Voice command local storage media, systems, and methods for use on an oil and gas platform ics

Abstract

The application discloses a voice instruction local storage medium, system and method suitable for an ICS of an oil and gas platform, relates to the technical field of an industrial control system of an oil and gas platform, and adopts a multi-layer architecture design resistant to a severe environment, integrates an industrial-grade solid-state hard disk and a flash memory array, cooperates with a special file system to realize hierarchical storage and quick indexing of voice instructions; the system is based on an edge computing architecture, fuses noise preprocessing, instruction hierarchical storage, breakpoint continuation synchronization and a security check mechanism, and realizes whole-process localized closed-loop management of voice instructions from collection, processing, storage to calling. Through the industrial-grade wide-temperature anti-vibration hardware design of the storage medium, the RAID-5 redundancy architecture and the bad block automatic shielding function, the complex working conditions of high humidity, high salt, extreme temperature and continuous vibration of the offshore oil and gas platform can be tolerated, the voice instruction storage integrity reaches 99.8%, and the hardware demand of long-term stable operation of the platform ICS system is met.

Description

Technical Field

[0001] This invention relates to the field of industrial control system technology for oil and gas platforms, and in particular to a local storage medium, system, and method for voice commands suitable for oil and gas platform ICS. Background Technology

[0002] Offshore oil and gas platforms, as core infrastructure for marine energy development, face challenges in their operation and maintenance, including high labor costs, significant safety risks, and limited operational efficiency. Intelligent voice interaction technology, as a natural human-machine collaboration solution that allows users to "talk without lifting a finger," has become a key technology for improving the safety and efficiency of platform operation and maintenance, and is widely used in scenarios such as equipment operation, fault handling, and data retrieval.

[0003] However, the unique high-noise, weak-network environment of offshore oil and gas platforms, along with the stringent requirements of ICS systems for data security and real-time performance, result in significant shortcomings in existing voice command storage solutions: First, general-purpose storage media cannot adapt to the harsh offshore environment of high humidity, high salinity, and strong vibration, leading to low data storage reliability and susceptibility to data loss or storage media damage. Second, existing storage systems largely rely on cloud storage, which cannot achieve real-time storage and retrieval of voice commands during network outages, resulting in significant delays in emergency response. Third, voice commands lack correlation with equipment operations and sensor data, making command tracing and fault analysis difficult. Fourth, there is a lack of security protection mechanisms specifically for oil and gas operation scenarios, posing security risks such as unauthorized command injection and data leakage. Summary of the Invention

[0004] To address the aforementioned problems, the purpose of this invention is to provide a local storage medium, system, and method for voice commands in oil and gas platform ICS, which can achieve highly reliable storage, low-latency invocation, and secure control of voice commands in harsh environments.

[0005] To achieve the above objectives, in a first aspect, the technical solution adopted by the present invention is: a local storage medium for voice commands suitable for ICS (Integrated Signal Processing) on ​​oil and gas platforms, characterized in that it comprises: Hardware storage layer: a dual storage unit consisting of an industrial-grade solid-state drive and a NAND flash memory array. The industrial-grade solid-state drive adopts a wide-temperature and vibration-resistant design, with an operating temperature range of -40℃ to 85℃ and an MTBF of ≥2 million hours. The NAND flash memory array adopts a RAID-5 redundant architecture, with a single array capacity of ≥1TB, and supports automatic bad block masking and data reconstruction. Storage control layer: integrates main control chip, cache controller and encryption module. The main control chip supports SATAIII and NVMe dual protocols, the cache controller is configured with ≥8GB DDR4 cache, and the encryption module uses AES-256 algorithm to encrypt the stored data in real time. File system layer: A dedicated file system based on the improved log structure file system, which supports dividing storage partitions according to the priority of emergency operation commands, regular operation commands, and query commands. Each partition can be configured with independent read / write policies and backup cycles.

[0006] Furthermore, the file system layer also includes an instruction index module, which adopts a three-dimensional index structure of timestamp, instruction type, and entity tag. The entity tag is generated based on the core entities of the knowledge graph in the oil and gas operation field, and supports fast retrieval by device name, operation type, and instruction priority.

