CN-121976916-A - Decision data analysis method and system based on intelligent perception

CN121976916ACN 121976916 ACN121976916 ACN 121976916ACN-121976916-A

Abstract

The invention discloses a decision data analysis method and a decision data analysis system based on intelligent perception, and relates to the technical field of decision data analysis; the method comprises the steps of obtaining running dynamic data of a plurality of groups of wind turbines, synchronously collecting environmental impact data of a scene where the wind turbines are located, fusing the running dynamic data of the wind turbines to obtain a running situation map of the wind turbines, carrying out state prediction in the running situation map of the wind turbines, positioning a plurality of abnormal risk wind turbines, matching a plurality of operation and maintenance treatment instructions, and sending the operation and maintenance treatment instructions to corresponding wind farm management and control terminals. The method solves the technical problems of low fault early warning accuracy and low operation and maintenance treatment efficiency caused by the lack of full-flow decision data analysis support of the new energy station wind turbine generator in the prior art, achieves the full-flow intelligent decision analysis of the new energy station wind turbine generator, and improves the technical effects of the fault early warning accuracy and the operation and maintenance treatment efficiency of the wind turbine generator.

Inventors

Ai Xianqiang
GAO XIANG
CUI YANSHEN
GAO WEI
SHUAI CHAO
ZHOU SHIJIE
HONG YUKAI
LIU XIAOTONG
JIN AN
ZHANG JIN
ZENG ZHIPING
LIU ZIDA
LI PEI
WANG HUAN
GAO CHUNYU
SUN XINNING

Assignees

江西大唐国际新能源有限公司
大唐可再生能源试验研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20251125

Claims (8)

