and an encoder-decoder was used to generate routes for vehicles departing from different depots iteratively. A multi-head attention strategy was employed for mining complex spatiotemporal correlations within time windows in the encoder. Then, a decoder with multi-agent was designed to generate solutions by optimizing reward and observing transition state. Meanwhile, a local search strategy was employed to improve the quality of solutions. The experiments results demonstrate that the proposed method can significantly outperform traditional methods in effectiveness and robustness.

The presence of structure heterogeneity in regional innovation networks reflects the complexity and diversity of knowledge diffusion and collaborative R& D relationships. This article introduces a... Show More

The presence of structure heterogeneity in regional innovation networks reflects the complexity and diversity of knowledge diffusion and collaborative R& D relationships. This article introduces a network model based on the multiple systems generating functions mathematical algorithm to analyze the resilience of interacting networks under different link patterns. The percolation threshold is illustrated at two different levels: the subcritical and supercritical states. The algorithm is then tested on both simulated networks and real-world networks. The results of the simulation study highlight the crucial role of linking between sub-networks and emphasize the effectiveness of a moderate degree protection strategy.

Aiming at the problem that the environmental planning path of intelligent logistics vehicles on urban roads and remote mountainous areas cannot fit the actual driving scene well. This study creates... Show More

Aiming at the problem that the environmental planning path of intelligent logistics vehicles on urban roads and remote mountainous areas cannot fit the actual driving scene well. This study creates the algorithm model that combines an ant colony algorithm with a dynamic window algorithm and a Bessel smoothing strategy. Compared to the traditional colony algorithm with the same parameters, this fusion algorithm makes the path smoother by 72.2% when used on an urban highway. It also follows the right-hand rule for right-turn intersections. When the vehicle's height is determined in a mountain environment, this fusion algorithm reduces the driving's mean square deviation of height by 81.5% and shortens the path distance by 38.7%. The fusion algorithm can plan the target path of intelligent logistics vehicles and has the characteristics of scenarios available, multiple factors coordinated, and driving safety. It has provided certain research value and ideas for the digital transformation of the logistics industry.