Wind power plant alarm process approval method based on intelligent decision and related device

Abstract

The invention provides a wind power plant alarm process approval method based on intelligent decision and a related device, and relates to the technical field of wind power plant operation and maintenance management, and the method comprises the following steps: collecting and extracting multi-dimensional features in real time when wind power plant equipment triggers an alarm; based on a pre-trained classification model, performing type division and priority evaluation on the alarms according to the multi-dimensional features to obtain alarm types and priorities; and according to the alarm type and the priority, determining a notification object and an examination and approval path, and then dynamically allocating processing personnel to realize wind power plant alarm process examination and approval. According to the invention, the problems of low automation degree and unreasonable distribution of processing personnel in the existing wind power plant alarm processing flow can be solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind farm operation and maintenance management, in particular to a wind farm alarm process approval method based on intelligent decision and related devices. BACKGROUND

[0002] In the operation and maintenance of a wind farm, a device monitoring system is usually responsible for monitoring the status of various devices. Once an alarm is triggered by a device, manual approval processing needs to be performed according to a pre-set process. The existing alarm processing method mainly relies on a fixed manual approval process. This process is often set at the initial stage of the construction of a wind farm. As the scale of the wind farm expands and the types of devices increase, the fixed approval process is difficult to adapt to the specific circumstances of different device alarms, resulting in low alarm processing efficiency.

[0003] The existing wind farm alarm processing process has the following significant defects: low automation: alarm notification and personnel allocation mostly rely on manual operation or simple automation rules, which cannot dynamically adjust the processing strategy according to the specific characteristics of the alarm, resulting in prolonged response time. Unreasonable personnel allocation: the traditional approval process may send all alarms to all relevant personnel in a predetermined order, rather than selecting the most suitable processing personnel according to the nature and urgency of the alarm, which not only wastes technical resources but also increases the workload of professional personnel, reducing overall efficiency. Rigidity of approval path: the fixed approval path does not take into account the actual situation of the alarm, which may result in important alarms being delayed and minor alarms occupying excessive attention, affecting the safety of device operation and the normal operation of the wind farm.

[0004] In summary, the existing wind farm alarm processing process has problems such as low automation, unreasonable personnel allocation, and rigid approval path, and there is an urgent need for a more intelligent and efficient alarm processing method to improve the operation and maintenance efficiency and safety of the wind farm. SUMMARY

[0005] The present application aims to provide a wind farm alarm process approval method based on intelligent decision and related devices to solve the problems of low automation and unreasonable personnel allocation in the existing wind farm alarm processing process.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solution: In a first aspect, a wind farm alarm process approval method based on intelligent decision includes the following steps: Real-time acquisition and extraction of multi-dimensional features when a wind farm device triggers an alarm; Based on a pre-trained classification model, the alarm is classified by type and priority based on the multi-dimensional features, resulting in alarm type and priority; According to the alarm type and priority, a notification object and an approval path are determined, and a processing personnel is dynamically allocated, so as to realize the wind farm alarm process approval.

[0007] In some embodiments, the method further comprises the following steps: After the wind farm alarm process approval is completed, the artificial confirmation information of the processing personnel and the wind farm equipment state information are obtained, and it is judged whether the alarm is ended, if the alarm is ended, the feedback data is collected and the knowledge base of the wind farm is updated.

[0008] In some embodiments, the multi-dimensional features include: equipment operation data, fault identification data, environmental parameters and equipment state data. The equipment operation data includes active power, fan speed, equipment load rate, current and voltage; the fault identification data includes fault code and fault occurrence time; the environmental parameters include bearing temperature, temperature rise rate, external wind speed, smoke concentration and humidity; and the equipment state data includes vibration amplitude, main frequency and equipment cumulative running time.

[0009] In some embodiments, the classification model adopts a gradient boosting tree model.

[0010] In some embodiments, according to the multi-dimensional features, the alarm is classified by type and evaluated by priority, and the steps of obtaining the alarm type and priority include: When the vibration amplitude > 100 and the main frequency is 2 kHz, the alarm type is mechanical failure; when the voltage fluctuation > ± 15%, the alarm type is electrical failure; and when the external wind speed > 25 m / s, the alarm type is environmental anomaly. If any one of the following conditions is met: the temperature > 80℃ and the smoke concentration > 0.5 mg / m 3 , the fan speed = 0 and the active power = 0, or the external wind speed > 30 m / s, the priority of the alarm is serious alarm; if the active power is 10%-20% lower than the rated value, or the vibration amplitude is 50 μm ~ 100 μm, the priority of the alarm is important alarm; and in other cases, the priority of the alarm is general alarm.

