Ultrasonic wind measurement instruments are widely used in meteorological monitoring, wind energy assessment, and environmental safety due to their simple structure, fast response, and strong adaptability. However, under complex wind field conditions, the shadow effect can introduce significant measurement errors. To reduce these errors, this article presents a novel ultrasonic wind measurement instrument. First, computational fluid dynamics (CFDs) was employed to optimize the probe geometry, transducer arrangement, and relative positioning, thereby minimizing instrument-induced disturbances to the ambient flow. Second, an error-correction model was developed by integrating CFD simulation data with a neural network algorithm. This model captures local disturbance characteristics and establishes a nonlinear mapping between wind speed and direction errors and the undisturbed flow field, enabling dynamic correction of instrument outputs. Finally, a wind-tunnel experimental platform was constructed to evaluate the proposed instrument. Experimental results show that the wind speed error was reduced to 0.061 m/s, and the wind direction error decreased to 0.221 degrees, confirming the effectiveness of the proposed design and correction approach.

Ground-based air temperature measurements are highly susceptible to solar and environmental long-wave radiation, often leading to temperature deviations of up to approximately 1 degrees C. This study proposes a low-uncertainty air temperature measurement instrument specifically designed to minimize such radiation-induced temperature deviations. Computational fluid dynamics (CFD) simulations were first performed to optimize the structural design of the instrument, thereby enhancing internal airflow and improving both radiation shielding efficiency and convective heat dissipation. Subsequently, the radiation-induced temperature deviations of the optimized structure were quantitatively analyzed under various environmental conditions. A multi-layer perceptron (MLP) neural network was then employed to develop a temperature deviation correction model using the CFD-generated dataset. Finally, field comparative experiments were conducted using a 076B aspirated temperature measurement instrument as a reference. Experimental results show that, before correction, the proposed instrument exhibited a maximum radiation-induced temperature deviation of 0.43 degrees C and a mean temperature deviation of 0.31 degrees C. The root mean square error (RMSE), mean absolute error (MAE), and correlation coefficient (r) between the experimental and predicted radiation-induced temperature deviations were 0.08 degrees C, 0.07 degrees C, and 0.99, respectively. After applying the correction model, the maximum radiation-induced temperature deviation decreased to 0.12 degrees C, and the mean deviation decreased to 0.02 degrees C, demonstrating excellent consistency. In conclusion, the proposed instrument achieves efficient ventilation and low-uncertainty temperature measurement without relying on mechanical components.

Solar radiation frequently causes traditional temperature sensors to deviate from the true ambient temperature, with errors reaching approximately 1 degrees C, thereby considerably affecting the reliability of meteorological data. To address this issue, this study presents a novel temperature sensor designed for natural ventilation, utilizing an optimized thermal design structure and computational fluid dynamics (CFD) simulation techniques to analyze its thermal performance under complex environmental conditions. Additionally, a radiative error correction model was developed using neural network algorithms to enhance measurement accuracy further. Field experimental results indicate that the sensor effectively reduces radiative error to below 0.1 degrees C, with root mean square error (RMSE) and mean absolute error (MAE) values of 0.047 degrees C and 0.038 degrees C, respectively. This research demonstrates the significant engineering applicability of this design in the field of atmospheric temperature measurement.

