Software and hardware security and credibility mutual recognition measurement mechanism and method for master control system of wind turbine generator

Abstract

The invention provides a wind turbine generator master control system software and hardware security and credibility mutual recognition measurement mechanism and method, and relates to the field of wind turbine generator control, and the method comprises the following steps: carrying out the hierarchical division of a wind turbine generator master control system, and carrying out the behavior measurement of all hierarchies according to the division result; the method comprises the following steps of: performing hierarchical bidirectional authentication on each hierarchy of a main control system of a wind turbine generator by combining a data link layer encryption technology on the basis of a cross-hierarchy credible mutual recognition mechanism; based on a cross-equipment-level credible mutual recognition mechanism, carrying out equipment-level identity bidirectional authentication between a main control system and a peripheral system of the wind turbine generator; and the change conditions of the behavior measurement result, the hierarchical authentication result and the equipment-level authentication result are tracked in real time, the credible state of the main control system of the wind turbine generator is evaluated, and a response strategy is dynamically executed according to the evaluation result. According to the method, the security problems of hardware tampering, software malicious injection, cross-device communication hijacking and the like can be solved, and meanwhile, the industrial control requirements of real-time performance and low resource occupation are met.

Description

Technical Field

[0001] This invention relates to the field of wind turbine generator control, and more specifically, to a mechanism and method for measuring the security, reliability, and mutual recognition of software and hardware in the main control system of a wind turbine generator. Background Technology

[0002] Currently, trusted computing products mainly take the form of hardware and operating system layers, often consisting of a CPU-embedded TPCM root of trust or an external CPU TPCM root of trust plus a TSB trusted software base deployed at the operating system layer; the specific implementation steps are as follows:

[0003] 1. Construct a hardware root of trust (TCM+TPCM);

[0004] A root of trust is based on cryptographic techniques;

[0005] A root of trust is an independently operating system;

[0006] The root of trust is isolated from the computing system;

[0007] 2. The trusted root method implements proactive trusted verification of computing components;

[0008] The root of trust is started before the host CPU;

[0009] The root of trust has full access to the host resources;

[0010] The root of trust transparently implements proactive measurement while the host is running;

[0011] 3. Trusted computing provides fundamental support for other security mechanisms;

[0012] Trusted roots provide a foundation of trust for security mechanisms;

[0013] Trusted computing supports the transformation of traditional security mechanisms into dynamic defense;

[0014] Trusted computing effectively enhances situational awareness and attack response capabilities;

[0015] Among them, the main control system hardware and upper-layer application software of large wind turbines lack the qualification of security mutual trust certification, which may face security threats such as supply chain attacks, malicious code injection, and communication protocol hijacking. Traditional static signature verification cannot cover dynamic risks during runtime.

[0016] The lack of a two-way trusted authentication mechanism between the hardware equipment (such as PLCs and sensors) and software components and application control programs (such as control algorithms) of the wind power main control system makes it vulnerable to man-in-the-middle attacks.

[0017] Existing technologies make it difficult to achieve low-overhead dynamic security monitoring in resource-constrained master control systems.

[0018] Currently, domestic industrial server hardware and operating system layers have taken the lead in adopting secure and trusted functions, using hardware TPM (Trusted Platform Module) or HSM (Security Module) to store keys. However, the main control system of wind turbine generators has not yet adopted a secure and trusted measurement two-way mutual recognition mechanism between hardware, operating system, application, and subsystems. This poses a risk of network attacks and malicious software tampering, and cannot reliably guarantee the integrity and authenticity of data transmission between the main control system's software and hardware. For example, wind turbine status data (such as speed and power) may be tampered with, leading to incorrect control commands, which makes it difficult to meet the high security requirements of new power systems.

[0019] In addition, the following related technical shortcomings exist: TPM-based trusted boot only covers the boot phase and does not extend to real-time control processes (such as the main control system application); traditional encryption algorithms (such as RSA) have high computational overhead in resource-constrained main control systems, affecting real-time performance; and cross-device data interaction between the main control system and the pitch system relies on a single protocol, lacking security verification and authentication of communication data, and lacking end-to-end trusted assurance.

[0020] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0021] In view of this, the present invention provides a software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system to solve the above-mentioned problems.

[0022] To solve the above problems, the specific technical solution adopted by the present invention is as follows:

[0023] The mechanism and method for measuring the security, trustworthiness, and mutual recognition of hardware and software in the main control system of wind turbine generators include the following steps:

[0024] S 1. Based on the transmission characteristics of the wind turbine main control system, the wind turbine main control system is divided into levels, and the behavior between each level is measured according to the division results to obtain the behavior measurement results.

[0025] S2. Based on the cross-level trusted mutual recognition mechanism and combined with data link layer encryption technology, perform bidirectional authentication at each level of the wind turbine main control system to obtain the level authentication result.

[0026] S3. Based on the cross-device-level trusted mutual recognition mechanism, perform two-way device-level identity authentication between the wind turbine main control system and the peripheral system to obtain the device-level authentication result.

[0027] S4. Track changes in behavior measurement results, hierarchical authentication results, and device-level authentication results in real time, assess the trustworthiness of the wind turbine main control system, and dynamically execute response strategies based on the assessment results.

[0028] Preferably, the step of dividing the wind turbine main control system into hierarchical levels based on the transmission characteristics of the wind turbine main control system, and measuring the behavior between each level according to the division results to obtain the behavior measurement results includes the following steps:

[0029] S 11. Based on the transmission characteristics of the wind turbine main control system, the wind turbine main control system is divided into a hardware layer, a firmware layer, a software layer and a communication layer to obtain a hierarchical structure.

[0030] S 12. Embed a security chip as a root of trust in the hardware layer of the hierarchical structure, generate a hardware layer trust mechanism, use the hardware layer trust mechanism to verify the integrity of the programmable logic controller firmware, real-time operating system and application at each level, and create a trusted execution environment.

[0031] S 13. Based on a trusted execution environment, dynamically measure and monitor the code segment and memory behavior of critical control processes in real time to obtain behavioral measurement results.

