Equipment intelligent guarantee system and method based on generative cognitive twinning

Abstract

The invention provides an equipment intelligent guarantee system and method based on generative cognitive twinning. According to the system, a generative artificial intelligence model is used as a cognitive kernel, multi-modal data such as sensor signals, images and texts and unstructured knowledge such as manual, logs and maintenance records are fused, and cognitive twin bodies with semantic comprehension, causal reasoning and adaptive evolution capabilities are constructed. The system comprises a cognitive kernel module, a digital mapping module, a cognitive evolution module, a guarantee decision module and a man-machine interaction module, realizes cross-modal perception and cognitive consistency correction, generates an equipment guarantee strategy through semantic reasoning, and feeds back an optimization decision in a closed loop. The method can be widely applied to intelligent operation, maintenance and guarantee of various kinds of complex equipment, and the accuracy, autonomy and continuous optimization capability of guarantee decision making are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent operation and maintenance and support of equipment, in particular to an equipment intelligent support system and method based on generative cognitive twin. BACKGROUND

[0002] With the development of modern complex equipment towards intelligence and digitization, a large amount of heterogeneous data (such as sensor measurement data, equipment images and text logs) is used to support the operation, maintenance and support of equipment. However, the existing equipment support system usually processes a single type of data, lacks comprehensive understanding and utilization of multi-source heterogeneous information, and the support decision still mainly relies on manual analysis and experience. In addition, although the traditional digital twin technology can realize the virtual mapping of the physical equipment state, it usually stays at the simulation and monitoring of the data level, and lacks the deep semantic understanding and causal reasoning ability of the equipment state and failure mechanism. The existing technology cannot fully mine the correlation between equipment operation data and historical knowledge (such as maintenance manual, fault log), cannot dynamically adapt to the changing working conditions and knowledge updates, and leads to the lack of intelligence and adaptability in the formulation of equipment support scheme.

[0003] In recent years, the development of generative artificial intelligence (such as large language models and deep learning generative models) has brought new opportunities for intelligent equipment support. Generative artificial intelligence models have the ability to learn semantics and knowledge from large-scale unstructured data, and can generate reasoning and suggestions with logical consistency according to the context. If generative artificial intelligence technology is introduced into the digital twin system and cognitive intelligence is given to the digital twin, it will have the ability of cross-modal semantic understanding, causal relationship reasoning and autonomous evolution, which will break through the bottleneck of the existing equipment support system. However, there is no "cognitive twin" solution for the general equipment support field, which uses generative artificial intelligence as the core of the twin to unify the processing of multi-modal data and knowledge, and realizes the intelligent decision-making loop. Therefore, a new technical solution is needed to combine generative cognitive intelligence and digital twin technology to improve the automation and intelligence level of the equipment support system. SUMMARY

[0004] In view of the deficiencies of the prior art, the purpose of the present application is to provide a general equipment intelligent support system, which can integrate multi-modal data and domain knowledge for semantic-level understanding and reasoning, thereby automatically generating efficient support decisions and continuously optimizing, solving the defects that the existing technology cannot fully mine the correlation between equipment operation data and historical knowledge (such as maintenance manual, fault log), cannot dynamically adapt to the changing working conditions and knowledge updates, and leads to the lack of intelligence and adaptability in the formulation of equipment support scheme. In order to achieve the above purpose, the present application provides an equipment intelligent support system based on generative cognitive twin, which includes the following modules: The cognitive kernel module internally generates a generative artificial intelligence model and utilizes the generative artificial intelligence model to perform semantic understanding and causal reasoning on multi-modal data and unstructured knowledge from the equipment, constructs a cognitive twin representation of the equipment, and outputs state diagnosis and prediction results of the equipment. The digital mapping module collects multi-source data of the equipment, and updates the digital twin model to reflect the real-time state of the equipment, to realize virtual-real synchronization of the physical equipment and the cognitive twin state; the multi-source data includes sensor signals, images, and operation logs. The cognitive evolution module trains and updates the artificial intelligence model in the cognitive kernel module according to new data in the equipment support process and feedback in the human-machine interaction process, optimizes the knowledge and model parameters of the cognitive twin, realizes adaptive evolution and continuous learning of the model, and corrects cognitive bias to maintain virtual-real cognitive consistency. The support decision module receives diagnosis and analysis results of the cognitive kernel module, generates corresponding equipment support strategies and decision suggestions through semantic reasoning, and evaluates the effectiveness of the strategies after execution according to feedback; the equipment support strategies and decision suggestions include fault diagnosis, maintenance scheme development, and performance optimization. The human-machine interaction module provides a user-system interaction interface to display the equipment support scheme given by the support decision module in a visual and natural language manner, so that maintenance personnel can query the equipment state, obtain support suggestions, receive user confirmation and correction opinions, and collect actual execution feedback and enter the cognitive evolution module to realize human-machine collaboration and decision-making closed loop.

