641 results about "Pantograph" patented technology

An urban rail train pantograph abrasion loss lightweight prediction method based on twin data is used for urban rail train pantograph abrasion loss prediction. The invention provides a lightweight model which is high in calculation efficiency and simple in structure aiming at the problem of wear loss prediction of a pantograph carbon contact strip. According to the model, the prediction robustness is ensured, and meanwhile, the wear loss prediction of the pantograph carbon contact strip can be quickly and accurately realized. In addition, twin data of pantograph abrasion are calculated through cyclic simulation by means of material parameters of the carbon slide plate, the weight and offset of the model are adjusted and optimized, and the prediction precision is improved. According to the method, firstly, collected wear data are preprocessed and then input into a lightweight network for pre-training; and then, performing parameter tuning on the pre-trained lightweight model by using twin data, and finally storing the adjusted parameters to complete prediction of the wear life. The method achieves the precise prediction of the wear loss of the pantograph carbon slide plate, and has the advantages of being small in parameter quantity and short in operation time.

The invention discloses a railway pantograph carbon slide plate wear detection method based on machine vision, and the method comprises the steps: firstly carrying out the illumination enhancement of a pantograph image through employing an SCI model which introduces a channel attention mechanism, and carrying out the super-resolution reconstruction of the pantograph image after the illumination enhancement through employing a Real-ESRGAN model; secondly, an improved M3Net model is adopted to carry out pantograph carbon slide plate area positioning, an improved M3Net model encoder adopts a residual error post-standardization structure and logarithm interval continuous relative position deviation method, a decoder adopts an MIB module and an MAB module, and then a DexiNed model introducing a CA block and a space attention mechanism is adopted to carry out carbon slide plate abrasion edge detection; and finally, judging whether uneven wear exists or not according to a wear edge detection result. According to the railway pantograph carbon slide plate wear detection method disclosed by the invention, the problem of large wear detection error of the carbon slide plate under the condition of poor image quality or complex background in the existing method is solved.

The invention relates to the technical field of anomaly detection, in particular to a pantograph arcing anomaly detection and early warning method based on fusion of ultraviolet light and infrared imaging, and the method comprises the steps: collecting an ultraviolet spectrum signal of a pantograph region in real time, and extracting arcing feature data; acquiring infrared temperature distribution images of the pantograph and the overhead line system, and extracting temperature abnormal region data; timestamp synchronization and space coordinate mapping are carried out on the arcing characteristic data and the temperature anomaly region data, and a fusion data set with time-space alignment is generated; on the basis of the fused data set, calculating the space-time similarity between the arcing feature data and the temperature anomaly region data, and outputting an arcing-temperature correlation index; constructing a Gaussian mixture model, and dynamically judging an abnormal level; and when the abnormal grade exceeds a preset threshold value, a grading early warning signal is triggered, and a fault positioning report including an arcing position and a temperature abnormal range is generated. The method has high engineering practicability, and has important significance for guaranteeing safe and stable operation of a rail transit power supply system.

The invention discloses a pantograph slide plate point cloud data splicing method and system based on a calibration block and an improved ICP algorithm, and the method specifically comprises the steps: firstly continuously outputting multi-frame slide plate contour data through a laser displacement sensor, and expanding the multi-frame slide plate contour data into three-dimensional point cloud data; processing the sliding plate data by adopting a hybrid filtering denoising algorithm to obtain a sliding plate contour data set, and converting two groups of laser point cloud data acquired by the two laser displacement sensors into the same coordinate system to form complete sliding plate surface point cloud data; thirdly, completing coarse registration on the sliding plate surface point cloud data by adopting a data rapid splicing method based on a calibration block; and finally, an improved ICP (Inductively Coupled Plasma) registration algorithm is adopted to complete rapid fine registration of the point cloud data to obtain complete sliding plate surface point cloud data. The method has high precision and real-time performance, can overcome the interference caused by the influence of the surface texture, color and dynamic measurement environment of the pantograph slide plate, and provides real-time data support for intelligent operation and maintenance of rail transit.