[0007] Secondly, the technical solution adopted by the present invention is as follows: a local storage system for voice commands applicable to ICS on oil and gas platforms, characterized in that it includes a local storage medium for voice commands applicable to ICS on oil and gas platforms as described in any one of claims 1-2, and an edge processing unit, a data interface module, a synchronization control module, and a security verification module that interface with the ICS system on oil and gas platforms, with each module working collaboratively: Edge processing unit: integrates an improved CRNN noise suppression model and a lightweight ASR recognition module. The noise suppression model is trained based on the ShipsEar ocean noise dataset and suppresses sea wave and mechanical noise by ≥25dB. The ASR recognition module has <5MB of parameters and supports offline keyword wake-up and command transcription. Data interface module: Supports MODBUS, MQTT, and OPCUA industry standard protocols, and can seamlessly connect with the DCS and ESD subsystems of ICS system, with a data transmission rate of ≥100Mbps; Synchronization control module: includes a local cache management unit and a breakpoint resume unit. The local cache management unit uses the LRU algorithm to cache high-frequency call instructions, with a cache hit rate of ≥90%. When the network is restored, the breakpoint resume unit synchronizes the locally stored data to the cloud according to the timestamp generated by the instructions. The synchronization process supports incremental transmission and data verification. Security verification module: integrates instruction integrity verification unit and access control unit. The integrity verification unit uses SHA-256 hash algorithm to verify instruction data integrity, and the access control unit allocates data read and write permissions based on role permissions and supports full auditing of operation logs.

[0008] Furthermore, the edge processing unit also includes a multimodal data fusion module. The fusion module receives voice command data, device sensor data, and maintenance log text data, dynamically adjusts the weights of each modality data through a gating attention mechanism, generates structured command data containing context information, and then stores it in the storage medium.

[0009] Furthermore, the storage optimization module, based on the instruction lifecycle management strategy, compresses and stores regular operation instructions and query instructions that have exceeded the preset retention period, with a compression ratio of ≥3:1; it permanently stores emergency operation instructions and generates multiple copies for backup.

[0010] Furthermore, the synchronization control module also includes a bandwidth adaptive unit, which monitors the satellite link bandwidth quality in real time. When the bandwidth is <100kbps, it only synchronizes key instruction information, instruction type, execution result, and timestamp; when the bandwidth is ≥1Mbps, it synchronizes complete instruction data and associated multimodal data.

[0011] Furthermore, the security verification module also includes an anomaly detection unit. Based on the entity relationship rules of the knowledge graph in the oil and gas operation field, the anomaly detection unit identifies illegal commands, including mismatched equipment operation permissions, violation of operation procedures, and parameters exceeding safety thresholds, and triggers local alarms and command interception.

[0012] Thirdly, the technical solution adopted by the present invention is: a local storage method for voice commands applicable to oil and gas platform ICS, implemented based on the aforementioned local storage system for voice commands applicable to oil and gas platform ICS, comprising: The received raw voice commands are processed to suppress marine environmental noise, and then offline keyword wake-up and command transcription are performed to output text commands. Text commands, equipment sensor data, and maintenance log text data are dynamically weighted and fused to generate structured command data containing contextual information. Perform integrity checks, access permission checks, and illegal instruction detection based on domain knowledge graphs on structured instruction data, and output legal instructions after passing the checks; Legitimate commands are stored in the aforementioned local storage medium for voice commands applicable to the oil and gas platform ICS, and different types of commands are stored differently according to the command lifecycle management strategy, including compressed storage of regular commands and permanent storage and multiple copy backups of emergency commands. It adopts a local caching and breakpoint resume mechanism, and adaptively selects to synchronize key information or complete data based on network bandwidth quality, and synchronizes the locally stored data to the cloud.

[0013] Furthermore, the integrity verification uses the SHA-256 hash algorithm to verify the integrity of the instruction data; the access permission verification assigns data read and write permissions based on role permissions and supports full auditing of operation logs. The illegal command detection based on domain knowledge graphs specifically involves identifying illegal commands, including those with mismatched equipment operation permissions, violations of operation procedures, and parameters exceeding safety thresholds, based on entity relationship rules in the oil and gas operation domain knowledge graph, and triggering local alarms and command interception.

[0014] Furthermore, the local cache uses the LRU algorithm to cache frequently called instructions, with a cache hit rate of ≥90%; the breakpoint resume mechanism synchronizes data in the order of timestamps generated by instructions when the network recovers, supporting incremental transmission and data verification; the satellite link bandwidth quality is monitored in real time. When the bandwidth is <100kbps, only the instruction type, execution result, and timestamp are synchronized. When the bandwidth is ≥1Mbps, the complete instruction data and associated multimodal data are synchronized.