[0005] Further, the cognitive kernel module includes a generative pre-training model trained on equipment domain data, which has cross-modal information fusion and semantic understanding capabilities, can process numerical sensor data, image signals, and text knowledge simultaneously, identify potential equipment failure modes, and perform causal relationship inference.

[0006] Further, the digital mapping module includes a data acquisition unit for acquiring equipment operation parameters and environmental parameters in real time, a data preprocessing unit for cleaning, format conversion, and time alignment of data from different sources, and a twin synchronization unit for updating the processed data to state variables of the cognitive twin model.

[0007] Further, the cognitive evolution module is provided with a cognitive consistency correction mechanism, which triggers model adjustment and / or data re-collection when the reasoning results output by the cognitive kernel module are inconsistent with actual observations and / or pre-stored domain knowledge, to correct cognitive bias of the cognitive twin; the cognitive evolution module incorporates new fault cases after each equipment maintenance through incremental training, to continuously improve the adaptability of the model to new problems.

[0008] Further, the guarantee decision module adopts knowledge graph reasoning, rule engine and reinforcement learning algorithm to make decision calculation on the equipment state information provided by the cognitive kernel module, and generates guarantee strategies including maintenance steps under fault diagnosis conclusion, preventive maintenance plan, spare parts scheduling scheme and operation optimization suggestion, and the execution effect is fed back for updating the decision logic.

[0009] Further, the human-computer interaction module includes a natural language processing unit, which enables maintenance personnel to interact with the system through voice and text to inquire and obtain explanatory equipment state explanations and guarantee suggestions; and the human-computer interaction module transmits the information fed back by the user to the cognitive evolution module after analysis, to perfect the knowledge base of the cognitive twin.

[0010] An equipment intelligent guarantee method based on generative cognitive twin, comprising the following steps: 1) Data acquisition and twin synchronization: the digital mapping module collects multi-modal operation data of the equipment, and combines it with historical knowledge base information to update the cognitive twin model of the equipment, forming a complete description of the equipment state; 2) Cognitive analysis: the cognitive kernel module performs semantic understanding and causal reasoning on the comprehensive information obtained in step 1), to obtain the health state evaluation, fault diagnosis conclusion or performance prediction result of the equipment; 3) Strategy formulation: according to the analysis result of step 2), the guarantee decision module generates corresponding equipment guarantee strategies and maintenance schemes, including required operation measures, resource arrangement and execution time; 4) Execution and interaction: the strategy scheme is provided to the equipment maintenance personnel through the human-computer interaction module for implementation, and the maintenance personnel can confirm and adjust the scheme according to the actual situation, and the system continuously monitors the equipment state and accepts supplementary information input by the human during the execution process; 5) Feedback learning: after the guarantee measures are executed, the actual results and feedback data are collected, and the model and knowledge of the cognitive twin are updated by the cognitive evolution module, and new fault phenomena and processing experience are included in the learning cycle, so as to enter the next round of equipment state data acquisition and analysis, and improve the intelligent optimization level of the guarantee strategy.

[0011] Further, in the cognitive analysis process of step 2), the different modal data obtained are first subjected to feature extraction and preliminary analysis: the sensor time series data is subjected to anomaly detection and trend analysis, the device image is subjected to target recognition and state evaluation, and the text log is subjected to natural language processing to extract key events and parameters; then the analysis results of each modal are fused through time and semantic association, and are uniformly input into the generative artificial intelligence model to output the final diagnosis and prediction result.

[0012] Further, in the feedback learning process of step 5), when the cognitive kernel module diagnoses an error and / or the strategy effect is poor, the cognitive evolution module extracts the corresponding case, which is used together with the original training data to incrementally train and / or fine-tune the generative artificial intelligence model, enhance the accuracy of the model in similar scenarios, and adjust the rule parameters of the decision-making module to avoid similar mistakes.

