Back Propagation Neural Network Technique Research Articles

Human face identification after plastic surgery using SURF, Multi-KNN and BPNN techniques

Facial identification for surgical and non-surgical datasets is getting popular. The reason behind this popularity is the growing need of a robust facial recognition system which is consistent to occlusion, spoofing attacks and last but most important plastic surgery effects. Plastic therapies are undertaken by individuals to beautify their external appearance but it is also undertaken by impostors to commit crimes and falsify their true identities. This research work aims at developing a facial recognition system which can identify genuine and impostor pairs. The projected methodology optimizes face detection via Back-Propagation Neural Network (BPNN) and dimensionality reduction by means of Speeded Up Robust Features followed by Multi-K-Nearest-Neighbor technique. The novelty is the production of an innovative-fangled T-Database which trains the BPNN. Thus, BPNN converges faster and achieves higher recognition. The proposed scheme is not applied till date on a medically altered dataset. We have applied five distance metrics and integrated them to acquire T-Dataset, which is fed to the BPNN. This scheme is tested on surgical and non-surgical datasets and it is deduced that higher recognition is achieved with non-surgical databases as compared to surgical ones. For both surgical and non-surgical datasets, the computational cost attained is the modest.

Particle Swarm Optimized Back Propagation Neural Network for State of Health Estimation of Lithium-ion Battery

The assessment and monitoring of battery health is very crucial for the maintenance and safety of battery-powered applications such as Electric vehicles (EVs). To conduct appropriate battery operation in EVs, the battery capacity should be estimated accurately. In this regard, the State of health (SOH) estimation is conducted for evaluating the battery aging status. This work proposes a hybrid backpropagation neural network (BPNN) and particle swarm optimization (PSO) technique for SOH estimation. A multi-feature input data framework is constructed with 31-dimensional features for the model training by using 4 battery datasets from NASA i.e. B5, B6, B7 and B18. The acquisition of the data samples has been performed with a systematic sampling technique. The presented work is conducted with a training testing ratio of 70:30 and validated with the MIT Stanford battery dataset. The experimental outcomes demonstrated high SOH estimation accuracy compared with the conventional BPNN model. In the case of battery B5, it was observed that RMSE, MSE and MAPE for the BPNN-PSO model are 0.6791, 0.0046, 0.3203 compared with the conventional BPNN model i.e. 0.8796, 0.0077, 0.4881 respectively. Furthermore, the significance of capacity regeneration in B7 and B18 results in high-performance metrics compared with other battery datasets. The research conducted would be beneficial to estimate the battery status regarding battery health i.e. SOH accurately in Battery System Management (BMS) based EV application.

State of charge estimation of lithium-ion batteries using improved BP neural network and filtering techniques

The state of charge (SOC) is a critical parameter in the battery management system (BMS), and its accurate estimation is essential for ensuring the safety and reliability of batteries. This paper presents a lithium-ion battery SOC estimation method that combines an improved neural network with a filtering algorithm. Firstly, the backpropagation (BP) algorithm is chosen as the architecture of the neural network in the hybrid method due to its strong nonlinear approximation ability, and the particle swarm optimization (PSO) algorithm is used to optimize it to avoid falling into local optimal solutions. By combining the search ability of PSO with the learning ability of the BP neural network, the accuracy of the neural network model is improved. The proposed method integrates the PSO-BP model with the extended Kalman filter based on minimum error entropy (MEE-EKF). PSO-BP is utilized as the measurement equation for MEE-EKF, while the ampere-hour integration method is employed as the state equation to achieve closed-loop SOC estimation. Finally, experimental validation is conducted under four typical operating conditions and one random condition across a wide temperature range. The results demonstrate that the proposed method achieves high accuracy across all conditions compared with the results of other algorithms, with a maximum absolute error of not exceeding 3.13%, a mean absolute error of less than 0.54%, and a root mean square error of no more than 0.66%.

