Research On Tri-wavelength Non-contact Black Ice Detection Technology

Black ice on the road has a great hazard to traffic safety,however,there are few studies on black ice detection.At present,the real-time weather collection equipment used on domestic highways is mainly foreign products,and they have no black ice detection function.Given this situation,to meet the needs of real-time monitoring of road surface conditions,this research proposed a new method for black ice detection on road surface.Additionally,this method can also be used to detect other road conditions such as dryness,wetness,water accumulation,icing,and snow accumulation.At the same time,it can also provide information on the thickness of accumulated water,the thickness of ice,and the thickness of snow.It lays a solid theoretical and technical foundation for the next development of new road weather condition monitoring equipment.This research proposes a new method to detect the state of black ice.This method uses three wavelengths of 1310 nm,1430 nm,and 1550 nm as detection wavelengths.The detector receives the backscattering power corresponding to the three wavelengths under different road conditions and,converts the received backscattering power to normalized reflectance.The normalized reflectance is a newly defined quantity for the convenience of comparison,which is defined as the ratio of the backscattering power under a certain road condition to the backscattering power under the dry road condition.Under the same road condition,the backscattering power detected by the detectors on different road sections is different.Therefore,it is impossible to directly compare the absolute magnitude of the backscattering power to determine the road conditions.The normalized reflectance solves this problem well,and the detection results of different road sections can be compared with each other at the same scale,to summarize the condition criterion,thereby enabling identification of black ice road conditions and the like.In this study,the theoretical basis for detecting road conditions such as black ice is divided into two parts.The first part is the effect of roughness on backscattering.The presence of water,ice,snow,and other coverings on the asphalt pavement will cause the surface roughness to change.The roughness of snow is the largest and the roughness of icy,stagnant roads is minimal.The roughness of the black ice road surface is between that of the icy road surface and dry road surface.The second part is that the absorption coefficient of water,ice,and snow at wavelengths of 1310 nm,1430 nm,and 1550 nm are different.Previously,in most experimental studies,the distinction between icy roads and wet roads was poor,and it was easy to make the distinction after using the tri-wavelength method with different absorption rates.Comprehensively considering the difference between the roughness theory and the absorption coefficient,the detection results of black ice and other road conditions can be predicted.The experiment shows that the theoretical derivation is consistent with the experimental results.The normalized reflectance of the black ice and icing state at the wavelength of 1430 nm is greater than the normalized reflectance at 1550 nm,and the water state is the opposite;the black ice condition is always greater than the normalized reflectance of the icing state,The normalized range of icing state is0.56（29）R_ice（1310）?0,0.48（29）R_ice（1430）?0,0.28（29）R_ice（1550）?0,beyond this range,it is regarded as black ice state.The normalized reflectance in the dry state is always around 1.The normalized reflectance of the snow state at a wavelength of 1310 nm is greater than 1.Based on these,dry,water,black ice,ice,snow states have a judgment basis.