[0013] The equipment intelligent support system and method based on generative cognitive twinning provided by the present application realize the transformation of equipment support from "passive response" to "active intelligence".

[0014] Compared with the prior art, the present application achieves the following technical effects: 1. Cross-modal cognitive ability: capable of fusing and analyzing multi-modal information such as sensor data, images and texts, comprehensively judging the equipment state and failure at the semantic level, avoiding information silos, and improving the comprehensiveness of decision-making basis.

[0015] 2. Cognitive consistency correction: introduce a cognitive consistency correction mechanism to automatically find and correct inconsistencies between the cognitive twin and the actual equipment state or known mechanism, ensuring the reliability and accuracy of the model reasoning results.

[0016] 3. Intelligent support decision: based on the semantic reasoning ability of generative artificial intelligence, automatically generate maintenance and repair schemes, fault handling measures and other support strategies, which are faster and more creative than manually formulated, and can provide optimized decisions for complex failures.

[0017] 4. Self-adaptive evolution optimization: through continuous learning of new data and feedback, the knowledge base and model parameters of the cognitive twin are constantly improved, and the support strategy is self-adaptively optimized with the evolution of the equipment state, realizing an intelligent support system that gets smarter with use.

[0018] 5. General applicability: the system and method do not rely on special models or rules of specific industries, and can be applied to the support fields of various complex equipment such as aviation, aerospace, vehicles and energy, and has wide applicability and promotional value. BRIEF DESCRIPTION OF DRAWINGS

[0019] For ease of illustration, the present application is described in detail by the following specific embodiments and drawings.

[0020] Figure 1 is a schematic diagram of an equipment intelligent support system architecture provided by the present application based on generative cognitive twinning; wherein the cognitive kernel module (1), the digital mapping module (2), the cognitive evolution module (3), the support decision-making module (4), and the human-computer interaction module (5); Figure 2 is a schematic diagram of an equipment intelligent support method flow provided by the present application. Detailed Implementation

[0021] The following are specific embodiments of this application, described in conjunction with the accompanying drawings, to further illustrate the technical solutions of this application. However, this application is not limited to these embodiments. Specific details, such as particular configurations, are provided in the following description merely to aid in a comprehensive understanding of the embodiments of this application. Therefore, those skilled in the art should understand that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application.

[0022] like Figure 1 As shown, an intelligent equipment support system based on generative cognitive twins includes five main functional modules: a cognitive kernel module (1), a digital mapping module (2), a cognitive evolution module (3), a support decision module (4), and a human-computer interaction module (5). Each module forms a closed-loop collaborative relationship through data and signal interaction, deeply integrating the generative artificial intelligence model with the digital twin. This enables the generative artificial intelligence model to possess cross-modal understanding, causal reasoning, and adaptive evolution capabilities, thereby achieving intelligent support for the equipment.

[0023] The cognitive kernel module (1), as the core of the generative cognitive twin equipment intelligent support system, incorporates generative artificial intelligence models, such as deep learning models trained on large-scale equipment data and technical documents. This module receives multimodal information input from the digital mapping module (2) and textual knowledge input from historical knowledge bases (such as maintenance manuals and maintenance records) to perform semantic understanding and analytical reasoning on the current state of the equipment. Through anomaly pattern recognition of sensor values, target and state recognition of images, and parsing of text descriptions, the cognitive kernel module (1) can infer possible causes of equipment failures, performance bottlenecks, or optimization space. During the reasoning process, the cognitive kernel module (1) adopts a causal reasoning model to establish a correlation between observed symptoms and potential causes, and generates corresponding diagnostic conclusions and prediction results. When multi-source information is contradictory or inconsistent, the module also actively inquires or retrieves more data or adjusts the internal model by referring to domain rules through a cognitive consistency correction mechanism to ensure that the cognitive results are consistent with the objective situation.

