Multi-redundancy safety control method and system for main control system of wind generating set

Abstract

The invention relates to the technical field of automatic control, and discloses a multi-redundancy safety control method and system for a main control system of a wind generating set, and the method comprises the steps: collecting the operation data of the wind generating set; predicting a potential equipment fault according to the operation data; and continuously detecting the communication link, and triggering edge control when the communication link fails. According to the method, high-precision monitoring of key parameters of the wind generating set is achieved through data collection and consistency verification of the multiple redundant sensors, even if individual sensors fail or communication is interrupted, the system can still obtain reliable data through the remaining sensors, and stable operation of equipment is ensured. Fault prediction is carried out in combination with the random forest model, the system can accurately judge the type and severity of the fault according to comparison of real-time data and historical data, and further damage to equipment is avoided. The edge control mechanism enables the system to still operate autonomously and trigger emergency response when a communication link fails, and the safety and reliability of the wind generating set are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of automatic control technology, and in particular to a multiple redundant safety control method and system for a main control system of a wind turbine generator set. Background Art

[0002] As a sustainable energy solution, wind power generation technology has been widely used and developed rapidly in recent years. Modern wind turbines mainly rely on the dynamic changes of wind resources, converting mechanical energy into electrical energy to provide stable clean energy for the power grid. With the vigorous development of the wind power industry, the automation control technology of wind turbines has gradually matured, especially in remote monitoring, real-time data acquisition and intelligent control. The main control system of the wind turbine can monitor and optimize the operating status based on sensor data. However, with the continuous expansion of the scale of wind farms and the increasingly complex operating environment of wind turbines, the demand for redundant safety control of the main control system has become increasingly important. Redundant control technology is usually used to ensure that the system can still operate stably in the event of communication interruption, sensor failure or environmental changes, which provides an important guarantee for improving the safety and reliability of wind turbines. Summary of the invention

[0003] In view of the above-mentioned problems, the present invention is proposed.

[0004] Therefore, the technical problem solved by the present invention is that the prior art lacks an effective localized emergency response mechanism.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: a multiple redundant safety control method for a wind turbine main control system, comprising: collecting operating data of the wind turbine; predicting potential equipment failures based on the operating data; continuously detecting communication links, and triggering edge control when the communication link fails.

[0006] As a preferred solution of the multiple redundant safety control method of the wind turbine main control system described in the present invention, the operating data includes but is not limited to wind speed, wind direction, blade speed, generator temperature, voltage, current, bearing vibration and blade stress.

[0007] As a preferred scheme of the multiple redundant safety control method of the wind turbine main control system described in the present invention, wherein: the prediction of potential equipment failures based on operating data includes: building a prediction model based on a random forest model; using a data analysis tool to extract the features of each data and combining them as input variables of the prediction model; inputting the input variables into the trained random forest model to obtain a first result.

[0008] As a preferred scheme of the multiple redundant safety control method of the main control system of a wind turbine generator set described in the present invention, wherein: after obtaining the first result, it also includes: evaluating according to the first result; presetting the first interval, the second interval and the third interval to compare with the first result to determine the triggering of the action; the action includes no action, the first action and the second action; the trigger includes, when the first result belongs to the first interval, it indicates that the equipment has no fault and no action is executed; when the first result belongs to the second interval, it indicates that the equipment has a minor fault, the first action is executed, the system issues an alarm and reduces the equipment load; when the first result belongs to the third interval, it indicates that the equipment has a serious fault, the second action is executed, and the system immediately performs a shutdown operation.

[0009] As a preferred solution of the multiple redundant safety control method of the wind turbine main control system described in the present invention, wherein: the continuous detection of the communication link includes, the main control system sends a PING request to the remote control center every time period, and records the response time, packet loss rate and signal strength of the PING; when the response time, packet loss rate and signal strength of the link are all within the normal range, the link is judged to be normal; when the link failure is triggered, switch to the redundant link, the link failure includes, the PING response time exceeds the maximum delay time, and the duration exceeds the preset threshold; PING has no response for n consecutive times; the signal strength is continuously lower than the set signal strength threshold.

[0010] As a preferred scheme of the multiple redundant safety control method of the main control system of the wind turbine generator set described in the present invention, wherein: triggering edge control when the communication link fails includes, when both the main communication link and the redundant communication link fail, triggering edge control; the edge control includes, after the edge control is triggered, adjusting the blade speed to the minimum limit to maintain normal operation; comparing the real-time data with the preset safety threshold, when the real-time data is within the safety threshold, performing no action; when the real-time data exceeds the safety threshold, triggering a recovery strategy, if the real-time data is within the safety threshold after the recovery strategy is triggered, continue to run, if the real-time data still exceeds the safety threshold, trigger an emergency shutdown; the recovery strategy includes but is not limited to reducing the output power of the generator and gradually slowing down the speed of the wind turbine blades.