[0032] Preferably, the step of embedding a security chip as a root of trust in the hierarchical hardware layer to generate a hardware-layer trust mechanism, and using the hardware-layer trust mechanism to verify the integrity of the programmable logic controller firmware, real-time operating system, and application programs step by step, and to create a trusted execution environment, includes the following steps:

[0033] S121. Deploy a security chip as a root of trust in the hardware layer of the wind turbine main control system, and store the key and the root of trust in the root of trust to generate a hardware layer trust mechanism.

[0034] S122. Obtain the status of the security chip, the status of the security chip includes startup status and running status;

[0035] S123. When the security chip is in the startup state, the central processing unit of the programmable logic controller firmware is measured and verified using a pre-deployed long short-term memory network model.

[0036] S124. When the security chip is in the running state, the operating system in the programmable logic controller firmware verifies the application program of the wind turbine main control system and creates a trusted execution environment after the verification is completed.

[0037] Preferably, when the security chip is in the enabled state, the measurement and verification of the central processing unit of the programmable logic controller firmware using a pre-deployed long short-term memory network model includes:

[0038] S 1231. Utilize the lightweight long short-term memory network model pre-deployed in the central processing unit within the programmable logic controller firmware of the wind turbine main control system to perform dynamic instruction flow analysis on the security chip boot code and generate a dynamic signature.

[0039] S1232. After startup, the security chip performs reverse analysis of the central processing unit through dynamic signature, and judges the security chip for instruction anomaly and trust measurement based on the analysis result to prevent side-channel attacks.

[0040] S1233. After the security chip passes the trust measurement, release the reset pin so that the security chip can start loading the operating system and read the bootloader in the embedded memory for measurement. After the measurement is successful, load the root file system.

[0041] S1234. The security chip verifies the digital signature or private key of the operating system in the central processing unit of the programmable logic controller, obtains the verification result, and records the verification result in the memory or hard disk of the programmable logic controller. After the security chip passes the verification, it enters the operation stage.

[0042] Preferably, when the security chip is in the running state, the operating system within the programmable logic controller firmware verifies the application program of the wind turbine main control system, and after verification, creates a trusted execution environment including the following steps:

[0043] S 1241. When the security chip is in the running state, the operating system in the programmable logic controller of the wind turbine main control system verifies the application version and private key of the wind turbine main control system.

[0044] S1242. After the application version and private key of the wind turbine main control system are verified, the application process of the wind turbine main control system is added to the verification whitelist.

[0045] S 1243. After the whitelist is successfully added, a trusted execution environment is created by means of secure isolation and protection of code or data access-free access. The trusted execution environment is used to isolate critical control processes to ensure that the key processes and data of the wind turbine main control system application and operating system run in a protected environment.

[0046] Preferably, the real-time dynamic measurement and monitoring of code segments and memory behavior of critical control processes based on a trusted execution environment includes:

[0047] The memory hash value of the code segment of the critical process;

[0048] Logical consistency between peripheral sensor data and preset physical thresholds;

[0049] A sliding window statistical model is used to detect abnormal communication command frequency. Within a time period based on the current time, the number of communication commands and the frame error rate are dynamically monitored.

[0050] Preferably, the method of performing bidirectional authentication at each level of the wind turbine main control system based on a cross-level trusted mutual recognition mechanism and combined with data link layer encryption technology, and obtaining the level authentication results include:

[0051] The mutual recognition between the hardware layer and the firmware layer, the mutual recognition between the firmware layer and the software layer, and the mutual recognition between the software layer and the communication layer;

[0052] The mutual recognition methods include:

[0053] To obtain mutually recognized data between levels, data link layer encryption technology is used, combined with hash values ​​with perceptual features generated by deep hash neural networks;

[0054] The generated hash value is encrypted using a quantum-safe key distribution encryption technology.

[0055] By using blockchain technology, the encrypted hash value is distributed and stored on multiple nodes, and compared with the pre-set whitelist hash value to obtain the hierarchical authentication result.

[0056] Preferably, the step of performing two-way device-level identity authentication between the wind turbine main control system and the peripheral system based on the cross-device-level trusted mutual recognition mechanism to obtain the device-level authentication result includes the following steps:

[0057] S31. Obtain the communication protocol between the wind turbine main control system and the peripheral system, and negotiate the cryptographic algorithm in the communication protocol;

[0058] S32. Based on the communication protocol with added identifier password algorithm, the wind turbine main control system and the peripheral system perform two-way authentication of device-level identity, and record and store the data in the authentication process to obtain the device-level authentication result.

[0059] Preferably, the real-time tracking of changes in behavior measurement results, hierarchical authentication results, and device-level authentication results, and the assessment of the trustworthiness of the wind turbine main control system, and the dynamic execution of response strategies based on the assessment results, includes the following steps:

[0060] S41. Utilize a preset monitoring engine to monitor the data flow information of behavior measurement results, hierarchical authentication results, and device-level authentication results in real time;

[0061] S42. Based on the monitored data stream information, assess the trust status of the wind turbine main control system. If the assessment result indicates that the trust status of the wind turbine main control system is normal, allow the wind turbine main control system to operate normally. Otherwise, trigger the hierarchical response mechanism and execute the dynamic safety response strategy.

[0062] Preferably, the hierarchical response mechanism includes:

[0063] Level 1 response is used to isolate and restart abnormal processes in the wind turbine's main control system.

[0064] Level 2 response is used to switch the wind turbine main control system to a preset backup control module;

[0065] Level 3 response is used to safely shut down the wind turbine's main control system and report the incident to the management platform.

[0066] The beneficial effects of this invention are as follows:

[0067] This invention ensures the security, reliability, and anti-attack capabilities of the main control system of large wind turbine units during startup, operation, and upgrade by constructing a hardware root of trust, dynamic integrity verification at the hardware chip level, operating system level, and application software, and cross-level trust transfer. Furthermore, this invention can solve security problems such as hardware tampering, malicious software injection, and cross-device communication hijacking, while meeting the industrial control requirements of real-time operation and low resource consumption. Attached Figure Description

[0068] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0069] Figure 1 This is a flowchart of the software and hardware security and trust mutual recognition measurement mechanism and method of the wind turbine main control system according to an embodiment of the present invention;

[0070] Figure 2 This is a schematic diagram of the hardware and software security and trust mutual recognition measurement mechanism and method of the wind turbine main control system according to an embodiment of the present invention. Detailed Implementation

[0071] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0072] According to embodiments of the present invention, a software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system is provided.

