Calculation Error Research Articles

Memristive Crossbar Array-Based Probabilistic Graph Modeling.

Modern graph datasets with structural complexity and uncertainties due to incomplete information or data variability require advanced modeling techniques beyond conventional graph models. This study introduces a memristive crossbar array (CBA)-based probabilistic graph model (C-PGM) utilizing Cu0.3Te0.7/HfO2/Pt memristors, which exhibit probabilistic switching, self-rectifying, and memory characteristics. C-PGM addresses the complexities and uncertainties inherent in structural graph data across various domains, leveraging the probabilistic nature of memristors. C-PGM relies on the device-to-device variation across multiple memristive CBAs, overcoming the limitations of previous approaches that rely on sequential operations, which are slower and have a reliability concern due to repeated switching. This new approach enables the fast processing and massive implementation of probabilistic units at the expense of chip area. In this study, the hardware-based C-PGM feasibly expresses small-scale probabilistic graphs and shows minimal error in aggregate probability calculations. The probability calculation capabilities of C-PGM are applied to steady-state estimation and the PageRank algorithm, which is implemented on a simulated large-scale C-PGM. The C-PGM-based steady-state estimation and PageRank algorithm demonstrate comparable accuracy to conventional methods while significantly reducing computational costs.

A novel method for measuring electrolytic conductivity with a polarization impedance control function

The article presents an idealized model of the differential method utilized in conductometry to measure electrolytic conductivity (EC). It identifies the main drawback of the widely used differential method: the inability to control the influence of polarization impedance on EC measurement results. A criterion for the absence of error caused by polarization effects is proposed, involving equality between the ratios of resistance differences and cell constant differences. A novel method for EC measurement, termed the double differential method, is introduced. This method relies on repeated differential impedance measurements and incorporates an additional third impedance measurement of a virtual liquid column. The new approach enables the identification of polarization impedance influence on EC measurements and the calculation of measurement errors for correction. A 3 times lower type A measurement uncertainty value can be achieved by averaging the corrected results.

Analisis penerapan akuntansi Pajak Penghasilan Pasal 21 pada karyawan PT. Manado Post

Manado Post is a Media Company under the Jawa Post Group. Manado Post has a large number of employees with the specifications of Permanent Employees and Non-Permanent Employees, so it has great potential in paying taxes, especially Article 21 Income Tax. In addition, there is the potential for errors in both calculating and recording Income Tax. This also affects accounting records, because errors in calculations will cause errors in accounting records. In addition, Manado Post on irregularities in both calculation and reporting of Income Tax caused by differences in views on the Income Tax Law. This also affects accounting records. Based on the reference background of this study, the subject matter can be formulated which will then be the object of discussion in this study. This study used qualitative descriptive method. A total of 55 employees were involved in the study, and of these, 15 employees were the subjects of the study. The qualitative research intended here is to illustrate how the accounting treatment of Article 21 Income Tax applied in Manado Post. The results showed that Manado Post did not fully follow the applicable Tax Law in calculating article 21 income tax on its employees. This causes an overpayment of Article 21 income tax by the company. In this case, Manado Post needs to improve understanding and compliance with applicable tax rules to avoid errors in tax calculations.

ВПЛИВ ПУЛЬСАЦІЙ ІМПУЛЬСНОГО ПЕРЕТВОРЮВАЧА НА ПОХИБКУ СЛІДКУВАННЯ ЗА ШВИДКІСТЮ БОРТОВОГО ЕЛЕКТРОПРИВОДА

An analysis of the tracking error was performed without taking into account the pulsations of the electric motor supply voltage and taking into account the pulsations of the electric motor supply voltage. The results of the calculation of the constant tracking error are presented, taking into account the influence of pulsations, which is a consequence of the pulse width modulation (PWM) process. Ref. 7, fig. 2.

