Lowest MSE Value Research Articles

Implementation of Backpropagation Artificial Neural Network for Electricity Load Forecasting in Jember District

The increase in population and various kinds of human activities in the world has made it possible for changes to increase the need for electrical power with demand that is not the same at any time. Based on this description, this research will propose research on the theme of electricity load forecasting as a preventive measure to determine future electricity load needs. Research was assisted using MATLAB data processing software to process research data. Three forecasting models were carried out, namely day, night and day-night conditions. From these three forecasting models, parameters such as epoch, number of input layers, number of hidden layers, activation function, and etc. The data is divided into two parts, training data and test data with a ratio of 70: 30. Test results using the backpropagation artificial neural network method show the highest MSE values for the three forecasting models, day, night, and day-night, are, 0.0039, 0.0041, and 0.002 while the lowest MSE values were in the three models are, 6.77E-04, 0.001, and 0.0011.

Stock price forecasting by supervised learning methods

Predicting stock prices has already long been a popular but challenging topic due to unavoidable uncertainties in various factors. With continuous exploration and development of machine learning algorithms over the years, they have become more and more popular in forecasting stock value changes. This paper aiming to predict future stock prices using returns from the past by applying three different supervised learning methods to the seemingly unpredictable dataset, namely Long Short-Term Memory (LSTM), Linear regression and Fully connected Neural Networks (FNN). R-squared and MSE were used as statistical indicators of the performance of each model. The calculated R-squared values for linear regression, LSTM and FNN were 0.876975, 0.898741 and 0.929504 respectively. In addition, the MSE of the corresponding models were 5.7786893, 0.0007542 and 0.0005237. As a conclusion, FNN performed the best in the sense that it put out the highest R-squared and lowest MSE value. As a result, FNN may take priority to the other manners in future predictions.

Test of Land Potential Sensor Tools Based on Biomass on Measurements of Temperature, Moisture, and Soil pH and Their Relationship to the Biomass Values in Marginal Land

This research aims to test the accuracy of biomass-based land potential sensor devices. The indicators tested were the results of measurements of growing environmental conditions including temperature, humidity and soil pH, as well as measurements of tree biomass. The analysis technique is carried out in 2 stages, namely the comparative measurement test stage and the biomass calculation application test in the field. The research results show that the average measurement of pH = 5.7493, humidity = 88.0616% and temperature = 27.016ºC is generally not significantly different at the 95% t-confidence interval from the average of the measurement results from the 3 comparison sensor devices. In the pH measurement test, the 3 comparison devices had the lowest MSE value and the lowest standard deviation was in the pH-1 device of 0.03346, the humidity test in the RH-1 device was 0.03 and the temperature in the Temp-1 device was 0.0142. This shows that the sensor device built has quite accurate and valid measurement values with a fairly low error rate. The results of the application of sensor devices to measure tree biomass values obtained an average value of 163.878 kg/ha with a standard deviation of ± 68.878 and provided a significant difference at the 0.001 level for the 11 villages measured, as well as the relationship between biomass values and the growing environment is not linear.

Investigation of goose breeding in Turkiye by linear and nonlinear regression models

In this study, the change in the number of geese breeding in Turkiye over the years was examined by linear and non-linear regression models. Among linear and non-linear regression models, linear, quadratic, cubic, logarithmic, and inverse regression models were used. R2 and MSE values were taken as criteria for comparing the models. As a result of the study, the cubic regression model with the highest R2 value and the lowest MSE value was found to be the best fitting model for the number of geese. According to the cubic regression model, the number of geese in Turkiye in 2023 and 2024 was estimated to be 1849304 and 2107588, respectively.