[0011] In some embodiments, according to the alarm type and priority, the notification object and the approval path are determined, and the processing personnel is dynamically allocated, and the steps include: The straight-line distances between all processing personnel and the fault equipment are obtained, and the processing personnel with the straight-line distance < 5 km are screened out. Count the current work order quantity of the processor within the straight-line distance <5km, and select the processor with the current work order quantity <=2 from the processor within the straight-line distance <5km, to obtain the processor within the straight-line distance <5km and the current work order quantity <=2; Calculate the processing success rate of the processor within the straight-line distance <5km and the current work order quantity <=2 for the alarm type; Select the processor within the straight-line distance <5km and the current work order quantity <=2 with the highest processing success rate as the optimal processor. In the above steps, the serious alarm and the important alarm are simultaneously notified to the wind power station person in charge and the operation and maintenance on-duty personnel, and the general alarm is notified to the operation and maintenance on-duty personnel.

[0012] In a second aspect, a wind farm alarm process approval based on intelligent decision-making includes: A multi-dimensional feature extraction module is configured to collect and extract multi-dimensional features in real time when a wind farm device triggers an alarm; An alarm type classification and evaluation module is configured to classify and evaluate the alarm type and priority based on a pre-trained classification model according to the multi-dimensional features. A dynamic approval and distribution module is configured to dynamically distribute processors after determining the notification object and approval path according to the alarm type and priority, to realize the wind farm alarm process approval.

[0013] In a third aspect, an electronic device includes a memory, a processor, and a computer program stored in the memory and executable in the processor, and the processor executes the computer program to realize the steps of the wind farm alarm process approval method based on intelligent decision-making.

[0014] In a fourth aspect, a computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the steps of the wind farm alarm process approval method based on intelligent decision-making according to any one of claims 1-6.

[0015] In a fifth aspect, a computer program product includes a computer program, and the computer program is executed by a processor to realize the steps of the wind farm alarm process approval method based on intelligent decision-making.

[0016] Compared with the prior art, the present application has the following beneficial effects: The application provides a wind farm alarm process approval method based on intelligent decision-making, which is based on a pre-trained classification model, and according to the multi-dimensional features, the alarm is classified by type and evaluated by priority to obtain the alarm type and priority; according to the alarm type and priority, the notification object and approval path are determined, and then the processing personnel are dynamically allocated to realize the wind farm alarm process approval. By using the real-time collected multi-dimensional features and the pre-trained classification model, the automatic alarm type classification and priority evaluation are realized, which helps to quickly respond and accurately handle various emergency situations, ensures the stable operation of the wind farm, and can significantly improve the efficiency of the wind farm in processing device alarms, reduces unnecessary waiting and manual judgment time.

[0017] Further, when dynamically allocating processing personnel, the application comprehensively considers the straight-line distance, the current number of work orders and the alarm processing success rate, which can optimize resource allocation, improve early warning response speed and service quality, which not only improves the maintenance efficiency, but also reduces the energy loss and economic cost caused by equipment failure. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a flowchart of a wind farm alarm process approval method based on intelligent decision-making provided by an embodiment of the application; Figure 2 is a structural diagram of a wind farm alarm process approval system based on intelligent decision-making provided by an embodiment of the application. DETAILED DESCRIPTION

[0019] In order for those skilled in the art to better understand the application scheme, the technical scheme of the application will be further described in detail below with reference to the drawings, and the content is an explanation of the application and not a limitation.

[0020] It should be noted that the terms "include" and "have" and any variations thereof in the specification and claims of the application are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device containing a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, systems, products or devices.

