When unmanned rescue vehicle (URV) performing rescue missions in the disaster area, URV will not be feasible to continue to use the built-in map for path planning, and even cause more serious consequences. Therefore, when a rescue vehicle works with a human being in a dynamic environment such as disaster recovery, it is necessary to quickly complete the task of adapting to the scene and learning to perform its duties. In this paper, the search and rescue robot first collects environmental information according to the camera sensor installed by itself, and then constructs the intelligent vehicle behavior decision model from the vehicle driving efficiency and optimal path. Secondly, the search and rescue robot estimates through the improved DQN network structure value function. And update the network parameters to get the corresponding Q value through the training network. Finally, the experimental results show that the algorithm can quickly generate a safe and smooth path that satisfies the kinematic constraints of the vehicle.
{"title":"Unmanned rescue vehicle navigation with fused DQN algorithm","authors":"Wenguan Cao, Xiaoci Huang, Fanglin Shu","doi":"10.1145/3366194.3366293","DOIUrl":"https://doi.org/10.1145/3366194.3366293","url":null,"abstract":"When unmanned rescue vehicle (URV) performing rescue missions in the disaster area, URV will not be feasible to continue to use the built-in map for path planning, and even cause more serious consequences. Therefore, when a rescue vehicle works with a human being in a dynamic environment such as disaster recovery, it is necessary to quickly complete the task of adapting to the scene and learning to perform its duties. In this paper, the search and rescue robot first collects environmental information according to the camera sensor installed by itself, and then constructs the intelligent vehicle behavior decision model from the vehicle driving efficiency and optimal path. Secondly, the search and rescue robot estimates through the improved DQN network structure value function. And update the network parameters to get the corresponding Q value through the training network. Finally, the experimental results show that the algorithm can quickly generate a safe and smooth path that satisfies the kinematic constraints of the vehicle.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123844746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The flapping-wing robotic aircraft achieve the purpose of flight by imitating the flapping of bird wings in nature, which has wide application prospect. According to the bionics principle, this paper designs a small flapping-wing robotic aircraft with high flexibility, and it is driven by three motors. The wing flapping and empennage torsion are realized by gear transmission, four-link mechanism, balance and steering mechanism respectively; the minimum transmission angle of space four-link mechanism is adopted to optimize the target from the design requirements of small flapping-wing robotic aircraft. The optimization design of flapping-wing robotic aircraft is carried out to achieve the minimum transmission angle of four-link mechanism based on NSGA-II, which makes the movement of flapping-wing robotic aircraft more efficient and stable. Finally, the change rule of flapping motion is obtained based on the kinematics simulation of ADAMS, and the feasibility of the scheme is verified.
{"title":"Optimization design and analysis of the flapping-wing robotic aircraft","authors":"Chun Zhao, Jun Zhong, Chao Wang","doi":"10.1145/3366194.3366286","DOIUrl":"https://doi.org/10.1145/3366194.3366286","url":null,"abstract":"The flapping-wing robotic aircraft achieve the purpose of flight by imitating the flapping of bird wings in nature, which has wide application prospect. According to the bionics principle, this paper designs a small flapping-wing robotic aircraft with high flexibility, and it is driven by three motors. The wing flapping and empennage torsion are realized by gear transmission, four-link mechanism, balance and steering mechanism respectively; the minimum transmission angle of space four-link mechanism is adopted to optimize the target from the design requirements of small flapping-wing robotic aircraft. The optimization design of flapping-wing robotic aircraft is carried out to achieve the minimum transmission angle of four-link mechanism based on NSGA-II, which makes the movement of flapping-wing robotic aircraft more efficient and stable. Finally, the change rule of flapping motion is obtained based on the kinematics simulation of ADAMS, and the feasibility of the scheme is verified.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116551616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The combustion-powered actuator shows a great performance on driving hopping robots. However, the analysis for driving hopping is limit due to its complex driving process interacting with dynamic and thermodynamic. In order that its analysis method could be obtained, two methods are conducted, and the merits of them are compared. First, the equations of combustion-powered actuators are obtained according to the combustion engine model. Second, the effects of the thermodynamic factors, including combustion duration, combustion quality index and combustion efficiency, are studied. The results show that the distance between the cylinder and piston changes little as these factors changes, when the pressure inside of the chamber is up to the maximum pressure. According to the conclusion, the combustion process during the driving process can be approximately considered as a constant volume combustion. Third, based on the constant volume combustion hypothesis, the pressure is expressed as the function of the stroke and diameter in a closed formation. To verify this hypothesis, the simulations with the two models are conducted, which shows that the results are consistency with each other. According to the constant volume hypothesis, the pressure can be expressed in the closed solution, which is convenient for designing the parameters of the actuator.