In this paper, a highly sensitive dual-offset optical fiber temperature sensor based on Vernier effect is proposed. The dual-offset optical fibers (MMF1 and MMF2) are fabricated using fiber fusion technology. Both MMF1 and MMF2 have an offset size of 47 mu m, with offset lengths of 847 mu m and 533 mu m, respectively. Secondly, SU-8 photoresist is applied into the microcavity of MMF2. The photoresist is then exposed to ultraviolet light, enabling it to completely encapsulate MMF2. Incident light is split into two parts: one enters MMF1, and the other leaks into the air. The light entering the MMF1 propagates forward and is split into two beams again. One beam (I1) propagates into the SU-8 photoresist, and the other beam (I2) propagates in MMF2. Similarly, the beam leaking into the air is also split into two beams with intensities I3 and I4, respectively. I1 and I2 form the first Mach-Zehnder interferometer (MZI), and I3 and I4 form the second MZI, forming a parallel MZI. Since the free spectral ranges of the interference spectra of the two MZIs are relatively close, a Vernier envelope is obtained. In the temperature range of 30-55 degrees C, the wave node of the Vernier envelope near 1488 nm is monitored in detail. The temperature sensitivity reaches 3.2586 nm/degrees C. The proposed temperature sensor features high sensitivity, low cost, and a simple preparation process, with broad application prospects in intelligent devices, material research, and tumor therapy.

High-accuracy surface air temperature measurements (typically within 0.05-0.1 °C) are essential for atmospheric research and climate science applications. However, such measurements are affected by multiple sources of uncertainty, including radiation-induced temperature deviations and sensor time response characteristics. To address these challenges, this study develops and experimentally evaluates a high-accuracy air temperature measurement instrument designed to minimize radiation-induced errors. The principal sources of measurement deviation-direct solar radiation, diffuse and reflected components, long-wave radiation, altitude-related air density, wind speed, and solar incident angle-are systematically analyzed. A novel instrument incorporating a streamlined air-guiding structure is proposed to enhance convective ventilation and reduce radiative heating. A computational fluid dynamics (CFD) model is established to quantify radiation-induced temperature deviations under various environmental conditions. Based on the CFD-generated dataset, a multilayer perceptron-based correction model is developed to perform multi-parameter nonlinear correction of radiation-induced temperature deviations. Comparative field experiments, conducted using a 076B fan-aspirated instrument as a reference, show that the proposed instrument achieves a root mean square error of 0.019 °C and a mean absolute error of 0.015 °C under the tested conditions. Radiation-induced temperature deviations are effectively constrained within 0.05 °C, indicating that the combined structural design and correction approach can significantly improve measurement accuracy.

A highly sensitive bullet-shaped fiber-optic Michelson temperature sensing probe with a wide measurement range is proposed in this paper, and it is fabricated using fiber fusion tapering technology and ultraviolet curing technology. First, a section of fiber taper is fused at the center of a multimode fiber (MMF) with a flat-cut end. The length and diameter of the fiber taper are 52 μm and 52 μm, respectively. Then, the fiber taper is completely encapsulated by SU-8 photoresist. The fiber taper and SU-8 photoresist form microcavity 1 and microcavity 2, respectively. The two microcavities are arranged in parallel at the end of the MMF, and a bullet-shaped fiber sensing probe is formed. The optical paths of two microcavities are close. According to the theory of the Vernier effect, a Vernier envelope appears in the interference spectrum of the proposed fiber-optic sensing probe. However, the Vernier envelope is not directly observed in the interference spectrum due to the presence of a microcavity 3. Microcavity 3 is formed by the superposition of microcavity 1 and microcavity 2. The distinct Vernier envelope can be observed by filtering out the interference spectrum of microcavity 3. The shifts of the Vernier envelope and the high-frequency peak with temperature are monitored. The sensitivity of this probe reaches 2.2627nm/°C within the temperature range of -10 °C to 62 °C. The proposed sensing probe features a ultra-compact structure, a wide temperature measurement range and high sensitivity, making it a promising breakthrough in the field of temperature monitoring.