[0021] As Figure 1As shown, the embodiment provides a wind farm alarm process approval method based on intelligent decision-making, including the steps of: real-time acquisition and extraction of multi-dimensional features when wind farm equipment triggers an alarm; based on a pre-trained classification model, type classification and priority assessment of the alarm according to the multi-dimensional features, obtaining the alarm type and priority; according to the alarm type and priority, determining the notification object and approval path, and then dynamically allocating the processing personnel, realizing the wind farm alarm process approval. This embodiment uses big data analysis and machine learning technology, which can quickly identify key features from massive equipment operation data, improving the accuracy and efficiency of alarm processing. By building a classification model to analyze multi-dimensional features, automatic identification of different alarm types and intelligent assessment of priority are achieved, which relies on the learning and pattern recognition capabilities of deep learning algorithms on historical data. The technology in this embodiment can significantly reduce manual intervention, speed up alarm response, effectively prevent wind farm equipment failure, and improve equipment operation stability. In other embodiments, multiple machine learning models such as neural networks and support vector machines can be integrated to solve the accuracy problem of alarm type identification in complex environments.

[0022] An aspect of the embodiment provides a further implementation, which further includes the steps of: after completing the wind farm alarm process approval, obtaining the manual confirmation information of the processing personnel and the wind farm equipment state information, and judging whether the alarm is ended, if the alarm is ended, collecting feedback data and updating the knowledge base of the wind farm. This embodiment introduces a closed-loop control mechanism to ensure the completeness and continuous optimization of alarm processing. In principle, by collecting the processing results and the latest state of the equipment, the effectiveness of the alarm processing is verified, and the knowledge base is improved to enhance the intelligent level of future alarm processing. The closed-loop management of alarm processing is realized, and the self-learning and adaptability of the system are enhanced. In other embodiments, more complex data models such as time series prediction models can be established to predict the future state of the equipment, to warn potential failures in advance, and to further improve the operation efficiency of the wind farm.

[0023] One aspect of this embodiment provides an implementation method in which multidimensional features include: equipment operation data, fault identification data, environmental parameters, and equipment status data; equipment operation data includes active power, fan speed, equipment load rate, current, and voltage; fault identification data includes fault codes and fault occurrence times; environmental parameters include bearing temperature, heating rate, external wind speed, smoke concentration, and humidity; and equipment status data includes vibration amplitude, main frequency, and cumulative equipment operating time. This embodiment comprehensively considers various factors affecting equipment operation and improves the comprehensiveness and accuracy of alarm identification through multidimensional data analysis. The model built using these features can more accurately capture changes in equipment operating status and promptly detect anomalies. This extraction and analysis of multidimensional features can effectively reduce false alarm rates and improve the targeting of alarm processing. In other embodiments, more sensors can be added to collect unstructured data such as sound signals and video surveillance, and deep learning technology can be used to mine fault patterns hidden in this data to further enrich alarm features.

[0024] One aspect of this embodiment provides an implementation method where the classification model employs a gradient boosting tree model. Gradient boosting trees are powerful machine learning models capable of handling high-dimensional features, exhibiting good generalization ability and prediction accuracy. By iteratively constructing a decision tree, each iteration fits the residual of the previous prediction, gradually improving the model's predictive ability. Using a gradient boosting tree model can accurately classify alarm types and evaluate priorities, reducing processing latency and improving fault response speed. In other embodiments, other advanced tree-based models such as random forests and XGBoost can also be used to solve classification problems in specific scenarios, improving the flexibility and robustness of alarm processing.

[0025] One aspect of this embodiment provides a method for classifying alarms into types and prioritizing them based on multidimensional features to obtain alarm types and priorities. Specifically, this includes: when the vibration amplitude is greater than 100... When the frequency is 2 kHz, the alarm type is mechanical failure; when the voltage fluctuation is greater than 15%, the alarm type is electrical failure; when the external wind speed is greater than 25 meters per second, the alarm type is environmental anomaly; if any one of the following conditions is met: the temperature is greater than 80 degrees Celsius and the smoke concentration is greater than 0.5 milligrams per cubic meter (fire alarm), the fan speed is equal to 0 and the active power is equal to 0 (equipment failure alarm), or the external wind speed is greater than 30 meters per second (extreme weather alarm), the priority of the alarm is serious alarm; if the active power is 10%-20% lower than the rated value (equipment performance degradation alarm but does not affect power generation), or the vibration amplitude is 50 μm ~100 μm, the priority of the alarm is important alarm; in other cases (including device state monitoring, environmental monitoring alarm), the priority of the alarm is general alarm. Through setting specific threshold conditions, the embodiment realizes rapid judgment of alarm types and reasonable division of priorities. In principle, based on the threshold of device operating characteristics and environmental parameters, the model can quickly identify the fault category and emergency degree, providing a basis for subsequent processing. In terms of effect, this accurate type division and priority evaluation helps to quickly mobilize resources, prioritize serious alarms, and avoid resource waste and processing delays. In other embodiments, by adjusting the threshold or using a dynamic threshold strategy, the operating characteristics of different wind farms can be adapted, improving the adaptability and accuracy of alarm processing.

