Evolutionary Programming Research Articles

Shuffled Frog-Leaping Programming for Solving Regression Problems

There are various automatic programming models inspired by evolutionary computation techniques. Due to the importance of devising an automatic mechanism to explore the complicated search space of mathematical problems where numerical methods fails, evolutionary computations are widely studied and applied to solve real world problems. One of the famous algorithm in optimization problem is shuffled frog leaping algorithm (SFLA) which is inspired by behaviour of frogs to find the highest quantity of available food by searching their environment both locally and globally. The results of SFLA prove that it is competitively effective to solve problems. In this paper, Shuffled Frog Leaping Programming (SFLP) inspired by SFLA is proposed as a novel type of automatic programming model to solve symbolic regression problems based on tree representation. Also, in SFLP, a new mechanism for improving constant numbers in the tree structure is proposed. In this way, different domains of mathematical problems can be addressed with the use of proposed method. To find out about the performance of generated solutions by SFLP, various experiments were conducted using a number of benchmark functions. The results were also compared with other evolutionary programming algorithms like BBP, GSP, GP and many variants of GP.

PSO-Based Oscillatory Stability Assessment by Using the Torque Coefficients for SMIB

This study discusses the evaluation of oscillatory stability based on the synchronizing K s and damping K d torque coefficients for a single-machine system connected to an infinite bus (SMIB). Particle swarm optimization (PSO) technique is used to determine K s and K d values and subsequently identify the oscillatory stability conditions in the SMIB. The ability of PSO is compared with those of evolutionary programming (EP) techniques and artificial immune system (AIS). The least square (LS) method is selected as the benchmark for K s and K d values determined by PSO, EP, and AIS. Simulation results show that PSO successfully estimated K s and K d values closest to LS compared with EP and AIS. PSO also uses lower computational time compared with those of the two other techniques. The proposed technique is suitable for solving oscillatory stability problems based on the determination of eigenvalues and minimum damping ratio.

Supervised evolutionary programming based technique for multi-DG installation in distribution system

<span lang="EN-US">Installing DG in network system, has supported the distribution system to provide the increasing number of consumer demand and load, in order to achieve that this paper presents an efficient and fast converging optimization technique based on a modification of traditional evolutionary programming method for obtain the finest optimal location and power loss in distribution systems. The proposed algorithm that is supervised evolutionary programming is implemented in MATLAB and apply on the 69-bus feeder system in order to minimize the system power loss and obtaining the best optimal location of the distributed generators. </span>

Evolving semantic object segmentation methods automatically by genetic programming from images and image processing operators

Even though numerous segmentation methods exist, the requirement of prior knowledge or parameter tuning makes them restricted to limited image domains. Without predefining solution models, genetic programming (GP) is able to solve complex problems by evolving computer programs automatically. In this paper, three new GP-based methods are designed to evolve segmentation algorithms automatically from images and primitive image processing operators (e.g., filters and histogram equalization). Specifically, a strongly typed representation, the cooperative coevolution technique and a two-stage evolution are introduced in GP, respectively, to form three new methods that can evolve solutions to conduct image preprocessing, segmentation and postprocessing automatically. The new methods are termed as StronglyGP, CoevoGP and TwostageGP, and standard GP-based algorithm (StandardGP) is employed as a reference method. The proposed methods are tested on two complicated datasets (i.e., Weizmann and Pascal datasets), which contain high variations in both objects and backgrounds. The results show that StronglyGP and StandardGP can evolve effective segmentors for the given complex segmentation tasks, while CoevoGP and TwostageGP perform worse than StronglyGP and StandardGP, which may be caused by the overfitting problem in deriving postprocessing solutions. In addition, compared with StandardGP, StronglyGP achieves better segmentation performance with smaller solution sizes. Moreover, compared with four widely used segmentation methods, StronglyGP and StandardGP can produce satisfactory results consistently on both Weizmann and Pascal datasets.

