|
| S. no. | Algorithm | Advantages | Disadvantages | Remarks |
|
| 1. | Simulated annealing (SA) | (i) Easy to implement [24]
(ii) Converge under appropriate conditions | (i) SA has a lower performance than GA [210]
(ii) The SA algorithm is inferior to alternative optimization algorithms like GA, PSO, and TLBO [24]
(iii) Slow convergence [28]
(iv) SA is not suitable for working at some temperature and irradiance values [28] | Although SA was tried for the optimization of SAHs, the present literature has shown that it is inferior to the alternative optimization algorithms like GA, PSO, and TLBO. |
| 2. | Particle swarm optimization (PSO) | (i) PSO considers different and better positions to update the population
(ii) Simple in concept
(iii) Higher convergence speed [14]
(iv) Easy to be implemented [14] | (i) In Ref. [84], PSO methodology did not show better optimization compared to TLBO
(ii) PSO depends on the inertial weight and learning constants [213] | PSO works well for determining the maximum optimal thermal efficiency of SAHs. However, there are other algorithms that work better than PSO. |
| 3. | Genetic algorithm (GA) | (i) Accurate and quick compared to traditional methods [203] | (i) In [202], the thermal efficiency of SAH was reported to be very poor. This was due to its low heat transfer capability, and it depends upon the various system and operating parameters
(ii) Generally, GA tends to be trapped into local minima
(iii) Difficult to converge
(iv) It can produce meaningless results | GA remains one of the most widely used optimization algorithms. According to the literature on SAH optimization, it still shows that some drawbacks like poor thermal efficiency exist. |
| 4. | Artificial bee colony (ABC) | (i) Accurate and quick compared to traditional methods [203]
(ii) In Ref. [203], the thermal efficiency of SAC obtained by the ABC algorithm was slightly higher than the thermal efficiency obtained by GA
(iii) Simple as the PSO algorithm, but with fewer restriction limits like the highest cycle count and colony extent [214] | (i) Some parameters can affect the algorithm efficiency [23]
(ii) Slow convergence speed [212] | According to the recent literature, despite some drawbacks, ABC has shown good results for the optimization of SAHs. |
| 5. | Teaching-learning-based optimization (TLBO) | (i) In Ref. [24], TLBO was found to be more acceptable than alternative optimization algorithms like GA, PSO, and SA
(ii) It is an efficient algorithm [24]
(iii) It is a promising algorithm for determining optimal designs [24]
(iv) In Ref. [84], the TLBO approach showed better results for maximizing the efficiency than the PSO technique
(v) It is a simple and low-cost method in terms of metaheuristic evaluation for thermohydraulic efficiency of roughened SAHs [84]
(vi) A fast convergence speed [215]
(vii) A high precision algorithm [215] | (i) The TLBO only requires tuning of the algorithm’s accepted regulating limits for its operation [201] | Until now, TLBO has shown good results for the optimization of SAHs compared to the other optimization methods. |
| 6. | Elitist teaching-learning-based optimization (ETLBO) | (i) In Ref. [200], ETLBO and TLBO algorithms were both found to be more acceptable than alternative optimization algorithms like GA, PSO, and SA
(ii) It is an efficient algorithm [200]
(iii) It is a promising algorithm for determining optimal designs [200] | (i) The impact of the ETLBO algorithm’s accepted governing metrics like the population size, generation number, and elite size on the algorithm’s efficiency must be considered and investigated in detail. But, the TLBO only requires tuning of the algorithm’s accepted supervising limits for its operation [201] | According to the recent literature, both the ETLBO and TLBO algorithms perform better than the other optimization algorithms considered by the previous researchers. Until recently, despite the mentioned drawbacks of this approach, there are not many research papers on SAH optimization applying this algorithm. |
|