An improved method for evaluating fracture density using pulsed neutron capture logging

Fracture density is a critical fracture evaluation parameter for the optimization and production prediction of hydraulic fracturing models. Nonradioactive tracer (NRT) techniques have successfully used a quantitative fracture density method based on neutron-induced gamma rays from formation elements. Furthermore, fracture density can be calculated using the formation capture cross section before and after hydraulic fracturing based on pulsed neutron capture logging. However, the response sensitivity of the macroscopic scattering cross section to the fractures decreases when the fracture density is high due to the neutron self-shielding phenomenon. Therefore, an improved fracture density evaluation method is applied, which combines the macroscopic capture cross section and the neutron self-shielding correction factor. In the new method, the peak area of titanium from the captured gamma spectrum is used to obtain a neutron self-shielding correction factor to improve the sensitivity of fracture density determination. Furthermore, the response of capture cross-section variation to fracture density at various tagged proppant concentrations and formation backgrounds is investigated. The findings indicate that the tagged proppant concentration influences the detection limit of fracture density and the sensitivity of fracture density identification. The accurate calculation range of fracture density using the new method has been extended from 5% to 10% under the condition that the tagged proppant concentration is 0.2%. Meanwhile, whereas water salinity significantly impacts capture cross-section variation, the effects of porosity, lithology, and fluid type on capture cross-section variation are negligible. A simulated fracturing example demonstrates the method’s applicability in various measurement environments. The results show that fracture density and height are consistent with the model settings after correcting for water salinity, and the fracture density calculation error is less than 3%. Therefore, our evaluation method for fracture density corrected for the neutron self-shielding effect improves response sensitivity and fracture density calculation accuracy.

Anti-tropical cyclone load reduction control of wind turbines based on deep neural network yaw algorithm

Rapid changes in the wind field of tropical cyclones can cause excessive loads and threaten the safety of offshore wind turbines. This paper designs a wind turbine yaw optimization strategy based on the deep neural network to reduce the structural loads of wind turbines caused by tropical cyclones. Firstly, the high-precision tropical cyclone data is used to analyze the characteristics of the wind field. Then, a pseudo-Monte Carlo experiment is designed to compensate for the incompleteness of the observed data. Furthermore, a robust nonlinear coupling model between the wind characteristics and the loads of the wind turbine is constructed by a deep neural network, and the optimal yaw angle is searched in real time based on this model. The simulation results show that the proposed deep neural network model based on pseudo-Monte Carlo scenario generation can robustly calculate the structural loads of wind turbines with a calculation error of less than 8 %. After applying this model to the real-time optimization control loop, the corresponding optimized yaw angle can be obtained according to the operating data of the wind turbine under tropical cyclone conditions so that the structural loads of the wind turbine are reduced. Compared with the no-yaw and traditional yaw strategies, the load suppression effect is 1 %–9 % under different working conditions. The proposed data-driven structural load model and load suppression algorithm will effectively improve the operating safety of wind turbines under tropical cyclone conditions.

Experimental and Numerical Simulation Research on Different Shapes of Flame-Stabilizing Baffles in the Furnace

The use of fully premixed combustion in small gas boilers can improve denitrification efficiency. On the basis of fully premixed combustion, adding flame stabilizers inside the boiler can further reduce production of oxides of nitrogen. Four types of flame-stabilizing baffles were added at a certain position inside the furnace after the fully premixed burner. Experiments were conducted separately on the Noporous baffle, and numerical calculations were performed for 36 operating conditions of the four baffles. The experimental results and numerical calculations indicate that under experimental conditions, NOx emissions were all below 40 mg/m3, and the net heat efficiency of the boiler was above 80%. Under a maximum firing rate, CO emissions are below 20 ppm, and the minimum error between the calculated and experimental values is 2.2%. The calculation error of CO emissions under various working conditions does not exceed 6.8%, indicating that the impact of different-shaped baffles on CO emissions is relatively small. When installing a Nonporous baffle, the error between the experimental and calculated exhaust temperature values under minimum firing rates is 6.6%, the error between the calculated and measured values under middle firing conditions is 2.9%, and the error between the calculated and measured values under maximum firing rate conditions is 3.0%. Among the four different partition conditions, the exhaust temperature of the Nonporous baffle is the lowest. Under the same excess air coefficient, the pressure in the furnace for the middle and maximum firing rate is higher than that of the minimum firing rate, and the experimental values are in good agreement with the calculated values. When installing the Strip baffle, the calculated CO2 emission is the lowest. The experimental results show that the NOx in the flue gas inside the furnace is mainly NO, and the NO content exceeds 75%, reaching a maximum of 85%. The experimental results show that the minimum NOx emission value is 26.9 mg/m3. The error between the measured and calculated NOx values when installing a Nonporous baffle is 20.4%. All of the above indicate that installing a flame-stabilizing baffle at an appropriate position in the furnace can further reduce NOx emissions, and the optimization amplitude is related to the shape of the baffle.

