In industrial processes, the accurate pinpointing of root causes is pivotal to ensure timely and precise interventions, ultimately culminating in the mitigation of underlying disruptions. While root cause diagnosis (RCD) methods endeavor to transcend the limitations of traditional fault diagnosis approaches by tracing origins, they often grapple with the labyrinthine causal interdependencies interspersed among multivariate time series. The presence of redundant, bidirectional, or even cyclic relationships further convolutes the diagnostic landscape. To address these challenges, we present an innovative RCD methodology designed specifically for intricate industrial processes. Our unique RCD framework integrates node state surveillance, grounded in a pre-fault causality graph, circumventing the customary delays associated with amassing sufficient post-fault samples in extant RCD techniques. Importantly, our method adeptly handles cyclic causality graphs, a recurrent challenge in the feedback controls of complex systems. For pre-fault causality construction, we propose asymmetric sparsemax-driven predictive modeling (ASPM), which refines the sparsity and asymmetry of causality graphs, bolstering the reliability and clarity of RCD outcomes. Central to our method is a pioneering graph structure learning module, skilled in removing unnecessary connections while preserving vital predictive interrelationships in multivariate time series. Empirical validation with the continuous stirred-tank reactor (CSTR) process and blast furnace (BF) ironmaking process confirms the enhanced efficacy of our proposed RCD approach in industrial settings.

Electric vehicles (EVs) serve as a critical link between energy and transportation networks (TNs). However, previous research on TN coordination has often overlooked integrated energy systems (IESs), focusing primarily on power networks, which is unable to ensure the secure operation of IES due to the complexities arising from the conversion among various energy sources, and it does not facilitate flexible multi-energy management for the IES. Furthermore, the lack of consideration for mobile load scheduling flexibility in the TN and collaborative operations compromises overall cost-effectiveness. To this end, this study proposes a coordinated IES and TN operation model to achieve economic scheduling and flexible multi-energy management with comprehensive secure operation constraints. In particular, the constituent networks of the IES and TN are modeled individually, after which a transactive relationship model is proposed to interconnect the networks. To lower the computational intractability, several approximation and relaxation methods are proposed to reformulate the model into a mixed-integer-quadratic programming (MIQP) problem. Three case studies are conducted to examine the collaborative operation of IES and TN, demonstrating its advantages over independent operation in terms of reducing consumption and alleviating congestion for flexible multi-energy management.

The increasing penetration of electric vehicles (EVs) gradually couples the traffic network (TN) and the power distribution network (PDN), which brings both opportunities and challenges for decarbonization. In this research, a coupled power-traffic networks operation model is proposed to achieve the goals of cost reduction and low-carbon emissions, as well as minimize each entity’s self-consumption. Thus, a bi-level game framework is adopted to model competitive behaviors. Specifically, the upper-level TN operator (TNO) determines monetary incentives (i.e., charging prices and carbon taxes) to guide vehicles, and the PDN operator (PDNO) cooperates with the TNO to optimize power flows. While the lower-level EV driver group (EVDG) and gasoline vehicle driver group (GVDG) make route choices with informed incentives to minimize consumption, respectively. A game theoretic approach is adopted to prove the existence of route selection solutions in the non-cooperative competitive behavior between EVDG and GVDG. Moreover, a carbon-tax-based pricing scheme is proposed to schedule vehicles in an economic and low-carbon manner. The bi-level model with four entities (i.e., PDNO, TNO, EVDG, and GVDG) is solved by a decentralized algorithm to identify the optimal operational state. Numerical results validate the effectiveness of the proposed scheme compared with several other schemes.

The rapid proliferation of electric vehicles (EVs) has promoted the process of electrified transportation, which deepened the interdependency of power and transportation networks (TNs). Considering the routing preference of EV travelers, this article proposes an environment-aware dispatch model to coordinate coupled power-TNs (CPTNs) toward higher economic benefits and renewable energy utilization. Specifically, discrete choice models with environmental factor are developed to characterize the user behavior. Based on the constructed user behavior models, an integrated optimal traffic-power flow (IOTPF) model is proposed and an environment-aware user equilibrium (EUE) can be achieved. A decentralized optimization algorithm based on Gauss-Seidel iterative process is developed to solve the equilibrium flows. Numerical results analyze the influence of environment-aware user behavior on the traveling cost and the utilization of renewable energy, the effectiveness of the IOTPF model in coordinating the optimal strategy of the coupled network is demonstrated.

