Nonlinear adaptive filtering using kernel‐based algorithms with dictionary adaptation

Nonlinear adaptive filtering has been extensively studied in the literature, using, for example, Volterra filters or neural networks. Recently, kernel methods have been offering an interesting alternative because they provide a simple extension of linear algorithms to the nonlinear case. The main drawback of online system identification with kernel methods is that the filter complexity increases with time, a limitation resulting from the representer theorem, which states that all past input vectors are required. To overcome this drawback, a particular subset of these input vectors (called dictionary) must be selected to ensure complexity control and good performance. Up to now, all authors considered that, after being introduced into the dictionary, elements stay unchanged even if, because of nonstationarity, they become useless to predict the system output. The objective of this paper is to present an adaptation scheme of dictionary elements, which are considered here as adjustable model parameters, by deriving a gradient‐based method under collinearity constraints. The main interest is to ensure a better tracking performance. To evaluate our approach, dictionary adaptation is introduced into three well‐known kernel‐based adaptive algorithms: kernel recursive least squares, kernel normalized least mean squares, and kernel affine projection. The performance is evaluated on nonlinear adaptive filtering of simulated and real data sets. As confirmed by experiments, our dictionary adaptation scheme allows either complexity reduction or a decrease of the instantaneous quadratic error, or both simultaneously. Copyright © 2015 John Wiley & Sons, Ltd.     

Fault estimation and accommodation for networked control systems with nonuniform sampling periods

This paper deals with the problem of fault estimation and accommodation for a class of networked control systems with nonuniform uncertain sampling periods. Firstly, the reason why the adaptive fault diagnosis observer cannot be applied to networked control systems is analyzed. Based on this analysis, a novel robust fault estimation observer is constructed to estimate both continuous‐time fault and system states by using nonuniformly discrete‐time sampled outputs. Furthermore, using the obtained states and fault information, a nonuniformly sampled‐data fault tolerant control law is designed to preserve the stability of the closed‐loop system. The proposed scheme can not only guarantee the impact of continuous‐time uncertainties and discrete‐time sampled estimation errors on the faulty system to satisfy a H _∞ performance index but also repress the negative effect of the unknown intersample behavior of continuous‐time fault by use of an inequality technique. Finally, simulation results are included to demonstrate the feasibility of the proposed method. Copyright © 2014 John Wiley & Sons, Ltd.     

An improved particle swarm optimisation for unit commitment in microgrids with battery energy storage systems considering battery degradation and uncertainties

Using electric storage systems (ESSs) is known as a viable strategy to mitigate the volatility and intermittency of renewable distributed generators (DGs) in microgrids (MGs). Among different electric storage technologies, battery energy storage (BES) is considered as the best option. In unit commitment (UC) module, the set of committed dispatchable DGs along with their power, power exported to/imported from macrogrid and status and power of ESS units are determined. In this paper, BES degradation is considered in UC formulation and an efficient particle swarm optimisation with quadratic transfer function is proposed for solving UC in BES‐integrated MGs, while the uncertainties of demand, renewable generation and market price are considered and dealt with robust optimisation. UC is formulated as a multi‐objective optimisation problem whose objectives are MG operation cost and BES degradation. The resultant multi‐objective optimisation problem is converted into a single‐objective optimisation problem and the effect of weight factors on MG operation cost and BES lifecycle are investigated. The results show that by consideration of BES degradation in objective function, BES lifecycle increases from 350 to 500 and the minimum depth of charge increases from 5.5% to 34%; however, MG operation cost increases from $8717 to $8910.2. The results also show that by consideration of uncertainties, MG's operation cost increases by 8.22%.     

The VALU3S ECSEL project: Verification and validation of automated systems safety and security

Manufacturers of automated systems and their components have been allocating an enormous amount of time and effort in R&D activities, which led to the availability of prototypes demonstrating new capabilities as well as the introduction of such systems to the market within different domains. Manufacturers need to make sure that the systems function in the intended way and according to specifications. This is not a trivial task as system complexity rises dramatically the more integrated and interconnected these systems become with the addition of automated functionality and features to them. This effort translates into an overhead on the V&V (verification and validation) process making it time-consuming and costly. In this paper, we present VALU3S, an ECSEL JU (joint undertaking) project that aims to evaluate the state-of-the-art V&V methods and tools, and design a multi-domain framework to create a clear structure around the components and elements needed to conduct the V&V process. The main expected benefit of the framework is to reduce time and cost needed to verify and validate automated systems with respect to safety, cyber-security, and privacy requirements. This is done through identification and classification of evaluation methods, tools, environments and concepts for V&V of automated systems with respect to the mentioned requirements. VALU3S will provide guidelines to the V&V community including engineers and researchers on how the V&V of automated systems could be improved considering the cost, time and effort of conducting V&V processes. To this end, VALU3S brings together a consortium with partners from 10 different countries, amounting to a mix of 25 industrial partners, 6 leading research institutes, and 10 universities to reach the project goal.     

Optimized techniques for driving and control of the switched reluctance motor to improve efficiency

This work presents modeling, driving and classical speed control techniques for the switched reluctance motor. The aim is to improve the computational model, the control response and the machine efficiency. A parametric regression model was used to find the inductance profile of the switched reluctance motor and from the new inductance profile model. The drive and control techniques are shown: (i) with speed control acting on the excitation voltage and fixed switching angles, (ii) with speed control acting on the switching angles and fixed excitation voltage and (iii) with speed control acting on the excitation voltage, in this case, with dynamic switching angles and controller parameters. The inductance profile is represented by expression and inserted into the machine computer model, allowing greater precision and low computational cost. The speed control acting on the excitation voltage with dynamic controller parameters and dynamic switching angles allowed: (i) shorter response time for a wide range of control, (ii) higher efficiency, (iii) low computational cost and (iv) simplified implementation and maintenance. The techniques proposed in this work obtained precision of the computational model with respect to the system (in workbench) and optimized parameters in a wide range of the speed control, allowing an improvement of switched reluctance motor efficiency.     

