Evolutionary Techniques Research Articles

An efficient optimization strategy for vapor compression–absorption-based cascaded refrigeration system using various evolutionary techniques

ABSTRACT A novel optimization strategy for a compression-absorption cascaded refrigeration system (CACRS) consisting of a single effect vapor absorption system (VARS) and vapor compression system (VCRS), which employs several absorbent solution-refrigerant-refrigerant pairs to consider the combination of absorbent solution-refrigerant in VARS and refrigerants in VCRS as an optimization variable, is successfully implemented using various evolutionary optimization techniques. This article analyzes three different absorbent solution-refrigerant in the VARS with eleven different refrigerants in the VCRS. A nonlinear objective function is formulated based on exergy and environmental analysis, which is minimized using 10 different evolutionary optimization techniques. The proposed novel strategy reduces the complexity of using optimization techniques to select the best optimum absorbent-refrigerant-refrigerant pair. Differential evolution with multi-population-based ensemble of mutation strategies determined the minimum total annual cost as 13,137.78 US$/year for (CaCl2-LiBr-LiNO3-)-H2O-R152a. Lower condenser temperature and higher temperature difference in the cascade-condenser resulted in a minimum value for the total annual cost.

A grammatical evolution approach to the automatic inference of P systems

P systems are a bio-inspired framework for defining parallel models of computation. Despite their relevance for both theoretical and application scenarios, the design and the identification of P systems remain tedious and demanding tasks, requiring considerable time and expertise. In this work, we try to address these problems by proposing an automated methodology based on grammatical evolution (GE)—an evolutionary computation technique—which does not require any domain knowledge. We consider a setting where observations of successive configurations of a P system are available, and we rely on GE for automatically inferring the P system, i.e., its ruleset. Such approach directly addresses the identification problem, but it can also be employed for automated design, requiring the designer to simply express the configurations of the P system rather than its full ruleset. We assess the practicability of the proposed method on six problems of various difficulties and evaluate its behavior in terms of inference capability and time consumption. Experimental results confirm our approach is a viable strategy for small problem sizes, where it achieves perfect inference in a few seconds without any human intervention. Moreover, we also obtain promising results for larger problem sizes in a human-aided context, paving the way for fully or partially automated design of P systems.

An evolutionary robust soft measurement technique via enhanced atom search optimization and outlier robust extreme learning machine for wastewater treatment process

Accurate and prompt measurement of certain key variables occurring during wastewater treatment, especially effluent ammonia nitrogen (NH4-N) and biological oxygen demand (BOD), is of great importance. In this study, a soft measurement model combining random forest (RF), enhanced atomic search optimization (EASO) and online sequential outlier robust extreme learning machine (OSORELM) is proposed for wastewater treatment. First, RF is used to select auxiliary variables with high correlation of NH4-N and BOD as model inputs. Second, a dynamic perturbation strategy and a generalized opposition-based learning are added to ASO to improve the algorithm performance. And EASO is used for model hyperparameter optimization. Then, the optimized OSORELM model is used for soft measurements of NH4-N and BOD. Finally, to verify the stability of the model, 5 %, 10 % and 15 % noise were added for further experiments. The results show that the proposed model has better prediction and stronger robustness in soft measurements of NH4-N and BOD.

Effectiveness of evolutionary-tuned neurofuzzy inference system in predicting wind turbine gearbox oil temperature

Monitoring the gearbox oil temperature of wind turbines has been one of the most effective ways of increasing their reliability, and reducing maintenance and operational costs. The temperature variation is affected by several factors, including meteorological variables. Unfortunately, this is less considered in the condition monitoring of wind turbine gearboxes. Also, while the neurofuzzy inference model offers a high prospect in improving the understanding of wind turbine gearbox oil temperature behaviour, tuning its hyperparameters for such an application has been less explored. Hence, this study investigates the effectiveness and efficiency of predicting wind turbine gearbox oil temperature using an adaptive neurofuzzy inference system optimized with population-based evolutionary techniques (the genetic algorithm and particle swarm optimization). Relevant meteorological variables were selected from a case study wind farm in South Africa and used as model inputs. The hyper-parameter tuning of the neurofuzzy inference system model was carried out using population-based techniques before training the models. The performances of these hybrid models were compared based on established statistical performance metrics. The genetic algorithm hybrid model outperformed the particle swarm optimization hybrid on the test dataset with a root mean square error of 3.79, mean absolute deviation of 2.93, mean absolute percentage error of 5.9, and relative mean bias error of 1.74. The genetic algorithm-based hybrid model showed higher accuracy and reliability for wind turbine gearbox oil temperature prediction in the case study wind farm and offers potential use for neighbouring wind farms in the same renewable energy development zone.