Intelligent Optimization Design of Distillation Columns Using Surrogate Models Based on GA-BP

The design of distillation columns significantly impacts the economy, energy consumption, and environment of chemical processes. However, optimizing the design of distillation columns is a very challenging problem. In order to develop an intelligent technique to obtain the best design solution, improve design efficiency, and minimize reliance on experience in the design process, a design methodology based on the GA-BP model is proposed in this paper. Firstly, a distillation column surrogate model is established using the back propagation neural network technique based on the training data from the rigorous simulation, which covers all possible changes in feed conditions, operating conditions, and design parameters. The essence of this step is to turn the distillation design process from model-driven to data-driven. Secondly, the model takes the minimum TAC as the objective function and performs the optimization search using a Genetic Algorithm to obtain the design solution with the minimum TAC, in which a life-cycle assessment (LCA) model is incorporated to evaluate the obtained optimized design solution from both economic and environmental aspects. Finally, the feasibility of the proposed method is verified with a propylene distillation column as an example. The results show that the method has advantages in convergence speed without sacrificing accuracy and can obtain an improved design solution with reduced cost and environmental impact. Compared with the original design using rigorous simulation, the TAC is reduced by 6.1% and carbon emission by 27.13 kgCO2/t.

An optimized back propagation neural network for automated evaluation of health condition using sensor data

Ships and other large equipment must meet strict standards for equipment integrity and operational dependability in order to perform missions. To meet this demand, one of the essential linkages is to guarantee the long-term safe and healthy functioning of their power transmission equipment. The Optimized Back Propagation Neural Network (OBPNN) technique used in this study introduces a unique method for monitoring sensor data and evaluating the health state, with the SVM being optimized using the fish swarm algorithm (FSA). A major problem that maintenance is facing nowadays is reliable fault prediction. One of the trickiest difficulties is arguably automatically modelling typical behaviour from condition monitoring data, particularly when there is little information about actual failures. A data-driven learning framework with the best bandwidth selection is suggested to address this challenge. It is based on nonparametric density estimation for outlier identification and OBPNN for normality modelling. The distance to the separating hyper plane's log-normalization is used to provide a health score that is also available. The algorithm's viability is shown by experimental findings while evaluating the progression of a major defect over time in a marine diesel engine. Improved prediction capabilities and low false positive rates on healthy data are realized.

On impact process of FGPM plate with the damaged interlayer: piezoelectricity-based interaction effect analysis and stress field prediction using BPNN

Revealing the piezoelectricity-based interaction effect on the impact process of functionally graded piezoelectric material(FGPM) plates with the damaged interlayer promotes the intelligent development of engineering structures. The extended neural network algorithm is conducive to the integrated design of the intelligent structure. The impact process of FGPM plate with damaged interlayer is investigated. In the framework of Reddy's higher-order shear plate theory, the reduced constitutive equations considering the piezoelectric effect and damage effect are established for the FGPM layer and damaged interlayer respectively. The contact forces in the loading and unloading processes are modeled by the modified Hertz contact theory. The impact dynamic equations are derived through the Hamiltonian variational principle. The high-order truncated double trigonometric series solution in the impact process is extended. Further, the back propagation neural network(BPNN) technique is developed to predict the stress state during the impact process. Ultimately, from the perspective of energy conversion, the interaction of piezoelectricity and damage effect, structural parameters and loading state on the impact process are systematically revealed. The developed BPNN technique achieves accurate prediction of the flexural stress fields.

Retrieval of atmospheric temperature and humidity profiles over a tropical coastal station from ground-based Microwave Radiometer using deep learning technique

Ground-based multi-frequency microwave radiometer Profiler (Microwave Radiometer Profiler-MRP) provides valuable information on the altitude distribution of atmospheric humidity and temperature in the troposphere (upto ∼10 km) with high temporal resolution (of ∼1 min) under all-weather condition. These information are vital inputs for characterization of the atmospheric boundary layer, convective cloud systems, local weather modification and for studying the atmospheric dynamics. Potential of such radiometers are well established over the mid-latitude region. Thiruvananthapuram, located in the Indian peninsular region, is one of the first stations in the equatorial region where such an instrument made continuously operational for seven years since 2010 at the tropical coastal station. The instrument measures the brightness temperatures (Tb in Kelvin) at Ka- and V-band frequencies with a temporal resolution of ∼1 min. A trained back-propagation neural network technique based on the radiosonde ascends data for more than one decade obtained by India Meteorological station at Thiruvananthapuram, was used to derive the atmospheric humidity and temperature profiles. This paper presents the implementation of another retrieval technique based on deep learning approach - batch normalization and neural network (BNN) to retrieve temperature and water vapour density (WVD) profiles from Tb values during clear sky conditions and validation of the retrieval accuracies. This inversion model, much faster in computation, consists of two hidden layers and the rectified linear unit (ReLU) is used as the activation function as it can overcome the problems of saturation and vanishing gradients. The Tbs observed by radiometer are corrected by reducing the bias between the simulated Tb, using forward radiative transfer model, and the observed Tb. The validation of retrieved profiles with the radiosonde demonstrates a good retrieval capability, showing a root-mean-square error of 1.8K for temperature, <2 g/m3 for WVD and ∼14% for RH.

