In order to more accurately detect the accuracy of word-wheel water meter digits, 2000 water meter pictures were produced, and an improved Faster-RCNN algorithm for detecting water meter digits was proposed. The impro...In order to more accurately detect the accuracy of word-wheel water meter digits, 2000 water meter pictures were produced, and an improved Faster-RCNN algorithm for detecting water meter digits was proposed. The improved Faster-RCNN algorithm uses ResNet50 combined with FPN (Feature Pyramid Network) structure instead of the original ResNet50 as the feature extraction network, which can enhance the accuracy of the model for small-sized digit recognition;the use of ROI Align instead of ROI Pooling can eliminate the error caused by the quantization process of the ROI Pooling twice, so that the candidate region is more accurately mapped to the feature map, and the accuracy of the model is further enhanced. The experiment proves that the improved Faster-RCNN algorithm can reach 91.8% recognition accuracy on the test set of homemade dataset, which meets the accuracy requirements of automatic meter reading technology for water meter digital recognition, which is of great significance for solving the problem of automatic meter reading of mechanical water meters and promoting the intelligent development of water meters.展开更多
This paper advances the viewpoints and methods of the rapid sample product trial manufacture technique for developing water meter new products by CAD and simulation, computer virtual assembling and optimizing, rapid m...This paper advances the viewpoints and methods of the rapid sample product trial manufacture technique for developing water meter new products by CAD and simulation, computer virtual assembling and optimizing, rapid machining process and measurement etc. as the design and sample product trial manufacture process of water meter new products are long in product development period, and low in product development efficiency in the present time.展开更多
The smart water meter in water supply network can directly affect water production and usage when faults occur.The traditional method of fault detection is inefficient with time lagging,which is not helpful for modern...The smart water meter in water supply network can directly affect water production and usage when faults occur.The traditional method of fault detection is inefficient with time lagging,which is not helpful for modernization of water supply system.The capability of automatic fault diagnosis of smart water meter is an important means to improve the service quality of water supply.In this paper,an automatic fault diagnosis method for the smart device is proposed based on BP neural network.And it was applied on Google Tensorflow platform.Fault symptom vectors were constructed using water meter status data and were used to train the neural network model.In order to improve the learning convergence speed and fault classification effect of the network,a method of weighted symptom was also employed.Experimental results show that it has good performance with a general fault diagnosis accuracy of 98.82%.展开更多
This paper presents an experimental study to investigate the effect of using the magnetic water conditioner on the properties of water. The water flows through a closed loop, while the pH, TDS, and hardness represent ...This paper presents an experimental study to investigate the effect of using the magnetic water conditioner on the properties of water. The water flows through a closed loop, while the pH, TDS, and hardness represent its properties. For magnetic water conditioner with flux density of 170 mT, results showed that pH increased by 15.65% for 820 minutes of non-stop circulation. The increase in pH is divided to 93.5% for the first 360 minutes, and 6.5% for the last 460 minutes. TDS and Hardness of water are not affected by the magnetic water conditioner. Water remembers and keeps the impact of passing through the magnetic field for several hours, and pH decreased by 0.642 in24 hours. While the results lead to introduce and create the magnetized water saturation curve and water memory meter.展开更多