An evolutionary programming technique for evaluating the effect of ambient conditions on the power output of open cycle gas turbine plants - A case study of Afam GT13E2 gas turbine

In this research a novel Evolutionary Programming (EP) approach combined with a Continual Learning Neural Network (EP-CLNN) technique is proposed for the evaluation of Open Cycle Gas Turbines (OCGTs) net power output considering ambient temperature conditions. Studies were performed comparatively considering the proposed EP-CLNN approach, state-of-the-art linear regressors of several Polynomial (Poly) orders (Poly-1, Poly-2, Poly-3, and Poly-4) and three relatively small-to-big datasets. These approaches were applied to two open source datasets reported in the literature and a case study small dataset acquired from the Afam gas power station plant (GT132E2). The results showed that the reported Root-Mean Squared Error (RMSE) and mean net power estimates for the EP-CLNN are indeed promising considering the constraint of sequential continual learning prediction requirement, reduced variable (single input variable), simpler function set, and relatively small datasets. For the case of the big open source dataset (Dataset-1), the proposed EP-CLNN approach gave the best solution with least RMSE of 4.9926 MW over the linear regressors - Poly-1, Poly-2, Poly-3, and Poly-4 with RMSE estimates of 5.4609 MW, 5.2305 MW, 5.0986 MW, and 5.0980 MW respectively. For the case of the small dataset (Dataset-2), the estimated mean net powers were 32,974.0000 MW, 59140.0000 MW, and 96.2143 MW for Poly-1, Poly-2, and the EP techniques respectively; when compared to the base (reference) net mean power of 97.52 MW, the EP technique is the better one. Furthermore, for dataset 3, the EP technique gave an RMSE of 0.3381 MW while the combined solution of EP-CLNN gave a Mean Absolute Percentage Error (MAPE) consensus estimate of 0.1377 from a synthesized dataset of 20 evolved symbolic expressions. Thus, the EP-CLNN represents a promising approach to real-time gas turbine power output modeling with better accuracy and unique consensus expression capability.

A Subtraction-Average-Based Optimizer for Solving Engineering Problems with Applications on TCSC Allocation in Power Systems.

The present study introduces a subtraction-average-based optimization algorithm (SAOA), a unique enhanced evolutionary technique for solving engineering optimization problems. The typical SAOA works by subtracting the average of searcher agents from the position of population members in the search space. To increase searching capabilities, this study proposes an improved SAO (ISAO) that incorporates a cooperative learning technique based on the leader solution. First, after considering testing on different standard mathematical benchmark functions, the proposed ISAOA is assessed in comparison to the standard SAOA. The simulation results declare that the proposed ISAOA establishes great superiority over the standard SAOA. Additionally, the proposed ISAOA is adopted to handle power system applications for Thyristor Controlled Series Capacitor (TCSC) allocation-based losses reduction in electrical power grids. The SAOA and the proposed ISAOA are employed to optimally size the TCSCs and simultaneously select their installed transmission lines. Both are compared to two recent algorithms, the Artificial Ecosystem Optimizer (AEO) and AQuila Algorithm (AQA), and two other effective and well-known algorithms, the Grey Wolf Optimizer (GWO) and Particle Swarm Optimizer (PSO). In three separate case studies, the standard IEEE-30 bus system is used for this purpose while considering varying numbers of TCSC devices that will be deployed. The suggested ISAOA's simulated implementations claim significant power loss reductions for the three analyzed situations compared to the GWO, AEO, PSO, and AQA.

