Mode Output Research Articles

Molecularly imprinted electrochemiluminescence-colorimetric dual-mode sensor based on Mn@NC nanozyme amplification for the detection of phoxim

The common detection methods of phoxim mostly rely on the output mode of single signal, which is easily disturbed by complex environmental media and causes errors. Herein, in order to improve the reliability of detection, we developed a dual-mode detection strategy based on nanozyme for the detection of phoxim. Firstly, manganese-nitrogen co-doped carbon nanozyme (Mn@NC) with peroxidase-like activity was synthesized by one-step pyrolysis, which was used to construct in the luminol-H2O2 electrochemiluminescence (ECL) system with molecularly imprinted polymer (MIP) technology. Among them, Mn@NC can act as a co-reaction promoter to catalyze H2O2 to produce large amounts of hydroxyl radicals (•OH), thereby significantly amplifying the cathode signal of luminol and improving the stability of the signal. Benefiting from the specific selection of phoxim by MIP, this molecularly imprinted electrochemiluminescence (MIECL) mode has considerable sensitivity and selectivity, with a low limit of detection (LOD) of 0.011 ng/mL (S/N = 3). Moreover, the •OH produced by Mn@NC-catalyzed H2O2 oxidized the colorless 3,3′,5,5′-tetramethylbenzidine (TMB) to blue oxidized TMB (oxTMB). Taking advantage of the fact that the simultaneous presence of acetylcholinesterase (AChE) and acetylthiocholine (ATCh) inhibited the activity of Mn@NC. A sensitive and simple colorimetric pesticide sensing platform was constructed by combining the nanozyme with the TMB-H2O2 system to achieve the colorimetric detection of phoxim, with the LOD as low as 1.27 ng/mL. The dual-mode sensor was successfully applied to the determination of phoxim in agricultural products with satisfactory results, showing its great potential in the monitoring of food hazards.

Spared ulnar nerve injury results in increased layer III–VI excitability in the pig somatosensory cortex

This study describes cortical recordings in a large animal nerve injury model. We investigated differences in primary somatosensory cortex (S1) hyperexcitability when stimulating injured and uninjured nerves and how different cortical layers contribute to S1 hyperexcitability after spared ulnar nerve injury. We used a multielectrode array to record single-neuron activity in the S1 of ten female Danish landrace pigs. Electrical stimulation of the injured and uninjured nerve evoked brain activity up to 3 h after injury. The peak amplitude and latency of early and late peristimulus time histogram responses were extracted for statistical analysis. Histological investigations determined the layer of the cortex in which each electrode contact was placed. Nerve injury increased the early peak amplitude compared with that of the control group. This difference was significant immediately after nerve injury when the uninjured nerve was stimulated, while it was delayed for the injured nerve. The amplitude of the early peak was increased in layers III–VI after nerve injury compared with the control. In layer III, S1 excitability was also increased compared with preinjury for the early peak. Furthermore, the late peak was significantly larger in layer III than in the other layers in the intervention and control group before and after injury. Thus, the most prominent increase in excitability occurred in layer III, which is responsible for the gain modulation of cortical output through layer V. Therefore, layer III neurons seem to have an important role in altered brain excitability after nerve injury.

HydraViT: Adaptive multi-branch transformer for multi-label disease classification from Chest X-ray images

Chest X-ray is an essential diagnostic tool in the identification of chest diseases given its high sensitivity to pathological abnormalities in the lungs. However, image-driven diagnosis is still challenging due to heterogeneity in size and location of pathology, as well as visual similarities and co-occurrence of separate pathology. Since disease-related regions often occupy a relatively small portion of diagnostic images, classification models based on traditional convolutional neural networks (CNNs) are adversely affected given their locality bias. While CNNs were previously augmented with attention maps or spatial masks to guide focus on potentially critical regions, learning localization guidance under heterogeneity in the spatial distribution of pathology is challenging. To improve multi-label classification performance, here we propose a novel method, HydraViT, that synergistically combines a transformer backbone with a multi-branch output module with learned weighting. The transformer backbone enhances sensitivity to long-range context in X-ray images, while using the self-attention mechanism to adaptively focus on task-critical regions. The multi-branch output module dedicates an independent branch to each disease label to attain robust learning across separate disease classes, along with an aggregated branch across labels to maintain sensitivity to co-occurrence relationships among pathology. Experiments demonstrate that, on average, HydraViT outperforms competing attention-guided methods by 1.9% AUC and 5.3% MAE, region-guided methods by 2.1% AUC and 8.3% MAE, and semantic-guided methods by 2.0% AUC and 6.5% MAE in multi-label classification performance.

