Integrated System Research Articles

Unlocking AI Creativity: A Multi-Agent Approach with CrewAI

By automating processes that traditionally require human intelligence, such as learning, reasoning, and problem-solving, Artificial Intelligence (AI) has transformed many industries. Even with these developments, System 1 thinking is quick, providing instinctive reactions, while System 2 thinking entails thorough analysis and reasoned decision-making. AI systems still have a long way to go before they can replicate System 2 thinking—deliberate, analytical, and essential for managing complex problems. This study investigates how Multi-Agent Systems (MAS) integrate System 2 thinking into AI, concentrating on CrewAI, a no-code framework designed to improve AI creativity and decision-making. Unlike TensorFlow Agents and OpenAI’s Gym, which are limited to single-agent reinforcement learning, CrewAI excels in handling multi-agent, real-world tasks through collaboration. This study explores practical applications of CrewAI, such as intelligent grid management, automated customer support, and advertising. These examples highlight how CrewAI promotes AI creativity and problem-solving through cooperative agent interactions, leveraging System 2 thinking. Problems like scalability and coordination are also addressed, with solutions such as dynamic role assignment and hierarchical task management. In summary, the integration of System 2 thinking into MAS frameworks like CrewAI signifies progress toward creating intelligent, dependable AI systems capable of tackling the complexities of real-world problems.

Multi-energy load forecasting for small-sample integrated energy systems based on neural network Gaussian process and multi-task learning

Multi-energy load forecasting forms the foundation of the operation and scheduling of integrated energy systems. Nevertheless, insufficient data and underutilization of the coupling relationship between the multi-energy load limit the accuracy of load forecasting. This paper presents a predictive model combining neural network Gaussian processes and multi-task learning. The approach is tailored to enhance forecasting accuracy in environments with small-sample datasets. This model capitalizes on the advantageous properties of infinitely wide neural networks for handling small-sample data. Simultaneously, the model effectively extracts the interconnected dynamics of cooling, heating, and electricity loads within the integrated energy system through multi-task learning. In addition, the model applies concrete dropout, enhancing robustness to irregular loads while maintaining the synergistic benefits of the multi-task framework. Furthermore, this paper employs a two-stage gradient descent approach to replace kernel matrix computations of Gaussian processes, reducing the computational cost of parameter optimization and yielding superior forecasting performance in shorter training durations. The simulation results indicate that the proposed model attains a mean accuracy of 97.93% for a 3-day forecasting horizon. Compared with alternative forecasting models, this model exhibits higher accuracy and enhanced generalization capabilities in multi-energy load forecasting for small-sample integrated energy systems.

A novel hybrid low-temperature thermal catalysis and radiative sky cooling system for day and night air purification and cooling

Thermal catalytic oxidation is an air purification technology that can efficiently and stably degrade volatile organic compounds. However, using thermal catalysis to purify indoor air in the summer can cause problems of indoor overheating. Radiative sky cooling is a passive cooling method that dissipates heat through reflection and radiation, which can provide sub-ambient cooling during day and night. In this study, a novel hybrid low-temperature thermal catalysis and radiative sky cooling system for day and night air purification and cooling was proposed, which combined thermal catalysis with radiative sky cooling, and used low-temperature driven thermal catalysts to solve the problem of indoor overheating in summer. A numerical model of the hybrid system was established to simulate the formaldehyde degradation and cooling performance of the system under different operating conditions. The results show that under summer conditions with an average daily temperature of 35°C and humidity levels between 70% and 80%, the system produced a total of 665.29 m³/m2 of clean air after running for an entire day. The average single-pass conversion rate of formaldehyde was 0.46, and the maximum temperature difference between indoor and outdoor air was 5°C. This provides guidance for the integration of the hybrid system with buildings.

Cooperative resource sharing and cost minimization in energy hub systems using an improved grasshopper optimization algorithm approach