Application of genetic algorithm in model music composition innovation

Abstract The genetic algorithm-assisted musical composition combines evolutionary computational techniques and artistic composition. In this article, the evolution of musical compositions based on traditional genetic algorithm compositions is combined with interactive genetic algorithm techniques, mainly in model music innovation, using a genetic algorithm to generate an initial population with real number encoding of all individuals. The article focuses on studying the fitness function used to automatically evaluate the automatically generated melodies, perform evolutionary operations on the generated melodies, and generate high-quality model music melodies based on the treatment of evolutionary operations. The evaluation time of the generated melodies by the improved model is reduced from 4 minutes to 0.002 seconds, which makes the composer much more efficient in performing model music composition. The evaluation data of 16 pieces before and after the genetic manipulation were also compared, and it was found that the system evaluated the music after the genetic manipulation significantly higher than the music without the genetic manipulation. The different forms of musical compositions created by the composing system developed based on the technical research of computer genetic algorithm-assisted model music innovation are also a useful supplement to composers’ creations and open a new chapter for the field of artificial intelligence.

Optimum Scenarios of EV Charging Infrastructure: A Case Study for the Saudi Arabia Market

The lack of an EV charging infrastructure is of the top five barriers preventing the adoption of EVs on a large scale. A long charging time is also one of the five barriers, according to the latest survey published by the IEA in 2021. The estimated increase in demand for EVs is a big challenge in many countries all around the world. This challenge exists in many EU and Middle East countries. The main reason for this problem is the requirement of huge funds to install enough public charging points that result in satisfactory charging services. Hence, the phase-out plans of internal combustion engine (ICE) vehicles can be carried out successfully and smoothly. Unfortunately, there is a trade-off between the cost of installing charging points and EV charging time. Therefore, it is important to optimize both factors simultaneously. This way, the charging services can be provided at the minimum possible cost and at a satisfactory level of quality. This study determines the optimum ratio of the number of chargers to the number of EVs in a certain province. The optimal number of chargers that are necessary to optimally serve a certain number of EVs has been determined. Two well-known evolutionary search techniques have solved the optimization problem: particle swarm optimization (PSO) and genetic algorithms (GA). Both algorithms have succeeded in providing many optimal charging infrastructure scenarios. Hence, the decision maker can select the most convenient scenario from several alternatives based on the available budgets and the most convenient charging time.

Presenting a Novel Evolutionary Method for Reserve Constrained Multi-Area Economic/Emission Dispatch Problem

Economic dispatch (ED) attempts to find the most cost-effective combination of power generation units while meeting operational constraints. Another problem that can’t be resolved by standard economic dispatch problems is figuring out the method of generating dispatch that would be most cost-effective in meeting the local demand without exceeding the tie-line capacity. Making a trade-off between fuel costs and environmental concerns, a contentious problem in industrialized countries, seems essential. As a result, this study introduces a multi-objective approach for different ED problems, such as multi-area emission economic dispatch (MAEED) and reserve constrained multi-area emission economic dispatch (RCMAEED), when there are real-world restrictions present, like the valve point effect (VPE), prohibited operating zones (POZs), multi-fuel operation (MFO), and ramp-rate (RR) restrictions. In this study, the generation cost and emissions are taken into consideration as objective functions. Since the MAED problem in the power system is inherently nonlinear, adding the aforementioned restrictions makes the problem even more challenging. To address the complexity of the multi-objective optimization problem, the modified grasshopper optimization (MGO) algorithm, based on the chaos mechanism, is proposed in this paper. The proposed method has been tested on a four-area power system with sixteen electrical generators, and the results are contrasted with those of previous evolutionary techniques. Based on the results, it can be concluded that using the proposed MGO method to solve the MAED and RCMAED problems will result in generation costs that are around $300 and $600 less than using the MPSO and PSO methods, respectively. Also, the proposed MGO method has reduced emission levels by roughly 30% as compared to the GO method in order to solve the RCMAEED problem.

