High Energy Efficiency Research Articles

Cryogenic in-memory computing using magnetic topological insulators.

Machine learning algorithms have proven to be effective for essential quantum computation tasks such as quantum error correction and quantum control. Efficient hardware implementation of these algorithms at cryogenic temperatures is essential. Here we utilize magnetic topological insulators as memristors (termed magnetic topological memristors) and introduce a cryogenic in-memory computing scheme based on the coexistence of a chiral edge state and a topological surface state. The memristive switching and reading of the giant anomalous Hall effect exhibit high energy efficiency, high stability and low stochasticity. We achieve high accuracy in a proof-of-concept classification task using four magnetic topological memristors. Furthermore, our algorithm-level and circuit-level simulations of large-scale neural networks demonstrate software-level accuracy and lower energy consumption for image recognition and quantum state preparation compared with existing magnetic memristor and complementary metal-oxide-semiconductor technologies. Our results not only showcase a new application of chiral edge states but also may inspire further topological quantum-physics-based novel computing schemes.

Control and islanding fault detection method of distributed photovoltaic EPG based on deep learning

ABSTRACT Distributed photovoltaic power generation system has many advantages, such as low production cost, rich raw material resources, environmental protection and pollution-free, long service life of power generation equipment, high energy efficiency conversion rate, etc., as an important supplement to centralized power supply mode, distributed photovoltaic power generation system has broad development space and great application value today when fuel resources are increasingly exhausted and the power load increases sharply during the peak period of power consumption. Because solar energy has great advantages, photovoltaic power generation technology is the most direct way to use solar energy in our country effectively. Solar energy has a good energy storage effect, wide distribution and no pollution. These advantages have caused photovoltaic power generation technology to get more attention, especially application of distributed photovoltaic power generation in small and medium-power roof photovoltaic systems. Due to the increasing shortage of traditional energy, the problem of environmental pollution needs to be improved, and public’s awareness of environmental protection needs to be awakened. Renewable energy, such as solar, wind, and bioenergy, has gradually developed into an alternative to traditional fossil energy. We can get better solutions and detection methods through in-depth learning and analysis of this method.

A Bionic Robotic Fish with a Dielectric Elastomer

Abstract Dielectric elastomer actuators (DEAs) are promising enabling devices that can be used in a wide range of robots, artificial muscles and microfluidics. They are characterized by high actuating strain, low cost and noise, and high energy density and efficiency. There are three main challenges for enabling DEs to become actuators: (1) developing suitable and compatible electrode materials; (2) effectively isolating the actuator electrode from the surrounding fluid; and (3) creating a rigid frame that usually requires prestraining of the dielectric layer. The use of robotic fish in water is an important application field of biomimetic soft robots. At present, most underwater robotic fish use spiral propulsion, which has several problems, including propulsion efficiency, position controllability and aquatic organism involvement. To provide solutions, the research and development of underwater robotic fish that imitate the fins and body propulsion of fish and the use of soft underwater robotic fish are in full adoption. This project involves the research and development of a bionic soft underwater robot fish with a software driver, which can imitate swimming via the tail fin and body of a fish, especially with respect to stable swimming propulsion, to successfully develop high-performance soft underwater robot fish. In addition, to imitate the unstable swimming movements of fish, such as turning and sharp acceleration and deceleration, robot fish that use DE drivers with good flexibility and high strain have been researched and developed.

Dehydration-enhanced Ion Recognition of Triazine Covalent Organic Frameworks for High-resolution Li+/Mg2+ Separation.

The precise and rapid extraction of lithium from salt-lake brines is critical to meeting the global demand for lithium resources. However, it remains a major challenge to design ion-transport membranes with accurate recognition and fast transport path for the target ion. Here, we report a triazine covalent organic framework (COF) membrane with high resolution for Li+ and Mg2+ that enables fast Li+ transport while almost completely inhibiting Mg2+ permeation. The remarkably high rejection of Mg2+ by the COF membrane is achieved via imposed ion dehydration and the construction of the energy well. The proper hydrophilic environment of the COF channel promotes the dissociation of Li+ from the negatively charged functional groups, allowing Li+ for hopping transport supported by the sulfonate side-chains to shorten the diffusion path of Li+. Under high-salinity electrodialysis conditions, the COF membrane demonstrates robust Li+/Mg2+ separation performance (No Mg2+ were detected in the collected solution), achieving efficient lithium recovery and high product purity (Li2CO3: 99.3%). This membrane design strategy enables high energy efficiency and powerful lithium extraction in the electrodialysis lithium extraction process, and can be generalized to other energy and separation related membranes.