Implementation of Optimal Protection Coordination for Microgrids with Distributed Generations Using Quaternary Protection Scheme

The flexible operation of microgrids, along with the availability of distributed generation (DG) units, causes a variety of changes in short-circuit current levels, magnitudes, and directions, which has undesirable effects on the operation of protection systems. Conventional protection schemes use typical directional overcurrent relays (DOCRs) with limited operating capability, unable to respond to microgrid operations in the manner of short-circuit current changes. In this paper, a quaternary protection scheme implemented with dual-directional overcurrent relays (dual-DOCR) and a protection control strategy is developed for protecting against faults in microgrids, taking into account the existence of DGs and connection and disconnection of DG units. The optimal dual-DOCRs setting and coordination are formulated as an optimization problem solved by evolutionary programming to minimize the relay operating times. The proposed protection scheme is implemented with a centralized protection control system based on the smart grid concept to increase the adaptability of the dual-DOCRs, which have multiple relay setting groups in accordance with system state changes. The simulation case studies are performed using the IEEE 14-bus test system, which is modified as a meshed microgrid operation. Test scenarios, including possible operations of microgrids, DGs availabilities, and different fault events, are analyzed and discussed. The comparative studies and simulation results show that the proposed scheme provides efficient coordination between the primary and backup relays and increases the responsibility of the protection system, which can be observed from the significant reduction in the relay operating times, resulting in the enhancement of selectivity, sensitivity, and speed of microgrid protection systems.

Multi cases optimal reactive power dispatch using evolutionary programming

<span>Evolutionary Programming (EP) is one of many types in Evolutionary Computation (EC) that used for optimization process. EP technique is used to find the optimal reactive power dispatch (ORPD) since it is one of the accessible options schemes that can be used on the system as a reactive power support. Sometimes, it is not necessary to operate all generators in order to perform ORPD to in achieve the objectives. Also, increment of reactive power load to the system will cause voltage decomposes with the increase in transmission loss in the system. Therefore, the proposed method decides the best grouping of generators that should be operated in system by bearing in mind the transmission loss reduction. ORPD will be used to minimize the transmission loss as well the increasing reactive power loading. This method conducted on IEEE 30-bus Test System with multi cases scenario. The best combination of operating generators determined and the transmission loss after optimization is smaller compared to the transmission loss before optimization resulted.</span>

Optimization of a Linear Fresnel Reflector Applying Computational Fluid Dynamics, Entropy Generation Rate and Evolutionary Programming

This work presents an optimization of a Linear Fresnel Reflector based on the Computational Fluid Dynamics, the Entropy Generation Rate and the Evolutionary Programming method. The objective function of the optimization process takes into account the maximization of the absorbed radiation solar flux on the receiver tube and the minimization of the total Entropy Generation Rate. A set of design equations were used to build the Linear Fresnel Reflector geometries of each one of the individuals per generation. The design equations consider, among others, a coupling between the angles and distances of the mirrors and the required geometrical parameters for the construction of the CPC secondary reflector. The Evolutionary Programming considers a small population of six individuals per generation and takes into account a search space for geometric parameters such as the aperture area, the width and the length of the mirrors. The mutation operator is applied to generate the individuals and the selection operator is applied to find the best individuals for the next generation. Seven generations were needed to find the optimal Linear Fresnel Reflector. The optimal Linear Fresnel Reflector (NN individual) presents an increase of 2.48% for the average absorbed radiation flux on the absorber tube and a decrease of 20% for the total Entropy Generation Rate, both in comparison with a prototype of a Linear Fresnel Reflector. For the absorbed radiation flux, both individuals presents the minimum values on the top side of the absorber tube (1,386 W m-2 and 1,982 W m-2 for the prototype and NN individual respectively), while the maximum values are located at the lower part of the absorber tube (7,180 W m-2 and 8,199 W m-2 for the prototype and NN individual respectively). In terms of local Entropy Generation Rate, the NN individual has a decrease of 14.6%, 60% and 36.8% of Entropy Generation Rate due to viscous dissipation, heat transfer and radiation respectively at the CPC zone in comparison with the prototype of a Linear Fresnel Reflector. Finally, the NN individual has an increase of 16.4% and 23.8%. for the thermal efficiency and the exergy efficiency, respectively.