In the realm of industrial process automation, the escalating complexity of systems and the surge of data from developments like the Industrial Internet of Things (IIoT) have underscored the urgent need for efficient and precise data-driven fault diagnosis. Traditional methods, despite advancements through deep learning, often struggle to differentiate between various fault categories. This issue is exacerbated in industrial contexts where acquiring and labeling fault data is a resource-intensive task. In response, our innovative approach presents an self-supervised framework, namely contrastive clustering-assisted discriminative feature learning (CCDFL), that ingeniously combines a temporal gate encoder (TGE) for efficient time-series data processing and feature extraction, with a contrastive clustering-assisted discriminator (CCD) leveraging contrastive learning for improved feature discrimination. This synergistic fusion is crucial for accurately identifying and classifying different fault types, marking a significant advancement in intelligent manufacturing and fault diagnosis. It is particularly adept at handling diverse and unforeseen faults, offering practical solutions in complex scenarios. The effectiveness of this method is further substantiated in the experimental section of our study.

With the increasing popularization of information and communication technologies, the development of smart grids has also emerged. However, the high reliance on information technology puts the grid at risk of malicious attacks. False Data Injection Attack (FDIA)have become a serious threat to the smart grid. While most of the current methods for detecting FDIA focus on determining whether the grid has been subjected to such attacks, they often overlook the importance of considering the topology and temporal correlations within the grid data and are unable to localize their attacks. To address this limitation, we have proposed a novel multi-label attack localization detection method based on spatio-temporal features. Our method considers both time and space and enables the localisation of attacks. Spatial feature extraction is conducted using a Convolutional Neural Network (CNN), which is subsequently subjected to feature filtering through an Attention Mechanism. The filtered features are then channeled into both Long-Term and Short-Term Memory network (LSTM) via a channel fusion process. This intricate process allows for the extraction of spatio-temporal features, ultimately facilitating the accurate localization of attacks. To evaluate the efficacy of our approach, we conducted extensive simulation experiments using the IEEE 14bus system and the IEEE 118-bus system. The results showcase remarkable performance with an overall accuracy of 99.6% for the former and 98.7% for the latter, underscoring the effectiveness of our proposed model.

Millimeter-wave (MMW) radar sensing, heralded as one of the most promising technologies for ensuring secure navigation in autonomous vehicles, is renowned for its high-resolution imaging capabilities and adaptability to diverse environmental and lighting conditions. This article presents a 3-D MMW imaging radar, which is specifically designed and employed for environmental mapping. It employs a full heterodyne detection technique, utilizing a sawtooth linear frequency modulated continuous wave (FMCW) approach with a central frequency of 94 GHz. Its minimum detectable radar cross section (RCS) is $-$ 82.7 dBm, and its range resolution is 15 cm. Experimental comparisons of outdoor scenarios employing different antennas are conducted, and their performances are meticulously analyzed. These findings offer valuable insights into the application of MMW technology in the realm of environmental mapping.

PID control is still the most important and popular method in industrial control at present. PID control is easy to achieve and it can improve the steady-state performance and dynamic performance of the system. PID controller can be used for all objects, however, it has some problems with parameter adjustment and control effect. The proportional integral differential coefficient of PID control is fixed, and it can't change when disturbed, so the stability of the system will be affected. Moreover, PID control is prone to overshoot and can't be used in specific systems. Reinforcement learning (RL) algorithms have developed rapidly from discrete action to continuous action in recent years. It has aroused the high interest of researchers in the field of automatic control. RL control performs better in the degree of intelligence and dynamic performance, however, the steady-state performance is poor. The sensitive response of RL control will damage the actuator. In this paper, an adaptive PID controller based on deep reinforcement learning is proposed. By designing reward values, the desired control effect is described. In this way, an agent is trained to provide parameters to the PID controller in real time. It can improve the response speed of the system, suppress overshoot, and have a certain anti-disturbance ability by training the agent to achieve real-time PID parameter adjustment.

