Journal of Circuits, Systems and Computers

In this paper, a dynamic parameters lightning search algorithm (DP-LSA) is proposed for optimal operation management of electric vehicle (EV) parking lots (PLs) and distributed generations (DGs) in a microgrid (MG). The MG comprises distributed generation sources such as diesel generators and fuel cells. The batteries of electric vehicles parked in the PL are also used for energy storage. A proposed objective function combines operating, maintenance, and replacement costs, considering inflation, motivation, and annual growth. Simulations are done in two scenarios, and the proposed algorithm results are compared with particle swarm optimization (PSO) algorithm results. Simulation results prove the better performance of the proposed LSA algorithm than the PSO algorithm in reducing the microgrid operation cost.

At present, most network models do not fully consider the prior information on construction project cost, and do not do much research to find out more information or effectively utilize the information the find on construction project cost; conversely, some previous networks mainly consider the depth of the network for the reconstruction results. In addition, there is no detailed discussion on the selection of loss functions for various algorithms, and some common error loss functions are directly used. In view of these shortcomings, a fast estimation model of construction cost based on an improved convolutional neural network is proposed. The construction project is decomposed according to the project characteristics. The project characteristic factors that have a greater impact on the project cost and the main project quantity are selected and analyzed. Then, the convolutional neural network principle is used to analyze the project characteristic factors, the project cost and main project quantities. We simulate and compare the established convolutional neural network models, verify the usability of the model, and finally get a model with high reliability. The establishment of this model provides a new method for the estimation of project cost and main economic indicators, which can meet the actual work requirements to a certain extent.

When gathering optical satellite pictures, light reflected from the surface due to water vapor, snow, fog, haze, and more tiny particles in the environment is generally seen as a gap in the propagation process. Haze has a greater number of suspended particles like aerosols and water droplets. These particles have absorption effects and scattering in the light. Although haze translucency grants a chance for image restoration, a well-organized and broadly relevant haze removal procedure for holding several hazes is quite an extensive provocation. To address this issue, this paper proposed an Optimal Contrast Limited Adaptive Histogram Equalization (OCLAHE) to capture further intricate features and patterns connected to haze, enabling more accurate haze detection and removal. The deeper network can learn complicated descriptions and recognize between hazy and nonhazy regions with higher precision. The proposed method is validated in I-Haze and O-Haze datasets, and its performance is quantified by various performance metrics such as MSE, SSIM, PSNR, WPSNR, and Running time. The experimental consequences demonstrate that the developed model performs better than other techniques and attains an MSE from both datasets as 0.0054 and 0.0051. Overall, the proposed model amends the accuracy and reliability of haze detection in images.

Currently, elevator group control systems containing multiple elevators are used in high-rise buildings, where elevator groups are centrally dispatched according to a set dispatching plan to provide vertical transportation services for the passengers. However, if these elevators work independently, once passengers send out a call signal, multiple elevators will respond to the same request at the same time, and the elevator group will repeatedly go back and forth, which seriously affects the system operation efficiency and also increases energy consumption, so it is a key issue to solve the dispatching problem in the elevator group control system. Elevator group control scheduling is an NP-hard problem with explosive combination characteristics. In this paper, a multi-objective model is formulated based on the criteria of average passenger travel time, average waiting time and system energy consumption for the scheduling problem. Then the particle swarm optimization algorithm is proposed to solve the scheduling problem. Finally, the performance of the proposed algorithm is compared with the genetic algorithm-based elevator group control scheduling, the simulation results show that the proposed algorithm with fast convergence while the waiting time of passengers is significantly decreased.

Considerable advances have been made recently in applying reinforcement learning (RL) to packing problems. However, most of these methods lack scalability and cannot be applied in dynamic environments. To address this research gap, we propose a hybrid algorithm called GraphPack to solve the strip packing problem. Two graph neural networks are designed to fully incorporate the problem’s structure and enhance generalization performance. SkylineNet encodes the geometry of free space as the context feature, while PackNet, supporting the symmetry of rectangles, chooses the next rectangle to pack from the remaining rectangles at each timestep. We conduct fixed-scale, variable rectangle number and variable strip width experiments to test our method. The experimental results show that our method outperforms classical heuristic methods as well as previous RL methods. Notably, our method exhibits strong generalization ability and produces stable results even when the number of rectangles or strip width differs from that during training.