Comparison study on nature-inspired optimization algorithms for optimization back analysis of underground engineering

Optimization back analysis is the most common approach to displacement back analysis for underground engineering. However, this is a non-convex problem that requires the use of nature-inspired global optimization algorithms. Therefore, the present study will investigate on the suitability of six state-of-the-art nature-inspired algorithms for elastic back analysis and elastic–plastic back analysis. These algorithms include improved genetic algorithm, immunized evolutionary programming, particle swarm optimization, continuous ant colony optimization, artificial bee colony and black hole algorithm. Numerical results indicate that immunized evolutionary programming is overall the best algorithm followed by the black hole algorithm; while, the improved genetic algorithm is the worst optimizer. Meanwhile, using elastic back analysis, the sensitivity analysis of the main input parameters for these nature-inspired optimization algorithms has been conducted. At last, the comparative results have been verified by using in one real underground roadway in Huainan coal mine of China.

Optimal TCSC Allocation via Chaotic Immune Symbiotic Organisms Search for Voltage Profile Improvement

As the load demand in a power system increases, power system operators struggle to maintain the power system to be operated within its acceptable limits. If no mitigation actions are taken, a power system may suffer from voltage collapse, which in turn leads to blackout. Flexible AC Transmission System (FACTS) devices can be employed to help improve the voltage profile of the power system. This paper presents the implementation of Chaotic Immune Symbiotic Organism Search (CISOS) optimization technique to solve optimal Thyristor Controlled Series Compensator (TCSC) in a power system for voltage profile improvement. Validation process are conducted on IEEE 26-bus RTS resulting in the capability of CISOS in solving the allocation problem with a better voltage profile. Comparative studies conducted with respect to Particle Swarm Optimization (PSO) and Evolutionary Programming (EP) has revealed the superiority of CISOS over PSO and EP in solving the optimal allocation problem by producing optimal solution with a better voltage profile. The results and information obtained from this study can help power system operator in terms of optimal compensation in power system as well as improving the operation of a power system.

Intelligent based technique for under voltage load shedding in power transmission systems

<p>The increasing demand of electric power energy and the presence of disturbances can be identified as the factors of voltage instability condition in a power system. A secure and reliable power system should be considered to ensure smooth delivery of electricity to the consumers. A power system may experience undesired event such as voltage instability condition leading to voltage collapse or cascading collapse if the system experiences lack of reactive power support. Thus, to avoid blackout and cascaded tripping, load shedding is the last resort to prevent a total damage. Under Voltage Load Shedding (UVLS) scheme is one of the possible methods which can be conducted by thepower system operators to avoid the occurrence of voltage instability condition. This paper presents the intelligent based technique for under voltage load shedding in power transmission systems. In this study, a computational based technique is developed in solving problem related to UVLS. The integration between a known computational intelligence-based technique termed as Evolutionary Programming (EP) with the under-voltage load shedding algorithm has been able to maintain the system operated within the acceptable voltage limit. Loss and minimum voltage control as the objective function implemented on the IEEE 30-Bus Reliability Test System (RTS) managed to optimally identify the optimal location and sizing for the load shedding scheme. Results from the studies, clearly indicate the feasibility of EP for load shedding scheme in loss and minimum voltage control in power system.</p>

Computational Intelligence Based Technique for Multi-DG Installation in Transmission System

The increasing electricity demand in transmission system has caused the power transmission system experiencing stress condition. This phenomenon has forced the system to need urgent additional supply to maintain system adequacy, in particular the de-regulated power system environment. Distributed generation (DG) has been identified as one of the possible solutions to address this issue. DG installation has the capability to reduce transmission loss and improve the voltage profile. This paper presents evolutionary programming (EP) technique for optimizing the sizing and locations in DG installation. In this study, several DGs have been installed to address the voltage profile improvement and loss minimization; implemented on the IEEE 30-Bus Reliability Test System (RTS). Results obtained from the study revealed that, installation of multi-DGs in a transmission system has significantly minimized the transmission loss along with voltage profile improvement.