Abstract As the most promising alternative to internal combustion engines (ICEs), electric vehicles (EVs) have an excellent development outlook. The charging route scheduling of EVs can simultaneously affect traffic congestion in the transportation network (TN) and power flow distribution in the power distribution network (PDN). The research on TN and PDN coupling networks based on the static traffic flow model is relatively mature; however, it ignores that the traffic flow will spread across periods in a short scheduling period. In this paper, a semi‐dynamic traffic flow model is proposed to represent the dynamic propagation characteristics of EVs and ICEs flow. Furthermore, the cost of carbon emission and system operation are combined as the overall goal of system optimisation. Since the model has become a more complex non‐linear model, this paper proposes to combine the heuristic sequential boundary tightening and binary expansion method to linearise the model. The study compared four cases and found that a 20% penetration rate of EVs can reduce carbon emissions by 4.2% while reducing the system's total cost by 10%. Moreover, the impact of network congestion on the spatiotemporal distribution of traffic flow and power flow in the coupled network is alleviated.

In this paper, we use millimeter wave radar to conduct outdoor experiments under different air quality conditions and analyze the degree of millimeter wave transmission attenuation. The radar we use operates at 94 GHz with a bandwidth of 1 GHz, a distance resolution of 15 cm, and a maximum detection distance of about 150 meters. We mainly use PM2.5 and PM10 indices as air quality criteria and find that the reflected signal strength of millimeter wave from 90m to 150m decreases by 0.38dB when the air quality rises from 136 to 250, indicating that the weaker the air quality, the weaker the millimeter wave signal strength. This provides an experimental basis for the field of millimeter wave communication.

Fault detection plays a pivotal role in ensuring safety and efficiency in process industries. Subspace learning-based fault detection methods have gained recognition for their effective data structure characterization and noise mitigation. However, harnessing the benefits of subspace learning for fault detection tasks and ensuring significant discrimination between normal and fault feature representations to enhance classifier accuracy remain underexplored areas. In this study, we introduce a temporal capsule network (CapsNet) encoder-assisted one-class classifier (TceOne) methodology that enables joint optimization of subspace learning and fault detection. We modify the CapsNet to preserve temporal correlations of multivariate time series in the subspace, thereby enhancing the discriminability between normal and fault subspace representations. Normal subspace representations are confined to a compact region by minimizing the discriminative hypersphere radius of the one-class classifier, leaving fault features sparsely distributed outside the hypersphere. We then establish a specific subspace distance metric that draws normal data closer to the center and distances fault data from it. This metric accounts for the properties of CapsNet instantiation parameters, integrating the variations in both direction and magnitude of subspace representation. We demonstrate the effectiveness and superiority of our proposed methodology through experiments conducted on the Tennessee Eastman (TE) process.

The growing penetration of electric vehicles (EVs) strengthens the interaction between the transportation network (TN) and the power distribution network (PDN). Such interaction brings opportunities for reducing consumption and emissions, while creating challenges for collaborative scheduling. This paper presents a multi-objective optimization strategy, which is formulated to optimize the economy and emissions of the coupled networks in a way that the optimal operations of TN and PDN are guaranteed simultaneously. Specifically, the detailed models of TN comprised of EVs and gasoline vehicles (GVs), and PDN, respectively, are formulated; thus, the coupling relationship can be clarified. Since two objectives compete with each other, an economic and low-carbon scheduling problem (ELCP) is proposed based on the Pareto efficiency theory. To facilitate the model's tractability, several approximation and relaxation methods are exploited to derive a mixed-integer quadratic problem (MIQP). Furthermore, a co-optimization method that combines the epsilon-constraint algorithm is developed to transform the multi-objective problem into its equivalent single-objective problem for obtaining scheduling results. Finally, two test systems are examined to evaluate the effectiveness of the proposed method, and numerical results demonstrate that the economic cost and emissions are effectively balanced.