[0073] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figures 1-2 As shown, the software and hardware security and trust mutual recognition measurement mechanism and method of the wind turbine main control system according to an embodiment of the present invention includes the following steps:

[0074] S 1. Based on the transmission characteristics of the wind turbine main control system, the wind turbine main control system is divided into levels, and the behavior between each level is measured according to the division results to obtain the behavior measurement results.

[0075] In a preferred embodiment, the step of dividing the wind turbine main control system into hierarchical levels based on the transmission characteristics of the wind turbine main control system, and then measuring the behavior between each level based on the division results to obtain the behavior measurement results includes the following steps:

[0076] S 11. Based on the transmission characteristics of the wind turbine main control system, the wind turbine main control system is divided into a hardware layer, a firmware layer, a software layer and a communication layer to obtain a hierarchical structure.

[0077] S 12. Embed a security chip as a root of trust in the hardware layer of the hierarchical structure, generate a hardware layer trust mechanism, use the hardware layer trust mechanism to verify the integrity of the programmable logic controller firmware, real-time operating system and application at each level, and create a trusted execution environment.

[0078] In a preferred embodiment, embedding a security chip as a root of trust in the hierarchical hardware layer to generate a hardware-layer trust mechanism, and using this hardware-layer trust mechanism to verify the integrity of the programmable logic controller firmware, real-time operating system, and applications step by step, and to create a trusted execution environment, includes the following steps:

[0079] S121. Deploy a security chip as a root of trust in the hardware layer of the wind turbine main control system, and store the key and the root of trust in the root of trust to generate a hardware layer trust mechanism.

[0080] S122. Obtain the status of the security chip, the status of the security chip includes startup status and running status;

[0081] S123. When the security chip is in the startup state, the central processing unit of the programmable logic controller firmware is measured and verified using a pre-deployed long short-term memory network model.

[0082] It should be noted that a security chip (supporting at least the national cryptographic SM2 / SM4 algorithm) is embedded in the PLC hardware of the wind turbine main control system to store the key and the trusted measurement benchmark value. The integrity of the PLC firmware, real-time operating system, and application program is verified step by step. Specific integrity verification steps include:

[0083] Startup Phase: A lightweight LSTM model is deployed on the CPU chip within the PLC of the wind turbine main control system. After dynamic analysis of the LSTM model instruction stream, the chip's startup code generates a dynamic signature or a true random number. During startup, the security chip reverse-engineers the CPU chip using the dynamic signature or true random number, effectively monitoring instruction anomalies and trust metrics in real time, thus preventing side-channel attacks. Once the trust metric passes, the reset pin is released, and the CPU chip successfully boots and loads the operating system. After startup, the CPU chip reads the uboot (bootloader) from the eMMC (embedded memory) for measurement. If other chips are present, they are measured sequentially using the same method. Upon successful measurement, the root file system is loaded. The security chip verifies the digital signature or private key of the operating system within the PLC's CPU chip and returns the verification result, recording it in the PLC controller's memory or hard drive. After successful verification, the system enters the operation phase.

[0084] Operation Phase: During the operation phase, the operating system within the PLC of the wind turbine main control system verifies the version of the main control system application and the private key, etc. After the verification is successful, the security chip adds the main control system application process to the verification whitelist. A Trusted Execution Environment (TEE) is created through security isolation, code protection, and data access exemption, isolating critical control processes (such as power regulation algorithms) to ensure that the wind turbine main control system application and the critical processes and data of the operating system run in a protected environment.

[0085] In a preferred embodiment, when the security chip is in the powered-on state, the measurement and verification of the central processing unit of the programmable logic controller firmware using a pre-deployed long short-term memory network model includes:

[0086] S 1231. Utilize the lightweight long short-term memory network model pre-deployed in the central processing unit within the programmable logic controller firmware of the wind turbine main control system to perform dynamic instruction flow analysis on the security chip boot code and generate a dynamic signature.

[0087] It should be noted that during the firmware development phase of the programmable logic controller (PLC), a lightweight long short-term memory (LSTM) network model is integrated into the firmware, and customized training is performed for the specific environment of the wind turbine main control system and the characteristics of the security chip boot code. The training data should contain a large number of normal and abnormal security chip boot code instruction stream samples, enabling the model to learn the characteristic patterns of normal instruction streams. Necessary computing resources, such as memory space and registers, are allocated to the model in the central processing unit (CPU), and initialization settings are performed, including setting the model's initial state and parameters, to ensure the model can operate normally.

[0088] Specifically, by utilizing a pre-deployed lightweight long short-term memory network model, dynamic instruction flow analysis is performed on the boot code of the security chip to generate a dynamic signature, including the following steps:

[0089] S 12311. When the security chip starts up, communication with the programmable logic controller central processing unit is established through a pre-set hardware interface, the security chip startup code instruction stream is captured in real time and preliminarily organized in chronological order, and the start and end times are recorded.

[0090] S12312. Preprocess the instruction stream after initial processing. First, use one-hot encoding to convert the instructions into numerical codes. Then, divide the sequence according to the input length of the lightweight long short-term memory network model and fill or truncate it. Finally, standardize the sequence data.

[0091] S12313. The preprocessed instruction sequence is input into a lightweight long short-term memory network model pre-deployed in the central processing unit according to time steps. The lightweight long short-term memory network model uses internal memory units and gating mechanisms to learn the timing characteristics of the instruction sequence and records the internal state and output results in real time.

[0092] S12314. Extract key features from the output layer or intermediate layer of the lightweight long short-term memory network model and integrate them into a feature vector. Use a trapdoor hash function to calculate a fixed-length hash value from the feature vector as a dynamic signature, store it in a designated area of ​​the central processing unit, and output it to the security chip.