The invention relates to a real-time pantograph anomaly detection method based on image processing, and the method comprises the steps: building a pantograph anomaly detection model, training the model, collecting the image data and operation data of a pantograph in a train operation process in real time, transmitting the image data and operation data of the pantograph to a mercuric chloride chip, and carrying out the collection of the image data and operation data of the pantograph. Image data are input into the convolutional neural network unit for feature extraction, and operation data are input into the full-connection neural network unit for analysis; a pantograph region in the image data of the pantograph is extracted, secondary detection of a specified key point is carried out, and the abnormal condition of the pantograph is judged; if it is judged that the pantograph is abnormal in the two times of detection, the corresponding position is reported to a defect alarm, and a maintainer repairs the corresponding position; and if only one of the two times of detection is judged to be abnormal, the corresponding image is put into a suspected defect library, and if the position is abnormal again, an abnormal alarm is given, so that the accuracy and the real-time performance of the abnormal detection of the pantograph are effectively improved.

The invention belongs to the technical field of life state monitoring and maintenance management of a pantograph carbon contact strip, and particularly discloses a pantograph carbon contact strip full-life cycle data management system, which comprises a multi-level gradient feature library, a pseudo edge filtering technology, multi-scale feature extraction and radial basis function classification, full-scale identification of defects such as cracks is realized, and the microcrack detection accuracy is improved; multi-source data collaborative acquisition and three-dimensional coordinate mapping are adopted, and self-adaptive sliding window analysis is combined, so that data integrity and timeliness under different working conditions are ensured; through impact energy parameter integration, pressure fluctuation index calculation and a thermal expansion correction coefficient, a partition life prediction model is established, and the wear prediction error of a key area is reduced; high-risk area automatic marking and maintenance work order generation are realized, the RFID tracing technology is combined, the maintenance efficiency is improved, and the service life is prolonged. According to the method, the defects of a traditional method in the aspects of defect detection, dynamic monitoring and service life prediction are effectively overcome.

The invention relates to the technical field of pantograph-catenary system monitoring, and discloses a nonlinear pantograph-catenary system state prediction method based on a physical information neural network, and the method comprises the steps: obtaining the structural parameters of a catenary and a pantograph in a nonlinear pantograph-catenary system, and the train speed; a multi-expert Mama network based on Fourier transform is constructed; utilizing the first full connection layer to map input features formed by the structural parameters of the contact net and the pantograph and the train speed to a high-dimensional feature space to obtain first high-dimensional features; frequency domain features are extracted from the first high-dimensional features by using a plurality of cascaded multi-expert Mamba network layers, and time sequence features are modeled at the same time; fusing the spliced frequency domain feature and the time sequence feature by using a second full connection layer to obtain a second high-dimensional feature fusing the time-frequency energy and the dynamic evolution information; and performing nonlinear pantograph-catenary system state prediction according to the second high-dimensional features. According to the invention, the state prediction accuracy of the nonlinear pantograph-catenary system can be improved.

The invention discloses a self-adaptive grinding control method and system for a rigid contact network, and belongs to the technical field of maintenance of rail transit power supply equipment. The method comprises the following steps: acquiring pantograph-catenary action historical data of each position of the contact network along a busbar direction, and constructing a contact line damage index distribution curve; identifying a high damage interval based on the damage index, extracting roughness, ripple amplitude and oxidation thickness parameters, constructing a damage intensity vector, and calculating a damage grade score of each position by adopting a fuzzy comprehensive evaluation method; a preset polishing strategy library is matched according to the scoring result, a corresponding polishing wheel combination, the rotating speed and the contact pressure are determined, and a control parameter set is generated; the grinding device dynamically adjusts the working state according to the control parameters in the advancing process. The method achieves full-process self-adaptive closed-loop control of state recognition, strategy decision, execution control and quality feedback of contact line grinding operation.

The invention discloses an intelligent handheld pantograph tension tester which comprises a fixing mechanism used for fixing a pantograph and a testing mechanism used for testing the tension of the pantograph, the testing mechanism comprises a tension sensor, a rope assembly, a locking assembly and a handheld controller, one end of the tension sensor is connected with the fixing mechanism, and the other end of the tension sensor is connected with the rope assembly. The other end of the tension sensor is connected with the rope assembly, the locking assembly is connected with the pantograph and the rope assembly, and the handheld controller is used for controlling the pantograph tension testing process. The locking assembly comprises a clamping block, a clamping groove is formed in one side of the clamping block, a limiting module is arranged on the part, located at the top of the clamping groove, of the clamping block, the limiting module is used for locking the whole locking assembly to the corresponding position of the pantograph, a testing module is arranged on the part, located at the bottom of the clamping groove, of the clamping block, and the testing module is used for testing the locking state. A winding module is arranged at the bottom of the clamping block and used for being matched with the rope assembly for winding. The accuracy in the pantograph tension detection process is improved.

