Design and Optimization of Bi-Planar Coils for Active Magnetic Compensation System Inside the MINI-Magnetically Shielded Room

The MINI-magnetically shielded room (MINI-MSR) offers a near-zero magnetic environment for the high performance of optically pumped atomic magnetometers (OPMs) in magnetoencephalography (MEG) tests. It has the advantages, such as being small-sized, portable, lightweight, offering efficient space utilization, and being cost-effective. However, owing to its small size and narrow internal space, the MINI-MSR can only provide a limited uniform region with larger gradients and irregular magnetic noise distributions. Moreover, it is affected by large external background fluctuations. Once the external interference exceeds the magnetic shielding capacity of the MINI-MSR, the precision of OPMs is compromised. Therefore, an active magnetic compensation system (AMCS) composed of biplanar coils (BCs) is introduced to enlarge the uniform region, maintaining the inhomogeneity errors below 1% to meet the measurement accuracy. However, conventional algorithms for designing BCs struggle with searching for the regularization coefficient ω that determines the coil's shape. The process lacks ergodicity and efficiency, posing challenges in obtaining maximum volume and uniformity within the target region. In this study, a new hybrid method (NHM) composed of the target field method (TFM), image method (IM), and novel artificial fish swarm algorithm (novel ASFA) is proposed to design BCs for the AMCS. The aim is to suppress background field fluctuations, enhancing the signal-to-noise ratio while addressing coupled errors to reduce larger gradients and magnetic noise levels. Experimental results show that the maximum inhomogeneity error of BCs is less than 1.23%, which is consistent with simulation results. The sensitivity of the AMCS at 0.01 Hz ranges from 180 to 0.16 pT/Hz 1/2 , realizing the noise suppression ratio of 30.51 dB. Based on the closed-loop AMCS, the signal-to-noise ratio in the alpha rhythm generation task is effectively improved.