A new sparse design method on phased array-based acoustic emission sensor for partial discharge detection

The acoustic detecting performance of a partial discharge (PD) ultrasonic sensor array can be improved by increasing the number of array elements. However, it will increase the complexity and cost of the PD detection system. Therefore, a sparse sensor with an optimization design can be chosen to ensure good acoustic performance. In this paper, first, a quantitative method is proposed for evaluating the acoustic performance of a square PD ultrasonic array sensor. Second, a method of sparse design is presented to combine the evaluation method with the chaotic monkey algorithm. Third, an optimal sparse structure of a 3 × 3 square PD ultrasonic array sensor is deduced. It is found that, under different sparseness and sparse structure, the main beam width of the directivity function shows a small variation, while the sidelobe amplitude shows a bigger variation. For a specific sparseness, the acoustic performance under the optimal sparse structure is close to that using a full array. Finally, some simulations based on the above method show that, for certain sparseness, the sensor with the optimal sparse structure exhibits superior positioning accuracy compared to that with a stochastic one. The sensor array structure may be chosen according to the actual requirements for an actual engineering application.