The increasing popularity of electric vehicles (EVs) has led to a growing need for electrified transportation, which has further deepened the interconnectedness of power and transportation networks. To account for EV drivers' preferences in route selection, this paper proposes an environment-aware dispatch model that coordinates the integrated power and transportation network to achieve greater economic benefits and increased utilization of renewable energy. The model employs discrete choice models with an environmental factor to represent user behavior. The constructed user behavior models serve as the basis for the integrated optimal traffic-power flow (IOTPF) model, which aims to optimize the flow of both power and traffic while minimizing the total cost of EV travel and maximizing the use of renewable energy sources. The paper employs a decentralized optimization algorithm to solve the equilibrium flows of the IOTPF model. Numerical results illustrate how the environment-aware user behavior impacts travel costs and the use of renewable energy, demonstrating the effectiveness of the IOTPF model in coordinating the optimal strategy of the integrated power-transportation network.

It is essential to conduct root cause diagnosis (RCD) to guarantee the safety of operations of chemical processes and suppress fault deterioration, yet RCD has encountered challenges in the context of the increasing complexity of industrial processes. Granger causality (GC) analysis is one of the most commonly used methods to construct process causal maps and identify root causes. However, its use is subject to a number of limitations for dynamic nonlinear industrial processes. Therefore, an optimized GC analysis is proposed to perform RCD in complex industrial processes based on the spatiotemporal coalescent-based prediction model (SCPM) and causality verification algorithm (CVA). The time series prediction model SCPM is presented as an alternative to GC’s conventional autoregressive model in order to avoid spurious regressive facing nonlinear processes, which stacks dilated convolutional neural networks (DCNN) and bidirectional gated recurrent unit (BiGRU) to coalesce time series nonlinear couplings and long-term dependence information. Moreover, considering data that may deviate from normal distribution after the fault occurs, a nonparametric causality test method CVA based on the Wilcoxon signed-rank test (WSRT) is constructed combined with SCPM to eliminate false causality discovery. Empirical results on the Tennessee Eastman process and the blast furnace process demonstrate the effectiveness of the proposed RCD method.

Integrated energy systems (IES) improve overall energy efficiency by synchronizing resources and networks across multiple regions, the high penetration level of electric vehicles (EVs) may pose a threat without coordinating with the IES. Thus, this paper proposes a coordinated IES and transportation network (TN) operation model toward economic scheduling, with the secure operation conditions are comprehensively considered. Specifically, the composition networks of IES and the TN are modeled separately, then a transactive relationship model is proposed to get the networks bridged. To reduce the computational intractability, several approximation and relaxation methods are utilized to transform the model into a mixed-integer-quadratic problem (MIQP). Results show that the coordinated scheduling of IES and TN can lower 4.3% (105.266k$) of consumption compared with the independent scheduling.

In this letter, we designed a hemispherical refractive lens using ice for millimeter wave (MMW) imaging in the environment of −15 °C. Guided by geometrical optics theory, we calculated a suitable thickness of the lens to compensate for the phase of aperture field emitted by the antenna. Then, the 3 dB beamwidth was reduced from 17.6° to 2.1°. Meanwhile, the antenna gain is increased from 17 to 24 dB, which raises the maximum radar detection distance by 18%. Finally, we conducted experiments for MMW imaging to test the ice lens under different road conditions. Measuring results demonstrated that the image error rate has been significantly reduced and the peak signal-to-noise ratio (PSNR) has increased by more than 5 dB, which are consistent with simulated results. It should be noted that the ice lens can improve imaging quality and increase imaging resolution, making it widely applicable in the fields of Pole expedition and snow disaster rescue.

