AI Intelligent Valve Actuator Training Method and System Based on Multimodal Large Model

Abstract

The present invention relates to the technical field of AI intelligent valve actuator training, and particularly to an AI intelligent valve actuator training method and system based on a multimodal large model. The system includes a training processor, a training error analysis unit, a progressive evaluation unit, a simulation evaluation unit, an autonomous training management unit, and a training management unit. The present invention initially analyzes from the perspective of the training reading rate during the valve actuator training process. On the one hand, it helps to ensure the effectiveness of the constructed valve dynamic response model. On the other hand, it helps to conduct safety supervision on the valve actuator training process to ensure the normal operation of the valve actuator, thereby reducing the failure rate of valve actuator training. Furthermore, it analyzes from the perspective of the versatility of the valve actuator to improve the current valve actuator's ability to accurately control intelligent valves in different scenarios. At the same time, it analyzes from the perspective of simulation testing to ensure that the output action results are always within the error range of the actual results.

Description

Technical Field

[0001] The present invention relates to the technical field of AI intelligent valve actuator training, and particularly to an AI intelligent valve actuator training method and system based on a multimodal large model. Background Art

[0002] A valve actuator is a control actuator for opening and closing a valve. Its structure generally includes a power unit, a control unit, and an execution unit. It is generally applied to fluid pipelines, especially liquid pipeline systems. In case of an emergency, relevant operators cannot reach the site to open and close the valve in a timely manner, or personnel cannot approach the valve site. At the same time, when the power unit loses power or gas source, the actuator can drive the valve to a pre-set safe state;

[0003] However, in the prior art, during the training process of an AI intelligent valve actuator, the training status of the AI intelligent valve actuator cannot be accurately analyzed, which increases the risk of training errors of the AI intelligent valve actuator. At the same time, it is not conducive to the precise control of the AI intelligent valve actuator in different scenarios, and the training accuracy of the AI intelligent valve actuator cannot be verified from the perspective of simulation, resulting in an increase in execution errors. At the same time, as the AI intelligent valve degenerates, the training cycle of the AI intelligent valve actuator cannot be flexibly adjusted independently, resulting in a decrease in the control accuracy of the AI intelligent valve actuator;

[0004] In view of the above technical deficiencies, a solution is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide an AI intelligent valve actuator training method and system based on a multimodal large model to solve the above-mentioned technical deficiencies. The present invention initially analyzes from the perspective of the training reading rate during the training process of the AI intelligent valve actuator. On the one hand, it helps to ensure the effectiveness of the constructed valve dynamic response model. On the other hand, it helps to conduct safety supervision on the training process of the AI intelligent valve actuator to ensure the normal operation of the AI intelligent valve actuator, thereby reducing the failure rate of the AI intelligent valve actuator training. Further analysis is carried out from the perspective of the versatility of the AI intelligent valve actuator to adjust the current training scope of the AI intelligent valve actuator to improve the precise control of the current AI intelligent valve actuator for the AI intelligent valve in different scenarios. At the same time, analysis is carried out from the perspective of simulation testing to ensure that the output action result always remains within the error range compared with the actual result. By deeply analyzing the comprehensive performance of the AI intelligent valve actuator, a reasonable independent training cycle adjustment is made based on the performance degradation of the AI intelligent valve to improve the continuous control quality and accuracy of the AI intelligent valve actuator.

[0006] The object of the present invention can be achieved by the following technical solutions: an AI intelligent valve actuator training system based on a multi-modal large model, including a training processor, a training error analysis unit, a progressive evaluation unit, a simulation evaluation unit, an autonomous training management unit, and a training management unit;

[0007] The training processor is used to retrieve the historical operation data of the AI intelligent valve actuator and send the historical operation data to the training error analysis unit for training error tracking and evaluation analysis to obtain an abnormal signal or a normal signal. When a normal signal is generated, the progressive evaluation unit is used to perform training discrimination processing on the output result of the adaptive training of the collected AI intelligent valve actuator to obtain a comprehensive signal or a defect signal, where the adaptive training includes closed-loop test training and fault tolerance training;

[0008] When a normal signal is generated, the simulation evaluation unit is used to perform simulation test feedback analysis on the measured error value of the obtained AI intelligent valve actuator to obtain a stable signal or a deviation signal. The autonomous training management unit is used to obtain the comprehensive performance score of the AI intelligent valve and at the same time perform autonomous training update cycle regulation analysis on the comprehensive performance score to obtain an autonomous training plan update cycle.

