Reinforcement Learning Based Quantization Strategy Optimal Assignment Algorithm for Mixed Precision

The quantization algorithm compresses the original network by reducing the numerical bit width of the model,which improves the computation speed. Because different layers have different redundancy and sensitivity to databit width. Reducing the data bit width will result in a loss of accuracy. Therefore, it is difficult to determinethe optimal bit width for different parts of the network with guaranteed accuracy. Mixed precision quantizationcan effectively reduce the amount of computation while keeping the model accuracy basically unchanged. In thispaper, a hardware-aware mixed precision quantization strategy optimal assignment algorithm adapted to low bitwidth is proposed, and reinforcement learning is used to automatically predict the mixed precision that meets theconstraints of hardware resources. In the state-space design, the standard deviation of weights is used to measurethe distribution difference of data, the execution speed feedback of simulated neural network accelerator inferenceis used as the environment to limit the action space of the agent, and the accuracy of the quantization model afterretraining is used as the reward function to guide the agent to carry out deep reinforcement learning training. Theexperimental results show that the proposed method obtains a suitable model layer-by-layer quantization strategyunder the condition that the computational resources are satisfied, and themodel accuracy is effectively improved.The proposed method has strong intelligence and certain universality and has strong application potential in thefield of mixed precision quantization and embedded neural network model deployment.

Operation optimization strategy of existing residential building energy-saving renovation market:From the perspective of subject behavior

The energy-saving renovation of existing residential buildings is a crucial measure to achieve the strategic goal of energy conservation and emission reduction in China and build ecologically livable cities.This article focuses on the perspective of subject behavior,starting from analyzing the current situation and difficulties of the operation of the energy-saving renovation market for existing residential buildings in China,drawing on the practical experience of the operation of the existing residential building energy-saving renovation market abroad.Based on principles such as systematicity,humanization,feasibility,and sustainability,the article constructs an operation optimization system of the existing residential building energy-saving renovation market from the perspective of subject behavior.In order to provide a reference for the healthy and orderly operation of the existing residential building energy-saving renovation market,this paper proposes implementation strategies for optimizing the operation of the existing residential building energy-saving renovation market.Suggestions are proposed from four aspects:optimizing the market environment,innovating the financing model,building the information sharing platform,and utilizing the synergies of the main subjects.

Optimized Strategy for Layout of Crop Production Areas in Hunan Province

The optimized strategy made a comprehensive consideration of resources, technology, market orientation, production scale, industry basis and layout based on the principle of crop security and farmers’ income increasing, and determined the general planning on layout and structure optimization of future crop production ar-eas, with present crop production, market outlook, future industry development, con-cluding crop production characteristics of the 4 crop regions, and proposing function orientation and highlights.

Implementation of Radial Basis Function Artificial Neural Network into an Adaptive Equivalent Consumption Minimization Strategy for Optimized Control of a Hybrid Electric Vehicle

Continued increases in the emission of greenhouse gases by passenger vehicles have accelerated the production of hybrid electric vehicles. With this increase in production, there has been a parallel demand for continuously improving strategies of hybrid electric vehicle control. The goal of an ideal control strategy is to maximize fuel economy while minimizing emissions. Methods exist by which the globally optimal control strategy may be found. However, these methods are not applicable in real-world driving applications since these methods require a priori knowledge of the upcoming drive cycle. Real-time control strategies use the global optimal as a benchmark against which performance can be evaluated. The goal of this work is to use a previously defined strategy that has been shown to closely approximate the global optimal and implement a radial basis function (RBF) artificial neural network (ANN) that dynamically adapts the strategy based on past driving conditions. The strategy used is the Equivalent Consumption Minimization Strategy (ECMS), which uses an equivalence factor to define the control strategy and the power train component torque split. An equivalence factor that is optimal for a single drive cycle can be found offline with a priori knowledge of the drive cycle. The RBF-ANN is used to dynamically update the equivalence factor by examining a past time window of driving characteristics. A total of 30 sets of training data (drive cycles) are used to train the RBF-ANN. For the majority of drive cycles examined, the RBF-ANN implementation is shown to produce fuel economy values that are within ±2.5% of the fuel economy obtained with the optimal equivalence factor. The advantage of the RBF-ANN is that it does not require a priori drive cycle knowledge and is able to be implemented in real-time while meeting or exceeding the performance of the optimal ECMS. Recommendations are made on how the RBF-ANN could be improved to produce better results across a greater array of driving conditions.