Risks and sustainability of outdoor ski resorts in China under climate changes

Global warming is jeopardizing the artificial snow making conditions, shortening outdoor ski resorts opening days and increasing operation cost, severely threatening the sustainability of outdoor ski industry. The sustainability of 772 outdoor ski resorts in China under RCP 4.5 climate scenarios in 2030 s and 2050 s had been analyzed. (1) The skiing sports developed prominently during 1996 to 2023 and will boom in China in the coming decades. (2) Accelerated global warming is the main threat to the sustainability of outdoor ski resorts of China. However, snowfall isn’t a critical influencing factor in the coming decades. (3) The outdoor ski resorts in south China are facing the most threats and are the most unsustainable resorts. We proposed a nexus of Government-Operator-Skier adaptation framework for adapting climate change threats and advocating the temporal small-scale ski resorts are more adaptive as their high water and energy efficiencies for saving water and electricity resources.

Analytical and numerical investigations of metal square origami foam-filled tube under axial compression

It is widely known that thin-walled structures are expected for numerous engineering devices, due to lightweight, high specific strength, and excellent energy absorption capacity. Thus, the new high-efficiency energy absorbing (EA) device is the eternal pursuit goal of designers. A novel high-efficiency energy-absorbing device of the metal square origami foam-filled tube (MSOFT) under axial compression is designed in this article, and axial crushing of the MSOFT is investigated analytically and numerically. An analytical model for axial crushing of MSOFT is established based on the conservation of energy and the hyper-folded unit theory, considering the energy absorbed by the deformation of metal origami tube, foam compression, and the interaction between foam and origami tube. Then, the finite element (FE) simulation of MSOFTs under axial compression are conducted by Abaqus/Explicit software, and found that theoretical results are in good consistency with the FE results. Then, the influences of yield strength, wall thickness, height and cross-section width of the tube, and foam strength of MSOFT under axial compression are discussed, and found that the force and energy increases when the above parameters increase. Finally, the mean crushing force (MCF) and the crushing force efficiency (CFE) among MSOFT, origami tube and square tube are compared. Compare with the origami tube and square tube, the MCF of MSOFT increases by 40.8–106.9% and 12.1–49.8%, and the CFE of MSOFT increases by 101.2–266.3% and 1.1–12.4%. It is indicated that for a wide range of geometric and material parameters the MSOFT exhibits predictable and stable collapse behavior.

High-entropy assisted capacitive energy storage in relaxor ferroelectrics by chemical short-range order

Next-generation advanced high/pulsed power capacitors rely heavily on dielectric ceramics with high energy storage performance. Although high entropy relaxor ferroelectric exhibited enormous potential in functional materials, the chemical short-range order, which is a common phenomenon in high entropy alloys to modulate performances, have been paid less attention here. We design a chemical short-range order strategy to modulate polarization response under external electric field and achieve substantial enhancements of energy storage properties, i.e. an ultrahigh energy density of ~16.4 J/cm3 with markedly improved efficiency ~90% at an electric field of 85 kV/mm in Nb doped (Bi0.2Na0.2K0.2La0.2Sr0.2) TiO3 system. Atomic-scale scanning transmission electron microscopy observations show that Nb exhibits a chemical short-range order structure, with Nb enriched regions displaying ultrasmall polar nanoregions and more flexible polarization configurations, which is conducive to achieving high maximum polarization and low residual polarization. Moreover, refined grain size of ~0.25μm, suppressed oxygen vacancies and enhanced bandwidth contribute to a high breakdown field strength. These collective factors result in exceptionally high energy storage density and efficiency. This short-range order strategy is expected to enhance the functional performances in other high entropy relaxor ferroelectrics.