[0009] Preferably, the training error tracking and evaluation analysis process is as follows:

[0010] Collect the training period of the AI intelligent valve actuator, set the training period of the AI intelligent valve actuator as a time threshold, obtain the historical operation data of the AI intelligent valve actuator within the time threshold, and the historical operation data includes pressure, flow rate, and stroke position. Based on the historical operation data and using an LSTM neural network to construct a valve dynamic response model;

[0011] Obtain the training reading rate curve of the AI intelligent valve actuator within the time threshold, obtain the maximum peak value and the minimum valley value from the training reading rate curve, obtain the difference between the maximum peak value and the minimum valley value, and perform discrimination processing on the difference between the maximum peak value and the minimum valley value. If the difference between the maximum peak value and the minimum valley value is less than a preset threshold, a feedback instruction is generated;

[0012] When a feedback instruction is generated, obtain the average value after adding the maximum peak value and the minimum valley value, and perform discrimination processing on the average value after adding the maximum peak value and the minimum valley value to obtain an abnormal signal or a normal signal.

[0013] Preferably, the training discrimination processing process is as follows:

[0014] Obtain the output results of the adaptive training of the AI intelligent valve actuator within the time threshold. The adaptive training includes closed-loop test training and fault tolerance training. The output results include qualified and unqualified. Obtain the number corresponding to the qualified output results of the adaptive training of the AI intelligent valve actuator, and perform discriminant processing on the number corresponding to the qualified output results of the adaptive training of the AI intelligent valve actuator to obtain a comprehensive signal or a defect signal.

[0015] Preferably, the simulation test feedback analysis process is as follows:

[0016] Build a dynamic physical model in the simulation environment. At the same time, obtain the visual image of the AI intelligent valve under the current set environment, and at the same time obtain the input text instruction. Generate a guidance instruction based on the visual image and the text instruction. Input the visual image, the text instruction, and the guidance instruction into the valve dynamic response model to obtain the action instruction output by the valve dynamic response model. Among them, the text instruction is the task objective for the AI intelligent valve actuator to execute the task.

[0017] Preferably, obtain the measured error value between the actual response action instruction of the AI intelligent valve actuator in the dynamic physical model and the action instruction output by the valve dynamic response model, and perform discriminant processing on the measured error value. If the measured error value is less than or equal to the preset measured error value threshold, generate a stable signal. If the measured error value is greater than the preset measured error value threshold, generate a deviation signal.

[0018] Preferably, the autonomous training update cycle regulation analysis process is as follows: Obtain the historical working condition data and sensor monitoring data of the AI intelligent valve within the time threshold. The historical working condition data includes the opening and closing times, torque value, and response time. The sensor monitoring data includes vibration and temperature. Perform preprocessing on the historical working condition data and sensor monitoring data. The preprocessing includes cleaning and data enhancement. Build a comprehensive performance scoring model based on the preprocessed historical working condition data and sensor monitoring data.

[0019] Preferably, obtain the comprehensive performance score of the AI intelligent valve within the time threshold based on the comprehensive performance scoring model, and perform discriminant processing on the comprehensive performance score. If the comprehensive performance score is less than the preset comprehensive performance score threshold, obtain the change amount of the comprehensive performance score, and perform discriminant processing on the change amount of the comprehensive performance score: If the change amount of the comprehensive performance score is less than the preset threshold, it is determined as first-level degradation. If the change amount of the comprehensive performance score is greater than or equal to the preset threshold, it is determined as second-level degradation. Obtain the preset cycle reduction value corresponding to the first-level degradation or second-level degradation. Obtain the autonomous training update cycle of the AI intelligent valve actuator within the time threshold. Set the value obtained by subtracting the product of the autonomous training update cycle and the preset cycle reduction value from the autonomous training update cycle as the autonomous training planning update cycle.

