Search

The insufficient generalisation capability of the conventional learning from demonstration (LfD) model necessitates redemonstrations. In addition, retraining the model can overwrite existing knowledge, making it impossible to perform previously acquired skills in new application scenarios. These are not economical and efficient. To address the issues, in this study, a broad learning system (BLS) and probabilistic roadmap (PRM) are integrated with dynamic movement primitive (DMP)-based LfD. Three key innovations are proposed in this paper: (1) segmentation and extended demonstration: a 1D-based topology trajectory segmentation algorithm (1D-SEG) is designed to divide the original demonstration into several segments. Following the segmentation, a Gaussian probabilistic roadmap (G-PRM) is proposed to generate an extended demonstration that retains the geometric features of the original demonstration. (2) DMP modelling and incremental learning updating: BLS-based incremental learning for DMP (Bi-DMP) is performed based on the constructed DMP and extended demonstration. With this incremental learning approach, the DMP is capable of self-updating in response to task demands, preserving previously acquired skills and updating them without training from scratch. (3) Electric vehicle (EV) battery disassembly case study: this study developed a solution suitable for EV battery disassembly and established a decommissioned battery disassembly experimental platform. Unscrewing nuts and battery cell removal are selected to verify the effectiveness of the proposed algorithms based on the battery disassembly experimental platform. In this study, the effectiveness of the algorithms designed in this paper is measured by the success rate and error of the task execution. In the task of unscrewing nuts, the success rate of the classical DMP is 57.14% and the maximum error is 2.760 mm. After the optimisation of 1D-SEG, G-PRM, and Bi-DMP, the success rate of the task is increased to 100% and the maximum error is reduced to 1.477 mm.

With the application of computer vision in the field of agricultural disease recognition, the convolutional neural network is widely used in grape leaf disease recognition and has achieved remarkable results. However, most of the grape leaf disease recognition models have the problem of weak generalization ability. In order to overcome this challenge, this paper proposes an image identification method for grape leaf diseases in different domains based on Fourier domain adaptation. Firstly, Fourier domain adaptation is performed on the labeled source domain data and the unlabeled target domain data. To decrease the gap in distribution between the source domain data and the target domain data, the low-frequency spectrum of the source domain data and the target domain data is swapped. Then, three convolutional neural networks (AlexNet, VGG13, and ResNet101) were used to train the images after style changes and the unlabeled target domain images were classified. The highest accuracy of the three networks can reach 94.6%, 96.7%, and 91.8%, respectively, higher than that of the model without Fourier transform image training. In order to reduce the impact of randomness, when selecting the transformed image, we propose using farthest point sampling to select the image with low feature correlation for the Fourier transform. The final identification result is also higher than the accuracy of the network model trained without transformation. Experimental results showed that Fourier domain adaptation can improve the generalization ability of the model and obtain a more accurate grape leaf disease recognition model.

A novel DC-assisted fast hot-pressing (FHP) powder sintering technique was utilized to prepare Al/Diamond composites. Three series of orthogonal experiments were designed and conducted to explore the effects of sintering temperature, sintering pressure, and holding time on the thermal conductivity (TC) and sintering mechanism of an Al-50Diamond composite. Improper sintering temperatures dramatically degraded the TC, as relatively low temperatures (≤520 °C) led to the retention of a large number of pores, while higher temperatures (≥600 °C) caused unavoidable debonding cracks. Excessive pressure (≥100 MPa) induced lattice distortion and the accumulation of dislocations, whereas a prolonged holding time (≥20 min) would most likely cause the Al phase to aggregate into clusters due to surface tension. The optimal process parameters for the preparation of Al-50diamond composites by the FHP method were 560 °C-80 MPa-10 min, corresponding to a density and TC of 3.09 g cm−3 and 527.8 W m−1 K−1, respectively. Structural defects such as pores, dislocations, debonding cracks, and agglomerations within the composite strongly enhance the interfacial thermal resistance (ITR), thereby deteriorating TC performance. Considering the ITR of the binary solid-phase composite, the Hasselman–Johnson model can more accurately predict the TC of Al-50diamond composites for FHP technology under an optimal process with a 3.4% error rate (509.6 W m−1 K−1 to 527.8 W m−1 K−1). The theoretical thermal conductivity of the binary composites estimated by data modeling (Hasselman–Johnson Model, etc.) matches well with the actual thermal conductivity of the sintered samples using the FHP method.