Sustainable liquid nitrogen fertilizer production via air plasma bubbles: insights into plasma-enabled N2 fixation chemistry

Abstract Liquid nitrogen fertilizers, such as potassium nitrate (KNO3) have been commonly used in modern agriculture, playing a crucial role in agricultural production. However, its production involves energy-intensive and environmentally unfriendly processes such as the Haber-Bosch process. This study demonstrated a new strategy for the sustainable and distributed production of liquid KNO3 fertilizer via air plasma bubbles. We investigated the effects of solution characteristics (initial liquid conductivity, pH) and discharge power on the nitrogen fixation performance of the air plasma bubble system. Using a strongly alkaline solution can induce the increase of vibrational temperature (Tvib) during air plasma discharges, thereby enhancing NOx yield together with favoring the NOx adsorption process. Moreover, through electrical characteristics and a simplified circuit diagram, we found a highly conductive liquid phase is not conducive to NOx generation due to the significant energy dissipation in the liquid before discharge. By further adjusting discharge power parameters and coupling the introduction of O3, the highest energy efficiency (58.5 mmol/kWh) of NOx production, with an excellent production rate (1687.4 μmol/h) is achieved. These findings provide an overall understanding of the effects of solution characteristics on gas-liquid plasma chemistry and pave the way for the optimized production of liquid nitrogen fertilizers.

Battery Health Monitoring and Remaining Useful Life Prediction Techniques: A Review of Technologies

Lithium-ion (Li-ion) batteries have become essential in modern industries and domestic applications due to their high energy density and efficiency. However, they experience gradual degradation over time, which presents significant challenges in maintaining optimal battery performance and increases the risk of unexpected system failures. To ensure the reliability and longevity of Li-ion batteries in applications, various methods have been proposed for battery health monitoring and remaining useful life (RUL) prediction. This paper provides a comprehensive review and analysis of the primary approaches employed for battery health monitoring and RUL estimation under the categories of model-based, data-driven, and hybrid methods. Generally speaking, model-based methods use physical or electrochemical models to simulate battery behaviour, which offers valuable insights into the principles that govern battery degradation. Data-driven techniques leverage historical data, AI, and machine learning algorithms to identify degradation trends and predict RUL, which can provide flexible and adaptive solutions. Hybrid approaches integrate multiple methods to enhance predictive accuracy by combining the physical insights of model-based methods with the statistical and analytical strengths of data-driven techniques. This paper thoroughly evaluates these methodologies, focusing on recent advancements along with their respective strengths and limitations. By consolidating current findings and highlighting potential pathways for advancement, this review paper serves as a foundational resource for researchers and practitioners working to advance battery health monitoring and RUL prediction methods across both academic and industrial fields.

Digital Mini-LED Lighting Using Organic Thin-Film Transistors Reaching over 100,000 Nits of Luminance.

This paper demonstrates the use of organic thin-film transistors (OTFTs) to drive active digital mini light-emitting diode (mini-LED) backlights, aiming to achieve exceptional display performance. Our findings reveal that OTFTs can effectively power mini-LED backlights, reaching brightness levels exceeding 100,000 nits. This approach not only enhances image quality but also improves energy efficiency. OTFTs offer a flexible and lightweight alternative to conventional silicon-based transistors, enabling innovative and versatile display designs. The integration of mini-LED technology with OTFTs produces displays with superior contrast ratios, enhanced color brightness, and lower power consumption. This technological advancement is poised to revolutionize high-dynamic-range (HDR) displays, including those in televisions, smartphones, and wearable devices, where the demand for high brightness and energy efficiency is paramount.

Preparation of sulfur-doped porous carbon from polyphenylene sulfide waste for photothermal conversion materials to achieve solar-driven water evaporation.

In recent years, solar-driven photothermal water evaporation technology for seawater desalination and wastewater treatment has developed rapidly, which is of great significance for addressing the issue of freshwater scarcity. However, due to the high costs associated with the manufacturing, maintenance, and operation of such devices, their application remains challenging in remote and resource-scarce regions. Due to its excellent light absorption capability in the near-infrared region, high hydrophilicity, and stable chemical properties, coupled with the low cost of recycling waste carbonized polyphenylene sulfide, this material is an excellent choice as a photothermal material for solar-driven water evaporation devices. Ordinary wood in nature usually has a highly regenerative porous structure, which is a natural water transport channel that facilitates the transport of water from the bottom to the top, allowing it to be rapidly converted into vapor. Based on this characteristic, this article innovatively proposes to prepare waste polyphenylene sulfide from porous carbonized materials (KCP) as the photothermal conversion material for novel photothermal water evaporation devices, achieving solar-driven water evaporation. This material efficiently facilitates the conversion between solar and thermal energies and exhibits excellent hydrophilicity, thereby enabling the rapid utilization of absorbed solar energy for water evaporation on the surface of the evaporator. In this study, a porous carbonized polyphenylene sulfide photothermal water evaporator (KCP-wood) was fabricated by using freeze-drying and in situ coating to load the photothermal conversion material onto a wood substrate. Under simulated one-sun irradiation, this evaporator achieved a water evaporation rate of 2.41 kg m-2 h-1 and a photothermal conversion efficiency of 91.3%. Additionally, a systematic study was conducted on the photothermal performance of various light-water evaporators, encompassing photothermal conversion efficiency, stability, thermal conductivity, and anti-fouling capabilities. Finally, the practical performance of the light-water evaporator under various environmental conditions was validated, demonstrating its excellent stability and durability. It is capable of effectively applying to high-efficiency water resource utilization and solar energy conversion fields.