[0020] An AI intelligent valve actuator training method based on a multimodal large model includes the following steps:

[0021] Step 1: The process of evaluating the training error of the valve dynamic response model in the early stage;

[0022] Step 2: Evaluate and process the training scope of the AI intelligent valve actuator from the perspective of versatility through an information progression method;

[0023] Step 3: The operation of supervising the execution error of the AI intelligent valve actuator based on the simulation environment;

[0024] Step 4: Construct a comprehensive performance scoring model for the AI intelligent valve and an autonomous training cycle regulation process based on the analysis of the comprehensive performance scoring model.

[0025] The beneficial effects of the present invention are as follows:

[0026] The present invention initially analyzes from the perspective of the training reading rate in the training process of the AI intelligent valve actuator. On the one hand, it helps to ensure the effectiveness of the constructed valve dynamic response model. On the other hand, it helps to conduct safety supervision on the training process of the AI intelligent valve actuator to ensure the normal operation of the AI intelligent valve actuator, thereby reducing the failure rate of the training of the AI intelligent valve actuator. Further analysis is carried out from the perspective of the versatility of the AI intelligent valve actuator to adjust the current training scope of the AI intelligent valve actuator, so as to improve the ability of the current AI intelligent valve actuator to accurately control the AI intelligent valve in different scenarios;

[0027] The present invention analyzes from the perspective of simulation testing to ensure that the output action result is always within the error range of the actual result. By deeply analyzing the comprehensive performance of the AI intelligent valve, a reasonable autonomous training cycle adjustment can be made according to the performance degradation of the AI intelligent valve, so as to improve the continuous control quality and accuracy of the AI intelligent valve actuator. Description of the Drawings

[0028] The following further describes the present invention with reference to the drawings;

[0029] Figure 1 is the system flow block diagram of the present invention;

[0030] Figure 2 is the reference analysis diagram of the method of the present invention. Detailed Embodiments

[0031] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Embodiment

[0032] Please refer to Figures 1 to 2 As shown, the present invention is an AI intelligent valve actuator training system based on a multimodal large model, including a training processor, a training error analysis unit, a progressive evaluation unit, a simulation evaluation unit, an autonomous training management unit, and a training management unit. The training processor is unidirectionally communicatively connected to the training error analysis unit, the training error analysis unit is bidirectionally communicatively connected to the progressive evaluation unit, the training error analysis unit is unidirectionally communicatively connected to both the simulation evaluation unit and the training management unit, the simulation evaluation unit is unidirectionally communicatively connected to both the autonomous training management unit and the training management unit, and the autonomous training management unit is unidirectionally communicatively connected to the training management unit;

[0033] The training processor is used to retrieve the historical operation data of the AI intelligent valve actuator and send the historical operation data to the training error analysis unit. The training error analysis unit is used to perform training error tracking and evaluation analysis on the received historical operation data, and perform discriminant processing on the mean value obtained by adding the maximum peak value and the minimum valley value to understand whether the construction of the valve dynamic response model is effective, so as to reduce the failure rate of training;

[0034] The specific process of training error tracking and evaluation analysis is as follows: collect the training period of the AI intelligent valve actuator and set the training period of the AI intelligent valve actuator as the time threshold;

[0035] Obtain the historical operation data of the AI intelligent valve actuator within the time threshold. The historical operation data includes time series data such as pressure, flow rate, and stroke position. Based on the historical operation data, construct a valve dynamic response model using an LSTM neural network;