The invention relates to a pantograph (1) for a rail vehicle, the pantograph comprising a structure and an articulated frame (5) mechanically connected to the structure, and a head mechanically connected to the articulated frame, wherein the pantograph is characterised in that it comprises a safety device (36) and an actuator (30) configured to act on the articulated frame via the safety device, and wherein the safety device is configured to selectively: - mechanically couple the actuator to the articulated frame, whereby the articulated frame can be moved, by the actuator, between a raised position in which the head is in contact with a catenary and a lowered position in which the head is brought closer to the structure; and - mechanically uncouple the actuator from the articulated frame, whereby the articulated frame moves into the lowered position, under its own weight, independently of the action of the actuator.

The invention discloses a pantograph arc monitoring method based on a multi-mode arc detection network, and belongs to the technical field of pantograph arc monitoring, and the method comprises the following steps: building a multi-mode arc detection network based on a de-noising diffusion probability model DDPM and a semantic decoder, and carrying out the self-supervision training through an unmarked pantograph image, a trained multi-mode arc detection network is obtained; and inputting a pantograph image in an actual line into the trained multi-mode arc detection network to obtain an arc prediction label, and completing pantograph arc monitoring. According to the method, the visible light image features and the infrared light image features are extracted by using the de-noising diffusion probability model DDPM, robust and discriminative features can be generated, and the dependence of a multi-mode arc detection network on the arc data labeled with labels is reduced; the semantic decoder is used for fusing the multi-modal data of the image, and the dependence of the multi-modal arc detection network on the labeled arc data is reduced through the audio-visual correlation of the arc characteristics.

The invention relates to the technical field of pantograph carbon contact strip abrasion, and particularly discloses a pantograph carbon contact strip abrasion self-adaptive detection system based on multi-sensor fusion, which comprises a carbon contact strip data acquisition module, a comprehensive abrasion degree generation module, a self-adaptive adjustment generation module, an abrasion risk judgment module and a database, according to the method, surface images, edge thickness, real-time temperature and dynamic contact pressure data are collected through a camera and a multi-source sensor, the direct abrasion degree and the indirect abrasion degree are calculated, then the comprehensive abrasion degree is obtained, the historical mean value and the standard deviation of the abrasion degree are calculated according to historical data, and the comprehensive abrasion degree is obtained. The current comprehensive abrasion degree deviation degree is combined to adjust the abrasion threshold sensitivity and generate a dynamic abrasion threshold, and the detection period is adjusted accordingly, so that the defects of singleness and fixed threshold of a traditional detection method are overcome, the detection accuracy and the operation and maintenance efficiency are improved, an intelligent solution is provided for state management of the pantograph carbon contact strip, and the detection efficiency is improved. And train operation safety is guaranteed.

The invention discloses a high-speed rail pantograph infrared thermal image intelligent monitoring system, which relates to the technical field of intelligent monitoring, and comprises the following steps: determining a monitoring target and carrying out data acquisition to obtain infrared thermal image data; performing heat map analysis and anomaly detection according to the infrared thermal image data to obtain a heat map analysis result; generating a fault early warning model by using a machine learning algorithm according to a heat map analysis result; performing real-time fault diagnosis and decision support according to the fault early warning model to obtain a fault diagnosis result; generating a system optimization and maintenance suggestion according to the fault diagnosis result; performing equipment maintenance and performance improvement according to the system optimization and maintenance suggestion to obtain a maintenance result; and performing long-term monitoring and intelligent decision making according to a maintenance result. According to the efficiency of the genetic algorithm in multi-objective optimization, the accuracy of equipment fault early warning and the rationality of a maintenance strategy are improved, the operation efficiency and safety of equipment are improved, and long-term stable operation of the high-speed rail pantograph is guaranteed.

The invention relates to the technical field of image data processing, and discloses an anomaly detection method and system based on a pantograph image, and the method comprises the steps: obtaining a series of first images of a pantograph region, calculating the shielding area and illumination uniformity of the first images, and obtaining the first images of the pantograph region; if any one of the shielding area or the illumination uniformity does not meet a preset condition, screening out the corresponding first image; and identifying each part of the pantograph in the first image, judging whether each part is abnormal or not according to a preset condition, and outputting a first judgment result. Therefore, due to the fact that the part, not meeting the requirement, of the first images in the series of first images is screened out, all the components are recognized, and a more accurate and reliable pantograph anomaly detection result can be obtained.