{"title":"Analysis Method of Combustion-powered Actuators for Hopping Robots","authors":"Yunguang Luan, Xiaoyong Wang, Haibo Yang, W. Xia","doi":"10.1145/3366194.3366218","DOIUrl":"https://doi.org/10.1145/3366194.3366218","url":null,"abstract":"The combustion-powered actuator shows a great performance on driving hopping robots. However, the analysis for driving hopping is limit due to its complex driving process interacting with dynamic and thermodynamic. In order that its analysis method could be obtained, two methods are conducted, and the merits of them are compared. First, the equations of combustion-powered actuators are obtained according to the combustion engine model. Second, the effects of the thermodynamic factors, including combustion duration, combustion quality index and combustion efficiency, are studied. The results show that the distance between the cylinder and piston changes little as these factors changes, when the pressure inside of the chamber is up to the maximum pressure. According to the conclusion, the combustion process during the driving process can be approximately considered as a constant volume combustion. Third, based on the constant volume combustion hypothesis, the pressure is expressed as the function of the stroke and diameter in a closed formation. To verify this hypothesis, the simulations with the two models are conducted, which shows that the results are consistency with each other. According to the constant volume hypothesis, the pressure can be expressed in the closed solution, which is convenient for designing the parameters of the actuator.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116781341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Due to the problems of sharp tool wear, large drilling force and high temperature during the actual processing of 15-5PH stainless steel, in order to improve the processing efficiency and tool life, this work uses the finite element simulation software AdvantEdge FEM studies the cemented carbide drilling process of 15-5PH stainless steel. The multi-factor orthogonal test method is adopted to analyze the results by range analysis method to study the influence of factors such as tool's point angle, clearance angle and helix angle on axial force and temperature. A brief analysis has presented a picture of change process of the axial force and temperature with time. The results show that the point angle has the greatest influence on the axial force, the clearance angle is the second, and the helix angle has the minimum impact. The helix angle has the greatest influence on the temperature, the point angle is the second, and the clearance angle has the minimum impact on the temperature. Finally, author optimizes the geometric parameters of the drill bit for 15-5PH stainless steel and gets the optimized parameter combination. The optimal combination of parameters are expressed as point angle φ = 118°, clearance angle γ = 14°, and helix angle β = 30°.
{"title":"Optimization of Bit Geometry Parameters for 15-5PH Stainless Steel Drilling Based on Finite Element Simulation","authors":"Xueyong Liu, Minghong Wang","doi":"10.1145/3366194.3366267","DOIUrl":"https://doi.org/10.1145/3366194.3366267","url":null,"abstract":"Due to the problems of sharp tool wear, large drilling force and high temperature during the actual processing of 15-5PH stainless steel, in order to improve the processing efficiency and tool life, this work uses the finite element simulation software AdvantEdge FEM studies the cemented carbide drilling process of 15-5PH stainless steel. The multi-factor orthogonal test method is adopted to analyze the results by range analysis method to study the influence of factors such as tool's point angle, clearance angle and helix angle on axial force and temperature. A brief analysis has presented a picture of change process of the axial force and temperature with time. The results show that the point angle has the greatest influence on the axial force, the clearance angle is the second, and the helix angle has the minimum impact. The helix angle has the greatest influence on the temperature, the point angle is the second, and the clearance angle has the minimum impact on the temperature. Finally, author optimizes the geometric parameters of the drill bit for 15-5PH stainless steel and gets the optimized parameter combination. The optimal combination of parameters are expressed as point angle φ = 118°, clearance angle γ = 14°, and helix angle β = 30°.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131286062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This article elaborates that the human brain structure determines that the people cannot really do "multitasking", but can consciously overcome the bottleneck of attention thinking, distribute the attention thinking rationally. Meanwhile, a lot of experimental cases show that people can finish 2 tasks at the same time if one is familiar with one task or by shifting the thinking quickly, thus "multitasking" is achieved. And intelligent interaction based on how the attention distribute has become one of the current development trends. According to 3 types of interactions based on the progressive decreasing of attention requirement, we provide a standardized definition to the interaction design, and extract the typical proposals. Attention is one factor when a user is thinking his/her preferred interaction. It is proved that, compared with the traditional "undivided attention", "multitasking" is more flexibility in dealing with problems. It is an essential thinking ability evolved for human to adapt to the development of science and technology and improve the social productivity.