[0036] Obtain the training reading rate curve of the AI intelligent valve actuator within the time threshold, obtain the maximum peak value and the minimum valley value from the training reading rate curve, obtain the difference between the maximum peak value and the minimum valley value, and perform discrimination processing on the difference between the maximum peak value and the minimum valley value. If the difference between the maximum peak value and the minimum valley value is less than the preset threshold, generate a feedback instruction. When the feedback instruction is generated, obtain the average value after adding the maximum peak value and the minimum valley value, and perform discrimination processing on the average value after adding the maximum peak value and the minimum valley value. If the average value after adding the maximum peak value and the minimum valley value is less than the preset threshold, generate an abnormal signal, and it will be determined that the construction of the valve dynamic response model is abnormal. If the average value after adding the maximum peak value and the minimum valley value is greater than or equal to the preset threshold, generate a normal signal, and it will be determined that the construction of the valve dynamic response model is normal. Send the abnormal signal or the normal signal to the training management unit. After receiving the abnormal signal or the normal signal, the training management unit immediately displays the preset warning text corresponding to the abnormal signal or the normal signal to ensure the effectiveness of the construction of the valve dynamic response model and reduce the failure rate of training at the same time. Embodiment

[0037] When the normal signal is generated, the progressive evaluation unit is used to collect the output results of the adaptive training of the AI intelligent valve actuator and perform training discrimination processing on the output results, so as to adjust the training range of the current AI intelligent valve actuator to improve the accuracy of the current AI intelligent valve actuator in controlling the AI intelligent valve in different scenarios, which helps to improve the versatility and comprehensiveness of the AI intelligent valve actuator. The specific training discrimination processing process is as follows: Obtain the output results of the adaptive training of the AI intelligent valve actuator within the time threshold. The adaptive training includes closed-loop test training, fault tolerance training, etc. The output results include qualified and unqualified. Obtain the number corresponding to the output result of the adaptive training of the AI intelligent valve actuator being qualified, and perform discrimination processing on the number corresponding to the output result of the adaptive training of the AI intelligent valve actuator being qualified:

[0038] If the number corresponding to the output result of the adaptive training of the AI intelligent valve actuator being qualified is equal to the preset threshold, generate a comprehensive signal;

[0039] If the number corresponding to the output result of the adaptive training of the AI intelligent valve actuator being qualified is equal to the preset threshold, generate a defect signal. Send the comprehensive signal or the defect signal to the training management unit through the training error analysis unit. After receiving the comprehensive signal or the defect signal, the training management unit immediately displays the preset warning text corresponding to the comprehensive signal or the defect signal to adjust the training range of the current AI intelligent valve actuator to improve the accuracy of the current AI intelligent valve actuator in controlling the AI intelligent valve in different scenarios, which helps to improve the versatility and comprehensiveness of the AI intelligent valve actuator;

[0040] Among them, closed-loop test training means testing and training the regulation accuracy of the AI intelligent valve actuator under different set load disturbances, and fault tolerance training means testing and training the operating state of the AI intelligent valve actuator when the upper computer instruction is lost;

[0041] When a normal signal is generated, the simulation evaluation unit is used to collect the measured error value of the AI intelligent valve actuator and conduct a simulation test feedback analysis on the measured error value of the AI intelligent valve actuator, that is, analyze from the perspective of simulation execution accuracy to determine whether the training accuracy of the AI intelligent valve actuator is qualified, so as to reduce the error risk by means of increasing training samples or reducing model complexity, etc. The specific simulation test feedback analysis process is as follows: construct a dynamic physical model in the simulation environment, and at the same time obtain the visual image of the AI intelligent valve under the current set environment, and at the same time obtain the input text instruction, generate a guiding instruction based on the visual image and the text instruction, and input the visual image, the text instruction and the guiding instruction into the valve dynamic response model to obtain the action instruction output by the valve dynamic response model. Among them, the text instruction is the task target for the AI intelligent valve actuator to execute the task;

[0042] It should be noted that the multi-modal content composed of visual images and text instructions is used as the input. The visual image gives the model the basic function of perceiving the execution environment of the AI intelligent valve actuator, while the text instruction is the task target for the AI intelligent valve actuator to execute the task, and the guiding instruction of the relevant callable API is used as the communication bridge between natural language and the execution control of the AI intelligent valve actuator, so as to realize controlling the action of the AI intelligent valve actuator by using the language ability of the execution large model;

[0043] In the process of understanding the task text instruction, multi-modal matching is of utmost importance. The AI intelligent valve actuator itself does not have the ability of semantic understanding. Therefore, for the object to be operated, the object needs to retrieve the corresponding image through text description, and then further obtain its three-dimensional real space position through methods such as camera calibration, and hand it over to the AI intelligent valve actuator for execution;