This study presents a cooperative paradigm for energy hub systems (EHSs) where a network of interconnected hubs cooperates in exploiting the resources with the purpose of economic saving. In such an architecture, each hub provided with various sources of energy, such as combined heat and power (CHP), hot water tanks, renewable sources, electric chillers, and absorption chillers, will integrate all these sources for more adaptability and efficiency to the system. Moreover, the integration of energy storage systems (ESSs) is considered to enhance the flexibility of the energy hub concerning power, heating, and cooling. Recognizing the complexity associated with incorporating multiple constraints, the improved grasshopper optimization algorithm (IGOA) is introduced to effectively address this challenge. By leveraging this algorithm, the study aims to overcome the intricacies involved in considering various constraints and achieve an optimal outcome. The IGOA improves the efficiency and effectiveness of local and national searches in solving complex energy hub optimization problems. Reducing the likelihood of getting stuck in suboptimal solutions, enhances the algorithm's ability to find optimal solutions considering multiple constraints, thereby enhancing the overall performance and cost-effectiveness of EHSs. The issue is defined as a planning challenge, and by collaborative efforts, the expenses associated with the network energy hubs are reduced, illustrating the efficacy of this concept. The findings indicate the influence of the suggested cooperative technique, with operating cost reductions of 19.09 %, 13.27 %, and 8.75 % for Hub 1, Hub 2, and Hub 3, respectively. Furthermore, the cooperative framework eradicates energy deficits and disruptions, in contrast to 1,198.21 kWh of unfulfilled demand and 22 interruptions in the non-cooperative scenario. These results underscore the significant advantages of the collaborative technique in improving cost-efficiency, reliability, and resource utilization.

How does the built environment affect intermodal transit demand across different spatiotemporal contexts?

Bus and metro are the two primary modes of public transportation in many megacities worldwide. Understanding their cooperation is crucial for the integration of the public transportation system. Despite extensive research on public transportation demand, studies focusing on bus-metro cooperation remain limited. Intermodal transit demand directly reflects the level of cooperation between the two modes in travel behavior. In this study, intermodal transit demand is extracted from smart card data in Beijing, China. The extreme gradient boosting algorithm is employed to investigate the determinants of intermodal transit demand considering spatiotemporal variation. The SHapley Additive exPlanations method further interprets these models. Findings reveal that (1) the relative spatial relationship between bus and metro service facilities significantly influences their cooperation; however, these influences gradually weaken as urban space expands from the core to the peripheral area; (2) in peripheral area, the characteristics of the bus network hold the highest average importance ranking; (3) extensive nonlinear relationships and threshold effects exist between the built environment and intermodal transit demand, with the magnitude, pattern, and direction of these impacts varying significantly across different spatiotemporal contexts; and (4) changes in the spatial layout of transportation service supplies impact their competition and cooperation, such as adequate bus service supplies potentially reducing the cooperation between bus and metro to some extent. These findings will assist planners and public transit operators in developing regulations that encourage cooperation between bus and metro, thereby increasing the attraction and competitiveness of the public transit system.

Techno-economic performance of different DC power distribution networks in official buildings based on a real-time analysis method

The integration of photovoltaic (PV) and energy storage systems into official buildings has garnered considerable attention, which are recognized as DC power sources. Notably, DC power can satisfy the majority of electrical demands in official buildings. The primary objective of this research is to enhance overall system efficiency by efficiently utilizing renewable energy through direct current (DC) power distribution compared to alternating current (AC) systems. This study proposes a techno-economic analysis framework to guide designers in selecting system configuration and voltage levels. The analysis covers various PV capacities, load compositions, and battery capacities, with the aim of optimizing power distribution efficiency and achieving a more competitive levelized cost of energy (LCOE). The proposed model integrates the dynamic efficiency of actual converters and wire gauges, enabling real-time computation of the energy supply–demand balance within the power distribution system, including associated conversion and transmission losses. The results indicate that conversion loss accounts for approximately 80% of the overall energy loss within power distribution networks. The implementation of DC power distribution significantly improves system efficiency and economic viability. Specifically, the DC375 system exhibits superior techno-economic performance in scenarios involving high PV and battery capacities. In scenarios with limited PV and battery capacity, the building power distribution system should employ a hybrid AC-DC topology to efficiently distribute the available PV output to building loads powered by DC

An integration of quantum systems using BB84 for enhanced security in aeroponic smart farming

An integration of quantum systems using BB84 for enhanced security in aeroponic smart farming

3-D impedance matching network (IMN) based on through-silicon via (TSV) for RF energy harvesting system

Radio Frequency Energy Harvesting (RFEH) system can harvest radio frequency (RF) energy from the natural environment, showing potential applications to play the role of the battery in specific settings, such as outdoors or human body, demonstrating vast potential applications on the scenarios with difficulty to renew the traditional battery. In the RFEH system, Impedance Matching Network (IMN) consumes substantial amounts of metal wire and chip area in standard CMOS process and is not easy to be integrated into the monolithic substrate. The planar structure shows the shortcoming of large area and prominent parasitic influences, limiting its integration in the REFH system. The development of 3-D integrated circuit technology based on Through-Silicon Vias (TSV) offers a possible solution for RF passive devices. In the paper, a novel TSV-based 3-D IMN is designed. The capacitor is implemented with the TSV structure, while the inductor is designed with Redistribution Layer (RDL) spiral pattern. The co-design of 3D LC matching circuit shows the merits of the proposed solution, including high performance and compact size. It can be used as a technical solution for the integration of IMN in a monolithic CMOS RFEH system.