Global temperatures are rising by approximately 0.1 degrees C per decade. Existing air temperature measurement systems often report temperatures higher than actual air temperature due to the effects of solar radiation, leading to errors of up to 1 degrees C. As a result, there is an urgent need for a new temperature measurement system with improved radiation protection and ventilation capabilities. Furthermore, a specialized temperature error correction model is essential for the new system. Computational fluid dynamics (CFD) software was employed to simulate the radiation shielding and ventilation performance of the new system. Temperature differences between the new system and actual air temperature under various environmental conditions were quantified using CFD software. Subsequently, a specialized temperature difference correction model, incorporating multiple environmental variables, was developed using a neural network algorithm. Finally, the measurement accuracy of the new system was evaluated through field comparison experiments. During the experiments, a 076B fan aspirated temperature measurement system with an error of less than 0.03 degrees C served as the reference system. Before correction, the new system exhibited a maximum temperature difference of 0.69 degrees C and an average temperature difference of 0.35 degrees C compared to the reference system. The mean absolute error, root mean square error, and correlation coefficient between the temperature differences from the correction model and the experimental data were 0.07 degrees C, 0.08 degrees C, and 0.9 degrees C, respectively. After correction, the average temperature difference decreased to 0.06 degrees C. These results indicate that the new system has significant potential for highaccuracy temperature measurement.

To ensure the accuracy and stability of electron density measurement by the laser-aided diagnostic system like polarimeter/interferometer (POINT) during the Experimental Advanced Superconducting Tokamak (EAST) experiment campaigns, a real-time laser frequency stabilization hardware solution is proposed in detail in this paper. The device is based on an incremental Proportional-Integral-Derivative (PID) control method tuned by the Back Propagation (BP) neural network, which stabilizes the intermediate-frequency (IF) of the lasers at the target value. Firstly, a real-time data acquisition and frequency calculation module is designed and implemented. Then, the BP neural network is used to calculate and output an appropriate control rate value for the current frequency. After appropriate adjustment according to this value, it is output to the digital-to-analog converter and finally applied to the piezo-electric transducer of the laser, so as to precisely adjust the length of the laser cavity and realize the stable control of the IF. The results indicate that the proposed method has achieved a rapid and stable adjustment of the POINT laser system. The control method meets the control requirements of stabilization with well self-adaptation ability and dynamic performance for POINT system on EAST.

Air temperature sensors often exhibit measurement inaccuracies due to solar radiation effects, leading to overestimations of actual air temperature by up to 1 degrees C. To address this issue, a novel temperature sensor with enhanced radiation shielding and ventilation was developed. The proposed design incorporates an S-shaped flow diversion device, two silver-plated mirrored aluminum plates, two hemispherical flow diversion structures, and a sensor probe. Computational fluid dynamics (CFDs) simulations were conducted to evaluate radiation shielding efficiency and airflow performance. Radiation-induced errors were assessed under varying environmental conditions, and a neural network algorithm was implemented to develop a correction model. Experimental results showed that, prior to correction, the average radiation-induced errors for the new sensor, the 41 003 model, and the 43 502 model were 0.443 degrees C, 0.533 degrees C, and 0.317 degrees C, respectively, with maximum errors of 0.996 degrees C, 1.42 degrees C, and 0.905 degrees C. After correction, the average radiation-induced error of the new sensor decreased to 0.147 degrees C, with a maximum of 0.485 degrees C. The 41 003 and 43 502 models achieved reductions to 0.171 degrees C (maximum: 0.598 degrees C) and 0.118 degrees C (maximum: 0.37 degrees C), respectively. The new sensor demonstrated a mean absolute error (MAE) of 0.434 degrees C, root mean square error (RMSE) of 0.351 degrees C, and correlation coefficient ( ${r}$ ) of 0.855 in corrected radiation-induced errors compared to experimental data. In contrast, the 41 003 model exhibited an MAE of 0.604 degrees C, RMSE of 0.471 degrees C, and r of 0.832, while the 43 502 model achieved an MAE of 0.268 degrees C, RMSE of 0.218 degrees C, and r of 0.901. These findings highlight the effectiveness of the new sensor design and correction model in mitigating radiation-induced measurement errors, providing improved accuracy for atmospheric temperature sensing applications.

