This paper proposes a new logic element model based on an Izhikevich (IZ) neuron and neural system that emulates two- and three-state logic behaviours. In a noise-free environment, with a periodic current of suitable amplitude and frequency, the IZ system is capable of performing logical AND and OR operations. Initially, a single IZ neuron demonstrates membrane dynamics in response to an input signal generated by combining two-state logic currents below the threshold. Subsequently, an IZ neural system model is introduced to enhance the reliability and resilience of the system. This model is characterised by electrical coupling with fast conduction and chemical coupling with a more adaptable structure. Each logic input independently influences each neuron within the system. Additionally, it has been observed that the reliability of the logic element is influenced by changes in synaptic strength, with a neural system lacking sufficient synaptic strength failing to generate logical output. Furthermore, the system displays a three-state logic behaviour under suitable forcing periodicity, thus enhancing the power efficiency of the logic element. The proposed IZ neuron and neural system are expected to significantly impact the development of brain-inspired logic elements.

A new safety assessment framework has been proposed to address the operational risks of the integration of wind power and photovoltaic grid, which integrates the characteristics of distributed power sources with the dynamic reconfiguration requirements of the distribution grid. The framework comprehensively considers the impacts of wind power and photovoltaic output uncertainties, as well as load fluctuations, on the stability of the distribution grid. It also evaluates the safety under different operational states of the distribution grid. Using Halton sequence sampling technology to accurately simulate the output of distributed power sources and the status of system components, combined with CPLEX optimisation for solving, a dynamic reconfiguration model is constructed to address potential faults in the distribution grid. Introducing the combined weighting method, a comprehensive risk assessment system for voltage violations, power flow violations, and load shedding has been constructed. The effectiveness of this method has been validated through simulations on the IEEE33 bus and IEEE118 bus systems, providing new insights to improve the safety and reliability of distribution grids.

The study addresses the challenge of bird collisions with wind turbines by developing an autonomous risk assessment method. The research uses data from the stereoscopic Bird Protection System (BPS) to anticipate potential collision threats by analysing flight parameters and distance from turbines. The danger factor depends on the flight characteristics of the identified bird species and the parameters of the wind turbine control system. The paper proposes an online quantitative risk assessment model that operates in real time, with the aim of minimising unnecessary turbine shutdowns while improving bird conservation. The model is validated through field data from bird flights. The findings suggest that adaptive management of turbine operations based on real-time bird flight data can significantly reduce collision risks without compromising energy production efficiency. The research underscores the balance between ecological considerations and the economic viability of wind energy, proposing an adaptive strategy that reduces unnecessary turbine stoppages while ensuring the safety of avian species.

This paper considers the robustness of the differential pulse-code modulation system with higher-order predictors. Special attention is paid to the robust parametric stability of the prediction filters with respect to the predictor coefficients. A generalisation of robustness in the classical sense is performed, and appropriate relations for calculating the probability of robustness are derived using Kharitonov principle. The proposed robustness estimation method is used for the third- and fourth-order prediction filters on speech signals, where the application of traditional methods is too difficult. For this reason, the Monte Carlo method is used to solve complex probability integrals. Verification and error analysis are performed for the previously considered second-order predictor. Satisfactory predetermined accuracy is achieved by increasing the number of samples. The results obtained could be very useful to design a system with suitable values for the predictor coefficients.

Parkinson’s disease is accepted as one of the most important diseases in the world. Parkinson’s disease can be diagnosed in various conventional techniques. Recently, these techniques have been replaced by artificial intelligence systems. This study proposes a feature selection and classification technique for Parkinson’s disease based on speech signals using a meta-heuristic algorithm. The proposed method selects the features from the data set including speech signal data that most accurately represent the problem using the efficient search strategies of the immune plasma algorithm (IPA). The experimental results are promising compared to other competing methods for diagnosing Parkinson’s disease in the literature.