Numerical solutions of the Wolbachia invasive model using Levenberg-Marquardt backpropagation neural network technique

The current study presents the numerical solutions of the Wolbachia invasive model (WIM) using the neural network Levenberg-Marquardt (NN-LM) backpropagation technique. The dynamics of the Wolbachia model is categorized into four classes, namely Wolbachia-uninfected aquatic mosquitoes (An∗), Wolbachia-uninfected adult female mosquitoes (Fn∗), Wolbachia-infected aquatic mosquitoes (Aw∗), and Wolbachia-infected adult female mosquitoes (Fw∗). A reference dataset for the proposed NN-LM technique is created by solving the Wolbachia model using the Runge-Kutta (RK) numerical method. The reference dataset is used for validation, training, and testing of the proposed NN-LM technique for three different cases. The obtained numerical results from the proposed neural network technique are compared with the results obtained from the RK method for accuracy, correctness, and efficiency of the designed methodology. The validation of the proposed solution methodology is checked through the mean square error (MSE), error histograms, error plots, regression plots, and fitness plots.

Comparative assessment of back propagation neural network and self organising feature map for iris-based access control system

Iris recognition is considered as one of the best biometric methods used for human identification and verification because of its unique features that differ from one person to another. Self-Organizing Feature Map (SOFM) and Back Propagation Neural Network (BPNN) are two techniques that have been used previously for iris recognition but their performance comparison has not been adequately investigated. This research, therefore, carried out a comparative analysis between SOFM and BPNN in iris recognition. Three hundred (300) iris images from one hundred and fifty (150) subjects were acquired from Covenant University Iris dataset. The acquired iris images were pre-processed by segmenting the iris portion using Hough Transform and were normalized using Daugman’s Rubber Sheet Model. Local Binary Pattern was used for feature extraction and dimension reduction of the iris images. Each of the two algorithms, SOFM and BPNN was used individually as iris image classifier, this technique was implemented using MATLAB (R2016a). The performance of the two iris classifiers was evaluated individually and compared at 0.75 threshold value based on Recognition Accuracy (RA), False Acceptance Rate (FAR), False Rejection Rate (FRR), Equal Error Rate (EER), Training Time (TT) and Recognition Time (RT). The SOFM and BPNN techniques were validated by carrying out a t-test to compare the differences between the two techniques at 5% significance level. The SOFM technique gave RA, FAR, FRR, EER, TT and RT of 97.14%, 1.67%, 3.75%, 3.18%, 104.47s and 81.98s, respectively, while the corresponding values for BPNN were 94.29%, 5.00%, 6.25%, 5.91%, 116.23s and 101.77s respectively. The P value between the SOFM and BPNN techniques was 0.002. This research outcome revealed that SOFM outperformed BPNN in iris recognition system with respect to RA, FAR, FRR, EER, TT and RT. The SOFM technique could be used for more robust iris recognition system than BPNN in security surveillance systems or other related systems.

Highly Efficient Inverse Design of Semiconductor Optical Amplifiers Based on Neural Network Improved Particle Swarm Optimization Algorithm

An artificial intelligent neural network improved particle swarm optimization algorithm is proposed for the inverse design of semiconductor optical amplifier. Seven input parameters, current-gain curve and saturation output power curve are selected to form the data set based on the physical model of semiconductor optical amplifier. The effectiveness of forecasting performance is improved by contrasting two back propagation neural network techniques (Scaled Conjugate Gradient and Levenberg-Marquardt) and operational settings (Central Processing Unit and Graphics Processing Unit). Higher accuracy is achieved through feedback analysis of neuron number optimization and test error. The addition of a unique backpropagation neural network can make the fitness of particle swarm algorithm mostly converge below <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$2\times 10^{-4}$</tex-math></inline-formula> . The relative difference between original performances and inverse predictions is close to 0%, which proves the effectiveness of parameter extraction. This method can take advantage of neural networks to improve accuracy and speed of particle swarm optimization algorithms for efficient semiconductor optical amplifier inverse design and multi-solution analysis.