[0011] As a preferred scheme of the multiple redundant safety control method of the wind turbine main control system described in the present invention, after the edge control is triggered, it also includes, in the edge control mode, continuously recording key data during the link failure, the key data including but not limited to the real-time data of the sensor, the result of the threshold comparison, the time point and specific conditions for triggering the emergency response; when the main communication link returns to normal, the key data is synchronized to the main control system.

[0012] A multi-redundant safety control system for a main control system of a wind turbine generator set adopting any of the methods described in the present invention, wherein: an acquisition module acquires operating data of the wind turbine generator set and performs preprocessing, and synchronously transmits the operating data to a main control module and an edge module; the main control module uses a prediction model to perform fault prediction based on the collected operating data, and determines the action to be performed based on the prediction result; and the edge module continuously detects the communication link with the main control module, and switches to edge control when the communication link fails.

[0013] A computer device comprises: a memory and a processor; the memory stores a computer program, comprising: the steps of implementing any one of the methods of the present invention when the processor executes the computer program.

[0014] A computer-readable storage medium stores a computer program, comprising: when the computer program is executed by a processor, the steps of implementing any one of the methods of the present invention are implemented.

[0015] Beneficial effects of the present invention: The method of the present invention realizes high-precision monitoring of key parameters of wind turbines through data collection and consistency verification of multiple redundant sensors. Even if individual sensors fail or communication is interrupted, the system can still obtain reliable data through the remaining sensors to ensure the stable operation of the equipment. Combined with the random forest model for fault prediction, the system can accurately determine the type and severity of the fault based on the comparison of real-time data and historical data to avoid further damage to the equipment. The edge control mechanism enables the system to operate autonomously and trigger emergency response when the communication link fails, effectively improving the safety and reliability of wind turbines. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0017] Figure 1 An overall flow chart of a multiple redundant safety control method for a wind turbine generator system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0019] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a multiple redundant safety control method for a main control system of a wind turbine generator set, comprising:

[0020] S1: Collect the operating data of wind turbines.

[0021] Furthermore, the system continuously collects key parameters of wind turbines through multiple redundant sensors, including but not limited to generator temperature, blade vibration, wind speed, rotation speed, voltage, current, etc. All data are transmitted to the main control system and edge devices in real time, and saved in the local storage unit for subsequent analysis.

[0022] Specifically, the setting of redundant sensors ensures that even if a sensor fails, the system can still obtain valid data. For example, at least three wind speed sensors are configured, located at the top, root and middle of the wind turbine tower, to obtain wind speed data at different heights and positions. Generator temperature sensors are installed at multiple key locations of the generator (such as the stator, rotor and bearing parts), and multiple sensors can be installed at each key location to ensure that the temperature of different components inside the generator can be fully captured.

[0023] For the data of redundant sensors, firstly, the consistency check is performed on the redundant sensor data of the same type, and the data uploaded by different sensors at the same location are compared. If the output data of a certain sensor deviates from the data of other sensors at the same location by more than 5% (or other set thresholds), the data is judged as invalid data and ignored. The data of other redundant sensors are continued to be used for analysis and control. If a sensor provides invalid data for multiple consecutive times, the sensor is judged to be faulty and a fault report is generated and sent to the staff.

[0024] The sensor data after consistency verification is processed by median filtering to comprehensively process the outputs of multiple sensors to obtain a more accurate and reliable final data.

[0025] S2: Predict potential equipment failures based on operating data.

[0026] Furthermore, after data preprocessing is completed, data analysis tools (such as Pandas or NumPy in Python) are used to extract the features of each data item and combine them as input variables of the model.

[0027] Specifically, for wind speed data, statistical features such as the calculated mean, variance, and extreme value are used to determine the impact of wind speed changes on blade speed; for blade speed, instantaneous speed, maximum speed, and speed fluctuation rate are extracted.

[0028] The prediction model is built based on the random forest model and trained with labeled historical data. The random forest model consists of multiple decision trees, each of which is trained with a random subset of the data. The final classification result is determined by majority voting.

