A multi-factor dynamic environment compensation method for a railway infrared axle temperature detection system

Abstract

The application discloses a kind of multi-factor dynamic environment compensation methods of railway infrared axle temperature detection system, in view of the existing railway infrared axle temperature detection system using fixed static compensation coefficient, there is big temperature measurement error, poor stability of hot axle prediction under the complex outdoor working conditions such as strong light, temperature and humidity, train passing airflow, the engineering problem of this application is analyzed by the coupling interference mechanism of environmental temperature, illumination intensity, wind speed, humidity to infrared temperature measurement signal, a trackside multi-environment parameter synchronous acquisition scheme is designed, a multi-factor weighted dynamic error correction mathematical model is constructed, relying on edge computing architecture to optimize the algorithm, realize trackside terminal local real-time compensation operation, avoid cloud transmission delay defects.The experimental results show that the present application can significantly reduce the temperature measurement error introduced by environment, improve the stability of hot axle prediction, provide a theoretical model and engineering landing reference for the intelligent iteration of trackside infrared monitoring equipment.

Description

Technical Field

[0001] This invention belongs to the field of environmental compensation technology, specifically relating to the design of a multi-factor dynamic environmental compensation method for a railway infrared axle temperature detection system. Background Technology

[0002] The Railway Infrared Axle Temperature Detection System (THDS) is a core trackside online monitoring device for preventing major train accidents such as axle hot-roller and axle breakage in the railway industry. It relies on infrared non-contact temperature measurement to collect axle box and railbed temperature data around the clock. The accuracy of hot-roller warnings directly determines the level of track safety management. Existing THDS optimization research mostly focuses on a single dimension: hardware circuitry or cloud-based big data algorithms. Current research has completed hardware optimization for front-end acquisition, including differential amplification, isolated power supply, and multi-level EMC protection, effectively suppressing signal drift distortion caused by electromagnetic interference from electrified traction. However, system temperature measurement errors caused by the coupling of multiple environmental factors such as outdoor temperature, light, airflow, rain, and snow cannot be completely eliminated by hardware alone.

[0003] Existing THDS (Heat Discharge Management) devices generally come with factory-fixed static compensation coefficients, failing to differentiate between various operating conditions such as day / night illumination, seasonal temperature variations, train piston wind, and rain / snow attenuation. This leads to frequent temperature measurement distortions under extreme weather conditions, directly causing false alarms and missed alarms for hot axles, significantly increasing the workload for maintenance personnel in on-site verification. Most existing temperature correction algorithms are deployed on remote backend servers, suffering from network transmission delays and offline failures, making it difficult to meet the millisecond-level real-time temperature monitoring and early warning requirements of high-speed trains. Summary of the Invention

[0004] The purpose of this invention is to address the problem of weak adaptability of the static compensation model of the existing railway infrared axle temperature detection system, and to propose a multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system.

[0005] The technical solution of this invention is: a multi-factor dynamic environmental compensation method for a railway infrared axle temperature detection system, comprising the following steps: S1. An integrated multi-environmental sensing and acquisition unit is configured next to the railway infrared axle temperature detection system to collect multi-source environmental data synchronized with the temperature measurement data of the railway infrared axle temperature detection system.

[0006] S2. A composite filtering algorithm of amplitude limiting filter and median average is used to remove the field pulse interference noise of the temperature measurement data to obtain the reference temperature measurement value.

[0007] S3. Calculate the environmental temperature error, illumination error, wind speed error, and air humidity error of the railway infrared axle temperature detection system based on multi-source environmental data.

[0008] S4. The accurate rail bottom temperature after environmental compensation is calculated based on the reference temperature measurement value, ambient temperature error, illumination error, wind speed error, and air humidity error.

