Search Results (1,475)

In the Internet of Things (IoT) era, Mobile Edge Computing (MEC) significantly enhances the efficiency of smart devices but is limited by battery life issues. Wireless Power Transfer (WPT) addresses this issue by providing a stable energy supply. However, effectively managing overall energy consumption remains a critical and under-addressed aspect for ensuring the network’s sustainable operation and growth. In this paper, we consider a WPT-MEC network with user cooperation to migrate the double near–far effect for the mobile node (MD) far from the base station. We formulate the problem of maximizing long-term computation rates under a power consumption constraint as a multi-stage stochastic optimization (MSSO) problem. This approach is tailored for a sustainable WPT-MEC network, considering the dynamic and varying MEC network environment, including randomness in task arrivals and fluctuating channels. We introduce a virtual queue to transform the time-average energy constraint into a queue stability problem. Using the Lyapunov optimization technique, we decouple the stochastic optimization problem into a deterministic problem for each time slot, which can be further transformed into a convex problem and solved efficiently. Our proposed algorithm works efficiently online without requiring further system information. Extensive simulation results demonstrate that our proposed algorithm outperforms baseline schemes, achieving approximately 4% enhancement while maintain the queues stability. Rigorous mathematical analysis and experimental results show that our algorithm achieves $\left[O(1/V),O\left(V\right)\right]$ trade-off between computation rate and queue stability. Full article

In this work, we explore the application of deep reinforcement learning (DRL) to algorithmic trading. While algorithmic trading is focused on using computer algorithms to automate a predefined trading strategy, in this work, we train a Double Deep Q-Network (DDQN) agent to learn its own optimal trading policy, with the goal of maximising returns whilst managing risk. In this study, we extended our approach by augmenting the Markov Decision Process (MDP) states with sentiment analysis of financial statements, through which the agent achieved up to a 70% increase in the cumulative reward over the testing period and an increase in the Calmar ratio from 0.9 to 1.3. The experimental results also showed that the DDQN agent’s trading strategy was able to consistently outperform the benchmark set by the buy-and-hold strategy. Additionally, we further investigated the impact of the length of the window of past market data that the agent considers when deciding on the best trading action to take. The results of this study have validated DRL’s ability to find effective solutions and its importance in studying the behaviour of agents in markets. This work serves to provide future researchers with a foundation to develop more advanced and adaptive DRL-based trading systems. Full article

Segmenting unseen object instances in unstructured environments is an important skill for robots to perform grasping-related tasks, where the trade-off between efficiency and accuracy is an urgent challenge to be solved. In this work, we propose a fast unseen object instance segmentation (Fast UOIS) method that utilizes predicted center offsets of objects to compute the positions of local maxima and minima, which are then used for selecting initial seed points required by the mean-shift clustering algorithm. This clustering algorithm that adaptively generates seed points can quickly and accurately obtain instance masks of unseen objects. Accordingly, Fast UOIS first generates pixel-wise predictions of object classes and center offsets from synthetic depth images. Then, these predictions are used by the clustering algorithm to calculate initial seed points and to find possible object instances. Finally, the depth information corresponding to the filtered instance masks is fed into the grasp generation network to generate grasp poses. Benchmark experiments show that our method can be well transferred to the real world and can quickly generate sharp and accurate instance masks. Furthermore, we demonstrate that our method is capable of segmenting instance masks of unseen objects for robotic grasping. Full article

Federated learning (FL) is a decentralized machine learning approach whereby each device is allowed to train local models, eliminating the requirement for centralized data collecting and ensuring data privacy. Unlike typical typical centralized machine learning, collaborative model training in FL involves aggregating updates from various devices without sending raw data. This ensures data privacy and security while collecting a collective learning from distributed data sources. These devices in FL models exhibit high efficacy in terms of privacy protection, scalability, and robustness, which is contingent upon the success of communication and collaboration. This paper explore the various topologies of both decentralized or centralized in the context of FL. In this respect, we investigated and explored in detail the evaluation of four widly used end-to-end FL frameworks: FedML, Flower, Flute, and PySyft. We specifically focused on vertical and horizontal FL systems using a logistic regression model that aggregated by the FedAvg algorithm. specifically, we conducted experiments on two images datasets, MNIST and Fashion-MNIST, to evaluate their efficiency and performance. Our paper provides initial findings on how to effectively combine horizontal and vertical solutions to address common difficulties, such as managing model synchronization and communication overhead. Our research indicates the trade-offs that exist in the performance of several simulation frameworks for federated learning. Full article