Optimal guidance strategy for flexible load based on hybrid direct load control and time of use

The time-of-use(TOU)strategy can effectively improve the energy consumption mode of customers,reduce the peak-valley difference of load curve,and optimize the allocation of energy resources.This study presents an Optimal guidance mechanism of the flexible load based on strategies of direct load control and time-of-use.First,this study proposes a period partitioning model,which is based on a moving boundary technique with constraint factors,and the Dunn Validity Index(DVI)is used as the objective to solve the period partitioning.Second,a control strategy for the curtailable flexible load is investigated,and a TOU strategy is utilized for further modifying load curve.Third,a price demand response strategy for adjusting transferable load is proposed in this paper.Finally,through the case study analysis of typical daily flexible load curve,the efficiency and correctness of the proposed method and model are validated and proved.

Motion simulation of moorings using optimized LSTM neural network

Mooring arrays have been widely deployed in sustained ocean observation in high resolution to measure finer dynamic features of marine phenomena.However,the irregular posture changes and nonlinear response of moorings under the effect of ocean currents face huge challenges for the deployment of mooring arrays,which may cause the deviations of measurements and yield a vacuum of observation in the upper ocean.We developed a data-driven mooring simulation model based on LSTM(long short-term memory)neural network,coupling the ocean current with position data from moorings to predict the motion of moorings,including single-step output prediction and multi-step prediction.Based on the predictive information,the formation of the mooring array can be adjusted to improve the accuracy and integrity of measurements.Moreover,we proposed the cuckoo search(CS)optimization algorithm to tune the parameters of LSTM,which improves the robustness and generalization of the model.We utilize the datasets observed from moorings anchored in the Kuroshio Extension region to train and validate the simulation model.The experimental results demonstrate that the model can remarkably improve prediction accuracy and yield stable performance.Moreover,compared with other optimization algorithms,CS is more efficient and performs better in simulating the motion of moorings.

Urban Drainage Network Scheduling Strategy Based on Dynamic Regulation: Optimization Model and Theoretical Research

With the acceleration of urbanization in China,the discharge of domestic sewage and industrial wastewater is increasing,and accidents of sewage spilling out and polluting the environment occur from time to time.Problems such as imperfect facilities and backward control methods are com-mon in the urban drainage network systems in China.Efficient drainage not only strengthens infrastructure such as rain and sewage diversion,pollution source monitoring,transportation,drainage and storage but also urgently needs technical means to monitor and optimize production and operation.Aiming at the optimal control of single-stage pumping stations and the coordinated control between two-stage pumping stations,this paper studies the modelling and optimal control of drainage network systems.Based on the Long Short Term Memory(LSTM)water level prediction model of the sewage pumping stations,and then based on the mechanism analysis of drainage pipe network,the factors that may cause the water level change of pumping station are obtained.Grey correlation analysis is carried out on these influencing factors,and the prediction model is established by taking the factors with a high correlation degree as input.The research results show that compared with the traditional prediction model,the LSTM model not only has higher prediction accuracy but also has better inflection point tracking ability.

Optimal Treatment Strategy for Infectious Diseases with Two Treatment Stages

In this paper, a disease transmission model with two treatment stages is proposed and analyzed. The results indicate that the basic reproduction number is a critical threshold for the prevalence of the disease. If the basic reproduction number is less than one, the disease free equilibrium is globally asymptotically stable. Otherwise, the endemic equilibrium is globally asymptotically stable. Therefore, besides the basic reproduction number, a new marker for characterizing the seriousness of the disease, named as dynamical final infective size, is proposed, which differs from traditional final size because the proposed model includes the natural birth and death. Finally, optimization strategies for limited medical resources are obtained from the perspectives of basic reproduction number and dynamical final infective size, and the real-world disease management scenarios are given based on these finding.

