Scholarlink

In recent times, the usage of roadways and heavy vehicles for bulk load transportations has rapidly increased for fulfilling the requirements of the cargo, heavy machineries, and equipment in various places of the country. Even through the network of heavy vehicles is very important for the transportations of goods, the equipment overloading of heavy vehicles may cause road collapse, accidents due to tire burst and brakes worn. This was taken to the consideration by the MoRT&H and Section 114 was added to the Motor Vehicle Act in 1988 which classifies the type of vehicle based on axles and Gross Vehicle Weight (GVW) and is clearly defined as rule to prevent the overloaded vehicles moving on roads to stop the causes because of it. But still the awareness was not up to the level among the transportation community on exceeding the limitation of loads in the vehicle. To stop such equipment overloading, prevention method may be done with the help of modern technologies using machine learning algorithms and latest sensors which can be implemented in toll gates. Overloading may cause tires of the vehicle to burst and break failures which leads to road accidents. The overloaded vehicles have expanded tires, and they are identified visually but it doesn’t work out all the time. To overcome this, ML algorithms can be used to classify the vehicles condition. Strain gauge sensors will mind the weight of individual axle, and gross vehicle weight was calculated with weights on all axles. The STM32F205RB microcontroller sends the acquired data to the Google Cloud, and the data was retrieved by the ML algorithm and it is used to overcome the external factors which affect the precision of the weighing and help in achieving maximum accuracy. In case of detection of overloaded vehicles, strain gauge sensors will convert the mechanical strain over it to the electrical resistance strain which will act as a mechanical-electrical converter. The mechanical strain was gained by using bending plates where the plates with bending nature was fixed on the surface and strain was determined according to the bend experienced when the vehicle passed over it. Maintenance of the model is minimized since the sensors are calibrated automatically using AI/ML algorithm.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning Algorithms for Identification of Equipment Overloading

  • 章节

  • 发表日期：

    2025年

查看全文
  • B. Meena Preethi,
  • S. Gokul,
  • M. Parameshvar,
  • B. Dharshini

摘要

In recent times, the usage of roadways and heavy vehicles for bulk load transportations has rapidly increased for fulfilling the requirements of the cargo, heavy machineries, and equipment in various places of the country. Even through the network of heavy vehicles is very important for the transportations of goods, the equipment overloading of heavy vehicles may cause road collapse, accidents due to tire burst and brakes worn. This was taken to the consideration by the MoRT&H and Section 114 was added to the Motor Vehicle Act in 1988 which classifies the type of vehicle based on axles and Gross Vehicle Weight (GVW) and is clearly defined as rule to prevent the overloaded vehicles moving on roads to stop the causes because of it. But still the awareness was not up to the level among the transportation community on exceeding the limitation of loads in the vehicle. To stop such equipment overloading, prevention method may be done with the help of modern technologies using machine learning algorithms and latest sensors which can be implemented in toll gates. Overloading may cause tires of the vehicle to burst and break failures which leads to road accidents. The overloaded vehicles have expanded tires, and they are identified visually but it doesn’t work out all the time. To overcome this, ML algorithms can be used to classify the vehicles condition. Strain gauge sensors will mind the weight of individual axle, and gross vehicle weight was calculated with weights on all axles. The STM32F205RB microcontroller sends the acquired data to the Google Cloud, and the data was retrieved by the ML algorithm and it is used to overcome the external factors which affect the precision of the weighing and help in achieving maximum accuracy. In case of detection of overloaded vehicles, strain gauge sensors will convert the mechanical strain over it to the electrical resistance strain which will act as a mechanical-electrical converter. The mechanical strain was gained by using bending plates where the plates with bending nature was fixed on the surface and strain was determined according to the bend experienced when the vehicle passed over it. Maintenance of the model is minimized since the sensors are calibrated automatically using AI/ML algorithm.