Object Detection and Regression Based Visible Spectrophotometric Analysis: A Demonstration Using Methylene Blue Solution

This study investigates the estimation of the concentration of methylene blue solutions to understand if visible spectrophotometry could be performed using a smartphone and machine learning. The presented procedure consists of taking photos, detecting test tubes and sampling region of interest (ROI) with YOLOv5, finding the hue, saturation, value (HSV) code of the dominant color in the ROI, and regression. 257 photos were taken for the procedure. The YOLOv5 object detection architecture was trained on 928 images and the highest mAP@05 values were detected as 0.915 in 300 epochs. For automatic ROI sampling, the YOLOv5 detect.py file was edited. The trained YOLOv5 detected 254 out of 257 test tubes and extracted ROIs. The HSV code of the dominant color in the exported ROI images was determined and stored in a csv file together with the concentration values. Subsequently, 25 different regression algorithms were applied to the generated data set. The extra trees regressor was the most generalizing model with 99.5% training and 99.4% validation R2 values. A hyperparameter tuning process was performed on the extra trees regressor and a mixed model was created using the best 3 regression algorithms to improve the R2 value. Finally, all three models were tested on unseen data and the lowest MSE value was found in the untuned extra trees regressor and blended model with values of 0.10564 and 0.16586, respectively. These results prove that visible spectrophotometric analysis can be performed using the presented procedure and that a mobile application can be developed for this purpose.

Using the Arima Method with Minitab Applications for Forecasting Work Order Projects of Casting Construction (Case Study: PT. Bumi Sarana Beton)

The ARIMA method is a non-stationary homogeneous time series model that uses the procedure for applying the Autoregressive model or scheme and the Moving Average in preparing its forecasts. The purpose of this study was to determine the application of the Autoregressive Integrated Moving Average (ARIMA) method and minitab application at PT. Bumi Sarana Beton in planning the estimated number of work orders. It can be concluded that for K-225 the best model is the ARIMA model (2,0,0) because it has the lowest MSE value and for K-400 the best model to use is the ARIMA model (2,0,2 ) because it has the smallest value. From the results of the research that has been done, it can be concluded that the production of K-225 in 2022 is 3573.50 m3, while using the ARIMA method in 2023 the total production is 3920.61 m3 and in 2024 the total production is 3573.50 m3. to 3824.36 m3. Production of K-400 in 2022 was 2015.00 m3, while using the ARIMA method in 2023 a total production of 1857.07 m3 was obtained and in 2024 a total production of 2045.89 m3 was obtained.

Utilization of Webcam Cameras as X-Ray Image Capture (kV Settings and Shutter Sensors)

The result of x-ray imagery is generally in the form of film sheets. To obtain an image, the film must pass through the process of processing with chemicals. Departing from these problems, efforts were made to develop digital x-ray detectors using much cheaper devices. By using voltage variations and also sensors that aim to find out the difference in the quality of the resulting image. The contribution of this study is that the system can display negative image results from webcam camera captures that have previously passed the image processing process with the Matrix Laboratory (MATLAB) Application. The measurement ranges used are 60, 65, and 70 kV. with a set mA of 25mA, a duration of irradiation of 1 second, and a moderate intensity of illumination light. From the measurement results, it shows that the X-ray Image Capture Tool can be compared with the image results from the Philip brand DR with the MSE value obtained, which is 34.8775 with parameters Phototransistor BPT1331, 70kV, and 25mA. And the lowest MSE value is 61.7615 with parameters LDR, 66kV, 25mA. The results of this study indicate that the tool can be used to capture X-rays.

Day Ahead Energy Consumption Forecasting Through Time-Series Neural Network

Demand response management through appliance scheduling can effectively decrease electricity bills. Similarly, it can decrease the peak demand of both consumers and the utility grid. However, prior knowledge of the load profile of the consumer is required for effective appliance scheduling. This work developed a novel time-series forecasting model within the shallow neural network framework to predict the load curve for optimal planning of demand response by a nonlinear autoregressive network with exogenous input network. The algorithms such as Levenberg-Marquard, scaled conjugate gradient, and Bayesian regularization was applied to train the model. The results were compared with conventional seasonal autoregressive integrated moving average statistical model and long short-term memory network. The results indicated that the optimum methodology is a nonlinear autoregressive network with exogenous inputs using the Bayesian regularisation algorithm, which has the lowest MSE value in the training and testing phases of 0.0031 and 0.0029, respectively. It is practical to continue designing artificial neural networks to analyze hourly load consumption in the context of the positive outcomes acquired.

