differential evolution algorithm

Water supply networks are one of the most important urban infrastructures for supplying water. Considering that currently water wastage is a global concern and on the other hand the demand for water is increasing, this issue has made it necessary to manage the demand and modify the consumption pattern. One of the important components of non-revenue water is leakage in the water supply network, and leak detection is one of the necessary measures to reduce non-revenue water and manage consumption. In this research, an optimization formulation has been developed for the purpose of leak detection in water networks assuming the lack of information on the number of leaks and pressure measurement data, and the search problem has been solved with the differential evolution algorithm. The performance of the developed model has been investigated by defining different scenarios in terms of location, amount and number of leaks. First, the location scenarios were examined in terms of the number and amount of leaks, including one, ten, and twenty leaks at the same time, and then the developed model was implemented for location scenarios with an unknown number of leaks and the uncertainty of nodal needs. The results showed that the success of the model in the case of the certainty of the input data and the existence of a node is 100%, and by considering the hourly changes in the nodal demand and increasing the number of leaks up to ten and twenty leak nodes, the success rate of the model in finding the exact leak points is 95% and 94.5% has been obtained. In the scenarios where the number of leaks was considered unknown, the success of the model to find the number of leaks is 94%. The success of the model in the case of uncertainty of nodal requirements with the number of known leaks reaches 91% with the increase of leaks and 86% with the number of unknown leaks.

Banks' inability to credit assessment and financial evaluation of customers and forecasting accurately the credit risk of borrowers has devastating effects on the global financial system and economic activity and have been the main causes of global financial crises in recent years.The purpose of this paper is to compile a credit forecasting model for legal customers of private banks by using meta-heuristic algorithms in the branches of Pasargad Bank in the northwest of Iran.

This research is base on the purpose of developmental research and based on the method of performing descriptive work. The statistical population of this study is in two sections of banking experts and legal customers of Pasargad Bank in the northwest of the Iran. The statistical sample size for the first community of 58 banking experts including managers, credit officials and heads of branches in with credit work experience in private banks and for the second community, 427 legal clients were selected based on targeted sampling. In order to collect data in this research, a questionnaire and documents of Pasargad Bank have been used. The validity of the questionnaire was investigated as content validity and based on the indicators of content validity ratio and content validity index. The reliability of the questionnaire was assessed using Cronbach's alpha coefficient. In order to analyze the research data, t-test, confirmatory factor analysis, multilayer neural network, genetically trained neural network, trained neural network with particle swarm optimization and trained neural network with differential evolution will be used.

 The research findings show that all four models are able to predict the credit predictions of the legal customers of private banks and the best way to predict the credit predictions of the legal customers of private banks is the neural network trained with differential evolution algorithm with the least amount of error compared to the other three methods.Originality/Value: In this research by using meta-heuristic algorithms, a new credit forecasting model produce for legal customers of private banks with the least amount of error.

Considering the importance of heat exchangers in operational units, in this research, the differential evolution (DE) algorithm was used to optimize compact heat exchanger aim to minimize the number of entropy generation units. The number of total entropy generation units of heat and pressure in the compact heat exchanger as an objective function and six decision variables including heat exchanger length, fin frequency, fin length, number of layers of flow on both sides of the offset strip fin, fin height, and fin thickness with a series of restrictions were Investigated. The results obtained from DE optimization were compared and validated by genetic algorithm (GA) and particle swarm optimization (PSO). The level of this algorithm in minimizing entropy generation (by 1.5 and 17.6% compared to the PSO method and GA) revealed high ability of the DE method in optimizing compact heat exchangers. Moreover, the effect of changes in control parameters in the DE algorithm on convergence and the degree of optimization of the objective function was studied

One of the main challenges in speech recognition is noise resistant feature extraction. In this paper, a new feature extraction algorithm, called Fractional and Adaptive Power Normalized Cepstral Coefficients Algorithm, has been proposed as a noise-resistant method for speech recognition. This proposed feature extraction method is based on a fractional short-term Fourier Transform. The selection of fractional conversion coefficient is important for proper analysis of multi-component signals like speech. Therefore, the proposed method obtains the optimum parameter of α for fractional Fourier Transform based on the noise class in the environment, adaptively by the Differential Evolution meta-heuristic algorithm. Moreover, TI Digit and Noisex-92 are used for evaluation of the resistance and accuracy of the recognition of the automatic speech recognition system. Simulation results show more resistance and higher recognition accuracy of the proposed feature extraction method rather than other methods in noisy and without noise environments. In the proposed ASR system, the Support Vector Machine (SVM) classifier with a nonlinear kernel has been used. Also, all the simulations are performed in MATLAB.

The Performance of shooting pool, in addition to the quality of the area in which the flow collides with it, depends to the height of the jet drop, the angle of the water flow, the depth of the jet and the concentration of the jet. By increasing the height of the jet drop, the fall velocity increases and subsequently the jetchr('39')s energy will be more intrusive. Different collision area from the scouring phenomenon that considered in the design step, may lead to damages to the structure of the dam. For this purpose, an Adaptive Neural Fuzzy Inference System (ANFIS) model and Differential Evolution Algorithm were used to optimize the geometry of the flip bucket spillway to reduce the scour depth of the jet. To construct a fuzzy inference system and calculate the scour depth based on available input/output patterns, the ANFIS fuzzy-inductive inference system was used. ANFIS model was used to train input and output data using laboratory data. Finally, using the differential evolution algorithm and also the ANFIS model output, the geometry of the flip bucket spillway has been optimized by defining the objective function for minimizing the depth of the scour hole in the downstream of the spillway.

With the rapid growth of wind power plants, different types of these resources are connected to the network at different voltage levels. A number of these resources have DFIG. The development of distributed generation has made problems such as an increased level of short circuit and the loss of protective devices coordination. During the fault occurrence, transient stability may be lost due to drastic changes in speed, power, and rotor angel of synchronous and asynchronous generators. Based on the dependence of resistance and temperature, high-temperature superconducting fault current limiter could decrease the fault current. This paper, after studying the ability of HTSFCL to reduce the fault current and to increase the stability level in PSCAD/EMTDC software, for the first time proposes objective function of HTSFCL optimal location by simultaneously considering security and protection coordination indices, voltage, and transient stability of synchronous and asynchronous generators in the networks connected to DFIG. Optimal location and size of HTSFCL in IEEE 30-bus test network are determined using differential evolution (DE) algorithm in MATLAB while considering single- and multi-objective indices. Also, indices of security, coordination of overcurrent relays, voltage stability, and rotor angel stability are compared in six modes. The achieved results show the applicability of the proposed method.

Weirs are used to adjust and measure the flow within open conduit. Also, weirs are applied in various shapes like rectangular, triangular, circular and triangular plan forms. In this study, the discharge coefficient of triangular plan form weirs was predicted using a hybrid model based on fuzzy systems (ANFIS) and differential evolution (DE) algorithm. Adaptive neuro Also, to examine the performance of hybrid models, the Monte Carlo simulations (MCs) were used. In this research, the k-fold Cross Validation with k equal to 5 was used for evaluation of the models ability. Then, the six hybrid models were defined using input parameters. According to simulation results, the superior model predicted the discharge coefficient in terms of Froude number, weir vertex angle, the ratio of weir length to weir height, the ratio of head over the crest of the weir to weir height and channel width to weir length. For superior model, the Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE) and correlation coefficient (R) were calculated equal to 1.644, 0.016 and 0.972, respectively. Also, the Froude number was identified as the most effective parameter to model the discharge coefficient of triangular plan form weirs