[0029] For each decision tree, Bootstrap sampling is used to randomly select samples for training, and random splits are performed in the feature space of the samples. Assuming that the random forest contains T trees, the training data set for each tree is:

[0030] D t ={(X t ,Y t )},t=1,2,…,T

[0031] Among them, X t represents a subset randomly selected from the total sample; Y t Represents the corresponding output label.

[0032] The output of each tree is a classification result h t (X), when all trees are trained, the random forest will combine the classification results of all decision trees and use the majority voting method to get the final classification result:

[0033]

[0034] Among them, c represents all possible classification categories.

[0035] Set the number of decision trees in the random forest, T, and the maximum depth of each tree, d (usually 10-15). Use the Gini coefficient as the splitting criterion for the decision tree, expressed as:

[0036]

[0037] Among them, p k represents the probability of category k.

[0038] The real-time wind turbine operation data is used as the input variable. After feature extraction, it is input into the trained random forest model. The model uses the prediction results of each decision tree to comprehensively obtain the fault classification result of the current equipment as the first result.

[0039] Furthermore, an evaluation is performed based on the first result, and a preset first interval, a second interval, and a third interval are compared with the first result to determine the triggering of the action.

[0040] Specifically, the first interval, the second interval, and the third interval represent the severity of equipment failures. Equipment failures include but are not limited to mechanical failures, temperature failures, electrical failures, and extreme weather. In addition to the failure type, the label output by the prediction model also determines the severity of the failure by comparing real-time data and historical data. Taking temperature failures as an example, the specific steps are as follows:

[0041] During the training process, the system uses historical data to establish a statistical model for the characteristic values ​​of different types of faults (such as wind speed, temperature, vibration, speed, etc.), and divides the fault severity intervals: temperatures below 60°C are normal operation (the first interval), 60°C to 80°C are slight overheating (the second interval), and above 80°C are severe overheating (the third interval).

[0042] When constructing a decision tree, each node not only determines whether a certain sensor data triggers a fault type, but also further determines which interval the data is in through a threshold. For example: the root node detects the temperature sensor data; the first-layer node determines whether TTT is greater than 80°C. If so, it outputs "severe overheating" (third interval); the second-layer node determines whether TTT is between 60°C and 80°C. If so, it outputs "slight overheating" (second interval); otherwise, it outputs "normal" (first interval).

[0043] Each decision tree outputs different fault types and severities based on the input feature data. Through majority voting, the results of all decision trees are finally combined to determine whether the current device is in the first interval, the second interval, or the third interval.

[0044] Furthermore, the action includes no action, a first action and a second action; the trigger includes, when the first result belongs to the first interval, it indicates that the equipment has no fault, and no action is performed; when the first result belongs to the second interval, it indicates that the equipment has a minor fault, and the first action is performed, the system issues an alarm and reduces the equipment load; when the first result belongs to the third interval, it indicates that the equipment has a serious fault, and the second action is performed, and the system immediately performs a shutdown operation.

[0045] It should be noted that the random forest model may not be able to simply classify the precise severity of the fault through a decision tree in some extreme cases (such as extreme weather or sensor noise). In order to improve the adaptability of the model, during the operation of the equipment, the output changes of various sensors are monitored in real time, and the division thresholds of different intervals are dynamically adjusted in combination with model judgment, or when extreme weather is detected, the staff is notified through early warning to conduct key monitoring. For example, in extreme environments, the temperature fluctuation may be greater. The system can adjust the classification of temperature-related faults through a regression model of historical trends, making the model more adaptable when responding to changes.

[0046] S3: Continuously detect the communication link and trigger edge control when the communication link fails.

[0047] Furthermore, the continuous detection of communication links includes the main control system using a system timer to trigger detection, setting multiple links, sending a PING request to the remote control center at fixed time intervals (such as 5 seconds), and the system records the PING response time, packet loss rate and signal strength of each link respectively.

[0048] Analyze based on the response of the remote control center: when the PING response time exceeds the set maximum delay threshold (such as 1 second), it is recorded as an abnormality; if there is no response to three consecutive PING requests (that is, the packet loss rate is 100%), it is determined that the link has failed; when the signal strength is lower than the set threshold (such as -80dBm), the link is unstable.

[0049] If the response time and packet loss rate of the link are within the normal range (for example, the response time is less than 1 second and the packet loss rate is less than 5%), the system considers the link status to be normal and continues to use the link.

[0050] When any of the following conditions is detected, the system considers the primary link to be unstable or failed and prepares to switch to the redundant link:

[0051] 1. PING or heartbeat packets are sent more than the preset number of times (such as 3 times) without any response.