[0093] Specifically, the process of extracting key features from the output or intermediate layers of a lightweight long short-term memory network model and integrating them into a feature vector, then using a trapdoor hash function to calculate a fixed-length hash value from the feature vector as a dynamic signature, storing it in a designated area of ​​the central processing unit, and outputting it to the security chip includes the following steps:

[0094] S 123141. Select key neuron nodes from the output layer or intermediate layer of a lightweight long short-term memory network model and extract their corresponding feature data.

[0095] It's important to note that Lightweight Long Short-Term Memory (LSTM) networks perform complex calculations and feature extraction across multiple neurons when processing data. The output layer contains the final processed results, while the intermediate layers contain feature representations of the data at different processing stages. Key neurons are those that play a crucial role in feature extraction and model decision-making. By selecting these nodes and extracting their corresponding feature data, we can obtain the most representative and critical information during the model's data processing. This information will serve as the basis for generating dynamic signatures later.

[0096] S123142. Integrate and process the extracted key feature data, and splice or fuse them into feature vectors according to preset rules;

[0097] Specifically, concatenation involves linking feature data from different dimensions sequentially to form a higher-dimensional vector; fusion may involve weighted averaging, summing, or other operations on different feature data to synthesize information from various features.

[0098] S123143. Use the trapdoor hash function to calculate the integrated feature vector and generate a fixed-length hash value as a dynamic signature.

[0099] It's important to note that a trapdoor hash function is a special type of hash function. It possesses characteristics such as one-way hashing and collision resistance, and it also contains a trapdoor (secret information). The integrated feature vector is input into the trapdoor hash function, and after calculation, a fixed-length hash value is generated. This hash value is unique and unpredictable, and can serve as a dynamic signature for data. The dynamic signature changes as the input data (feature vector) changes, thus reflecting the characteristics and state of the data in real time.

[0100] S123144. Store the generated dynamic signature in a designated area of ​​the central processing unit and output it to the security chip.

[0101] S1232. After startup, the security chip performs reverse analysis of the central processing unit through dynamic signature, and judges the security chip for instruction anomaly and trust measurement based on the analysis result to prevent side-channel attacks.

[0102] It should be noted that dynamic signatures are feature representations generated by the central processing unit (CPU) after processing data based on a lightweight long short-term memory (LSTM) network model. The purpose of reverse parsing is to reconstruct key information from the signature during the CPU's data processing, such as the data's characteristic patterns and the state of the processing flow, in order to understand the current operating status of the CPU.

[0103] The security chip compares the information obtained through reverse engineering with pre-set normal instruction patterns and behavior rules. If the analysis result deviates from the normal pattern, such as an unexpected instruction sequence or abnormal instruction execution frequency, an instruction anomaly is determined to exist.

[0104] Furthermore, based on the reverse engineering results, the security chip assesses the trustworthiness of the central processing unit (CPU). It checks whether the CPU's hardware and software status meet expected security standards, such as whether the hardware circuitry is functioning correctly and whether the operating system kernel has been tampered with. By considering multiple factors, it determines whether the CPU is in a trustworthy state.

[0105] S1233. After the security chip passes the trust measurement, release the reset pin so that the security chip can start loading the operating system and read the bootloader in the embedded memory for measurement. After the measurement is successful, load the root file system.

[0106] It should be noted that once the security chip completes the trust measurement of the central processing unit and passes the determination, it indicates that the central processing unit is in a safe and trustworthy state. At this time, the security chip will release the reset pin. The reset pin is usually used to control the initialization and restart operation of the device. Releasing the reset pin is equivalent to giving the security chip itself a start signal, allowing it to enter the normal startup process.

[0107] Furthermore, when reading the bootloader from embedded memory for measurement, a complete integrity measurement is required. Specifically, the security chip uses a pre-selected hash algorithm (such as SHA-256, SHA-3, etc.) to calculate a fixed-length hash value from the read bootloader code. The security chip compares the calculated hash value of the bootloader with the expected hash value stored in the memory. If the two hash values ​​are the same, it means that the bootloader code has not been tampered with, and its integrity is guaranteed; if the hash values ​​are different, it indicates that the bootloader may have been maliciously modified, and the security chip will take corresponding security measures, such as stopping the boot process or issuing an alarm.

[0108] S1234. The security chip verifies the digital signature or private key of the operating system in the central processing unit of the programmable logic controller, obtains the verification result, and records the verification result in the memory or hard disk of the programmable logic controller. After the security chip passes the verification, it enters the operation stage.

[0109] S124. When the security chip is in the running state, the operating system in the programmable logic controller firmware verifies the application program of the wind turbine main control system and creates a trusted execution environment after the verification is completed.

[0110] In a preferred embodiment, when the security chip is in the running state, the operating system within the programmable logic controller firmware verifies the application program of the wind turbine main control system, and after verification, creates a trusted execution environment including the following steps:

[0111] S 1241. When the security chip is in the running state, the operating system in the programmable logic controller of the wind turbine main control system verifies the application version and private key of the wind turbine main control system.

[0112] S1242. After the application version and private key of the wind turbine main control system are verified, the application process of the wind turbine main control system is added to the verification whitelist.

[0113] S 1243. After the whitelist is successfully added, a trusted execution environment is created by means of secure isolation and protection of code or data access-free access. The trusted execution environment is used to isolate critical control processes to ensure that the key processes and data of the wind turbine main control system application and operating system run in a protected environment.

[0114] S 13. Based on a trusted execution environment, dynamically measure and monitor the code segment and memory behavior of critical control processes in real time to obtain behavioral measurement results.

[0115] As a preferred embodiment, the real-time dynamic measurement and monitoring of code segments and memory behavior of critical control processes based on a trusted execution environment includes:

[0116] The memory hash value of the code segment of the critical process;

[0117] Logical consistency between peripheral sensor data and preset physical thresholds;

[0118] A sliding window statistical model is used to detect abnormal communication command frequency. Within a time period based on the current time, the number of communication commands and the frame error rate are dynamically monitored.

[0119] S2. Based on the cross-level trusted mutual recognition mechanism and combined with data link layer encryption technology, perform bidirectional authentication at each level of the wind turbine main control system to obtain the level authentication result.