[0026] An aspect of the embodiment provides a step of dynamically allocating a processing personnel according to an alarm type and a priority after determining a notification object and an approval path, and specifically comprises: acquiring a straight-line distance of all processing personnel from a fault device, and screening out processing personnel with a straight-line distance less than 5 kilometers; counting a current work order quantity of the processing personnel with a straight-line distance less than 5 kilometers, and preferentially selecting processing personnel with a current work order quantity less than or equal to 2 from the processing personnel with a straight-line distance less than 5 kilometers to obtain processing personnel with a straight-line distance less than 5 kilometers and a current work order quantity less than or equal to 2; calculating a processing success rate of the processing personnel with a straight-line distance less than 5 kilometers and a current work order quantity less than or equal to 2 on the alarm type; and selecting processing personnel with the highest processing success rate as optimal processing personnel. In the above steps, the serious alarm and the important alarm are simultaneously notified to a wind power station person in charge and an operation and maintenance on-duty personnel, and the general alarm is notified to the operation and maintenance on-duty personnel. The embodiment adopts a geographic information system (GIS) technology and a task scheduling algorithm to realize efficient matching and dynamic scheduling of the processing personnel. By calculating the distance of the processing personnel from the fault device and the work order burden, and combining the historical processing success rate, the task allocation strategy is dynamically adjusted to ensure the timeliness and effectiveness of alarm processing. The dynamic allocation mechanism based on the position and the work order load can significantly shorten the alarm response time and improve the fault processing efficiency. In other embodiments, an artificial intelligence assistant can be introduced to automatically plan an optimal route, reduce the moving time of the processing personnel, and further improve the response speed.

[0027] As shown in Figure 2 The embodiment also provides a wind power plant alarm process approval system based on intelligent decision-making, which comprises: A multi-dimensional feature extraction module is configured to collect and extract multi-dimensional features in real time when an alarm of a wind power plant device is triggered. An alarm type classification and evaluation module is configured to classify and evaluate the alarm type and priority based on a pre-trained classification model according to the multi-dimensional features. A dynamic approval and allocation module is configured to dynamically allocate a processing personnel according to the alarm type and priority after determining a notification object and an approval path to realize the wind power plant alarm process approval.

[0028] The division of the modules in the embodiment is illustrative, and is merely a logical function division. In actual implementation, another division mode can be used. In addition, the function modules in each embodiment of the present application can be integrated in one processor, or can be physically separated, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software function module.

[0029] The embodiment also provides a computer device, which comprises a processor and a memory for storing a computer program (the computer program in the embodiment comprises a computing component and an iteration component, and can perform model computing and model updating), the computer program comprises program instructions, and the processor is used for executing the program instructions stored in the computer storage medium. The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components and the like, which are the computing core and control core of the terminal, and are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to implement a corresponding method flow or a corresponding function; the processor in the embodiment can be used for the operation of the wind farm alarm process approval method based on intelligent decision.

[0030] The embodiment also provides a storage medium, specifically a computer readable storage medium (Memory), which is a memory device in the computer device, and is used for storing programs and data. It can be understood that the computer readable storage medium herein can include an internal storage medium in the computer device, and of course can also include an expansion storage medium supported by the computer device. The computer readable storage medium provides a storage space, and the storage space stores an operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that the computer readable storage medium herein can be a high-speed RAM memory, or a non-volatile memory such as at least one disk memory. One or more instructions stored in the computer readable storage medium can be loaded and executed by the processor to implement the corresponding steps of the wind farm alarm process approval method based on intelligent decision in the above embodiment.

[0031] The embodiment also provides a computer program product, which comprises a computer program, and when the computer program is executed by the processor, the corresponding steps of the wind farm alarm process approval method based on intelligent decision in the above embodiment are implemented.