[0024] The digital mapping module (2) is responsible for data mapping between physical equipment and digital twin. This module connects various sensors and data acquisition devices to obtain real-time equipment operating parameters (such as temperature, pressure, vibration signals, etc.), operation logs, environmental parameters, and device image / video monitoring screens, etc. The digital mapping module (2) pre-processes and formats the obtained data, and updates it to the cognitive twin model, so that the cognitive kernel module (1) always has the latest information reflecting the current state of the equipment. Through virtual-real synchronization, this module ensures that the state parameters of the digital twin reflect the actual equipment status in a timely manner, providing a reliable data foundation for subsequent cognitive analysis.

[0025] The cognitive evolution module (3) is used to continuously improve the intelligence level of the cognitive twin. This module updates and trains the generative artificial intelligence model in the cognitive kernel module (1) regularly or in real time according to the execution results fed back by the support decision module (4) and the human-machine interaction module (5). For example, when the conclusion of a certain fault diagnosis is inconsistent with the actual maintenance result, the cognitive evolution module will use this difference as a training sample to adjust the model parameters or add new knowledge to correct the model cognitive bias. The cognitive evolution module (3) enables the system to have self-learning ability, and the newly obtained sensing data, fault cases and expert feedback will be used to enrich the model's knowledge base, thereby realizing the self-adaptive evolution of the cognitive twin. Through this continuous optimization mechanism, the system can more and more accurately predict faults and develop optimization strategies, and keep synchronized with the actual state of the equipment and the latest field knowledge.

[0026] The support decision module (4) is responsible for converting cognitive results into specific support action plans. This module receives diagnosis and analysis results from the cognitive kernel module (1), combines pre-defined support goals (such as maximizing reliability, minimizing downtime, etc.), and generates corresponding support strategies through semantic reasoning and decision algorithms. For example, when the cognitive kernel judges that a component may experience performance degradation, the support decision module can generate a preventive maintenance plan, suggesting the replacement of specific parts or the adjustment of usage parameters; if a fault is identified, the module will provide suggestions for fault location and repair steps. In addition, the support decision module (4) uses feedback information to continuously correct decision rules: when a suggested support measure does not work well, the system will adjust the strategy generation logic with the support of the cognitive evolution module (3). The plans output by the support decision module are presented to the user through the human-machine interaction module (5) and can be implemented by interfacing with external maintenance execution systems.

[0027] The human-computer interaction module (5) provides an interface for information exchange between the system and the user. The human-computer interaction module (5) supports multiple interaction methods, including a graphical user interface, a mobile terminal application, and a voice / text dialogue interface. Maintenance personnel can query the equipment digital twin state (such as key component health indicators, current fault warnings, etc.) through the human-computer interaction module, and obtain the maintenance plan and operation guidance given by the maintenance decision-making module (4). The natural language processing function of the human-computer interaction module (5) allows users to ask about equipment conditions or seek decision suggestions in spoken or written language. The cognitive kernel module (1) of the system understands the query intent and gives the corresponding reply. At the same time, the human-computer interaction module (5) records the actual measures taken by the user and the feedback results, such as whether the maintenance is successful, whether the fault is eliminated, etc. These feedback data will be transmitted back to the cognitive evolution module (3) for model updating, forming a closed-loop optimization of the system.

[0028] Through a friendly human-computer interface and intelligent interaction means, the equipment intelligent support system based on generative cognitive twin provided by the application is convenient for frontline support personnel to use, and enhances the human-computer collaboration effect.

[0029] In actual application, the equipment intelligent support system based on generative cognitive twin of the application works collaboratively, and realizes the complete process of equipment intelligent support. As shown in Figure 2 The method flow includes the following steps: 1. Data acquisition and fusion: The real-time running data and environmental parameters of the equipment are collected through the digital mapping module (2), and the relevant historical maintenance records and technical materials are obtained from the knowledge base. The multi-source information is cleaned, analyzed and fused to form a comprehensive description of the equipment state.

[0030] 2. Cognitive analysis and diagnosis: The cognitive kernel module (1) performs in-depth analysis on the fused information, including anomaly detection, pattern recognition and causal reasoning, to generate the current health state evaluation, potential fault diagnosis result and performance trend prediction of the equipment.

[0031] 3. Strategy generation: The maintenance decision-making module (4) formulates corresponding support strategies according to the cognitive analysis results. This includes determining the maintenance priority, selecting the best maintenance plan, arranging spare parts and resources, and optimizing the operation parameters. The strategy formulation takes into account historical successful experience and established support constraints.