[0120] It should be noted that the wind turbine main control system employs bidirectional authentication between the hardware layer, firmware layer, software layer (also known as the application layer), and communication layer: The hardware layer of the wind turbine main control system mutually recognizes information with the firmware layer operating system through the security chip within the PLC controller. This information includes: PLC hardware information and operating system security identifiers, and information acquired from the main control system application software image. The firmware layer and software layer mutually recognize information including: operating system security identifiers, data acquired from the main control system application software image, and data transmitted from the communication layer. The software layer and communication layer mutually recognize information including: data acquired from the main control system application software image and data transmitted from the communication layer. The authentication method employs data link layer encryption technology, combined with a hash value generated by a deep hash neural network. The perceptual features include system time, private key, transmission data check bit, and data information. Simultaneously, it utilizes quantum encryption technology such as quantum secure key distribution, providing a higher level of security and resisting malicious tampering and attacks caused by obfuscation and distortion. Blockchain node transmission is used, distributing and storing data across multiple nodes to avoid single points of failure. Block hash calculation is performed and compared with pre-stored whitelist hash values ​​to prevent unauthorized modification, achieving cross-level trusted mutual recognition of the wind turbine main control system.

[0121] As a preferred implementation, the method of performing bidirectional authentication at each level of the wind turbine main control system based on a cross-level trusted mutual recognition mechanism and combined with data link layer encryption technology, and obtaining the level authentication results include:

[0122] The mutual recognition between the hardware layer and the firmware layer, the mutual recognition between the firmware layer and the software layer, and the mutual recognition between the software layer and the communication layer;

[0123] The mutual recognition methods include:

[0124] (1) Obtain mutually recognized data between layers, using data link layer encryption technology, and combining it with hash values ​​with perceptual features generated by deep hash neural networks; perceptual features include system time, private key, transmission data check bits and data information;

[0125] It's important to note that a deep hashing neural network is a model that combines deep learning and hashing algorithms. It can extract information with perceptual features from mutually recognized data and generate fixed-length hash values. Compared to traditional hashing algorithms, hash values ​​generated by deep hashing neural networks are more collision-resistant and discriminative, better reflecting the characteristics of mutually recognized data. For example, it can identify subtle changes in mutually recognized data and generate different hash values ​​accordingly.

[0126] Specifically, after acquiring mutually recognized data between different levels, the data is encrypted using a pre-set encryption algorithm and a pre-allocated data link layer encryption key to generate encrypted data. The encrypted data is then input into a trained deep hash neural network. After forward propagation calculation, the node values ​​of the output layer are the generated hash values ​​with perceptual features.

[0127] (2) The generated hash value is encrypted using quantum-safe key distribution encryption technology;

[0128] Specifically, the encryption technology using quantum-safe key distribution to encrypt the generated hash value includes the following steps:

[0129] Step 1: Build a quantum key distribution (QKD) network among the authentication nodes at each level of the wind turbine main control system. Based on the no-cloning theorem of quantum states, use the entanglement-based Gaussian modulation coherent state QKD protocol to generate initial quantum keys and store them in the quantum key pools corresponding to each node.

[0130] It should be noted that in the main control system of a wind turbine, authentication nodes at different levels (such as hardware, firmware, and software layers) need to communicate securely and reliably to achieve mutual recognition. To achieve this, a quantum key distribution network needs to be built between these nodes. Specifically, this can be done by installing corresponding quantum communication devices, such as quantum key distribution terminals, on each node and connecting them through quantum channels (such as optical fibers) to form a network architecture capable of quantum key transmission.

[0131] Among them, the entanglement-based Gaussian modulation coherent state QKD protocol is a specific quantum key distribution protocol. In this protocol, the sending end (such as an authentication node) generates quantum entangled states through an entanglement source (EPR source) and measures the position and momentum of these entangled states. Then, the generated entangled state particles are transmitted to the receiving end (another authentication node) through an untrusted channel. The receiving end performs corresponding measurement operations on the received quantum states. Through a series of quantum operations and measurement processes, the two parties can generate an initial set of quantum keys.

[0132] Step 2: Select a suitable quantum key from the quantum key pool, and preprocess the key based on the remaining key quantity and key freshness of the key pool to meet subsequent encryption requirements.

[0133] It should be noted that the remaining key quantity in the link refers to the number of quantum keys that are still available for distribution in the current link. If the remaining key quantity in the link is small, it is necessary to prioritize encryption schemes that consume less key, or to trigger the key generation and distribution mechanism in a timely manner to supplement key resources and ensure the smooth progress of encryption operations.

[0134] Key freshness reflects information such as the generation and usage time of the key. It measures how new or old the key is. As time goes by, the security of the key may gradually decrease because attackers may have more time to try to crack the key. Therefore, when selecting a key, it is necessary to give priority to keys with higher freshness, that is, keys that have been generated more recently and used less often, in order to improve the security of encryption.

[0135] In addition, based on the selection principles and the key pool state information mentioned above, the selected quantum keys are preprocessed, including key format conversion and encoding. For example, the key is converted into a format suitable for encryption algorithm input, or specific encoding operations are performed on the key to enhance its resistance to attacks.

[0136] Step 3: Using the KNN (K-Nearest Neighbors) algorithm, taking into account parameters such as the remaining key quantity in the link, the key pool capacity, and the average link contribution, calculate the weighted distance of each path, and select the optimal route to distribute the quantum key to the node that needs to encrypt the hash value.

[0137] It's important to note that the K-Nearest Neighbors (KNN) algorithm is a commonly used machine learning algorithm applied to routing in quantum key distribution networks. The basic idea of ​​KNN is to find the K nearest neighbors to the target point based on the distance metric between data points, and then classify or predict based on the characteristics of these neighbors. In quantum key distribution networks, the KNN algorithm can be used to select the optimal routing path based on the characteristic distance metric between key distribution nodes.

[0138] In calculating the weighted distance of each path, the Euclidean distance equidistant metric is used to calculate the weighted distance of each candidate path. The weighted distance comprehensively considers multiple factors such as the remaining key quantity in the link, the key pool capacity, and the average link contribution, which can more accurately reflect the quality of each path. By calculating the weighted distance, each path can be quantitatively evaluated.

