For the wireless charging systems for electric vehicles (EVs), the misalignment phenomenon due to inaccurate parking is the most significant issue, which causes non-negligible negative impacts on the power efficiency and amount. That is because the positional misalignment between coils leads to significant changes in parameters such as mutual inductance, which in turn causes dramatic fluctuations in the system’s output voltage and efficiency. It potentially prevents the system from functioning correctly or even damages it. Therefore, research on the anti-misalignment capability of EV wireless charging systems is crucial. Current research focuses on high-frequency inverter control, coupling mechanism design, and compensation topology design. However, these methods fail to maintain a constant output voltage when both coil misalignment and large variations in load occur. This paper proposes a novel hybrid compensation topology based on the QRQP coil.

This paper uses finite element simulation software to investigate a QRQP coil and its misalignment and coupling characteristics. To reduce output voltage fluctuations caused by coil misalignment and large variations in load, a novel hybrid topology is introduced based on the QRQP coil. This topology leverages the principle of opposing output characteristics between S-LCC, LCC-S, LCC-LCC, and SS topologies. Detailed design guidelines for the parameters and optimization strategies are proposed. Meanwhile, the system’s anti- misalignment capability under different parameter selections is analyzed. Finally, optimal system parameters are selected and analyzed. The optimized system can maintain a constant output voltage under various misalignment angles and load variations within a specific range. When the receiving coil is removed, the primary-side current can be effectively limited, which ensures the system's safety.

The proposed topology has been validated through a 1 kW laboratory prototype. Experimental results show that when the load resistance varies from 20 Ω to 100 Ω, the system maintains output voltage fluctuations of less than 5% under X-axis misalignment from -140 mm to +140 mm, Y-axis misalignment from -105 mm to +105 mm, and diagonal misalignment along the XY-axis from -200 mm to +200 mm. When the load resistance varies from 20 Ω to 100 Ω and the coils’ vertical distance changes from -35 mm to 70 mm, the output voltage fluctuation can be kept within 8%. Furthermore, since the system exhibits capacitive behavior after misalignment and has no compensating inductance, it can operate with high efficiency. Analysis under extreme conditions shows that when the receiving coil is removed, the optimized hybrid topology effectively limits the primary-side current surge, preventing system damage.

The following conclusions can be drawn. (1) The proposed optimization theory for the novel hybrid topology is consistent with the experimental results. (2) The optimized hybrid compensation topology based on the QRQP coil can effectively reduce output voltage fluctuations when coil misalignment and large variations in load occur simultaneously. (3) When removing the receiving coil, the optimized hybrid topology effectively limits the primary-side current surge, preventing system damage.

Renewable energy systems have gained continuous attention for achieving “carbon peak” and “carbon neutrality”, especially DC conversion technologies for renewable energy conversions. Multi-port converter (MPC) has been widely applied in renewable energy systems and electric vehicles due to the characteristics of low cost, high efficiency, and high power density. The non-isolated MPC suffers poor stability due to insufficient electrical isolation between ports. In contrast, isolated converters are often more complex and less flexible. Wireless power transfer (WPT) technology offers convenience, safety, flexibility, and the ability to charge multiple devices, effectively achieving electrical isolation between input and load ports. Thus, combined with WPT and MPC technologies, this paper proposes a three-port DC-DC converter with integrated wireless power transfer capability. The proposed topology facilitates DC power transfer between multiple DC sources with the same or different voltage levels. It enables wireless power transfer between DC sources and load by introducing WPT coupling technologies. The system achieves non-contact hot plug & play between DC loads and the power grid side, which indirectly isolates the impact of the load on the power grid.

The system employs a hybrid power flow control method, with dual half-bridge micro-inverters providing the dual input ports. The load port is wirelessly coupled through an LCL-LCL-type resonant coupling network connected to a full-bridge rectifier. This three-port topology is simple and highly flexible, allowing free power transmission between dual input sources, with the two sources sharing one LCL resonant tank for power transmission to the load without any additional circuit components. System control strategies can be divided into two phases: Phase1: pulse width modulation (PWM) controls the power flow between two DC sources by controlling the average DC offset current in the LCL resonant tank, enabling bidirectional power transmission; Phase 2: phase shift modulation (PSM) control method adjusts the wireless output power for DC load. These two control loops can operate independently or be combined for comprehensive control. The absence of coupling between these methods enhances the stability and effectiveness of each control function. Additionally, the system allows for dual input ports with unbalanced voltage levels.