The invention relates to the technical field of pantographs, in particular to an electrified highway pantograph control and state monitoring method, which comprises the following steps of: acquiring operation data of pantograph contact pressure, voltage, current and vehicle speed based on the real-time operation state of the pantograph, segmenting according to the number of a vehicle driving path, calculating the change trend of the contact pressure, and calculating the state of the pantograph according to the change trend. And obtaining path gradient index data. According to the invention, through segment data acquisition of contact pressure, voltage, current and vehicle speed and track serial number association processing, diversity of monitoring data and track pressure change capture capability are improved, and through cooperative association analysis of pantograph slide plate state, current change and pressure fluctuation, a fault range is effectively reduced, and fault diagnosis accuracy is improved. The method is advantaged in that fault positioning accuracy is enhanced, potential offset areas and unbalance points in the track can be identified, and through distribution characteristic analysis, contact point fluctuation marking precision is optimized.

The invention provides a training method of a pantograph fault judgment model, a fault judgment method and a train, and relates to the technical field of rail trains. The training method comprises the steps of obtaining a fault training set; and iteratively executing operation 1-operation 3: operation 1, inputting the modal information into an initial neural network corresponding to the modal information, and outputting modal features of the modal information; operation 2, the multiple modal features of the training sample are processed through a classification fusion network, a state prediction result is obtained, the state prediction result represents whether the pantograph breaks down or not, and the classification fusion network is constructed based on multiple target attention networks and a target classifier; 3, adjusting model parameters of an initial judgment model according to the state prediction result and the state label, wherein the initial judgment model comprises a plurality of initial neural networks and a classification fusion network; and under the condition of model parameter convergence, determining the initial judgment model as a trained pantograph fault judgment model.

The invention discloses a system for simulating the contact condition of a pantograph and a contact net after an earthquake. Relates to the field of railway bridge contact net anti-seismic tests. The system comprises a contact network simulation assembly, a train simulation assembly and a vibration table assembly. The contact network simulation assembly comprises a winch, a contact rope and two supporting columns. The two supporting columns are fixedly arranged at intervals, the winch is arranged at the top ends of the supporting columns and used for driving the contact rope to slide between the two supporting columns, the train simulation assembly comprises a train body, wheel parts arranged below the train body and a pantograph arranged on the train body, and the pantograph is in contact with the contact rope; the vibration table assembly comprises a table body and a vibrator used for vibrating the table body. The table body comprises a plurality of table tops which are independently arranged; the wheel part is placed on the table body; the vibration table assembly is adopted, each wheel below the train head is placed on the corresponding vibration table, and compared with the situation that a large vibration table top is placed below each wheel of a front bogie and a rear bogie, the real situation that the vibration states of all the wheels are different when a train runs on an unsmooth track after an earthquake is better met, the requirement for laboratory equipment conditions is lower, and the working efficiency is improved. Theoretically, test simulation is easier to realize, and the test result is more real, accurate and comprehensive.

The invention discloses a pantograph-contact network system operation performance evaluation method and system. The method comprises the following steps: firstly, collecting data in real time, normalizing the collected data, and screening key hidden dangers which have power-off risks and need speed reduction operation through a two-stage risk rapid judgment mechanism; after hidden dangers are eliminated, an analytic hierarchy process is adopted to calculate index weights and initial weights of all parameters, then information entropies of all the parameters are calculated based on an improved entropy weight method to determine the dispersion degree, and real-time dynamic adjustment of the parameter weights is achieved through a combined weight method; and finally, by applying a fuzzy comprehensive evaluation method and combining the dynamically adjusted parameter weights and the membership degrees of the parameters, calculating the operation performance score of each index, then obtaining the comprehensive score of the operation performance of the pantograph-catenary system, and judging the performance grade according to the score interval. The method and the system provided by the invention are suitable for accurate evaluation of the operation performance of the icing galloping section pantograph-catenary system, and can quickly identify the operation risk and scientifically judge the performance grade.

A drive system for raising or lowering a sliding piece of a current collector for rail vehicles. The sliding piece is disposed on a positioning device of the current collector, and the sliding piece is configured to be brought into contact with an overhead line using contact pressure. The drive system includes a bellows drive which is actuated by compressed air to actuate the positioning device, and a pneumatic control unit having a control valve for supplying or withdrawing compressed air to / from the bellows drive. The drive system includes a detection line and a lowering valve for venting the bellows drive when pressure drops in the detection line, the lowering valve being a 3 / 2-way valve disposed in a pneumatic supply line between the control valve and the bellows drive.

