Developed Model Research Articles

DEVELOPMENT AND SIMULATION OF SPIKE NEURAL NETWORK ARCHITECTURE

A review of different models of spike neural networks has been conducted. A spike type neural network is developed in the article. The mathematical model of exchange of neurotransmitters between neurons is proposed. An algorithm for network operation and neuron interaction is proposed, as well as an approach to training the network based on the formation of new connections from dendrites and increasing the signal transmission coefficient. The model of neural network was created in Python using the developed algorithm and mathematical model. To speed up the calculations, they were paralleled using the Numba library. The library Matplotlib was used for visual modelling and plotting the number of neurotransmitters on neurons. Experimental studies of the developed model of biological type network were carried out. It was shown that the developed network after training responds faster to the signals that were applied to it in the process of training.

Dual-encoder architecture for metal artifact reduction for kV-cone-beam CT images in head and neck cancer radiotherapy.

During a radiotherapy(RT) course, geometrical variations of target volumes, organs at risk, weight changes (loss/gain), tumor regression and/or progression can significantly affect the treatment outcome. Adaptive RT has become the effective methods along with technical advancements in imaging modalities including cone-beam computed tomography (CBCT). Planning CT (pCT) can be modified via deformable image registration (DIR), which is applied to the pair of pCT and CBCT. However, the artifact existed in both pCT and CBCT is a vulnerable factor in DIR. The dose calculation on CBCT is also suggested. Missing information due to the artifacts hinders the accurate dose calculation on CBCT. In this study, we aim to develop a deep learning-based metal artifact reduction(MAR) model to reduce the metal artifacts in CBCT for head and neck cancer RT. To train the proposed MAR model, we synthesized the kV-CBCT images including metallic implants, with and without metal artifacts (simulated image data pairs) through sinogram image handling process. We propose the deep learning architecture which focuses on both artifact removal and reconstruction of anatomic structure using a dual-encoder architecture. We designed four single-encoder models and three dual-encoder models based on UNet (for an artifact removal) and FusionNet (for a tissue restoration). Each single-encoder model contains either UNet or FusionNet, while the dual-encoder models have both UNet and FusionNet architectures. In the dual-encoder models, we implemented different feature fusion methods, including simple addition, spatial attention, and spatial/channel wise attention. Among the models, a dual-encoder model with spatial/channel wise attention showed the highest scores in terms of peak signal-to-noise ratio, mean squared error, structural similarity index, and Pearson correlation coefficient. CBCT images from 34 head and neck cancer patients were used to test the developed models. The dual-encoder model with spatial/channel wise attention showed the best results in terms of artifact index. By using the proposed model to CBCT, one can achieve more accurate synthetic pCT for head and neck patients as well as better tissue recognition and structure delineation for CBCT image itself.

Adjustable compression wraps: back to the future

Adjustable Compression Wraps (ACWs) are used to treat a variety of conditions, in particular venous leg ulcers, edema and lymphedema. Publications on the use of these devices are ultimately few and far between, with limitations or biases (small sample sizes, methodological weaknesses, non-comparative studies, lack of interface pressures). A development model like that used in the study of a drug could be more rational. We propose five phases: phase 1) discovery and development, phase 2) in vitro testing, phase 3) clinical trials, phase 4) real-life data analysis, medico-economic investigation using Artificial Intelligence (AI), and phase 5) preparation of regulatory dossiers. This lack of coherence in development could be an obstacle to widespread clinical use. Using AI to analyse large quantities of data could reduce the cost and number of clinical trials needed to generate more robust evidence. This would make it possible to better define the clinical effects of ACWs in specific situations (e.g., leg wounds in the elderly, diabetic wounds, mixed leg ulcers, secondary lymphedema of the upper limb, efficacy, and comfort...) and to compare treatment costs between bandages and ACWs. If we don't explore the future with AI, we'll never know what will be useful for the present - hence the title “Back to the Future”.