Confronted with the fast social development, the speed of garbage generation has increased. Therefore, from every aspect, especially staying green and healthy, it is urgent and vital to ensure accurate and stable garbage classification. In the field of garbage classification, household garbage has a large volume and the classification process still requires manual intervention. Besides, most of the current studies focus on industrial waste and medical waste, while the existing studies on household garbage are not effective enough. Furthermore, these studies are mostly trained and validated in self-built datasets, making the experimental results insufficiently representative and severely limiting the development of research. Aiming to facilitate the study of household garbage classification, this paper proposes a simple yet effective network model (FConvNet) to improve classification performance. In this paper, an effective convolution block with fused convolution is designed to enhance the potential correlation of the features in the spatial dimension, and then a lightweight symmetric structure is constructed to infer more information about the feature representation. The experiments on public datasets have justified the effectiveness of FConvNet with accuracy rates of 95%, 95% and 97%, achieving better performance on public datasets compared to baselines. Additionally, an online system based on the WeChat mini program framework is designed to provide convenient household garbage classification services and enable the collection of uploaded images to perform model optimizations.

This work describes voltage-mode (VM), current-mode (CM) PID controllers and an electrically controllable floating inductance (FI) simulator circuit. Only a single current-controlled current backward transconductance amplifier (CC-CBTA) and a grounded capacitor are used in the proposed FI simulator circuit. One CBTA, one CC-CBTA, two capacitors and one resistor are used in the proposed PID controllers. All of the proposed circuits have appropriate input and output impedance values, allowing them to be utilized in cascade connections without requiring a buffer circuit. There is also no matching constraint between the active and passive values of the proposed circuit. The SPICE software environment and experimental results were used to validate theoretical derivations and associated outcomes. The layout of the CC-CBTA has been built, and the post-layout simulations are carried out using the netlist extracted from the layout. Comparisons were made with similar circuits found in the professional literature.

Since the leakage currents are dramatically increasing while the MOS transistors scale down to deep-submicron processes, the $I$–$V$ characteristic of the channel is varied unintentionally. As a result, the analog integrated circuit (IC) designs based on it (at above 100-nm technologies) will malfunction. In this paper, a novel low-leakage double-body MOSFET (LLDB-M), as a circuit-level scheme in answering to the need for the current-mode analog IC design in deep-submicron processes, is proposed. In this technique, the sub-threshold and gate-oxide tunneling leakage currents (as two major parts) are reduced via lowering the gate-source voltage and increasing the threshold voltage of the MOS transistor. An LLDB-M consists of two typical transistors — the first one is the main transistor and the latter implements the shift-voltage to reduce the gate-source voltage and floor of the channel leakage currents. The drain, gate, and body of the main transistor as well as the body and source of the second one, organize the terminals of an LLDB-M. The proposed LLDB-M transistors are replaced with large-leakage transistors in a translinear loop in the weak inversion region and then the commonly-used circuits such-as the current mirror, one-quadrant multiplier-divider (at both up-down and stack topologies), the true RMS-DC converter, and the log-domain low-pass-filter are designed. The effectiveness and performance of the proposed LLDB-M technique and circuits are evaluated using HSPICE software in 22-nm BSIM4 CMOS process and Cadence Virtuoso tool in 65-nm TSMC CMOS technology. Post-layout simulation results show that the proposed circuit-level method by considerably reducing leakage currents could ensure the effective function of the current-mode analog IC designs in deep-submicron technologies. Also, comprehensive simulations have been performed to prove the robustness and reliability of the proposed circuits against process, voltage and temperature (PVT) changes.

This paper presents a novel jitter reduction technique for a voltage-controlled ring oscillator (VCRO). This technique is helpful in employing VCRO-based circuits like Analog to Digital Converter (ADC), Phase Locked Loop (PLL) and time-based circuits whose performance is severely degraded by high jitter. In the proposed technique, the jitter is extracted using Excess Phase Discrimination (EPD) block and is transformed into Pulse Width Modulated (PWM) signal, which is then fed into the Phase Compensator (PC) block. This block reduces the jitter by about half its value after enabling the Single Ended Injection Locking (SEIL). The proposed work is designed in SCL 180 nm CMOS technology at 1.8 V supply. On enabling the proposed circuit, the Root Mean Square (RMS) jitter reduces from 5.94 to 2.58 ps (about 56%) for VCRO running at 58 MHz and consumes a total power of 375.84 $\mu$W. The overall performance of the proposed circuit does not vary much in the post-layout simulations, hence demonstrating the effectiveness of the proposed technique.

