Search

High-precision load forecasting is crucial for the power system planning and electricity market transactions. Recently, deep learning models have been widely used due to their powerful data mining capabilities. However, the existing research mainly focus on model structure adjustment and input feature selection, ignoring the influence of model ensemble on prediction. A single deep learning model is not yet able to address the various complex challenges that arise in short-term load forecasting. To improve the accuracy of short-term load forecasting, this paper proposes a novel multi-scale ensemble method and multi-scale ensemble neural network. This neural network uses long short-term memory, gate recurrent units, and temporal convolutional network as the basic model. By coupling the stochastic weight averaging ensemble method and differential evolution ensemble method, these deep learning networks were assembled from single-model scale and multi-model scale, respectively, thereby effectively improving the model prediction accuracy. For predicting the power load of Hubei Province in China, meteorological features and time features were in consideration. The proposed model was trained and compared with eleven intelligent short-term load forecasting models, including machine learning, deep learning and ensemble deep learning models. Simulations show that the proposed model has the best comprehensive prediction performance. This study highlights the power of ensemble deep learning models coupled with multiple ensemble techniques and the promising prospect of our proposed model in short-term load forecasting.

Advanced engineering systems possess a large number of components with complicated failure dependencies. To accurately assess the system reliability, the degradation models of components should be known in advance and the model parameters should be accurately estimated via a large quantity of historical time-to-failure data. In real-world situations, due to limited data, lack of knowledge, and vague judgments from experts, components’ degradation model parameters are, however, inevitably encountered with epistemic uncertainty and oftentimes quantified as evidential variables. In this article, upper and lower bounds of system reliability, termed as reliability-box, are estimated when components’ degradation model parameters are elicited from experts and quantified by evidential variables. In the first place, the constrained optimization model is leveraged to assess the reliability-box of each component by giving the evidential variable of the component’s degradation model parameters. Next, based on the system structure, the evidential network of the system is constructed to propagate the epistemic uncertainty from the component level to the system level. Therefore, the focal elements of the evidential variable of system reliability, i.e., the system reliability bounds, can be assessed via the belief and plausibility functions to the mass function of the leaf node of the evidential network. The effectiveness of the proposed methods is demonstrated by a rolling system in the chip cutting detection module.

This paper investigates the transient aerodynamic interference of rotors imposed on fuselage for a quadcopter through Computational Fluid Dynamics (CFD) simulations. The numerical study of transient effects due to rotor rotation is enabled by sliding mesh which defines the rotation domains encompassing rotor blades. The results show that the interference effects of rotor change the aerodynamic forces of the fuselage, causing about 67% in-crease in lift, 13% increase in drag, and 90% increase in pitching up moment on average. The variations of fuselage lift are associated with the pressure distribution changes due to rotors rotation, the high-pressure areas and low-pressure areas over the rotor projects on the arms of the quadcopter causing periodical abrupt changes on the lift, drag and pitching moment.

Reliability of safety instrumented systems (SISs) is a critical measure to ensure production safety of many industries. This paper focus on low-demand SISs. The reliability of these SISs is quantified by evaluating their probability of failure on demand (PFD). However, due to lack of knowledge, and/or vague judgments from experts, epistemic uncertainty associated with the parameters of components' degradation models is inevitable. Meanwhile, common cause failure (CCF) of two or more components caused by shared environments, often exists in SISs, reliability assessment for a SIS, therefore, becomes a challenging task. In this paper, fuzzy reliability assessment for SISs is conducted by taking account of the CCF among components of a SIS. The fuzzy Markov model is utilized to characterize the degradation process of components under fuzzy environment. The fuzzy state probability distribution of components is, then, calculated by formulating a set of constrained optimization models. Based on the fault tree of a SIS, the fuzzy PFD of the entire SIS with CCFs is formulated by using the $\beta $ -factor to quantify the CCFs. The fuzzy PFD at any $\alpha $ -cut level is, therefore, computed by a constrained optimization model. According to the optimization of parameters of SIS, the lower bound and upper bound of fuzzy PFD of SIS can be determined. Finally, we can quantify the effectiveness of CCF event for assessing the fuzzy PFD of SIS.

Hydropower, as a kind of clean energy, has the characteristics of rapid response to load changes, and is often used to undertake peak regulation and frequency regulation tasks. However, due to the uncertainties of reservoir inflow and runoff, the relative lag of power grid outgoing channels, and the interconnection of intermittent new energy, such as wind power and solar energy, the hydropower system is facing enormous peak regulation pressure and the continuous prominent problem of discarding water. In the study, aiming at the problem of risk control of discarding water in power generation process of hydropower station, a practical calculation method of risk of discarding water in power generation is proposed based on the analysis of frequency of runoff. Taking the Ertan hydropower station as an example, the corresponding water level processes of different discarding water risks under the control of pre-flood falling water level, flood limit water level and storage water level after flood season are deduced, and in the routine dispatching process, the water level operation control guidance corresponding to different discarding risk is introduced. The simulation results show that the proposed discarding water risk control method is practical and effective, and can provide guidance for discarding water risk control in the actual dispatching process of hydropower station.