Analysis of Enterprise Marketing Strategy Optimization from the Perspective of New Media

With the advent of the economic era of“Internet+,”new media has become a new means of enterprise marketing by virtue of its own advantages,including fast communication speed,diversified communication channels,low cost,and novel content.Enterprises should actively integrate into the new media era,constantly improve their cultural soft power and new media marketing ability,build new marketing systems,set up professional new media marketing teams,and further improve their marketing ability;innovate new media marketing content,attract consumers’attention,and expand the audience group;open up new media marketing channels,carry out diversified marketing,comprehensively enhance their marketing ability,and succeed in the fierce market competition.

Cognitive interference decision method for air defense missile fuze based on reinforcement learning

To solve the problem of the low interference success rate of air defense missile radio fuzes due to the unified interference form of the traditional fuze interference system,an interference decision method based Q-learning algorithm is proposed.First,dividing the distance between the missile and the target into multiple states to increase the quantity of state spaces.Second,a multidimensional motion space is utilized,and the search range of which changes with the distance of the projectile,to select parameters and minimize the amount of ineffective interference parameters.The interference effect is determined by detecting whether the fuze signal disappears.Finally,a weighted reward function is used to determine the reward value based on the range state,output power,and parameter quantity information of the interference form.The effectiveness of the proposed method in selecting the range of motion space parameters and designing the discrimination degree of the reward function has been verified through offline experiments involving full-range missile rendezvous.The optimal interference form for each distance state has been obtained.Compared with the single-interference decision method,the proposed decision method can effectively improve the success rate of interference.

A review of Al-based material dopants for high-performance solid state lithium metal batteries

As the world transitions to green energy, there is a growing focus among many researchers on the requirement for high-efficient and safe batteries. Solid-state lithium metal batteries(SSLMBs) have emerged as a promising alternative to traditional liquid lithium-ion batteries(LIBs), offering higher energy density, enhanced safety, and longer lifespan. The rise of SSLMBs has brought about a transformation in energy storage, with aluminum(Al)-based material dopants playing a crucial role in advancing the next generation of batteries. The review highlights the significance of Al-based material dopants in SSLMBs applications, particularly its contributions to solid-state electrolytes(SSEs), cathodes, anodes,and other components of SSLMBs. Some studies have also shown that Al-based material dopants effectively enhance SSE ion conductivity, stabilize electrode and SSE interfaces, and suppress lithium dendrite growth, thereby enhancing the electrochemical performance of SSLMBs. Despite the above mentioned progresses, there are still problems and challenges need to be addressed. The review offers a comprehensive insight into the important role of Al in SSLMBs and addresses some of the issues related to its applications, endowing valuable support for the practical implementation of SSLMBs.

A Review of Anode Materials for Dual‑Ion Batteries

Distinct from"rockingchair"lithium-ion batteries(LIBs),the unique anionic intercalation chemistry on the cathode side of dual-ion batteries(DIBs)endows them with intrinsic advantages of low cost,high voltage,and ecofriendly,which is attracting widespread attention,and is expected to achieve the next generation of large-scale energy storage applications.Although the electrochemical reactions on the anode side of DIBs are similar to that of LIBs,in fact,to match the rapid insertion kinetics of anions on the cathode side and consider the compatibility with electrolyte system which also serves as an active material,the anode materials play a very important role,and there is an urgent demand for rational structural design and performance optimization.A review and summarization of previous studies will facilitate the exploration and optimization of DIBs in the future.Here,we summarize the development process and working mechanism of DIBs and exhaustively categorize the latest research of DIBs anode materials and their applications in different battery systems.Moreover,the structural design,reaction mechanism and electrochemical performance of anode materials are briefly discussed.Finally,the fundamental challenges,potential strategies and perspectives are also put forward.It is hoped that this review could shed some light for researchers to explore more superior anode materials and advanced systems to further promote the development of DIBs.