[0009] Furthermore, the integrated multi-environment sensing and acquisition unit in step S1 includes an industrial integrated digital temperature and humidity sensor, a wide-range photosensitive sensor, and a low-start-up wind speed tube sensor. The industrial integrated digital temperature and humidity sensor is arranged in the ventilated and light-shielded area on the same side as the infrared probe of the railway infrared axle temperature detection system to collect ambient temperature and humidity. The installation height and tilt angle of the wide-range photosensitive sensor are exactly the same as those of the infrared probe to collect ambient light intensity. The low-start-up wind speed tube sensor is arranged on the side of the track bottom detection section to capture the near-field wind speed generated by the piston turbulence of the passing train.

[0010] Furthermore, in step S1, the integrated multi-environmental sensing acquisition unit shares an embedded hardware local clock with the infrared probe of the railway infrared axle temperature detection system. When the train's magnetic steel triggers sampling, it synchronously latches a set of temperature measurement data and a set of multi-source environmental data to form a one-to-one matching dataset.

[0011] Furthermore, step S2 includes the following sub-steps: S21. Construct the original temperature measurement sequence based on the temperature measurement data. ,in This represents the first [number] in the original temperature measurement sequence. Temperature sample value of the frame, , Indicates the number of sampling frames.

[0012] S22, For the original temperature measurement sequence The temperature sample value of each frame is used to determine whether it meets the requirements. If so, the temperature sample value of the current frame is retained; otherwise, the temperature sample value of the current frame is discarded as an outlier, and the temperature sample value of the previous frame is used to obtain the temperature measurement sequence after amplitude limiting and filtering. This represents the threshold for temperature mutation in a single frame.

[0013] S23. Sort the temperature measurement sequences after amplitude limiting and filtering in ascending order, and calculate the reference temperature value based on the sorted temperature measurement sequences. : in This indicates the first temperature measurement in the ascending order sequence. Temperature sample value of the frame.

[0014] Furthermore, the ambient temperature error in step S3 The calculation formula is: in Indicates the temperature error coefficient. Indicates the real-time ambient temperature. This indicates the laboratory standard calibration temperature.

[0015] Furthermore, the illumination error in step S3 The calculation formula is: in Indicates the illumination error coefficient. This indicates the real-time ambient light intensity.

[0016] Furthermore, the wind speed error in step S3 The calculation formula is: in This represents the wind speed error coefficient. This indicates the instantaneous near-field wind speed.

[0017] Furthermore, the air humidity error in step S3 The calculation formula is: in Indicates the humidity error coefficient. This indicates the relative humidity of the air.

[0018] Furthermore, the precise orbital bottom temperature after environmental compensation in step S4 The calculation formula is: in Indicates the reference temperature value. Indicates ambient temperature error. Indicates illumination error, Indicates wind speed error. Indicates air humidity error. , , and These represent the normalized dynamic weights of ambient temperature error, illumination error, wind speed error, and air humidity error, respectively.

[0019] Furthermore, the normalized dynamic weights satisfy the normalization constraint. .

[0020] The beneficial effects of this invention are: (1) The present invention analyzes the multi-dimensional coupling interference mechanism of temperature, light, wind speed and humidity, and uses a trackside integrated multi-environmental sensing and acquisition unit to obtain matching measured samples, constructs a normalized weighted adaptive error correction mathematical model, and performs all algorithms locally on the trackside embedded terminal, thus solving the transmission delay problem of traditional cloud algorithms.

[0021] (2) The dual verification through laboratory simulation and long-term real vehicle test on heavy-load lines shows that the integrated hardware and software optimization scheme of the present invention can significantly reduce the temperature measurement error introduced by the environment and significantly reduce the frequency of hot axle false alarms. At the same time, the present invention is compatible with the existing THDS equipment main architecture, has controllable cost, strong implementation and can be widely used to upgrade existing line monitoring equipment. It provides a complete theoretical and engineering reference for the optimization of railway infrared axle temperature detection system. In the future, online machine learning strategies can be introduced to realize the periodic autonomous iteration of weights and further improve the compensation and adaptive capability under extreme and variable working conditions. Attached Figure Description

[0022] Figure 1 The diagram shown is a flowchart of a multi-factor dynamic environmental compensation method for a railway infrared axle temperature detection system provided in an embodiment of the present invention. Detailed Implementation

[0023] Exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be understood that the embodiments shown and described in the drawings are merely exemplary and are intended to illustrate the principles and spirit of the invention, and are not intended to limit the scope of the invention.