Cloud based cyber-physical systems: Network evaluation study

In recent years, the Industry 4.0 concept brings new demands and trends in different areas; one of them is distributing computational power to the cloud. This concept also introduced the Reference Architectural Model for Industry 4.0 (RAMI 4.0). The efficiency of data communications within the RAMI 4.0 model is a critical issue. Aiming to evaluate the efficiency of data communication in the Cloud Based Cyber-Physical Systems (CB-CPS), this study analyzes the periods and data amount required to communicate with individual hierarchy levels of the RAMI 4.0 model. The evaluation of the network properties of the communication protocols eligible for CB-CPS is presented. The network properties to different cloud providers and data centers’ locations have been measured and interpreted. To test the findings, an architecture for cloud control of laboratory model was proposed. It was found that the time of the day; the day of the week; and data center utilization have a negligible impact on latency. The most significant impact lies in the data center distance and the speed of the communication channel. Moreover, the communication protocol also has impact on the latency. The feasibility of controlling each level of RAMI 4.0 through cloud services was investigated. Experimental results showed that control is possible in many solutions, but these solutions mostly cannot depend just on cloud services. The intelligence on the edge of the network will play a significant role. The main contribution is a thorough evaluation of different cloud providers, locations, and communication protocols to provide recommendations sufficient for different levels of the RAMI 4.0 architecture.     

皮肤T细胞淋巴瘤局部用药和物理治疗的研究进展

皮肤T细胞淋巴瘤(cutaneous T cell lymphoma,CTCL)大多属于惰性淋巴瘤,病情进展缓慢,患者确诊时多处于疾病早期。局部用药和物理治疗在CTCL治疗中占重要地位,包括局部外用皮质类固醇激素、免疫抑制剂、维A酸,放疗及光疗等。近年来,有研究对CTCL传统局部用药和物理治疗方法的有效性和安全性进行了重新评估。同时,新型CTCL局部用药和物理治疗策略也不断涌现,如局部外用他扎罗汀、瑞喹莫特,紫外线A联合新型光敏剂的光化学疗法和光动力治疗等。本文对CTCL局部用药和物理治疗的研究进展作一综述。     

Association between patient psychosocial characteristics and receipt of in‐center nocturnal hemodialysis among prevalent dialysis patients

Introduction: Compared to traditional in‐center hemodialysis (HD), in‐center nocturnal dialysis (INHD) is characterized by longer sessions and nighttime administration, which may lead to better outcomes for some patients. Given the importance of patient choice in the decision to initiate INHD, we explored associations between patients’ psychosocial characteristics and their receipt of INHD. Methods: Among hemodialysis patients at a medium‐sized dialysis organization, we identified INHD patients as those for whom ≥80% of dialysis sessions were INHD sessions—starting at 6:30 pm or later and lasting ≥5 hours—over the 3 months (≥20 sessions total) after their first INHD session. We extracted dialysis session data from electronic medical records and psychosocial data from social worker assessments. We tested associations of patients’ psychosocial characteristics—as well as demographic and clinical characteristics—with INHD receipt among all hemodialysis patients (INHD and HD) in bivariate analyses and multivariable logistic regression models. Findings: Among 759 patients with complete data, we identified 47 (6.2%) as INHD patients. On average, these patients were more likely than HD patients to be employed (full‐time 10.6% vs. 5.2%; part‐time 17.0% vs. 4.2%; P < 0.001), and they were significantly less likely to require ambulatory assistance (14.9% vs. 39.6%, P < 0.001). In multivariable regressions, we found that part‐time employment (versus being unemployed) was associated with a 7.1 percentage‐point higher likelihood of being an INHD patient (P = 0.01), and the negative association with ambulatory assistance needs approached statistical significance (P = 0.056). No other psychosocial factors included in this main regression analysis were statistically significantly associated with INHD patient status. Discussion: Researchers comparing the outcomes of patients undergoing INHD versus other treatment modalities will need to account for differences in employment status—and other factors like requiring ambulatory assistance and age which may predict the ability to work—between INHD users and comparison patients to avoid bias in estimates.     

A decision fusion based methodology for fault Prognostic and Health Management of complex systems

Prognostic and Health Management (PHM) represents an active field of research and a major scientific challenge in many domains. It usually focuses on the failure time or the Remaining Useful Life (RUL) prediction of a system.This paper presents a generic framework, based on a discrete Bayesian Network (BN), particularly tailored for decision fusion of heterogeneous prognostic methods. The BN parameters are computed according to the fixed prognostic objectives.The effectiveness of the proposed decision fusion based methodology for the prognostic is demonstrated through the RULs estimation of turbofan engines. The application highlights the ability of the approach to estimate RULs which overpasses the performance of most other published results in the literature.     

具有双通道数据包丢失的Delta算子系统故障检测

研究同时存在双通道数据包丢失和时变时延的Delta算子网络控制系统（NCSs）故障检测问题.假定数据包丢失发生在控制器到执行器、传感器至控制器的数据传输过程中，并且利用两个相互独立的伯努利随机变量描述是否发生丢包.将上述的NCSs建模为网络切换系统，提出任意切换律下故障检测滤波器的设计方法.利用线性矩阵不等式（LMIs）方法、Lyapunov-Krasovskii泛函和平均驻留时间等得出所考虑的网络切换系统具备指数均方稳定性的充分条件.证明了所用的网络切换系统满足H_∞性能，并推导出了滤波器参数的显式表达.数值仿真结果验证了所提方法的有效性.