Land degradation vulnerability modelling for landscape restoration planning in Eastern DR Congo using the analytical hierarchy process (AHP) and geospatial techniques

Land degradation is a major issue for attainment of sustainable development in Eastern DR Congo. This study aims to assess and model the spatial pattern of land degradation vulnerability (LDV) in this region through the application of the Analytical Hierarchy Process (AHP) and geospatial techniques by using the Kalehe territory as a case study. LDV has been defined as the resultant of three components: the exposure to erosion risk, the sensitivity associated with biophysical factors, and the adaptability associated with socioeconomic factors of vulnerability. These factors were weighted using the EasyAHP plugin and multisource open spatial data to obtain the LDV model. This model was validated using the frequency ratio of physical land degradation features. The LDV map was obtained through the application of the weighted overlay technique in QGIS. The results indicated that about 32% of the territory is prone to high to very high vulnerability, representing the hotspot of LDV. The developed model has an overall accuracy of 77.8% in predicting the area with high to very high LDV. Thus, it can be used during land conservation planning to identify the priority areas for implementation of landscape restoration initiatives at the territorial level in Eastern DR Congo.

TRUTH OR DARE GAME MEDIA DEVELOPMENT TO PROVIDE AN UNDERSTANDING OF FAMILY FUNCTION

This research aims to develop the truth or dare game media to provide an understanding of family functions. The method used is Research and Development (R&D), using the Analysis, Design, Development, Implementation, Evaluation (ADDIE) development model. Data collection techniques by conducting expert feasibility tests on products that have been developed. The data processing technique uses a Likert scale in the form of quantitative and qualitative data. Feasibility test results with experts in the field, media experts 82.5%, material experts 93.3%, language experts 88.6%. The research subjects were 10 students from the Class of 2021 Department of which means it is very suitable for application and the truth or dare game media can increase students' understanding of family functions.

Study on the synergistic evolution of environment and industry-a case study of provinces along the yellow river basin

The ecological protection and high-quality development of the provinces along the Yellow River Basin have become a major national strategy in China. However, the synergistic development of the environment and industry within the basin still faces numerous challenges. Research studies on the synergistic evolution of industry with the environment are important to guide future development of the provinces. This paper combines panel data from eight provinces in the Yellow River Basin from 2011 to 2020, utilizes the Entropy Weight TOPSIS method to evaluate the growth index of subsystems, and then employs the L-V model to measure the interaction force and synergistic degree of the environmental and industrial subsystems in the provinces along the Yellow River Basin. Finally, it simulates the synergistic evolution paths of each province under different scenarios. The study shows that: 1. From 2011 to 2020, the synergy effects in the upper and middle reaches of the Yellow River Basin exhibited an upward trend, while the synergy effects in the lower reaches demonstrated a downward trend. In the upper reaches, the shift was from mild non-synergy to mild synergy, in the middle reaches from mild non-synergy to strong synergy, and in the lower reaches from strong synergy to strong non-synergy. 2. The development model in the upper and middle reaches transitioned from a weakly beneficial relationship between industry and the environment to a mutually beneficial symbiotic model. In contrast, the lower reaches shifted from a mutually beneficial symbiotic model to a state of weak industrial benefits and strong environmental harm, ultimately reaching a state where both industry and the environment mutually harmed each other. 3. According to the forecast of industrial and environmental synergy for the next 30 years, Qinghai, Inner Mongolia, Ningxia, and Shanxi provinces would benefit the most from adopting a mutually beneficial symbiotic model. However, as the development of a symbiotic model requires more external support and incurs higher costs, it is essential for each province to select its development model based on its own economic conditions. Based on this, this paper provides a theoretical basis for the development of the environment and industry in the provinces along the Yellow River Basin.

Integration of cost-effective datasets to improve predictability of strategic noise mapping in transport corridors in Delhi city, India.

Assessing exposure to environmental noise levels at transport corridors remains complex in conditions where no standardized noise prediction model is available. In planning and policy implementation for noise control, noise mapping is an important step. In the present study, land use regression model has been developed to predict the environmental noise levels in Delhi city, India, using previously developed approaches along with machine learning techniques, however improved using new datasets. Lday, Lnight, LAeq,24h, and Ldn were modeled at daily resolution by utilizing an annual noise levels dataset from 31 locations in Delhi city. The noise-monitored data was integrated with travel time data, nighttime light data along with common parameters including land use, road networks, and meteorological parameters. The developed LUR models showed good fit with R2 of 0.72 for Lday, 0.55 for Lnight, 0.71 for LAeq,24h, and 0.61 for Ldn, which was further improved up to 0.88 for Lday, 0.79 for Lnight, 0.86 for LAeq,24h, and 0.81 for Ldn by integrating machine learning approaches. The developed models were validated through tenfold cross validation and by comparison to a separate noise level dataset. The average travel time variable was observed to be the most influential predictor of LUR models for Lday and LAeq,24h, which signifies the crucial impact of road traffic congestion on environmental noise levels. The study also analyzed the parametric sensitivity of various infrastructural factors reported in the study, which shall be helpful for planning for smart cities.