{"title":"The interactive ability design of artificial intelligence based on the attention allocation mechanism of interactive perception","authors":"He Xuejiao","doi":"10.1145/3366194.3366300","DOIUrl":"https://doi.org/10.1145/3366194.3366300","url":null,"abstract":"This article elaborates that the human brain structure determines that the people cannot really do \"multitasking\", but can consciously overcome the bottleneck of attention thinking, distribute the attention thinking rationally. Meanwhile, a lot of experimental cases show that people can finish 2 tasks at the same time if one is familiar with one task or by shifting the thinking quickly, thus \"multitasking\" is achieved. And intelligent interaction based on how the attention distribute has become one of the current development trends. According to 3 types of interactions based on the progressive decreasing of attention requirement, we provide a standardized definition to the interaction design, and extract the typical proposals. Attention is one factor when a user is thinking his/her preferred interaction. It is proved that, compared with the traditional \"undivided attention\", \"multitasking\" is more flexibility in dealing with problems. It is an essential thinking ability evolved for human to adapt to the development of science and technology and improve the social productivity.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131626208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A certain type of multi-blade centrifugal ventilating fan is taken as the research object, from the perspective of CFD, the distribution of the three-dimensional flow field inside the ventilating fan is numerically simulated, the root cause of the internal problem of the ventilating fan is determined, and the impeller is further analyzed from the flow of the ventilating fan. The rationality of the internal flow channel. The reliability of the simulation results is proved by comparison with the experimental data. The numerical simulation of the flow field has a good reference for the optimization of the ventilation fan structure and the research of noise reduction.
{"title":"Pneumatic Flow Field Analysis of Multi-blade Centrifugal Ventilation Fan","authors":"W. Deng","doi":"10.1145/3366194.3366313","DOIUrl":"https://doi.org/10.1145/3366194.3366313","url":null,"abstract":"A certain type of multi-blade centrifugal ventilating fan is taken as the research object, from the perspective of CFD, the distribution of the three-dimensional flow field inside the ventilating fan is numerically simulated, the root cause of the internal problem of the ventilating fan is determined, and the impeller is further analyzed from the flow of the ventilating fan. The rationality of the internal flow channel. The reliability of the simulation results is proved by comparison with the experimental data. The numerical simulation of the flow field has a good reference for the optimization of the ventilation fan structure and the research of noise reduction.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132184497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sub-voice (crying, laughter, sigh, etc.) carries a large amount of effective information of speakers, and has a huge auxiliary role in emotion recognition, behavior recognition, physiological and psychology research. Correct detection and recognition of subvoice is the premise of research and application. The method is divided into two phases: sub-voice detection and sub-voice recognition. The high-efficiency hybrid audio segmentation algorithm based on likelihood ratio and model pre-judgment is used to realize sub-voice detection. After detecting sub-voice, we extract grayscale spectrograms, and input them into the PCANET network to automatically extract features. The SVM model is then used for identification. The experimental results show that the detection accuracy of the proposed detection method is as high as 94.2%, and the proposed recognition method is 7.7% higher than the traditional artificial statistical feature recognition method.