Multi-Objective Evolutionary Programming for Developing Recommender Systems based on Collaborative Filtering

In the era of internet, several online platforms offer many items to users. Users could spend a lot of time to find (or not) some items they are interested, sometimes, they will probably not find the desired items. An effective strategy to overcome this problem is a recommender system, one of the most popular applications of machine learning. Recommender systems select most appropriate items to an specific user based on previous information between items and users, and they are developed using diffeent approaches. One of the most successful approach for developing recommender systems is collaborative filtering, which can filter out items that a user might like based on reactions of users with similar profiles. Often, traditional recommender systems only consider precision as evaluation metric of performance, however, others metrics (like recall, diversity, novelty, etc) are also important. Unfortunately, some metrics are conflicting, e.g., precision impacts negatively on other metrics. This paper presents a multi-objective evolutionary programming method for developing a recommender system, which is based on a new collaborative filtering technique, while maximizes the recall for a given precision, The new collaborative filtering technique uses three components for recommending an item to a user: 1) clustering of users; 2) a previous memory-based prediction; and 3) five decimal parameters (threshold average clustering, threshold penalty, threshold incentive, weight attached to average clustering and weight attached to Pearson correlation). The multiobjective evolutionary programming optimizes the clustering of users and the five decimal parameters, while, it searches maximizes both similarity precision and recall objectives. A comparison between the proposed method and a previous nonevolutionary method shows that the proposed method improves precision and recall metric on a benchmark database.

Two New Improved Variants of Grey Wolf Optimizer for Unconstrained Optimization

Grey wolf optimization (GWO) algorithm is a relatively recent and novel optimization approach. GWO showed performance improvement over all competing algorithms. However, the relevant literature identified that the primary GWO due to its position update equation shows superiority in exploitation, but is inefficient in exploration. It shows slow convergence and low precision, too. Motivated by the outlined issues in the primary GWO, this work presents two new and improved GWO algorithms. The first proposed variant modifies all the three models, encircling model of prey, position update equation and the hunting equation of canonical GWO. Further, a new parameter is introduced to scale the encircling and position update equations. As a result, the exploration issue of the algorithm is tackled. Unlike the first variant, the second proposed variant does not modify the position update models, but it incorporates Minkowski's information into GWO. To the best of our knowledge, no such modifications to GWO have been done before. The proposed modified versions of GWO are tested on a well-known test functions suit and then compared with different population-based algorithms, including fast evolutionary programming and particle swarm optimization. It was identified from the simulation results that proposed algorithms outperform different algorithms in comparison on majority of problems. The sensitivity study of the proposed algorithms to their various parameters is also provided.

Multi-area economic dispatching using improved grasshopper optimization algorithm

In this study a new optimization algorithm is presented to solve the multi-area economic dispatching (MAED) problem. The studied system includes the tie-line constraints including transmission losses, prohibited operating zones, multiple fuels, and valve-point loading. For optimized solving the MAED algorithm, grasshopper optimization algorithm (GOA) is utilized. GOA is a newly introduced swarm-based optimization algorithm that is inspired by the swarming behavior of the grasshopper insects. For validating the proposed method, it is compared with three different case studies and the results have been compared with some different meta-heuristics such as particle swarm optimization, artificial bee colony, evolutionary programming, and differential evolution to demonstrate the capability of the presented system to solve the MAED problems.

Multi-Objective Evolutionary Programming for Solving Economic Dispatch Problem

Multi-Objective Evolutionary Programming for Solving Economic Dispatch Problem

Simulated Annealing Meta-heuristic for Addition Chain Optimization

In this work, a simulated annealing (SA) algorithm is implemented in the Python programming language with the aim of minimizing addition chains of the "star-chain" type. The strategies for generating and mutating individuals are similar to those used by the evolutionary programming (EP) and genetic algorithms (GA) methods found in the literature [1]-[3]. The proposed variant is the acceptance mechanism that is based on the simulated annealing meta-heuristic (SA). The hypothesis is that with the proposed acceptance mechanism, diversity is obtained in the search-space through a simple strategy that allows finding better solutions compared to the deterministic method Optimized Window. The simulations were performed with exponents in the range 218-234 and were compared with the results reported in [3], where a GA is proposed to get optimal addition chains. It is concluded that the proposed algorithm is able to find chains of shorter length than those found with the Optimized Window method and with a performance similar to that of the GA proposed in [3].

