Generalized Regression Neural Network Model Research Articles

Forecasting regional water demand using multi-fidelity data and harris hawks optimization of generalized regression neural network models – A case study of Heilongjiang Province, China

Accurate forecasting of water demand plays a crucial role in decision makers to effectively allocate regional water resources and promotes the rational development and utilization of water resources. To address the challenges posed by limited raw data availability and the difficulty in selecting parameters for water demand forecasting models, this study proposes a coupled model called MF-HHO-GRNN (Multi-Fidelity Data, Hybrid Harris Hawks Optimization, and Generalized Regression Neural Network). Initially, multi-fidelity data was constructed employing water demand factor data for Heilongjiang Province, China (1999–2020) as high-fidelity data, alongside data from Liaoning and Jilin Provinces of China as low-fidelity data. Adaptive boosting algorithms, data concatenation, and data dimensionality reduction techniques were employed to construct multi-fidelity data, thereby improving the availability of the original data. Subsequently, Harris hawk optimization algorithms were implemented to optimize the smoothing factor of the GRNN to resolve the challenge of selecting suitable parameters for water demand forecasting models. Finally, the MF-HHO-GRNN model was compared with other prediction models to analyze the prediction accuracy. The optimized MF-HHO-GRNN model meeting accuracy requirements was utilized to forecast agricultural irrigation, industrial, domestic, ecological water demand, and total water demand in Heilongjiang Province for the next 5 years. Results demonstrated that the proposed MF-HHO-GRNN model attained predictive accuracy of 98 %, 98 %, and 97 % for total water demand under three distinct time horizons (1 year, 3 years, and 5 years), proving excellent predictive capabilities. Furthermore, a comprehensive comparison was made between the predicted total water demand and user-specific water demands for the next 5 years in Heilongjiang Province and the regional development plans. The results showed that the predictions of the coupled model aligned with the future development trends of Heilongjiang Province, validating the practicality of the MF-HHO-GRNN model. This study provides decision-makers with an effective method for forecasting water demand, contributing to the future realization of more intelligent and accurate water resource management and decision-making.

Precision positioning based on temperature dependence self-sensing magnetostrictive actuation mechanism

The self-sensing actuator that integrates both actuation and sensing functions is an effective way to simplify the structure of electromechanical systems. This paper proposes a temperature-dependent self-sensing actuation strategy based on a self-sensing giant magnetostrictive actuator (SSGMA) by sensing online stiffness. Utilizing the developed model considering temperature-dependent hysteresis nonlinearity, the self-sensing output characteristics of SSGMA are first evaluated. The self-sensing based control system is then designed in detail. The relationship between the self-sensing signal and the output displacement of SSGMA at different temperatures is constructed by General Regression Neural Network (GRNN) model for fast recognition by the controller. On this basis, a composite control scheme that takes into account rate-dependent asymmetric hysteresis nonlinearities and combines an adaptive feedforward controller with PID feedback controller is proposed for precision actuation of SSGMA. Finally, a temperature-controlled experimental platform is assembled for verification. Experimental results demonstrate that, based on the developed model, the SSGMA is capable of accurate self-sensing output at temperatures of 22 °C, 40 °C, and 70 °C, respectively. The SSGMA with the proposed control method enables the synchronous and precise self-sensing positioning of step and multiple-frequency sinusoidal expectations at different temperatures.

Modeling Xanthan Gum Foam's Material Properties Using Machine Learning Methods.