A Low-Power General Matrix Multiplication Accelerator with Sparse Weight-and-Output Stationary Dataflow.

General matrix multiplication (GEMM) in machine learning involves massive computation and data movement, which restricts its deployment on resource-constrained devices. Although data reuse can reduce data movement during GEMM processing, current approaches fail to fully exploit its potential. This work introduces a sparse GEMM accelerator with a weight-and-output stationary (WOS) dataflow and a distributed buffer architecture. It processes GEMM in a compressed format and eliminates on-chip transfers of both weights and partial sums. Furthermore, to map the compressed GEMM of various sizes onto the accelerator, an adaptable mapping scheme is designed. However, the irregular sparsity of weight matrices makes it difficult to store them in local buffers with the compressed format; denser vectors can exceed the buffer capacity, while sparser vectors may lead to the underutilization of buffers. To address this complication, this work also proposes an offline sparsity-aware shuffle strategy for weights, which balances the utilization of distributed buffers and minimizes buffer waste. Finally, a low-cost sparse computing method is applied to the WOS dataflow with globally shared inputs to achieve high computing throughput. Experiments with an FPGA show that the proposed accelerator achieves 1.73× better computing efficiency and 1.36× higher energy efficiency than existing approaches.

Influence of Synthesis Conditions on the Capacitance Performance of Hydrothermally Prepared MnO2 for Carbon Xerogel-Based Solid-State Supercapacitors.

In this study, the potential to modify the phase structure and morphology of manganese dioxide synthesized via the hydrothermal route was explored. A series of samples were prepared at different synthesis temperatures (100, 120, 140, and 160 °C) using KMnO4 and MnSO4·H2O as precursors. The phase composition and morphology of the materials were analyzed using various physicochemical methods. The results showed that, at the lowest synthesis temperature (100 °C), an intercalation compound with composition K1.39Mn3O6 and a very small amount of α-MnO2 was formed. At higher temperatures (120-160 °C), the amount of α-MnO2 increased, indicating the formation of two clearly distinguished crystal structures. The sample obtained at 160 °C exhibited the highest specific surface area (approximately 157 m2/g). These two-phase (α-MnO2/K1.39Mn3O6) materials, synthesized at the lowest and highest temperatures, respectively, and containing an appropriate amount of carbon xerogel, were tested as active mass for positive electrodes in a solid-state supercapacitor, using a Na+-form Aquivion® membrane as the polymer electrolyte. The electrochemical evaluation showed that the composite with the higher specific surface area, containing 75% manganese dioxide, demonstrated improved characteristics, including 96% capacitance retention after 5000 charge/discharge cycles and high energy efficiency (approximately 99%). These properties highlight its potential for application in solid-state supercapacitors.

Efficient algorithm for resource optimization in optical communication networks.

Beyond 5 G and 6 G, communication systems should be able to deliver high throughput, low latency, high dependability, and high energy efficiency services. The creation of hybrid systems that can meet and largely satisfy these needs is promised by the merging of systems based on optical communication and radio frequency (RF). Smart devices may work together to cooperatively train Machine Learning (ML) models in a distributed fashion using Federated Learning (FL), all without disclosing personal information to a central server. This paper proposes a new solution to optimize the network resources in optical-RF communication network. The main idea is to optimize user selection, transmission power and channel estimation based on multilayer perception. Then, the loss function is minimized through joint optimization of user selection and transmission power. Simulation results show that, the proposed algorithm has better performance as compared with existing algorithms.

Multifunctional Cu3BiS3-BP@PEI Radiosensitizer with Enhanced Reactive Oxygen Species Activity for Multimodal Synergistic Therapy.