Lateral kinematic properties of offshore pipe piles embedded in saturated soil considering soil plug effect

AbstractThis study establishes a theoretical framework for analyzing the lateral oscillation of marine pipe piles. The additional mass model is introduced herein to consider the inertial fluctuation effect of the soil plug. Analytical mathematical methods are used to determine the complex impedance variation of the pile over a range of frequency effects. An investigation is performed to determine how the presence of soil plugs changes the lateral complex stiffness and natural frequency of pipe piles. Additionally, comparisons of the applicability of the plane strain model and continuous medium model have been conducted to enable the easy use of the theoretical model. The main conclusions can be drawn as (1) if the fluctuation inertia effect of the soil plug is not taken into consideration, the dynamic active length and the dynamic stiffness of the pipe pile will be underestimated; (2) for the soft soil, the plane strain model may give rise to substantial calculation errors attributed to them regardless of the vertical continuity of the soil, nevertheless, the calculation error decreases rapidly with the increase of soil shear modulus and vibration frequency.

Calculation of Porosity and CO2 Content in CO2-Bearing Gas Formations Based on Density and Neutron Logging Forward and Inverse Methods.

When the gas reservoir contains CO2, the logging response tends to be complex. When calculating porosity with neutron and density curves, the accuracy of porosity calculation is reduced due to the uncertainty of fluid parameters, which in turn affects the evaluation of CO2 content. It increases the uncertainty of reserve evaluation. In this paper, a forward modeling model of density logging is established based on the equivalent volume model, and the method of neutron logging is formed by putting forward the nonlinear formula of the reciprocal of the comprehensive deceleration length. The key parameters and their variation rules of neutron logging and density logging forward modeling under different temperature and pressure conditions are obtained by experimental means and Monte Carlo method, respectively. The variation law of the influence of different CO2 content on neutron logging and density logging curves is simulated. The research shows that with the increase of CO2 content, the neutron logging value of gas reservoir decreases, while the density logging value increases, and the greater the porosity of gas reservoir, the more obvious this change trend is. Compared with CH4, CO2 has a more significant impact on the excavation effect of neutron logging. On this basis, combined with the conductivity model of the study area, the porosity and CO2 content of CO2 bearing gas reservoir are solved by the least-square method. The practice shows that the relative error of porosity calculation is within ±4% and the error of CO2 content calculation is within ±10%, which proves the reliability of this method.

Coarse mesh finite difference acceleration of the random ray method

This paper presents the development and evaluation of the Coarse Mesh Finite Difference (CMFD) acceleration method to The Random Ray Method (TRRM). We demonstrate its effectiveness in accelerating the convergence of the fission sources and reducing the real statistical error in neutron transport calculations. TRRM treats the angular variable of the neutron flux stochastically and performs batch sampling of characteristic rays to transform the Method of Characteristics (MOC) into a stochastic process, that is capable of making significant strides in memory efficiency and computational performance for some problems. Despite its advantages, TRRM exhibits a potential challenge with a large number of inactive cycles and inherent inter-cycle correlation much like the Monte Carlo method. To address this, the CMFD acceleration method is explored and demonstrated to dramatically reduce the number of required inactive cycles and diminish inter-cycle correlation. Results from the numerical analysis of a 2D C5G7 core problem indicate that the application of CMFD leads to enhanced convergence, with the integration of a CMFD acceleration step every cycle offering the most substantial reduction in statistical noise and error. The study reveals that applying CMFD with every cycle effectively resolves the issue of needing inactive cycles and significantly lowers the inter-cycle correlation, thereby providing a more accurate estimation of standard deviation for pin power distributions. We conclude that using CMFD not only minimizes the number of inactive cycles of TRRM – much like normal Monte Carlo transport – but also lowers real statistical error effectively. For a targeted maximum standard deviation of 0.1% in the pin power, the addition of CMFD can decrease the number of necessary active cycles by 41% compared to standard TRRM, as demonstrated by the 2D C5G7 benchmark analysis.