[0024] This invention provides a multi-factor dynamic environmental compensation method for a railway infrared axle temperature detection system, such as... Figure 1 As shown, it includes the following steps S1~S4: S1. An integrated multi-environmental sensing and acquisition unit is configured next to the railway infrared axle temperature detection system to collect multi-source environmental data synchronized with the temperature measurement data of the railway infrared axle temperature detection system.

[0025] THDS railbed infrared thermometry relies on infrared probes to receive infrared radiation from the target and convert it into photoelectric signals. Outdoor environmental parameters directly change the intensity of infrared radiation propagation and the thermal radiation base of the equipment itself, forming four independent sources of interference: First, changes in ambient temperature throughout the day and seasons alter the thermal base of the chassis and track, causing an overall shift in the temperature measurement baseline; second, direct sunlight during the day adds extra infrared radiation energy, resulting in falsely high temperature readings; third, high-speed train traffic generates piston airflow that accelerates heat dissipation from the railbed metal, leading to lower temperature readings; and fourth, rain, snow, and high humidity attenuate infrared transmission energy, causing signal distortion.

[0026] Based on this, the integrated multi-environment sensing and acquisition unit in this embodiment of the invention includes an industrial integrated digital temperature and humidity sensor, a wide-range photosensitive sensor, and a low-start-up wind speed tube sensor.

[0027] The industrial integrated digital temperature and humidity sensor is positioned in a ventilated, light-shielded area on the same side as the infrared probe of the railway infrared axle temperature detection system, away from heat-generating components such as the power supply enclosure. It is used to collect ambient temperature and humidity data. The sensor has a range of -40℃ to 70℃ and 0 to 100%RH, with a temperature accuracy of ±0.2℃ and a humidity accuracy of ±2%RH. It connects to an embedded main controller via an RS485 bus and performs zero-point calibration during each maintenance window to compensate for long-term temperature drift of the components.

[0028] The wide-range photosensor is installed at the same height and tilt angle as the infrared probe, providing unobstructed coverage of the track surface detection area for collecting ambient light intensity. With a measurement range of 0–20000 Lux, the wide-range photosensor fully covers all scenarios, including nighttime, cloudy days, and midday strong light, quantifying the positive temperature measurement error caused by sunlight superposition.

[0029] The low-start-speed tubular sensor is positioned on the side of the detection section at the bottom of the rail to capture the near-field wind speed generated by the piston turbulence as the train passes. The low-start-speed tubular sensor has a range of 0–30 m / s. The instantaneous wind speed is synchronously latched at the moment the magnet triggers the temperature measurement, characterizing the negative temperature measurement deviation caused by airflow heat dissipation.

[0030] In this embodiment of the invention, the integrated multi-environmental sensing and acquisition unit shares an embedded hardware local clock with the infrared probe of the railway infrared axle temperature detection system. When the train's magnetic steel triggers sampling, a set of temperature measurement data and a set of multi-source environmental data are synchronously latched, forming a one-to-one matching dataset. The raw data is cached locally, and exported in batches during each track opening for model fitting, completely eliminating modeling distortion caused by timing misalignment.

[0031] S2. A composite filtering algorithm of amplitude limiting filter and median average is used to remove the field pulse interference noise of the temperature measurement data to obtain the reference temperature measurement value.

[0032] Step S2 includes the following sub-steps S21 to S23: S21. Construct the original temperature measurement sequence based on the temperature measurement data. ,in This represents the first [number] in the original temperature measurement sequence. Temperature sample value of the frame, , Indicates the number of sampling frames.