The utilisation of numerical process simulation has greatly facilitated the challenging task of liquid composite moulding (LCM) process optimisation, providing ease of solution evaluation at a significantly reduced cost compared to complete reliance on physical prototyping. However, due to the process complexity, such process simulation is still considerably expensive at present. In this paper, cost-saving approaches to minimising the computational cost of simulation-based optimisation for LCM processes are compiled and discussed. Their specific applicability, efficacy, and suitability for various optimisation/moulding scenarios are extensively explored in detail. The comprehensive analysation and assimilation of their operation alongside applicability for the problem domain of interest are accomplished in this paper to further complement and contribute to future simulation-based optimisation capabilities for composite moulding processes. The importance of balancing the cost-accuracy trade-off is also repeatedly emphasised, allowing for substantial cost reductions while ensuring a desirable level of optimization reliability.

Halides play important roles in human health and environmental monitoring. However, different halides interfere with each other in current measurement methods. Simultaneous sensing of multiple halides in a fast and low-cost manner remains a challenge. Here, we report a fluorometric multi-halide sensing method by using a single citrate-based fluorophore, CA-Cys, on a custom-made portable device. The fluorescence emitted by CA-Cys is quenched due to the dynamic quenching of halide ions; the sensitivities vary from halide types and pH, providing the capability to obtain multiple Stern-Volmer equations at various pH values. The concentration of each halide can then be obtained by solving the resultant set of equations. A mM scale detection limit is demonstrated, which is suitable for halide wastewater monitoring. A proof-of-concept smartphone-based portable device is also fabricated and tested. The results from the fluorometer and portable device indicated that our multi-halide system is promising for real-world multi-halide sensing applications. This work represents a new direction in developing portable, low-cost, and simultaneous multi-halide sensing technologies.

The magnetic domains of non-oriented electrical steel bearing cumulative thermal compressions made by a Gleeble 3500 Thermal System were observed using an atomic force microscope (AFM). The component forces, comprising the magnetic forces between the AFM probe and magnetic domains of the samples, along the freedom direction of the probe, were measured, and they formed the value fluctuation of the magnetic domains. The fluctuations of the magnetic domains were analyzed by examining the power spectral density (PSD) curves. The hysteresis curves of the samples were measured using a highly sensitive magnetic measurement system. An analysis of the magnetic force microscope (MFM) maps suggested that some magnetic domains were compressed into crushed and fragmented shapes, similar to the microstructure of deformed grains. Meanwhile, some were reconstructed within the thermal compressions, like dynamic recrystallization microstructures. Meaningfully, the MFM probe moved and deformed the proximal magnetic domains of tested samples within the region of its weak magnetic field. The peak positions of the magnetic domains with a high deformation rate were shifted and moved during the measuring processes by the weakly polarized probe. Both windward and leeward sides simultaneously expressed a slope towards each co-adjacent valley in the MFM maps and induced a statistical throbbing within a narrow band in the PSD curves. Thus, the MFM scanning mode was also analyzed and improved to obtain accurate MFM maps with low disturbances from the weak magnetic field of the probe. Swapping the order positions of the middle processes in the MFM scanning and adding a gliding step between them could offset the peak skewing of magnetic domains caused by the weakly polarized probe during MFM measurement process without incurring excessive replacement costs. Accumulative compression at a high rate (10 s−1) would crush magnetic domains into irregularly decreasing sizes with messy boundaries. This investigation provides an example of the complete relationships among deformations, magnetic domains, and magnetic properties.