Optimization Strategies of Na_(3)V_(2)(PO_(4))_(3) Cathode Materials for Sodium‑Ion Batteries

Na_(3)V_(2)(PO_(4))_(3)(NVP)has garnered great attentions as a prospective cathode material for sodium-ion batteries(SIBs)by virtue of its decent theoretical capacity,superior ion conductivity and high structural stability.However,the inherently poor electronic conductivity and sluggish sodium-ion diffusion kinetics of NVP material give rise to inferior rate performance and unsatisfactory energy density,which strictly confine its further application in SIBs.Thus,it is of significance to boost the sodium storage performance of NVP cathode material.Up to now,many methods have been developed to optimize the electrochemical performance of NVP cathode material.In this review,the latest advances in optimization strategies for improving the electrochemical performance of NVP cathode material are well summarized and discussed,including carbon coating or modification,foreign-ion doping or substitution and nanostructure and morphology design.The foreign-ion doping or substitution is highlighted,involving Na,V,and PO_(4)^(3−)sites,which include single-site doping,multiple-site doping,single-ion doping,multiple-ion doping and so on.Furthermore,the challenges and prospects of high-performance NVP cathode material are also put forward.It is believed that this review can provide a useful reference for designing and developing high-performance NVP cathode material toward the large-scale application in SIBs.

Toward innovative veterinary nanoparticle vaccines

Nanoparticles are significant for veterinary vaccine development because they are safer and more effective than conventional formulations.One promising area of research involves self-assembled protein nanoparticles(SAPNs),which have shown potential for enhancing antigen-presenting cell uptake,B-cell activation,and lymph node trafficking.Numerous nanovaccines have been utilized in veterinary medicine,including natural self-assembled protein nanoparticles,rationally designed self-assembled protein nanoparticles,animal virus-derived nanoparticles,bacteriophagederived nanoparticles,and plant-derived nanoparticles,which will be discussed in this review.SAPN vaccines can produce robust cellular and humoral immune responses and have been shown to protect against various animal infectious diseases.This article attempts to summarize these diverse nanovaccine types and their recent research progress in the field of veterinary medicine.Furthermore,this paper highlights their disadvantages and methods for improving their immunogenicity.

Strategic insights into the cultivation of pancreatic cancer organoids from endoscopic ultrasonography-guided biopsy tissue

BACKGROUND The frequent suboptimal efficacy of endoscopic ultrasound-guided fine-needle biopsy(EUS-FNB)to culture pancreatic cancer(PC)organoids(PCOs)poses a major challenge in the advancement of personalized medicine for advanced PC.AIM To explore how to obtain appropriate puncture tissues from EUS-FNB and optimize the strategy for efficiently constructing PCOs,providing an efficient tool for the advancement of personalized medicine.METHODS Patients who underwent EUS-FNB for the diagnosis of PC tissue were prospectively enrolled.We refined the endoscopic biopsy procedures and organoid cultivation techniques.All tissue specimens verified by on-site pathological assessment were cultured in a semi-suspended medium in a microfluidic environment.We assessed differences in PCOs cultured beyond and below five generations examining patient demographics,specimen and organoid attributes,and the sensitivity of organoids to a panel of clinical drugs through cell viability assays.RESULTS In this study,16 patients with PC were recruited,one sample was excluded because onsite cytopathology showed no tumor cells.Successful organoid generation occurred in 93.3%(14 of 15)of the EUS-FNB specimens,with 60%(9 of 15)sustaining over five generations.Among these patients,those with a history of diabetes,familial cancer,or larger tumors exhibited enhanced PCO expandability.The key factors influencing longterm PCOs expansion included initial needle sample quality(P=0.005),rapid initiation of organoid culture postisolation(P≤0.001),and high organoid activity(P=0.031).Drug sensitivity analysis revealed a partial response in two patients following therapeutic intervention and surgery and stable disease in four patients,indicating a moderate correlation between organoid response and clinical outcomes.CONCLUSION Optimal initial needle sampling,rapid and precise biopsy sample processing,process isolated samples as soon as possible,and sufficient cellular material are crucial for successful cultivating PCOs.High organoid activity is an important factor in maintaining their long-term expansion,which is essential for shortening the time of drug sensitivity analysis and is the basis of PC research.