Continuous optimization method of micro geometry for unparallel beveloid gears

To enhance the bearing capacity and minimize the loaded transmission error for unparallel beveloid gears, a continuous optimization method of micro geometry is proposed. Considering that the conjugate characteristic of tooth surfaces is destroyed by misalignments and modification, an unload contact analysis model is first developed. This includes a contact tooth sequence determination model and an unloaded transmission error calculation model for multi-tooth contact. To determine contact due to deformation, a potential contact point matching model of non-contact tooth surfaces is introduced. Further, a numerical loaded contact analysis model based on the influence coefficient method for unparallel beveloid gear is developed. For multi-tooth contact of modified tooth surfaces, the transmission error compatibility condition is introduced. Based on the loaded contact model, a continuous optimization model for the comprehensive performance of contact pressures and loaded transmission error within a meshing cycle is established. To improve optimization efficiency, a calculation strategy for continuous optimization is developed. The feasibility of the proposed method is validated using a numerical example.

Investigation on saturated vapor and two-phase refrigerant injection in the rotary compressor for the extreme hot and cold conditions

Vapor injection has been widely applied in the rotary compressor to improve the heating capacity and reduce the discharge temperature in the cold condition. However, under the evaporation temperature below −20 °C or the condensation temperature above 60 °C, the discharge temperature is still relatively high, which limits compressor performance and its continuous operation time. To keep the optimal performance and further reduce discharge temperature, this paper focuses on the two-phase refrigerant injection and proposes a mathematical method to investigate the effect of injection vapor quality on the compressor performance. Based on the proposed method, some performance parameters including cooling capacity, heating capacity, EER(COP), power, discharge mass flow rate and discharge temperature have been calculated and an experiment is conducted to test these performance parameters. The calculation errors between calculated and tested results are within 4% and it indicates that the proposed method is accurate to investigate saturated vapor and two-phase refrigerant injection characteristics. The simulated results show that the use of two-phase refrigerant injection results in the trade of a slight drop in cooling capacity, heating capacity and EER(COP) for a large drop in discharge temperature, which is valuable and meaningful for engineering application under the extreme hot and cold conditions.

Local heat transfer measurements of the fin on the blade crown by infrared thermography

Current research on heat transfer of the labyrinth surface mainly focuses on the whole structure made of the same material, but the heat conduction from the tip to the root will affect the measurement of the heat transfer coefficient. To minimize the influence of heat conduction, copper strips are embedded on the surface of the fin made of bakelite. The inverse heat transfer problem is solved using the temperatures collected by an infrared thermal imager to determine the local convective heat transfer coefficient. The error of the lumped parameter method is smaller compared to the differential solution. Constant heat flux experiments are used to validate the experimental results, and numerical analysis is performed to analyze the heat transfer mechanism. The flow field, the Nu number distribution, and fitting correlations of the fins are obtained. By comparing and analyzing the heat transfer coefficients obtained by the three methods, it is found that the lumped parameter method can eliminate the influence of heat conduction on the surface. Taking the lumped parameter method as a reference, the errors of stable heat flow and numerical calculation were 21 % and 4.7 %, respectively. Due to the influence of the vortex direction, the Nu number distribution along the height of the windward surface of the second fin is different from the others. The Nu number distribution along the x-direction on the top surface of the fin is similar to the tip of the windward surface. Along the flow direction, the average Nu number on the top surface of the first fin is 1.2 times that of the second one.

A Study on the Propulsion Performance of Hybrid-Driven Underwater Glider Equipped with a Kappel Tip Rake Propeller

In order to solve the problem of the lack of maneuverability of the conventional underwater glider, this paper proposes a hybrid-driven underwater glider equipped with a Kappel tip rake propeller, analyzes the propulsion performance of different types of Kappel tip rake propellers in the wake field of the hybrid-driven underwater glider, optimizes the overall propulsion performance of the hybrid-driven underwater glider, and realizes self-propulsion and gliding with high efficiency and low energy consumption. In the research process, the Schnerr–Sauer cavitation model and the cavitation simulation strategy of VOF two-phase flow were adopted, coupled with the SST k-ω and γ transition turbulence model, and the control calculation error was not more than 3%. Based on the hydrodynamic performance study of the Kappel tip rake propeller, the self-propelled simulation was carried out under the working conditions of 6 kn, 5 kn, 4 kn, and 3 kn, and the gliding simulation was carried out under the working conditions of 1 kn, 0.5 kn, and a glide angle of 12°. The propulsion performance of the hybrid-driven underwater glider with different models of Kappel tip rake propellers was analyzed. It was found that the maximum open water propulsion efficiency of the propeller Kap05 had the largest improvement, which was 3.07% higher than that of the reference base propeller. Under the self-propelled condition, the hybrid-driven underwater glider with the propeller Kap05 had the lowest wake fraction, and the propellers Kap04 and Kap05 had the best propulsion performance in the wake field of the hybrid-driven underwater glider. In the gliding condition, the form of the folding paddle can reduce the gliding resistance generated by the propeller by more than 45% and the gliding negative lift by more than 68%. A moderate tip rake can effectively improve the propulsion efficiency of the Kappel tip rake propeller in the wake field of the hybrid-driven underwater glider, reduce energy loss, and improve the overall performance of the hybrid-driven underwater glider.

