Traditional Expert Research Articles

Optimal load distribution control for airport terminal chiller units based on deep reinforcement learning

The building sector consumes substantial energy, with HVAC systems accounting for nearly half of the total energy use. Optimizing chiller unit operations is crucial for reducing energy consumption. This research introduces a novel model-free control method for optimizing central chiller load distribution using deep reinforcement learning (DRL). The aim is to address the challenges posed by the high-dimensional complexity and parameter tuning in traditional meta-heuristic algorithms. The proposed method leverages deep neural networks combined with reinforcement learning to optimize chiller control based on real-world data. Case studies conducted in a large airport terminal demonstrate that this approach achieves a 7.71 % energy savings, closely matching the performance of model-based control methods with only a 0.26 % difference. The results indicate that the DRL method not only outperforms traditional expert control systems but also offers a feasible alternative for optimal control in complex, variable environments with limited data. This study contributes to the field by filling a research gap in applying DRL for optimal chiller load (OCL) control and demonstrates its potential for large-scale public building applications. The novelty of this research lies in its model-free approach, which provides a robust solution for energy optimization in dynamic and uncertain environments.

Expert consensus on quality control indicators for integrative medicine in ankylosing spondylitis (2023 version)

AbstractThe integrated treatment with Chinese and Western medicine plays an important role in the treatment of ankylosing spondylitis (AS), which can significantly relieve the patient's symptoms and improve quality of life. However, as a result of the lack of standardization of clinical management in some regions, the quality of AS treatment in China remains unoptimistic. In 2021, initiated by the Center for Integrative Medicine of China‐Japan Friendship Hospital, led by the traditional Chinese medicine department of rheumatology of China‐Japan Friendship Hospital, relying on the rheumatology branch of the Beijing Association of the integrating of traditional and Western medicine, and using the Delphi method, an expert consensus on “Quality Control Indicators for Integrative Medicine in Ankylosing Spondylitis” was established (Beijing Association of the Integrating of Traditional and Western Medicine Standardization Expert Consensus Filing Number: 2021Z031A2). The consensus revision updates the 2021 version by adding four items, which will provide a reference for the quality control of integrated traditional Chinese and Western medicine for AS in China and help improve the current diagnosis and treatment status of AS.

End-to-End Autonomous Driving Decision Method Based on Improved TD3 Algorithm in Complex Scenarios.

The ability to make informed decisions in complex scenarios is crucial for intelligent automotive systems. Traditional expert rules and other methods often fall short in complex contexts. Recently, reinforcement learning has garnered significant attention due to its superior decision-making capabilities. However, there exists the phenomenon of inaccurate target network estimation, which limits its decision-making ability in complex scenarios. This paper mainly focuses on the study of the underestimation phenomenon, and proposes an end-to-end autonomous driving decision-making method based on an improved TD3 algorithm. This method employs a forward camera to capture data. By introducing a new critic network to form a triple-critic structure and combining it with the target maximization operation, the underestimation problem in the TD3 algorithm is solved. Subsequently, the multi-timestep averaging method is used to address the policy instability caused by the new single critic. In addition, this paper uses Carla platform to construct multi-vehicle unprotected left turn and congested lane-center driving scenarios and verifies the algorithm. The results demonstrate that our method surpasses baseline DDPG and TD3 algorithms in aspects such as convergence speed, estimation accuracy, and policy stability.

A deep‐learning framework for enhancing habitat identification based on species composition