This study presents a novel radiation measurement system capable of simultaneously measuring solar and upward longwave radiation, with the goal of achieving measurement accuracy within +/- 5% under the tested experimental conditions. A multiphysics heat transfer analysis based on computational fluid dynamics (CFDs) was first conducted to quantify the influence of key environmental factors on the thermal response of the sensing elements. Subsequently, an environmental correction model was developed using a multilayer perceptron (MLP) neural network to compensate for the nonlinear effects of meteorological variables. Finally, a field comparison platform was constructed to assess the system's performance. During the experiments, solar radiation data from a Kipp and Zonen CMP10 pyranometer and longwave radiation values derived from the Stefan-Boltzmann law were used as reference standards. The results showed that the relative errors for solar and longwave radiation measurements ranged from -3.66% to 3.69% and -3.86% to 3.81%, respectively. The root mean square errors (RMSEs) between the estimated and measured values were 15.4 W/m2 for solar radiation and 16.7 W/m2 for longwave radiation, with corresponding mean absolute errors (MAEs) of 9.8 and 11.4 W/m2. The correlation coefficients were 0.98 and 0.96, respectively, indicating a strong agreement with the reference data. These results demonstrate the high accuracy and robustness of the proposed system, highlighting its potential for applications in energy balance analysis, climate monitoring, and agroecological research.

Purpose Precise temperature measurements are crucial for understanding Earth’s energy balance and for accurately predicting future climate change. Therefore, atmospheric temperature observations using radiosonde sensors require enhanced accuracy, targeting measurements with a precision of 0.1 K or better. Design/methodology/approach First, temperature errors of radiosonde sensors were simulated using computational fluid dynamics (CFD) from sea level up to an altitude of 32 km. These simulations accounted for a range of environmental factors, including solar radiation intensity, solar radiation angle, air velocity and altitude (air density). A neural network algorithm was then applied to learn and model the CFD-derived temperature errors. Based on this, a temperature error correction algorithm for radiosonde sensors was developed. Findings Experimental results demonstrated that the average absolute error between the measured temperature errors and the values corrected using the algorithm was 0.019 K, with a root mean square error of 0.018 K and a correlation coefficient of 0.99. These findings suggest that the temperature error correction algorithm effectively reduces measurement errors to approximately 0.05 K. Social implications The widespread adoption of this technology can impact various aspects of society, including enhancing the overall quality of meteorological observation networks and providing more accurate meteorological data support for multiple fields, such as agriculture, disaster early warning, and public health. Originality/value This study focuses on developing a correction algorithm for radiation-induced errors in sounding temperature sensors by integrating CFD with neural network algorithm. This approach aims to enhance the accuracy of temperature observations from sounding sensors, minimizing biases caused by solar radiation. The improved precision in temperature measurements will contribute to more reliable historical temperature data, thereby supporting research in climate change by providing accurate datasets for long-term climate analysis.

PurposeAccurate atmospheric temperature measurement is crucial for climate research and weather forecasting, but radiosonde sensors suffer from radiation-induced errors especially at high altitudes. Current methods inadequately address multi-physics interactions, leading to uncertainties in climate data. This study aims to propose a high-precision correction framework for these sensors.Design/methodology/approachThis study develops a hybrid mechanistic-data-driven modeling framework that combines multi-physics computational fluid dynamics (CFD) simulations with neural network learning. The CFD model simulates radiation-induced temperature errors of radiosonde sensors over altitudes ranging from 0 to 32 km, accounting for solar radiation, airflow characteristics and convective-radiative heat transfer mechanisms. A backpropagation neural network is trained on the CFD-generated data set to capture the complex nonlinear relationships between environmental variables and sensor error responses. Experimental validation is performed on a custom-built platform featuring a solar simulator and a low-pressure wind tunnel, designed to emulate stratospheric environmental conditions.FindingsThe algorithm reduces errors to 0.025 K (mean absolute error) and 0.05 K (root mean square error), with a correlation coefficient of 0.998. Altitude and solar irradiance dominate errors, while increased airflow suppresses deviations by up to 82.5%, highlighting convective cooling loss as the primary error driver.Originality/valueThis work pioneers the integration of multi-physics CFD and machine learning for radiosonde error correction, achieving sub-Kelvin accuracy. By combining physical interpretability with the flexibility and efficiency of machine learning, the hybrid approach establishes a novel paradigm for atmospheric sensor calibration and high-precision environmental monitoring.

