Novel Sparse Planar Array Synthesis Model for Microwave Power Transmission Systems With High Efficiency and Low Cost

Abstract

A novel algorithm is developed to realize the optimal synthesis of a sparse nonuniformamplitude nonuniform-distribution planar array (SNANDPA) in microwave power transmission (MPT) systems. The dual compression factor particle swarm optimization (DCFPSO) algorithm and the subarray partition technique are adopted to realize the optimal synthesis of SNANDPA. The DCFPSO algorithm first optimizes beam collection efficiency (BCE ) and side-lobe level outside the receiving region (CSL) of SNANDPA which ensure efficient energy transmission of an MPT system and suppress the influence of electromagnetic wave radiated by antenna array on the environment. The subarray partition technique then simplifies the feed network to minimize the system cost. SNANDPA parameters including transmitting aperture, receiving aperture, BCE, CSL, power pattern, element weight, and element distribution, can be obtained efficiently via the proposed method. Representative numerical cases under the different numbers of subarray and elements conditions are analyzed and compared with those of other two traditional MPT array models. Experimental results show that, when the transmitting aperture is 4.5λ × 4.5λ and the square receiving region u 0 = v 0 = 0.2, BCE and CSL are 94.96% and −17.09 dB, respectively, and only 64 elements and 8 amplifiers are required. We conclude that the proposed model can be used to create an efficient and low-cost MPT system.

Novel Subarray Partition Algorithm for Solving the Problem of Too Low Beam Collection Efficiency Caused by Dividing a Few Subarrays

“…The second planar array model is a Nonuniform Distributed Rectangular Planar Array (NDRPA), such an overall model has higher optimization freedom and may get better results. If readers want to know the physical model of NDRPA, you can refer to the model of SNANDPA in [9]. The third array model optimizes only the globally optimal individuals (only-gbest) in SQSRA.…”

“…[9] Ref. [13] According to Table 4, it can be concluded that the BCE achieved by the proposed method is 1.95% higher than the result in [9] (BCE| N DRP A = 93.04% vs. BCE| [9] = 91.09%). Although the array in [13] is a fully populated and uniformly excited array (γ a =1), the BCE only reaches 91%.…”

“…[12] Ref. [9] Ref. [13] According to Table 4, it can be concluded that the BCE achieved by the proposed method is 1.95% higher than the result in [9] (BCE| N DRP A = 93.04% vs. BCE| [9] = 91.09%).…”

“…To further illustrate the effectiveness of the MPMP-DWPSO-SP algorithm, we compared the results of optimizing the three models by using the one-step method with SNANDPA by using the previous two-step method in [9]. The comprehensive comparison result is shown in Fig.…”

“…Previous studies have divided subarrays by using excitation [8]. The optimization of the subarray partition is divided into two steps: the first step is optimizing the position of the element and the second step is dividing subarrays [9], or the first step is optimizing the structure of the subarray, and the second step is optimizing the element position [10][11]. The obtained BCE by using this method can achieve satisfactory results when it is divided into multiple subarrays (more than 8), but when the number of subarrays is small (less than 4), the BCE is exceptionally low.…”