Development of radiosensitizers with high-energy deposition efficiency, electron transfer, and oxidative stress amplification will help to improve the efficiency of radiotherapy. To overcome the drawbacks of radiotherapy alone, it is also crucial to design a multifunctional radiosensitizer that simultaneously realizes multimodal treatment and tumor microenvironment modulation. Herein, a multifunctional radiosensitizer based on the Cu3BiS3-BP@PEI nanoheterostructure (NHS) for multimodal cancer treatment is designed. Cu3BiS3-BP@PEI NHS is able to deposit a high radiation dose into cancer cells, enhancing the radiotherapy effect. Due to the heterostructure and the synergistic effect of Cu3BiS3 and black phosphorus (BP), significantly boosted 1O2 and •OH generation is obtained under X-ray irradiation, which is promising for extremely efficient radiodynamic therapy. More importantly, the acidic tumor microenvironment (TME) can induce the cycle conversion of Cu2+ to Cu+, oxidizing glutathione (GSH) and catalyzing intracellular overproduction of H2O2 into highly toxic •OH, which thus further enhances reactive oxygen species (ROS) production and reduces GSH-associated radioresistance. Furthermore, Cu3BiS3-BP@PEI NHS has an excellent photothermal effect and can effectively transform light into heat. The outcomes of the in vitro and in vivo research confirm that the as-prepared Cu3BiS3-BP@PEI NHS has a high synergistic therapeutic efficacy at a low radiation dose. This work provides a viable approach to constructing a multifunctional radiosensitizer for deep tumor treatment with TME-triggered multiple synergistic therapies.

Electrochemical Ammonia Synthesis: The Energy Efficiency Challenge.

We discuss the challenges associated with achieving high energy efficiency in electrochemical ammonia synthesis at near-ambient conditions. The current Li-mediated process has a theoretical maximum energy efficiency of ∼28%, since Li deposition gives rise to a very large effective overpotential. As a starting point toward finding electrocatalysts with lower effective overpotentials, we show that one reason why Li and alkaline earth metals work as N2 reduction electrocatalysts at ambient conditions is that the thermal elemental processes, N2 dissociation and NH3 desorption, are both facile at room temperature for these metals. Many transition metals, which have less negative reduction potentials and thus lower effective overpotentials, can dissociate N2 at these conditions but they all bind NH3 too strongly. Strategies to circumvent this problem are discussed, as are the other requirements for a good N2 reduction electrocatalyst.

Life Cycle Assessment of Methanol Production from Municipal Solid Waste: Environmental Comparison with Landfilling and Incineration

Inadequate waste management strategies play a significant role in exacerbating environmental challenges, such as increased greenhouse gas emissions, resource depletion, and other adverse ecological impacts. These issues are aggravated by the global rise in municipal solid waste (MSW) generation, surpassing the rate of population growth. Simultaneously, there is an urgent demand for sustainable energy solutions to combat climate change and its wide-ranging impacts. In response, this study addresses a critical question: is methanol production from MSW, a waste-to-chemical (WtC) alternative based on circular economy principles, a more environmentally sustainable approach compared to traditional waste-to-energy (WtE) methods like landfilling with biogas recovery and incineration? To answer this, this study evaluates the environmental performance of MSW-to-methanol technologies using life cycle assessment (LCA), focusing on key indicators such as global warming potential, resource depletion, and impacts on human health and ecosystem quality. The results reveal that methanol production from MSW significantly reduces global warming potential (GWP) by 87% compared to landfilling and 56% compared to incineration. Additionally, the process demonstrates high energy efficiency in electricity generation, achieving 80% of the output of incineration. These findings position MSW-to-methanol as a promising alternative for advancing sustainable waste management and renewable energy transitions. While the technology is still in its developmental stages, this research highlights the need for further advancements and policy support to enhance feasibility and scalability. By providing a comparative environmental analysis, this study contributes to identifying innovative pathways for addressing pressing waste management and energy sustainability challenges.

Introducing Sulfur in VNi-Layered Double Hydroxide Enables Efficient Electrocatalytic Oxidation of Benzylamine with High Current Densities.