{"title":"Sub-voice Detection and Recognition based on Hybrid Audio Segmentation and Deep Learning","authors":"Xiaolei Zhao, Chenyin Wang, Xibin Xu","doi":"10.1145/3366194.3366219","DOIUrl":"https://doi.org/10.1145/3366194.3366219","url":null,"abstract":"Sub-voice (crying, laughter, sigh, etc.) carries a large amount of effective information of speakers, and has a huge auxiliary role in emotion recognition, behavior recognition, physiological and psychology research. Correct detection and recognition of subvoice is the premise of research and application. The method is divided into two phases: sub-voice detection and sub-voice recognition. The high-efficiency hybrid audio segmentation algorithm based on likelihood ratio and model pre-judgment is used to realize sub-voice detection. After detecting sub-voice, we extract grayscale spectrograms, and input them into the PCANET network to automatically extract features. The SVM model is then used for identification. The experimental results show that the detection accuracy of the proposed detection method is as high as 94.2%, and the proposed recognition method is 7.7% higher than the traditional artificial statistical feature recognition method.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134549522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Magnetic resonance imaging (MRI) is now commonly used for the examination and diagnosis of joints. A key step is to segment the bones of interest in MRI. This paper presents an algorithm for automatic segmentation of shoulder joint images based on a joint convolutional neural network model, which can accurately segment glenoid and humeral head in the shoulder image. This method includes two collaborative deep learning networks. The first network uses Mask R-CNN segmentation model to perform preliminary instance segmentation of glenoid and humeral head. The second network uses the probability maps of voxel belonging to the different objects (glenoid, humeral head, and background) as the constraint of the spatial location; thereby more accurate segmentation can be obtained. There are 50 groups of MRI which are used to train and test, the accuracy of Dice Coefficient, Positive Predicted Value (PPV), and Sensitivity for glenoid and humeral head reached 0.91±0.02, 0.95±0.01, 0.94±0.02 and 0.88±0.01, 0.91±0.02, 0.90±0.02 respectively, exceeding the current advanced segmentation algorithms.
{"title":"Shoulder Joint Image Segmentation Based on Joint Convolutional Neural Networks","authors":"Yunpeng Liu, Renfang Wang, Ran Jin, Dechao Sun, Hui-xia Xu, Chen Dong","doi":"10.1145/3366194.3366235","DOIUrl":"https://doi.org/10.1145/3366194.3366235","url":null,"abstract":"Magnetic resonance imaging (MRI) is now commonly used for the examination and diagnosis of joints. A key step is to segment the bones of interest in MRI. This paper presents an algorithm for automatic segmentation of shoulder joint images based on a joint convolutional neural network model, which can accurately segment glenoid and humeral head in the shoulder image. This method includes two collaborative deep learning networks. The first network uses Mask R-CNN segmentation model to perform preliminary instance segmentation of glenoid and humeral head. The second network uses the probability maps of voxel belonging to the different objects (glenoid, humeral head, and background) as the constraint of the spatial location; thereby more accurate segmentation can be obtained. There are 50 groups of MRI which are used to train and test, the accuracy of Dice Coefficient, Positive Predicted Value (PPV), and Sensitivity for glenoid and humeral head reached 0.91±0.02, 0.95±0.01, 0.94±0.02 and 0.88±0.01, 0.91±0.02, 0.90±0.02 respectively, exceeding the current advanced segmentation algorithms.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125651203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A self-calibration method of the depth-camera based on homography matrix is proposed to improve the accuracy and ease of use in calibrating the depth-camera. Firstly, the shortcomings of the existing self-calibration methods are pointed out, and the internal conversion model of the depth-camera is established. Then, the proposed self-calibration mathematical model of the depth-camera is deduced according to the imaging relationship of image pairs, and the internal and external parameters of the depth-camera are obtained by solving the essential matrix and the corresponding matrix. Finally, the distortion parameters of the depth-camera are obtained by an immune clonal selection algorithm being used for optimizing the distortion model of the depth image. Compared with Zhang's calibration method, the relative errors of the focal length and the coordinates obtained by the proposed self-calibration calibration method are smaller, in addition it has a good restoration effect for the edge distortions of the depth images.
