Early leak detection in water distribution networks is essential to minimize losses and improve operational efficiency. This systematic review analyzes 53 studies published between 2018 and 2025 that employed machine learning, deep learning, and hybrid approaches. The results show that pressure is the most widely used and most sensitive input variable for identifying hydraulic anomalies. Most datasets originate from EPANET-generated simulations, while experimental and field data are less common due to their high costs and operational complexity. Machine learning models, particularly SVMs, achieve accuracies between 94 and 100%, demonstrating stability with noisy data and low computational cost, while in deep learning, CNNs are most effective for multiclass classification and localization, typically reaching 95–99% accuracy. Hybrid approaches that combine automatic feature extraction (e.g., CNNs or autoencoders) with conventional classifiers (such as SVMs or LSSVMs) yield the best results, surpassing 97% accuracy and achieving localization errors below 0.2 m. Based on these findings, a theoretical model is proposed using a hybrid CNN + SVM approach to enhance accuracy, robustness, and adaptability in real-time monitoring systems.

This review presents a synthesis of fuzzy logic-based (FL) controllers applied to photovoltaic (PV) systems over the last decade, with a specific focus on maximum power point tracking (MPPT) and power management. These subsystems are critical for improving the efficiency of PV energy conversion, as they directly address the nonlinear, time-varying, and uncertain behavior of solar generation under dynamic environmental conditions. FL-based control has proven to be a powerful and versatile tool for enhancing MPPT accuracy, inverter performance, and hybrid energy management strategies. The analysis concentrates on three main categories, namely, Mamdani, Takagi–Sugeno (T-S), and Type-2, highlighting their architectures, operational characteristics, and application domains. Mamdani controllers remain the most widely adopted due to their simplicity, interpretability, and effectiveness in scenarios with moderate response time requirements. T-S controllers excel in real-time high-frequency operations by eliminating the defuzzification stage and approximating system nonlinearities through local linear models, achieving rapid convergence to the maximum power point (MPP) and improved power quality in grid-connected PV systems. Type-2 fuzzy controllers represent the most advanced evolution, incorporating footprints of uncertainty (FOU) to handle high variability, sensor noise, and environmental disturbances, thereby strengthening MPPT accuracy under challenging conditions. This review also examines the integration of metaheuristic algorithms for automated tuning of membership functions and hybrid architectures that combine fuzzy control with artificial intelligence (AI) techniques. A bibliometric perspective reveals a growing research interest in T-S and Type-2 approaches. Quantitatively, Mamdani controllers account for 54.20% of publications, T-S controllers for 26.72%, and Type-2 fuzzy controllers for 19.08%, reflecting the balance between interpretability, computational performance, and robustness to uncertainty in PV-based MPPT and power management applications.

This study compares bio-inspired optimization algorithms for enhancing an ANN-based Maximum Power Point Tracking (MPPT) forecast system under partial shading conditions in photovoltaic systems. Four algorithms—grey wolf optimizer (GWO), particle swarm optimization (PSO), squirrel search algorithm (SSA), and cuckoo search (CS)—were evaluated, with the dataset augmented by perturbations to simulate shading. The standard ANN performed poorly, with 64 neurons in Layer 1 and 32 in Layer 2 (MSE of 159.9437, MAE of 8.0781). Among the optimized approaches, GWO, with 66 neurons in Layer 1 and 100 in Layer 2, achieved the best prediction accuracy (MSE of 11.9487, MAE of 2.4552) and was computationally efficient (execution time of 1198.99 s). PSO, using 98 neurons in Layer 1 and 100 in Layer 2, minimized MAE (2.1679) but had a slightly longer execution time (1417.80 s). SSA, with the same neuron count as GWO, also performed well (MSE 12.1500, MAE 2.7003) and was the fastest (987.45 s). CS, with 84 neurons in Layer 1 and 74 in Layer 2, was less reliable (MSE 33.7767, MAE 3.8547) and slower (1904.01 s). GWO proved to be the best overall, balancing accuracy and speed. Future real-world applications of this methodology include improving energy efficiency in solar farms under variable weather conditions and optimizing the performance of residential solar panels to reduce energy costs. Further optimization developments could address more complex and larger-scale datasets in real-time, such as integrating renewable energy sources into smart grid systems for better energy distribution.