[0139] Step 4: Using the distributed quantum key as the encryption key, the AES symmetric encryption algorithm is used to encrypt the hash value with perceptual features generated by the deep hash neural network to ensure the security of hash value transmission and storage.

[0140] It should be noted that, since the quantum key has been successfully distributed to the nodes that need to encrypt the hash value, the distributed quantum key will now be used as the encryption key for subsequent encryption operations.

[0141] Among them, AES (Advanced Encryption Standard) is a widely used symmetric encryption algorithm with advantages such as high encryption efficiency and strong security. In symmetric encryption algorithms, the same key is used for encryption and decryption, which makes the encryption and decryption process relatively simple and efficient. Choosing the AES algorithm to encrypt hash values ​​can improve the encryption and decryption speed while ensuring security, thus meeting the real-time requirements of the wind turbine main control system.

[0142] (3) Using blockchain technology, the encrypted hash value is distributed and stored on multiple nodes, and compared with the pre-set whitelist hash value to obtain the hierarchical authentication result.

[0143] S3. Based on the cross-device-level trusted mutual recognition mechanism, perform two-way device-level identity authentication between the wind turbine main control system and the peripheral system to obtain the device-level authentication result.

[0144] It should be noted that the wind turbine main control system has two-way authentication with peripheral systems such as the pitch system, converter, SCADA system, and EMS system: the communication between the main control system and peripheral devices such as the pitch controller, converter, SCADA system, and EMS system is based on a custom communication protocol. The protocol is enhanced with a custom identifier cryptographic algorithm or a mature identifier cryptographic algorithm (such as SM9) to achieve two-way authentication of device-level identity.

[0145] Among them, lightweight encryption (such as SM4-CTR mode) and message authentication codes (such as SM3-HMAC) are integrated into the mainstream communication protocols used in wind turbine control systems (such as Modbus TCP, CAN, RS485, CANopen, Profibus, OPC UA, 104, etc.).

[0146] Furthermore, blockchain technology is required to record supplier information and software version hashes for hardware components (such as sensors), supporting full lifecycle traceability. Supply chain traceability uses blockchain to record supplier information and software version hashes for hardware components.

[0147] As a preferred implementation, the step of performing two-way device-level identity authentication between the wind turbine main control system and the peripheral system based on a cross-device-level trusted mutual recognition mechanism to obtain the device-level authentication result includes the following steps:

[0148] S31. Obtain the communication protocol between the wind turbine main control system and the peripheral system, and negotiate the cryptographic algorithm in the communication protocol;

[0149] S32. Based on the communication protocol with added identifier password algorithm, the wind turbine main control system and the peripheral system perform two-way authentication of device-level identity, and record and store the data in the authentication process to obtain the device-level authentication result.

[0150] S4. Track changes in behavior measurement results, hierarchical authentication results, and device-level authentication results in real time, assess the trustworthiness of the wind turbine main control system, and dynamically execute response strategies based on the assessment results.

[0151] It should be noted that the runtime monitoring engine monitors the data information flow in real time and can dynamically adjust the measurement strategy according to the needs of different scenarios and different models (dynamically monitor the logical consistency between the code segment memory hash value of the key process and the data of the external sensors and the preset physical threshold; adopt a dynamic threshold algorithm or timing prediction model; transmit using dynamic address and dynamic value by specifying a preset physical threshold, and attach timing information; both sides of the device dynamically monitor the instruction timing and frequency anomalies of the communication protocol). The measurement results are recorded and evaluated, and the dynamic adjustment measurement strategy is adjusted based on the evaluation results.

[0152] As a preferred implementation, the real-time tracking of changes in behavior measurement results, hierarchical authentication results, and device-level authentication results, and the assessment of the trustworthiness of the wind turbine main control system, and the dynamic execution of response strategies based on the assessment results, includes the following steps:

[0153] S41. Utilize a preset monitoring engine to monitor the data flow information of behavior measurement results, hierarchical authentication results, and device-level authentication results in real time;

[0154] It should be noted that the preset monitoring engine is a tool specifically designed for real-time monitoring of data related to the wind turbine main control system. It possesses efficient data acquisition and processing capabilities, enabling real-time acquisition of data streams for behavioral measurement results, hierarchical authentication results, and device-level authentication results. The behavioral measurement result data stream includes behavioral characteristic data of various processes and components within the system, such as process startup time, running frequency, and resource usage. The hierarchical authentication result data stream reflects the authentication status between different levels, such as whether authentication between hardware and software is successful, and whether communication between the network layer and the application layer is legitimate. The device-level authentication result data stream records the authentication information of individual devices, such as device identification, authentication time, and authentication method.

[0155] S42. Based on the monitored data stream information, assess the trust status of the wind turbine main control system. If the assessment result indicates that the trust status of the wind turbine main control system is normal, allow the wind turbine main control system to operate normally. Otherwise, trigger the hierarchical response mechanism and execute the dynamic safety response strategy.

[0156] It should be noted that after obtaining the monitored data stream information, a comprehensive assessment of the reliability status of the wind turbine main control system is required based on this information. The assessment process can employ various methods, such as rule-based assessment and model-based assessment. Rule-based assessment methods can judge various indicators in the data stream according to pre-defined rules. If all indicators meet the rule requirements, the system is considered to be in a normal reliability status; otherwise, the system is considered to be abnormal. Model-based assessment methods can construct a system reliability status model, input the data stream information into the model, and obtain the system reliability status assessment result through model calculation and analysis.

[0157] When the assessment result indicates that the wind turbine main control system is in a normal trusted state, it means that no obvious security threats have been found in the system, and the wind turbine main control system is allowed to continue to operate normally to ensure the power generation efficiency and stability of the wind turbine. However, if the assessment result shows that the system's trusted state is abnormal, a graded response mechanism needs to be triggered immediately. The graded response mechanism divides the response strategy into different levels, such as Level 1 response, Level 2 response, and Level 3 response, based on the severity and scope of the system anomaly.

[0158] As a preferred embodiment, the hierarchical response mechanism includes:

[0159] Level 1 response is used to isolate and restart abnormal processes in the wind turbine's main control system.