Combined scanning tunnelling and atomic force microscopy using a qPlus sensor enables the measurement of electronic and mechanic properties of two-dimensional materials at the nanoscale. In this work, we study hexagonal boron nitride (h-BN), an atomically thin 2D layer, that is van der Waals-coupled to a Cu(111) surface. The system is of interest as a decoupling layer for functional 2D heterostructures due to the preservation of the h-BN bandgap and as a template for atomic and molecular adsorbates owing to its local electronic trapping potential due to the in-plane electric field. We obtain work function (Φ) variations on the h-BN/Cu(111) superstructure of the order of 100 meV using two independent methods, namely the shift of field emission resonances and the contact potential difference measured by Kelvin probe force microscopy. Using 3D force profiles of the same area we determine the relative stiffness of the Moiré region allowing us to analyse both electronic and mechanical properties of the 2D layer simultaneously. We obtain a sheet stiffness of 9.4 ± 0.9 N·m−1, which is one order of magnitude higher than the one obtained for h-BN/Rh(111). Using constant force maps we are able to derive height profiles of h-BN/Cu(111) showing that the system has a corrugation of 0.6 ± 0.2 Å, which helps to demystify the discussion around the flatness of the h-BN/Cu(111) substrate.

The hydroturbine runner is the core of the whole hydroelectric generating unit, which is employed to transform water energy into mechanical energy. In the process of service, the runner frequently suffers from abrasion due to erosion and cavitation. Defects are usually repaired by welding. To acquire suitable weld cladding repair process parameters, a combination of experimental and numerical simulation was applied to investigate the temperature and weld residual stress distribution in the repair zone under the different welding repair approaches. The results illustrate that the temperature and welding residual stress distribution of the blade and the shroud are out of symmetry, the temperature conduction rate is faster on the blade side, and the high-stress zone is predominantly concentrated in the weld and its adjacent area. When the preheating temperature is up to 150 °C, the peak value of welding residual stress reaches a minimum of 796.29 MPa. The welding sequence can adjust the distribution trend of welding residual stresses. The welding sequence of three-stage welding can effectively reduce the welding residual stresses near the shroud at the water outlet side of the blade. The results of the study will provide theoretical guidance for the welding repair of hydraulic turbine runners.

Emotion recognition is receiving significant attention in research on health care and Human-Computer Interaction (HCI). Due to the high correlation with emotion and the capability to affect deceptive external expressions such as voices and faces, Electroencephalogram (EEG) based emotion recognition methods have been globally accepted and widely applied. Recently, great improvements have been made in the development of machine learning for EEG-based emotion detection. However, there are still some major disadvantages in previous studies. Firstly, traditional machine learning methods require extracting features manually which is time-consuming and rely heavily on human experts. Secondly, to improve the model accuracies, many researchers used user-dependent models that lack generalization and universality. Moreover, there is still room for improvement in the recognition accuracies in most studies. Therefore, to overcome these shortcomings, an EEG-based novel deep neural network is proposed for emotion classification in this article. The proposed 2D CNN uses two convolutional kernels of different sizes to extract emotion-related features along both the time direction and the spatial direction. To verify the feasibility of the proposed model, the pubic emotion dataset DEAP is used in experiments. The results show accuracies of up to 99.99% and 99.98 for arousal and valence binary classification, respectively, which are encouraging for research and applications in the emotion recognition field.

In this paper, a digital hydraulic variable control method based on multi-way valve spool displacement feedback is proposed, which combines the advantages of low cost and high reliability of the loader fixed displacement pump hydraulic system, and the distinguished energy saving effect of the loader variable pump hydraulic system. Based on the principle of hydraulic resistance, the mathematical model of flow-pressure of multi-way valve was established. Meanwhile, the theoretical structure of a valve spool suitable for digital hydraulic system was deduced. On this basis, an improvement scheme of the valve spool was proposed in combination with machining feasibility. The simulation model of multi-way valve was established, the correctness of which was verified by experiment with the significance test method, and the valve port parameters of the multi-way valve were optimized by genetic algorithm. The simulation model before and after optimization was analyzed by AMESim software. The simulation results showed that the digital hydraulic system can basically realize the function of a variable system, and the energy consumption was significantly reduced compared with the fixed displacement pump hydraulic system. The optimized valve spool structure of the digital hydraulic system had more stable load characteristics than the fixed displacement pump system and a larger speed range. The increase range accounts for about 27% of the total stroke.