Firstly, a dual-sided LCL resonant coupling network model is established based on the AC impedance method to analyze its frequency limitations under constant voltage and constant current output characteristics. Secondly, the system topology’s various operating states are analyzed based on switching modes. The overall system model is developed using time-domain analysis, and a small-signal model of the resonant coupling network is established to determine the primary-side PWM control and secondary-side PSM control strategies. Thirdly, a simulation model is built in PSIM to verify the system’s functionality. Matlab/Simulink is used to optimize the parameters of the compensation network. Finally, an experimental platform is set up in a microgrid and energy storage interconnected system to evaluate the system's dynamic characteristics under different voltage levels and load conditions, efficiency variations, steady-state control performance of the closed-loop controller, and dynamic response characteristics.

Experimental results show that under dual inputs of DC 36 V with only wireless output, the system achieves a peak efficiency of 93.6% and load-independent constant current output performance. The system effectively controls the power flow direction and magnitude between the primary-side energy ports, and the designed controller maintains stable load power even under sudden changes in load resistance and voltage levels at the dual half-bridge energy ports. The controller also demonstrates good robustness and dynamic response performance.

This paper proposes a fault diagnosis method based on an improved dung beetle optimization algorithm (IDBO) to solve the problem of low accuracy of mechanical fault diagnosis of high-voltage circuit breakers. Tent chaotic mapping, a golden sine strategy, and adaptive t-distribution perturbation are incorporated to optimize a deep hybrid kernel extreme learning machine (DHKELM).

Firstly, this paper takes a TY-1S-12/630-16 single-phase vacuum high-voltage circuit breaker as the research object and builds a platform for collecting high-voltage circuit breaker closing vibration signals. Five operating conditions are simulated: normal state, cushion spring fatigue, base looseness, insulator looseness, and drive shaft jam. The laser vibrometer’s sampling time and frequency are set to 1 000 ms and 78 125 Hz. 60 groups of samples for each condition of the high-voltage circuit breaker are collected, totaling 300 sets of samples.

Secondly, the successive variational modal decomposition (SVMD) is used to decompose the acquired signals, the seven IMF components with different center frequencies are obtained after decomposition, and the power spectral entropy of each IMF component is extracted to construct the feature vector matrix. Data dimensionality reduction of the feature vectors is carried out using the t-distribution-stochastic neighborhood embedding algorithm (t-SNE) to obtain 300 by 3-dimensional feature vectors. After dimensionality reduction by t-SNE, the samples of the same state show clear clustering characteristics, while the samples of different states are separated in the mapping results of t-SNE. Hence, the problems of information redundancy and high- dimensional data are avoided.

Then, by introducing three optimization strategies-fusion Tent chaotic mapping, golden sine strategy, and adaptive t-distribution perturbation, the improved dung beetle optimization (IDBO) algorithm is proposed. The IDBO algorithm optimizes the parameters of the DHKELM for constructing the IDBO-DHKELM high-voltage fault diagnosis model. The unimodal and multimodal functions from the CEC2005 test suite are selected for performance testing. The improved IDBO algorithm is compared with traditional PSO, WOA, and DBO algorithms, verifying its superior convergence speed, optimization precision, and stability in finding the optimal solution.

Finally, a platform is built to simulate mechanical failures of high-voltage circuit breakers. The fault diagnosis results show that the proposed method’s fault diagnosis accuracy reaches 98.33%, and the average accuracy of the classification of the DHKELM model is improved by 11.67%, 5.83%, and 3.33%, respectively, compared with that of the traditional SVM, ELM, and CNN models. The DHKELM model improves the average classification accuracy by 9.16% and 7.5% compared with PSO-DHKELM and DBO-DHKELM models, and the precision rate, recall rate, and F_1-score are greatly improved.