MmSTCT: spatial–temporal convolution transformer network considering driving intention for multimodal vehicle trajectory prediction of highway

Due to the dynamic nature of traffic flow and the uncertainty of drivers' maneuvers, multimodal trajectory prediction remains challenging. In this paper, a multimodal vehicle trajectory prediction model is proposed based on the encoder-decoder structure, utilizing a spatial-temporal dilated causal convolutional transformer network for the hierarchical extraction of driving intention and spatial interaction features. Specifically, temporal information is extracted from the initial motion states by utilizing a bi-directional long short-term memory network. Subsequently, the social interaction module and spatial-temporal dependency module are introduced to model the interactive influences among surrounding vehicles and the target vehicle at the same timestamp as well as at different timestamps. Then adaptive fusion of driving intention is performed to generate multiple queries for accommodating the output of multimodal trajectory prediction module in a non-autoregressive manner. The effectiveness of the proposed model is demonstrated through experimental evaluations on the NGSIM, HighD and CitySim datasets.

Research on a novel non-uniform and asymmetric ultrasonic vibration system

Abstract A new type of non-uniform and asymmetric ultrasonic vibration system is proposed and applied to the turning process. The research results show that the design theory of the non-uniform and non-symmetric ultrasonic vibration system is accurate and reliable.The theoretical design frequency and modal simulation frequency error of Ultrasonic Vibration Turning System (UVTS) is only 2.5%. The designed non-uniform and non-symmetric ultrasonic vibration system has excellent vibration performance and stability, and the established simulation model can predict surface microtexture accurately. The vibration mode output by the non-uniform and non-symmetric ultrasonic vibration system is longitudinal-bending vibration, and the actual vibration performance is excellent. Compared with conventional cutting, the addition of longitudinal-bending ultrasonic vibration under the same conditions can give the machined surface a microscopic textured structure, reduce surface roughness and water contact angle, and improve surface wettability.The minimum surface roughness obtained in the experiment is 1.005 μm.The orthogonal experiment results show that there is a correlation between the machining parameters, surface micro-structure parameters, surface wettability, and surface roughness.

Dunhuang mural inpainting based on reference guidance and multi‐scale fusion

AbstractIn response to the inadequate utilization of prior information in current mural inpainting processes, leading to issues such as semantically unreliable inpaintings and the presence of artifacts in the inpainting area, a Dunhuang mural inpainting method based on reference guidance and multi‐scale feature fusion is proposed. First, the simulated broken mural, the mask image, and the reference mural are input into the model to complete the multi‐level embedding of patches and align the multi‐scale fine‐grained features of damaged murals and reference murals. Following the patch embedding module, a hybrid residual module is added based on hybrid attention to fully extract mural features. In addition, by continuing the residual concatenation of outputs of the hierarchical embedding module improves the ability of the model to represent deeper features, and improves the robustness and generalisation of the model. Second, the encoded features are fed into the decoder to generate decoded features. Finally, the convolutional tail is employed to propagate them and complete the mural painting. Experimental validation on the Dunhuang mural dataset demonstrates that, compared to other algorithms, this model exhibits higher evaluation metrics in the inpainting of extensively damaged murals and demonstrates overall robustness. In terms of visual effects, the results of this model in the inpainting process exhibit finer textures, richer semantic information, more coherent edge structures, and a closer resemblance to authentic murals.

Reconfigurable mode converter based on a Sb2Se3 phase change material and inverse design

In this paper, we combine the inverse design with a silicon-Sb2Se3 hybrid platform to design an on-chip mode converter that converts basic modes to higher-order modes. Firstly, we present a 1 × 2 mode converter with dimensions of 4.8 × 2.7 µm2 that enables TE0 mode input, TE0 or TE1 output in the C-band (1530 nm to 1565 nm) with an insertion loss (IL) of less than 0.8 dB and a crosstalk (CT) of less than -13 dB. Secondly, the device is extended to a 1 × 3 switchable three-mode converter. Using two controllable phase change regions as drivers, it can flexibly control the switching from TE0 mode input to three modes of TE0, TE1, or TE2 outputs, which enables mode switching and signal routing. The device can be switched between three modes and has broad application potential in broadband optical signal processing for mode division multiplexing systems, as well as optical interconnections. Finally, the device is extended to a 1 × 2 controllable (mode and power) beam splitter, which can control the power ratio between output modes. By modulating the crystallinity of Sb2Se3, the simulation achieves a multilevel switching of 36 levels (> 5-bit). These devices pave the way for high integration densities in future photonic chips.