[0015] The present invention has the following advantages due to the adoption of the above technical solutions: 1. This invention provides a local storage medium and system for voice commands in an oil and gas platform ICS. The storage medium adopts an industrial-grade wide-temperature and vibration-resistant hardware design, combined with a RAID-5 redundant architecture and automatic bad block shielding function. It can withstand the complex operating conditions of high humidity, high salinity, extreme temperature and continuous vibration of offshore oil and gas platforms. The voice command storage integrity reaches 99.8%, meeting the hardware requirements for long-term stable operation of the platform ICS system.

[0016] 2. This invention provides a local storage medium and system for voice commands in ICS of oil and gas platforms. Based on an edge computing architecture, it realizes the local processing of voice commands from acquisition, noise reduction, transcription to storage and retrieval. It can still work normally without cloud network. The local command call response latency is ≤800ms. There is no data loss within ≤72 hours of network interruption. It is perfectly adapted to weak network / network outage scenarios at sea and greatly improves the efficiency of emergency operation response.

[0017] 3. This invention provides a local storage medium and system for voice commands in an oil and gas platform ICS. It integrates AES-256 real-time encryption, SHA-256 integrity verification, role-based access control, and an abnormal command detection mechanism based on an oil and gas knowledge graph. It can effectively intercept illegal operation commands, prevent data tampering and leakage, and support full auditing of operation logs. It avoids security risks of the platform ICS system from data storage to command execution.

[0018] 4. This invention provides a local storage medium and system for voice commands in oil and gas platform ICS. It adopts instruction priority partitioning storage, lifecycle compression management and LRU high-frequency caching strategy, which improves storage resource utilization by more than 3 times. It also supports mainstream industrial protocols such as MODBUS, MQTT and OPCUA, and can be seamlessly connected with the DCS, ESD and other subsystems of oil and gas platform ICS system. At the same time, it is suitable for voice interaction scenarios of human / unmanned platforms. It does not require major modification of the existing system, and the engineering implementation and subsequent operation and maintenance costs are low. Attached Figure Description

[0019] Figure 1This is a flowchart illustrating the entire process of local storage of voice commands in this invention. Figure 2 This is a flowchart illustrating the local caching of instructions during network anomalies and the resume transmission after network recovery in this invention. Figure 3 This is a flowchart illustrating the security verification process for local voice command invocation according to the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0021] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0022] Example 1: In this embodiment of the present invention, a local storage medium for voice commands suitable for an oil and gas platform ICS is provided. In this embodiment, as... Figure 1 As shown, the local storage medium for voice commands includes: Hardware storage layer: a dual storage unit consisting of an industrial-grade solid-state drive and a NAND flash memory array. The industrial-grade solid-state drive adopts a wide-temperature and vibration-resistant design, with an operating temperature range of -40℃ to 85℃ and an MTBF of ≥2 million hours. The NAND flash memory array adopts a RAID-5 redundant architecture, with a single array capacity of ≥1TB, and supports automatic bad block masking and data reconstruction. Storage control layer: integrates main control chip, cache controller and encryption module. The main control chip supports SATAIII and NVMe dual protocols, the cache controller is configured with ≥8GB DDR4 cache, and the encryption module uses AES-256 algorithm to encrypt the stored data in real time. File system layer: A dedicated file system based on the improved log structure file system, which supports dividing storage partitions according to the priority of emergency operation commands, regular operation commands, and query commands. Each partition can be configured with independent read / write policies and backup cycles.

[0023] In the above embodiments, the file system layer also includes an instruction index module. The instruction index module adopts a three-dimensional index structure of timestamp, instruction type, and entity tag. The entity tag is generated based on the core entities of the knowledge graph in the oil and gas operation field and supports fast retrieval by device name, operation type, and instruction priority.

[0024] Specifically, in this embodiment, the hardware storage layer uses an industrial-grade NVMe SSD and three 1TB NAND flash memory modules to form a RAID-5 array; the storage control layer uses an Intel Xeon D-2700 main control chip, configured with 8GB DDR4 cache, and integrates an AES-256 hardware encryption module; the file system layer is based on an improved Linux LFS, divided into high, medium, and low priority partitions, with a backup cycle of 30 days for high-priority partitions, 7 days for medium-priority partitions, and 1 day for low-priority partitions; the instruction index module adopts a three-dimensional structure of "timestamp, instruction type, and entity tag".