The magnetic field's compactness can better increase energy's transmission distance in a near-filed wireless power transfer (WPT) system. This paper proposes a non-centrosymmetric excitation unit (NEU) design method for the WPT system with a matrix coupling mechanism to improve the magnetic flux density in the central region. The proposed method enhances the magnetic focusing performance with a flat two-dimensional structure while maintaining its misalignment tolerance without any other auxiliary coil or circuit. The self-inductance value of the transmitting coil in this design method is even smaller than that of the conventional design method under the same size and number of turns. In addition, the proposed method is applicable to all regular matrix coils. Then, a detailed design method of the coupler is given based on the circuit analysis. Finally, taking the LCL-LCC compensation WPT system as an example, the feasibility of the proposed design method is verified. Compared to traditional design methods, the proposed design method can increase the induced voltage by 24.7% at the optimal position.

The current mobile robot UAV (unmanned aerial vehicle)/AGV (automated guided vehicle) plays an important role in industrial inspection, office logistics, agricultural and forestry plant protection, and military reconnaissance and surveillance fields. A multi-level, multi-modal power supply demonstration application has been designed in intelligent parks and industrial inspections to achieve charging docking between unmanned devices. However, most power supply methods are fixed-point charging, which is complicated in the energy transmission process and faces energy loss problems. In special charging environments such as wilderness and underwater, if there is no suitable parking charging platform condition, UAV hovering charging technology has become the most necessary and perfect charging solution, and it is also the key to building wireless power transmission network environments. Based on the well-established technology of fixed-point hovering and hovering following for unmanned inspection equipment in the air-to-air, ground-to-ground, and air-to-ground scenarios, hovering and alignment energy replenishment is undoubtedly a highly efficient and cutting-edge design concept applied to unmanned inspection equipment. Nowadays, to further enhance the flexibility and reliability of wireless charging systems, more and more wireless charging systems are designed based on matrix coils. The matrix-based multi-excitation wireless charging system has gained more favor in charging applications due to its reconstruction mechanism of the magnetic field at the transmitting end and strong tolerance for voltage and current stress.

In summary, a matrix coil design method based on non-centrosymmetric excitation units is proposed, combined with an evaluation of the system's spatial field transmission capability to improve the optimization design steps and the focusing ability of the magnetic field. This design method, which has a two-dimensional planar structure, enhances the magnetic flux density at the center position without any auxiliary coils or circuits, avoiding the offset tolerance weakening of the matrix coil itself as much as possible. It reduces the self-inductance of the coil unit to improve mutual inductance utilization.

Magnetically coupled resonant wireless power transfer (MCR-WPT) technology has received significant attention due to its ability to realize mid-range power transfer. However, the transmission characteristics of MCR-WPT systems are susceptible to variations in coupling coefficients and loads. Parity-time (PT) symmetry has been introduced into the WPT system (PT-WPT) to achieve constant power and high-efficiency transmission over medium distances. This paper provides a comprehensive review of the PT-WPT technology.

First, the paper introduces the PT-WPT system’s basic structure and operating mechanism. It analyzes how the system balances energy gain and loss through the nonlinear saturated negative resistor, allowing it to maintain stable power transmission under varying coupling conditions. Coupled-mode and circuit models are used to construct the PT-WPT system. The two models’ similarities and differences in the energy transmission mechanism, PT symmetry conditions, and system characteristics are described. In addition, PT-WPT can be considered a novel wireless power transfer technology.

Next, the paper discusses the construction methods of nonlinear saturated negative resistors, which can be divided into two categories based on the components used: operational amplifiers and power converters. While operational amplifiers provide a simple and low-cost solution, they are limited in power output. In contrast, power converters, such as half-bridge, full-bridge, and class E inverters, enable higher power output and efficiency but require more complex control strategies. Then, the advantages and disadvantages of these methods are discussed, and directions for improving the design of negative resistors are given.

This paper introduces the different types of coupling mechanisms and the implementation of charging functions. Among the topologies of PT-WPT systems, single-transmitter-single-receiver is the most basic structure; high-order compensation networks and the introduction of relay coils are commonly used to extend the transmission distance of the PT-WPT system. Multi-transmitter/multi-receiver can also improve the system’s reliability and realize stable power transmission in multi-load systems. Furthermore, the charging control strategies are investigated to realize the constant power and constant current/voltage functions independent of the coupling coefficient and load variation, further promoting the practicalization of PT-WPT systems.

Finally, this paper summarizes the existing research on PT-WPT systems and future research issues. PT-WPT technology is expected to find broader applications in the future and promote the development of wireless charging technology.