Planning and Optimisation of Renewable Energy Systems for Decarbonising Operations of Oil Refineries

AbstractGiven the urgency to transition to low carbon future, oil refineries need to identify feasible strategies for decarbonisation. One way to address this is by integrating renewable energy systems. However, the high initial costs and intermittency appeared to be the key barriers for the adoption of renewable energy technologies. Hence, a multi-period optimisation model is developed via mixed integer linear programming in this work to determine the optimal renewable energy system in terms of cost and its optimal energy storage technology to enhance its flexibility for oil refinery operations. This model aims to minimise the costs of the renewable energy system while considering its ability to accommodate the varying energy demands across the time periods. An oil refinery case study is used to demonstrate the effectiveness of the developed model. The developed model is expected to propose an optimal renewable energy system that meets the energy demands and, at the same time, achieves the decarbonisation goal. Based on the results, the optimal renewable energy system comprises cost-effective technologies to generate various energy outputs including electricity, hydrogen, high-pressure and medium-pressure steam to meet energy demands. Additionally, the result of the case study shows that the integration of renewable energy systems achieves a reduction of 5,353 tonnes of carbon dioxide. Apart from that, the incorporation of energy-efficient energy storage results in a 10% reduction in the total cost of the optimal renewable energy system. Compressed hydrogen gas storage and battery were used to store excess hydrogen and electricity during periods with low demands and subsequently consumed during peak demand periods. This can, therefore, reduce the technological capacity required. With the aid of storage facilities, the flexibility of the renewable energy system is elevated in meeting varied demands, which otherwise would incur additional expenses.

High-resolution cyclic framework for the Songliao Basin in northeastern China, and its implications for sedimentation and organic matter enrichment

The study of fine-grained sedimentation has consistently concentrated on investigating the mechanisms and principles governing the enrichment of organic matter. However, the lack of unified stratigraphic framework has always existed as fine-grained sedimentation covers two distinct grain-size grades, namely, mud and silt, which has impeded the progress of subsequent production research. This study exemplified this issue by analyzing the first member of the Qingshankou Formation in the southern Songliao Basin. We established reconstructed gamma and density curves that mitigated filter noise interference, integrated high-resolution sequence results with astronomical cycle divisions, and created a high-frequency isochronous stratigraphic framework for clastic fine-grained sedimentation by leveraging the weak sensitivity of sandstone density curves and the robust stability in eccentricity cycle extraction. This approach addresses the inconsistencies in stratigraphic division methodologies and mismatched outcomes stemming from the use varying techniques to delineate mud and silt components within clastic fine-grained sedimentary sequences. Furthermore, it elucidates how tectonic-scale variations in sediment supply coupled with potential accommodation changes dictate macroscopic stacking patterns within strata, whereas climate fluctuations on orbital time scales govern sand-mud progradation degrees within these layers, culminating in periodic rhythmic characteristics characterized by vertical sand-mud interbedding. A model for stratigraphic development pertaining to lake delta systems constrained by a “synchronous heterotopy” paradigm is proposed for the southern Songliao Basin. The organic matter enrichment pattern aligns with its filling dynamics, indicating an “overfilling” type developmental pattern at lower strata levels where organic material predominantly originates from terrestrial plant debris external to the basin; this material accumulates primarily within silty zones along layers—with areas exhibiting heightened enrichment values slightly lagging behind short-eccentric maxima positions. In contrast, under an upper “balanced filling” type developmental framework, sources of organic matter are derived both internally and externally relative to the basin—exhibiting substantial heterogeneity—and regions marked by elevated organic matter concentrations are directly associated with locations identified as short-eccentric maxima.

Addressing the Sense of School Belonging Among All Students? A Systematic Literature Review

The sense of school belonging plays an important role in students’ academic, behavioural, and psychological outcomes. Based on a systematic review, following the PRISMA 2020 guidelines and examining 86 studies conducted between 1990 and February 2023, the article addresses two research questions: (a) what are the predictors of the sense of school belonging at the individual, micro, meso, exo, macro, and chrono levels of the bioecological model of human development; (b) do these predictors differ based on students’ individual characteristics, and if so, how. The findings reveal individual factors as important predictors of school belonging and indicate the lack of studies that take into consideration the interplay of different (micro, meso, exo, macro, chrono) levels in addressing the sense of school belonging. Considering the complexity and multi-factorial nature of the sense of school belonging, it calls upon further research, which would support the development of evidence-based interventions for fostering school belonging among different groups of students, particularly those who are at risk of feeling alienated from school, and thus promote equity in education.