Development of a software platform for automatic train operation: Integrating meso- and microscopic layers

This article presents a comprehensive study on the development and integration of the mesoscopic and microscopic layers within the control architecture of railway systems, enabling autonomous operation. A key outcome of this research is developing a software platform that provides the foundation for improving autonomous train control algorithms. The mesoscopic layer, defined as the high-level component of the automatic train operation (ATO) subsystem, focuses on generating setpoints to control the traction system, which are subsequently transmitted to the low-level component of the ATO subsystem. A simplified model is employed in the mesoscopic layer to enhance planning capabilities and facilitate efficient decision-making processes, while a detailed model represents the physical system. Simulations evaluate the effectiveness of this integrated system, considering various disturbances to thoroughly assess the performance of the mesoscopic layer algorithm and the microscopic layer control system. The results confirm that by successfully integrating these layers, precise and efficient control of train motion is achieved while considering all constraints imposed by the macroscopic layer. Furthermore, this software platform is set for future development, including implementing intelligent control techniques to enhance autonomous train control.

Recyclable polyurethane from castor oil based on dynamic disulfide bonds and multiple hydrogen bonds as adhesive and photothermal conversion materials

As an important polymer material, polyurethane brings convenience to daily life but also causes environmental problems, and the manufacturing of bio-based repairable, re-processable and sturdy materials can effectively reduce environmental pressure. Herein a low temperature recyclable polyurethane (PU) was developed with castor oil (CO) by combining isophorone diisocyanate (IPDI), dynamic disulfide bonds and hydrogen bonds. The castor oil-based PU showed impressive tensile strength (16.1MPa) remarkable elongation at break (1055.8%), and high bonding power (up to ~6MPa) with bonding wood chips. The dynamic disulfide bonds and hydrogen bonds imparted the bio-based PU with outstanding elastic recovery, impressive self-healing capability (up to ~90%), short relaxation time (5~6min at 180°C), favorable shape memory behavior, and multiple recyclability. By mixing different proportions of carbon nanotubes (CNTs), recyclable and stretchable conductive composites are realized. In addition, an integrated system of high-efficiency bio-based solar photovoltaic generator is demonstrated for simulating the ambient sunlight-heat-electricity conversion, which provides some guidance for the efficient use of solar energy.

Enhancing energy security and efficiency through a regional integration policy: The case of the Southern Cone

This paper addresses the knowledge gap concerning foreign exchange risk in energy transactions within the Regional Initiative Energy Integration System of the Southern Cone Countries (SIESUR) of the Latin American region. Focusing on the cross-border electricity, it investigates the impact of exchange rate fluctuations on the transaction model in SIESUR: US dollar-denominated cross-border energy contracts with a time lag between invoicing and payment. The objective of this paper is to estimate foreign exchange risk exposure and propose mitigation solutions. Analyzing foreign exchange risk of energy trading among SIESUR member countries reveals substantial financial costs hindering electricity integration. Energy importing countries, facing greater exposure of their local currencies and weakened bargaining power, are particularly vulnerable. A Monte Carlo simulation model, using historical exchange rate data, highlights significant currency volatility, emphasizing the substantial foreign exchange risk exposure within SIESUR. Notably, cost overruns of energy imports due to foreign exchange risk exposure can reach 12.6% of transaction value. Through a global analysis, potential strategies for risk management are explored to foster electricity exchange expansion. Efforts to mitigate foreign exchange risk are crucial for advancing electricity integration within the SIESUR region. This study underscores policy implications for enhancing regional energy security and promoting sustainable development.

The integrated smart system to assist elderly at home

Some elders prefer to live independently rather than in a nursing home. However, they need to be assisted because of health and safety reasons. We developed an integrated system consisting of a smart home, a mobile robot, and a wearable device, using message queuing telemetry transport (MQTT) protocol to communicate with each other. The smart home can be monitored and controlled via a mobile application. We use the telepresence robot equipped with a light detection and ranging (LIDAR) sensor to perform simultaneous localization and mapping (SLAM) and autonomous navigation. The wearable device application is used to detect older adults' heart rate, the steps taken, and the burnt calories. We also add pose detection and location estimation using a depth camera powered by an efficient deep-learning algorithm. We also developed an Android application as a dashboard that monitors and controls all system components. Our system can facilitate communication between its various components.

