Herein, to accurately predict tool wear, we proposed a new deep learning network—that is, the IE-Bi-LSTM—based on an informer encoder and bi-directional long short-term memory. The IE-Bi-LSTM uses the encoder part of the informer model to capture connections globally and to extract long feature sequences with rich information from multichannel sensors. In contrast to methods using CNN and RNN, this model could achieve remote feature extraction and the parallel computation of long-sequence-dependent features. The informer encoder adopts the attention distillation layer to increase computational efficiency, thereby lowering the attention computational overhead in comparison to that of a transformer encoder. To better collect location information while maintaining serialization properties, a bi-directional long short-term memory (Bi-LSTM) network was employed. After the fully connected layer, the tool-wear prediction value was generated. After data augmentation, the PHM2010 basic dataset was used to check the effectiveness of the model. A comparison test revealed that the model could learn more full features and had a strong prediction accuracy after hyperparameter tweaking. An ablation experiment was also carried out to demonstrate the efficacy of the improved model module.

A novel simulation method of microtopography for grinding surface was proposed in this paper. Based on the theory of wavelet analysis, multiscale decomposition of the measured topography was conducted. The topography was divided into high frequency band (HFB), theoretical frequency band (TFB), and low frequency band (LFB) by wavelet energy method. The high-frequency and the low-frequency topography were extracted to obtain the digital combination model. Combined with the digital combination model and the theoretical topography obtained by geometric simulation method, the simulation topography of grinding surface can be generated. Moreover, the roughness parameters of the measured topography and the simulation topography under different machining parameters were compared. The maximum relative error of Sa, Sq, Ssk and Sku were 1.79%, 2.24%, 4.69% and 4.73%, respectively, which verifies the feasibility and accuracy of the presented method.

Lithium-ion capacitors (LICs) have been widely explored for energy storage. Nevertheless, achieving good energy density, satisfactory power density, and stable cycle life is still challenging. For this study, we fabricated a novel LIC with a NiO-rGO composite as a negative material and commercial activated carbon (AC) as a positive material for energy storage. The NiO-rGO//AC system utilizes NiO nanoparticles uniformly distributed in rGO to achieve a high specific capacity (with a current density of 0.5 A g⁻¹ and a charge capacity of 945.8 mA h g⁻¹) and uses AC to provide a large specific surface area and adjustable pore structure, thereby achieving excellent electrochemical performance. In detail, the NiO-rGO//AC system (with a mass ratio of 1:3) can achieve a high energy density (98.15 W h kg⁻¹), a high power density (10.94 kW kg⁻¹), and a long cycle life (with 72.1% capacity retention after 10,000 cycles). This study outlines a new option for the manufacture of LIC devices that feature both high energy and high power densities.

A large headframe is the core structure of a mine hoisting system. In the traditional design, only the static analysis under load is considered, resulting in the resonance phenomenon of the large headframe in later applications. In order to restrain the resonance phenomenon, a novel method for dynamic characteristic analysis and structural optimization design of a large headframe is proposed. First, the eigenfrequencies and vibration modes of the large headframe were obtained through modal analysis. The results showed that the numerical values of the multi-order eigenfrequencies of the system are relatively close. When subjected to alternating loads of similar frequencies, a large headframe is prone to the resonance phenomenon. Second, the steady-state vibration response of the large headframe was obtained through harmonic response analysis. The results showed that when the frequency of the alternating load is close to the first-order eigenfrequency, the vibration amplitude increases. Meanwhile, the fourth-order and the fifth-order eigenfrequencies are very close. When subjected to alternating loads of similar frequencies, the fourth-order and the fifth-order vibration modes of the headframe will be excited simultaneously. At this time, the headframe will have a strong resonance, which may cause structural damage and other problems. Finally, based on the above analysis, nine different structural optimization schemes are proposed in this paper. Through modal analysis and harmonic response analysis, the nine schemes were compared and analyzed, and the optimal scheme was eventually determined as scheme 9. The method proposed in this paper provides a new concept for the structural optimization design of a large mining headframe, and it has great significance for restraining the resonance phenomenon and ensuring the safety of mining operations.

In order to simulate the grinding surface more accurately, a novel modeling method is proposed based on the ubiquitiform theory. Combined with the power spectral density (PSD) analysis of the measured surface, the anisotropic characteristics of the grinding surface are verified. Based on the isotropic fractal Weierstrass–Mandbrot (W-M) function, the expression of the anisotropic fractal surface is derived. Then, the lower bound of scale invariance δ_min is introduced into the anisotropic fractal, and an anisotropic W-M function with ubiquitiformal properties is constructed. After that, the influence law of the δ_min on the roughness parameters is discussed, and the δ_min for modeling the grinding surface is determined to be 10⁻⁸ m. When δ_min = 10⁻⁸ m, the maximum relative errors of Sa, Sq, Ssk, and Sku of the four surfaces are 5.98%, 6.06%, 5.77%, and 4.53%, respectively. In addition, the relative errors of roughness parameters under the fractal method and the ubiquitiformal method are compared. The comparison results show that the relative errors of Sa, Sq, Ssk, and Sku under the ubiquitiformal modeling method are 5.36%, 6.06%, 5.84%, and 4.53%, while the maximum relative errors under the fractal modeling method are 23.21%, 7.03%, 83.10%, and 7.25%. The comparison results verified the accuracy of the modeling method in this paper.

A novel three-dimensional fractal model for normal contact stiffness is proposed in this paper. First of all, a hypothetical surface based on axisymmetric cosinusoidal asperity is established. Then, based on the hypothetical surface, the analytical expressions for the contact stiffness and contact load are derived by combining the three-dimensional fractal theory with the contact mechanics theory. In addition, the simulation results of the presented model and the Pan model are compared with the experimental results. The comparison results show that the maximum relative error of the Pan model is 29.58%, while the maximum relative error of the presented model is 4.35%. Ultimately, the influence of different fractal parameters on contact stiffness is discussed. Under the same contact load, the normal contact stiffness first increases and then decreases with the increase of the fractal dimension D, while the normal contact stiffness monotonically decreases with the increase of scale coefficient G. The results are explained from the perspective of the shape of the asperity. This study provides a novel model for the calculation of normal contact stiffness, which provides a model basis for the study of contact properties for the mechanical interface.