Large-scale grids have gradually become the dominant trend in power systems, which has increased the importance of solving the challenges associated with large-scale economic dispatch (LED). An increase in the number of decision variables enlarges the search-space scale in LED. In addition to increasing the difficulty of solving algorithms, huge amounts of computing resources are consumed. To overcome this problem, we proposed a surrogate-assisted adaptive bat algorithm (GARCBA). On the one hand, to reduce the execution time of LED problems, we proposed a generalized regression neural network surrogate model based on a self-adaptive “minimizing the predictor” sampling strategy, which replaces the original fuel cost functions with a shorter computing time. On the other hand, we also proposed an improved hybrid bat algorithm (RCBA) named GARCBA to execute LED optimization problems. Specifically, we developed an evolutionary state evaluation (ESE) method to increase the performance of the original RCBA. Moreover, we introduced the ESE to analyze the population distribution, fitness, and effective radius of the random black hole in the original RCBA. We achieved a substantial improvement in computational time, accuracy, and convergence when using the GARCBA to solve LED problems, and we demonstrated this method’s effectiveness with three sets of simulations.
17 May 2024 in International Journal of Electrical Power & Energy Systems
During the agricultural season, the proportion of irrigation load in rural microgrid increases rapidly. However, the energy can hardly be managed effectively due to the surge in irrigation load, rapid fluctuations of renewable energy, and inaccurate electrical parameters of aging equipment. In this article, an energy management model for rural microgrid including irrigation, hydrogen, electric heating, and power systems is proposed. Further, formidable challenges arise from inaccurate water-electricity conversion coefficient of water pump and non-convex dynamic efficiency function of hydrogen energy storage system. A deep reinforcement learning (DRL) method is proposed incorporating a Kolmogorov-Arnold network (KAN) structure and a two-stage implementation deployment method, which maintains robustness against discrepancies between the controlled training environment and the dynamic application scenario. The results of simulation experiments demonstrate adaptability and robustness of the KAN-based method in handling deviations of equipment parameters. The effectiveness and superiority of KAN-based DRL approach are validated through a comparative analysis with the state-of-the-art algorithms.
1 January 2024 in Journal of Modern Power Systems and Clean Energy
Multi-area combined economic/emission dispatch (MACEED) problems are generally studied using analytical functions. However, as the scale of power systems increases, existing solutions become time-consuming and may not meet operational constraints. To overcome excessive computational expense in high-dimensional MACEED problems, a novel data-driven surrogate-assisted method is proposed. First, a cosine-similarity-based deep belief network combined with a back-propagation (DBN+BP) neural network is utilized to replace cost and emission functions. Second, transfer learning is applied with a pretraining and fine-tuning method to improve DBN+BP regression surrogate models, thus realizing fast construction of surrogate models between different regional power systems. Third, a multi-objective antlion optimizer with a novel general single-dimension retention bi-objective optimization policy is proposed to execute MACEED optimization to obtain scheduling decisions. The proposed method not only ensures the convergence, uniformity, and extensibility of the Pareto front, but also greatly reduces the computational time. Finally, a 4-area 40-unit test system with different constraints is employed to demonstrate the effectiveness of the proposed method.
1 January 2024 in International Journal of Swarm Intelligence Research
In this article, a novel hybrid binary bat algorithm named HBBA is proposed for global optimization problems. First, to avoid simultaneous updating of bat velocity's dimensional components, i.e., elements of velocity vector, a random black hole model is modified to adapt to binary algorithm for updating in unknown spaces for each dimensional component individually. Through this way, the search ability of bats around the current group best is increased greatly. Second, a time-varying v-shaped transfer function, rather than a time-invariant one as in closely related works, is proposed to map velocity in continuous search space to a binary one. This accelerates the speed to switch individuals' positions, i.e., solutions in binary space. Third, a chaotic map is utilized to replace monotonous parameters in original binary bat algorithm, which is beneficial for avoiding premature convergence. Simulation results demonstrate the effectiveness of the proposed algorithm by three types of benchmark functions and unit commitment problem.