Modeling and Measurement of Tool Wear During Angular Positioning of a Round Cutting Insert of a Toroidal Milling Tool for Multi-Axis Milling

The aim of this study was to develop a concept for an angular positioning method for a round cutting insert in a torus cutter body dedicated to the multi-axis milling process under high-speed machining cutting conditions. The method concept is based on a developed wear model using a non-linear estimation method adopting a quasi-linear function. In addition, a tool life model was developed, taking into account the cutting blade work angle parameter, the laser marking method for the round cutting insert, and a wear measurement methodology. The developed tool wear model provides an accuracy of 90% in predicting the flank wear of the cutting blade. The developed procedure for angular positioning of the round cutting insert enables the entire cutting edge to be fully utilized, extending the total tool life. In addition, the measured largest defect values between the worn cutting edge and the nominal outline of the round cutting insert indicate the location of notching-type wear.

Research on Ecological Restoration Strategies for Abandoned Mines Based on Ecology-Oriented Development

The ecological restoration of abandoned mines poses a significant challenge within China’s broader efforts toward the protection and restoration of the nation’s ecosystems. Based on the ecology-oriented development (EOD) model, a differential game framework for the ecological restoration of abandoned mines was constructed in this study, involving both government and corporate entities as players. Our analysis focused on various factors, including investment levels in ecological restoration quality, the quality of ecological restoration itself, investment levels in environmentally sensitive industry development, the quality of environmentally sensitive industries, environmentally sensitive product pricing strategies, and profit dynamics for both players across four distinct modes. Our findings reveal several key insights. First, extending the government’s concession period incentivizes enterprises to increase investment in both the quality of ecological restoration of abandoned mines and the development of environmentally sensitive industries. Second, the government should prefer the cooperative mode to introduce social capital for the ecology-oriented development of abandoned mines. Third, in noncooperative mode, for technology-intensive industries, the government’s industrial support policy is more mutually beneficial; conversely, for labor-intensive industries, the ecological restoration compensation policy yields higher mutual gains. Lastly, differentiated incentive policies implemented by the government across various modes prompt the input of enterprises in ecological restoration quality and environmentally sensitive industry development. Based on these findings, we offer specific policy recommendations to guide governmental efforts in promoting EOD for the ecological restoration of abandoned mines. Funding: This work was supported by the Sichuan Office of Philosophy and Social Science [Grant SC22C004] and the National Social Science Fund of China [Grant 22XKS015].

A class of mixed thinning threshold integer-valued autoregressive model for the COVID-19 data

To better capture the features of time series of counts such as the counting of elements of variable character and piecewise phenomenon, this paper proposes a class of mixed-thinning threshold integer-valued autoregressive model where the distribution of the innovation sequence is unknown. Strict stationarity ergodicity property is proved. The unknown parameters in the model are estimated by conditional least squares and modified quasi-likelihood methods, as well as their asymptotic normality, are established for the case that the threshold variable is known. The performances of the estimators are investigated in simulation, which manifests that the modified quasi-likelihood estimation performs better. Finally, a set of COVID-19 data on critical cases imported from outside China is analyzed to illustrate the application of developed model.

The Development of Machine Learning Models for Radial Compressor Monitoring With Instability Detection

Abstract Modern radial compressors are designed to operate with high performance and a wide range of applications while minimizing their environmental impact. It is necessary to study the machine near the stability limit and gain a comprehensive understanding of the fluid dynamic mechanisms that trigger the instability in the system to extend the operating range at high efficiency. Machine learning aids in developing pattern identification models for detecting compressor instability. In a prior study, a two-stage radial compressor for refrigerant gas underwent extensive simulation, capturing unsteady RANS conditions and generating a substantial dataset of pressure signals from multiple probes. Selected signals, coupled with detailed computational fluid dynamics (CFD) post-processing, revealed fluid dynamic structures near surge conditions. This study utilizes all pressure signals to assess the stage's operational state (stable, transient, or unstable/surge). An additional goal is to develop an algorithm predicting flow patterns in different stage sections with minimal pressure signals at the rotor inlet. This is a preliminary step toward the development of a smart monitoring and diagnostic model for the compressor installed in a plant. The developed model consists of three submodels, trained with CFD results obtained from unsteady Reynolds averaged Navier–Stokes (URANS) simulations. The three submodels are a regression submodule, a classification submodule, and a forecasting algorithm. The regression submodule predicts diffuser static pressure fields, the classification submodule categorizes whether the compressor is in a critical condition or not, and the forecasting algorithm predicts the inducer pressure signals in the future impeller rotations according to the actual operation history. These sub-modules work together to forecast whether the compressor, according to the operation strategy, is going into instability conditions.