[0032] 4. Scheme execution and interaction: The generated support scheme is presented to the user through the human-computer interaction module (5). Maintenance personnel can view the specific measures and steps suggested, and confirm or adjust the execution plan. During the execution process, the system can continue to monitor the equipment state through sensors, and the user can also report additional conditions through the interface.

[0033] 5. Feedback and model updating: After the maintenance action is completed, the system collects actual results and feedback data. For example, record the replaced parts, solved problems, and unsolved residual problems, etc. The cognitive evolution module (3) uses these feedbacks to update the model parameters or knowledge base of the cognitive kernel module (1). If a new failure mode is found, it will be included in the training, thereby improving the system's ability to predict and respond to similar problems. Return to step 1 in the subsequent operation and continue the above process.

[0034] As an embodiment, the equipment intelligent support system based on generative cognitive twin provided by the present application is applied to the intelligent support of an aero-engine. When the sensor detects abnormal vibration during engine operation, and relevant alarm information appears in the log record, the cognitive kernel module (1) combines sensor data and maintenance manual knowledge to infer and judge that it may be caused by imbalance due to fouling or wear of a certain compressor blade. The support decision module (4) generates a maintenance scheme according to the problem obtained by inference and judgment, and suggests checking and cleaning the corresponding part at the next shutdown window, and pre-allocates the required spare parts. Through the human-computer interaction module (5), the engineer receives this suggestion and performs the operation. After completing the maintenance, the system records that the engine vibration index returns to normal, verifies the effectiveness of the decision, and feeds back the fouling fault case and solution to the cognitive evolution module (3). The system updates according to the fouling fault case and solution, and in the future, if similar vibration signal characteristics appear, the system will diagnose and propose an optimized support strategy more quickly. This closed-loop process reflects the intelligent and adaptive advantages of the equipment intelligent support system based on generative cognitive twin provided by the present application in complex equipment support.

[0035] In summary, the present application realizes a new architecture of the equipment intelligent support system by integrating generative artificial intelligence into digital twin technology. The system can autonomously understand multi-source information, infer complex faults, and optimize decision schemes, significantly improving the intelligent level of equipment operation and maintenance support. It should be noted that the above embodiments are only used to illustrate the principles and advantages of the present application. Those skilled in the art can make various modifications or supplements to the specific implementation methods or use similar ways to replace them without deviating from the concept of the present application or exceeding the scope defined by the appended claims.

[0036] Those skilled in the art can make various modifications or supplements to the specific embodiments described or use similar ways to replace them without deviating from the concept of the present application or exceeding the scope defined by the appended claims.

Claims

1. An equipment intelligent support system based on generative cognitive twin, characterized in that, Includes the following modules: The cognitive kernel module has a built-in generative artificial intelligence model and uses it to perform semantic understanding and causal reasoning on multimodal data and unstructured knowledge from the equipment, constructs a cognitive twin representation of the equipment, and outputs the equipment's state diagnosis and prediction results. The digital mapping module collects multi-source data from the equipment and updates the digital twin model based on the multi-source data to reflect the real-time status of the equipment, thereby achieving virtual-real synchronization between the physical equipment and the cognitive twin status; the multi-source data includes sensor signals, images, and operation logs. The cognitive evolution module trains and updates the artificial intelligence model in the cognitive kernel module based on new data during equipment support and feedback from human-computer interaction, optimizes the knowledge and model parameters of the cognitive twin, realizes the adaptive evolution and continuous learning of the model, and corrects cognitive biases to maintain consistency between virtual and real cognition. The support decision module receives the diagnostic and analysis results from the cognitive kernel module, generates corresponding equipment support strategies and decision suggestions based on the diagnostic and analysis results from the cognitive kernel module through semantic reasoning, and evaluates the effectiveness of the strategies based on feedback after execution. The equipment support strategies and decision-making recommendations include fault diagnosis, maintenance plan development, and performance optimization. The human-computer interaction module provides a user interface for interacting with the system, displaying the equipment support plan given by the support decision module in a visual and natural language manner. Maintenance personnel can query equipment status, obtain support suggestions, receive user confirmation and correction opinions, and collect actual execution feedback and enter the cognitive evolution module to realize human-computer collaboration and decision-making closed loop.