[0052] 2. Continuous high latency is recorded, exceeding the threshold (such as response time greater than 1 second for more than 30 seconds).

[0053] 3. The signal strength continues to be lower than the set minimum threshold (such as -80dBm for more than 30 seconds).

[0054] When the system determines that the main communication link is unstable or failed, the system checks the status of the backup link. If the backup link is available, the system cuts off the main communication link and starts data transmission on the backup link. After switching, the system continues to monitor the new link status through PING. The backup links include but are not limited to wireless links, wired links and satellite links.

[0055] When the primary communication link returns to normal (response time, packet loss rate, and signal strength return to normal range), the system can choose to restore to the primary communication link or continue to operate on the redundant link.

[0056] Furthermore, when the primary communication link and the redundant communication link detect faults at the same time and cannot be recovered (such as delay exceeding the threshold, packet loss rate is too high, or there is no response at all), the system automatically triggers edge control, and the edge computing device will take over the control and monitoring tasks of the entire wind turbine.

[0057] The edge control includes that after the edge control is triggered, the system first adjusts the blade speed to the minimum limit to maintain normal operation to ensure that the equipment operates within a safe range and reduce energy consumption and equipment wear. The minimum speed is set based on historical data and the minimum safe power output requirement of the equipment.

[0058] The edge computing device compares the real-time collected data (such as blade speed, wind speed, generator temperature, vibration frequency, etc.) with the preset safety threshold. The preset threshold is based on the safety range of the equipment under normal operating conditions, such as: generator temperature: 70℃ (normal range), 90℃ (exceeding the limit, triggering an emergency shutdown).

[0059] When the real-time data is within the safety threshold, no action is taken; when the real-time data exceeds the safety threshold, the recovery strategy is triggered. If the real-time data is within the safety threshold after the recovery strategy is triggered, the system continues to run. If the real-time data still exceeds the safety threshold, an emergency shutdown is triggered.

[0060] The recovery strategy includes, but is not limited to, reducing the output power of the generator and gradually slowing down the rotation speed of the wind turbine blades.

[0061] The emergency shutdown includes, but is not limited to, stopping the operation of the generator to ensure that the mechanical parts are no longer subject to additional loads; stopping the rotation of the blades to avoid further damage to the blades or bearings; and cutting off the power supply to protect the electrical system from overload or overheating.

[0062] In edge control mode, key data during link failure is continuously recorded, including but not limited to real-time data of sensors, results of threshold comparison, time point and specific conditions for triggering emergency response.

[0063] The system generates a detailed fault log that records all emergency response processes, including each power adjustment, speed control, and emergency shutdown trigger. When the primary or redundant communication link is restored, the edge device automatically synchronizes the recorded fault log, real-time data, and device status to the main control system.

[0064] The present embodiment also provides a multi-redundant safety control system for a wind turbine main control system, including: an acquisition module, which collects and pre-processes the operating data of the wind turbine, and synchronously transmits the operating data to the main control module and the edge module; the main control module, which uses a prediction model to predict faults based on the collected operating data, and decides on the action to be performed based on the prediction results; and the edge module, which continuously detects the communication link with the main control module, and switches to edge control when the communication link fails.

[0065] Example 2: The following is an embodiment of the present invention, which provides a multiple redundant safety control method for a main control system of a wind turbine generator set. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0066] In order to verify the effectiveness of the multiple redundant safety control method of the main control system of the wind turbine generator set, a simulation experiment based on actual working conditions was carried out. The experimental object is a wind turbine generator set, whose sensors include wind speed, blade vibration, generator temperature, voltage, current and speed. In order to ensure the reliability and accuracy of data acquisition, the system is equipped with redundant sensors, which are arranged at key positions such as the top of the tower, the root and the middle of the blade. At least 3 sensors of each type are configured to capture data at different positions.

[0067] During the experiment, the operating data under different working conditions were first collected from the redundant sensors of the wind turbine generator set. These data were sent to the main control system and edge devices through real-time transmission and stored in the local storage unit. In order to simulate equipment failure and communication loss, the generator temperature was manually modified, causing the sensor to display an overly high temperature value. At the same time, in order to simulate the communication loss, the generator temperature data was manually deleted from the data of sensor 3.

[0068] Through the design of redundant sensors, the system can still obtain valid data from other sensors and detect that the temperature of the generator is too high (exceeding the safety threshold of 70°C). In the event of communication loss, the system automatically ignores the failed sensor and performs fault diagnosis through data from other available sensors. The obtained data is shown in Table 1.