Evolutionary Ensemble Learning Pathloss Prediction for 4G and 5G Flying Base Stations With UAVs

The usage of unmanned aerial vehicles (UAVs) as flying base stations (FBSs) for expanding coverage and assisting the terrestrial cellular networks constitutes a promising technology for 5G and beyond. A crucial parameter affecting cellular network design is path loss prediction. An alternative to the accurate, though time-consuming, propagation prediction with deterministic Ray Tracing models could be Machine Learning (ML) based predictions. Ensemble Learning techniques are used in order to optimally combine the predictions of standalone models. That is, they combine the best-performing individual models into a better-performing meta-model. Our proposed method of the evolutionary tuned stacked ensemble optimizes the ensemble as a whole, instead of optimizing its individual base learners. To the best of our knowledge, this is the first time that an evolutionary technique is applied in order to mutually tune an ensemble’s base learners for a path loss modeling problem in electromagnetics. Moreover, we present a model that works in more than one frequency. As opposed to the standard implementation of ensemble learning, our method offers a significant performance boost with low complexity.

Design Approach for Evolutionary Techniques Using Genetic Algorithms: Application for a Biped Robot to Learn to Walk and Rise after a Fall

In this research, a biped robot is programmed to learn to walk using genetic algorithms. The biped robot has the ability to rise after a fall. The system is made according to human genetics, where a child tries to learn to walk, takes a few steps, falls, and then stands up somehow after a fall. After this process, the system restarts and tries to walk again. The biped robot evolves after each generation, and after each generation it will have a more natural walking posture. The system is unpredictable, because the biped robot sometimes evolves faster and sometimes evolves slower, like in human genetics where some children are faster and other children are slower to walk. The biped robot is considered to have evolved enough after having a walking posture similar to that of a human and does not fall.

Optimum Generation Scheduling of Thermal and Hydro Power System with Renewable Power Sources Using Enhanced Grey Wolf Optimization Algorithm

In this work, an algorithm of enhanced grey wolf optimization (EGWO) is developed to resolve the scheduling problem of a power system, which is having thermal, hydro, wind, and solar power plants with a battery energy storing system (BESS). The planning period covers 24 equal intervals of a day. In this scheduling problem, the impact on valve point loading of thermal plants, the time coupling effect in cascaded reservoirs of hydro units, and various constraints imposed on the power network are taken into consideration. The grey wolf optimization (GWO) technique is used, which is framed on the group attitude of grey wolves such as governance ranking and hunting activity. In this proposed algorithm, GWO algorithm is enhanced using three techniques. First, the quasi-oppositional learning approach is exploited to obtain the optimal solution quickly. Second, the elite mutation operator is applied to maximize the diversity of the swarm. Finally, by implementing the elastic-ball strategy, infeasible solutions are modified into feasible one. The potential of the EGWO technique is ascertained by employing it in two experimental settings. From the simulation outcome, it is inferred that the implemented technique bestows less operating cost with minimal computation time as compared to other evolutionary techniques.

Dynamic and Model Predictive Controllers for Frequency Regulation of an Isolated Micro—Grid with Electrical Vehicles and the ESS Integration

Electric vehicles (EVs) improve the performance of Energy Storage Systems (ESS). The inclusion of electric vehicles in the microgrid system has attracted considerable attention due to the emerging potential of vehicle-to-grid options. The proper coordination among ESS & EVs has a high capacity to tolerate frequency discrepancies and the functioning of the MG (Micro-Grid). In this research, two innovative control techniques-To address the issue of frequency regulation in the standalone MG, Model Predictive Control (MPC) and Dynamical Droop Control (D2C) are linked to affect the structure of ESS and EVs following extensive RES inclusion or a significant change in load demand. The D2C protects EV energy in line with system demands by preserving the least amount of power for anticipated EV demand. Combined MPC and D2C settings are tuned using a complex evolutionary technique to further improve performance. In MATLAB/Simulink, a single MG (micro-Grid) is modeled and evaluated using the controlling strategies indicated above. As well, a range of case papers is taken into account to confirm the integration of the D2C and MPC for a single MG's modulation scheme. Moreover, the MPC outperforms both fuzzy-based proportional-integral (FPI) and proportional-integral (PI) controllers in terms of performance results.