1. The decision data analysis method based on intelligent perception is characterized by comprising the following steps: the key components of the wind turbine generator system of the new energy station are used as cores, and an integrated brain-like sense sensing device array is deployed; The brain-like sense of calculation integrated sensing equipment array acquires communication interaction with equipment through a sensor to obtain a plurality of groups of wind turbine running dynamic data, wherein the wind turbine running dynamic data comprise voiceprint signals, vibration data, video images, tension data and equipment running parameters; the method comprises the steps of synchronously collecting environmental impact data of a scene where a wind turbine is located through an environment sensing unit, wherein the environmental impact data comprise temperature, humidity, wind speed and air pressure parameters; Fusing the operation dynamic data of the plurality of groups of wind turbine generators to obtain an operation situation map of the wind turbine generators; carrying out state prediction in the running situation map of the wind turbine by combining the environmental impact data, and positioning a plurality of abnormal risk wind turbines; Matching a plurality of operation and maintenance treatment instructions according to a plurality of operation risk characteristics of the plurality of abnormal risk wind turbines; And sending the operation and maintenance treatment instructions to corresponding wind farm management and control terminals through an industrial Ethernet communication link.
2. The decision data analysis method based on intelligent perception according to claim 1, wherein the key components of the wind turbine generator system of the new energy station are used as cores, and an integrated brain-like sense sensing device array is deployed, and the method comprises: the method comprises the steps of interactively obtaining model parameters and a core component list of the wind turbine generator, and determining a reference sensing device type according to the model parameters and the core component list; Carrying out load intensity quantification according to the historical operation load data of the wind turbine generator, and matching and sensing coverage area according to the load intensity; Positioning a sensing monitoring area according to the sensing coverage area by taking the installation position of the key component of the wind turbine generator as the center; and deploying the brain-like integrated sensing equipment array in the sensing monitoring area according to the reference sensing equipment type and the sensing coverage range.
3. The intelligent perception based decision data analysis method of claim 2, wherein the brain-like integrated sensory device array is deployed in the sensory monitoring area according to the reference sensory device type and sensory coverage, the method comprising: dividing regional sensing requirements according to peripheral layout features and terrain conditions of the wind turbine generator to obtain a plurality of regional sensing equipment configurations; the historical fault data of the wind turbine generator are obtained interactively, and fault high-speed component information is extracted; Supplementing the type of the sensing equipment according to the fault high-transmitting part information to obtain an updated type of the sensing equipment; And deploying the brain-like integrated sensing device array in the sensing monitoring area based on the updated sensing device type and the multiple regional sensing device configurations.
4. The intelligent perception-based decision data analysis method according to claim 2, wherein the running dynamic data of the plurality of groups of wind turbines are fused to obtain a running situation map of the wind turbines, and the method comprises: Aggregating the plurality of groups of wind turbine generator running dynamic data based on the wind turbine generator component ID to obtain Q time sequence running data of Q core components; after space-time reference unification is carried out on the Q time sequence operation data through a time synchronization protocol, fitting and outputting Q component operation situations; And aligning the Q component operation situations in the sensing monitoring area through a coordinate mapping space, and outputting the wind turbine generator operation situation map.
5. The intelligent perception based decision data analysis method of claim 4, wherein, in combination with the environmental impact data, state prediction is performed in the wind turbine operational situation map, and a plurality of abnormal risk wind turbines are located, the method comprising: Constructing an equipment state prediction model based on a pulse convolution neural network, wherein the equipment state prediction model is embedded into the mechanism knowledge of the wind turbine; inputting the environmental impact data and Q component operation situations into the equipment state prediction model to obtain Q extended operation situations; connecting the Q component operation situations and the Q extension operation situations in the wind turbine generator operation situation map, and positioning a plurality of abnormal risk nodes; and extracting wind turbine generator identifiers from the abnormal risk nodes to obtain the abnormal risk wind turbine generators.
6. The intelligent awareness based decision data analysis method of claim 5 wherein a plurality of operation and maintenance treatment instructions are matched according to a plurality of operation risk features of the plurality of abnormal risk wind turbines, the method comprising: Performing running situation backtracking on the plurality of abnormal risk wind turbines along the plurality of abnormal risk nodes to obtain a plurality of instantaneous risk running situations; according to the fault characteristics of the instantaneous risk running situations, combining the power industry knowledge graph and the historical operation and maintenance cases, and matching and outputting a plurality of targeted operation and maintenance operations; And performing instruction encoding on the plurality of targeted operation and maintenance operations, and outputting the plurality of operation and maintenance treatment instructions.
7. The intelligent awareness based decision data analysis method of claim 6 wherein the plurality of operation and maintenance handling instructions are sent to corresponding wind farm management terminals over an industrial ethernet communication link, the method comprising: Quantizing the execution difficulty coefficients of the plurality of targeted operation and maintenance operations, wherein the execution difficulty coefficients are determined according to the number of operation steps and the technical requirement level; Calling fault deterioration prediction time lengths corresponding to the multiple instantaneous risk running situations; According to the weighted calculation results of the execution difficulty coefficient and the fault deterioration prediction duration, the operation and maintenance treatment instructions are subjected to priority ranking, and a plurality of operation and maintenance priorities are output; And according to the operation and maintenance priorities, the operation and maintenance treatment instructions are sent to corresponding wind farm management and control terminals or operation and maintenance execution units in a grading mode.
8. A smart perception based decision data analysis system for implementing the smart perception based decision data analysis method of any one of claims 1-7, the system comprising: the equipment array deployment module is used for deploying the brain-like sense-of-feel integrated sensing equipment array by taking key components of the wind turbine generator set of the new energy station as cores; The dynamic data acquisition module is used for acquiring and interacting the brain-like sense integrated sensing equipment array with equipment through a sensor to obtain a plurality of groups of wind turbine running dynamic data, wherein the wind turbine running dynamic data comprise voiceprint signals, vibration data, video images, tension data and equipment running parameters; The environment influence data acquisition module is used for synchronously acquiring environment influence data of a scene where the wind turbine generator is located through the environment sensing unit, wherein the environment influence data comprise temperature, humidity, wind speed and air pressure parameters; the running situation map acquisition module is used for fusing the running dynamic data of the plurality of groups of wind turbine generators to obtain a running situation map of the wind turbine generators; The risk wind turbine generator positioning module is used for carrying out state prediction in the wind turbine generator running situation map by combining the environmental impact data to position a plurality of abnormal risk wind turbine generators; The operation and maintenance treatment instruction matching module is used for matching a plurality of operation and maintenance treatment instructions according to a plurality of operation risk characteristics of the plurality of abnormal risk wind turbines; And the operation and maintenance treatment instruction sending module is used for sending the operation and maintenance treatment instructions to the corresponding wind farm management and control terminal through the industrial Ethernet communication link.