The replacement of the thermodynamically unfavorable anodic oxygen evolution reaction (OER) with a more favorable organic oxidation reaction, such as the anodic oxidation of benzylamine, has garnered significant interest in hybrid water electrolyzer cells. This approach promises the production of value-added chemicals alongside hydrogen fuel generation, improving overall energy efficiency. However, achieving high current density for benzylamine oxidation without interference from OER remains a challenge, limiting the practical efficiency of the electrolyzer cell. In this study, we investigated a room temperature method for sulfur introduction in VNi-layered double hydroxide (LDH) catalyst and its application for electrocatalytic benzylamine oxidation. The S-introduction in VNi-LDH was found to modulate the electronic states of nickel and vanadium, increasing the number of active sites, electrochemical surface area, and charge transfer properties. The resulting S-VNi-LDH catalyst achieved a high current density of 400 mA cm-2 at only 1.39 V vs RHE potential for benzylamine oxidation, avoiding interference from oxygen evolution. The catalyst demonstrated 100% selectivity (Faradaic Efficiency = 98.6%) for the conversion of benzylamine into benzonitrile within 2.5 h of the reaction. In a two-electrode electrolysis system, S-VNi-LDH achieved a current density of 400 mA cm-2 at a cell voltage of 1.50 V when OER was substituted with benzylamine oxidation. The S-VNi-LDH showed energy consumption of 4.67 kWh/m3 H2 for OER and 1.31 kWh/m3 H2 during benzylamine oxidation, indicating a high energy efficiency with exceptional stability over five cycles, maintaining 98.6 ± 0.4% FE and consistent voltage. The S-VNi-LDH also oxidized various amines, including substituted benzylamines and secondary amines, achieving high conversion (95-97%) and faradaic efficiency (85.8-98%). This study presents an eco-friendly, room-temperature method for S-doping in VNi-LDH, which out performed the reported catalysts in the literature.

Photo-Controllable Förster Resonance Energy Transfer Based on Dynamic Chiral Self-Assembly of Sequence-Defined Amphiphilic Alternating Azopeptoids.

Endowing biomimetic sequence-controlled polymers with chiral functionality to construct stimuli-responsive chiral materials offers a promising approach for innovative chiroptical switch, but it remains challenging. Herein, it is reported that the self-assembly of sequence-defined chiral amphiphilic alternating azopeptoids to generate photo-responsive and ultrathin bilayer peptoidosomes with a vesicular thickness of ≈1.50nm and a diameter of around ≈290nm. The photoisomerization of azobenzene moiety facilitates a reversible structural transformation from isotropic peptoidosomes to anisotropic 1D helical nanoribbons (≈80nm width) under the alternating irradiation with UV and visible lights, consequently leading to the chirality expression and transfer from chiral asymmetric center to achiral azobenzene units. As a biomimetic model with deformation-induced energy transfer, a non-invasive azobenzene-based Förster resonance energy transfer system is unprecedentedly constructed via the introduction of a fluorescent donor of pyrene derivatives and sequentially photo-regulated the donor/acceptor ratio, displaying a reversible gradient fluorescent color variation from blue to yellow (a broad Stokes shift of ≈200nm) and a high-efficient energy transfer efficiency of 97.2%. The photo-controllable photoluminescence phenomenon endows these chiral aggregates with a proof-of-concept application on multi-colored information encryption. This work provides a prospective strategy to fabricate stimuli-responsive chiral biomimetic materials with a potential on the light-controllable switches.

Reinforcement Learning-Based Scheduling Optimization for DNN Accelerators

This paper presents a novel reinforcement learning (RL)-based framework for optimizing dataflow scheduling in Deep Neural Network (DNN) accelerators. As DNNs grow increasingly complex, efficient hardware accelerators such as TPUs and custom-designed ASICs are essential to meet high performance and energy efficiency demands. However, optimizing dataflow scheduling remains challenging due to the vast design space and dynamic hardware constraints. The proposed framework uses Proximal Policy Optimization (PPO) to dynamically adjust scheduling strategies. After the RL agent selects the rows to optimize, a brute-force search is employed to find the optimal solutions for these rows, ensuring that the scheduling satisfies both DNN parameters and hardware resource constraints. We validated the framework on various DNN models, including YOLO v3, Inception v4, MobileNet v3, and ResNet-50, across multiple accelerators like Eyeriss, TPU v3, and Simba. The experimental results show substantial improvements, with RL-Scheduling achieving up to a 65.6% reduction in execution cycles and a 59.7% improvement in energy efficiency over existing scheduling algorithms. Additionally, the method demonstrates superior algorithm execution efficiency compared to most existing approaches.