The present article aims to determine the green economic policies of an inventory model for non-instantaneous deteriorating items under practical scenarios. These scenarios involve specific maximum lifetimes for items with deteriorations controllable through investments in preservation technologies, which can affect the period without deterioration. Additionally, carbon is emitted due to energy-related costs, prompting retailers to invest in green technology investments to reduce carbon emissions concurrently under the carbon tax policy and the carbon cap-and-trade policy simultaneously. Meanwhile, when a retailer is required to make a prepayment, the purchase discount policy is contingent on the number of installments offered. This means that the retailer prepays off the entire purchasing cost with a single installment, thereby receiving a maximum percentage of price discount. Otherwise, the retailer prepays a certain fraction of the purchasing cost with multiple installments, and the percentage of the price discount will be contingent on the number of  identical installments. In this context, we present theoretical results for optimal solutions, and a salient algorithm is presented, which is derived from these theoretical findings within a sustainable inventory system. To better illustrate the proposed mathematical problems, several numerical examples are presented, followed by sensitivity analysis for different scenarios. Full article

In the field of geomatics, artificial intelligence (AI) and especially machine learning (ML) are rapidly transforming the field of geomatics with respect to collecting, managing, and analyzing spatial data. Feature selection as a building block in ML is crucial because it directly impacts the performance and predictive power of a model by selecting the most critical variables and eliminating the redundant and irrelevant ones. Random forests have now been used for decades and allow for building models with high accuracy. However, finding the most expressive features from the dataset by selecting the most important features within random forests is still a challenging question. The often-used internal Gini importances of random forests are based on the amount of training examples that are divided by a feature but fail to acknowledge the magnitude of change in the target variable, leading to suboptimal selections. Shapley values are an established and unified framework for feature attribution, i.e., specifying how much each feature in a trained ML model contributes to the predictions for a given instance. Previous studies highlight the effectiveness of Shapley values for feature selection in real-world applications, while other research emphasizes certain theoretical limitations. This study provides an application-driven discussion of Shapley values for feature selection by first proposing four necessary conditions for a successful feature selection with Shapley values that are extracted from a multitude of critical research in the field. Given these valuable conditions, Shapley value feature selection is nevertheless a model averaging procedure by definition, where unimportant features can alter the final selection. Therefore, we additionally present Conditional Feature Selection (CFS) as a novel algorithm for performing feature selection that mitigates this problem and use it to evaluate the impact of model averaging in several real-world examples, covering the use of ML in geomatics. The results of this study show Shapley values as a good measure for feature selection when compared with Gini feature importances on four real-world examples, improving the RMSE by 5% when averaged over selections of all possible subset sizes. An even better selection can be achieved by CFS, improving on the Gini selection by approximately 7.5% in terms of RMSE. For random forests, Shapley value calculation can be performed in polynomial time, offering an advantage over the exponential runtime of CFS, building a trade-off to the lost accuracy in feature selection due to model averaging. Full article

The climate crisis necessitates a global shift to achieve a secure, sustainable, and affordable energy system toward a green energy transition reaching climate neutrality by 2050. Because of this, renewable energy sources have come to the forefront, and the research interest in microgrids that rely on distributed generation and storage systems has exploded. Furthermore, many new markets for energy trading, ancillary services, and frequency reserve markets have provided attractive investment opportunities in exchange for balancing the supply and demand of electricity. Artificial intelligence can be utilized to locally optimize energy consumption, trade energy with the main grid, and participate in these markets. Reinforcement learning (RL) is one of the most promising approaches to achieve this goal because it enables an agent to learn optimal behavior in a microgrid by executing specific actions that maximize the long-term reward signal/function. The study focuses on testing two optimization algorithms: logic-based optimization and reinforcement learning. This paper builds on the existing research framework by combining PPO with machine learning-based load forecasting to produce an optimal solution for an industrial microgrid in Norway under different pricing schemes, including day-ahead pricing and peak pricing. It addresses the peak shaving and price arbitrage challenges by taking the historical data into the algorithm and making the decisions according to the energy consumption pattern, battery characteristics, PV production, and energy price. The RL-based approach is implemented in Python based on real data from the site and in combination with MATLAB-Simulink to validate its results. The application of the RL algorithm achieved an average monthly cost saving of 20% compared with logic-based optimization. These findings contribute to digitalization and decarbonization of energy technology, and support the fundamental goals and policies of the European Green Deal. Full article