With the extensive penetration of distributed renewable energy and self-interested prosumers, the emerging power market tends to enable user autonomy by bottom-up control and distributed coordination. This paper is devoted to solving the specific problems of distributed energy management and autonomous bidding and peer-to-peer (P2P) energy sharing among prosumers. A novel cloud-edge-based We-Market is presented, where the prosumers, as edge nodes with independent control, balance the electricity cost and thermal comfort by formulating a dynamic household energy management system (HEMS). Meanwhile, the autonomous bidding is initiated by prosumers via the modified Stone-Geary utility function. In the cloud center, a distributed convergence bidding (CB) algorithm based on consistency criterion is developed, which promotes faster and fairer bidding through the interactive iteration with the edge nodes. Besides, the proposed scheme is built on top of the commercial cloud platform with sufficiently secure and scalable computing capacity. Numerical results show the effectiveness and practicability of the proposed We-Market, which achieves 15% cost reduction with shorter running time. Comparative analysis indicates better scalability, which is more suitable for larger-scale We-Market implementation.

Terahertz (THz) imaging has drawn significant attention because THz wave has a unique capability to transient, ultra-wide spectrum and low photon energy. However, the low resolution has always been a problem due to its long wavelength, limiting their application of fields practical use. In this paper, we proposed a complex one-shot super-resolution (COSSR) framework based on a complex convolution neural network to restore superior THz images at 0.35 times wavelength by extracting features directly from a reference measured sample and groundtruth without the measured PSF. Compared with real convolution neural network-based approaches and complex zero-shot super-resolution (CZSSR), COSSR delivers at least 6.67, 0.003, and 6.96% superior higher imaging efficacy in terms of peak signal to noise ratio (PSNR), mean square error (MSE), and structural similarity index measure (SSIM), respectively, for the analyzed data. Additionally, the proposed method is experimentally demonstrated to have a good generalization and to perform well on measured data. The COSSR provides a new pathway for THz imaging super-resolution (SR) reconstruction below the diffraction limit.

The growing penetration of electric vehicles (EVs) strengthens the interaction between the transportation network (TN) and power distribution network (PDN). The randomness of EV charging behavior brings both challenges and opportunities to the coupled network. The goals of cost reduction, efficiency enhancement, and low-carbon emission are expected to be achieved by optimizing the dispatch of traffic flow and power flow. The economy has been studied in the existing research, whereas, environmental protection is ignored. Thus, an economic and low-carbon scheduling problem (ELCP) is formulated to collaboratively optimize the economy and emissions. Specifically, the detailed models of the TN and PDN are formulated, respectively, then the coupling relationship can be clarified. Besides, a multi-objective optimization problem is proposed, whose objectives optimize the overall economy and emissions of the coupled network, respectively. Furthermore, a solution method that combines the epsilon-constraint algorithm with the extreme points of the feasible solution space is proposed to obtain various solutions. A fourteen-link/five-bus TN-PDN is applied to evaluate the proposed method. Simulation results show that the optimal traffic and power flow dispatch schemes can be obtained by solving the ELCP. At the compromise solution point, the economic cost can be reduced by 14.25%, while the corresponding carbon emission only increases by 4.02%.

The increasing penetration of electric vehicles (EVs) gradually evolves the urban transportation network (UTN) and power distribution network (PDN) from being independent to being coupled. To analyze the impact of drivers' behavior on the coupled network, and reveal the relationship between the two networks, a coupled traffic-distribution (CTD) model is proposed, in which the total traveling cost of UTN and the power service cost of PDN is minimized. In the CTD model, A stochastic user equilibrium traffic assignment problem (SUE-TAP) with elastic traffic demand and discrete path selection of drivers is formulated to capture the traffic flow distribution comprised of EVs and gasoline vehicles (GVs). An alternative current optimal power flow (ACOPF) model of PDN is utilized to provide the optimal charging price and scheduling plan. Besides, the SUE-TAP with nonlinear functions is difficult to solve, a novel piecewise linear approximation method is presented to deal with the nonlinear items of the SUE-TAP caused by drivers' traveling cost and traveling tendency incorporating perception. Incorporating the behavioral theory, a distributed coordination method derived from optimal condition decomposition is proposed for the coupled traffic-power network equilibrium. The proposed distributed scheme is examined using a coupled network, which consists of a 20-road UTN and a 33-bus PDN. Numerical results demonstrate the effectiveness and practicability of the proposed model and the distributed coordination method.