Predictive Voltage Control in Multi-Modular Matrix Converters under Load Variation and Fault Scenario

This paper presents a model predictive control (MPC) strategy to regulate output voltages in a multi-modular matrix converter topology for isolated loads. The converter system harnesses power from a six-phase permanent magnet synchronous generator (PMSG) to deliver sinusoidal voltages to a three-phase load, with LC filters positioned at the output of each MC module within the multi-modular scheme. The proposed MPC approach ensures that the output voltages remain within acceptable ranges of magnitude, phase, and frequency, even under load variations and system faults. This control strategy is particularly suitable for uninterruptible power supply systems, microgrids or other applications where voltage regulation is critical. Experimental studies validate the effectiveness of the control strategy under various load conditions, reference voltage changes, and simulated system fault scenarios. The results highlight the robustness and reliability of the proposed voltage control using the multi-modular matrix converter.

OxcarNet: sinc convolutional network with temporal and channel attention for prediction of oxcarbazepine monotherapy responses in patients with newly diagnosed epilepsy

Objective.Monotherapy with antiepileptic drugs (AEDs) is the preferred strategy for the initial treatment of epilepsy. However, an inadequate response to the initially prescribed AED is a significant indicator of a poor long-term prognosis, emphasizing the importance of precise prediction of treatment outcomes with the initial AED regimen in patients with epilepsy.Approach. We introduce OxcarNet, an end-to-end neural network framework developed to predict treatment outcomes in patients undergoing oxcarbazepine monotherapy. The proposed predictive model adopts a Sinc Module in its initial layers for adaptive identification of discriminative frequency bands. The derived feature maps are then processed through a Spatial Module, which characterizes the scalp distribution patterns of the electroencephalography (EEG) signals. Subsequently, these features are fed into an attention-enhanced Temporal Module to capture temporal dynamics and discrepancies. A channel module with an attention mechanism is employed to reveal inter-channel dependencies within the output of the Temporal Module, ultimately achieving response prediction. OxcarNet was rigorously evaluated using a proprietary dataset of retrospectively collected EEG data from newly diagnosed epilepsy patients at Nanjing Drum Tower Hospital. This dataset included patients who underwent long-term EEG monitoring in a clinical inpatient setting.Main results.OxcarNet demonstrated exceptional accuracy in predicting treatment outcomes for patients undergoing Oxcarbazepine monotherapy. In the ten-fold cross-validation, the model achieved an accuracy of 97.27%, and in the validation involving unseen patient data, it maintained an accuracy of 89.17%, outperforming six conventional machine learning methods and three generic neural decoding networks. These findings underscore the model's effectiveness in accurately predicting the treatment responses in patients with newly diagnosed epilepsy. The analysis of features extracted by the Sinc filters revealed a predominant concentration of predictive frequencies in the high-frequency range of the gamma band.Significance. The findings of our study offer substantial support and new insights into tailoring early AED selection, enhancing the prediction accuracy for the responses of AEDs.

Ultra‐Wide Modulation and Reversible Reconfiguration of a Flexible Organic Crystalline Optical Waveguide Between 645 and 731 nm

Flexible organic crystalline optical waveguides, i.e., delivering input or self‐emit lights through various dynamic organic crystals, have attracted increasing attentions in the past decade. However, the modulation of waveguide outputs relies on chemical design and substituent modification, being time‐consuming and laborious. Here we report an elastic organic crystal that displays long‐distance light transducing capability up to 2.0 cm and ultra‐wide modulation of crystalline optical waveguides between red (645 nm) and near infrared (731 nm) in both the pristine and the elastically bent states based on a pre‐designed self‐absorption effect. The flexible organic crystalline optical waveguides can be precisely and reversibly reconfigured by controlling irradiation point. In addition, deep red amplified spontaneous emissions (ASE) that are able to transduce through a 5.0 mm bent crystal with an ultra‐low optical loss coefficient of 0.092 dB/mm has been attained. To the best of our knowledge, this is the first report of flexible organic ASE waveguides. The present study not only provides a simple yet effective strategy to remarkably modulate flexible organic crystalline optical waveguides but also demonstrates the superiority of laser over normal emission as flexible optical communication elements.