Comparative study of genetic programming-based algorithms for predicting the compressive strength of concrete at elevated temperature

The elevated temperature severely influences the mixed properties of concrete, causing a decrease in its strength properties. Accurate proportioning of concrete components for obtaining the required compressive strength (C-S) at elevated temperatures is a complicated and time-taking process. However, using evolutionary programming techniques such as gene expression programming (GEP) and multi-expression programming (MEP) provides the accurate prediction of concrete C-S. This article presents the genetic programming-based models (such as gene expression programming (GEP) and multi-expression programming (MEP)) for forecasting the concrete compressive strength (C-S) at elevated temperatures. In this regard, 207 C-S values at elevated temperatures were obtained from previous studies. In the model’s development, C-S was considered as the output parameter with the nine most influential input parameters, including; Nano silica, cement, fly ash, water, temperature, silica fume, superplasticizer, sand, and gravels. The efficacy and accuracy of the GEP and MEP-based models were assessed by using statistical measures such as mean absolute error (MAE), correlation coefficient (R2), and root mean square error (RMSE). Moreover, models were also evaluated for external validation using different validation criteria recommended by previous studies. In comparing GEP and MEP models, GEP gave higher R2 and lower RMSE and MAE values of 0.854, 5.331 MPa, and 0.018 MPa respectively, indicating a strong correlation between actual and anticipated outputs. Thus, the GEP-based model was used further for sensitivity analysis, which revealed that cement is the most influencing factor. In addition, the proposed GEP model provides simple mathematical expression that can be easily implemented in practice.

PR controller-based droop control strategy for AC microgrid using Ant Lion Optimization technique

Microgrids are now an integral part of the modern power system to maintain the stability of the power distribution process. Ensuring the frequency and voltage stability in the autonomous mode is quite a challenging task indeed. These have now reached the extent of utilizing evolutionary intelligence techniques leading to the concept of intelligent microgrids. In this work, a novel droop control strategy is implemented with a proportional resonant controller, whose proportional gain is optimized by the Ant Lion Optimization algorithm (A.L.O). The efficacy of the A.L.O algorithm for its application in the primary level droop control for voltage stability is investigated in this work. The superiority of the A.L.O optimized PR controller over the conventional PI controller is compared in terms of error indices, timed domain specifications, speed of response, voltage regulation and T.H.D levels. It could ensure a higher speed of response, better voltage stability and higher power quality as against its PI counterpart. The performance of the proposed A.L.O optimized PR controller is validated by comparing it with the performance achieved by controllers tuned with two other algorithms namely, particle swarm optimization (P.S.O) and Satin bower bird optimization (S.B.O). The proposed controller could ensure better steady-state stability compared to the other two.

Machine learning assisted optimization of tribological parameters of Al–Co–Cr–Fe–Ni high-entropy alloy

ABSTRACT High-entropy alloys (HEA) have become increasingly important in the materials world due to their ability to combine multiple principal elements in an alloy system to achieve desired mechanical and tribological properties. In this study, Al–Co–Cr–Fe–Ni HEA alloy is synthesized using vacuum induction melting and its tribological properties are analyzed under dry conditions, which is important since dry sliding is a common mode of wear in many engineering applications. To enhance its tribological performance, evolutionary data-driven modeling and optimization techniques are applied to determine the best possible solutions for the loading condition at different frequency levels. A neural net-based algorithm (EvoNN) is utilized in the data-driven modeling process to generate and train models, which are then used in the optimization process. The reference vector-based algorithm (cRVEA) is incorporated with the EvoNN to evaluate the optimum wear, coefficient of friction (COF), and surface roughness results in multidimensional hyperspace. The results obtained from this study fall within the experimental data range and satisfy the tribological requirements under optimum conditions. The excellent wear resistance of HEAs due to their high hardness, strength, and resistance to deformation makes them suitable for use in various wear applications, including cutting tools, bearings, and gears. Overall, this study highlights the potential of HEAs and data-driven modeling techniques for the development of advanced materials with superior properties and performance.

