As one of the emerging green energy sources, wind power has developed rapidly in the world, especially in the past ten years, with the rapid development of wind power in China and the sharp increase in the number of wind turbines. Compared with traditional thermal power, wind power is characterized by scattered units and a large number of units. How to improve the utilization rate of wind turbines, reduce the occurrence rate and failure time of equipment failures, and avoid sudden failures of equipment at the same time have become the main daily operation and maintenance target of wind farms. With the continuous expansion of the installed wind power capacity of each power generation group, as the owner of the asset, due to the lack of informatized data support and a scientific measurement index system, it is impossible to accurately quantify the loss level of the power generation efficiency of the asset, and thus cannot effectively identify the power generation efficiency of the asset. It is also impossible to correctly evaluate and effectively motivate the production operation and maintenance system team. With the concept of big data and cloud computing, it has been integrated into the wind power industry's ability to catch wind, aerodynamic efficiency, wind energy conversion ability and other technical links of interconnection. Wind farms will truly realize intelligent operation, maintenance and management.


For a long time, the management methods of wind farms in China have been extensive and scattered: there is a serious asymmetry between the management personnel and the information of wind farms and equipment; the professional skills of personnel are uneven, and there is a lack of technical support; due to the backward operation and maintenance methods, 24-hour on-duty maintenance consumes a lot of labor costs. All of these make it difficult for wind farms to operate in a good state, lose a lot of power generation potential, fail to eliminate hidden troubles in time, and greatly increase hidden costs. It is urgent to use big data and AI technology for smart operation and maintenance of wind farms. 1. How to realize the background monitoring and full life cycle management of the remote fan to ensure the optimal operation of the fan? 2. How to timely and accurately carry out fault prediction and early warning, so as to reduce the asset loss caused by equipment failure? 3. How to make accurate wind farm power prediction, reduce power grid assessment and lower limit losses, and improve grid friendliness?


Core advantages description


As an application running on the terminal of the smart wind farm platform, DataFocus brings together the software data of various business systems such as wind turbine equipment, wind farm management information system, and inspection system to break the information island. Use innovative search-based analysis to quickly build a large screen for thematic data analysis and visualization. Use the algorithm integration platform of DataFocus, combined with SCADA data to train models such as fault warning and power plant power prediction.


Smart wind farm operation is an organic whole, which requires the intelligence of wind turbines, the informatization of wind farm management, and the intelligence of operation and maintenance. No one store can provide you with the best turnkey solution that includes both hardware and software. More often, it is necessary to carry out intelligent transformation of existing wind farms. As a highly flexible professional software product focusing on data analysis and algorithm integration, DataFocus can be easily integrated with various operation and maintenance software platforms.


Taking Company D as an example, it operates multiple wind farms with a total power generation of nearly 200 MW, and the huge investment in wind farm construction has been completed. It's all pure profit. By introducing the DataFocus data analysis platform, it breaks the information islands, collects the full amount of operating data of the wind farm in a unified way, and conducts business scenario planning. Specifically, Company D used DataFocus to solve the following three key problems:

1. Optimal state calibration of wind turbines, in a specific external environment and other states, to find the relevant variables and the range of change for the optimal state of the machine, for example: the external environment such as temperature and humidity, air pressure, time, divided into different stages according to a certain range, the optimal state of the machine such as the maximum power generation, etc. Turn the angle, etc., to find which of these variables is most relevant and the range of fluctuations.

2. By mining and modeling the big data generated by the SCADA system, we can predict and diagnose serious equipment problems such as fan blade icing and toothed belt failure, so that the stressful maintenance in the past can be avoided. It can effectively improve the utilization rate and operation and maintenance cost of wind power equipment.

In general, deep digitization and intelligence have greatly improved the efficiency of operation and maintenance, and Company D's wind farm has been significantly improved in four aspects.

1. Optimize the performance of wind turbines through machine learning and environmental self-adaptation, increase wind farm power generation by 5% on average, and increase revenue;

2. Improve operation and maintenance efficiency through intelligent algorithms , MTBF increased by 12%, and the cost was greatly reduced;

3. The grid friendliness was increased through collaborative control, and the power grid assessment was reduced by about 50%;

4. The operation management level was improved through digitalization, saving more than $1 million in labor costs every year.

Finally realize the background monitoring and full life cycle management of remote wind turbines to ensure the optimal operation status of wind turbines.

Recommended for you
Personal version for students, teachers or researchers
Intelligent search analysis
Chart Visualization
Import Excel and other local data
Open source datasets
Free Trial >>