Intelligent swarm-based optimization technique for oscillatory stability assessment in power system

<p>This paper discussed the prediction of oscillatory stability condition of the power system using a particle swarm optimization(PSO) technique. Indicators namely synchronizing(<em>K<sub>s</sub></em>)and damping(<em>K<sub>d</sub></em>) torque coefficients is appointed to justify the angle stability condition in a multi-machine system. PSO is proposed and implemented to accelerate the determination of angle stability. The proposed algorithm has been confirmed to be more accurate with lower computation time compared with evolutionary programming(EP) technique. This result also supported with other indicators such as eigenvalues determination, damping ratio and least squares method. As a result, proposed technique is achievable to determine the oscillatory stability problems.</p>

Multi-verse optimization based evolutionary programming technique for power scheduling in loss minimization scheme

<span style="font-size: 9pt; font-family: 'Times New Roman', serif;" lang="EN-US">The growth of computational intelligence technology has witnessed its application in numerous fields. Power system study is not left behind as far as computational intelligence trend is concerned. In power system community, optimization process is one of the crucial efforts for most remedial action to maintain the power system security. Basically, power scheduling refers to prior to fact action(such as scheduling generators to generate certain powers for next week). Power scheduling process is one of the most important routines in power systems. Scheduling of generators in a power transmission system is an important scheme; especially its offline studies to identify the security status of the system. This determines the cost effectiveness in power system planning. This paper investigates the performance of multi-verse based evolutionary programming(lowest EP) technique in the application of power system scheduling to ensure loss is gained by the system. Losses in the system can be controlled through this implementation which can be realized through the validation on a chosen reliability test system as the main model. Validation on </span><span style="font-size: 10.0pt; font-family: 'Times New Roman','serif'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;" lang="EN-US">IEEE 30-Bus Reliability Test System resulted that both techniques are reliable and robust in addressing this issue.</span><p class="MsoTitle"> </p>

Dynamic economic dispatch: a comparative study for differential evolution, particle swarm optimization, evolutionary programming, genetic algorithm, and simulated annealing

This paper presents a comparative study for five artificial intelligent (AI) techniques to the dynamic economic dispatch problem: differential evolution, particle swarm optimization, evolutionary programming, genetic algorithm, and simulated annealing. Here, the optimal hourly generation schedule is determined. Dynamic economic dispatch determines the optimal scheduling of online generator outputs with predicted load demands over a certain period of time taking into consideration the ramp rate limits of the generators. The AI techniques for dynamic economic dispatch are evaluated against a ten-unit system with nonsmooth fuel cost function as a common testbed and the results are compared against each other.

Multi Gene Genetic Program Modelling on L-Asparaginase Activity of Bacillus Stratosphericus

Abstract The current study focuses on maximization of L-Asparaginase production from Bacillus stratosphericus isolated from Ocimum tenuiflorum. Optimization study followed by modelling using Artificial Neural Network (ANN) was performed. The experimental data obtained from Response Surface Methodology (RSM) was further studied by an evolutionary algorithm Genetic Programming (GP) to find the prediction equation. GP does not require prior knowledge of the data sets. GP is an extension of Genetic Algorithm (GA), where the results are represented in the form of trees. Multi gene genetic programming (MGPP) is a variant of GP used to solve non-linear mathematical models. The prediction equation obtained from the GP analysis is represented in the form of tree. Each tree represents single gene. Best fit individuals obtained at each generation by using genetic operators were selected to get better regression co-efficient value. The predicted and experimental data showed good significance with R2 = 0.99956.