[0044] Obtain the measured error value between the actual response action instruction of the AI intelligent valve actuator in the dynamic physical model and the action instruction output by the valve dynamic response model, and perform discriminant processing on the measured error value. If the measured error value is less than or equal to the preset measured error value threshold, a stable signal is generated. If the measured error value is greater than the preset measured error value threshold, a deviation signal is generated. Send the stable signal or deviation signal to the training management unit. After receiving the stable signal or deviation signal, the training management unit immediately determines the preset warning text corresponding to the stable signal or deviation signal, so as to reduce the error risk by means such as increasing training samples or reducing model complexity, thereby helping to improve the control accuracy of the AI intelligent valve actuator and ensuring that the output action result and the actual result always remain within the error range. Embodiment

[0045] When a stable signal or a comprehensive signal is generated, the autonomous training management unit is used to obtain the comprehensive performance score of the AI intelligent valve, and at the same time, perform autonomous training update cycle regulation analysis on the comprehensive performance score, so as to reasonably adjust the training cycle according to the performance degradation of the AI intelligent valve, in order to improve the continuous control quality and accuracy of the AI intelligent valve actuator. The specific process of autonomous training update cycle regulation analysis is as follows: Obtain the historical working condition data and sensor monitoring data of the AI intelligent valve within the time threshold. The historical working condition data includes the number of opening and closing times, torque value, response time, etc., and the sensor monitoring data includes vibration, temperature, etc. Perform preprocessing on the historical working condition data and sensor monitoring data. The preprocessing includes cleaning, data enhancement, etc. Build a comprehensive performance score model based on the preprocessed historical working condition data and sensor monitoring data;

[0046] Based on the comprehensive performance scoring model, the comprehensive performance score of the AI intelligent valve within the time threshold is obtained. The comprehensive performance score is discriminated. If the comprehensive performance score is less than the preset comprehensive performance score threshold, the change amount of the comprehensive performance score is obtained, and the change amount of the comprehensive performance score is discriminated: if the change amount of the comprehensive performance score is less than the preset threshold, it is determined as first-level degradation; if the change amount of the comprehensive performance score is greater than or equal to the preset threshold, it is determined as second-level degradation. It should be noted that the performance degradation degrees corresponding to the first-level degradation and the second-level degradation increase in turn. The preset cycle reduction value corresponding to the first-level degradation or the second-level degradation is obtained, and the autonomous training update cycle of the AI intelligent valve actuator within the time threshold is obtained. The value obtained by subtracting the product of the autonomous training update cycle and the preset cycle reduction value from the autonomous training update cycle is set as the autonomous training planning update cycle, and the autonomous training planning update cycle is sent to the training management unit. After receiving the autonomous training planning update cycle, the training management unit conducts autonomous training on the AI intelligent valve actuator according to the autonomous training planning update cycle, that is, reasonably adjusts the training cycle according to the performance degradation of the AI intelligent valve to improve the control quality and accuracy of the AI intelligent valve actuator. Embodiment

[0047] An AI intelligent valve actuator training method based on a multi-modal large model includes the following steps:

[0048] Step 1: The process of evaluating the training error of the valve dynamic response model in the early stage, that is, tracking and evaluating the training error of the historical operation data, and discriminating the mean value obtained by adding the maximum peak value and the minimum valley value to determine whether the training of the AI intelligent valve actuator is normal to ensure the effectiveness of the valve dynamic response model;

[0049] Step 2: Evaluate the training range of the AI intelligent valve actuator from the perspective of versatility through the way of information progression, that is, conduct training discrimination on the output result of the adaptive training of the AI intelligent valve actuator to judge whether the versatility and comprehensiveness of the training of the AI intelligent valve actuator are qualified;

[0050] Step 3: The operation of supervising the execution error of the AI intelligent valve actuator in the simulation environment, that is, conduct simulation test feedback analysis on the measured error value of the AI intelligent valve actuator to judge whether the training accuracy of the AI intelligent valve actuator is qualified;