[0025] Example 2: In this embodiment of the present invention, a local storage system for voice commands suitable for an oil and gas platform ICS is provided. In this embodiment, as... Figure 2 , Figure 3 As shown, the system includes: the local storage medium for voice commands suitable for oil and gas platform ICS as described in Embodiment 1 above, and an edge processing unit, data interface module, synchronization control module, and security verification module that interface with the oil and gas platform ICS system. These modules work collaboratively. Edge processing unit: integrates an improved CRNN noise suppression model and a lightweight ASR recognition module. The noise suppression model is trained based on the ShipsEar ocean noise dataset and suppresses sea wave and mechanical noise by ≥25dB. The ASR recognition module has <5MB of parameters and supports offline keyword wake-up and command transcription. Data interface module: Supports MODBUS, MQTT, and OPCUA industry standard protocols, and can seamlessly connect with the DCS and ESD subsystems of ICS system, with a data transmission rate of ≥100Mbps; Synchronization control module: includes a local cache management unit and a breakpoint resume unit. The local cache management unit uses the LRU algorithm to cache high-frequency call commands, with a cache hit rate of ≥90%. When the network is restored, the breakpoint resume unit synchronizes the locally stored data to the cloud according to the timestamp generated by the command. The synchronization process supports incremental transmission and data verification. Security verification module: integrates instruction integrity verification unit and access control unit. The integrity verification unit uses SHA-256 hash algorithm to verify the integrity of instruction data, and the access control unit assigns data read and write permissions based on role permissions, and supports full auditing of operation logs.

[0026] In the above embodiments, the edge processing unit further includes a multimodal data fusion module. The fusion module receives voice command data, device sensor data, and maintenance log text data, dynamically adjusts the weights of each modal data through a gating attention mechanism, generates structured command data containing context information, and then stores it in the storage medium.

[0027] In the above embodiments, the storage optimization module compresses and stores regular operation instructions and query instructions that have exceeded the preset retention period based on the instruction lifecycle management strategy, with a compression ratio of ≥3:1; it permanently stores emergency operation instructions and generates multiple copies for backup.

[0028] In the above embodiments, the synchronization control module further includes a bandwidth adaptive unit. The bandwidth adaptive unit monitors the satellite link bandwidth quality in real time. When the bandwidth is <100kbps, it only synchronizes key instruction information, instruction type, execution result, and timestamp. When the bandwidth is ≥1Mbps, it synchronizes complete instruction data and associated multimodal data.

[0029] In the above embodiments, the security verification module further includes an anomaly detection unit. The anomaly detection unit identifies illegal instructions based on the entity relationship rules of the knowledge graph in the oil and gas operation field, including mismatched equipment operation permissions, violation of operation procedures, and parameters exceeding the safety threshold, and triggers local alarms and instruction interception.

[0030] Example 3: In this embodiment of the present invention, a method for local storage of voice commands for ICS on oil and gas platforms is provided, based on the local storage system for voice commands for ICS on oil and gas platforms described in Example 2 above. In this embodiment, the method includes the following steps: 1) The received raw voice commands are processed to suppress marine environmental noise, and then offline keyword wake-up and command transcription are performed to output text commands; 2) By dynamically weighting and fusing text commands, equipment sensor data, and maintenance log text data, structured command data containing contextual information is generated; 3) Perform integrity checks, access permission checks, and illegal instruction detection based on domain knowledge graphs on structured instruction data, and output legal instructions after passing the checks; 4) Store the legitimate commands to the local storage medium for voice commands applicable to the oil and gas platform ICS in Example 1, and perform differentiated storage for different types of commands according to the command lifecycle management strategy, including compressed storage of regular commands and permanent storage and multiple copy backups of emergency commands; 5) Adopting a local caching and breakpoint resume mechanism, it adaptively selects to synchronize key information or complete data based on network bandwidth quality, and synchronizes locally stored data to the cloud.

[0031] In step 1) above, the noise suppression process uses an improved CRNN noise suppression model, which is trained based on the ShipsEar ocean noise dataset and suppresses sea wave and mechanical noise by ≥25dB; the instruction transcription uses a lightweight ASR recognition module with less than 5MB of parameters, which supports offline keyword wake-up and instruction transcription.

[0032] In step 2) above, the multimodal data fusion module dynamically adjusts the weights of voice command data, device sensor data and maintenance log text data through a gating attention mechanism to generate structured command data containing contextual information.