Excellent architectural design, energy-efficient control systems, and smart home technologies need to take into account the influence of solar radiation. Therefore, there is a necessity for high-precision measurement of solar radiation. However, existing solar radiation instruments are susceptible to environmental factors such as wind speed, air temperature, and air density, resulting in significant measurement errors. Therefore, this paper proposes the design of a solar radiation measurement instrument based on the thermoelectric effect. By integrating neural network algorithms, this instrument can mitigate the influence of environmental factors on solar radiation measurement. First, employing computational fluid dynamics (CFD) for multi-physics simulations of the instrument yielded solar radiation values under various environmental parameters. Subsequently, employing neural network algorithms to train and learn from the CFD simulation results, a quantitative relationship between solar radiation values and environmental parameters was established. This formed a radiation measurement error correction algorithm to mitigate the influence of environmental parameters on solar radiation observation results. Finally, constructing a radiation observation platform validated the measurement accuracy of the instrument. The experimental results indicate that the maximum radiation error of the new instrument is -3.97%, with an average radiation error of -0.16%, and the full-scale radiation error is less than 3.88%.

Global temperatures are rising at approximately 0.1 degrees C per decade. Existing air temperature observation sensors often measure temperatures higher than the actual values due to solar radiation effects, leading to errors of up to 1 degrees C, which significantly affects the accuracy of meteorological observations. Traditional naturally ventilated temperature sensors exhibit significant limitations in reducing radiation errors, making it challenging to meet the precision requirements of 0.05 degrees C or even higher in atmospheric science research. To address this challenge, this paper proposes and designs a novel naturally ventilated temperature sensor. The core sensing element of the sensor employs a Pt100 thin-film platinum resistor, and its performance is optimized through computational fluid dynamics (CFD) methods and a multi-layer perceptron (MLP) network to reduce radiation errors. External field comparative experiments with the 076B artificially ventilated temperature monitoring device, using its measurements as temperature references, have validated the effectiveness of the new sensor in reducing radiation errors. Experimental results indicate that the new sensor has a root mean square error (RMSE) of 0.047 degrees C, a mean absolute error (MAE) of 0.039 degrees C, and a correlation coefficient (r) of 0.999. The average radiation error of the calibrated sensor is 0.03 degrees C. These findings fully demonstrate the significant advantages of this sensor in improving the accuracy of temperature measurements.

Computational Fluid Dynamics (CFD) has been employed to study the effects of sensor structure, materials, and environmental factors on radiation-induced error and to explore their inherent variation patterns for correcting radiation-induced errors in radiosonde temperature data. Initially, CFD was used to assess the impact of different sensor structures, sizes, ascent speeds, altitudes, solar radiation intensity, and radiation directions on radiationinduced error. Subsequently, a four-wire sounding temperature sensor was designed. Simulation results indicate that the four-wire sensor exhibits excellent radiation thermal balance in three-dimensional space under various solar radiation directions. Therefore, the challenging factor of solar radiation direction can be disregarded in the radiation-induced error correction model. A multi-variable input radiation-induced error correction model for the four-wire sensor was developed using a neural network algorithm. Finally, experimental studies on the fourwire sensor and its radiation-induced error correction model were conducted using a low-pressure wind tunnel and a full-spectrum solar simulator.