[0051] Step 4: Construct a comprehensive performance scoring model for the AI intelligent valve and an autonomous training cycle regulation process based on the analysis of the comprehensive performance scoring model, that is, preprocess the historical working condition data and sensor monitoring data to construct a comprehensive performance scoring model, and conduct autonomous training update cycle regulation analysis based on the comprehensive performance score output by the comprehensive performance scoring model, and perform autonomous training on the valve actuator according to the obtained autonomous training plan update cycle;

[0052] In summary, the present invention initially analyzes from the perspective of the training reading rate in the training process of the AI intelligent valve actuator. On the one hand, it helps to ensure the effectiveness of the constructed valve dynamic response model. On the other hand, it helps to conduct safety supervision on the training process of the AI intelligent valve actuator to ensure the normality of the AI intelligent valve actuator, thereby reducing the failure rate of the AI intelligent valve actuator training. Further analysis is carried out from the perspective of the versatility of the AI intelligent valve actuator to adjust the current training scope of the AI intelligent valve actuator to improve the accuracy of the current AI intelligent valve actuator in precisely controlling the AI intelligent valve in different scenarios. At the same time, analysis is carried out from the perspective of simulation testing to ensure that the output action result always remains within the error range compared with the actual result. By deeply analyzing the comprehensive performance of the AI intelligent valve, reasonable autonomous training cycle adjustment is carried out according to the performance degradation of the AI intelligent valve to improve the continuous control quality and accuracy of the AI intelligent valve actuator.

[0053] The setting of the threshold value is for the convenience of comparison. Regarding the size of the threshold value, it depends on the amount of sample data and the base number set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantified value.

[0054] The size of the coefficient is a specific value obtained by quantifying each parameter for subsequent comparison. Regarding the size of the coefficient, it depends on the amount of sample data and the corresponding operating coefficient initially set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantified value.

[0055] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and all should be covered within the protection scope of the present invention.

Claims

1. An AI intelligent valve actuator training system based on a multimodal large model, characterized in that, It includes a training processor, a training error analysis unit, a progressive evaluation unit, a simulation evaluation unit, an autonomous training management unit, and a training management unit; The training processor is used to retrieve the historical operation data of the AI intelligent valve actuator and send the historical operation data to the training error analysis unit for training error tracking and evaluation analysis to obtain an abnormal signal or a normal signal. When a normal signal is generated, the progressive evaluation unit is used to perform training discrimination processing on the output result of the adaptive training of the collected AI intelligent valve actuator to obtain a comprehensive signal or a defect signal, where the adaptive training includes closed-loop test training and fault tolerance training; The comprehensive signal or the defect signal is sent to the training management unit through the training error analysis unit. After receiving the comprehensive signal or the defect signal, the training management unit immediately displays the preset warning text corresponding to the comprehensive signal or the defect signal so as to adjust the current training range of the AI intelligent valve actuator; When a normal signal is generated, the simulation evaluation unit is used to perform simulation test feedback analysis on the measured error value of the obtained AI intelligent valve actuator to obtain a stable signal or a deviation signal. The autonomous training management unit is used to obtain the comprehensive performance score of the AI intelligent valve and at the same time perform autonomous training update cycle regulation analysis on the comprehensive performance score to obtain the autonomous training plan update cycle; The process of the training error tracking and evaluation analysis is as follows: Collect the training period of the AI intelligent valve actuator, set the training period of the AI intelligent valve actuator as the time threshold, and obtain the historical operation data of the AI intelligent valve actuator within the time threshold. The historical operation data includes pressure, flow, and stroke position. Based on the historical operation data and using the LSTM neural network to construct a valve dynamic response model; Obtain the training reading rate curve of the AI intelligent valve actuator within the time threshold, obtain the maximum peak value and the minimum valley value from the training reading rate curve, obtain the difference between the maximum peak value and the minimum valley value, and perform discrimination processing on the difference between the maximum peak value and the minimum valley value. If the difference between the maximum peak value and the minimum valley value is less than the preset threshold, a feedback instruction is generated; When a feedback instruction is generated, obtain the mean value after adding the maximum peak value and the minimum valley value, and perform discrimination processing on the mean value after adding the maximum peak value and the minimum valley value. If the mean value after adding the maximum peak value and the minimum valley value is less than the preset threshold, an abnormal signal is generated, and it is determined that the construction of the dynamic response model is abnormal. If the mean value after adding the maximum peak value and the minimum valley value is greater than or equal to the preset threshold, a normal signal is generated, and it is determined that the construction of the dynamic response model is normal.