Performance Evaluation of Selected Classification Algorithms for Iris Recognition System

Quite a lot of techniques proven to be resourceful have been espoused to develop iris recognition system. Nearly hybridized, supervised and unsupervised artificial neural network techniques have been used individually in iris recognition system and other pattern recognitions but have not been compared based on some performance metrics. Counter Propagation Neural Network (CPNN) is a hybridized technique, Self-Organizing Feature Map (SOFM) is an unsupervised learning technique and Back Propagation Neural Network (BPNN) is a supervised learning technique. This research conducted a performance comparison of CPNN, SOFM and BPNN techniques to recognize iris dataset and establish the more efficient among the three techniques. A database of Three hundred (300) iris images was acquired from LAUIRIS dataset from LAUTECH Biometric Research Group database. The original images of 640*360 dimensions were resized to 200*200 without any alteration in the image using 80% for training and 20% for testing. Hough transform was applied to segment locate the iris region of eye image. Daugman’s Rubber Sheet Model was used to create a dimensionally consistent representation. Principle Component Analysis was applied for feature extraction and dimensionally reduction. Finally, classification and matching were done by using CPNN, SOFM and BPNN techniques. This was implemented using MATLAB (Matrix Laboratory) R2016b. The performance metrics used for classification were False Positive Rate (FPR), Sensitivity, Specificity, Re, cognition Accuracy and Recognition Time at 0.70 threshold value.The Recognition Accuracy (RA), Recognition Time (RT), False Acceptance Rate (FAR), Sensitivity and Specificity of the three selected techniques (CPNN, SOFM and BPNN) resulted in values of 95.17%, 177.48s, 6.33% and 93.67% for CPNN 92.50%, 179.69s, 9.00%, 94.00% and 90.99% for SOFM while BPNN had 91.17%, 187.88s, 10.33%, 92.67% and 89.67% respectively.This paper showed that CPNN classification technique performed best for iris recognition system in terms of RA and recognition time. This research output will serve as a basis to pre-inform and guide researchers in choosing an efficient kernel based feature extraction technique.

Classification of Longan Types Using The Back-Propagation Neural Network Algorithm Based on Leaf Morphology With Shape Characteristics

In Indonesia, longan fruit is abundantly accessible. The Longan fruit comes in a number of kinds, and they vary in terms of their morphology, including the characteristics of their leaves. The varieties of longan fruit are to be categorized in this study based on the shape of the leaves. The author uses the RGB color extraction function, the Grey Level Co-occurrence Matrix (GLCM), and the Shape feature to get data for each cultivar. The accuracy value is then processed using the Back-Propagation Neural Network (BPNN) technique to determine the accuracy value that will be used as a determinant of the categorization of the Longan leaf image. The eccentricity and metric parameters are key components of the method. The BPNN algorithm demonstrated its usefulness for categorizing various kinds of longan fruit leaves during testing by obtaining an accuracy of 70%.

Short Term Load Forecasting under extensive Power Outages using Domestic Energy Meter Load Profile; A case study

Short-term load forecasting (STLF) is an obligatory and vibrant part of power system planning and dispatching. It is utilized for short and running targets in power system planning. Electricity consumption has nonlinear patterns due to its reliance on factors like time, weather, geography, culture, and some random and individual events. This research work emphasizes STLF through utilized load profile data from domestic energy meter and forecasts it by Multiple Linear Regression (MLR) and Cascaded Forward Back Propagation Neural Network (CFBP) techniques. First, simple regression statistical calculations were used for prediction, later the model was improved by using a neural network tool. The performance of both models compared with Mean Absolute Percent Error (MAPE). The MAPE error for MLR was observed as 47% and it was reduced to 8.9% for CFBP.