An improved prediction of frictional pressure drop using an accurate shear rate equation for high-yield stress drilling fluids at the surface and downhole conditions

This paper evaluates a more accurate and relatively new shear rate equation based on the phenomenological model for the high yield stress drilling fluids to estimate frictional pressure drop in laminar and turbulent flow regimes under surface and downhole conditions. Fann viscometer readings are used to determine shear rates using the above phenomenological model that differ from the conventional shear rates, i.e., γ˙(s−1)=1.7N, where N = Fann viscometer rotor rpm. The new shear rates are used to predict model parameters of Bingham plastic (BP), power-law (PL), Herschel–Bulkley (HB), and Robertson-Stiff (RS) models, and corresponding frictional pressure drops. The efficacy of the above rheology models with improved model parameters is tested by validating the literature-reported experimental pressure drop data of low to high yield stress water-based drilling fluids. The validation results are consistent and accurately predict frictional pressure drops using HB and RS models with the lowest MSE-values. The above methodology is applied to predict frictional pressure drops in a pipe for ten test high-pressure, high-temperature (HPHT) drilling fluids of varying base fluids at the surface and HPHT conditions in laminar and turbulent flow regimes. For this, isothermal compressibility and thermal decomposition temperature of drilling fluids are measured to forecast the temperature- and pressure-dependent densities and shear viscosities. The proposed density model is experimentally validated and found to predict density at downhole conditions adequately with mean absolute- and standard-deviation, i.e., 1.67 and 1.36%, respectively. Temperature- and pressure-dependent viscosity and density have been used to predict pressure drop under downhole conditions using HB and RS models with improved model parameters from the phenomenological model. The HB model is found to be more accurate than the RS model for predicting pressure drop. The proposed frictional pressure drop estimation method in the present study is more general and accurate under the surface and downhole conditions than the conventional practice.

Energy consumption forecast of Turkey using artificial neural networks from a sustainability perspective

ABSTRACT As industrialization, population and living standards increase, energy becomes increasingly crucial. Therefore, modeling and forecasting of energy use is critical for countries. The studies carried out for energy consumption estimation were examined and indicators used in the literature were classified according to the three main dimensions of sustainability, social, economic, and environmental factors. Turkey's energy consumption estimates were made by four sustainability-based models. Input variables are GDP, electricity sales prices, import, population and temperature, while output variable is gross energy consumption. The models were tested and compared with Artificial Neural Networks (ANN). The best-performing Social-Economic-Environmental model has lowest MSE value (0.0001114) and highest R² value (0.99912). This model was used to forecast Turkey's energy use between 2021 and 2025.

Feature Selection for financial data – comparison

Data analysis is currently one the key for the success of good condition of the companies. Feature selection as a preprocessing of data method the estimator accuracy scores can be improved, as well the performance on very high-dimensional data set can be boosted.The paper presents an application of the filter Feature Selection method to check how the data set after the application of the selected method will look like. The goal of this article is to compare the feature selection method with the application without usage of Feature Selection about the set of prepared financial data. For this purpose, the experiments have been executed with usage of Jupyter Notebook with Python as a programming language. The data source for the data set was Stooq, which is an open source website for tracking of changes on the financial markets. For this purpose, several joint-stock companies of various sectors from Polish Stock Exchange have been selected. The Feature Selection has been applied before the data has been entered into the Artificial Neural Networks model. The results have been compared using the MSE value. The lowest MSE value has been obtained where no Feature Selection method has been applied, however MSE value for Filter Method is one order of magnitude smaller. For the experiments without Feature Selection 112 input variables have been used, for the Filter Method only 4 variables have been incorporated.

