To analyse the stable operation range with varied input and output voltages, bifurcation behaviours and operation region of current-mode-controlled single-inductor dual-output (SIDO) boost converter are investigated. The inductor current borders and one-dimensional discrete-mapping model are deduced firstly. On the basis of the model, the bifurcation diagram and corresponding Lyapunov exponent spectrum are analysed. Two transition conditions and boundary equations of the converter are derived, where its working state transits from stability to instability, and its operation mode shifts from continuous conduction mode (CCM) to discontinuous conduction mode (DCM). The operation region with different circuit parameters can be estimated and divided through the parameter space maps, which is significant for the design of circuit parameters and stability control. Time-domain simulations and experimental results are presented to verify the theoretical analysis results. The research results show that the three kinds of bifurcation behaviours including period-doubling bifurcation, border collision bifurcation, and tangential bifurcation exist in the current-mode-controlled SIDO boost converters. Different from the previous works, the SIDO boost converter has a particular bifurcation trend when border collisions occur. The trend includes that operation mode shifts from CCM to DCM and then backs to CCM, and system period increases or decreases doubly.

The ability to perform autonomous exploration is essential for unmanned aerial vehicles (UAV) operating in unstructured or unknown environments where it is hard or even impossible to describe the environment beforehand. However, algorithms for autonomous exploration often focus on optimizing time and coverage in a greedy fashion. That type of exploration can collect irrelevant data and wastes time navigating areas with no important information. In this paper, we propose a method for exploiting the discovered knowledge about the environment while exploring it by relying on a theory of robustness based on Probabilistic Metric Temporal Logic (P-MTL) as applied to offline verification and online control of hybrid systems. By maximizing the satisfaction of the predefined P-MTL specifications of the exploration problem, the robustness values guide the UAV towards areas with more interesting information to gain. We use Markov Chain Monte Carlo to solve the P-MTL constraints. We demonstrate the effectiveness of the proposed approach by simulating autonomous exploration over Amazonian rainforest where our approach is used to detect areas occupied by illegal Artisanal Small-scale Gold Mining (ASGM) activities. The results show that our approach outperform a greedy exploration approach (Autonomous Exploration Planner) by 38% in terms of ASGM coverage. 1 .

The discrete mapping model of current-mode controlled buck converter with constant current load, taking account of composite output capacitors (parallel connection of two different types of capacitor branches, i.e. electrolytic capacitors and ceramic capacitors), is established. Based on the model, dynamical effects of varying output capacitance and equivalent series resistance (ESR) are investigated by bifurcation behaviours. The period of low-frequency oscillation among coexisting fast-slow scale instability is derived by exploring the loci of eigenvalues, while the operating regions are estimated. Time-domain simulation and experimental waveforms are provided for verification of the theoretical analysis, indicating the existences of subharmonic oscillation and coexisting fast-slow scale instability in the converter with variation of output capacitance and ESR. Research results reveal that the low-frequency oscillation can be eventually eliminated due to a relatively large (or small) ESR and the capacitance in the same branch presents to identical tendency of dynamical effects on the converter. Moreover, the interaction effects between two parallel capacitor branches are demonstrated. It illustrates that the low-frequency oscillation can be removed with smaller (or larger) ESR or capacitance in one branch of the composite output capacitors while larger (or smaller) ESR or capacitance in the other branch.

As a time-shifting load that is gradually popularized in the northern region, electric heating load has great adjustment potential. Because the electric heating operation characteristics are affected by many non-linear factors, the traditional equivalent thermal parameters model cannot accurately evaluate the regulation capability of individual electric heating load. Aiming at this problem, this paper proposes an evaluation method for the regulation capability of individual electric heating load based on radial basis function neural network. Firstly, electric heating load control experiments were carried out in a typical room of a residential quarter in winter and relevant experimental data were collected. Then, based on the operation data, the radial basis function neural network is used to evaluate the regulation capability of the individual electric heating load. Finally, the evaluation results based on radial basis function neural network are compared with those based on back propagation neural network and equivalent thermal parameters model. The results show that the proposed method has the least evaluation error and can more accurately evaluate the regulation capability of individual electric heating load.

The CP tensor decomposition is used in applications such as machine learning and signal processing to discover latent low‐rank structure in multidimensional data. Computing a CP decomposition via an alternating least squares (ALS) method reduces the problem to several linear least squares problems. The standard way to solve these linear least squares subproblems is to use the normal equations, which inherit special tensor structure that can be exploited for computational efficiency. However, the normal equations are sensitive to numerical ill‐conditioning, which can compromise the results of the decomposition. In this paper, we develop versions of the CP‐ALS algorithm using the QR decomposition and the singular value decomposition, which are more numerically stable than the normal equations, to solve the linear least squares problems. Our algorithms utilize the tensor structure of the CP‐ALS subproblems efficiently, have the same complexity as the standard CP‐ALS algorithm when the input is dense and the rank is small, and are shown via examples to produce more stable results when ill‐conditioning is present. Our MATLAB implementation achieves the same running time as the standard algorithm for small ranks, and we show that the new methods can obtain lower approximation error.