As terahertz (THz) phased arrays antennas (PAA) scale up, the accompanying increase in power density and heat generation poses significant challenges for thermal management. The aim of this work is to explore effective solutions for achieving an ideal uniform temperature distribution and lower peak temperature in the context of large-scale small heat sources. Considering the high-efficiency heat transfer and surface temperature uniformity of microchannel heat sinks, this paper proposes a novel composite microchannel heat sink structure based on traditional microchannel heat sinks. Using peak temperature, pressure drop, temperature uniformity, thermal stress, and thermal deformation as key indicators, a comprehensive numerical simulation analysis of the novel composite microchannel heat sink and traditional microchannel heat sinks under different Reynolds numbers (Re) were conducted based on computational fluid dynamics (CFD) and elasticity mechanics. The results show that the novel composite microchannel heat sink exhibits superior fluid flow and heat transfer performance with better temperature uniformity. At Re = 1500, it achieves improvements of 11.2 %, 9.2 %, 14.6 %, and 8.2 % compared to the traditional microchannel heat sinks. Moreover, it can achieve the same peak temperature as traditional microchannel heat sinks with lower pumping power. Furthermore, it was found that the novel composite microchannel heat sink can effectively reduce the pressure drop in microfluidic systems. At Re = 1500, the pressure drop is reduced by 37.5 %, 39 %, 31.9 %, and 35.6 % compared to the corresponding traditional microchannel heat sink. Overall, the novel composite microchannel heat sink outperforms traditional microchannel heat sinks in both flow characteristics and temperature uniformity.

In this paper, a Mach-Zehnder fiber-optic temperature sensor based on core-offset is proposed. The sensor is prepared using fiber-optic splicing technology and UV curing technology, and the fabrication process is simple. First, the impact of the fiber offset-spliced length and offset-spliced size on spectral quality is studied. Second, the fiber structure with the offset-spliced size of 42 μm and the offset-spliced length of approximately 700 μm is further investigated. The SU-8 photoresist with high thermal expansion coefficient and high thermo-optical coefficient is coated with the offset-spliced area. Finally, a fiber Mach-Zehnder temperature sensor is obtained. Within the temperature range of 30 ℃ to 60 ℃, the temperature characteristics are investigated. The fast Fourier transform is done, and the wavelengths of the interference peak near 1300 nm is measured. Experimental results show that the sensitivity of the proposed sensor reaches 0.72096 nm/℃. Moreover, SU-8 has good acid and alkali resistance, making the proposed fiber-optic temperature sensor promising for applications in biomedicine, chemistry and other fields.

Temperature variations directly affect the electrical performance and reliability of high-power semiconductor devices. Therefore, to prevent semiconductor devices from overheating and to ensure stable operation, the design of efficient heat sinks to enhance the thermal management of microchips is particularly critical. This work focuses on high-power insulated-gate bipolar transistor (IGBT) modules as a representative application and addresses the challenges of high thermal-failure risk and stringent temperature uniformity under high power-density operation by proposing a novel liquid-cooled plate (LCP) heat sink integrating serpentine channels, cooling fins, and staggered turbulence promoters. The serpentine channel increases the contact time between the coolant and the IGBT, enhancing heat exchange, while the additional fins enlarge the heat transfer area, further improving dissipation. Additionally, various turbulence promoters are introduced into the flow channels to disrupt laminar flow and induce turbulence, thereby increasing the heat transfer coefficient and thermal exchange efficiency. Computational fluid dynamics (CFDs) simulations are employed to analyze the thermal performance of the LCP heat sink, and experimental comparisons are conducted to evaluate the effects of different turbulence promoter structures and distributions on temperature uniformity. Simulation results indicate that the inclusion of turbulence promoters reduces the IGBT maximum temperature from 369.48 to 343.94 K, while the alternating arrangement of promoters significantly improves temperature uniformity, maximizing thermal performance. Experimental results indicate that increasing the coolant flow rate can lower the maximum IGBT temperature by up to 22 K, with performance stabilizing at 8.5 L/min. Moreover, when the spacing of the heat dissipation fins is 3.5 mm, the temperature of the LCP reaches its minimum, approximately 3 K lower than that observed with other spacing configurations.