In the dynamic and uncertain stock market, precise forecasting and decision-making are crucial for profitability. Traditional deep neural networks (DNN) often struggle with capturing long-term dependencies and multi-scale features in complex financial time series data. To address these challenges, we introduce CMR-DQN, an innovative framework that integrates discrete wavelet transform (DWT) for multi-scale data analysis, temporal convolutional network (TCN) for extracting deep temporal features, and a GRU–LSTM–Attention mechanism to enhance the model’s focus and memory. Additionally, CMR-DQN employs the Rainbow DQN reinforcement learning strategy to learn optimal trading strategies in a simulated environment. CMR-DQN significantly improved the total return rate on six selected stocks, with increases ranging from 20.37% to 55.32%. It also demonstrated substantial improvements over the baseline model in terms of Sharpe ratio and maximum drawdown, indicating increased excess returns per unit of total risk and reduced investment risk. These results underscore the efficiency and effectiveness of CMR-DQN in handling multi-scale time series data and optimizing stock market decisions. Full article

The traditional centralized optimization method encounters challenges in representing the interaction among multi-agents and cannot consider the interests of each agent. In traditional low-carbon scheduling, the fixed carbon quota trading price can easily cause arbitrage behavior of the trading subject, and the carbon reduction effect is poor. This paper proposes a two-layer dynamic community integrated energy system (CIES) low-carbon collaborative optimization operation method. Firstly, a multi-agent stage feedback carbon trading model is proposed, which calculates carbon trading costs in stages and introduces feedback factors to reduce carbon emissions indirectly. Secondly, a two-layer CIES low-carbon optimal scheduling model is constructed. The upper energy seller (ES) sets energy prices. The lower layer is the combined cooling, heating, and power (CCHP) system and load aggregator (LA), which is responsible for energy output and consumption. The energy supply and consumption are determined according to the ES energy price strategy, which reversely affects the energy quotation. Then, the non-dominated sorting genetic algorithm embedded with quadratic programming is utilized to solve the established scheduling model, which reduces the difficulty and improves the solving efficiency. Finally, the simulation results under the actual CIES example show that compared with the traditional centralized scheduling method, the total carbon emission of the proposed method is reduced by 16.34%, which can improve the income of each subject and make the energy supply lower carbon economy. Full article

The role of pumped storage in global energy structure transformation is becoming increasingly prominent. This article introduces a flexible excitation system based on fully controlled device converters into pumped storage units (PSUs). It can address the issues of insufficient excitation capacity and limited stability associated with traditional thyristor excitation systems. The study focuses on linear active disturbance rejection control (LADRC) for the flexible excitation control system of pumped storage units and utilizes intelligent optimization algorithms to optimize the controller parameters. This addresses the inherent problem of traditional PID controllers, which are unable to alleviate the trade-off between response speed and overshoot. At the same time, the robustness and anti-interference of the control system are improved, effectively enhancing the performance of the pumped storage flexible excitation control system. Simulation verifies the feasibility and superiority of the proposed method. Full article

In recent years, global trade volume has been increasing, and marine transportation plays a significant role here. In marine transportation, the choice of transportation route has been widely discussed. Minimizing fuel consumption, minimizing voyage time, and maximizing voyage security are concerns of the International Maritime Organization (IMO) regarding Maritime Autonomous Surface Ships (MASS). These goals are contradictory and have not yet been effectively resolved. This paper describes the ship path-planning problem as a multi-objective optimization problem that considers fuel consumption, voyage time, and voyage security. The model considers wind and waves as marine environmental factors. Furthermore, this paper uses an improved Whale Optimization Algorithm to solve multi-objective problems. At the same time, it is compared to three advanced algorithms. Through seven three-objective test functions, the performance of the algorithm is tested and applied in path planning. The results indicate that the algorithm can effectively balance the fuel consumption, voyage time, and voyage security of the ship, offering reasonable paths. Full article

Clique counts are crucial in applications like detecting communities in social networks and recurring patterns in bioinformatics. Counting k-cliques—a fully connected subgraph with k nodes, where each node has a direct, mutual, and symmetric relationship with every other node—becomes computationally challenging for larger k due to combinatorial explosion, especially in large, dense graphs. Existing exact methods have difficulties beyond k = 10, especially on large datasets, while sampling-based approaches often involve trade-offs in terms of accuracy, resource utilization, and efficiency. This difficulty becomes more pronounced in dense graphs as the number of potential k-cliques grows exponentially. We present Boundary-driven approximations of k-cliques (BDAC), a novel algorithm that approximates k-clique counts without using recursive procedures or sampling methods. BDAC offers both lower and upper bounds for k-cliques at local (per-vertex) and global levels, making it ideal for large, dense graphs. Unlike other approaches, BDAC’s complexity remains unaffected by the value of k. We demonstrate its effectiveness by comparing it with leading algorithms across various datasets, focusing on k values ranging from 8 to 50. Full article