In underwater applications, such as underwater robots, autonomous underwater vehicles (AUVs), and remotely operated vehicles (ROVs), magnetic-field coupled wireless power transfer (MC-WPT) enables the transmission of electrical energy without electric contact, improving the flexibility and security of power transfer. Underwater electrical devices and base stations must achieve long-distance and high-power wireless power transfer while realizing high-speed bidirectional wireless information exchange to enable command transmission, data feedback, and closed-loop control. Many scholars have researched shared-channel magnetic-field coupled underwater simultaneous wireless power and information transfer (MC-USWPIT) technology. However, there is still a gap between the transmission distance, power transfer capacity, information transfer speed, and the requirements of engineering applications. Therefore, this paper proposes an underwater simultaneous wireless power and information transfer system with a coplanar double-coil coupler. The research focuses on rapid wireless power replenishment and high-speed bidirectional information transmission for AUVs in seawater. The goal is to achieve high-power energy transfer and high-speed bidirectional information transmission over long transmission distances.

The coupler with a coplanar double-coil and the MC-USWPIT system topology are proposed. Using the relay coil for information transmission reduces the voltage stress on the information transmission circuit and helps mitigate the crosstalk between the power and information transfer channels. By employing an injecting information method with a series of LC circuits in the information transmission channel, the LC circuit is fully compensated at the power transmission frequency. Furthermore, smaller capacitance-blocking capacitors further reduce the crosstalk between the power transmission channel and the information transmission channel, as well as the voltage stress on the information transmission channel, thereby reducing the difficulty of system design.

Subsequently, the system is analyzed and modeled, and equivalent circuit models for the power and information transfer channels are provided. A parameter design method for the MC-USWPIT system is proposed. The method reduces the eddy current losses induced by the seawater and minimizes the impact of high-power energy transmission on the information transfer speed. It enables the simultaneous improvement of transmission distance, power transfer capacity, and information transfer speed in a frequency-division multiplexed MC- USWPIT system.

Finally, a 5 kW experimental setup in simulated seawater is constructed. In an environment with a seawater conductivity of 4.15 S/m, the system achieved a transmission distance of 50 cm, an output power of 5.33 kW, and an information transfer speed of 5.68 Mbit/s. Furthermore, under varying seawater conductivities (4, 5, and 6 S/m) and transmission distances (30, 40, and 50 cm), the system still demonstrates good power transfer performance and high information transfer speed. The experimental results confirm that the proposed MC-USWPIT system and method can effectively improve the transmission distance, power transfer capability, and bidirectional information transfer speed in simulated seawater.

As the demand for flexibility and efficiency in modern industrial equipment increases, motors often operate under variable speed conditions in real-world industrial applications. This poses challenges for traditional time-domain and frequency-domain fault diagnosis methods. These challenges arise primarily due to the non-linear and non-stationary characteristics of signals under variable speed conditions, which can affect fault feature extraction. Single deep learning models generally require training and test data to follow the same distribution, and domain adaptation or multi-source domain generalization methods are difficult to apply in the absence of target domain and multi-source domain data, limiting their ability to enhance the generalization of single-source domain models. To address these challenges, this study proposes a motor rolling bearing fault transfer diagnosis method that integrates angular domain resampling and feature enhancement.

First, to mitigate the issue of time-frequency characteristic offsets in vibration signals under different rotational speeds, angular domain resampling is employed. This technique processes vibration signals at varying speeds, obtaining angular domain vibration signals to minimize the offsets caused by speed changes. Second, to address the generalization limitations of deep learning models, fault data from constant speed conditions are used as the source domain for training the neural network. Covariance loss is introduced to amplify the feature differences among various classes in the source domain data. This allows the network to focus on more informative features for the classification task, thereby improving its generalization capability. Finally, the angular domain vibration signals under variable speed conditions are input into the trained model for fault classification.

The effectiveness of the proposed method is validated through several experiments. Initially, the time-frequency characteristics of vibration signals from an actual bearing inner ring fault are examined before and after angular domain resampling. Before resampling, the vibration signal intervals under variable speed conditions show significant variability. However, after resampling, the variability in the vibration intervals is significantly reduced. Furthermore, using t-SNE visualization, the study observes that networks without feature enhancement show slow gradient updates and minimal changes in feature distribution. In contrast, networks with feature enhancement exhibit continuous changes in feature distribution, even as the classification loss decreases, with increasing feature distances. The study also conducts four cross-working condition fault diagnosis experiments, comparing the proposed method with other methods. The results demonstrate that the proposed method improves fault identification accuracy by 35.04% compared to methods without angular domain resampling, especially in rolling element fault identification. When compared to methods without feature enhancement, the proposed method improves accuracy by 7.45%. Additionally, in transfer diagnosis tasks under different load conditions, the proposed method demonstrates high accuracy, recall, and F1 scores.