Reservoir optimal operation (ROO) needs to coordinate various profit-making objectives, which is a typical multiobjective optimization problem (MOP) with complex constraints. With the development of multiobjective evolutionary algorithms (MOEAs) in the past decades, more and more research has focused on MOEAs to solve MOP. Considering that multiobjective ROO is also a typical multi-stage Markov decision-making problem, this paper introduces the application of multiobjective dynamic programming (MODP) for multiobjective ROO in detail. On this basis, an improved MODP with selection mechanism of non-dominated solutions based on reference lines (MODP-BRL) is proposed to improve the convergence efficiency of MODP. The experimental results show that the proposed MODP-BRL is a reliable and effective tool in solving multiobjective ROO. In addition, MODP-BRL has better performance in convergence effect and efficiency in comparison experiments with NSGAII, NSGAIII, and SPEA2. It is noteworthy that MODP and MODP-BRL are very sensitive to the discrete step. With the decrease of the discrete step (the higher the discrete precision), the computing time increases nonlinearly. The appropriate discrete step of the state variable is key presets to balance the superiority and computational efficiency of non-dominated solutions with the application of MODP and MODP-BRL.

Obtaining optimal operation rules for cascade reservoirs is crucial for making the most of the comprehensive benefits of reservoirs. However, the operation of cascade reservoirs is generally complex, which involves with challenges including various decision variables, multiple conflicting objectives and constraints. In this paper, a new algorithm named multi-objective tangent algorithm (MOTA) is proposed for optimizing operation rules of cascade reservoirs with the objectives of hydropower generation, ecology and navigation. The performance of MOTA is firstly validated through several well-known benchmark problems. Then it is applied to a case study of cascade reservoirs optimization in China's Pearl River. Using observed inflow data (1956-2015) of cascade reservoirs, a uniform Pareto front is obtained eventually via MOTA after 1000 generations. The optimal operation rules fully considers the comprehensive benefits of hydropower generation, ecology and navigation in China's Pearl River. The optimal operation rules can be used as a guidance tool for decision makers, through the objectives' tradeoff without having to embed a priori preferences in the decision process. Finally, this paper use observed inflow data (2016) of cascade reservoirs for examining the operational rule to comprehend the analysis under different optimal operation rules. The obtained results show that MOTA can be a viable alternative for generating optimal operation rules for cascade reservoirs planning and management.

Laser-assisted simulation is a fast, simple, and effective method for studying dose-rate effects. However, in some compound radiation environments (e.g., the coexistence of transient $\gamma $ -rays and neutrons), the effectiveness of laser simulation has not been studied. This article investigates the feasibility of laser-assisted simulation of the dose-rate effects of neutron-irradiated bipolar transistors by using a sequential radiation method. The accuracy of laser simulation is calculated by using peak photocurrent and charge collection as the equivalent factor, respectively. According to the comparison of the transient parameters (e.g., transient rise/fall time, transient duration time, peak magnitude, collected charge, correlation coefficients) between laser and $\gamma $ -rays, it is proved that the laser-assisted simulation method still works. This means a laser with proper parameters can be used to construct a compound radiation environment with neutron sources, which can provide a powerful solution for the study of the transient co-interaction process of the ionization effects and displacement effects in materials and devices. [ABSTRACT FROM AUTHOR]

A pulsed laser has been effectively applied in dose rate effects simulation for its stable pulse output, low cost, simple operation, and lack of ionizing risks for users. The conversion factor (CF) was proposed to describe the equivalence between the responses generated by pulsed laser and pulsed ionizing radiation (e.g., γ-ray). Previously, the peak photocurrent was commonly used to calculate the CF and establish the equivalent relationship. But there are a number of shortcomings such as narrow effective equivalent range, inadequate information of waveform, as well as low accuracy of simulations. In this article, an improved method for obtaining the CF by using the charge collection as the equivalent factor is proposed, which considers not only the peak photocurrent but also the temporal information of the waveform. The advantages of the proposed method are proved both in theory and in experiments, including broader equivalent range and higher simulation accuracy. [ABSTRACT FROM AUTHOR]

Recently, wearable devices have become increasingly popular in our lives because of their neat features and stylish appearance. However, their tiny sizes bring about new challenges to human-device interaction such as texts input. Although some novel methods have been put forward, they possess different defects and are not applicable to deal with the problem. As a result, we propose an acoustic-based texts-entry system, i.e., EchoWrite, by which texts can be entered with a finger writing in the air without wearing any additional device. More importantly, different from many previous works, EchoWrite runs in a training-free style which reduces the training overhead and improves system scalability. We implement EchoWrite with commercial devices and conduct comprehensive experiments to evaluate its texts-entry performance. Experimental results show that EchoWrite enables users to enter texts at a speed of 7.5 WPM without practice, and 16.6 WPM after about 30-minute practice. This speed is better than touch screen-based method on smartwatches, and comparable with previous related works. Moreover, EchoWrite provides favorable user experience of entering texts. [ABSTRACT FROM AUTHOR]

By processing a large dataset composed of daily trajectories of thousands of students in Singapore, we find that, instead of simply picking up students from their homes, an optimal school shuttle planning system needs to learn the real transportation usage and plan across all potential pickup locations for every student to generate need-satisfying routes. It is challenging, however, to perform route planning over a large number of students each having multiple potential pickup locations. We develop a graph-based data structure that embeds potential pickup locations of all students with the awareness of real-world constraints and existing public transits. Based on the graph structure, we prove that the optimal last-mile school shuttle planning problem is NP-hard and thereafter design a Tabu-based expansion algorithm to solve the problem, which strikes at a proper balance between the savings of students’ commute time and the total cost of operating the shuttle buses. Extensive experiments with large-scale real-world crowdsensed trajectory data demonstrate that our last-mile school shuttles can save the traveling time for most students by over 20% and the savings can be up to 65% for 10% of the students. [ABSTRACT FROM AUTHOR]