The application of hybrid cuckoo search-grey wolf optimization algorithm in optimal parameters identification of solid oxide fuel cell

In the present study, dependable and accurate variables of solid oxide fuel cell (SOFC) are identified using an accurate and remarkable convergence pace optimization algorithm, so-called cuckoo search grey wolf optimization (CSGWO). The mentioned hybrid algorithm is acquired from the integration of cuckoo search (CS) and grey wolf optimization (GWO) algorithms. To confirm the CSGWO performance, it is compared with well-known optimization algorithms. The proposed CSGWO hybrid algorithm is applied for a 5-kW physically-based dynamic tubular stack. Along with this, the modeling is performed for different temperatures and pressures. The results of the CSGWO integrated algorithm reveal the supremacy of the considered method compared to other algorithms with the lowest values of MSE and corroborate precision, robustness, and excellent convergence pace compared to different optimization algorithms. The statistical results revealed that the developed CSGWO algorithm has the lowest MSE values by 1.3%, 0.1%, 0.3%, 0.1%, and 0% for the operational pressures of 1, 2, 3, 4, and 5atm, respectively. All in all, the experimental records confirm that the CSGWO integrated algorithm could be introduced as a favorable substitute to the SOFC models’ parameter estimation.

Image Enhancement Algorithm Based on Image Spatial Domain Segmentation

This paper proposes an image enhancement algorithm based on the theory of image segmentation and image frequency. First, a mathematical model corresponding to the pixel frequency is established by using the difference of pigment values between the pixels in the image and the surrounding pixels (i.e., pixel receptive field). Then, the image is divided into the low frequency region (background area), low-medium frequency region (foreground area), medium-high frequency region (target area) and high frequency region (detail area) by a pixel frequency characteristic graph. Gamma correction, MSRCR, MSR, top hat+bottom hat are used for image enhancement for each area, and then the parts are merged. Three indicators of PSNR, SSIM, and MSE are introduced to evaluate the quality of the enhanced image. The results show that the image enhanced by this algorithm has the highest PSNR and SSIM values and the lowest MSE value, indicating that the enhancement effect of this algorithm is better. Compared with traditional algorithms, the image enhancement algorithm in this paper produces higher image quality and richer details.

The Implementation of First Order and Second Order with Mixed Measurement to Identify Farmers Satisfaction

The purpose of this research is to develop structural modeling with metric and nonmetric measurement scales. Also, this study compares the level of efficiency between the first order and second-order models. The application of structural modeling in agriculture is the satisfaction of farmers in East Java. The data used in this study are about perceptions by distributing questionnaires to farmers in East Java Province in 2020. The respondents in this study were 155 districts in East Java Province. Therefore, the sampling technique chosen is probability sampling, which is a proportional area random sampling. The results are obtained that the first-order model is better than the second-order model because it has the lowest MSE value and the highest R2. The results of the path analysis for the first order and second-order models produce the same results that there is a significant positive effect between the gratitude variables on the farmer satisfaction variable. That is, the more gratitude felt by farmers, the satisfaction will be increased by East Java Farmers. On the other hand, the test results showed that demographic variables did not significantly influence gratitude variables.