Xanthan gum is commonly used in the pharmaceutical, cosmetic, and food industries. However, there have been no studies on utilizing this natural biopolymer as a foam material in the insulation and packaging sectors, which are large markets, or modeling it using an artificial neural network. In this study, foam material production was carried out in an oven using different ratios of cellulose fiber and xanthan gum in a 5% citric acid medium. As a result of the physical and mechanical experiments conducted, it was determined that xanthan gum had a greater impact on the properties of the foam material than cellulose. The densities of the produced foam materials ranged from 49.42 kg/m3 to 172.2 kg/m3. In addition, the compressive and flexural moduli were found to vary between 235.25 KPa and 1257.52 KPa and between 1939.76 KPa and 12,736.39 KPa, respectively. Five machine-learning-based methods (multiple linear regression, support vector machines, artificial neural networks, least squares methods, and generalized regression neural networks) were utilized to analyze the effects of the components used in the foam formulation. These models yielded accurate results without time, material, or cost losses, making the process more efficient. The models predicted the best results for density, compression modulus, and flexural modulus achieved in the experimental tests. The generalized regression neural network model yielded impressive results, with R2 values above 0.97, enabling the acquisition of more quantitative data with fewer experimental results.

Building models of technological processes based on neuro-fuzzy technology

The work considers the issues of formalization of the extraction process in the form of a generalized regression neural network model, which are the basis for solving the problem of analysis and synthesis of the extraction process control system for obtaining petroleum products. An adaptive learning algorithm for a neural network model has been developed that is characterized by high speed and accuracy. A comparative analysis of the developed model with existing ones was made, which showed the effectiveness of the proposed algorithm for building the architecture of neural network models and learning the weight coefficients of the model.

Predicting and optimizing reactive oxygen species metabolism in Punica granatum L. through machine learning: role of exogenous GABA on antioxidant enzyme activity under drought and salinity stress

BackgroundDrought and salinity stress have been proposed as the main environmental factors threatening food security, as they adversely affect crops' agricultural productivity. As a potential solution, the application of plant growth regulators to enhance drought and salinity tolerance has gained considerable attention. γ-aminobutyric acid (GABA) is a four-carbon non-protein amino acid that accumulates in plants as a response to stressful conditions. This study focused on a comparative assessment of several machine learning (ML) regression models, including radial basis function, generalized regression neural network (GRNN), random forest (RF), and support vector regression (SVR) to develop predictive models for assessing the effect of different concentrations of GABA (0, 10, 20, and 40 mM) on various physio-biochemical traits during periods of drought, salinity, and combined stress conditions. The physio-biochemical traits included antioxidant enzyme activities (superoxide dismutase, SOD; peroxidase, POD; catalase, CAT; and ascorbate peroxidase, APX), protein content, malondialdehyde (MDA) levels, and hydrogen peroxide (H2O2) levels. The non‑dominated sorting genetic algorithm‑II (NSGA‑II) was employed for optimizing the superior prediction model.ResultsThe GRNN model outperformed the other ML algorithms and was therefore selected for optimization by NSGA-II. The GRNN-NSGA-II model revealed that treatment with GABA at concentrations of 20.90 mM and 20.54 mM, under combined drought and salinity stress conditions at 20.86 and 20.72 days post-treatment, respectively, could result in the maximum values for protein content (by 0.80 and 0.69), APX activity (by 50.63 and 51.51), SOD activity (by 0.54 and 0.53), POD activity (by 1.53 and 1.72), CAT activity (by 4.42 and 5.66), as well as lower MDA levels (by 0.12 and 0.15) and H2O2 levels (by 0.44 and 0.55), respectively, in the ‘Atabaki’ and ‘Rabab’ cultivars.ConclusionsThis study demonstrates that the GRNN-NSGA-II model, as an advanced ML algorithm with a strong predictive ability for outcomes in combined stressful environmental conditions, provides valuable insights into the significant factors influencing such multifactorial processes.