Several hardware Trojan (HT) detection techniques are available today to ensure the security of hardware systems. However, the existing pre-silicon HT detection techniques have problems such as difficulties in capturing HT path features and poor applicability. To address these challenges, this paper proposes a gate-level HT detection scheme based on a deep learning model. We parse the circuit gate-level netlist and develop an algorithm to extract circuit path sentences based on the signal propagation rule. Path sentences consisting of gate names are extracted as experimental datasets. We apply the theory of natural language processing (NLP) to the task of HT detection and use three neural networks to filter the length of path sentences. Then, based on the deep learning model text convolutional neural network (TextCNN), we propose PS-TextCNN for HT detection. Our approach is verified on seven benchmark circuits of the RS232-series and eight benchmark circuits of the s-series. We achieve an average true positive rate (TPR) of 88.9%. The TPR of the RS232-series reaches a high score of 99.5%. The TPR of the s-series is 79.5%, which is significantly higher than that of the existing gate-level HT detection techniques.

The computational intelligence-based digital forecasting for social systems has been a novel tendency. This work takes forecasting of regional economy as the problem scenario, and introduces radial basis function neural network (RBFNN) to deal with this concern. Hence, an RBFNN-based fast forecasting model for regional economy is constructed in this paper. First, the economic flow data are encoded into sequential format, and RBFNN is employed to establish a sequential forecasting model that fits flow data. In addition, a fuzzy clustering method is further utilized to complete missing data in flow sequence, in order to improve the data quality for model calculation. In simulation procedure, the proposal is implemented on real economic operation data for assessment, and mean absolute error (MAE) is introduced to measure the forecasting performance. Several typical regression-based forecasting methods are introduced as the baseline. The experimental results show that MAE of the proposal is about 20%–30% lower than that of baseline methods, showing a better forecasting performance.

Traditionally, 3D segmentation tasks have operated in silos, focusing separately on semantic and instance segmentation. However, this disjointed approach lacks interoperability and fails to fully unleash the potential of a more integrated, multitask solution. To overcome this limitation, we introduce TUS-Net, an innovative transformer-based architecture meticulously crafted for both semantic and instance segmentation of point clouds. Our model introduces two pivotal advancements: First, it employs a superpoint-based pre-processing step that minimizes computational overhead without compromising on precision. Second, we leverage a dual-branch design within the transformer architecture, allowing it to adapt to the nuances of both segmentation tasks dynamically. Through extensive experimentation on the ScanNet dataset, our findings demonstrate that TUS-Net surpasses prevailing specialized models by a substantial margin and maintains remarkable computational efficiency. Notably, we achieve a 5.7% enhancement in mean Average Precision (mAP), for instance, segmentation, while striking an optimal balance between accuracy and runtime for semantic segmentation. These outcomes underscore the versatility, efficiency and high-performance attributes of TUS-Net, positioning it as an indispensable framework for robust 3D point cloud segmentation.

A charge and discharge management scheme is proposed. The stored electric energy in distributed storage devices will converge to the consistent. The consistency of the stored electric energy of the devices helps to maintain the maximum load capacity and maximum consumption capacity of distributed storage devices. The charging and discharging process is constructed as a time-varying optimization problem, and the proposed algorithm can respond to the time-varying parameters of the distributed storage devices in real time. The time-varying neurodynamic algorithms can obtain time-varying optimal solution trajectories to give the optimal charging and discharging strategy in real time. In addition, the proposed approach in this paper focuses on the privacy protection of device data. Each device can calculate the power of discharging or charging by communicating with the partially connected nodes. Numerical simulations of the proposed scheme in the paper are given to verify the effectiveness of the scheme. Numerical simulations show that our scheme can make the electric energy stored in each storage device converge and maintain the maximum load capacity or maximum consumption capacity of the whole distributed storage device.