In conclusion, the study finds that: (1) Angular domain resampling effectively reduces time-frequency distribution differences caused by speed variations, proving its applicability and rationality in data preprocessing at different speeds. (2) The feature enhancement strategy, by increasing covariance loss between different class features, amplifies feature differences between various health status signals, enabling the network to capture more distinctive features and significantly improving generalization capability. (3) The proposed method, without requiring target domain data, achieves fault identification accuracy of up to 97.29% under variable speed conditions, demonstrating good robustness under variable load conditions.

With the increasing scale of urban power grids and the application of various advanced power equipment, the comprehensive efficiency of the power system depends on the cooperation and synergy between various power equipment. Aiming at the physical entities with dynamic and real-time changes in the state of power equipment, constructing an efficient mathematical model is essential in system-level simulation of power systems and integrated design of large electromagnetic equipment. Taking the magnetically-saturated controllable reactor (MSCR) as the object, this paper proposes a nonlinear dynamic electromagnetic network model, considering the accuracy of theoretical analysis and the efficiency of parameter calculation.

Firstly, according to the structural characteristics and magnetic field distribution characteristics of MSCR, the MSCR solution domain is meshed by domain discretization. Considering the nonlinearity of the iron core, the magnetization curve model of the MSCR iron core is established by the piecewise interpolation method. The nonlinear grid parameters are calculated according to the principle of the flux tube. The MSCR equivalent magnetic network model is generated based on the loop current method.

Secondly, the electromagnetic model of circuit-magnetic circuit separation is established. The electromagnetic coupling equivalent circuit is established using the controlled source to realize the coupling connection between the circuit and the magnetic circuit. The nonlinear dynamic electromagnetic network model of MSCR is generated. Combined with the nonlinear iterative solution of the chord-cut method, the MSCR winding current and the core flux under different magnetic saturations are calculated.

Finally, a three-dimensional finite element model of MSCR is established based on the finite element method, and field-circuit coupling joint simulation is carried out. Experimental measurements are also performed on the MSCR winding current. The MSCR nonlinear dynamic electromagnetic network model is compared with the three-dimensional finite element model and experimental measurements. The MSCR nonlinear dynamic electromagnetic network model is verified.

(1) The proposed model's calculated winding current and core flux agree with the finite element model and experimental results under different magnetic saturations. (2) The calculation speed and the storage space of the proposed model in electromagnetic parameter calculation are approximately 50~240 times and 1/10 000~1/7 000 of the finite element model. The model can improve calculation efficiency and reduce cost while meeting computational accuracy. It has unique advantages in the initial design of controllable reactors and system-level simulation of power systems.

The rapid development of power electronics technology and wireless power transmission (WPT) has broadened application prospects in consumer electronics and traditional fields like electric vehicles, implantable medical devices, and autonomous underwater vehicles. The unique nature of wireless charging scenarios often results in significant variations in transmission distance, causing rapid drops in coupling coefficient and transmission efficiency. Therefore, coil compensation is an important research area of WPT. This paper proposes a segmented coil compensation method using inter-turn capacitance to solve increased internal voltage gradients in coils caused by traditional external capacitor compensation methods. By closely fitting the adjacent turns of the receiving coil, the capacitance of the closely fitting section is increased, thereby achieving coil compensation. The coil is divided into several segments, with inter-turn capacitance to compensate for each segment.

First, the advantages of segmented coil compensation are derived using rigorous circuit theory. Next, equivalent modeling and calculation of the Litz wire wound coil are performed to analyze the factors influencing the size of the inter-turn capacitance. Finally, the overall coil with inter-turn capacitance segmented compensation is modeled, and the port impedance of the entire coil is calculated. The inter-turn capacitance of the closely fitting coil segments increases significantly, effectively replacing the external capacitors for compensation. Compared to external capacitor compensation topologies, the proposed segmented compensation topology can reduce the internal voltage gradient of the coil, voltage loss, and energy dissipation. Accordingly, the energy reception efficiency of the coil is improved. Additionally, this structure is compact, with a small size and cost.