Time-series prediction of settlement deformation in shallow buried tunnels based on EMD-SSA-GRNN model

Tunnel settlement deformation monitoring is a complex task and can result in nonlinear dynamic changes. To overcome the disturbances caused by historical data and the difficulty in selecting input parameters during deformation prediction, a decomposition, reconstruction and optimization method for tunnel settlement deformation prediction is proposed. First, empirical mode decomposition (EMD) is used to decompose the in-situ monitoring data and reduce the interactions among information at different scales in sequences. Then, the monitoring data are decomposed into intrinsic mode functions (IMFs). Secondly, the smoothing factor of the generalized regression neural network (GRNN) is optimized by using the sparse search algorithm (SSA). An EMD-SSA-GRNN deformation prediction model is developed using the optimized GRNN algorithm and is used to predict the changes in the decomposed IMFs. Finally, using the measured deformation data from a shallowly buried tunnel along the Kaizhou-Yunyang Highway in Chongqing, China, the reliability and accuracy of different models are analysed. The results show that tunnel settlement deformation exhibited a trend and a slow change in the early stage, a rapid change in the middle stage and a slow change in the late stage, and the rate of change was significantly influenced by the excavation time and the upper and lower geological layers. The prediction accuracy of the EMD-SSA-GRNN model after EMD improved from 19.2 to 59.4% relative to that of the SSA-GRNN and single GRNN models. Moreover, we find that the three error evaluation indicators of the EMD-SSA-GRNN model are lower than those of the other models and that the results of the proposed model and are more strongly correlated with measured data.

Evaluating the efficacy of recyclable nanostructured adsorbents for rapid removal of methylparaben from aqueous solutions

In this study, we evaluate the efficiency of two novel nanostructured adsorbents – chitosan-graphitic carbon nitride@magnetite (CS-g-CN@Fe3O4) and graphitic carbon nitride@copper/zinc nanocomposite (g-CN@Cu/Zn NC) – for the rapid removal of methylparaben (MPB) from water. Our characterization methods, aimed at understanding the adsorbents’ structures and surface areas, informed our systematic examination of influential parameters including sonication time, adsorbent dosage, initial MPB concentration, and temperature.We applied advanced modeling techniques, such as response surface methodology (RSM), generalized regression neural network (GRNN), and radial basis function neural network (RBFNN), to evaluate the adsorption process. The adsorbents proved highly effective, achieving maximum adsorption capacities of 255 mg g−1 for CS-g-CN@Fe3O4 and 218 mg g−1 for g-CN@Cu/Zn NC. Through genetic algorithm (GA) optimization, we identified the optimal conditions for the highest MPB removal efficiency: a sonication period of 12.00 min and an adsorbent dose of 0.010 g for CS-g-CN@Fe3O4 NC, with an MPB concentration of 17.20 mg L−1 at 42.85 °C; and a sonication time of 10.25 min and a 0.011 g dose for g-CN@Cu/Zn NC, with an MPB concentration of 13.45 mg L−1 at 36.50 °C.The predictive accuracy of the RBFNN and GRNN models was confirmed to be satisfactory. Our findings demonstrate the significant capabilities of these synthesized adsorbents in effectively removing MPB from water, paving the way for optimized applications in water purification.

Prediction of NOx emission from two-stage combustion of NH3–H2 mixtures under various conditions using artificial neural networks

This study aims to predict NOx emission from the combustion of ammonia (NH3) and hydrogen (H2) mixtures using Artificial Neural Network (ANN) models. Chemical Reactor Network (CRN) models were utilized to calculate the NOx emission during the Rich-Quench-Lean (RQL) two-stage combustion of NH3–H2 mixtures under various conditions, such as fuel pressures, inlet temperatures, primary air humidification fractions, total equivalent ratios and primary equivalent ratios. The simulation results served as a data set for training and validation of three different ANN models: Backpropagation (BP), Radial Basis Function (RBF), and Generalized Regression Neural Network (GRNN). The results show that all three ANN models can effectively replace the CRN model in accurately predicting NOx concentrations. In particular, the BP model, optimized with 37 hidden neurons, showcased superior performance, achieving a correlation coefficient (R) value of 0.99826 with the simulated results. It had the lowest Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) values of 61.3 and 96.1, respectively. In addition, it achieved the highest coefficient of determination (R2) and Model Efficiency (ME) values of 0.9950 and 0.9908 respectively, surpassing the RBF and GRNN models. These results suggest that the BP model has significant potential as an alternative to the CRN model for predicting NOx emissions from two-stage combustion of NH3–H2 mixtures, demonstrating both accuracy and efficiency.