Key Technology Innovation Mode of New Energy Industry Ecological Integration System Based on Particle Swarm Optimization Algorithm

Key Technology Innovation Mode of New Energy Industry Ecological Integration System Based on Particle Swarm Optimization Algorithm

A novel cascade heat design for a geothermal energy-based combined power plant in integration with water electrolysis and ammonia synthesis processes: Comprehensive thermo-environ-economic analyses

The present paper introduces an innovative integrated system for simultaneous cooling, heating, power, and ammonia production utilizing geothermal energy. The novel configuration incorporates a geothermal-based single-flash power plant, a modified Kalina cycle, an organic Rankine cycle, an electrochemical unit, and an ammonia synthesis process. The Kalian cycle has undergone modifications aimed at enhancing its production capacity, while the electrochemical unit provides hydrogen for the ammonia synthesis process. The research encompasses a thorough assessment of the proposed system's potential with regard to thermodynamic, environmental impact, and economic considerations. Additionally, a parametric assessment is conducted using the flash pressure of the geothermal fluid and the proportion of unreacted gases to the ammonia synthesis reactor in order to analyze the performance variables of the system. The proposed system demonstrates a power generation capacity of 15513.18 kW, a heating load of 24880 kW, a cooling load of 11540 kW, and an ammonia production capacity of 0.282 kg/s. Moreover, the process exhibits irreversibility with a total value of 16977 kW. Consequently, the energy and exergy efficiencies amount to 34.55% and 57.17%, correspondingly. Besides, geothermal energy offers environmental benefits, as its non-polluting nature results in reduced CO2 emissions in comparison to biomass, coal-based, and hybrid energy systems. Eventually, the economic analysis demonstrates that the total cost rate and sum unit cost of products are 825 $/h and 9.57 $/GJ, respectively.

Renewable-driven hybrid refrigeration system for enhancing food preservation: Digital twin design and performance assessment

This study presents a new method for sustainable cooling systems using a hybrid refrigeration system powered by hybrid renewable energy sources. The system comprises a modular unit of vertical wind turbines integrated with bio-photovoltaic films to provide sustainable energy. The hybrid refrigeration system combines evaporative and solar thermal-driven adsorption cooling systems. In addition, a finite volume of soil is proposed for thermal energy storage. Experimental data inform the development of a digital twin for an integrated system, soil thermophysical characteristics, wind turbine performance, and technical specifications for other system components. This sustainable cooling package is cost-effective and space-efficient, particularly in remote or off-grid locations. Notably, the evaporative cooler and chilled water coil contribute to a cooling effect of 20.4 kW, and solar power generation reaches 12.38 kW at an intensity of 1053 W/m2. The annual electrical output averages 1.7 kW at a wind speed of 3.5 m/s. Under best conditions, wind power can surge to 7.99 kW at 9.88 m/s. The ratio of power generated by wind to solar energy ranges from 1.1 to 1.3. The system effectively meets a peak thermal energy demand of approximately 74 GJ/month, facilitated by solar collectors, underground thermal storage, and a renewable energy-fed auxiliary heater. This study paves the way for future techno-economic optimisation and advancements in sustainable energy solutions for remote cold storage facilities.

Grid-connected double-stage PV system with space vector PWM inverter and MPPT

Because of the increasing demand for non-conventional energy sources, photovoltaic system integration into the electrical grid is becoming more crucial. It is advised to employ a double-stage grid-connected PV system. This can be employed with the main parts boost converter, which helps to boost the DC output from the PV channels. And next one is the space vector pulse width modulation (SVPWM) inverter which can provide better harmonic performance and higher efficiency compared to traditional PWM methods. The SVPWM technique is employed to modulate the inverter output, generating a sinusoidal AC waveform that matches the grid frequency and voltage. By adjusting the pulse width to achieve effective and trustworthy grid integration. To maximize power extraction takes place by using maximum power point tracking (MPPT). The SVPWM inverter is crucial in converting DC power to AC power and enabling grid connection. Utilizing SVPWM allows exact control of voltage magnitude, frequency, and phase angle to provide high-quality sinusoidal AC output voltage. The power quality of the electricity pumped into the grid is improved by this precise management, which also ensures adherence to grid standards and reduces harmonic interference.