Robust fractional-order load frequency control for hybrid power system concerning high wind penetration

Hybrid power systems concerning wind farms have witnessed significant dynamic characteristic changes due to unpredicted generation output and integration mode conversion between frequency-sensitive wind plants and non-scheduled ones. This work presents a robust fractional-order load frequency control (LFC) method for a multi-area hybrid power system in the presence of wind penetration uncertainties. To characterize the injection effect of wind farms, the power system is described by a model with an uncertain wind penetration factor in a pre-estimated range. Then, an enhancement of Levy’s identification method is developed to convert this model into a fractional-order reduced structure to design the load frequency controller, which is in a form of fractional-order PID with a filter for ease of engineering implementation. The closed-loop stability is analyzed by the stability boundary locus (SBL) method under different wind penetration levels. It is shown that superior frequency response performance and system robustness can be obtained if the controller is designed in terms of the model with the worse-case wind penetration. Illustrative examples including a three-area hybrid power system under a variety of wind penetration variations are given to show the effectiveness of the proposed method.

Efficient validation of Boson Sampling from binned photon-number distributions

In order to substantiate claims of quantum computational advantage, it is crucial to develop efficient methods for validating the experimental data. We propose a test of the correct functioning of a boson sampler with single-photon inputs that is based on how photons distribute among partitions of the output modes. Our method is versatile and encompasses previous validation tests based on bunching phenomena, marginal distributions, and even some suppression laws. We show via theoretical arguments and numerical simulations that binned-mode photon number distributions can be used in practical scenarios to efficiently distinguish ideal boson samplers from those affected by realistic imperfections, especially partial distinguishability of the photons.

A Comparative Study of Artificial Neural Network Training Algorithms to Predict Photovoltaic Module Output Power

Introduction: Solar energy is a crucial component and contributes a large portion of renewable resources, the demand for which has recently increased. During previous years, it was difficult to predict the amount of energy obtained from photovoltaic systems. The angle of inclination of the solar panel, the amount of radiation, the speed of the wind, the amount of humidity, and the temperature of the weather are major factors that effectively affect the production of electrical energy. Scientists have used many strategies to predict the power generated by PV modules accurately, but each method has different pros and cons. Method: This study tested three training algorithms for artificial neural networks (ANN): scaled conjugate gradient (SCG), Levenberg Marquardt (LM), and Bayesian regularization (BR), determining which one performed best in terms of prediction speed and accuracy. Twenty-eight thousand two hundred ninety-six samples of experimental data for primary influencing environmental factors were fed into the artificial neural network, which consists of 15 hidden layers. Before training the network, we preprocessed the data to remove factors that have a secondary effect. Result: The analytical results showed that the artificial neural network trained according to the LM algorithm is the best in terms of accuracy and speed of predicting the resulting photovoltaic energy. Conclusion: The results showed that although the regression evaluation and MSE values for the LM, SCG, and BR algorithms are close (98.129%, 0.0622, 97.849%, 0.0587, 98.151%, and 0.0585, respectively), the LM training algorithm is the best in terms of speed of calculation and display of results.

Photoperiod, food restriction and memory for objects and places in mice

The suprachiasmatic nucleus (SCN) contains a population of cell-autonomous circadian oscillators essential for entrainment to daily light–dark (LD) cycles. Synchrony among SCN oscillators is modified by photoperiod and determines functional properties of SCN clock cycling, including its amplitude, phase angle of entrainment, and free running periodicity (τ). For many species, encoding of daylength in SCN output is critical for seasonal regulation of metabolism and reproduction. C57BL/6 mice do not show seasonality in these functions, yet do show photoperiodic modulation of SCN clock output. The significance of this for brain systems and functions downstream from the SCN in these species is largely unexplored. C57BL/6 mice housed in a long-day photoperiod have been reported to perform better on tests of object, spatial and fear memory compared to mice in a standard 12 h photoperiod. We previously reported that encoding of photoperiod in SCN output, evident in τ in constant dark (DD), can be blocked by limiting food access to a 4 h mealtime in the light period. To determine whether this might also block the effect of long days on memory, mice entrained to 18 h:6 h (L18) or 6 h:18 h (L6) LD cycles were tested for 24 h object memory (novel object preference, NOP) and spatial working memory (Y-maze spontaneous alternation, SA), at 4 times of day, first with food available ad libitum and then during weeks 5–8 of daytime restricted feeding. Photoperiod modified τ as expected, but did not affect performance on the NOP and SA tests, either before or during restricted feeding. NOP performance did improve in the restricted feeding condition in both photoperiods, eliminating a weak time of day effect evident with food available ad-libitum. These results highlight benefits of restricted feeding on cognitive function, and suggest a dose–response relationship between photoperiod and memory, with no benefits at daylengths up to 18 h.