2. The equipment intelligent support system based on generative cognitive twin according to claim 1, wherein, The cognitive kernel module includes a generative pre-trained model trained on equipment domain data. This model has cross-modal information fusion and semantic understanding capabilities, and can simultaneously process numerical sensor data, image signals and textual knowledge to identify potential equipment failure modes and perform causal inference to understand the causes of potential equipment failures.

3. The equipment intelligent support system based on generative cognitive twin according to claim 1, wherein, The digital mapping module includes: a data acquisition unit for real-time acquisition of equipment operating parameters and environmental parameters; a data preprocessing unit for cleaning, format conversion, and time alignment of data from different sources; and a twin synchronization unit for updating the processed data to the state variables of the cognitive twin model.

4. The equipment intelligent support system based on generative cognitive twins according to claim 1, characterized in that, The cognitive evolution module is equipped with a cognitive consistency correction mechanism. When the reasoning result output by the cognitive kernel module is inconsistent with the actual observation and / or pre-stored domain knowledge, it triggers model adjustment and / or data supplementation to correct the cognitive bias of the cognitive twin. The cognitive evolution module incorporates new fault cases after each equipment maintenance through incremental training, continuously improving the model's adaptability to new problems.

5. The equipment intelligent support system based on generative cognitive twins according to claim 1, characterized in that, The support decision module uses knowledge graph reasoning, rule engine and reinforcement learning algorithm to make decision calculations on the equipment status information provided by the cognitive kernel module. The generated support strategy includes maintenance steps, preventive maintenance plan, spare parts scheduling scheme and operation optimization suggestions under the fault diagnosis conclusion, and feeds back the execution effect to update the decision logic.

6. The equipment intelligent support system based on generative cognitive twins according to claim 1, characterized in that, The human-computer interaction module includes a natural language processing unit, enabling maintenance personnel to interact with the system via voice and text to obtain explanatory equipment status descriptions and support suggestions; and the human-computer interaction module parses the information fed back by the user and transmits it to the cognitive evolution module to improve the knowledge base of the cognitive twin.

7. A method for intelligent equipment support based on generative cognitive twins, characterized in that, Includes the following steps: 1) Data acquisition and twin synchronization: Collect multimodal operation data of equipment through digital mapping module, combine it with historical knowledge base information, update the cognitive twin model of equipment, and form a complete description of equipment status; 2) Cognitive analysis: The cognitive kernel module performs semantic understanding and causal reasoning on the comprehensive information obtained in step 1) to obtain the equipment's health status assessment, fault diagnosis conclusions, or performance prediction results; 3) Strategy Formulation: Based on the analysis results of step 2), the support decision module generates corresponding equipment support strategies and maintenance plans, including the required operational measures, resource arrangements, and execution timing; 4) Execution and Interaction: The strategy and solution are provided to the equipment maintenance personnel through the human-computer interaction module. The maintenance personnel can confirm and adjust the solution according to the actual situation. During the execution process, the system continuously monitors the equipment status and accepts supplementary information input by the user. 5) Feedback Learning: After the safeguard measures are implemented, the actual results and feedback data are collected. The cognitive evolution module updates and trains the cognitive twin's model and knowledge, incorporating new fault phenomena and handling experiences into the learning cycle, thereby entering the next round of equipment status data acquisition and analysis, and improving the intelligent optimization level of the safeguard strategy.

8. The intelligent equipment support method based on generative cognitive twins according to claim 7, characterized in that, In the cognitive analysis process of step 2), feature extraction and preliminary analysis are performed on the different modal data obtained: anomaly detection and trend analysis are performed on the sensor time series data, target recognition and status assessment are performed on the device images, and natural language processing is applied to the text logs to extract key events and parameters. The results of each modality analysis are then fused through temporal and semantic correlation and uniformly input into the generative artificial intelligence model to produce the final diagnostic and prediction results.

9. The intelligent equipment support method based on generative cognitive twins according to claim 7, characterized in that, In the feedback learning process of step 5), when the cognitive kernel module diagnoses an error and / or the strategy is ineffective, the cognitive evolution module extracts the corresponding case and uses it together with the original training data to incrementally train and / or fine-tune the generative artificial intelligence model, thereby enhancing the model's accuracy in similar scenarios and adjusting the rule parameters of the safeguard decision module to avoid similar errors.

Images