AbstractAimsThe accurate classification of habitats is essential for effective biodiversity conservation. The goal of this study was to harness the potential of deep learning to advance habitat identification in Europe. We aimed to develop and evaluate models capable of assigning vegetation‐plot records to the habitats of the European Nature Information System (EUNIS), a widely used reference framework for European habitat types.LocationThe framework was designed for use in Europe and adjacent areas (e.g., Anatolia, Caucasus).MethodsWe leveraged deep‐learning techniques, such as transformers (i.e., models with attention components able to learn contextual relations between categorical and numerical features) that we trained using spatial k‐fold cross‐validation (CV) on vegetation plots sourced from the European Vegetation Archive (EVA), to show that they have great potential for classifying vegetation‐plot records. We tested different network architectures, feature encodings, hyperparameter tuning and noise addition strategies to identify the optimal model. We used an independent test set from the National Plant Monitoring Scheme (NPMS) to evaluate its performance and compare its results against the traditional expert systems.ResultsExploration of the use of deep learning applied to species composition and plot‐location criteria for habitat classification led to the development of a framework containing a wide range of models. Our selected algorithm, applied to European habitat types, significantly improved habitat classification accuracy, achieving a more than twofold improvement compared to the previous state‐of‐the‐art (SOTA) method on an external data set, clearly outperforming expert systems. The framework is shared and maintained through a GitHub repository.ConclusionsOur results demonstrate the potential benefits of the adoption of deep learning for improving the accuracy of vegetation classification. They highlight the importance of incorporating advanced technologies into habitat monitoring. These algorithms have shown to be better suited for habitat type prediction than expert systems. They push the accuracy score on a database containing hundreds of thousands of standardized presence/absence European surveys to 88.74%, as assessed by expert judgment. Finally, our results showcase that species dominance is a strong marker of ecosystems and that the exact cover abundance of the flora is not required to train neural networks with predictive performances. The framework we developed can be used by researchers and practitioners to accurately classify habitats.

A dual-track feature fusion model utilizing Group Shuffle Residual DeformNet and swin transformer for the classification of grape leaf diseases

Grape cultivation is important globally, contributing to the agricultural economy and providing diverse grape-based products. However, the susceptibility of grapes to disease poses a significant threat to yield and quality. Traditional disease identification methods demand expert knowledge, which limits scalability and efficiency. To address these limitations our research aims to design an automated deep learning approach for grape leaf disease detection. This research introduces a novel dual-track network for classifying grape leaf diseases, employing a combination of the Swin Transformer and Group Shuffle Residual DeformNet (GSRDN) tracks. The Swin Transformer track exploits shifted window techniques to construct hierarchical feature maps, enhancing global feature extraction. Simultaneously, the GSRDN track combines Group Shuffle Depthwise Residual block and Deformable Convolution block to extract local features with reduced computational complexity. The features from both tracks are concatenated and processed through Triplet Attention for cross-dimensional interaction. The proposed model achieved an accuracy of 98.6%, the precision, recall, and F1-score are recorded as 98.7%, 98.59%, and 98.64%, respectively as validated on a dataset containing grape leaf disease information from the PlantVillage dataset, demonstrating its potential for efficient grape disease classification.

Maternal and Child Health Services Utilization During COVID-19 in Eastern Ethiopia.

Objectives: This study aimed to describe the impact of COVID-19 on maternal and child health service utilization in the Eastern part of Ethiopia. Methods: Comparative analysis was used to examine 2years of maternal and child health service utilization. Data were extracted from client registers. A traditional Expert Modeler and one-way Analysis of Variance were used to compare service utilization before and during the COVID-19 pandemic. Results: A total of 34,576 client records were reviewed, of which 17,100 (49.5%) and 17,476 (50.5%) had visited the MCH service before and during the COVID-19 pandemic, respectively. The total client visit has shown a 2% percentage point increase. However, postnatal care and child immunization services showed a decrease. Moreover, there was a significant difference between service visits before and during COVID-19 (f = 4.6, p < 0.04). Conclusion: Mothers and children have missed or unattended facility appointments due to protective impositions or fear of getting infected with COVID-19, which might suggest a higher proportion of MCH issues were not addressed during the pandemic. The health system should therefore improve its resilience and strengthen its access at the lowest health care inlets.

Consumer flash fiction: A methodology to support the early sensing of far-future innovation opportunities

Consumer-centered innovation with a far-future time frame is challenging because of the ineffectiveness of traditional market research tools in that context. Drawing on scholarly research that has identified creative action as a means of capturing the visionary knowledge of non-experts in scenario development, we propose consumer flash fiction as a methodology that allows firms to access consumer visions about the far future. Flash fiction stories are very short science fiction stories of 150-300 words and flow-written by non-experts in workshops designed specifically for that purpose. However, we find that the involvement of non-experts such as consumers in scenario development is rare. This limits our understanding about the type(s) of insights that farfuture flash fiction-based consumer scenarios can provide to strategists during the early opportunity sensing phase of innovation. To address this gap, we present and analyze the outcomes of three studies that collected about 859 consumer flash fiction stories on different far-future topics. In addition, we contribute to the conceptualization of the so-far lacking understanding of the complementarity between the traditional expert opinion-based science fiction scenario development approaches and consumer flash fiction.