Lembar Kerja Peserta Didik Berdasarkan Model Pembelajaran Generatif untuk Materi Gelombang Cahaya Kelas XI SMA

This research aims to develop LKPD based on a generative learning model on light wave material for class XI SMA that is valid and practical. The generative learning model was chosen because it is able to encourage students to actively build their understanding through interaction, reflection and discussion. This model has proven effective in improving critical thinking and problem-solving skills in physics learning. The developed worksheet is designed following the syntax of the generative learning model: orientation, cognitive conflict, disclosure, construct, application, and reflection, to support in-depth learning and link theory with real application. This research uses the Research and Development (R&D) method with the 4D development model (define, design, develop, disseminate). The define stage includes needs analysis and problem identification. At the design stage, the LKPD was designed by taking into account the content, structure, and aesthetic components. The development stage involves validity testing by experts, well as practicality testing by teachers and students. The dissemination stage is planned for wider application. The results showed that this LKPD has a very high level of validity with a validation index value of 0.94 in the content aspect and 0.86 in the structure aspect, which means that this LKPD meets the eligibility standards of the Merdeka curriculum. The practicality test shows that the LKPD is very practical with a teacher assessment score of 95% and a student assessment of 87%, so it is feasible to use it in the real field because it has been declared valid and practical.

Investigation on novel ultralight weight thermally insulated fireproof composite

Abstract This study highlights the performance of ultra-lightweight fireproof composite. Polyacrylonitrile based carbon fiber (CF) has been reinforced in Polyetherimide (PEI) polymer to develop the composite. The Surfac2e of CF and PEI film was modified by low-pressure plasma to improve the bonding strength between matrix and reinforcement. The polymeric composite was fabricated by compression molding with a pressure of 2 bar, temperature of 380 °C and holding time of 30 min. CF/PEI composite was used to make a hybrid composite by layering of silicone foam in between the layers. The hybrid composite was exposed to a Bunsen burner under sustained flame for a duration of 10 min. The composite panel’s flame-facing side reached 676.2 °C after 10 min of fire exposure, while the temperature on the other side only reached 58.2 °C. The fabricated hybrid composite was exposed to very low temperature in order to test its ability of thermal insulation under extreme cold temperature. Over the specific period of testing, the temperature of the dry ice decreased from 25 °C to −3.1 °C. After exposure to fire, only minimal loss of material was observed. The hybrid composite of carbon fiber and PEK film, sandwiched between silicone foam, exhibits excellent fire resistance due to its high limiting oxygen index. This composite is considered to be among the best thermally insulating and fire-resistant materials. Thermogravimetric analysis of carbon fiber and PEI-Carbon fiber composite was performed to determine the optimal processing temperature of compression molding for the composite, upon heating, it showed a modest weight decrease of 6.053%. The composite shows a significant improvement of impact resistance, compressive strength and thermal stability. A simulation model was developed under Ansys fluent software for both heating and cooling. The analysis of developed model also shows similar results.

The econophysical model of innovation diffusion

Analysis and evaluation of innovation efficiency require the development of tools to model their dissemination process within the industry. This paper presents a model of innovation diffusion based on physical approaches, describing stages of accelerating and decelerating growth. An exponential growth is described using a diffusion model, while a logarithmic one employs an electrical engineering model. The paper presents the correspondence of physical parameters with their economic counterpart: size of a company; characteristic of speed of information exchange between firms; company’s willingness to innovate; inter-firm influence and the breakthrough level of innovation. The theoretical model obtained was tested on historical data of innovation implementation in the fuel and energy complex, followed by adjustments of coefficients depending on the branch of innovation implementation. The developed model is applicable for describing the process of innovation dissemination in any industry in the country, as well as for investment and business planning in companies and decision-making on investments in innovation projects. When applied in industries with low levels of innovation activity, an increase in the level of high-tech production and the share of organizations implementing technological innovations is predicted. Using the example of Russia’s fuel and energy sector, rising in the technological level of enterprises and a decrease in import dependence are forecasted.