[0032] Those skilled in the art will appreciate that embodiments of the application can be devised for a method, a system, or a computer program product. Accordingly, the present application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0033] The present application is described in reference to the flowchart and / or block diagrams of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0034] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0035] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0036] Finally, it should be noted that the above-mentioned embodiments are merely intended for describing the technical solutions of the present application, but not for limiting it. Although the present application is described in detail with reference to the above embodiments, those skilled in the field should understand that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A wind farm alarm process approval method based on intelligent decision-making, characterized in that, Includes the following steps: Real-time acquisition and extraction of multi-dimensional features when wind farm equipment triggers alarms; Based on the pre-trained classification model, the alarms are classified and prioritized according to the multi-dimensional features to obtain the alarm type and priority. Based on the alarm type and priority, the notification recipients and approval paths are determined, and the processing personnel are dynamically assigned to realize the alarm process approval of wind farms.

2. The wind farm alarm process approval method based on intelligent decision-making according to claim 1, characterized in that, It also includes the following steps: After completing the wind farm alarm process approval, obtain the manual confirmation information of the processing personnel and the status information of the wind farm equipment, and determine whether the alarm has ended. If the alarm has ended, collect feedback data and update the wind farm's knowledge base.

3. The wind farm alarm process approval method based on intelligent decision-making according to claim 1, characterized in that, The multidimensional features include: equipment operation data, fault identification data, environmental parameters, and equipment status data; The equipment operation data includes active power, fan speed, equipment load rate, current and voltage; the fault identification data includes fault code and fault occurrence time; the environmental parameters include bearing temperature, heating rate, outside wind speed, smoke concentration and humidity; the equipment status data includes vibration amplitude, main frequency and cumulative equipment running time.

4. The wind farm alarm process approval method based on intelligent decision-making according to claim 1, characterized in that, The classification model uses a gradient boosting tree model.

5. The wind farm alarm process approval method based on intelligent decision-making according to claim 3, characterized in that, The steps of classifying and prioritizing alarms based on the multidimensional features to obtain alarm types and priorities specifically include: When the vibration amplitude is >100 Furthermore, when the main frequency is 2kHz, the alarm type is mechanical fault; when the voltage fluctuation is >±15%, the alarm type is electrical fault; when the outside wind speed is >25m / s, the alarm type is environmental anomaly. If the temperature is >80℃ and the smoke concentration is >0.5mg / m³, then... 3 If any one of the following conditions is met: the fan speed is 0 and the active power is 0, or the outside wind speed is > 30m / s, then the alarm priority is a critical alarm; if the active power is 10%-20% lower than the rated value, or the vibration amplitude is 50μm~100μm, then the alarm priority is an important alarm; in other cases, the alarm priority is a general alarm.

6. The wind farm alarm process approval method based on intelligent decision-making according to claim 5, characterized in that, The steps of dynamically assigning processing personnel after determining the notification recipients and approval paths based on the alarm type and priority specifically include: Obtain the straight-line distance between all personnel handling the faulty equipment and filter out personnel whose straight-line distance is less than 5km; The number of current work orders for the processing personnel within a straight-line distance of <5km is counted, and among the processing personnel within a straight-line distance of <5km, those with a current work order count of ≤2 are selected first, thus obtaining the processing personnel with a straight-line distance of <5km and a current work order count of ≤2. Calculate the success rate of processing personnel for the alarm type when the straight-line distance is <5km and the current number of work orders is ≤2. Select the processing personnel with the highest success rate and a straight-line distance of less than 5km and a current number of work orders of ≤2 as the optimal processing personnel; In the above steps, the critical alarm and important alarm are simultaneously notified to the wind farm site manager and the operation and maintenance personnel, and the general alarm is notified to the operation and maintenance personnel.

7. A wind farm alarm process approval based on intelligent decision-making, characterized in that, include: The multi-dimensional feature extraction module is used to collect and extract multi-dimensional features when wind farm equipment triggers alarms in real time; The alarm type classification and evaluation module is used to classify and prioritize alarms based on the multidimensional features according to the pre-trained classification model, so as to obtain alarm type and priority. The dynamic approval and allocation module is used to dynamically allocate processing personnel after determining the notification object and approval path based on the alarm type and priority, thereby realizing the alarm process approval of wind farms.

8. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable in the processor. When the processor executes the computer program, it implements the steps of the wind farm alarm process approval method based on intelligent decision-making as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the wind farm alarm process approval method based on intelligent decision-making as described in any one of claims 1 to 6.

10. A computer program product, the computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the wind farm alarm process approval method based on intelligent decision-making as described in any one of claims 1 to 6.

Images