Integrated traditional Chinese and western medicine expert consensus on the diagnosis and treatment of combined allergic rhinitis and asthma syndrome

Combined allergic rhinitis and asthma syndrome (CARAS) refers to a common respiratory disease that occurs simultaneously with clinical or subclinical allergic symptoms of the upper respiratory tract (allergic rhinitis) and the lower respiratory tract (asthma). The incidence of CARAS is high and the quality of life of the patients is greatly affected. At present, treatment of this comprehensive disease is often carried out separately in the otorhinolaryngology and respiratory departments. One of the reasons is a lack of coordinated treatment consensus on the comprehensive management of this disease. As a common respiratory disease, this syndrome also has a profound clinical basis of traditional Chinese medicine in its diagnosis and treatment. Therefore, the Allergy Committee of Chinese Association of Integrative Medicine organized domestic experts in respiratory medicine, otolaryngology, allergy, pediatrics, traditional Chinese Medicine internal medicine and other related fields to discuss and summarize the etiology and anatomical characteristics, pathophysiology and pathogenesis, laboratory examination, diagnostic evaluation and differential diagnosis as well as treatment of both traditional Chinese medicine and western medicine, in order to provide integrated diagnosis and treatment opinions for this common integrative disease of upper and lower respiratory system in clinical practice.

Expert consensus on quality control indicators of integrative medicine in primary Sjögren's syndrome (2023 version)

AbstractTreatment integrating traditional Chinese and Western medicine for primary Sjögren's syndrome (pSS) has yielded positive results and gained wide acceptance. However, the absence of standardized clinical management in certain regions has led to suboptimal outcomes in the diagnosis and treatment of pSS in China. In response, the Center for Traditional Chinese and Western Medicine at the China‐Japan Friendship Hospital, in collaboration with the Beijing Association of Traditional Chinese and Western Medicine Rheumatology Professional Committee, initiated a project led by the Traditional Chinese Medicine Department of Rheumatism. The project (Beijing Association of the Integrating of Traditional and Western Medicine standardization expert consensus record number: 2021Z031A3) aims to develop a set of quality control indicators for the integrated treatment of pSS using traditional Chinese and Western medicine. Based on the 2021 version, the consensus revision incorporates seven new items, focusing on essential assessment areas for pSS patients, treatment response, and healthcare cost management. The 2023 expert consensus update seeks to enhance pSS diagnosis and treatment practices in China, offering a benchmark for medical quality control.

The Intelligent Layout of the Ship Piping System Based on the Optimization Algorithm

The ship piping layout is one of the essential tasks in the detailed design stage of a ship. Traditional manual expert design has disadvantages such as low efficiency, reliance on experience, and subjective influence. Therefore, this paper systematically proposes an intelligent arrangement method for ships’ single, parallel, and branch pipelines. Firstly, the traditional genetic algorithm is improved and combined with the A* algorithm to solve the intelligent arrangement problem of a single pipeline in ships. Then, the parallel pipeline and branch pipeline are split into multiple single pipelines by combining with the connection point strategy to solve the arrangement problem of parallel pipeline and branch pipeline. Finally, the optimized A*-genetic algorithm proposed in this paper is compared with the A* algorithm, particle swarm algorithm, and the labyrinth-genetic algorithm used in previous research through simulation experiments. The results show that the A*-genetic algorithm of this paper is optimal in six indexes, including length, number of elbows, energy value, fitness value, number of optimal solutions, and average number of convergence generations, in the arrangement of the single pipeline. In solving the parallel pipeline and branch pipeline arrangement problems, the all-around performance of this paper’s algorithm is better than that of A*-genetic algorithm and maze–genetic algorithm, respectively. The A*-genetic algorithm of this paper considers the quality of pipeline arrangement and the solution’s efficiency. It verifies the adaptability and superiority of the algorithm for the intelligent arrangement of various types of pipelines in ship pipelines.