Description

Decision data analysis method and system based on intelligent perception Technical Field The invention relates to the technical field of decision data analysis, in particular to a decision data analysis method and system based on intelligent perception. Background In the field of operation and maintenance of wind turbines in new energy stations, the traditional operation and maintenance mode gradually exposes various limitations along with the expansion of the scale of the wind turbines and the improvement of the operation complexity of the wind turbines. In the prior art, the data are collected by relying on manual inspection or a single sensor, the operation state of key components of the wind turbine generator is difficult to be covered comprehensively, the operation dynamic data collection dimension is incomplete, the influence of environmental factors such as temperature, humidity and the like on the operation of the wind turbine generator is often ignored, and the data association analysis is insufficient. Meanwhile, due to the lack of decision data analysis support of a system, the operation data and the environment data are not effectively fused, and an accurate unit operation situation map is difficult to construct, so that equipment state prediction is delayed, and abnormal risk positioning is inaccurate. In addition, operation and maintenance treatment aiming at abnormal risks is based on experience judgment, a data-driven targeted instruction matching and priority ordering mechanism is lacked, operation and maintenance resource allocation is unreasonable, fault treatment efficiency is low, and requirements of efficient and safe operation of the wind turbine generator in the new energy station cannot be met. In the prior art, a new energy station wind turbine generator system lacks full-flow decision data analysis support, so that the technical problems of low fault early warning accuracy and low operation and maintenance treatment efficiency are caused. Disclosure of Invention The application provides a decision data analysis method and a decision data analysis system based on intelligent perception, which are used for solving the technical problems of low fault early warning accuracy and low operation and maintenance treatment efficiency caused by the lack of full-flow decision data analysis support of a new energy station wind turbine generator in the prior art. In view of the above problems, the present application provides a decision data analysis method and system based on intelligent perception. In a first aspect of the present application, there is provided a decision data analysis method based on intelligent perception, the method comprising: The method comprises the steps of taking key components of a wind turbine generator in a new energy station as a core, deploying a brain-like sense integrated sensing device array, collecting and interacting with devices through sensors to obtain multiple groups of wind turbine generator operation dynamic data, wherein the wind turbine generator operation dynamic data comprise voiceprint signals, vibration data, video images, tension data and device operation parameters, synchronously collecting environment influence data of a scene where a wind turbine generator is located through an environment sensing unit, wherein the environment influence data comprise temperature, humidity, wind speed and air pressure parameters, fusing the multiple groups of wind turbine generator operation dynamic data to obtain a wind turbine generator operation situation map, carrying out state prediction in the wind turbine operation situation map by combining the environment influence data to locate multiple abnormal risk wind turbines, matching multiple operation and maintenance treatment instructions according to multiple operation risk characteristics of the multiple abnormal risk wind turbines, and sending the multiple operation and maintenance treatment instructions to corresponding wind turbine generator management and control terminals through an industrial Ethernet communication link. In a second aspect of the present application, there is provided an intelligent perception based decision data analysis system, the system comprising: The system comprises a device array deployment module, a dynamic data acquisition module, an operation and maintenance instruction matching module, an operation and maintenance instruction sending module and an operation and maintenance instruction transmission module, wherein the device array deployment module is used for taking key components of a wind turbine generator system of a new energy station as a core, deploying a brain-like sense integrated sensing device array, the dynamic data acquisition module is used for acquiring and interacting with devices through sensors to obtain a plurality of groups of wind turbine generator system operation dynamic data, the wind turbine generator system operation dynamic data comprises voicep