Supercritical CO2 (S-CO2) Brayton cycles have become an effective way in utilizing solar energy, considering their advantages. The presented research discusses a parametrized analysis and systematic comparison of three S-CO2 power cycles coupled with parabolic trough collectors. The effects of turbine inlet temperature and pressure, compressor inlet temperature, and pressure on specific work, overall efficiency, and cost of core equipment of different S-CO2 Brayton cycles are discussed. Then, the two performance criteria, including specific work and cost of core equipment, are compared, simultaneously, between different S-CO2 cycle layouts after gaining the Pareto sets from multi-objective optimizations using genetic algorithm. The results suggest that the simple recuperation cycle layout shows more excellent performance than the intercooling cycle layout and the recompression cycle layout in terms of cost, while the advantage in specific work of the intercooling cycle layout and the recompression cycle layout is not obvious. This study can be useful in selecting cycle layout using solar energy by the parabolic trough solar collector when there are requirements for the specific work and the cost of core equipment. Moreover, high turbine inlet temperature is recommended for the S-CO2 Brayton cycle using solar energy.

Supercritical carbon dioxide (SCO2) is of great use in miniature power systems. It obtains the characteristics of high density and low viscosity, which makes it possible to build a compact structure for turbomachinery. For a turbine design, an important issue is to figure out the appropriate solidity of the rotor. The objective of this research is to present the aerodynamic performance and provide the design reference for SCO2 and air radial-inflow turbines considering different solidity structures. For the low solidity case of SCO2 turbine, new splitter structures are proposed to improve its performance. The automatic design and simulation process are established by batch modes in MATLAB. The numerical investigation is based on a 3D viscous compressible N-S equation and the actual fluid property of SCO2 and air. The distributions of flow parameters are first presented. Rotor blade load and aerodynamic force are then thoroughly analyzed and the aerodynamic performances of all cases are obtained. The SCO2 turbine has larger power capacity and higher efficiency while the performance of the air turbine is less affected by rotor solidity. For both SCO2 and air, small solidity can cause the unsatisfactory flow condition at the inlet and the shroud section of the rotor, while large solidity results in the aerodynamic loss at the trailing edge of rotor blade and the hub of rotor outlet. A suction side offset splitter can greatly improve the performance of the low solidity SCO2 turbine.

Research on concentrated solar power (CSP) plants has been increasing in recent years. Supercritical carbon dioxide (S-CO2) has been applied to solar power plants due to its promising physical properties. S-CO2 has a relatively low critical temperature of 31.1 °C and owns high density in the supercritical region. Hence, it is a vital working fluid in the application of low temperature heat source and miniature power equipment. Due to the fact that solar power system has a constantly changing heat source according to season and weather, a satisfactory off-design performance is necessary for the turbine in a solar power system. In this work, a S-CO2 radial-inflow turbine based on CSP is designed. A thorough numerical analysis of the turbine is then performed. To investigate the off-design performance of this turbine, three types of nozzle profiles with different leading edge diameters are adopted. Mach number, temperature and pressure distribution are covered to present the off-design effect with different nozzle profiles. Moreover, the relation of output power, mass flow rate and efficiency with different leading edge diameter (LED) are analyzed. Results show that different LED has a vital influence on the aerodynamic characteristics and off-design performance of the S-CO2 turbine based on CSP. In addition, the designed turbine with LED = 4 mm can obtain the highest mass flow rate and output power. While the turbine with LED = 10 mm provides slightly better off-design efficiency for CSP plants.

To meet the challenge of large-scale renewable energy penetration and take full advantage of existing AC infrastructure, the bipolar DC distribution system is of interest. In this article, the system structure and characteristics of the bipolar DC distribution network are proposed. The three-level Neural Point Clamped Converter (NPC) is used in the proposed system to construct the bipolar DC system. To optimize the DC system performance, an improved cooperative control and energy management strategy is proposed mainly to mitigate DC voltage fluctuation and balance the positive and negative phase voltage. The improved strategy consists of (1) 2-degree of freedom (2DOF) PID controller in traditional voltage control loop; (2) cooperative controller to take full advantage of storage system; (3) voltage equalization controller to balance two-phase voltages; and (4) the energy management system to dispatch the response job to batteries and supercapacitors. Experiments and simulations are performed to validate the effectiveness of the proposed strategy.