Physical fingerprint transformation of herbal medicines powders using near-infrared spectroscopy

To achieve the rapid determination of physical parameters of herbal medicine. A method based on near-infrared (NIR) spectroscopy was proposed. The potential of direct standardization, partial least squares regression and generalized regression neural network (GRNN) for physical fingerprint transformation of Paeoniae Radix Alba, Scutellariae Radix, Sinomenii Caulis, and Pueraria Lobatae Radix, were investigated. The results revealed that the predictive capacity of GRNN models was the best. Except for a few parameters in the validation samples of Pueraria Lobatae Radix and Sinomenii Caulis, the mean absolute deviations of all other physical parameters were less than 0.5. The similarity between the actual and predicted physical fingerprints of all validation samples and test samples was high when using the GRNN models (cosine coefficient > 0.99, Mahalanobis distance < 1.30). Additionally, the simplified GRNN models based on the 40 selected variables of hygroscopicity and angle of repose still showed the ideal predictive capacity.

Compliance Prediction for Structural Topology Optimization on the Basis of Moment Invariants and a Generalized Regression Neural Network.

Topology optimization techniques are essential for manufacturing industries, such as designing fiber-reinforced polymer composites (FRPCs) and structures with outstanding strength-to-weight ratios and light weights. In the SIMP approach, artificial intelligence algorithms are commonly utilized to enhance traditional FEM-based compliance minimization procedures. Based on an effective generalized regression neural network (GRNN), a new deep learning algorithm of compliance prediction for structural topology optimization is proposed. The algorithm learns the structural information using a fourth-order moment invariant analysis of the structural topology obtained from FEA at different iterations of classical topology optimization. A cantilever and a simply supported beam problem are used as ground-truth datasets, and the moment invariants are used as independent variables for input features. By comparing it with the well-known convolutional neural network (CNN) and deep neural network (DNN) models, the proposed GRNN model achieves a high prediction accuracy (R2 > 0.97) and drastically shortens the training and prediction cost. Furthermore, the GRNN algorithm exhibits excellent generalization ability on the prediction performance of the optimized topology with rotations and varied material volume fractions. This algorithm is promising for the replacement of the FEA calculation in the SIMP method, and can be applied to real-time optimization for advanced FRPC structure design.

A CNN–LSTM Machine-Learning Method for Estimating Particulate Organic Carbon from Remote Sensing in Lakes

As particulate organic carbon (POC) from lakes plays an important role in lake ecosystem sustainability and carbon cycle, the estimation of its concentration using satellite remote sensing is of great interest. However, the high complexity and variability of lake water composition pose major challenges to the estimation algorithm of POC concentration in Class II water. This study aimed to formulate a machine-learning algorithm to predict POC concentration and compare their modeling performance. A Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM) algorithm based on spectral and time sequences was proposed to construct an estimation model using the Sentinel 2 satellite images and water surface sample data of Chaohu Lake in China. As a comparison, the performances of the Backpropagation Neural Network (BP), Generalized Regression Neural Network (GRNN), and Convolutional Neural Network (CNN) models were evaluated for remote sensing inversion of POC concentration. The results show that the CNN–LSTM model obtained higher prediction precision than the BP, GRNN, and CNN models, with a coefficient of determination (R2) of 0.88, a root mean square error (RMSE) of 3.66, and residual prediction deviation (RPD) of 3.03, which are 6.02%, 22.13%, and 28.4% better than the CNN model, respectively. This indicates that CNN–LSTM effectively combines spatial and temporal information, quickly captures time-series features, strengthens the learning ability of multi-scale features, is conducive to improving estimation precision of remote sensing models, and offers good support for carbon source monitoring and assessment in lakes.