In robotic harvesting, maneuvering around obstacles to position manipulators is challenging, especially in unstructured environments. This study proposes a method to detect the relative position of tomato bunches to the main stem position using the BlendMask-BiFPN algorithm. Initial comparative tests between full-stem and partial-stem labelling strategies revealed that the latter produced more complete peduncle masks, which guided our choice for subsequent experiments. Significant modifications to the BlendMask algorithm included the integration of a ResNet-101-BiFPN backbone, which improved the feature fusion network of the model. The revised model demonstrated high efficiency in pinpointing the relative positions of clustered tomatoes, achieving 91.3 % ARmask 50 and 84.8 % APmask 50 for the detection of tomato bunches. Comparisons with Mask RCNN, YOLACT, YOLACT++, and YOLOv8 showed that the BlendMask-BiFPN model outperforms these alternatives, suggesting its potential for more effective robotic harvesting in complex agricultural scenarios.

This paper presents a novel method for detecting and classifying faults in dynamic control systems empowered with DC motors, operating under laboratory conditions. The approach employs a convolutional neural network model enhanced with an artificial endocrine influence to evaluate the condition of the rotating motor shaft by analysing information from the vibration sensors mounted on the shaft itself. The trained network effectively classifies the level of unbalance in the system into three categories based on the vibrations: optimal (no unbalance), first and second degree of unbalance. To validate the efficiency of the proposed model, its performance was compared with the performance of deep learning algorithms commonly recommended for time-series classification: default convolutional neural network, fully convolutional neural network, and residual network. The new model was shown to perform classification tasks with the highest accuracy, proving to be an efficient fault diagnosis tool with a viable potential to be applicable in industrial predictive maintenance processes.

This paper introduces a decoupled unknown input observer (DUIO) for Takagi-Sugeno (T-S) systems, designed specifically for the synchronous reluctance motor (SynRM). The proposed DUIO method demonstrates enhanced robustness and accuracy in state estimation by effectively decoupling the influence of unknown inputs from the estimation error dynamics. Furthermore, the DUIO exhibits superior performance compared to the proportional integral observer (PIO) and the proportional multi-integral observer (PMIO) presented in previous studies, without the need for prior knowledge of the unknown input form or assumptions regarding its boundedness. Stability conditions, achieved using the quadratic Lyapunov function, are expressed as linear matrix inequalities (LMIs), which ensure asymptotic convergence of the estimation error. The effectiveness of the DUIO method is further validated in various scenarios through hardware-in-the-loop (HIL) implementation. This innovative approach significantly enhances the accuracy and reliability of SynRM state estimations and unknown input detections.

There are many instances of intellectual property rights violations due to the common usage of digital data on the Internet, including unauthorised use, copying, and theft of digital content. Intellectual property rights of digital photos must be upheld, as they are very valuable materials. Digital watermarking is a more modern method to do this. By using a watermark (WM), the owner's information is included into the content, which may then be shared or saved. When required, this technology will retrieve the encoded WM information to prove ownership. Different technologies have been investigated and created on the basis of existing technologies, fields of use, etc. This paper proposes a novel approach to digital watermarking based on a neural network. First, the trigger data set and noise data set are generated from the binary encoding and random cutting of the original training samples. Then, the pattern with higher watermark trigger accuracy is obtained from the trigger set. Simulation results show that the proposed algorithm performs better in terms of accuracy and computing time cost compared to existing algorithms

This paper compares different maximum power point tracking (MPPT) control strategies in microgrid dynamics, focussing on perturb and observe (P&O), adaptive neuro-fuzzy inference system (ANFIS), particle swarm optimisation (PSO), and PSO sliding mode controller techniques. The study investigates their performance under varying microgrid conditions, considering factors like weather and load variations. The simulation results provide a detailed comparative analysis of the power at the point of common coupling (PCC) for MPPT techniques at different time intervals. Both the P&O and PSO sliding mode recorded a power output of 287 kW, while PSO-ANFIS achieved a slightly higher power output of 294 kW. At 2.5 seconds, the P&O method recorded a power output of 712 kW, while the PSO sliding mode and the PSO-ANFIS techniques achieved 717 kW and 738 kW, respectively. Overall, the PSO-ANFIS technique consistently outperformed the other methods in terms of power output, demonstrating its effectiveness in maximising energy extraction and adaptability to dynamic conditions. These findings provide valuable insights for designing and implementing MPPT controllers in microgrid systems, emphasising the efficiency of the hybrid PSO-ANFIS technique in enhancing the overall performance and stability of renewable energy systems.