A new optimal space vector modulation with DTC switching strategy for induction motor control

Efforts to achieve swift and precise dynamic torque control have been central in AC drive research. Recent advancements in embedded computer systems have highlighted direct torque control (DTC) and field-oriented control (FOC) as key methods for enhancing torque dynamics, both utilizing space vector modulation (SVM) to optimize voltage source inverter positioning. This study introduces a novel synthesis by integrating DTC with SVM to address limitations in conventional DTC, which suffers from limited voltage vector availability, leading to undesirable torque behavior and significant current fluctuations. The primary goal is to develop an optimal switching modulator for the fastest torque response through the combined application of DTC and SVM. The proposed strategy optimizes DC bus usage, reduces torque fluctuations, minimizes total harmonic distortion in AC motor current, decreases switching losses, and ensures seamless digital system integration. Simulations using MATLAB/SIMULINK demonstrate significant torque, current, and flux linkage ripple reductions, validating the approach's effectiveness. This integration overcomes established limitations, extending the capabilities of motor control methodologies and offering enhanced performance and operational integrity in induction motor drive systems.

AI-powered financial crime prevention with cybersecurity, IT, and data science in modern banking

Financial crime in modern banking has evolved significantly with the digital transformation of financial services, presenting unprecedented challenges to traditional prevention methods. This comprehensive review examines the integration of artificial intelligence (AI), cybersecurity frameworks, and data science methodologies in combating financial crime within the banking sector. We analyze the current state of AI-powered solutions, including machine learning models, real-time detection systems, and advanced analytics frameworks that have transformed financial crime prevention. The review synthesizes findings from recent studies and industry implementations, highlighting the synergistic relationship between AI technologies and cybersecurity measures in creating robust defense mechanisms. Our analysis reveals that while AI-powered solutions demonstrate superior detection rates and reduced false positives compared to traditional methods, significant challenges remain in areas of data privacy, regulatory compliance, and system integration. The paper concludes by identifying critical research gaps and proposing future directions for enhancing the effectiveness of AI-based financial crime prevention systems. This review provides valuable insights for researchers, banking professionals, and policymakers working at the intersection of AI, cybersecurity, and financial crime prevention.

Digital Transformation in Talent Acquisition: An Investigation into the Integration of Big Data Analytics and Management Information Systems for Optimal Hiring Strategies

The primary goal was to empirically investigate the relationships amongst Big Data Analytics (BDA), Management Information System (MIS), optimal hiring strategies, leadership support, and talent acquisition process. This research aimed to fill crucial gaps in the literature by examining how technology adoption and leadership support influence the formulation and effectiveness of optimal hiring strategies, finally impacting talent acquisition results in the Chinese business landscape. Quantitative research was employed, targeting experts and decision-makers in human resource departments across numerous industries in China. The research used a random sampling approach to acquire facts through an online questionnaire, examining responses from 168 participants. Regression and mediation analyses, facilitated by SPSS, were applied to discover the relationships and mechanisms among BDA, MIS, optimal hiring strategies, leadership support, and talent acquisition success. The results discovered a significant and positive relationship between BDA and MIS with the optimal hiring techniques. Moreover, optimal hiring techniques have proven an advantageous association with talent acquisition success. The research similarly showed the mediating role of optimal hiring strategies in translating the advantages of BDA and MIS into talent acquisition success. Leadership support emerged as a moderator, amplifying the relationship between optimum hiring techniques and talent acquisition success. This study contributes empirical proof to the underexplored intersection of leadership support, technology, and talent hiring strategies. The findings offer practical insights for companies aiming to navigate the complexities of recruitment, emphasizing the significance of strategic decision-making, technology integration, and management endorsement in achieving talent acquisition success.

Consumer Attitudes towards AI-based Financial Advice: Insights for Decision Support Systems (DSS) and Technology Integration

As financial technology (fintech) rapidly transforms the financial services landscape, the integration of artificial intelligence (AI) in providing financial advice becomes a focal point. This study delves into the intricate dynamics of AI-based financial advice adoption, focusing on technology integration, decision support systems, and the mediating role of perceived utility. The primary purpose of this research is to unravel the complex relationships between technology integration, Decision Support System (DSS), perceived utility, and their collective influence on consumer attitudes toward AI-based financial advice and technology adoption. Adopting a cross-sectional design, we target the Chinese population and utilize an online questionnaire for data collection. A sample size of 259 participants is determined using the rule of thumb technique, with random sampling ensuring representativeness. Data analysis will be conducted using the AMOS software, allowing for an in-depth examination of the relationships between variables. The study findings demonstrate how integrating technology enhances customers' perceptions and acceptance of it, especially when it comes to data security and compatibility. The study also demonstrates how customer attitudes and the uptake of AI-driven financial aid are significantly impacted by DSS aspects like decision transparency and predictive analytics precision. The study is unique because of its cross-cultural approach and perceived value as a mediating component. While this study focuses on the Chinese population, the researchers acknowledge the importance of cultural differences in shaping user perceptions and behaviors. While the findings may offer valuable insights into the Chinese context, caution is advised in generalizing the results to other cultural contexts.