To address the critical role of atmospheric temperature in climate change and disaster monitoring, enhancing measurement accuracy to 0.1 degrees C is essential. Current instruments are susceptible to radiation interference, resulting in errors of approximately 1 degrees C. This study introduces a novel temperature sensor that improves accuracy by combining natural ventilation with forced ventilation. Silver-coated aluminum plates (95 % reflectivity) and white-coated deflectors (87 % reflectivity) minimize solar radiation errors. A neural network algorithm, along with CFD simulations, further corrects radiation errors under varying weather conditions. Field tests based on the 076B ventilation device demonstrate that this new sensor reduces the average radiation error to 0.02 degrees C, achieving a RMSE of 0.034 degrees C and a MAE of 0.028 degrees C. The correlation coefficient (r) with the reference temperature reached 0.999, demonstrating the sensor's high precision and providing an effective solution for reducing temperature measurement errors to below 0.1 degrees C. Um die entscheidende Rolle der atmosph & auml;rischen Temperatur im Klimawandel und bei der Katastrophen & uuml;berwachung zu adressieren, ist eine Verbesserung der Messgen & auml;uigkeit auf 0,1 degrees C unerl & auml;sslich. Aktuelle Instrumente sind anf & auml;llig f & uuml;r Strahlungsst & ouml;rungen, was zu Fehlern von etwa 1 degrees C f & uuml;hrt. Diese Studie stellt einen neuartigen Temperatursensor vor, der die Genauigkeit durch die Kombination von nat & uuml;rlicher Bel & uuml;ftung und erzwungener Bel & uuml;ftung verbessert. Silberbeschichtete Aluminiumplatten (95 % Reflektivit & auml;t) und wei ss beschichtete Abweiser (87 % Reflektivit & auml;t) minimieren Strahlungsfehler. Ein neuronales Netzwerk-Algorithmus sowie CFD-Simulationen korrigieren zus & auml;tzlich Strahlungsfehler unter variierenden Wetterbedingungen. Feldtests basierend auf dem 076B Bel & uuml;ftungsger & auml;t zeigen, dass dieser neue Sensor den durchschnittlichen Strahlungsfehler auf 0,02 degrees C reduziert und einen RMSE von 0,034 degrees C sowie einen MAE von 0,028 degrees C erreicht. Der Korrelationskoeffizient (r) mit der Referenztemperatur betrug 0,999, was die hohe Pr & auml;zision des Sensors demonstriert und eine effektive L & ouml;sung zur Reduzierung der Temperaturmessfehler auf unter 0,1 degrees C bietet.

Atmospheric temperature is fundamental information for various industries, such as production, life, and scientific research. The temperature error induced by the solar rays can reach 1 °C or even higher. A hemispherical shell-shaped atmospheric temperature measuring instrument that can reduce heat pollution and increase air velocity was designed. First, the instrument was optimized using computational fluid dynamics (CFD) software packages. Then, the CFD software packages were employed to quantify the temperature errors of the instrument with varying situations. A neural network model was employed to develop a temperature error correction model that can be targeted for multi-variable changes. This model provides accurate correction data when the influencing factors change continuously. Finally, field experiments were performed. The experimental data analysis indicates that the mean temperature error and the maximum error of the instrument before correction are 0.08 and 0.25 °C, respectively. The root mean square error, the mean absolute error, and the correlation coefficient between measured temperature errors from experiments and corrected temperature errors from the correction model are 0.099, 0.016, and 0.952 °C, respectively. By utilizing a temperature error correction model, the measuring error of the instrument can be minimized to a range between −0.05 and 0.04 °C. Consequently, the instrument is anticipated to enhance temperature measurement accuracy to ∼0.1 °C.