Artificial Intelligence (AI) techniques leverage the justice system in terms of effectiveness and efficiency. AI-empowered multi-label crime classification can facilitate the precise and expedient categorization of various legal documents. Multi-label classification within the justice system has an indispensable role in achieving accurate legal categorization that invigorates case analysis, optimizes resource distribution and refines the contours of legal processes. To support this critical function, this paper proposes a temporal convolutional network (TCN)-bidirectional encoder representations from transformers (BERT)-coupled model for multi-label crime classification. The proposed method fuses the temporal formation and semantic information in the model to obtain a high-quality result. The experimental results show that the proposed method achieved the best accuracy in comparison to existing methods on a public dataset.

In this paper, a novel image reconstruction method for printmaking is proposed via the combination of nonlinear diffusion filtering and convolution neural network. Nonlinear diffusion filtering based on a partial differential equation abstracts the input image to extract color features and texture features of images by nonlinear structured tensors. The image color features and texture features are combined to obtain a feature vector, and the feature vector is fed into the model as the input data. Then, the pre-trained deep learning model VGG-19 is utilized as the backbone network for further feature representation, and the extracted feature maps of each layer of the VGG19 model can be visualized for image reconstruction. The quality of the reconstructed images of prints is improved with the powerful feature extraction and combination ability of neural networks, which has the advantages of fast reconstruction speed and high quality of reconstructed images. Finally, the simulation experiments can demonstrate that the proposed reconstruction approach can achieve relatively ideal reconstruction performance in scenes of printmaking.

March algorithms are widely used in Memory Built-In Self-Test (MBIST) on-chip memory testing, providing linear test complexities that reduce the test time and cost. However, studies show that March algorithms with complexities lower than 18N have poor coverages of faults that have emerged with the advent of the nanometer process technologies and are more relevant to nowadays memories. New March AZ1 and March AZ2 algorithms, with 13N and 14N complexities, respectively, were introduced to provide optimum coverage of those faults and to produce a shorter test than an 18N-complexity test algorithm with a lesser area overhead, thus reducing chip manufacturing costs. This paper presents the implementation and validation of MBIST controllers that applied the March AZ1 and March AZ2 algorithms in a Field-Programmable Gate Array (FPGA) device. They were implemented in the Intel Max 10 DE10-Lite FPGA Development Board. A test generator was built in FPGA, as an alternative to the external tester, to provide test vectors required in initiating the test on the memory model using the implemented MBIST. The FPGA experimental tests demonstrated that they function correctly as the expected test sequences were observed. In addition, their fault detection abilities were also validated through tests on a fault-injected memory model, which shows that the implemented March AZ1 and March AZ2 provide 80.6% and 83.3% coverage of the intended faults, respectively, which outperform any other existing 14N-complexity March algorithms.

The price of carbon emission rights in the market fluctuates greatly due to various factors from economy, finance, and climate. For an enterprise that needs to conduct carbon trading activities, it is extremely necessary to fully grasp the price of carbon trading in the future period. Thus in this paper, a deep neural network-based assistive decision model for financial risk prediction in carbon trading market is proposed for this purpose. Specifically, the dynamic risk spillover effects of domestic and international carbon trading markets are studied, and a frontier time-varying model is utilized to measure the risk spillover effects. Then, the deep neural network is used for quantitative research and to construct an intelligent decision scheme that outputs financial risk prediction results. Four perspectives, energy price, climate environment, carbon market price, and macroeconomics, are selected as the input to analyze the influence factors of carbon emission rights price. Finally, several linear regression models are adopted as the baseline methods for comparison, and experimental results show that the proposed method can achieve better prediction performance compared with baseline methods.

In this paper, we construct a deep framework for full-reference image quality assessment (FR-IQA) by combining convolution and self-attention features; this approach effectively uses multiscale features to mimic the image evaluation in human eyes. We achieve the integration of information from local to global. First, convolutional neural network (CNN) and encoder of Swin transformer are used to extract multiscale paired features. Second, for each type of features, we fuse them after converting them to a fixed number of channels. For the fused features, we use the square of the difference between two pixel values at the corresponding channel positions to represent the features of the distortion degree. Then, we introduce a recurrent neural network (RNN) to capture the global features. Finally, for the two types of features, we use the full connection (FC) layer to regress two scores and then add the weights to compute the ultimate perceived score. By training and testing on publicly available FR-IQA datasets, experimental results further validate the superiority of our approach.