Development of Steg with Concentrating System

Solar cells, also known as photovoltaic cells, convert light energy into electrical energy through the photovoltaic effect. Integrating a thermoelectric generator into a photovoltaic system enhances power output and efficiency by utilizing waste heat. Concentrators are employed to further improve solar panel efficiency. This study aims to develop a solar thermoelectric generator (STEG) with a concentrating system that includes a heat sink cooling system, thermoelectric generator, and concentrator. Three conditions will be tested to observe the power output of each solar module: PV standalone, STEG with a heat sink, and STEG with a heat sink and concentrator. The efficiency appears to improve from 9.21% for PV standalone to 10.11% for STEG with a heat sink and 10.23% for STEG with a concentrating system. The results of these studies aim to establish a benchmark for enhancing the efficiency of future STEG systems.

Design and performance analysis of an embedded amplified piezoelectric jetting dispensing valve

Abstract In order to satisfy the market's demand for high-consistency, micro-dispensing of colloids, an embedded lever-amplified piezoelectric dispensing valve based on flexure hinge is designed. This structure has stable jetting performance, strong reliability, and can achieve micro-dispensing of medium and low viscosity glue. Theoretical analysis shows that the output displacement of the piezoelectric stack can meet the displacement requirement of dispensing after being amplified by this amplification mechanism. And the output displacement and mode of the amplification mechanism are calculated and analyzed by simulation. The rationality of the simulation model is verified by experiments, and the effect law of driving voltage, signal duty cycle and working frequency on the mass of glue droplets is obtained. The results show that under the conditions of driving voltage 100 V, duty cycle 50%, glue supply pressure 0.6 MPa, working frequency 100 Hz, the consistency deviation of single dispensing amount is ± 2.01%, as a whole, 0.34-0.38 mg of uniform tiny glue dots can be obtained. The experimental results have verified the stability and micro-dispensing performance of the embedded lever amplification piezoelectric dispensing valve, providing reference for the subsequent application and research of jet dispensing.

Design of a Portable Electronic Nose for Identification of Minced Chicken Meat Adulterated With Soybean Protein Isolate

ABSTRACTThe study aimed to develop a portable electronic nose system for detecting adulteration with soybean protein isolate (SPI) in chicken meat. The system mainly consisted of three parts: the gas sensor array, the DSP28335 control board, and the upper computer. The DSP28335 control board, developed using C language, included analog to digital converter (ADC) module, digital output (DO) module, pulse width modulation (PWM) module, controller area network (CAN) module, power module, drive circuit, and so forth. The upper computer, developed using LabVIEW, facilitated user interaction with the user by primarily handling CAN configuration and monitoring, displaying and storing sensor data, temperature and flow data, and sending and monitoring electronic nose commands. The feasibility of the proposed electronic nose for characterizing adulterated chicken meat was tested on six classes of chicken meat that had been adulterated with varied quantities of SPI. The mass fractions of SPI were 0%, 5%, 10%, 15%, 20%, and 25%, respectively. On the basis of odor data from the electronic nose, K‐nearest neighbor (KNN), linear discriminant analysis (LDA), and support vector machine (SVM) were applied to qualitatively distinguish minced chicken meat with different adulteration ratios. The results showed that the SVM model had the best recognition effect. When the best parameters (c, g) were c = 16 and g = 1, the accuracy of SVM model was 97.22% and 93.75% in the training and testing sets, respectively. These results demonstrated that the portable electronic nose designed in this paper effectively identifies minced chicken meat under various adulteration conditions, enabling rapid and nondestructive detection of chicken meat adulteration.