Efficient implementation of equivalent medium parameterization in finite-difference seismic wave simulation methods

Summary Grid point discretization of the model has a significant impact on the accuracy of finite-difference seismic waveform simulations. Discretizing the discontinuous velocity model using local point medium parameters can lead to artifact diffraction caused by the stair-step representation and inaccuracies in calculated waveforms due to interface errors, particularly evident when employing coarse grids. To accurately represent model interfaces and reduce interface errors in finite-difference calculations, various equivalent medium parametrization methods have been developed in recent years. Most of these methods require volume-integrated averaging calculations of the medium parameter values within grid cells. The simplest way to achieve this volume averaging is to apply numerical integration averaging to all grid cells. However, this approach demands considerable computational time. To address this computational challenge, we propose employing a set of auxiliary grids to identify which grid cells intersected by the welded interface and perform volume averaging only on these specific cells, thereby reducing unnecessary computational overhead. Additionally, we present a three-dimensional tilted transversely isotropic equivalent medium parameterization method, which effectively suppresses interface errors and artefact diffraction under the application of coarse grids. We also provide an approach for computing the normal direction of the interface, which is essential for the tilted transversely isotropic equivalent medium parameterization. Numerical tests validate the accuracy of the tilted transversely isotropic equivalent medium parameterization method and demonstrate the practicality of the implementation proposed in this paper for complex models.

Investigation of the accuracy of integral models of the oculo-motor system

For mathematical modeling of the human oculomotor system (OMS), integral nonlinear models are used, which simultaneously take into account the nonlinear and inertial properties of the research object. Based on the data of experimental studies of the OMS "input-output", transient and diagonal intersections of transient functions of the second and third orders are determined. To obtain experimental data, an innovative eye tracking technology is used, which allows recording eye responses to test visual stimuli. Thus test signals are displayed on the computer monitor at different distances from the starting position in the horizontal direction. The aim of the research is to study the accuracy of OMS identification using eye-tracking data by evaluating the calculation errors of multidimensional transient functions when using methods of nonlinear dynamic identification based on models in the form of Volterra series and polynomials. The object of the study is the process of nonparametric identification of the OMS based on Volterra models in the time domain. The subject of the research is algorithmic and software tools for calculating the dynamic characteristics of OMS based on eye-tracking data, analyzing the accuracy of the obtained models using two identification methods: the approximation method and the least squares method (LSM). The means of nonlinear dynamic identification of the human OMS based on Volterra series and polynomials were developed in the Python programming environment. The accuracy estimates of various OMS (linear, quadratic, and cubic) models were obtained based on data from three responses to test signals of different amplitudes. For the same test signals, the same models in the form of the Volterra series and polynomials were obtained, as these models coincide in the region of convergence of the Volterra series. The analysis of errors in the assessment of the dynamic characteristics of the OMS demonstrated that the model in the form of an integral polynomial of the second degree, which was built using the LSM based on three responses, has an accuracy twice as high as the accuracy of similar models built using the data of two responses. Thus, in further studies of the psychophysiological state of a person based on nonlinear dynamic models of the OMS according to the data of three responses, it is advisable to use a model in the form of a quadratic Volterra polynomial. Figs.: 17. Tabl. 3. Refs.: 10 titles.