2. The AI intelligent valve actuator training system based on the multi-modal large model according to claim 1, characterized in that The process of the training discrimination processing is as follows: Obtain the output result of the adaptive training of the AI intelligent valve actuator within the time threshold. The output result includes qualified and unqualified. Obtain the number corresponding to the qualified output result of the adaptive training of the AI intelligent valve actuator, and perform discrimination processing on the number corresponding to the qualified output result of the adaptive training of the AI intelligent valve actuator to obtain a comprehensive signal or a defect signal.

3. The AI intelligent valve actuator training system based on a multimodal large model according to claim 2, characterized in that, The simulation test feedback analysis process is as follows: A dynamic physical model is constructed in the simulation environment. At the same time, the visual image of the AI intelligent valve under the current set environment is obtained, and the input text instruction is obtained. A guidance instruction based on the visual image and the text instruction is generated. The visual image, the text instruction, and the guidance instruction are input into the valve dynamic response model to obtain the action instruction output by the valve dynamic response model. Among them, the text instruction is the task objective for the AI intelligent valve actuator to execute the task.

4. The AI intelligent valve actuator training system based on a multimodal large model according to claim 3, characterized in that, The measured error value between the actual response action instruction of the AI intelligent valve actuator in the dynamic physical model and the action instruction output by the valve dynamic response model is obtained, and the measured error value is judged and processed. If the measured error value is less than or equal to the preset measured error value threshold, a stable signal is generated. If the measured error value is greater than the preset measured error value threshold, a deviation signal is generated.

5. The AI intelligent valve actuator training system based on a multimodal large model according to claim 2, characterized in that, The autonomous training update cycle regulation analysis process is as follows: The historical working condition data and sensor monitoring data of the AI intelligent valve within the time threshold are obtained. The historical working condition data includes the opening and closing times, torque value, and response time. The sensor monitoring data includes vibration and temperature. The historical working condition data and sensor monitoring data are preprocessed. The preprocessing includes cleaning and data enhancement. A comprehensive performance scoring model is constructed based on the preprocessed historical working condition data and sensor monitoring data.

6. The AI intelligent valve actuator training system based on the multi-modal large model according to claim 5, characterized in that, Based on the comprehensive performance scoring model, the comprehensive performance score of the AI intelligent valve within the time threshold is obtained, and the comprehensive performance score is judged and processed. If the comprehensive performance score is less than the preset comprehensive performance score threshold, the change amount of the comprehensive performance score is obtained, and the change amount of the comprehensive performance score is judged and processed: If the change amount of the comprehensive performance score is less than the preset threshold, it is determined as first-level degradation. If the change amount of the comprehensive performance score is greater than or equal to the preset threshold, it is determined as second-level degradation. The preset cycle reduction value corresponding to the first-level degradation or second-level degradation is obtained. The autonomous training update cycle of the AI intelligent valve actuator within the time threshold is obtained. The value obtained by subtracting the product of the autonomous training update cycle and the preset cycle reduction value from the autonomous training update cycle is set as the autonomous training plan update cycle.

7. An AI intelligent valve actuator training method based on a multimodal large model, which is applied to the AI intelligent valve actuator training system based on the multimodal large model described in any one of claims 1-6, and is characterized in that, It includes the following steps: Step 1: The process of evaluating the training error of the valve dynamic response model in the early stage; Step 2: The process of evaluating the training range of the AI intelligent valve actuator from the perspective of versatility in a way of information progression; Step 3: The operation of supervising the execution error of the AI intelligent valve actuator under the simulation environment; Step 4: Constructing the comprehensive performance scoring model of the AI intelligent valve and the autonomous training cycle regulation process based on the analysis of the comprehensive performance scoring model.

Images