[0033] In step 3) above, the integrity verification uses the SHA-256 hash algorithm to verify the integrity of the instruction data; the access permission verification assigns data read and write permissions based on role permissions and supports full auditing of operation logs.

[0034] The illegal command detection based on domain knowledge graphs specifically involves identifying illegal commands, including those with mismatched equipment operation permissions, violations of operation procedures, and parameters exceeding safety thresholds, based on entity relationship rules in the oil and gas operation domain knowledge graph, and triggering local alarms and command interception.

[0035] In step 4) above, the compression ratio of compressed storage is ≥3:1, and it is only for regular operation instructions and query instructions that have exceeded the preset retention period; emergency operation instructions are permanently stored and multiple copies are generated for backup.

[0036] In step 5) above, the local cache uses the LRU algorithm to cache high-frequency call instructions, with a cache hit rate of ≥90%; the breakpoint resume mechanism synchronizes data according to the timestamp generated by the instructions when the network is restored, supporting incremental transmission and data verification; the satellite link bandwidth quality is monitored in real time. When the bandwidth is <100kbps, only the instruction type, execution result, and timestamp are synchronized. When the bandwidth is ≥1Mbps, the complete instruction data and associated multimodal data are synchronized.

[0037] Example 4: This example describes the deployment and testing of the storage system. The system of this invention is deployed on an unmanned maritime platform ICS system. The edge processing unit uses an NVIDIA Jetson Orin NX edge computing device, integrating an improved CRNN noise suppression model and a lightweight ASR module. The data interface module interfaces with the platform's DCS system via the OPCUA protocol. The synchronization control module sets the bandwidth adaptive threshold to 100kbps and the breakpoint resume timeout to 1 hour. The security verification module configures three types of role permissions, and the anomaly detection rules are built based on the platform's SOP and the device manual.

[0038] Test results show that: Environmental adaptability: The storage medium exhibits no data loss and stable read / write performance under extreme temperatures of -40℃ and 85℃ and vibration of 10g. Real-time performance: Local storage latency for voice commands ≤200ms, local call response latency ≤750ms, meeting the requirement of <1 second real-time interaction; Security: Successfully intercepted 98% of illegal commands; data was encrypted during storage and transmission with no risk of leakage. Reliability: During a 72-hour network outage, the system successfully stored 1200 commands, and the success rate of resuming transmissions after network recovery was 100%. Search efficiency: The average time for searching by device name is 35ms, and the average time for searching by operation type is 42ms, with a search accuracy of 99.2%.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A local storage medium for voice commands suitable for ICS (Integrated Control System) on oil and gas platforms, characterized in that, include: Hardware storage layer: a dual storage unit consisting of an industrial-grade solid-state drive and a NAND flash memory array. The industrial-grade solid-state drive adopts a wide-temperature and vibration-resistant design, with an operating temperature range of -40℃ to 85℃ and an MTBF of ≥2 million hours. The NAND flash memory array adopts a RAID-5 redundant architecture, with a single array capacity of ≥1TB, and supports automatic bad block masking and data reconstruction. Storage control layer: integrates main control chip, cache controller and encryption module. The main control chip supports SATAIII and NVMe dual protocols, the cache controller is configured with ≥8GB DDR4 cache, and the encryption module uses AES-256 algorithm to encrypt the stored data in real time. File system layer: A dedicated file system based on the improved log structure file system, which supports dividing storage partitions according to the priority of emergency operation commands, regular operation commands, and query commands. Each partition can be configured with independent read / write policies and backup cycles.

2. The local storage medium for voice commands suitable for ICS on oil and gas platforms as described in claim 1, characterized in that, The file system layer also includes an instruction index module, which adopts a three-dimensional index structure of timestamp, instruction type, and entity tag. The entity tag is generated based on the core entities of the knowledge graph in the oil and gas operation field, and supports fast retrieval by device name, operation type, and instruction priority.