Multiagent Systems on Virtual Games: A Systematic Mapping Study

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Context</i> : Games are a well-established scenario to test AI and Multi-Agent Systems (MAS) proposals due to their popularity and defiance. However, there is no big picture of the application of this technology to games, the evolution of the kind of problem tackled, or the game scenarios in which agents have been experimented. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Objective</i> : To perform a systematic mapping to characterise the state of the art in the field of MAS applied to virtual games and to identify trends, strengths, and gaps for further research. Method: A Systematic Mapping Study has been conducted to find primary studies in the field. A search was performed on title, abstracts, and keywords, whilst classification, data extraction, and further analysis were performed according to specific criteria focused on MAS papers with experimentation and evidence in a game scenario. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Results</i> : 78 studies published between 1998 and 2021 were found. Studies have been classified according to the MAS problem faced and the agent reasoning strategy. We detect that Machine Learning is the most common AI technique for MAS in games, considering both reinforcement learning and evolutionary techniques. MAS are used in a variety of gaming genres, especially in Real-Time Strategy (RTS), Sports and Simulation. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Conclusions</i> : RTS and Sports games are well-suited for concrete MAS problems such as multi-agent planning and task allocation. Expanding evidence and experimentation on other aspects related to scalability and usability issues is discussed. Those MAS problems and experiments that remain slightly modelled on games or are not thoroughly studied yet have been also identified.

Evolutionary Techniques for the Solution of Bio-Heat Equation Arising in Human Dermal Region Model

The proposed research work analyzes the bio-inspired problem through artificial neural networks with a feed-forward approach utilized to approximate the numerical results for singular nonlinear bio-heat equation (BHE) with boundary conditions based on four different scenarios created on the variation of environmental temperature to illustrate the effects of temperature on the human dermal region. The log-sigmoid function is used to construct the fitness function, while the optimization solvers: pattern search and genetic algorithm, are then hybridized with the active set technique, interior point technique, sequential quadratic programming for accurate and reliable results of the proposed BHE with various scenarios where the convergence of the numerical results is also analyzed. Moreover, a comparison of the proposed technique is expressed through residual error that reveals the nature of the numerical results and their efficiency. Additionally, a comprehensive statistical analysis is presented for the designed technique to better illustrate the accuracy, reliability, and efficiency of the obtained results.

Designing a hybrid PV/FC/wind for economic study using modified ruddy turnstone optimization algorithm

This paper outlines the optimal design of a hybrid system that includes wind turbines, solar panels, and a fuel cell to successfully meet the load's requirements. All cost components of the system, including load loss and expenditure for operation and maintenance, have been taken into consideration. A modified version of Ruddy Turnstone Optimization Algorithm is applied to determine the ideal equipment capacity for the hybrid system. Comparison and analysis are done between the optimization results and the outcomes obtained through evolutionary techniques. Moreover, the proposed technique is tested on a case study from Egypt and a comparison is made with existing Hybrid Firefly/Harmony Search method from literature. According to the results, both modified Ruddy Turnstone Optimization and Hybrid Firefly/Harmony Search methods have produced satisfactory outcomes. While the production cost slightly increased using the Hybrid Firefly/Harmony Search methodology, the system's dependability level has improved significantly, thereby ensuring reliable load supply. The results of a loss of power supply possibility comparison using 0.5%, 1%, and 2% of the modified Turnstone Optimization algorithm versus Hybrid Firefly/Harmony Search method approach yields a minimum cost of energy of 0.0798, 0.0739, and 0.0573 $/kWh, respectively.

1,2, 3

In this research paper, a swarm evolutionary modeling technique called the Updated Multi-Objective Ant-Lion Optimizer (UMO-ALO) is applied to the modeling of Open Circuit Voltage (OCV) state-of-charge (SOC) of Lithium-ion cells with cathode-anode composition (LiMnO2/Li4TiO). The model approximates the battery SOC by considering several battery cell internal physical parameters and a linear fitness function governed by two objectives involving the charge and discharge response of the battery cell. Bound limited coefficients of the physical parameters are used in the optimization process. The UMO-ALO uses a special update procedure to reduce the computational run-time of standard MO-ALO and hence speeds up the generic optimizer. Simulations were performed using a very small real laboratory data obtained from the relevant field studies and for different trial-run configurations of UMO-ALO. The results from these simulations show very good fitness responses close to zero margins for both trial-run configurations.