The Physiological Restorative Role of Soundscape in Different Forest Structures

Natural soundscape is considered a dominant type of hearing in forested areas and contributes to health and recovery effects from exposure to the biophilic outdoor environment. This study focuses on the different forest structures, and aims to explore the relationship between perceived soundscape and acoustical parameters, observe physiological indicators, and model the physiological restorative role of soundscape. Questionnaires and measuring equipment were used to gather psychophysical and physiological information at 20 observation sites in urban forested areas. Back-propagation neural network techniques were conducted to determine the forecasting model from psychophysical to physiological parameters. Our results suggested that LAeq and L10 are important factors that influence questionnaire responses. Our findings also showed that electromyogram (EMG) signals were the most obvious and sensitive in physiological parameters. Additionally, we found that L10–90 played the most important role among all physical parameters in the physiological restorativeness soundscape model. This can facilitate the understanding of the physiological restorative role of soundscape in different forest structures when proposing suitable forest-based health care strategies.

Neuro-Fuzzy-AHP (NFAHP) Technique for Copper Exploration Using Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and Geological Datasets in the Sahlabad Mining Area, East Iran

Fusion and analysis of thematic information layers using machine learning algorithms provide an important step toward achieving accurate mineral potential maps in the reconnaissance stage of mineral exploration. This study developed the Neuro-Fuzzy-AHP (NFAHP) technique for fusing remote sensing (i.e., ASTER alteration mineral image-maps) and geological datasets (i.e., lithological map, geochronological map, structural map, and geochemical map) to identify high potential zones of volcanic massive sulfide (VMS) copper mineralization in the Sahlabad mining area, east Iran. Argillic, phyllic, propylitic and gossan alteration zones were identified in the study area using band ratio and Selective Principal Components Analysis (SPCA) methods implemented to ASTER VNIR and SWIR bands. For each of the copper deposits, old mines and mineralization indices in the study area, information related to exploration factors such as ore mineralization, host-rock lithology, alterations, geochronological, geochemistry, and distance from high intensity lineament factor communities were investigated. Subsequently, the predictive power of these factors in identifying copper occurrences was evaluated using Back Propagation Neural Network (BPNN) technique. The BPNN results demonstrated that using the exploration factors, copper mineralizations in Sahlabad mining area could be identified with high accuracy. Lastly, using the Fuzzy-Analytic Hierarchy Process (Fuzzy-AHP) method, information layers were weighted and fused. As a result, a potential map of copper mineralization was generated, which pinpointed several high potential zones in the study area. For verification of the results, the documented copper deposits, old mines, and mineralization indices in the study area were plotted on the potential map, which is particularly appearing in high favorability parts of the potential map. In conclusion, the Neuro-Fuzzy-AHP (NFAHP) technique shows great reliability for copper exploration in the Sahlabad mining area, and it can be extrapolated to other metallogenic provinces in Iran and other regions for the reconnaissance stage of mineral exploration.

Construction of Mobile Internet Financial Risk Cautioning Framework Based on BP Neural Network

With the emergence of the 21st-century global economy, the international financial system faces economic risks. A competitive cautioning model for financial management is required to mitigate risks and losses in the financial sector. The financial losses of the banking industry have been categorized and analyzed using the Internet of Things (IoT) and big data technologies to minimize the economic risk of commercial banks in mobile internet finance (MIF). This article proposes a new financial risk cautioning framework (FRCF) based on the IoT, big data, and back propagation-neural network (BP-NN) to ensure steady growth of MIF in the long term. In this article, a big data technology-based approach for data recognition and mining has been suggested. A BP-NN-based method for risk identification and assessment in MIF is also presented. The BP-NN technique calculates each neural network (NN) layer’s node count, transfer functions, learning rate, and other characteristics. The proposed FRCF has been developed through the proper construction, analysis, and testing of many information samples. A conceptual understanding of the use of IoT, big data, and artificial intelligence (AI) technologies through NN models in the financial industry has been described in the article. The proposed FRCF can predict the MIF risks associated with the MIF lending infrastructure with a 98.2% accuracy.