An experimental coil is compared with a coil using external capacitor compensation. Under the same input and load conditions, the receiving power of the coil with the proposed segmented compensation is increased by 54.3%, and efficiency is improved by 27.6%, which verifies the proposed method.

The following conclusions can be drawn. (1) The tight alignment length of the turns and the dielectric constant of the wire insulation layer influence the inter-turn capacitance. When Litz wire is used for equivalent analysis, the inter-turn capacitance increases significantly with the tight alignment length and shows a linear growth trend. The capacitance also increases significantly as the equivalent dielectric constant of the insulation layer increases, effectively compensating the coil. (2) Inter-turn capacitance compensation provides adequate compensation. Compared to traditional concentrated compensation methods, the proposed segmented compensation reduces the internal voltage gradient of the coil, decreases voltage losses and energy dissipation, and enhances the energy reception efficiency of the coil.

The low temperature and high humidity environment in winter can easily cause wind turbine blades to freeze, seriously affecting the actual power output and safe operation of wind turbines. To avoid problems such as increased fatigue load and vibration of unit components caused by icing, wind farms need to implement shutdown strategies in a timely manner based on the icing situation of the blades. Therefore, accurate identification of blade icing status has become one of the key points in maintaining the safe operation of winter wind turbines. However, current ice diagnosis methods rely on a large amount of time series data for modeling and prediction. In practical work, due to equipment and working conditions, it is difficult to collect sufficient ice sample monitoring data, which leads to the widespread problem of data imbalance and has a continuous impact on the improvement of ice diagnosis accuracy. To solve this problem, this paper proposes a fusion diagnostic model based on conditional generative adversarial network (CTGAN) and light gradient boosting machine (LightGBM), aiming to achieve high-performance wind turbine blade ice diagnosis using a small number of training samples.

Firstly, based on the sliding window algorithm, new mixed features are further constructed on the basis of the original features. Secondly, the CTGAN model is used to learn the data distribution of real samples, and Nash equilibrium is achieved through adversarial training with generators and discriminators, generating new samples that are similar to real samples. Then, the synthesized samples are input into LightGBM to extract effective features and diagnose icing, and the LightGBM model is modified by introducing a focus loss function to improve its ability to distinguish confusing samples. Finally, the attribution theory based on shapley additive explanetions (SHAP) was used to analyze the factors affecting icing.

The simulation results on actual wind farm data show that the diagnostic accuracy of all algorithms has a certain improvement effect after using mixed features, and the average diagnostic accuracy of each model can reach 0.979. Due to the introduction of sample expansion algorithms, the accuracy of each model has improved to varying degrees compared to when data is lacking. When the sample imbalance rate is 30%, the accuracy of the traditional Logistic regression classification model is improved by 11.02%. At the same time, the accuracy of LightGBM (Focal Loss) is 0.982, which is close to the accuracy when the sample is sufficient. As the sample imbalance rate decreases and the actual number of ice-covered samples further decreases, the advantages of the sample expansion algorithm gradually become apparent. When the sample imbalance rate is 10%, compared to the unexpanded samples, the accuracy of Logistic regression model is improved by 13.55%. When the sample imbalance rate is 5% and the actual number of ice-covered samples is only 15, compared to the unexpanded samples, the accuracy of Logistic regression, KNN, XGBoost, and LightGBM models has improved by 35.85%, 4.52%, 9.32%, and 9.18%, respectively. This indicates that CTGAN has good sample generation ability and can effectively learn the distribution of real samples even when the sample data is small.

From the simulation analysis, the following conclusions can be drawn: (1) The mixed features constructed based on the sliding window algorithm in this paper can significantly improve the classification ability of each model. At the same time, the LightGBM model combined with mixed feature information has obvious advantages compared to other models. (2) The sample generation model CTGAN can effectively learn the distribution of real samples, and compared to other data augmentation methods, it can generate new samples that are more similar to real samples. (3) By using the Focal loss function to modify the LightGBM model, the model's ability to distinguish easily confused samples has been increased. In addition, based on the SHAP attribution theory, the importance of each icing factor was analyzed, and the quantitative impact of key features on the diagnostic results was quantified, improving the credibility of the model's diagnostic results.