[0069] Table 1 Experimental data table

[0070]

[0071]

[0072] Among them, 1-4 represent the four redundant sensors corresponding to each parameter. From the experimental data, it can be observed that the temperature sensors 1, 2, and 4 of the generator temperature show temperatures exceeding the safety threshold (70°C). The system verifies the average temperature through consistency check and finds that it is 86°C, which has exceeded the threshold and is judged as a "serious fault". At the same time, the generator temperature data of sensor 3 is missing, simulating the situation where the communication link is lost. Based on the remaining redundant sensor data, the system can still correctly judge the operating status of the equipment and make corresponding decisions.

[0073] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0074] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0075] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0076] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0077] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multiple redundant safety control method for a main control system of a wind turbine generator set, characterized in that: include: Collecting the operating data of wind turbines; Predict potential equipment failures based on operational data; Continuously detect the communication link and trigger edge control when the communication link fails.

2. The multiple redundant safety control method of the wind turbine generator master control system according to claim 1, characterized in that: The operating data includes, but is not limited to, wind speed, wind direction, blade speed, generator temperature, voltage, current, bearing vibration and blade stress.

3. The multiple redundant safety control method of the wind turbine generator master control system according to claim 2, characterized in that: The predicting of potential equipment failures based on the operating data includes building a prediction model based on a random forest model; Use data analysis tools to extract the characteristics of each data and combine them as input variables for the prediction model; Feed the input variables into the trained random forest model and obtain the first result.

4. The multiple redundant safety control method of the wind turbine generator master control system according to claim 3, characterized in that: After obtaining the first result, the method further includes: Performing an evaluation based on the first result; The preset first interval, second interval and third interval are compared with the first result to determine the triggering of the action; The actions include no action, a first action, and a second action; The trigger includes: When the first result belongs to the first interval, it means that the device has no fault and no action is performed; When the first result belongs to the second interval, it indicates that the equipment has a minor fault, and the first action is executed, the system issues an alarm and reduces the equipment load; When the first result belongs to the third interval, it indicates that the equipment has a serious fault, and the second action is executed, and the system immediately performs a shutdown operation.

5. The multiple redundant safety control method of the wind turbine generator master control system according to claim 4, characterized in that: The continuous detection of the communication link includes the main control system sending a PING request to the remote control center every time period, and recording the response time, packet loss rate and signal strength of the PING; When the response time, packet loss rate, and signal strength of the link are within the normal range, the link is considered normal; When a link failure is triggered, switch to a redundant link, wherein the link failure includes: The PING response time exceeds the maximum delay time and the duration exceeds the preset threshold; PING has no response for n consecutive times; The signal strength continues to be lower than the set signal strength threshold.

6. The multiple redundant safety control method of the wind turbine generator master control system according to claim 5, characterized in that: Triggering edge control when the communication link fails includes triggering edge control when both the primary communication link and the redundant communication link fail; The edge control includes, after the edge control is triggered, adjusting the blade rotation speed to the minimum limit to maintain normal operation; Compare the real-time data with the preset safety threshold. When the real-time data is within the safety threshold, no action is taken. When the real-time data exceeds the safety threshold, the recovery strategy is triggered. If the real-time data is within the safety threshold after the recovery strategy is triggered, the system will continue to run. If the real-time data still exceeds the safety threshold, an emergency shutdown will be triggered. The recovery strategy includes, but is not limited to, reducing the output power of the generator and gradually slowing down the rotation speed of the wind turbine blades.

7. The multiple redundant safety control method of the wind turbine generator master control system according to claim 6, characterized in that: After the trigger edge control, it also includes: In edge control mode, key data during link failure is continuously recorded, including but not limited to real-time data of sensors, results of threshold comparison, time point and specific conditions for triggering emergency response; When the main communication link returns to normal, the key data will be synchronized to the main control system.

8. A multi-redundant safety control system using a wind turbine generator master control system according to any one of the methods of claims 1 to 7, characterized in that: include, The acquisition module collects and pre-processes the operating data of the wind turbine generator set, and synchronously transmits the operating data to the main control module and the edge module; The main control module uses the prediction model to predict faults based on the collected operating data and decides on the actions to be performed based on the prediction results; The edge module continuously detects the communication link with the main control module and switches to edge control when the communication link fails.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that when the processor executes the computer program, the steps of the multiple redundant safety control method of the wind turbine generator set main control system as described in any one of claims 1-7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multiple redundant safety control method for a wind turbine generator set main control system as described in any one of claims 1 to 7 are implemented.

Images