[0033] S22, For the original temperature measurement sequence The temperature sample value of each frame is used to determine whether it meets the requirements. If so, the temperature sample value of the current frame is retained; otherwise, the temperature sample value of the current frame is discarded as an outlier, and the temperature sample value of the previous frame is used to obtain the temperature measurement sequence after amplitude limiting and filtering. This represents the threshold for temperature mutation in a single frame.

[0034] S23. Sort the temperature measurement sequences after amplitude limiting and filtering in ascending order, and calculate the reference temperature value based on the sorted temperature measurement sequences. : in This indicates the first temperature measurement in the ascending order sequence. Temperature sample value of the frame.

[0035] In this embodiment of the invention, the algorithm for calculating the reference temperature value has a very small computational load, making it suitable for low-computing-power embedded real-time operation of trackside MCUs.

[0036] S3. Calculate the environmental temperature error, illumination error, wind speed error, and air humidity error of the railway infrared axle temperature detection system based on multi-source environmental data.

[0037] In this embodiment of the invention, the ambient temperature error The calculation formula is: in Indicates the temperature error coefficient. Indicates the real-time ambient temperature. This indicates the laboratory standard calibration temperature.

[0038] Illumination error The calculation formula is: in Indicates the illumination error coefficient. This represents the real-time ambient light intensity (Lux).

[0039] Wind speed error The calculation formula is: in This represents the wind speed error coefficient. The value represents the instantaneous near-field wind speed (m / s). The negative sign in the formula indicates that the measured temperature value is lower when the wind speed increases.

[0040] air humidity error The calculation formula is: in Indicates the humidity error coefficient. This indicates the relative humidity of the air (%RH).

[0041] S4. The accurate rail bottom temperature after environmental compensation is calculated based on the reference temperature measurement value, ambient temperature error, illumination error, wind speed error, and air humidity error.

[0042] In this embodiment of the invention, the precise orbital bottom temperature after environmental compensation The calculation formula is: in Indicates the reference temperature value. Indicates ambient temperature error. Indicates illumination error, Indicates wind speed error. Indicates air humidity error. , , and These represent the normalized dynamic weights of ambient temperature error, illumination error, wind speed error, and air humidity error, respectively, which satisfy the normalization constraint. .

[0043] In this embodiment of the invention, the precise rail bottom temperature can be used as a reference. The system determines the shaft temperature threshold and outputs a hot shaft warning signal, while simultaneously storing the data locally and uploading it to the backend.

[0044] The effects of the present invention will be further described below with a specific experimental example.

[0045] The control group used a traditional fixed compensation algorithm and original single-end acquisition hardware, while the experimental group used the dynamic weighting algorithm and optimized shielded differential hardware provided in this embodiment of the invention. Laboratory multi-condition simulation tests and 6 months of field vehicle testing on heavy-duty electrified lines were conducted, comparing three core indicators: average temperature measurement error, data pass rate, and hot axle false alarm rate.

[0046] Among them, the laboratory multi-condition simulation test set up a comprehensive simulation test bench with controllable temperature, light, wind speed, and humidity to simulate various conditions such as cloudy days, strong midday sunlight, rain and snow, and high and low temperature coupling. Under the superposition of multiple interferences, the average temperature measurement error of the control group was 2.1℃, while the average temperature measurement error of the experimental group was only 1.3℃, with a 38.1% improvement in temperature measurement accuracy. The fluctuation range of extreme conditions was significantly reduced, and the anti-environment coupling interference performance was significantly better than that of traditional solutions.

[0047] Over 8,200 train samples were collected during on-site vehicle testing. The statistical results showed that the temperature measurement pass rate of the control group was 92.3%, and the hot axle false alarm rate was 4.9%. The temperature measurement pass rate of the experimental group increased to 98.5%, the hot axle false alarm rate decreased to 1.2%, and the overall average temperature measurement error decreased by 6.2%. The equipment operated continuously for a long time without baseline drift or failure to adapt to operating conditions, significantly reducing the workload of manual on-site verification and adapting to the complex operating environment of heavy-haul railways in the field.

