Industrial AI

Bohua Technology provides the following solutions for industrial AI
Based on machine learning and domain knowledge, combined with vibration and process data, provide early warning services for mobile equipment (large units, diesel engines, pumps, etc.)
Industrial equipment is characterized by complex operating mechanisms and various failure manifestations. Based on the current level of technology, it is difficult to accurately predict failures and analyze failures using pure big data analysis methods. Therefore, it is necessary to combine domain knowledge and solidify expert experience into the algorithm. Only through the comprehensive analysis of vibration data and process data and the extraction of key parameters can we provide early warning services for industrial equipment, build dynamic early warning models based on equipment operating conditions and loads, and further improve early warning and accuracy of equipment failures.
Provide expert system services for moving equipment (reciprocators, large units, pumps, etc.) based on machine learning and domain knowledge, combined with vibration and process data
For faults with rare samples and complex causes such as "rare faults" and "difficult faults", the system automatically implements encrypted collection and storage of abnormal data (minimum storage interval 1ms), combined with vibration data (acceleration, velocity, displacement) and pressure, Temperature, speed, flow rate, current, voltage, lubricating oil and other process data, through various professional analysis maps, provide expert diagnostic services and maintenance recommendations;
· AI algorithms are embedded in edge computing devices to achieve edge intelligence in near-field processing
Based on the accumulated data collection experience of 50,000 industrial equipment (as of November 2018) for 12 consecutive years, and the analysis of data (including waveforms and spectrum) for different types of equipment, Bohua Technology has developed an intelligent early warning system based on artificial intelligence. The model and intelligent maintenance decision model are dynamically optimized based on the data collected in real time. They have been implanted in a series of intelligent edge computing gateways independently developed by Bohua Technology to implement near-field early warning and processing of equipment failures. At present, the intelligent edge computing gateway has been It is used in batches in industrial sites to provide services for the stable operation of enterprise equipment.

1.AI-based intelligent edge computing gateway
The intelligent edge computing gateway completes the encrypted collection, processing, transmission, and storage of data before and after the fault point through intelligent pre-judgment of abnormal conditions. It also marks important priorities of data. When bandwidth or storage resources are insufficient, priority is given to important ones. Level transmission or storage of early warning status and abnormal data, reducing bandwidth and storage resource usage by 90%.

1) BH5000 IoT vibration gateway

Figure BH5000 Intelligent Edge Computing Gateway
2) BH7000 IoT vibration gateway

Figure BH7000 Intelligent Edge Computing Gateway
2. Predictive maintenance based on intelligent early warning model
The intelligent gateway uploads key data to the cloud in real time, and the cloud analyzes and processes the massive data of the same model and different devices, as well as different models and different devices. Use the time domain and frequency domain feature extraction methods to study the fault mechanism, use a combination of parametric and non-parametric methods to extract fault characteristics, establish a data-based mechanism knowledge rule and statistical information database, update and optimize the model in real time, and The optimized model is delivered to the corresponding intelligent gateway to complete the optimization and update of the intelligent threshold alarm model integrated in the cloud. At the same time, according to the previous technology accumulation, combined with the knowledge of equipment mechanism and digital twin technology, the granularity of the model is refined to a specific single device, and a digital model from the model to the specific device is established in real time, and the actual operating conditions of the equipment are associated to achieve the unit Intelligent predictive maintenance.

Figure six characteristics of the subdivision fields (pictures from "Tencent Research Institute" "artificial intelligence + manufacturing" industrial development research report ")


Figure Predictive maintenance based on artificial intelligence (picture from "Tencent Research Institute" "Artificial Intelligence + Manufacturing" Industrial Development Research Report)

来个网址稳定的看片的_a 毛视频在线免播放观看 亚洲在人线播放器_亚洲无线码_亚洲日产国码 亚洲日产国码_2019久久视频这里有精品15_久久2019精品免费视频 在线a观看v视频网站_久久2019最新视频大全_久久爱国产视频在线久 色情网站_2019最新在线观看的a_久久精品频在线2019 草莓视频官网_国产成人自拍_久久精品2019在线观看30