A Protocol for Electron Probe Microanalysis (EPMA) of Monazite for Chemical Th-U-Pb Age Dating

A protocol for the monazite (LREE,Y,Th,U,Si,Ca)PO4 in situ Th-U-Pb dating by electron probe microanalyser (EPMA) involves a suitable reference monazite. Ages of several potential reference monazites were determined by TIMS-U-Pb isotope analysis. The EPMA protocol is based on calibration with REE-orthophosphates and a homogeneous Th-rich reference monazite at beam conditions of 20 kV, 50 nA, and 5 µm for best possible matrix matches and avoidance of dead time bias. EPMA measurement of samples and repeated analysis of the reference monazite are performed at beam conditions of 20 kV, 100 nA, and 5 µm. Analysis of Pb and U on a PETL crystal requires YLg-on-PbMa and ThMz-on-UMb interference corrections. Offline re-calibration of the Th calibration on the Th-rich reference monazite, to match its nominal age, is an essential part of the protocol. EPMA-Th-U-Pb data are checked in ThO2*-PbO coordinates for matching isochrones along regressions forced through zero. Error calculations of monazite age populations are performed by weighted average routines. Depending on the number of analyses and spread in ThO2*-PbO coordinates, minimum errors <10 Ma are possible and realistic for Paleozoic monazite ages. A test of the protocol was performed on two garnet metapelite samples from the Paleozoic metamorphic Zone of Erbendorf-Vohenstrauß (NE-Bavaria, western Bohemian Massif).

The Impact of Renewable Energy on Low Energy Consumption in Civil Engineering Construction

As the main representative of new energy, photovoltaic energy is widely used in civil engineering construction, its purpose is to reduce the energy consumption in the project, there is a deviation in the integration of new energy and civil engineering, and it cannot effectively play its own advantages. Continuously analyze the changes in the energy consumption values of civil engineering. Then, the influencing factors of building energy consumption were introduced, and the covariance matrix was constructed, and the engineering construction benefit was taken as the ultimate goal. When calculating energy consumption, the dual benefits of building benefits are fully considered, and environmental protection and engineering energy consumption are estimated to reduce the error of energy consumption calculation. The results show that new energy can reduce the high energy consumption value in the project, improve the light transmittance, enhance the conversion rate of sunlight, and provide support for the transportation, lighting and thermal insulation of the building, with an increase rate of 10~25%, and the new energy can reduce the cost of civil engineering, with a reduction rate of 19~26%, and improve the economic and social benefits, with an increase of more than 10%. Therefore, new energy can reduce the energy and energy consumption of civil engineering, and expand the scope and application depth of new energy development.

Experimental validation study of equivalent bare charge calculation model for two shelled warheads

After the explosion of the warhead shell fragments will consume a portion of the charge release energy, a method of calculating the shockwave overpressure value of the explosion of the warhead is to calculate the shockwave overpressure value by firstly converting the actual charge C of the warhead into the charge CEB of the bare charge, and then converting CEB into TNT equivalent. Experimental results were used to compare the calculation error size of the two models for calculating the equivalent bare charge of the warhead, and the results show that the calculation results of Chi Jiachun's modified model are in better conformity with the experimental results; by analyzing the influence of the empirical constant in Chi Jiachun's modified model on the calculation results, the value of the empirical constant is given as a suggestion.

Evaluation of Sulphur Dioxide Hourly Prediction Using Long Short-term Memory for Summer and Winter Season

Increasing air pollution has necessitated the prediction of pollutants over time. Deterministic, statistical, and machine-learning methods have been adopted to predict and forecast pollutant levels. It aids in planning and adopting measures to overcome the adverse effects of air pollution. This study employs long short-term memory (LSTM). This study used the hourly data from a meteorological station in a low-town area, Mohammedia City, Morocco. The model prediction accuracy was evaluated based on hourly, weekly, and seasonal (summer and winter) readings for the summer and winter of 2019, 2020 and 2021. Root mean square error (RMSE), mean absolute error (MAE) and mean arctangent absolute percentage error (MAAPE) were calculated to evaluate the accuracy of the developed LSTM model. The MAE value of 0.026 was observed to be less in winter than 0.029 during summer in 2019. Also, it was observed that MAE values decreased from Year 2019-2021, indicating increased prediction accuracy. MAAPE also observed a similar trend. However, RMSE values indicated the opposite for 2019 and 2020; in 2021, the RMSE value was 0.21 for summer and 0.14 for winter for hourly readings. Based on the error calculation, the study found weekly hourly readings were the most accurate for predicting SO2 concentration. Also, the LSTM model was more accurate in predicting winter SO2 concentration than in the summer season. Further studies must incorporate local incidences affecting the SO2 concentration into the LSTM model to increase its accuracy.