[0048] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A multi-factor dynamic environmental compensation method for a railway infrared axle temperature detection system, characterized in that, Includes the following steps: S1. An integrated multi-environmental sensing and acquisition unit is configured next to the railway infrared axle temperature detection system to collect multi-source environmental data synchronized with the temperature measurement data of the railway infrared axle temperature detection system. S2. Use a composite filtering algorithm of amplitude limiting filter and median average to remove the field pulse interference noise from the temperature measurement data to obtain the reference temperature measurement value; S3. Calculate the environmental temperature error, illumination error, wind speed error, and air humidity error of the railway infrared axle temperature detection system based on the multi-source environmental data; S4. The accurate rail bottom temperature after environmental compensation is calculated based on the reference temperature measurement value, ambient temperature error, illumination error, wind speed error, and air humidity error.

2. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 1, characterized in that, The integrated multi-environment sensing and acquisition unit in step S1 includes an industrial integrated digital temperature and humidity sensor, a wide-range photosensitive sensor, and a low-start-up wind speed tube sensor. The industrial integrated digital temperature and humidity sensor is arranged in a ventilated and light-shielded area on the same side as the infrared probe of the railway infrared axle temperature detection system to collect ambient temperature and humidity. The installation height and tilt angle of the wide-range photosensitive sensor are exactly the same as those of the infrared probe to collect ambient light intensity. The low-start-up wind speed tube sensor is arranged on the side of the track bottom detection section to capture the near-field wind speed generated by the piston turbulence of the passing train.

3. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 1, characterized in that, In step S1, the integrated multi-environmental sensing acquisition unit shares an embedded hardware local clock with the infrared probe of the railway infrared axle temperature detection system. When the train's magnet triggers sampling, it synchronously latches a set of temperature measurement data and a set of multi-source environmental data to form a one-to-one matching dataset.

4. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 1, characterized in that, Step S2 includes the following sub-steps: S21. Construct the original temperature measurement sequence based on the temperature measurement data. ,in Represents the first in the original temperature measurement sequence Temperature sample value of the frame, , Indicates the number of sampling frames; S22, Regarding the original temperature measurement sequence The temperature sample value of each frame is used to determine whether it meets the requirements. If so, the temperature sample value of the current frame is retained; otherwise, the temperature sample value of the current frame is discarded as an outlier, and the temperature sample value of the previous frame is used to obtain the temperature measurement sequence after amplitude limiting and filtering. Indicates the threshold for temperature abrupt change in a single frame; S23. Sort the temperature measurement sequences after amplitude limiting and filtering in ascending order, and calculate the reference temperature value based on the sorted temperature measurement sequences. : in This indicates the first temperature measurement in the ascending order sequence. Temperature sample value of the frame.

5. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 1, characterized in that, The ambient temperature error in step S3 The calculation formula is: in Indicates the temperature error coefficient. Indicates the real-time ambient temperature. This indicates the laboratory standard calibration temperature.

6. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 1, characterized in that, The illumination error in step S3 The calculation formula is: in Indicates the illumination error coefficient. This indicates the real-time ambient light intensity.

7. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 1, characterized in that, Wind speed error in step S3 The calculation formula is: in This represents the wind speed error coefficient. This indicates the instantaneous near-field wind speed.

8. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 1, characterized in that, Air humidity error in step S3 The calculation formula is: in Indicates the humidity error coefficient. This indicates the relative humidity of the air.

9. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 1, characterized in that, The precise railbed temperature after environmental compensation in step S4 The calculation formula is: in Indicates the reference temperature value. Indicates ambient temperature error. Indicates illumination error, Indicates wind speed error. Indicates air humidity error. , , and These represent the normalized dynamic weights of ambient temperature error, illumination error, wind speed error, and air humidity error, respectively.

10. The multi-factor dynamic environmental compensation method for the railway infrared axle temperature detection system according to claim 9, characterized in that, The normalized dynamic weights satisfy the normalization constraint. .

Images