Parking Generating Rate Prediction Method Based on Grey Correlation Analysis and SSA-GRNN

The parking generating rate model is commonly used in parking demand forecasting. However, the key indicators of the parking generating rate are generally difficult to determine, especially its future annual value. The parking generating rate is affected by many factors. In order to more accurately predict the urban parking generating rate, this paper establishes a parking generating rate prediction model based on grey correlation analysis and a generalized regression neural network (GRNN) optimized by a sparrow search algorithm (SSA). Gross domestic product (GDP), urban area, urban population, motor vehicle ownership, and land use type are selected as input variables of the GRNN via grey correlation analysis. The SSA is used to optimize network weights and thresholds, and a model based on the SSA to optimize the GRNN is constructed to predict the parking generating rate of different cities. The results show that, after SSA optimization, the maximum absolute error of the GRNN model in predicting the parking generating rate is reduced, and the prediction accuracy of the model is effectively improved. This model can provide technical support for solving urban parking problems.

Application of General Regression Neural Network Algorithm in Data Mining for Predicting Glass Sales and Inventory Quantity

FF Jaya Glass is a shop that supplies and installs 3 mm to 12 mm glass. The store obtained glass from suppliers to be processed in shape and size according to customers’ order. After completing the customer's order, the shop worker will install the glass at the requested location. Unfortunately, currently stores do not utilize sales data to predict sales either manually or by utilizing technology. As a result, the store cannot predict when the number of glass orders will increase or decrease. In addition, errors often occur when ordering glass for the next period. As a result, stores often run out of glass supplies due to the large number of glass orders so that the achievement of profits is not optimal. This study aims to identify sales variables in glass sales data and build a general regression neural network model as a data mining method. In addition, this study aims to iterate to find the best value in the sales data training process, design and create applications according to user needs, and conduct system validation tests. The general regression neural network method is used to predict sales. The results of this study indicate that the application of general regression neural networks can be used to predict sales. This will make it easier for the store to provide glass supplies in the coming months with an accuracy of 98.1%.

Finite element and generalized regression neural network modelling of multiple cracks growth under the influence of multiple crack parameters

PurposeThis study presents finite element (FE) and generalized regression neural network (GRNN) approaches for modeling multiple crack growth problems and predicting crack-growth directions under the influence of multiple crack parameters.Design/methodology/approachTo determine the crack-growth direction in aluminum specimens, multiple crack parameters representing some degree of crack propagation complexity, including crack length, inclination angle, offset and distance, were examined. FE method models were developed for multiple crack growth simulations. To capture the complex relationships among multiple crack-growth variables, GRNN models were developed as nonlinear regression models. Six input variables and one output variable comprising 65 training and 20 test datasets were established.FindingsThe FE model could conveniently simulate the crack-growth directions. However, several multiple crack parameters could affect the simulation accuracy. The GRNN offers a reliable method for modeling the growth of multiple cracks. Using 76% of the total dataset, the NN model attained an R2 value of 0.985.Research limitations/implicationsThe models are presented for static multiple crack growth problems. No material anisotropy is observed.Practical implicationsIn practical crack-growth analyses, the NN approach provides significant benefits and savings.Originality/valueThe proposed GRNN model is simple to develop and accurate. Its performance was superior to that of other NN models. This model is also suitable for modeling multiple crack growths with arbitrary geometries. The proposed GRNN model demonstrates its prediction capability with a simpler learning process, thus producing efficient multiple crack growth predictions and assessments.

Experimental Study and GRNN Modeling of Shrinkage Characteristics for Wax Patterns of Gas Turbine Blades Considering the Influence of Complex Structures