With the increasing global concern over climate change, reducing greenhouse gas emissions has become a universal goal for governments and enterprises. For oversize and heavyweight cargo (OHC) transportation, multimodal transportation has become widely adopted. However, this mode inevitably generates carbon emissions, making research into effective emission reduction strategies essential for achieving low-carbon economic development. This study investigates the optimization of multimodal transportation paths for OHC (OMTP-OHC), considering various direct carbon pricing policies and develops models for these paths under the ordinary scenario—defined as scenarios without any carbon pricing policies—and two carbon pricing policy scenarios, namely the emission trading scheme (ETS) policy and the carbon tax policy, to identify the most cost-effective solutions. An enhanced genetic algorithm incorporating elite strategy and catastrophe theory is employed to solve the models under the three scenarios. Subsequently, we examine the impact of ETS policy price fluctuations, carbon quota factors, and different carbon tax levels on decision-making through a case study, confirming the feasibility of the proposed model and algorithm. The findings indicate that the proposed algorithm effectively addresses this problem. Moreover, the algorithm demonstrates a small impact of ETS policy price fluctuations on outcomes and a slightly low sensitivity to carbon quota factors. This may be attributed to the relatively low ETS policy prices and the characteristics of OHC, where transportation and modification costs are significantly higher than carbon emission costs. Additionally, a comparative analysis of the two carbon pricing policies demonstrates the varying intensities of emission reductions in multimodal transportation, with the ranking of carbon emission reduction intensity as follows: upper-intermediate level of carbon tax > intermediate level of carbon tax > lower-intermediate level of carbon tax = ETS policy > the ordinary scenario. The emission reduction at the lower-intermediate carbon tax level (USD 8.40/t) matches that of the ETS policy at 30%, with a 49.59% greater reduction at the intermediate level (USD 50.48/t) compared to the ordinary scenario, and a 70.07% reduction at the upper-intermediate level (USD 91.14/t). The model and algorithm proposed in this study can provide scientific and technical support to realize the low-carbonization of the multimodal transportation for OHC. The findings of this study also provide scientific evidence for understanding the situation of multimodal transportation for OHC under China’s ETS policy and its performance under different carbon tax levels in China and other regions. This also contributes to achieving the goal of low-carbon economic development. Full article

Lenders writing loans in the peer-to-peer market carry risk with the anticipation of an expected return. In the current implementation, many lenders do not have an exit strategy beyond holding the loan for the full repayment term. Many would-be lenders are deterred by the risk of being stuck with an illiquid investment without a method for adjusting to overall economic conditions. This risk is a limiting factor for the overall number of loan transactions. This risk prevents funding for many applicants in need, while simultaneously steering capital towards other more liquid and mature markets. The underdeveloped valuation methods used presently in the peer-to-peer lending space present an opportunity for establishing a model for assigning value to loans. We provide a novel application of an established model for pricing peer-to-peer loans based on multiple factors common in all loans. The method can be used to give a value to a peer-to-peer loan which enables transactions. These transactions can potentially encourage participation and overall maturity in the secondary peer-to-peer loan trading market. We apply established valuation algorithms to peer-to-peer loans to provide a method for lenders to employ, enabling note trading in the secondary market. Full article

In recent years, the energy industry has increased the proportion of renewable energy sources, which are sustainable and carbon-free. However, the increase in renewable energy sources has led to grid instability due to factors such as the intermittent power generation of renewable sources, forecasting inaccuracies, and the lack of metering for small-scale power sources. Various studies have been carried out to address these issues. Among these, research on Virtual Power Plants (VPP) has focused on integrating unmanaged renewable energy sources into a unified system to improve their visibility. This research is now being applied in the energy trading market. However, the purpose of VPP aggregators has been to maximize profits. As a result, they have not considered the impact on distribution networks and have bid all available distributed resources into the energy market. While this approach has increased the visibility of renewables, an additional method is needed to deal with the grid instability caused by the increase in renewables. Consequently, grid operators have tried to address these issues by diversifying the energy market. As regulatory method, they have introduced real-time energy markets, imbalance penalty fees, and limitations on the output of distributed energy resources (DERs), in addition to the existing day-ahead market. In response, this paper proposes an optimal scheduling method for VPP aggregators that adapts to the diversifying energy market and enhances the operational benefits of VPPs by using two Mixed-Integer Linear Programming (MILP) models. The validity of the proposed model and algorithm is verified through a case study analysis. Full article