The invention discloses a spring pantograph force adjusting system which comprises a bottom frame, a spring system, a force adjusting mechanism and an electric drive system. The bottom frame is of a frame welding structure, located at the bottom of the pantograph and used for supporting connecting components, the spring system is used for driving the lower arm to enable the pantograph to rise, and the force adjusting mechanism and the electric drive system are used for adjusting force. According to the spring pantograph force adjusting system, a force adjusting mechanism, an electric drive system and a spring drive system are integrally designed at the position of a pantograph underframe, the purpose that the spring force is adjustable in the pantograph running process is achieved, and the problems that the spring force of a high-speed spring pantograph cannot be adjusted, and the pantograph net following performance under high-speed running cannot be guaranteed are solved. And the high-speed operation current collection performance of the spring pantograph is ensured.

The invention relates to the technical field of electrified highway power receiving, and particularly discloses a stable pantograph system suitable for an electrified highway and a design method of the stable pantograph system. In order to solve the problem of unstable power receiving of the pantograph caused by road bumping and vehicle body swinging in a road driving environment, the pantograph is stably attached to a linear rail by designing a stacked pantograph transverse sliding mechanism and a multi-dimensional compensation system and combining a radar / infrared sensing technology. The system can tolerate one-side deviation of a vehicle body by + / -40-50 cm, solves the technical problems of up-and-down fluctuation and left-and-right drifting of the highway pantograph, fills the blank of design of an electrified highway pantograph net system, and has remarkable practical value and application prospect.

The invention discloses a bow net offline electromagnetic interference multichannel data acquisition and analysis method and device, and belongs to the technical field of electromagnetic interference test and analysis. A transient electromagnetic signal of pantograph-catenary discharge is acquired through cooperation of multiple antennas, multiple probes and multiple channels, and strict alignment of multi-source data under the same discharge event is realized through unified triggering and time base line unification; in the off-line stage, denoising, feature extraction and multi-event statistical modeling are carried out on data generated by triggering, so that multi-path and multi-mode electromagnetic characteristics of pantograph-catenary discharge and statistical laws of the multi-path and multi-mode electromagnetic characteristics of pantograph-catenary discharge are obtained; the device comprises a data acquisition unit, a synchronous trigger unit, an analysis processing unit and an interactive display unit, and can improve data processing efficiency and analysis depth while ensuring high sampling rate and broadband coverage; the device solves the problems of insufficient space coverage, poor synchronism and limited analysis capability in the existing pantograph-catenary discharge test, and is suitable for pantograph-catenary discharge mechanism research, electromagnetic interference evaluation and related engineering application.

The invention discloses a pantograph control system and method, and relates to the technical field of pantograph catenary contact, the pantograph control system comprises a gas circuit, a pressure sensor, an actuator, a pressure fluctuation device and a control device, the gas circuit comprises a pantograph head gas circuit branch surrounding a pantograph head and a pantograph head support gas circuit branch surrounding a pantograph head support; the pressure fluctuation device is arranged in the pantograph head gas path branch and is used for moving in the pantograph head gas path branch when the pantograph vibrates, so that the pressure of the gas path fluctuates; the pressure sensor is connected with the pantograph head bracket gas path branch and the control device and is used for detecting the current pressure of the gas path in real time and transmitting the current pressure to the control device; the control device is connected with the actuator and is used for determining a vibration suppression signal to be transmitted to the actuator at the current moment according to the current pressure received in a preset time period before the current moment; and the actuator is in mechanical contact with the pantograph head bracket gas circuit branch and is used for adjusting the current vibration state of the pantograph to a preset vibration state according to the vibration suppression signal.

The invention discloses a pantograph conductive structure without a supporting insulator. The pantograph conductive structure comprises a conductive rod, a sliding rail group, an upper arm plate and two flexible wire groups, the upper end of the conducting rod is hinged with the upper arm plate, and the lower end is connected to the slide rail group; the upper arm plate is connected with the conducting rod through a flexible wire set, the conducting rod is connected with the sliding rail set through another flexible wire set, and movable connection with the pantograph is achieved. The problem that a current fixed connection transmission mode of a traditional pantograph with a supporting insulator cannot meet the movable current transmission requirement of the pantograph without the supporting insulator is solved, the pantograph can be self-adapted to various states of working or falling of the pantograph, effective current transmission of the pantograph without the supporting insulator is achieved, and electric energy is provided for a vehicle traction system.