Use of microbiological data for the purposes of forensic medical examination

The study objective was to describe the formation of forensic microbiological examination as an analysis of a new type, defined as the detection and registration of reliably measured environmental and physiological changes within the microbial community of corpse in order to substantiate the possibility of using microbiological parameters to establish the prescription of death coming. It has been determined that the knowledge of the patterns of interaction of a human and his corpse with endogenous and exogenous flora provides the basis for solving a number of traditional and new application-oriented expert tasks and the allocation of such a variety of forensic examination as forensic microbiological examination. Endogenous and exogenous human flora and its interaction with living and dead biological tissues are the objects of this kind of examination, and the dynamic patterns of such interaction are the subject of study. One of the initial relevant tasks of forensic microbiological examination consists in development of methods, adequate for the expert task to be solved, choice of the research «target», «models» for comparative analysis and medium, adequate for task in hand, as well as certification of these methods and standardization of assessment criteria for the obtained results.

A novel way to build expert systems with infinite-valued attributes

&lt;abstract&gt;&lt;p&gt;An expert system is a computer program that uses the knowledge of an expert to solve problems in a specific domain. Expert systems are used in a wide variety of fields, such as medicine, financial diagnosis and engineering. The attributes of an expert system are the characteristics of the problems that the system can solve. In traditional expert systems, attributes typically have a finite number of possible values. However, in scenarios where an attribute can assume a value from an infinite (or significantly large finite) set, the expert system cannot be represented using propositional logic. Until now, no method had been identified to implement such a system on a Computer Algebra System. Here, we break new ground by presenting a model that not only addresses this gap but also provides a fresh perspective on previous results. In fact, these prior results can be viewed as specific instances within the broader framework of our proposed solution. In this paper, we put forth an algebraic approach for the development of expert systems capable of handling attributes with infinite values, thereby expanding the problem-solving capacity of these systems.&lt;/p&gt;&lt;/abstract&gt;

Automated text analysis methods to identify the individual structure of motivation for sports and a healthy lifestyle

In this study, we set two goals: develop and test a method for assessing a person’s motivational profile, based on psychological and content analysis of the text and applicable to motivation for sports and a healthy lifestyle and test automated text analysis programs based on artificial intelligence and compare their performance with each other and with traditional expert content analysis. The selection of programs was made from 65 text analysis programs and 12 artificial intelligence-based chatbots. A total of 12 texts were used for the study, including three long interviews and speeches, and 9 short (2-6 sentences) posts on social media and online media. The evaluated texts contain 5,787 words, of which 91% are originally in Chinese and 9% in English. The content analysis methodology was substantiated in detail in our previous publications.

Designing meaningful surveys: Engaging experts-by-lived experience in the development of a patient-reported experience measure (PREM) of integrated home and community care.

Background: People want to live well in their homes and communities for as long as possible. To support this, some home care services in Ontario, Canada are restructuring to integrate medical and personal care, with community-based social care and services (e.g., friendly visiting, meals, and transportation). Measures of client experience, referred to as ‘patient-reported experience measures’ (PREMs) are important for guiding health system improvements in a way that is meaningful to clients. A current challenge is that existing PREMS are insufficient for measuring client experience of new models of home and community care.&#x0D; Aims: This study engaged clients, family/friend caregivers, and health and social care providers in a participatory process to develop relevant items for a PREM for integrated home and community care. This engagement process aimed to provide opportunities for client feedback to support integrated care and ensure the new PREM will collect data that can inform health system improvements in ways that are most meaningful to clients and caregivers.&#x0D; Methods: Guided by Streiner et al.’s (2015) approach, we engaged experts-by-lived experience in content and face validity testing of an item pool matrix to determine relevance and coverage of domains and items. The matrix consisting of 3 domains, 14 categories, and 72 items was based on a literature review of home and community PREMs (32 PREMs with 550+ items) and interviews with healthcare leader experts (n=6). To conduct content and face validity testing we held three focus groups with home care recipients and family/friend caregivers (n=17) as well as individual interviews with health and social care providers working in home care (n=15). We analyzed transcripts line-by-line to generate themes related to face and content validity of each item. &#x0D; Results: Participants agreed that client experience of innovative home and community care is well captured by three domains: equity, continuity, and life care. They also agreed the corresponding 72 proposed items had face validity. Suggestions to improve content validity included that items be adapted to recognize the role of caregivers; the role of primary providers and/or coordinators in delivering well organized care; and shifting from a focus on self-management to having needed supports to stay in the home, and collaboratively developing care plans. Participants excluded several items due to being vague or not meaningful, such as asking, “my providers understood my needs”. Based on feedback, 22 items were removed from the matrix.&#x0D; Learnings: Engaging experts-by-lived experience in development of PREMs helps ensure the data generated from PREMs will be meaningful and useful for guiding health system improvements that matter to clients. Relying on existing literature and traditional expert opinion alone did not guarantee relevance or coverage of items. &#x0D; Next steps: The feedback from participants will be used to further refine and scale the PREM items. Clients of home care and family/friend caregivers will then be re-engaged in cognitive testing to identify issues related to answering the questions on the scale. Once complete, the PREM will undergo psychometric testing.&#x0D;