Research progresses on cathode materials of aqueous zinc-ion batteries

Electrochemical energy storage and conversion techniques that exhibit the merits such as high energy density,rapid response kinetics,economical maintenance requirements and expedient installation procedures will hold a pivotal role in the forthcoming energy storage technologies revolution.In recent years,aqueous zinc-ion batteries(AZIBs)have garnered substantial attention as a compelling candidate for large-scale energy storage systems,primarily attributable to their advantageous featu res encompassing cost-effectiveness,environmental sustainability,and robust safety profiles.Currently,one of the primary factors hindering the further development of AZIBs originates from the challenge of cathode materials.Specifically,the three mainstream types of mainstream cathode materials,in terms of manganese-based compounds,vanadium-based compounds and Prussian blue analogues,surfer from the dissolution of Mn~(2+),in the low discharge voltage,and the low specific capacity,respectively.Several strategies have been developed to compensation the above intrinsic defects for these cathode materials,including the ionic doping,defect engineering,and materials match.Accordingly,this review first provides a systematic summarization of the zinc storage mechanism in AZIBs,following by the inherent merit and demerit of three kind of cathode materials during zinc storage analyzed from their structure characteristic,and then the recent development of critical strategies towards the intrinsic insufficiency of these cathode materials.In this review,the methodologies aimed at enhancing the efficacy of manganese-based and vanadium-based compounds are emphasis emphasized.Additionally,the article outlines the future prospective directions as well as strategic proposal for cathode materials in AZIBs.

Optimization Strategies of Beijing Elderly Care Service Stations Based on Questionnaire Survey Method: A Case Study of Zhanlan Road Street of Xicheng District

4 elderly care service stations in Zhanlan Road Street,Xicheng District,Beijing are selected,and questionnaires are designed and distributed to the surrounding elderly population to understand their needs and satisfaction with the station environment.By observing elderly care service stations on site,the characteristics,obstacles,and shortcomings of the environment are recorded,and relevant data are collected and analyzed,such as the characteristics of the elderly population being interviewed,the planning and design data of the station environment,and the distribution of service facilities.The overall characteristics of the spatial environment of elderly care stations are summarized,and renovation measures and optimization suggestions are provided for the current shortcomings,thereby providing some basis for the spatial design of community elderly care service stations in the future.

Construction and Practice of Education and Teaching Quality Assurance Systems in Applied Colleges and Universities under the Background of Digital Intelligence

This paper discusses the optimization strategy of education and teaching quality assurance systems in applied colleges and universities under the background of digital intelligence.It first summarizes the relevant theories of digital intelligence transformation and analyzes the impact of digital intelligence transformation on higher education.Secondly,this paper puts forward the principles of constructing the quality assurance system of applied colleges,including strengthening the quality assurance consciousness,improving teachers’digital literacy,and implementing digital intelligence governance.From the practical perspective,this paper expounds on strategies such as optimizing educational teaching resource allocation,constructing a diversified evaluation system of teaching quality,strengthening the construction and training of teaching staff,and innovating teaching management methods.Specific optimization measures are put forward,such as improving policies,regulations,and system guarantees,strengthening cooperation between schools and enterprises,integrating industry,school,and research,building an educational information platform,and improving the monitoring and feedback mechanism of educational quality.

Teaching Mode Reform of the Electrical Automation Control Course

Under the background of the new era,higher requirements are put forward for colleges and universities to carry out teaching reform.Teachers of electrical automation control courses should update their teaching ideas regularly,innovate teaching methods,and take novel and effective measures to carry out related work.Because electrical automation control courses tend to be technology-oriented,teachers can help students consolidate basic knowledge.They should also focus on developing practical skills so students can easily adapt to future job positions.However,there are many problems in the actual teaching process,which hampers the improvement of the teaching quality to a certain extent.In view of this,this paper presents an in-depth exploration based on theories and practical experiences.It starts with an analysis of the current teaching status of electrical automation control courses in colleges and universities,followed by suggestions to improve them based on their characteristics and students’needs.

Research Status,Practical Dilemmas,and Optimization Strategies of Blended Teaching in Universities

Blended teaching has emerged as a prominent subject in the recent reform and innovation of higher education.It has become imperative and guiding for colleges and universities to embrace a mixed teaching approach that aligns with the evolving needs of education and teaching in the new era.This paper aims to provide a comprehensive overview of the research status surrounding blended teaching,encompassing fundamental issues,teaching design,practical guidance,teaching effectiveness,and evaluation.By critically examining the current challenges associated with blended teaching,this study proposes optimization strategies including enhancing student participation and interaction,promoting deep learning,improving teachers’preparedness,teaching technologies,and curriculum design capabilities,strengthening top-level design,and perfecting evaluation and incentive mechanisms.These strategies provide new directions for the reform of blended teaching.