The invention discloses a pantograph running state monitoring and analyzing method and system, and relates to the field of rail train monitoring. According to the main scheme, vibration data, contact force data, contact point temperature data and abrasion data are comprehensively analyzed; whether a pantograph pressure control instruction is triggered or not, whether a train speed limiting instruction is triggered or not and whether the pantograph has an operation fault or not are judged, a passive early warning strategy is changed into active regulation and control, the operation state of the pantograph can be evaluated more comprehensively, the frequency and cost of manual detection are reduced through the intelligent monitoring and prediction technology, and the working efficiency is improved. The evaluation accuracy is improved, and meanwhile, the maintenance cost and the safety risk caused by fault expansion are avoided.

The invention provides a pantograph crack detection method and system based on multi-sensor data fusion, and the method comprises the steps: obtaining detection data transmitted by a sensor installed on a pantograph, and carrying out the normalization processing of the detection data, so as to obtain normalized data; performing clustering processing on the normalized data to obtain a membership matrix; determining a graph structure based on the normalized data, and fusing the normalized data based on the graph structure to obtain fusion features; according to the method, the data of the pantograph can be detected in real time, fusion analysis is carried out on multi-source data through a machine learning model, early recognition, three-dimensional positioning and quantitative evaluation of cracks are achieved, and the method has the advantages of being high in accuracy, high in accuracy and high in reliability. The problems of low efficiency, poor real-time performance, high leak detection risk and the like in the prior art are solved.

The invention provides a pantograph image detection method, a training method, a train, a device and equipment. The pantograph image detection method comprises the following steps: acquiring a pantograph image acquired in the running process of a target train; the method comprises the following steps: identifying a pantograph image, and determining a to-be-detected path for the pantograph image; wherein the to-be-detected path indicates task types of at least two to-be-detected tasks needing to be executed for the pantograph image; according to the at least two task types, processing the pantograph image by using the target model, and generating an initial detection result for each task type; and based on the at least two to-be-detected tasks, performing fusion analysis on the initial detection results of the task types to generate a target detection result.

The invention provides a model training method and device, a pantograph-catenary system simulation method and device, and a medium, and relates to the technical field of pantograph-catenary simulation, and the method comprises the steps: obtaining a training sample set, constructing a target loss function comprising a physical loss function, constraining the output of a pantograph-catenary simulation model to accord with a physical rule through the physical loss function, and improving the simulation performance of the pantograph-catenary system. And finally, training the pantograph-catenary simulation model according to the training sample set, and obtaining the trained pantograph-catenary simulation model by minimizing the target loss function. By adding the physical loss function into the target loss function, the physical rule constraint output by the pantograph-catenary simulation model can be realized, and the effect of improving the prediction accuracy of the pantograph-catenary simulation model can be achieved.

The invention discloses a train pantograph-catenary system fault detection method and system based on YOLOv11, and relates to the image recognition technology, and the method comprises the steps: employing a wavelet convolution module to process feature extraction of different stages of a backbone network, carrying out the multi-scale decomposition of an input signal through wavelet transformation, and extracting high-frequency details and low-frequency contour features; adjusting weights of different brightness or color interval features by using an attention mechanism module, and simultaneously paying attention to local and global information by using parallel double-path attention; a loss function module is used for determining a proper threshold value for the sample, and the loss weight is automatically adjusted according to the characteristics of the sample and model output; the wavelet convolution module better retains image details, achieves the effect of picture deblurring, and effectively solves the problem of loss of feature problem information; the attention mechanism module reduces the influence of feature degradation caused by complex weather and background, and the loss function module enables the model to pay more attention to difficult samples and solves the problem of low detection precision of some samples.

The invention discloses a method for optimizing pneumatic lifting force of a pantograph head. An airflow adjusting convex plate used for improving a flow field structure so as to form a pressure difference between the upper side and the lower side of a carbon slide plate to improve the mean value of the pneumatic lifting force of the pantograph head is additionally arranged at the carbon slide plate of the pantograph head. The airflow adjusting convex plate is arranged at the carbon sliding plate of the pantograph head, the flow field structure is improved, the pneumatic lifting force mean value of the pantograph head is increased, the pantograph net following performance can be improved, the requirement of a higher-speed train for pantograph net contact force can be met, the current collection quality of the pantograph is improved, and due to the fact that the total lifting force target value of the pantograph head is fixed at the same speed, the current collection quality of the pantograph head is improved. And furthermore, the workload of the pantograph airbag can be reduced, the service life is prolonged, the fault occurrence frequency is reduced, and safe and stable operation of a train is ensured.