Decontamination of methylene Blue from simulated wastewater by the mesoporous rGO/Fe/Co nanohybrids: Artificial intelligence modeling and optimization

Decontamination of dye wastewater by nanomaterials has attracted an increasing attention. The mesoporous rGO/Fe/Co nanohybrids prepared by the co-precipitation method were used to decontaminate methylene blue (MB) from simulated wastewater. The obtained products were systematically characterized with X-ray diffraction, scanning electron microscopy, Raman spectroscopy, energy dispersive spectroscopy, N2-sorption, and X-ray photoelectron spectroscopy. It was found that these nanohybrids possessed a surface area of 108.445 m²/g and a narrow pore size distribution with a center of 3.94 nm. The effect of initial pH, temperature, contact time, initial concentration on the extent of decontamination by the material was investigated and modeled with artificial intelligence (AI). AI has become an important driving force for a new round of scientific and technological revolution and industrial transformation, which can achieve the overall improvement of social productivity. The optimum conditions were predicted and the maximum decontamination efficiency obtained by response surface methodology and AI, i.e. artificial neural network-particle swarm (ANN-PSO), artificial neural network-genetic algorithm (ANN-GA). Among the three models developed, the ANN-PSO model has the highest R2 value and the lowest MSE value, thus this model has the best performance. The experimental equilibrium data were analyzed with Langmuir, Freundlich, Tempkin and D-R isothermal models. It was exhibited that the data were well presented by Langmuir model with a maximum adsorption capacity of 909.10 mg/g at 298 K for the model. The kinetic study demonstrated that the decontamination processes could be well described by the pseudo-second-order model. In addition, thermodynamic parameters were calculated, which include Gibbs free energy change (ΔG°), entropy change (ΔS°), and enthalpy change (ΔH°). Overall, this material could be potentially used as a low cost and fast decontaminant to treat organic contaminants in wastewater or other pollutants.

Classification of alcohols obtained by QCM sensors with different characteristics using ABC based neural network

Alcohols with different structures are used frequently in hygiene products and cosmetics. It is desirable to classify these alcohols to evaluate their potential harmful effects using less costly methods. In this study, five different types of alcohol are classified using five QCM sensors with different structures. The main idea of the study is to determine the QCM sensor that makes the most successful classification. All the five of the QCM sensors gave successful results, but QCM12-constructed using only NP-was the most successful. ABC-based ANN is used for the classification, and the lowest MSE value in test dataset is obtained as 1.41E−16. The results of 300 different scenarios showed that different alcohols can be classified successfully by using ANN-ABC on the sensor data from QCM12.

Noise effect analysis on edge detection in detecting digits with bilateral filter

Recognizing characters such as digits in natural images is a challenge in image processing. Edge detection is the first approach in image recognition. Edge detection using Bilateral Filter can improve image quality. However, in image processing, noise often occurs in the image which causes objects in the image to be more difficult to detect. In this paper, we conducted experiments to detect edges by using Bilateral filters on each edge detection operator against digits on SVHN which added noise and analyzes the effect of noise on edge detection. In the proposed method, edge detection of SVHN with Speckle noise does not produce many edges compared to Salt & Pepper noise so digits in SVHN can still be seen. The performance of the edge detection operator with Bilateral Filter can also be seen from the MSE and PSNR values. The lowest MSE value and the highest PSNR value is on Canny when detecting SVHN edges with Speckle noise.

PELACAKAN DAN SEGMENTASI OBJEK BERGERAK MENGGUNAKAN METODE K-MEANS CLUSTERING BERBASIS VARIASI JARAK

In computer vision tracking and object segmentation is one important step in video processing. Accuracy in object tracking is important in video processing, where accurate object tracking is a thing that continues to be done by many researchers. there are still many problems that are often experienced when tracking objects in terms of lighting, noise up to a high level of error. Many methods can be used in research, one of which is clustering method. Clustering method is a method that is widely used in grouping data, one of which is often used is Kmeans clustering. This method is very flexible, and is able to classify large amounts of data. Besides that, Kmeans is also able to work adeptly and segment the image well. For this study using 5 distance approaches (cambera, chebychef, mahattan, minkowski, Euclidean) distance approach which is expected to improve the results of better accuracy. From the results of the research produced a mahatan distance approach has the best accuracy results with a PNSR value of 16,34399 and the lowest MSE value with a value of 1521,793. Compared to the use of standard models with Euclidean, the approach of high distance accuracy increases