Pulsed eddy current (PEC) has been widely employed in the metal industry for estimating the electromagnetic parameter and thickness of steel plates, owing to its unique advantages, such as rapid response, strong penetration ability, and richer information, in time domain and frequency domain. The slope of the last phase signal on a logarithmic scale is a crucial indicator in PEC testing. A simplified analytical expression for estimating the slope of the last phase signal is presented for the first time in this article. The slope can be computed based on the thickness, electrical conductivity, and magnetic permeability of the tested specimen. The simultaneous nonlinear mapping from measured signals to the thickness, electrical conductivity, and magnetic permeability of the tested specimen is constructed by the deep learning model based on the Transformer network, which is physics-informed through the deriveD simplified analytical expression, incorporating the last phase signal slope on a logarithmic scale to enhance test accuracy and speed. Numerical simulations and experiments were conducted to assess the proposed method for estimating electromagnetic parameters and thickness across different plate materials. The method provides real-time, accurate parameters estimation of the steel plate, and all estimated parameters have a relative error below 3%.

With the development of the Industrial Internet of Things, industrial control systems have become more open and intelligent. However, large numbers of unknown protocols exist in IIoT, threatening the security of IIoT devices and systems. Protocol reverse engineering extracts the grammar and semantics of the protocol by monitoring and analyzing the traffic trace or the execution process of instructions, without the need for protocol description. As the executable programs are mainly integrated into the IIoT devices and the communication traffic is relatively sparse, the traditional protocol analyzing method is not suitable for the IIoT environment. This paper proposes an improved sparse protocol parsing method of IIoT protocol based on the BPSO-vote-HMM hybrid model. The binary particle swarm optimization algorithm is introduced to expand the captured IIoT protocol message sequence, solving the problems of sparse samples in IIoT and the low efficiency of the GA-based data expansion model. Besides, we improve on the parameter training part to improve the efficiency and get better model parameters by dividing the training set into several sub-sets, conducting the parameter update parallel, and inputting the results into a voter to generate the final parameter of HMM, which is used in protocol field prediction. Finally, by combining the BPSO-based data expansion model and the protocol field parsing model based on vote-HMM, a hybrid analytical model is constructed to improve the analytical accuracy in a gradual evolutionary manner. Through a series of comparative experiments, the improved protocol field parsing model has better performance on IIoT protocol.

As the intelligentization of Industrial Internet of Things (IIoT) broke the relatively closed and credible industrial environment, IIoT faces increasingly serious security problems. The commonly used vulnerability discovery method is protocol reverse engineering. However, it is difficult to analyze IIoT protocols with existing protocol reverse engineering approaches, as they influence the normal operation or have spare sample data. In this paper, a sparse protocol parsing method for IIoT protocols is proposed. The parsing method expands the samples of the captured IIoT protocol message sequences using a genetic algorithm (GA), which designs its fitness function based on the protocol response data to select high-quality samples. By combining the GA with the hidden Markov model (HMM) with lower algorithm complexity, a hybrid parsing model is constructed to improve accuracy in a gradual evolution way. Through comparison experiments on various IIoT protocols, our HMM hybrid model has better performance than RNN hybrid models under sparse samples.