With the continuous increase in power demand in aerospace, shipping, electricity, and other industries, a series of manufacturing requirements such as high precision, complex structure, and thin wall have been put forward for gas turbines. Gas turbine blades are the key parts of the gas turbine. Their manufacturing accuracy directly affects the fuel economy of the gas turbine. Thus, how to improve the manufacturing accuracy of gas turbine blades has always been a hot research topic. In this study, we perform a quantitative study on the correlation between process parameters and the overall wax pattern shrinkage of gas turbine blades in the wax injection process. A prediction model based on a generalized regression neural network (GRNN) is developed with the newly defined cross-sectional features consisting of area, area ratio, and some discrete point deviations. In the qualitative analysis of the cross-sectional features, it is concluded that the highest accuracy of the wax pattern is obtained for the fourth group of experiments, which corresponds to a holding pressure of 18 bars, a holding time of 180 s, and an injection temperature of 62 °C. The prediction model is trained and tested based on small experimental data, resulting in an average RE of 1.5% for the area, an average RE of 0.58% for the area ratio, and a maximum MSE of less than 0.06 mm2 for discrete point deviations. Experiments show that the GRNN prediction model constructed in this study is relatively accurate, which means that the shrinkage of the remaining major investment casting procedures can also be modeled and controlled separately to obtain turbine blades with higher accuracy.

Comparative Study of Suspended Sediment Load Prediction Models Based on Artificial Intelligence Methods

Quantification of suspended load sediment is crucial for maintaining the ecosystem and quality of water/river bodies that serve as the habitat for many living organisms. Because the influencing factors are nonlinearly related to the suspended load sediment, it is a challenge to apply linear statistical models to predict accurately. To address such a problem, this study appliedartificial intelligence methods to simulate and predict suspended load sediment. The artificial intelligence methods are robust andcan handle adequately issues related to nonlinearity in modeling. In the present study, four artificial intelligence methods were developed to predict suspended sediment load distribution. The methods include a backpropagation neural network, group method of data handling, least squares support vector machine, and generalised regression neural network. In developing the respective models, drainage areas, river slopes, and length of rivers served as predictor variables while suspended sediment load was the response variable. The models were evaluated using the metrics of root mean square error, percentage root mean square error, uncertainty at 95%, root mean square error observations standard deviation ratio, and Legates and McCabe index. According to the results, the generalised regression neural network model achieved higher prediction accuracy than the other competing methods. The performance of the generalised regression neural network model can be attributed to its ability to calibrate and generalise appropriately to the training and testing data set. Hence, in practice, the generalised regression neural network model is proposed for suspended sediment load prediction for the study area which can be useful to policymakers and managers of water resources.

Experimental study, numerical simulation and GRNN model for predicting the flexural loading capacity of corroded reinforced concrete beams

The present article is an effort to quantify the flexural strength of corroded reinforced concrete (RC) beams. Eight RC beams with dimensions of 150 × 200 × 2200 mm were tested in four-point bending, including six beams corroded in a chloride environment with average corrosion of longitudinal reinforcement in the range of 5% to 15%. The remaining two beams were used as control beams. The ultimate strength, failure mode and particularly the relationship between crack width and corrosion level are analyzed and discussed. A two-dimensional non-linear finite element model has been presented to simulate the steel bar corrosion considering the deterioration of concrete strength and steel-concrete bond stress. The generalised regression neural network (GRNN)-based model for the ultimate flexural strength of corroded RC beams has been established based on 123 experimental data collected from previous studies. The experimental results were used to separately validate the non-linear finite element model and novel GRNN model. Experimental results indicate that if the overall corrosion rate is used, the corrosion crack width can be a function of the percentage of section loss. The GRNN model provides superior predictive results compared to previous empirical methods with the highest correlation coefficient (r), and the smallest normalised root mean square error.

A Generalized Regression Neural Network Model for Accuracy Improvement of Global Precipitation Products: A Climate Zone-Based Local Optimization