Research on Performance Evaluation Index System and Assessment Methods for Microgrid Operation in the Port Area

The vertical integration of ‘Generation-Grid-Load-Storage’ in microgrids for port areas is a prevailing trend. To comprehensively and accurately assess the operational efficiency of microgrids and develop an effective means for promoting the sustainable and scalable development of microgrids in port areas, an applicable evaluation index system and comprehensive evaluation method are essential. Addressing the issues of subjectivity in quantitative evaluation decision-making for different configuration and allocation schemes in port areas and the lack of a comprehensive evaluation system for microgrids in port areas, this study aims to scientifically and reasonably evaluate and select the most optimal configuration and energy dispatching schemes for microgrids in port areas. In this paper, the performance indicators of microgrids in port areas are hierarchically structured and classified into five dimensions: economic, energy efficiency, environmental, system reliability, and safety. A comprehensive evaluation index system for microgrid systems in port areas is constructed. Furthermore, an evaluation model for microgrid operational efficiency based on improved CRITIC-TOPSIS is proposed. The coefficient of variation is introduced to measure the relative strength and relative variation of indicators among dimensions, reducing the impact of correlation between indicators and improving the accuracy and objectivity of the evaluation results. Then, the TOPSIS method is used to calculate the comprehensive evaluation values and rank the optimal microgrid schemes in port areas. Finally, the feasibility of the evaluation index system and evaluation method for the operational efficiency of microgrids in port areas is verified through case analysis. The results indicate that the evaluation method based on improved CRITIC-TOPSIS can objectively and quantitatively evaluate the operational efficiency of microgrids in different port areas. The proposed method reasonably avoids the strong subjectivity in weight calculation by traditional expert judgment during the planning phase of microgrids in port areas, providing a new and scientifically effective engineering evaluation and analysis method for the evaluation of microgrid operational efficiency in port areas.

Knowledge Base Question Answering via Semantic Analysis

Knowledge Question Answering is one of the important research directions in the field of robot intelligence. It is mainly based on background knowledge to analyze users’ questions and generate answers. It is one of the important application methods of knowledge graph technology. Compared with the traditional expert system of question and answer, it has the advantage of a large-scale background knowledge base and the traceability and interpretability of the question-answering process. Compared with the current ChatGPT (Chat Generative Pre-trained Transformer) technology, it has advantages in the proprietary segmentation field. Aiming at the problem of the accuracy of existing knowledge question-answering methods being low, this paper studies the method of semantic analysis for knowledge question-answering under the support of a knowledge database, proposes a knowledge question-answering method based on the superposition of multiple neural network models, and conducts experimental verification on the publicly available NLPCC2016KBQA(Knowledge Q&amp;A Tasks in the 2016 Natural Language Processing and Chinese Computing Conference) data set. The experimental results show that the F1 value of this method is higher than that of the baseline model.