Detection of multiple abnormalities of breast cancer in mammograms using a deep dilated fully convolutional neural network

Breast cancer is one of the serious health issues which are quite frequent among women and the mortality rate due to it can be reduced if detected and diagnosed at the early stage. In this paper, a new generalized deep learning architecture, deep dilated fully convolutional neural network (DDFCNN), is designed to detect multiple abnormalities like mass, microcalcification, and architectural distortion in mammograms. The proposed DDFCNN architecture consists of a feature and a dilation module. In feature module, convolutional kernels of various sizes are used to capture information at different scales. Since there is no loss of resolution in the output of the dilation module, it renders more details which helps in localizing the anomaly properly. Several experiments are conducted on a widely used databases — DDSM and mini-MIAS where an accuracy of 95.33%, 91.42%, and 92.67% is measured for detection of mass, microcalcification, and architectural distortion, respectively with a FPs/I of 0.43, 0.50, and 0.46 for DDSM and similar assessment for mini-MIAS database are 96.56%, 89.63%, and 95.07% with a FPs/I of 0.29, 0.51, and 0.31. Moreover, the overall performance in detection of all the abnormalities in combination is 93.14% with 0.46 as FPs/I for DDSM whereas for mini-MIAS is 93.75% with 0.37 as FPs/I. The performance is further compared with the state-of-the-other methods where the proposed approach takes an edge.

BrainSuite BIDS App: Containerized Workflows for MRI Analysis.

There has been a concerted effort by the neuroimaging community to establish standards for computational methods for data analysis that promote reproducibility and portability. In particular, the Brain Imaging Data Structure (BIDS) specifies a standard for storing imaging data, and the related BIDS App methodology provides a standard for implementing containerized processing environments that include all necessary dependencies to process BIDS datasets using image processing workflows. We present the BrainSuite BIDS App, which encapsulates the core MRI processing functionality of BrainSuite within the BIDS App framework. Specifically, the BrainSuite BIDS App implements a participant-level workflow comprising three pipelines and a corresponding set of group-level analysis workflows for processing the participant-level outputs. The Anatomical Pipeline extracts cortical surface models from a T1-weighted (T1w) MRI. It then performs surface-constrained volumetric registration to align the T1w MRI to a labeled anatomical atlas, which is used to delineate anatomical regions of interest in the MRI brain volume and on the cortical surface models. The Diffusion Pipeline processes diffusion-weighted imaging (DWI) data, with steps that include coregistering the DWI data to the T1w scan, correcting for susceptibility-induced geometric image distortion, and fitting diffusion models to the DWI data. The Functional Pipeline performs fMRI processing using a combination of FSL, AFNI, and BrainSuite tools. It coregisters the fMRI data to the T1w image, then transforms the data to the anatomical atlas space and to the Human Connectome Project's grayordinate space. The outputs of each pipeline can then be processed during group-level analysis. The outputs of the Anatomical Pipeline and the Diffusion Pipeline are analyzed using the BrainSuite Statistics Toolbox in R (bstr), which provides functionality for hypothesis testing and statistical modeling. The outputs of the Functional Pipeline can be analyzed using atlas-based or atlas-free statistical methods during group-level processing. These analyses include the application of BrainSync, which synchronizes the time-series data temporally and enables comparison of resting-state or task-based fMRI data across scans. We also present the BrainSuite Dashboard quality control system, which provides a browser-based interface for reviewing the outputs of individual modules of the participant-level pipelines across a study in real-time as they are generated. BrainSuite Dashboard facilitates rapid review of intermediate results, enabling users to identify processing errors and make adjustments to processing parameters if necessary. The comprehensive functionality included in the BrainSuite BIDS App provides a mechanism for rapidly deploying the BrainSuite workflows into new environments to perform large-scale studies. We demonstrate the capabilities of the BrainSuite BIDS App using structural, diffusion, and functional MRI data from the Amsterdam Open MRI Collection's Population Imaging of Psychology dataset.

Power Dispatching Strategy Considering the Health Status of Multi-Energy Conversion Equipment in Highway Power Supply Systems

In order to extend the service life of a highway power supply system and the level of new energy consumption, a power dispatching strategy considering the health status of multi-energy conversion equipment is proposed in this paper. Firstly, the energy and load forms of the highway power supply system are introduced, and the structure of the multi-energy conversion equipment, the topological structures of the DC–DC and DC–AC modules, and the operating characteristics are analyzed. Secondly, the module temperatures and output voltages are used as main parameters to establish the health indexes of DC–DC and DC–AC modules, and then the health index of the multi-energy conversion equipment is further calculated. Thirdly, the new energy consumption index is defined, and a multi-objective optimization model for power dispatching of highway power supply systems is established with the goal of improving the health index of multi-energy conversion equipment and the new energy consumption index. The case study shows that the power dispatching strategy in this paper can better control the temperature of each module, improve the health status of multi-energy conversion equipment, and have a high level of new energy consumption.