The ability to obtain accurate precipitation data from various geographic locations is crucial for many applications. Various global products have been released in recent decades for estimating precipitation spatially and temporally. Nevertheless, it is extremely important to provide reliable and accurate products for estimating precipitation in a variety of environments. This is due to the complexity of topographic, climatic, and other factors. This study proposes a multi-product information combination for improving precipitation data accuracy based on a generalized regression neural network model using global and local optimization strategies. Firstly, the accuracy of ten global precipitation products from four different categories (satellite-based, gauge-corrected satellites, gauge-based, and reanalysis) was assessed using monthly precipitation data collected from 1896 gauge stations in Iran during 2003–2021. Secondly, to enhance the accuracy of the modeled precipitation products, the importance score of effective and auxiliary variables—such as elevation, the Enhanced Vegetation Index (EVI), the Land Surface Temperature (LST), the Soil Water Index (SWI), and interpolated precipitation maps—was assessed. Finally, a generalized regression neural network (GRNN) model with global and local optimization strategies was used to combine precipitation information from several products and auxiliary characteristics to produce precipitation data with high accuracy. Global precipitation products scored higher than interpolated precipitation products and surface characteristics. Furthermore, the importance score of the interpolated precipitation products was considerably higher than that of the surface characteristics. SWI, elevation, EVI, and LST scored 53%, 20%, 15%, and 12%, respectively, in terms of importance. The lowest RMSE values were associated with IMERGFinal, TRMM3B43, PERSIANN-CDR, ERA5, and GSMaP-Gauge. For precipitation estimation, these products had Kling–Gupta efficiency (KGE) values of 0.89, 0.86, 0.77, 0.78, and 0.60, respectively. The proposed GRNN-based precipitation product with a global (local) strategy showed RMSE and KGE values of 9.6 (8.5 mm/mo) and 0.92 (0.94), respectively, indicating higher accuracy. Generally, the accuracy of global precipitation products varies depending on climatic conditions. It was found that the proposed GRNN-derived precipitation product is more efficient under different climatic conditions than global precipitation products. Moreover, the local optimization strategy based on climatic classes outperformed the global optimization strategy.

State-of-Charge Prediction Model for Ni-Cd Batteries Considering Temperature and Noise

The accurate prediction of the state of charge (SOC) of Ni-Cd batteries is critical for developing battery management systems for high-speed trains. To address the challenges of the large floating charge voltage of Ni-Cd batteries and the vulnerability of a battery’s SOC to environmental factors such as temperature, this paper proposes an adaptive adjustment mechanism-based particle swarm optimization (APSO) generalized regression neural network (GRNN) model. The proposed model introduces the concept of the particle aggregation degree to quantify the convergence of the particle swarm optimization (PSO) algorithm. Furthermore, the speed weight of the particle swarm is adaptively adjusted using a comprehensive loss function to optimize the parameters of the GRNN model. To validate the proposed method, simulation experiments are conducted under test conditions using Ni-Cd batteries, and the prediction accuracies of various algorithms are compared. The experimental results demonstrate that the APSO-GRNN model significantly reduces the model’s prediction error. In addition, under the influence of different temperatures and noises, this method demonstrates strong robustness and high practical application value by accurately predicting the SOC, even with limited data samples.

ANFIS-based soft computing models for forecasting effective drought index over an arid region of India

Abstract Drought is a natural hazard that is characterized by a low amount of precipitation in a region. In order to evaluate the drought-related issues that cause chaos for human well-being, drought indices have become increasingly important. In this study, the monthly precipitation data from 1964 to 2013 (about 50 years) of the Jodhpur district in the drought-prone Rajasthan state of India was used to derive the effective drought index (EDI). The machine learning models hybridized with evolutionary optimizers such as the genetic algorithm adaptive neuro-fuzzy inference system (GA-ANFIS) and particle swarm optimization ANFIS (PSO-ANFIS) were used in addition to the generalized regression neural network (GRNN) to predict the EDI index. Using the partial autocorrelation function (PACF), models for forecasting the monthly EDI were constructed with 2-, 3- and 5-input combinations to evaluate their outcomes based on various performance indices. The results of the different combination models were compared. With reference to 2-input and 3-input combination models, both GA-ANFIS and PSO-ANFIS show better performance results with R2 = 0.75, while among the models with 5-input combination, GA-ANFIS depicts better performance results compared to other models with R2 = 0.78. The results are presented suitably with the aid of scatter plots, Taylor's diagram and violin plots. Overall, the GA-ANFIS and PSO-ANFIS models outperformed the GRNN model.