Differing perceptions of risk and the implications for the delivery of harm prevention programmes in Cambodia

ABSTRACT In 2005, Cambodia implemented harm prevention as a guiding principle in their National Strategic approach. In this practice-based research, social and community work practitioners at a community-based non-government organisation (NGO) delivered harm prevention workshops to 18 villages in rural Cambodia. Findings indicated that traditional expert understanding of risk did not always match how communities perceived risk. By engaging with the community utilising a Participation Action Research approach, Children’s Future International (CFI) found their expert position shifted to one where the communities became the experts, empowering them to be actively engaged in risk reduction strategies. These findings have implications for community engagement, development and targeting of harm prevention workshops.

Predict the thyroid abnormality particular disease likelihood of the symptoms’ certainty factor value and its confidence level: A regression model analysis

The traditional expert system (TES) in the medical field commonly uses a certainty factor (CF) rule-based algorithm that can be calculated several symptoms to determine the inference solutions. The main issue for this TES included a prediction for some particular disease likelihood in the cases of new patients. CF is calculated based on symptoms related to clinical signs in patients’ diagnoses. For some reason, this TES probably won’t predict uncertain things, such as particular disease likelihood of some diseases. So, supervised learning, such as linear regression, can solve this problem. We tried to analyse the existing TES for thyroid disorders due to modelling the regression equation to predict the thyroid abnormality particular disease likelihood, based on the symptoms’ CF value and its confidence level. We used multiple linear regression (MLR) and multiple polynomial regression (MPR) to analyse the best regression model to solve the problem. The results show that the MPR model indicates the best regression model for predicting particular disease likelihood of thyroid abnormality, supported by R-squared 94.7%, R-squared adjusted 94.4%, F-value 265.925, and p-value &lt; 0.05, which are higher than MLR model. Our study proposed a foundation for expert system development by focusing more on machine learning expert system (MLES) analysis approaches than TES.

Global maximum flow based time-domain simulation method for evaluating power system resilience

With climate change, extreme weather is occurring frequently, which can cause serious damage to energy supply systems. It is of great significance to study the performance of power transmission system under extreme weather for prevention and reinforcement before and after disasters. This paper proposes a time-domain simulation model for quantifying power system resilience under extreme typhoon scenario. Firstly, the power system resilience assessment method based on global maximum flow (GMF) is presented, which is applied to map the overall system operation performance via creating a network indicator while considering component fragility, system topology, and power flow. A specific resilience expression (RES) is proposed to describe percentage improvement of each component in the resilience metrics by considering the overall system performance. Secondly, power system vulnerability model and time-domain simulation methodology are established. The proposed method is demonstrated by four case studies using the IEEE-30 node test system under specific typhoon scenarios. From the GMF based RES index, the component repair strategy is formulated accordingly. The proposed restoration strategy, compared with the traditional expert strategy can effectively measure the resilience contribution of each component to the whole power system, and minimizes the power system performance losses due to extreme typhoon. The proposed approach quantitatively evaluates real-time system resilience to extreme typhoons, and it can also identify the system critical areas and assess the benefits of different resilience enhancement strategies. The resilience assessment framework can also be applied to other extreme scenarios and contribute to planning and construction of strong smart grid in the future.

Active Learning Guided Drug Design Lead Optimization Based on Relative Binding Free Energy Modeling.

In silico identification of potent protein inhibitors commonly requires prediction of a ligand binding free energy (BFE). Thermodynamics integration (TI) based on molecular dynamics (MD) simulations is a BFE calculation method capable of acquiring accurate BFE, but it is computationally expensive and time-consuming. In this work, we have developed an efficient automated workflow for identifying compounds with the lowest BFE among thousands of congeneric ligands, which requires only hundreds of TI calculations. Automated machine learning (AutoML) orchestrated by active learning (AL) in an AL-AutoML workflow allows unbiased and efficient search for a small set of best-performing molecules. We have applied this workflow to select inhibitors of the SARS-CoV-2 papain-like protease and were able to find 133 compounds with improved binding affinity, including 16 compounds with better than 100-fold binding affinity improvement. We obtained a hit rate that outperforms that expected of traditional expert medicinal chemist-guided campaigns. Thus, we demonstrate that the combination of AL and AutoML with free energy simulations provides at least 20× speedup relative to the naïve brute force approaches.