{"title":"Depth-Camera Calibration Optimization Method Based on Homography Matrix","authors":"Dongdong Li, Guangzhu Chen, Chunjiang Li, Xiaofeng Huang","doi":"10.1145/3366194.3366198","DOIUrl":"https://doi.org/10.1145/3366194.3366198","url":null,"abstract":"A self-calibration method of the depth-camera based on homography matrix is proposed to improve the accuracy and ease of use in calibrating the depth-camera. Firstly, the shortcomings of the existing self-calibration methods are pointed out, and the internal conversion model of the depth-camera is established. Then, the proposed self-calibration mathematical model of the depth-camera is deduced according to the imaging relationship of image pairs, and the internal and external parameters of the depth-camera are obtained by solving the essential matrix and the corresponding matrix. Finally, the distortion parameters of the depth-camera are obtained by an immune clonal selection algorithm being used for optimizing the distortion model of the depth image. Compared with Zhang's calibration method, the relative errors of the focal length and the coordinates obtained by the proposed self-calibration calibration method are smaller, in addition it has a good restoration effect for the edge distortions of the depth images.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126033958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In order to break the limitations on field applications of steady-state thermal conductivity measurement techniques, a new measuring system using the method termed point-heating steady state thermal conductivity measurement method is developed in this work. A corresponding 3D thermal transport model has been built to correlate the surface temperature rise with the incident heat flux, sample's thermal conductivity, and the location for temperature probing. The surface temperature is monitored by an infrared camera, which is easily affected by ambient temperature and humidity. BP neural network model is then employed to predict the influence of ambient temperature and humidity on the measuring instrument of thermal conductivity. The generalization and robustness of the BP neural network model are further verified by comparison with outputs of linear fitting and nonlinear fitting. The prediction model of F (x, y, z) based on BP neural network has good accuracy, and the error is between -0.17 and +0.17, which also improves the speed of measuring the thermal conductivity of the measuring instrument.
为了突破稳态导热系数测量技术在现场应用中的局限性,本文开发了一种新的测量系统,即点加热稳态导热系数测量法。建立了相应的三维热输运模型,将表面温升与入射热流密度、样品导热系数和测温位置联系起来。表面温度由红外摄像机监控,容易受到环境温度和湿度的影响。然后利用BP神经网络模型预测环境温度和湿度对导热系数测量仪的影响。通过与线性拟合和非线性拟合结果的比较,进一步验证了BP神经网络模型的泛化和鲁棒性。基于BP神经网络的F (x, y, z)预测模型具有较好的精度,误差在-0.17 ~ +0.17之间,提高了测量仪器导热系数的测量速度。
{"title":"Prediction on the influence of ambient temperature and humidity to measuring instrument of thermal conductivity based on BP neural network","authors":"Yuting Tang, Shen Xu, Ziyang Wang","doi":"10.1145/3366194.3366294","DOIUrl":"https://doi.org/10.1145/3366194.3366294","url":null,"abstract":"In order to break the limitations on field applications of steady-state thermal conductivity measurement techniques, a new measuring system using the method termed point-heating steady state thermal conductivity measurement method is developed in this work. A corresponding 3D thermal transport model has been built to correlate the surface temperature rise with the incident heat flux, sample's thermal conductivity, and the location for temperature probing. The surface temperature is monitored by an infrared camera, which is easily affected by ambient temperature and humidity. BP neural network model is then employed to predict the influence of ambient temperature and humidity on the measuring instrument of thermal conductivity. The generalization and robustness of the BP neural network model are further verified by comparison with outputs of linear fitting and nonlinear fitting. The prediction model of F (x, y, z) based on BP neural network has good accuracy, and the error is between -0.17 and +0.17, which also improves the speed of measuring the thermal conductivity of the measuring instrument.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126057011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}