针对光照不均匀和水表表盘雾化的指针式水表在读数检测时出现漏检、误检等问题,提出一种基于改进YOLOv5s的指针式水表读数检测方法。首先,采用Mosaic、Mixup等数据增强方法,提高模型的泛化能力;其次,引入加权双向特征金字塔网络(bilater...针对光照不均匀和水表表盘雾化的指针式水表在读数检测时出现漏检、误检等问题,提出一种基于改进YOLOv5s的指针式水表读数检测方法。首先,采用Mosaic、Mixup等数据增强方法,提高模型的泛化能力;其次,引入加权双向特征金字塔网络(bilateral feature pyramid network, BiFPN)实现更高层次的特征融合使得水表图像的深层特征图和浅层特征图充分融合,提高网络的表达能力;然后,嵌入卷积注意力机制(convolutional block attention module, CBAM),在通道和空间双重维度上强化指针式水表子表盘示数特征;最后将完全交并比损失函数(complete intersection over union loss, CIoU-Loss)替换为SIoU_Loss(scylla intersection over union loss),提升边界框的回归精度。改进算法的mAP@0.5达到97.8%,比YOLOv5s原始网络提升了3.2%。实验结果表明:该算法能有效提高指针式水表的读数检测精度。展开更多
文摘In order to more accurately detect the accuracy of word-wheel water meter digits, 2000 water meter pictures were produced, and an improved Faster-RCNN algorithm for detecting water meter digits was proposed. The improved Faster-RCNN algorithm uses ResNet50 combined with FPN (Feature Pyramid Network) structure instead of the original ResNet50 as the feature extraction network, which can enhance the accuracy of the model for small-sized digit recognition;the use of ROI Align instead of ROI Pooling can eliminate the error caused by the quantization process of the ROI Pooling twice, so that the candidate region is more accurately mapped to the feature map, and the accuracy of the model is further enhanced. The experiment proves that the improved Faster-RCNN algorithm can reach 91.8% recognition accuracy on the test set of homemade dataset, which meets the accuracy requirements of automatic meter reading technology for water meter digital recognition, which is of great significance for solving the problem of automatic meter reading of mechanical water meters and promoting the intelligent development of water meters.
文摘This paper advances the viewpoints and methods of the rapid sample product trial manufacture technique for developing water meter new products by CAD and simulation, computer virtual assembling and optimizing, rapid machining process and measurement etc. as the design and sample product trial manufacture process of water meter new products are long in product development period, and low in product development efficiency in the present time.
基金the Huaihua University Double First-Class initiative Applied Characteristic Discipline of Control Science and Engineeringthe Educational Cooperation Program of Ministry of Education of China(No.201801006090)the Hunan Provincial Natural Science Foundation of China(No.2017JJ3252).
文摘The smart water meter in water supply network can directly affect water production and usage when faults occur.The traditional method of fault detection is inefficient with time lagging,which is not helpful for modernization of water supply system.The capability of automatic fault diagnosis of smart water meter is an important means to improve the service quality of water supply.In this paper,an automatic fault diagnosis method for the smart device is proposed based on BP neural network.And it was applied on Google Tensorflow platform.Fault symptom vectors were constructed using water meter status data and were used to train the neural network model.In order to improve the learning convergence speed and fault classification effect of the network,a method of weighted symptom was also employed.Experimental results show that it has good performance with a general fault diagnosis accuracy of 98.82%.
文摘This paper presents an experimental study to investigate the effect of using the magnetic water conditioner on the properties of water. The water flows through a closed loop, while the pH, TDS, and hardness represent its properties. For magnetic water conditioner with flux density of 170 mT, results showed that pH increased by 15.65% for 820 minutes of non-stop circulation. The increase in pH is divided to 93.5% for the first 360 minutes, and 6.5% for the last 460 minutes. TDS and Hardness of water are not affected by the magnetic water conditioner. Water remembers and keeps the impact of passing through the magnetic field for several hours, and pH decreased by 0.642 in24 hours. While the results lead to introduce and create the magnetized water saturation curve and water memory meter.
文摘针对光照不均匀和水表表盘雾化的指针式水表在读数检测时出现漏检、误检等问题,提出一种基于改进YOLOv5s的指针式水表读数检测方法。首先,采用Mosaic、Mixup等数据增强方法,提高模型的泛化能力;其次,引入加权双向特征金字塔网络(bilateral feature pyramid network, BiFPN)实现更高层次的特征融合使得水表图像的深层特征图和浅层特征图充分融合,提高网络的表达能力;然后,嵌入卷积注意力机制(convolutional block attention module, CBAM),在通道和空间双重维度上强化指针式水表子表盘示数特征;最后将完全交并比损失函数(complete intersection over union loss, CIoU-Loss)替换为SIoU_Loss(scylla intersection over union loss),提升边界框的回归精度。改进算法的mAP@0.5达到97.8%,比YOLOv5s原始网络提升了3.2%。实验结果表明:该算法能有效提高指针式水表的读数检测精度。