Vibration analysis has proven to be effective for detecting failures in rotating elements. The objective of this study is to identify different failure conditions in a gear within an automotive gearbox by inducing failures in the gear and subsequently processing the vibration signals in the frequency domain using the Fast Fourier Transform. Two accelerometers were used to acquire data from the automotive transmission under different operating frequencies, to identify patterns indicative of failure in the damaged gear. The results show that the method can help identify the failure in the gear according to the amplitude value, differentiating between failure due to fracture and eccentricity. The increase in vibration levels is directly related to the severity of the failure present in the gear.

Data-driven models for embedded control must account for acquisition filtering and experimental uncertainty while retaining dependence among identified parameters. This work develops a framework connecting identification in filtered-state coordinates, joint fuzzy uncertainty propagation, and fixed-gain stability assessment. A structured affine discrete-time model is refined through finite-horizon prediction error, and complete bootstrap deviations define a joint radial parameter family. Set inclusion ensures nested prediction envelopes, while endpoint quadratic inequalities certify stability over continuous parameter segments. Evaluation uses 30 identification runs and a main campaign of 60 physical closed-loop trials on a DC motor (30 per controller), complemented by two further campaigns of ten trials per controller: one with an inserted series resistor and one with a sinusoidal reference. Separate calibration and testing quantify predictive coverage and interval width, while marginal-preserving permutation demonstrates the influence of parameter pairing on prediction bands and sampled stability. All 435 retained training families receive continuous stability certificates. Quantitative comparisons with N4SID, SUBNET, and a TSK fuzzy predictor identify complementary predictive strengths, and nominal versus guaranteed-cost control evaluations characterize the trade-off between tracking accuracy and voltage effort. The implemented nominal integral LQR achieves a tracking RMSE of $0.7969\mathrm{rad}$. The contribution links empirical joint uncertainty to nested propagation and verifiable stability properties within the retained model families, supported by physical control experiments. The certificates apply to the modeled linear dynamics; experimental validation remains limited to the tested platform and operating conditions.

Objective: This observational cross-sectional study aimed to comprehensively evaluate sensory–cognitive performance in children diagnosed with Attention-Deficit/Hyperactivity Disorder (ADHD), with a focus on auditory processing, visual–perceptual abilities, visual–motor integration, and oculomotor function. The study further examined how hyperactivity, age, and gender may influence these domains. Methods: A total of 70 non-medicated children with clinically diagnosed ADHD (mean age = $9.1\pm 2.4$ years; 67.1% male), all with normal visual acuity, were assessed using four standardized instruments: the Test of Auditory Processing Skills, Third Edition (TAPS-3), the Test of Visual Perceptual Skills, Fourth Edition (TVPS-4), the Beery–Buktenica Developmental Test of Visual–Motor Integration, Sixth Edition (VMI-6), and the Developmental Eye Movement (DEM) Test. Statistical analyses included one sample and independent samples t-tests, one-way ANOVA, and Pearson correlation coefficients. Results: Participants demonstrated significantly above-average performance in auditory processing (TAPS-3: $\mu =108.4$, $std=7.8$), average visual–perceptual abilities (TVPS-4: $\mu =100.9$, $std=7.2$), slightly below-average visual–motor integration (VMI-6: $\mu =97.1$, $std=9.0$), and marked deficits in oculomotor efficiency (DEM ratio: $\mu =87.3$, $std=18.1$). Statistically significant differences were observed across these domains (t-values ranging from 2.9 to 7.2, $p<0.01$). Children with hyperactive-impulsive presentations exhibited lower horizontal DEM scores ($\mu =73.4$, $std=12.3$) compared to inattentive counterparts ($\mu =82.9$, $std=16.2$; $p=0.009$). Age and sex influenced specific subtest scores, with boys and children aged 8–9 years achieving higher outcomes in word memory ($p=0.042$) and visual discrimination ($p=0.034$), respectively. Moderate correlations were identified between auditory and visual–perceptual skills ($r=0.32$, $p=0.007$), and between visual–perceptual and oculomotor performance ($r=0.25$, $p=0.035$). Conclusions: The findings from this sample reveal a distinct sensory–cognitive profile in children with ADHD, characterized by relatively preserved auditory processing and pronounced oculomotor deficits. These results underscore the value of a multimodal assessment protocol that includes oculomotor and visual efficiency evaluations. The conclusions pertain specifically to the cohort studied and should not be generalized to all populations with ADHD without further validation.