Heuristic strategy using hybrid deep learning with transfer learning for oral cancer detection

Oral cancer becomes the most disastrous ailment that affects the oral cavity parts of the mouth. Oral cancer diagnosis is the main challenge in the medical field. It becomes expensive and less capable of classifying oral cancer. In some cases, it may cause unnecessary morbidity and mortality. Recently, the detection of malignant and premalignant oral lesions has been a critical process owing to their low image resolution and lower acquisition time. Thus, a novel hybrid deep learning with meta-heuristic-based optimization is proposed. The pre-processing occurs by median filtering and Contrast Limited Adaptive Histogram Equalization (CLAHE). The CLAHE method is utilized to reduce unwanted noise. Finally, the classification is done by a proposed hybrid-based deep learning model termed as Recurrent Deep Belief Network (RDBN), in which the Deep Belief Network (DBN) is incorporated with the Recurrent Neural Network (RNN). Here, the RDBN helps to increase the performance classification. Furthermore, the hyperparameters of the RDBN model, such as learning rate, epochs and hidden neurons, are tuned using the proposed Hybrid Beetle-Barnacle Swarm Optimization (HBBSO) algorithm, where the Barnacles Mating Optimizer (BMO) is superimposed with Beetle Swarm Optimization (BSO) algorithms. In the given proposed model, the selected features are extracted by the optimization algorithms. Here, the parameters are fine-tuned to get the better optimal solution. From the experimental outcome, the developed model has acquired 5.2% better than PSO-RDBN, 5.6% improved than GWO-RDBN, 3.2% enhanced than BSO-RDBN and 3.5% superior to BMO-RDBN regarding accuracy. Thus, the proposed model achieves higher results for detecting oral cancer with enhanced classification performance than the existing approaches.

NEED ANALYSIS FOR PROBLEM BASED LEARNING E-COMIC ON EARTH AND SOLAR SYSTEM TOPIC

The aim of this study was to analyse the need for the development of science e-comic as learning media on earth and solar system topic. This research was conducted in 3 junior high schools in Sutera sub-district, based on the results of observations it is known that the school applies the kurikulum merdeka, and uses lecture and discussion methods in the learning process. However, there are obstacles that must be faced by teachers, namely the current package books/printed books are quite limited, so students have enough difficulty if they want to repeat learning material at home. This type of research is Research and Development (R&D) using the 4-D development model. This research instrument consists of a validation questionnaire, a practicality questionnaire, and an effectiveness questionnaire. The data obtained were analyzed using a formula to analyze the results of the validity test, practicality test, and effectiveness test. Based on the above problems, researchers found the right and appropriate media to be developed and can be used easily by students and teachers, namely problem-based science e-comic learning media on earth and solar system material.

Machine-learning-aided biochar production from aquatic biomass

AbstractModeling hydrothermal carbonization (HTC) and pyrolysis carbonization (PLC) for the conversion of biomass into high-quality biochar for various applications shows promise. Unlike the extensive modeling studies on lignocellulosic biomass, research on aquatic biomass (AB) had not been reported until now. In this study, we compiled 586 data points from existing literature and trained five tree-based models to predict the yields of hydrochar and pyrochar and their properties, including nitrogen recovery degree, energy density, energy recovery degree, and residual sulfur degree, based on 10 feedstock and process parameters. The random forest regression (RFR) model demonstrated the highest predictive accuracy among these models. It achieved R2 values ranging from 0.89 to 0.98 for hydrochar yield, nitrogen recovery degree of hydrochar, energy recovery degree of hydrochar, and residual sulfur degree of hydrochar. The extreme gradient boosting (XGB) model also showed exemplary performance, with R2 values between 0.84 and 0.94 for energy density of hydrochar, pyrochar yield, and nitrogen recovery degree of pyrochar. Results on feature importance highlighted that, beyond the well-documented impact of process parameters, the properties of biochar were significantly influenced by the elemental compositions, such as nitrogen and sulfur contents of the feedstock. The relationship between these factors was further elucidated using partial dependence plots. Finally, we used RFR model for hydrochar yield and XGB model for pyrochar yield as examples, to test generalization ability of developed models with new data, further explaining their application methods. Overall, this study provided valuable insights into predicting and understanding the HTC and PLC processes of AB to produce high-quality biochar for various applications using low resources and time costs. Besides, we presented an iterative learning application method where the developed models demonstrated exceptionally high performance with new data. This method is highly versatile and can be adopted across various directions in the field of machine learning. Graphical Abstract