3. A local storage system for voice commands suitable for ICS (Integrated Control System) of oil and gas platforms, characterized in that, Includes a local storage medium for voice commands applicable to oil and gas platform ICS as described in any of claims 1-2, and an edge processing unit, data interface module, synchronization control module, and security verification module that interface with the oil and gas platform ICS system, with each module working collaboratively: Edge processing unit: integrates an improved CRNN noise suppression model and a lightweight ASR recognition module. The noise suppression model is trained based on the ShipsEar ocean noise dataset and suppresses sea wave and mechanical noise by ≥25dB. The ASR recognition module has <5MB of parameters and supports offline keyword wake-up and command transcription. Data interface module: Supports MODBUS, MQTT, and OPCUA industry standard protocols, and can seamlessly connect with the DCS and ESD subsystems of ICS system, with a data transmission rate of ≥100Mbps; Synchronization control module: includes a local cache management unit and a breakpoint resume unit. The local cache management unit uses the LRU algorithm to cache high-frequency call instructions, with a cache hit rate of ≥90%. When the network is restored, the breakpoint resume unit synchronizes the locally stored data to the cloud according to the timestamp generated by the instructions. The synchronization process supports incremental transmission and data verification. Security verification module: integrates instruction integrity verification unit and access control unit. The integrity verification unit uses SHA-256 hash algorithm to verify instruction data integrity, and the access control unit allocates data read and write permissions based on role permissions and supports full auditing of operation logs.

4. A local voice command storage system for an oil and gas platform ICS according to claim 3, characterized in that: The edge processing unit also includes a multimodal data fusion module. The fusion module receives voice command data, device sensor data, and maintenance log text data, dynamically adjusts the weights of each modality data through a gating attention mechanism, generates structured command data containing context information, and then stores it in the storage medium.

5. A local voice command storage system for an oil and gas platform ICS according to claim 3, characterized in that: The storage optimization module, based on the instruction lifecycle management strategy, compresses and stores regular operation instructions and query instructions that have exceeded the preset retention period, with a compression ratio of ≥3:1; it permanently stores emergency operation instructions and generates multiple copies for backup.

6. A local voice command storage system for an oil and gas platform ICS according to claim 3, characterized in that: The synchronization control module also includes a bandwidth adaptive unit, which monitors the satellite link bandwidth quality in real time. When the bandwidth is <100kbps, it only synchronizes key instruction information, instruction type, execution result, and timestamp; when the bandwidth is ≥1Mbps, it synchronizes complete instruction data and associated multimodal data.

7. A local voice command storage system for oil and gas platform ICS according to claim 3, characterized in that: The security verification module also includes an anomaly detection unit. Based on the entity relationship rules of the knowledge graph in the oil and gas operation field, the anomaly detection unit identifies illegal commands, including mismatched equipment operation permissions, violation of operation procedures, and parameters exceeding safety thresholds, and triggers local alarms and command interception.

8. A method for local storage of voice commands for oil and gas platform ICS, implemented based on the local storage system for voice commands for oil and gas platform ICS as described in any one of claims 3-7, characterized in that: include: The received raw voice commands are processed to suppress marine environmental noise, and then offline keyword wake-up and command transcription are performed to output text commands. Text commands, equipment sensor data, and maintenance log text data are dynamically weighted and fused to generate structured command data containing contextual information. Perform integrity checks, access permission checks, and illegal instruction detection based on domain knowledge graphs on structured instruction data, and output legal instructions after passing the checks; Valid commands are stored in the local storage medium for voice commands applicable to oil and gas platform ICS as described in any of claims 1-2, and different types of commands are stored differently according to the command lifecycle management strategy, including compressed storage of regular commands and permanent storage and multiple copy backups of emergency commands. It adopts a local caching and breakpoint resume mechanism, and adaptively selects to synchronize key information or complete data based on network bandwidth quality, and synchronizes the locally stored data to the cloud.

9. The method for local storage of voice commands for ICS on oil and gas platforms according to claim 8, characterized in that: Integrity verification uses the SHA-256 hash algorithm to verify the integrity of instruction data; access permission verification assigns data read and write permissions based on role permissions and supports full auditing of operation logs. The illegal command detection based on domain knowledge graphs specifically involves identifying illegal commands, including those with mismatched equipment operation permissions, violations of operation procedures, and parameters exceeding safety thresholds, based on entity relationship rules in the oil and gas operation domain knowledge graph, and triggering local alarms and command interception.

10. The method for local storage of voice commands for ICS on oil and gas platforms according to claim 8, characterized in that: The local cache uses the LRU algorithm to cache frequently called instructions, with a cache hit rate of ≥90%; the breakpoint resume mechanism synchronizes data in the order of timestamps generated by instructions when the network is restored, and supports incremental transmission and data verification; the quality of satellite link bandwidth is monitored in real time. When the bandwidth is <100kbps, only the instruction type, execution result and timestamp are synchronized. When the bandwidth is ≥1Mbps, the complete instruction data and associated multimodal data are synchronized.

Images