Fault classification of power plants using artificial neural network

ABSTRACT In order to run power plant operations smoothly, power plant faults need to be detected, located, and classified quickly. For this, artificial neural network approaches are considered significant tool in related application of power system. This research paper focuses on power plant faults detection and classification using intelligent approach of artificial neural network. There is frequent nature of power plant faults, in particular transmission line faults. The key intention of this research is to use ANN approach to detect and classify power plant faults, and in the end to compare their performance and accuracy. In this proposal, the three-phase current and voltage are use as input. Backpropagation (BPNN) technique is used in developing the learning algorithm for multi-layer neural network that can accurately and reliably classify faults. The multi-layer neural network such as Levenberg-Marquardt and Bayesian regularization are used for offline training of the data. Findings from this research show that both Levenberg-Marquardt and Bayesian regularization show good accuracy to detect and classify faults in transmission line. However, LM algorithm trained data much faster than BR. LM shows fastest detection of faults with overall mean square error value of 6.13785e-3 at the 91 epoch. Whereas, BR trained network at the 717 epochs with overall mse of 4.67123e-3. It is to be noted that the accuracy of trained network to classify all the fault types is found to be 81.1%, which is higher than previous studies using Backpropagation neural network technique. Training and test results show that these neural network-based methods can effectively detect and classify faults and faults types and have adequate performance. Matlab is used in this research because it has toolbox, which has effective features to train required data and achieve effective result.

Short Term Renewable Energy Forecasting Based on Feed Forward Back Propagation Neural Network Strategy

The fundamental inputs used as a renewable energy source are wind speed and solar radiation. Both parameters are very nonlinear and depending on their surroundings. As a result, reliable prediction of these characteristics is required for usage in a variety of agricultural, industrial, transportation, and environmental applications since they reduce greenhouse gas emissions and are environmentally benign. In this study, we used a Feed Forward Back Propagation Neural Network (FFBPN) technique to predict proper data such as temperature, relative moisture, sun radiations, rain, and wind speed. The FFBPN will be trained in such a way that it can conduct hybrid forecasting with little changes to the programming codes, ranging from hourly (short term forecasting) to daily forecasting (medium term forecasting). This feature is one of the significant improvements, showing the suggested hybrid renewable energy forecasting system's high robustness. Because the hybrid forecasting system is a unique approach, the system's accuracy will be determined by comparing the findings to the corresponding values of the persistent model, a stand-alone forecasting model. Finally, the completely created system package could be sold and/or used in future research initiatives to help researcher’s analyses, validate, and illustrate their models across a variety of areas.

Standalone Smart Irrigation System: An Analysis of Reliability Incorporating BPNN Technique under Various Failures

Observing the huge requirements of water and energy for traditional irrigation and contracting resources, the utility of a smart irrigation system has become necessary. A system that is capable of saving water depends on renewable energy resources for power requirements and runs automatically. Water-saving automatic irrigation systems powered by renewable energy could be a solution to India's rising demand for energy and water. In the current paper, the author has proposed an analysis of a standalone smart irrigation system for its reliability and functional behaviour under various operational events of failures and repairs. The authors have implemented an intelligent computational method because of the increased degradation in system function due to the complex structure of the proposed irrigation system. The system's mathematical formulation is stated in terms of neural networks, and a back propagation neural network technique is applied for fast and comparatively better outcomes. The author has determined state probabilities and other reliability parameters by means of neural networks in the proposed smart irrigation system. To improve the accuracy and consistency of reliability parameters, a Feed Forward Back Propagation Neural Network (FFBPNN) is applied. FFBPNN's learning mechanism can optimize the values of parameters by modifying neural weights. The MATLAB codes are used by the authors to demonstrate the numerical examples, and iterations are repeated until the precision in error tends to 0.0001. The sensitivity and cost of the system are also analyzed, which can help in managing the real-time operations of the system.

Backpropagation and Radial Basis Function Methods for Predicting Rainfall in Sukabumi City Using Artificial Neural Networks: A Comparative Analysis

The weather has a substantial impact on the ability to live organisms to carry out everyday activities, particularly outside activities. Weather data is helpful in various fields, including marine, aviation, and agriculture. The maritime domain is beneficial for establishing the optimal navigation time for a fisherman, the aviation domain helps reduce climate-related mishaps, and the agriculture sector uses weather information to develop harvest season models for agricultural products. Indonesia is a tropical nation with heavy precipitation. Utilized for various objectives, rainfall forecasting models seek the utmost precision, particularly in specialized areas such as flood control. This study is based on two techniques: the Radial Basis Function Neural Network (RBFNN) and Backpropagation Neural Network (BPNN) techniques using multiple training functions. The RBFNN approach yields less accurate results for predicting precipitation, but the multi-practice BPNN method yields more accurate results.