[0160] Level 2 response is used to switch the wind turbine main control system to a preset backup control module;

[0161] Level 3 response is used to safely shut down the wind turbine's main control system and report the incident to the management platform.

[0162] To better understand the above technical solutions of the present invention, the following analysis of specific embodiments will further illustrate the above technical solutions of the present invention.

[0163] Example 1;

[0164] (1) Hardware trusted chain construction: a security chip that supports national cryptographic SM2 / SM4 is selected as the trusted root and adapted and integrated into the main control system PLC.

[0165] After the system is powered on, the security chip verifies the Bootloader signature (using the SM2 signature algorithm). If the verification fails, startup is prohibited.

[0166] After starting the RTOS (Real-Time Operating System), the control application calculates hash values ​​(using the SM3 algorithm) in blocks or outputs ciphertext of a custom fixed length and compares them with the pre-stored values.

[0167] (2) Dynamic measurement: A monitoring agent is deployed in a trusted execution environment to collect the hash value of the code segment of the pitch control process every 10ms. If a change in hash value is detected or the sensor speed data exceeds the threshold, protection is carried out according to the preset safety response mechanism (e.g., if the abnormal data of rated speed exceeds 200%, a three-level response mechanism will be triggered).

[0168] (3) Cross-device authentication: The main control system application can exchange identity certificates with the pitch system controller, converter controller and other sub-communication systems through the SM9 algorithm, but not limited to the SM9 algorithm; the communication data is encrypted in the SM4-CTR mode, but not limited to the SM4-CTR mode, and each frame is attached with a specific authentication code, such as the SM3-HMAC authentication code.

[0169] Example 2;

[0170] (1) Hardware Trust Construction: Embed a security chip (supporting at least the national cryptographic SM2 / SM4 algorithm) in the main control system hardware of the wind turbine, and store the key and trust measurement benchmark value. Verify the integrity of the PLC firmware, real-time operating system and application program step by step. The specific integrity verification steps include:

[0171] During the startup phase, the security chip sequentially verifies the PLC firmware digital signature, operating system digital signature, or private key, and returns the verification result based on the verification status, which is then recorded in the PLC controller's memory or hard drive; after successful verification, the system enters the operation phase.

[0172] During the operation phase, the operating system verifies the main control system application version key, etc. After the verification is successful, the security chip adds the main control system application process to the verification whitelist. It creates a Trusted Execution Environment (TEE) through security isolation, code protection and data access protection, etc., to isolate critical control processes (such as power regulation algorithms) and ensure that the wind turbine main control system application and the key processes and data of the operating system run in a protected environment.

[0173] (2) Static and dynamic metrics, including static metrics and dynamic metrics;

[0174] Static metrics perform block hash calculations on the application software image and compare them with pre-stored whitelist hash values ​​to prevent unauthorized modifications.

[0175] Dynamic metrics, using a lightweight monitoring agent (deployed in TEE) to detect the following behaviors in real time:

[0176] The code segment memory hash value of critical processes is collected every 20ms, along with the code segment hash values ​​of core control processes such as pitch and inverter.

[0177] If a change in hash value is detected or the sensor speed data exceeds the threshold, protection will be provided according to the preset safety response mechanism (e.g., if the abnormal rated speed data exceeds 200%, a level 3 response mechanism will be triggered).

[0178] A time range is defined based on the current real-time time, and anomalies in the command frequency of the communication protocol are monitored and statistically analyzed within this time range (based on a sliding window statistical model).

[0179] (2) Cross-level trusted mutual recognition, including device-level two-way authentication, protocol-level security enhancement, and supply chain traceability;

[0180] Device-level two-way authentication: A custom communication protocol is used between the main control system and peripheral devices such as pitch controllers and converters. The protocol is enhanced with a custom identifier cryptographic algorithm or a mature identifier cryptographic algorithm (such as SM9) to achieve device-level two-way authentication.

[0181] Protocol-level security enhancements include adding lightweight encryption (such as SM4-CTR mode) and message authentication codes (such as SM3-HMAC) to each frame of mainstream communication protocols used in wind turbine control systems (such as Modbus TCP, CAN, RS485, CANopen, Profibus, OPC UA, 104, etc.).

[0182] For supply chain traceability, data collected and transmitted between devices uses encryption technology to generate unique hash values. The data is transmitted using blockchain nodes, distributing and storing the data across multiple nodes to avoid single points of failure. The data hash value, along with metadata (such as timestamps), is written into the blockchain to ensure data integrity.

[0183] This ensures that once data is written to the blockchain, it cannot be arbitrarily altered. Each traceability record contains a timestamp written to a block and linked to the previous block, forming a chain arranged in chronological order, allowing for accurate tracing back to the source.

[0184] In summary, by utilizing the above-mentioned technical solutions of this invention, the present invention ensures the security, reliability, and anti-attack capabilities of the main control system of large wind turbine units during startup, operation, and upgrade by constructing a hardware root of trust, dynamic integrity verification at the hardware chip level, operating system level, and application software, and cross-level trust transfer. Furthermore, this invention can solve security problems such as hardware tampering, malicious software injection, and cross-device communication hijacking, while meeting the industrial control requirements of real-time operation and low resource consumption.

[0185] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0186] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A mechanism and method for secure and reliable mutual recognition measurement of hardware and software in the main control system of a wind turbine generator, characterized in that, Includes the following steps: S1. Based on the transmission characteristics of the wind turbine main control system, the wind turbine main control system is divided into levels, and the behavior between each level is measured according to the division results to obtain the behavior measurement results. S2. Based on the cross-level trusted mutual recognition mechanism and combined with data link layer encryption technology, perform bidirectional authentication at each level of the wind turbine main control system to obtain the level authentication result. S3. Based on the cross-device-level trusted mutual recognition mechanism, perform two-way device-level identity authentication between the wind turbine main control system and the peripheral system to obtain the device-level authentication result. S4. Track changes in behavior measurement results, hierarchical authentication results, and device-level authentication results in real time, assess the trustworthiness of the wind turbine main control system, and dynamically execute response strategies based on the assessment results.

2. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 1, characterized in that, The process of dividing the wind turbine main control system into hierarchical levels based on its transmission characteristics, and then measuring the behavior between each level based on the division results, to obtain the behavior measurement results, includes the following steps: S11. Based on the transmission characteristics of the wind turbine main control system, the wind turbine main control system is divided into a hardware layer, a firmware layer, a software layer and a communication layer to obtain a hierarchical structure. S12. Embed a security chip as a root of trust in the hardware layer of the hierarchical structure, generate a hardware layer trust mechanism, use the hardware layer trust mechanism to verify the integrity of the programmable logic controller firmware, real-time operating system and application at each level, and create a trusted execution environment. S13. Based on a trusted execution environment, dynamically measure and monitor the code segment and memory behavior of critical control processes in real time to obtain behavioral measurement results.

3. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 2, characterized in that, The process of embedding a security chip as a root of trust in the hierarchical hardware layer to generate a hardware-layer trust mechanism, and using this mechanism to verify the integrity of the programmable logic controller firmware, real-time operating system, and application programs at each level, and to create a trusted execution environment, includes the following steps: S121. Deploy a security chip as a root of trust in the hardware layer of the wind turbine main control system, and store the key and the root of trust in the root of trust to generate a hardware layer trust mechanism. S122. Obtain the status of the security chip, the status of the security chip includes startup status and running status; S123. When the security chip is in the startup state, the central processing unit of the programmable logic controller firmware is measured and verified using a pre-deployed long short-term memory network model. S124. When the security chip is in the running state, the operating system in the programmable logic controller firmware verifies the application program of the wind turbine main control system and creates a trusted execution environment after the verification is completed.

4. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 3, characterized in that, When the security chip is in the enabled state, the measurement and verification of the central processing unit of the programmable logic controller firmware using a pre-deployed long short-term memory network model includes: S1231. Utilize the lightweight long short-term memory network model pre-deployed in the central processing unit within the programmable logic controller firmware of the wind turbine main control system to perform dynamic instruction flow analysis on the security chip boot code and generate a dynamic signature. S1232. After startup, the security chip performs reverse analysis of the central processing unit through dynamic signature, and judges the security chip for instruction anomaly and trust measurement based on the analysis result to prevent side-channel attacks. S1233. After the security chip passes the trust measurement, release the reset pin so that the security chip can start loading the operating system and read the bootloader in the embedded memory for measurement. After the measurement is successful, load the root file system. S1234. The security chip verifies the digital signature or private key of the operating system in the central processing unit of the programmable logic controller, obtains the verification result, and records the verification result in the memory or hard disk of the programmable logic controller. After the security chip passes the verification, it enters the operation stage.

5. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 3, characterized in that, When the security chip is in the running state, the operating system within the programmable logic controller firmware verifies the application program of the wind turbine main control system, and after verification, creates a trusted execution environment including the following steps: S1241. When the security chip is in the running state, the operating system in the programmable logic controller of the wind turbine main control system verifies the application version and private key of the wind turbine main control system. S1242. After the application version and private key of the wind turbine main control system are verified, the application process of the wind turbine main control system is added to the verification whitelist. S1243. After the whitelist is successfully added, a trusted execution environment is created by means of secure isolation and protection of code or data access-free access. The trusted execution environment is used to isolate critical control processes to ensure that the key processes and data of the wind turbine main control system application and operating system run in a protected environment.

6. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 2, characterized in that, The real-time dynamic measurement and monitoring of code segments and memory behavior of critical control processes based on a trusted execution environment includes: The memory hash value of the code segment of the critical process; Logical consistency between peripheral sensor data and preset physical thresholds; A sliding window statistical model is used to detect abnormal communication command frequency. Within a time period based on the current time, the number of communication commands and the frame error rate are dynamically monitored.

7. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 2, characterized in that, The cross-level trusted mutual recognition mechanism, combined with data link layer encryption technology, performs bidirectional authentication at each level of the wind turbine main control system, resulting in the following authentication results: The mutual recognition between the hardware layer and the firmware layer, the mutual recognition between the firmware layer and the software layer, and the mutual recognition between the software layer and the communication layer; The mutual recognition methods include: To obtain mutually recognized data between levels, data link layer encryption technology is used, combined with hash values ​​with perceptual features generated by deep hash neural networks; The generated hash value is encrypted using a quantum-safe key distribution encryption technology. By using blockchain technology, the encrypted hash value is distributed and stored on multiple nodes, and compared with the pre-set whitelist hash value to obtain the hierarchical authentication result.

8. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 1, characterized in that, The method for performing two-way device-level identity authentication between the wind turbine main control system and peripheral systems based on a cross-device-level trusted mutual recognition mechanism to obtain the device-level authentication result includes the following steps: S31. Obtain the communication protocol between the wind turbine main control system and the peripheral system, and negotiate the cryptographic algorithm in the communication protocol; S32. Based on the communication protocol with added identifier password algorithm, the wind turbine main control system and the peripheral system perform two-way authentication of device-level identity, and record and store the data in the authentication process to obtain the device-level authentication result.

9. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 1, characterized in that, The process of tracking changes in real-time behavior metrics, hierarchical authentication results, and device-level authentication results, assessing the trustworthiness of the wind turbine main control system, and dynamically executing response strategies based on the assessment results includes the following steps: S41. Utilize a preset monitoring engine to monitor the data flow information of behavior measurement results, hierarchical authentication results, and device-level authentication results in real time; S42. Based on the monitored data stream information, assess the trust status of the wind turbine main control system. If the assessment result indicates that the trust status of the wind turbine main control system is normal, allow the wind turbine main control system to operate normally. Otherwise, trigger the hierarchical response mechanism and execute the dynamic safety response strategy.

10. The software and hardware security and trust mutual recognition measurement mechanism and method for wind turbine main control system according to claim 9, characterized in that, The graded response mechanism includes: Level 1 response is used to isolate and restart abnormal processes in the wind turbine's main control system. Level 2 response is used to switch the wind turbine main control system to a preset backup control module; Level 3 response is used to safely shut down the wind turbine's main control system and report the incident to the management platform.

Images